USF Libraries

Comparative biogeochemistry of modern, fossil, and artificially aged molluscs using protein, amino acid, stable isotopic...

MISSING IMAGE

Material Information

Title:
Comparative biogeochemistry of modern, fossil, and artificially aged molluscs using protein, amino acid, stable isotopic and ultrastructural methods
Physical Description:
ix, 103 leaves : ill. ; 29 cm
Language:
English
Creator:
Andrews, Samantha Debra
Publisher:
University of South Florida
Place of Publication:
Tampa, Florida
Publication Date:

Subjects

Subjects / Keywords:
Quahogs   ( lcsh )
Polinices duplicatus   ( lcsh )
Biomineralization   ( lcsh )
Dissertations, Academic -- Geology -- Masters -- USF   ( fts )

Notes

General Note:
Thesis (M.S.)--University of South Florida, 1998. Includes bibliographical references (leaves 80-87).

Record Information

Source Institution:
University of South Florida
Holding Location:
Universtity of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 025502992
oclc - 40742137
usfldc doi - F51-00132
usfldc handle - f51.132
System ID:
SFS0044183:00001


This item is only available as the following downloads:


Full Text

PAGE 1

COMPARATIVE BIOGEOCHEMISTRY OF MODERN. FOSSIL. AND ARTIFICIALLY AGED MOLLUSCS USING PROTEIN AMINO ACID. STABLE ISOTOPIC AND ULTRASTRUCTURAL METHODS by SAMANTHA DEBRA ANDREWS A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology University of South Florida August 1998 Major Profe s sor: Lisa L. Robbins Ph.D

PAGE 2

Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Master's Thesis This is to certify that the Master's Thesis of SAMANTHA DEBRA ANDREWS with a major in Geology has been approved by the Examining Committee on May 18, 1998 as satisfactory for the thesis requirement for the Master of Science degree Examining Committee: ( < I Lisa L. Robbins Ph.D. Membe/ Peter J. Harries, Ph.D. Member: Carl I. Ste/fel, Ph.D.

PAGE 3

ACKNOWLEDGEMENTS I would like to express sincere thanks to Dr. Lisa Robbins my major advisor and mentor for her unyielding support ( both financially and emotionally) inspiration. a nd g uidance on this project. I would also like to thank her for her grea t patience during HPLC emergencies and other technical difficulties. I also wish to thank Dr. Peter Harries and Dr. Carl Steefel for their valuable contributions as members of m y committee. Dr. Pegg y Ostrom Michigan State University, kindly ran carbon and nitrogen isotope analyses. I am also grateful to the Tampa Bay Fossil Club for their generous grant in aid of research. This work would not have been possible without any of them I'd like to thank Matthew Schmidt for everything, sociall y and academically Kristin Yale for years of memories both in Connecticut and Florida Bradley Raiche for computer usage and formatting support pa s t and present students of the Molecular Paleontology and Biomineralization Laboratory for k ee ping me sane in times of distres s and most importantly, all of the above for their friendship. Finally my family deserves great thanks for always showing enormous support, interest, and love. My parent s were always there through the trials and tribulations of this th esis.

PAGE 4

TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT INTRODUCTION Purpose ofThis Study/Hypothesis Significance BACKGROUND Shell Ultrastructure Bivalves Mercenaria Gastropods Polinices Conditions that Enhance Shell Matrix Protein Preservation Protein Role in Biomineralization Degradation 813C /81 5N Previous Work Temperature Studie s METHODS Sample Collection and Preparation Permeability Heating Experiments SEM RPHPLC Anal y sis and Fraction Collection 81 3C / 815N ofRPHPLC Fractions Amino Acid Analysis RESULTS SEM Gastropods 111 IV Vll I 2 4 6 7 7 8 8 9 10 14 15 17 19 21 24 24 25 27 28 29 30 31 33 33 33

PAGE 5

Bivalves Comparison RPHPLC Amino Acid Analysis Stable Carbon and Nitrogen Isotopic Analysis Permeabilit y DISCUSSION Description of Proteins RPHPLC Amino Acid Analysis Stable Carbon and Nitrogen Isotopic Analysis Permeability Future Directions Conclusions LIST OF REFERENCES APPENDICES APPENDIX 1. APPENDIX2. APPENDIX 3. APPENDIX4. Sample weights, tests run and weight percent. Sample ID's description collection time data and other notes for isotope analysis. Amino Acid Abbreviations. All amino acid data in nanomoles. II 34 34 44 45 5 7 58 65 65 68 69 71 75 76 78 80 88 89 91 96 97

PAGE 6

Table 1. Table 2. Table 3. Table 4. Table 5. LIST OF TABLES Averages of isotope data for modern and fossil Polin ices and Mercenaria. Permeability results. Weight percent averages. Isotope results for samples decalcified with HCl. 813C values (in %o) for total shell organic hydrolyzate for Strombus and Polinices. iii 57 63 66 74 75

PAGE 7

LIST OF FIGURES Figure 1. Preserved organic composition of fossil molluscs. 6 Figure 2. The general structure of a bivalve shell. 7 Figure 3. The simple prismatic structure of a bivalve shell. 8 Figure 4. The nacreous layer of a bivalve shell. 8 Figure 5. The crossed lamellar structure of a gastropod shell. 9 Figure 6 The nacreous structure of a gastropod shell. 9 Figure 7. Schematic representation of a composite section of one individual matrix sheet bounded on both sides by mineral. 11 Figure 8. Summary of process of hydrolysis. 16 Figure 9. Alanine was heated to various temperatures to determine whether it could persist for long periods 22 Figure 10. Flow chart of major steps in sample preparation and analysis. 32 Figure 11. SEM Photomicrograph of PM, cross-sectional view 35 Figure 12. SEM Photomicrograph of PM, cross-sectional view. 35 Figure 13. SEM Photomicrograph of PM outer surface. 36 Figure 14. SEM Photomicrograph of PM, inside surface. 36 Figure 15. SEM Photomicrograph ofPM, inside surface. 37 Figure 16. SEM Photomicrograph of PM. 37 Figure 17. SEM Photomicrograph of MM, outer surface. 38 Figure 18. SEM Photomicro g raph of MM 38 iv

PAGE 8

Figure 19. SEM Photomicrograph of MM. 39 Figure 20. SEM Photomicrograph of MM closer magnification of cross-sectional view. 39 Figure 21. SEM Photomicrograph of MF, outside surface. 40 Figure 22. SEM Photomicrograph of MF extreme top of photo is insid e s urfac e rest of v iew i s a cross-section. 40 F i gure 23. SEM Photomicrograph of PF outside surface. 41 F i g ur e 24. SEM Photomicrograph ofPF, cross-sectional view. 41 Figure 25. SEM Photomicrograph ofPF, cross-sectional view. 42 Figure 26. SEM Photomicrograph ofPF, cross-sectional view. 42 Figure 27 SEM Photomicrograph of PF, inside s urface. 43 Figure 28. SEM Photomicrograph ofPF, inside surface. 43 Figure 29. Typical RPHPLC graph of soluble organic fraction from modern Mercenaria. 46 Figure 30. T y pical RPHPLC graph of so luble organic fraction from fossil Mercenaria. 46 Figure 31. RPHPLC g raph of soluble organic fraction from modern Mercenaria. 47 Figure 32. RPHPLC graph of soluble organic fraction from modem Mercenaria. 47 Figure 33. Typica l RPHPLC g raph of soluble organic fraction from modern Polinices. 48 Figure 34 Typical RPHPLC g raph of soluble organic fraction from fossil Polinices. 48 Figure 35. RPHPLC graph of so lubl e organic fraction from modern Polini ces 49 Figure 36 RPHPLC graph of soluble organic fraction from modern Polinices. 49 v

PAGE 9

Figure 37. Modem Mercenaria amino acid compositions. 51 Figure 38. Fossil J\.1ercenaria amino acid compositions. 51 Figure 39. Modem Polinic es amino acid compositions. )_ Figure 40. Fossil Polinices amino acid compositions. 52 Figure 41. Modem vs. fossil Mercenaria amino acid compositions. 53 Figure 42. Modern vs. fossil Polinices amino acid compositions 53 Figure 43. Heated modem Mercena r ia (MM-2.2 heated dry) amino acid compositions. 55 Figure 44. Heated modem Mercenaria (MM -2.1 heated wet) amino acid compositions 55 Figure 45 Heated modem Polinices (PM-5, heated dry) amino acid compositions. 56 Figure 46. Heated modem Polinices (PM-3 heated wet) amino acid compositions. 56 Figure 47. RPHPLC chromatogram of MM used for carbon and nitrogen isotope analysis 59 Figure 48. RPHPLC chromatogram of MF used for carbon and nitrogen isotope analysis 60 Figure 49. RPHPLC chromatogram of PM used for carbon and nitrogen isotope analysis. 61 Figure 50. RPHPLC chromatogram of PF used for carbon and nitrogen isotope analysis. 62 Figure 51. Comparison of permeabilities of modem and fossil bivalves and gastropods. 64 Figure 52. Correlation of sample weight and permeability. 77 vi

PAGE 10

COMPARATIVE BIOGEOCHEMISTRY OF MODERN, FOSSIL. AND ARTIFICIALLY AGED MOLLUSCS USING PROTEI N. AMINO ACID. STABLE ISOTOPIC AND ULTRASTRUCTURAL METHODS by SAMANTHA DEBRA ANDREWS An Abstract Of a thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology University of South Florida August 1998 Major Professor: Lisa L. Robbins Ph D VII

PAGE 11

ABSTRACT Data obtained from organic separations of ancient and modem Polinices duplicatu s and Mercenaria mercenaria indicate that the ethylenediaminetetraacetic acid (EDT A) soluble organic shell-matrix proteins are composed of heterogeneous assemblages of proteins and polypeptides. Samples from specimens of Polinices and Mercenaria illustrated overall similar reverse phase high performance liquid chromatography (RPHPLC) chromatograms with hydrophilic and hydrophobic fractions as well as overall similar amino acid compositions although in different concentrations of the major protein-containing fractions At the bulk protein level large isotopic differences were observed between hydrophobic and hydrophilic proteins of both modern and fossil shells indicating some diagenesis had occurred, and possibly reflecting differences in diet and metabolism Heated modem samples showed that some diagenetic alteration occurred during artificial aging as indicated by HPLC and amino acid analysis suggesting that temperature studies may be useful in modeling degradation. Ultrastructurally, many interesting similarities and differences were observed between specimens The present study was able to illustrate that the gastropod's ultrastructure is more intact than the bivalve because of its structurally intact closely packed crystallites, and its lack of original pores in the modern specimen. The bivalve on the other hand shows wider packages of crystals and has porosity within its shell structure The viii

PAGE 12

gastropod also showed evidence of a thicker nacreous la ye r when compared to th e bivalve although both predominantly consisted of either a crossed lamellar (gastropod) or prismatic (bivalve) structure. In both genera, fossil specimens demonstrated significant crystal alteration suggesting that ancient intracrystallin e organics ma y be adversely affected. Mercenaria demonstrated a slightly lower permeability than Po!inices. and fossil specimens showed a higher permeability than modern specimens. C haracteri za tion of shell proteins and their associated calcium carbonate will aid in studies of biomineralization and crystal formation paleoenvironmental conditions and diagenesis. Abstract Approved_:_ ."""" --r= -=.::;...-...,;..-=:;.__ .;......;;..____;;. ________ Major Professor : Lisa L. Robbins Ph. D. Associate Professor Department of Geolo gy i x

PAGE 13

INTRODUCTION Genetic (Paabo 1989) biochemical (Robbins and Brew 1990) immunological (Lowenstein 1980 ; Muyzer et al., 1988) and other geochemical methods such as stable isotope analyses (Ostrom et al. 1994) are some of the approaches which can be used to investigate organic matter in shell and bone. While genetic techniques based on DNA or RNA are considered more direct biogeochemical approaches based on organic matter encapsulated in the calcium carbonate of the shell have the potential for use on fossil as well as living species. Although the use of ancient DNA in molecular paleontology is becoming increasingly advanced (Herrmann and Hummel, 1994), "classical methods of biogeochemistry are still better developed for use on very ancient material. Only in exceptionally well preserved cases can significant amounts of DNA be retrieved. Molecular data from shell matrix proteins provides new possibilities for describing diagenetic pathways from modern to fossil shells. The preservation of amino acid sequences from shell glycoproteins can offer the unique opportunity to compare these fossils with their modem counterparts. Whereas in the past only shell structure was available to provide information on evolutionary trends today fossil shell proteins may contribute to our understanding of molecular evolution. Because of the enormous amount of information in an amino acid sequence proteins give substantially more phylogenetic information than can be obtained from morphology alone (Robbins et al., 1993b). By

PAGE 14

comparin g the structure of modem and fossil molluscs with their protein and amino acid si g natur es. evolutionary questions may be addres s ed It ma y be possible to compar e protein s at v arious stages of their evolution; and because shell protein s pla y a major rol e in s hell morphology, it also ma y be possible to compare the evolution of shell morphology with that of shell proteins (Weiner et al., 1976). Because gastropod s h ave a more complex st ructure than bivalves (Nakahara et al. 1982 ; Andrew s et al.. 1985). these difference s in morpholog y s hould be reflected b y their protein signatures. By isolating and characterizing fossil shell protein matter the steps in how individual proteins degrade over geologic time can be better understood (Robbins et al. 1993b). Owing to the complexity of diagenetic reactions, the mechanisms involved in the diagene sis of proteins have been largely speculative To accurately interpret the results of paleo biochemical investigations a better comprehension of the dia ge netic his tor y of prot e in aceo u s material in carbonate shells i s esse ntial. A thorou g h understandin g of the various components of the shells will provide insight into the preservational pot e ntial of the sys tem" as a whole. Therefore I initiated a multipronged investigation to analy ze s tructural and biochemical components of s hell s. Purpose of This Study/Hypothesis Shells of a modem and fossil gastropod Polinices duplicatus and bivalve Merc e naria m erce naria were analyzed and compared for ultrastructural differences s tructural permeabilit y, protein amino acid, and stable carbon and nitro gen i so topic compo s ition s to la y th e foundation for under s tanding dia g enetic r es i s tance and pro te in indi gene it y in f o ssils. Th ese t wo clas ses of mollusc s provide useful comparison s 2

PAGE 15

because gastropods have been shown to have a mechanically stronger structure than bivalves (Andrews et al., 1985), yet there are few studies which analyze various components (structural and biochemical) to evaluate pathways of degradation. A subsample of each shell was measured for permeability to help evaluate the potential transport of molecule s through the shell. The hypothesis that shells with higher permeability have a higher chance for in situ degradation was tested Although such a hypothesis is simplistic there are no previous studies to document this trend. Furthermore, the amount of organic material and extent of permeability were compared to determine the effect on the preservation of the proteins and their potential indigeneity Additionally samples were heated to simulate diagenetic alteration and elucidate pathways of diagenesis. Isotopic analysis of individual proteins and polypeptides has been suggested as a way to provide an additional criterion in assessing indigeneity (Ostrom et al. 1994). As tracers of protein degradation o13C and o1 5N were measured on individual fractions of modem and fossil material. Finally scanning electron microscopy (SEM) was used to document shell ultrastructural differences between: 1) the classes, and 2) a -2.5 million year old fossil and a modem shell. The goals for this study are as follows: 1) to characterize the soluble matrix proteins of these two types of molluscs; 2) to test the hypothesis that greater shell permeability will result in altered organic matter; 3

PAGE 16

3) to determine if the soluble matrix proteins and their amino acid s are sufficiently distinct as to justify clas s -level distinction in the se two species ; and 4) to analyze the o13C and o1; N of bulk proteins to document difference s between classe s of proteins and to determine the extent of diagenesis. Significance Many researchers have assumed that increased shell permeability will show a decrease in organic material (Andrews et al. 1985 ; Curry, 1987b, 1988), but no baseline data exist s to support thi s assumption. For example Andrews et al. (1985) noted the impermeable nature ofthe molluscan shell matrix both to losses of primary amino acids and incorporation of secondary amino acids. Permeability data sets may provide ways of comparing intergeneric dia ge nesi s The a mino acid compos ition of single proteins from foss il shell matrice s can pro v ide information on the genetic relatednes s (Robbins and HealyWilliam s, 1991 ). Proteins reflect the genetic makeup of an organism because the amino acid sequence is encoded b y th e DNA (Robbins and HealyWilliams 1991 ). Becau s e the amino acid signature i s so di s tinct s m all fragments may be assigned to a genus and possibly even a spec ie s, ba se d on amino acid chromatography (Andrews et al. 1985) Phylogenetic tree s have be
PAGE 17

to geochronology biomineralization and diagenesi s (King 1977) Evolutionar y lineage s can now be traced by studying molecular and biochemical as well as morphological. variations through geologic time (King and Hare 1972a) I s otopic analysis of bulk organic matter can provide an under s tanding of paleoecolo g ical, taxonomic and most importantly diagenetic relationships. In addition to as s essing i s otopic variabilit y among different organic fractions from modem organism s, isotope studies can be used for assessing the indigeneity of ancient organic fractions that have experienced alteration (Ostrom eta!., 1994). The ability to document geochemical changes in individual proteins as a function of time will provide insight into mechanism s that occur during diagenesis. 5

PAGE 18

BACKGROUND Shell proteins have the potential to show exceptional preservation in ultrastructurally well-preserved specimens as far back as the Jurassic (Figure 1. Robbin s et a!., 1993 b). B y combining the study of shell u l trastructure and organic matrix preservation, a better understanding of protein and i n organic diagenesis can be reached (Hudson, 1967). A 500 E 400 0> Qj 0. E 300 E "' '0 i3 200 ., 0 c 100 E <: 0 B 500 400 E 300 0> a; 0. E 200 100 0 q \ a. 'q -o- Amino acids 0 o surviving ' P leist Plio Mio Olig Cret Jur Increasing Geological Age P leist Ill lnsolubles 0 Soluble peplides Free am1no acids Plio Mio Otig Cret Jur Increasing Geological Age 30 25 'U m m 20 "' c 15 < :;:: !E.. 10 :s 3 n; 5 0 0 Figure 1 Preserved organic composition of fossil molluscs. A) Time-dependent decrease of amino acids from fossil molluscs. B) Relative proportions of free amino acids, so luble peptides and insoluble res idues in P l e i stocene P l iocene Miocene Oligocene, Cretaceous, and Jurassic molluscs (After Curry, 1988). 6

PAGE 19

Shell Ultrastructure Bivalve s In genera l the s hell ultrastructure of bivalves consists of an inner prismati c layer overlain by a nacreou s layer. w hich i s s ub seque ntl y overlain b y an outer prismatic la yer (Carter 1990 ; Figure 2). Th e inner layer consists of a homogeneou s structure (crossed acicular and /or a fine complex crossed lamellar st ructure). The middle la yer is homogeneou s or crossed lamellar. The outer layer is crossed lamellar or composite prismatic s tructure. The s imple prismatic layer (Figure 3) contains parallel prisms oriented vertically with the long axes perpendicular to the shell surface. These could be either aragonite or calcite The aragonitic pri s matic crystals are elongate with a fanlike form. The nacreous layer (Figure 4) contains lamellar or sheet like structures that form one layer at a time and it i s a lw ays com po sed of aragonite (Carter 1990) I I; I i .II: I i I !OUTER PRISMATIC LAYER 'I ,l i NACRE ... ,.\ ,\\.t t'l: ,'I I ' J_t L L r =--"_.........-'' PRISMATIC LAYER / ,. I'; t F i g ure 2. T h e ge neral structure of a bivalve shell (After Lowenstam and Weiner 1989). 7

PAGE 20

Figure 3. The s imple pri s m atic structure of a bivalve shell (After Carter and Clark, 1985) Mercenaria Sheet F i gu re 4. The nacre o u s la yer of a bivalve shell (After Carter and Clark, 1985) Mercenaria (superfamily Veneracea) co nt a in s an outer and inner aragonitic pri s matic st ru cture bounding a l am inar nacreous l a y er (Carter 1 990) T h e middl e shell laye r consist s o nl y of a homogeneous struct ur e ear l y in g rowth and a crossed lam ellar s tructure i s sec re te d temporarily in later growt h s ta ges near the boundary between th e middl e and outer shell layer s (Carter 1990) The crossed lamellar structure doe s not form a di stinc t la ye r. This c h ange from hom ogeneo u s t o crossed lam ellar s tructu re i s a grad u a l change. Gastropods Gast r opo d ultr as tru c ture s include a crosse d lamellar s tructur e (which i s homologou s t o t h e pr i s matic structure ofth e bivalve ; Figure 5) composed oftwo differen t alternat in g layers of lam ellae (Carte r 1 990) whic h are mineralogically di sti nct. T h e first-8

PAGE 21

order lamellae are rod-shaped and oriented perpendicular to the s hell surface. The second-order lamellae are differentiated by dip angle away from the s hell margins. Higher dipping angles are composed of aragonite ; whereas lower dipping angles are calcite. The structure of the nacreous layer (Figure 6) is composed of stacks of columnar structures that form vertically rather than laterally as it does in bivalves. Columns form a stacking pattern first and then join laterally into tablet forms (Carter 1990). Simple crossed lamellar Figure 5. The crossed lam ellar structure of a gastropod shell (After Carter and C lark 1985). Polin ices C. Columnar F i gure 6. The nacreous structure of a gastropod shell (After Carter and Clark 1985). Polinices (family Naticidae) has a subovate-to egg shape, and a low elevated spire of about four whorls (Olsson et al., 1953). The shell is composed of aragonite (Robbins and Ostrom, 1995) and is non-nacreous. It is necessary to have an understanding of shell structure because even if two different species have the same mineralogy different crysta l structures can subject the 9

PAGE 22

organ i c matter to different diagenetic resistances. different amounts of crystal di sso luti ons and th e refore contamination ( Robbin s and Ostrom. 1 995). Conditions that Enhance Shell Matrix Protein Preservation Numerous factors includin g she ll material are known to increase the preservation of s hell matrix proteins. Fo r example if the protein is completely enclosed i n the calcite l att ic e dissolution and hydroly s is will b e limited Proteins in shell are naturall y protected by the close relationship of the protein with the mineral matrix retarding decay (Ambler and Daniel 1991 ) The crys tals comprising the shell provide an intracrystalline micro environment where organic material i s protected from contamination (Curry 1987 a) Onl y if the s hell is broken or recrystalli z ed will th e proteins be exposed to water and un dergo h y d ro l ys i s (Robbins eta!., 1 993b). The protein structure and organic materia l can a l so play a role in preservation of the protein Factors such as the primary seco ndary and tertiary structure of the protein w ill affect the pathway of its degradation (Robbins et al. 1993a ). The term primary structure refers to the fixed amino aci d se quence ofthe poly peptide chain (o r chains) m aki n g up the backbone of the molecule. The seco ndary structure i s based on the helical co iling of proteins stabilized by hydrogen bond s Stabilization of the helical coiling require s the presence of di s ulfide brid ges or the tertiary structure (Anfinsen 1959 ). The t e rtiary s tructure of the protein i s likely t o pla y a major role in determining th e res istanc e to modific a tion dur ing fossiliz ation and diagenesis (Eg lin ton and Logan 1991 ). 10

PAGE 23

The location of organics within the shell structure is important, and may play a role in the preservation of these proteins These locations vary between taxa. but u suall y the organics are located between and within the carbonate crystals of the shell. forming layered structures (Robbins eta!.. 1993b). Most molluscs have la yered organic and inorganic structures with intraand extra-crystalline organics (Crenshaw, 1 972: Wheeler eta!., 1988). The majority of crystals are enclosed in an organic matrix (usually comprised of glycoproteins) (We iner and Hood, 1975; Weiner and Traub, 1980 1984 ; Figure 7). This organic matrix forms prior to mineralization, and it is involved in the formation of the shell (Weiner and Hood, 1975; Weiner and Traub, 1980). After most of the shell formation is complete the entombed organic material will contribute to the mechanical properties of the shell (Weiner and Traub, 1984) -D m ineral acid macromolecules D silk-fibroin-like proteins Figure 7 Schematic representation of a composite section of one individual organic matrix sheet bounded on both sides by mineral. The total thickness of individual matrix sheets may vary between 30 nm and 300 nm (After Weiner and Traub, 1984). 11

PAGE 24

The organic matrix of a molluscan shell appears to be very complex with re s p ec t to the number and compo s ition of components within a single species, between species. and even between different crystal layer s of a s in g le bivalve shell ( Wilbur and Manya k. 1984). An intimate relation s hip exi s ts between the cry s tals and thi s organic matri x (Watabe 1965). The majority of skeletal organic molecules are intercrystalline and occur between individual crystal s in the skeleton (Curry 1988). Within th e skeletal elem e nt s trapped bet wee n the cry s tal are intracrystalline protein s, which make up 0 .01 to 0 1 weight percent of the mineral (Addadi and Weiner 1989; Albeck et al., 1993 ) In fossil shells, the concentration of the organic matter initially decreases rapidly as the intercrystalline material is lost. The slowly decomposing intracrystalline organic matter preserved within a protected micro-environment within the mineralized phase shows a s lo wer rate of decline an d th ere fore ha s a greater s urviv a l pot e ntial (C urr y. 1988) The organic matri x consists mainl y of so luble and insoluble protein fraction s. The in so luble matrix (IM) and soluble matri x (SM) are distinct entitie s. The IM i s the structural framework of the matrix and the SM interacts with the crystal (Wheeler et al. 1988 ) The SM is thought to b e intracrystalline, while the IM is intercry s talline (Wilbur a nd Man yak 1984). The so luble protein of the organic matri x o f mollusc s consi s t s of sulfated g lycoprotein s rich in a cidic amino acids (Weiner and Traub 1984) and i s therefore usuall y composed of repeating sequences of aspartic acid separated by either g lycine or se rine (a lternatin g aspartic acid residues). The regularl y spaced negatively ch a r ge d a s p a rtic acid could function as a templ a t e on which the minerali za tion occurs. T hi s i s acco mpli s h e d by binding the po s itiv e l y char ge d calcium t o the ne g ativel y charged aspa rti c acid. 1 2

PAGE 25

Partially degraded organics may be preserved as coherent proteinaceous sheets or they can exist as free amino acids derived from the breakdown of the original protein (Hudson. 1967) An organic matrix may be observed even in recrystallized shells. but they will not necessarily retain their original organic composition Furthermore. porou s sa mple s in warm oxidizing environments are poor preservers of organic matter (Hudson. 1967). Fossils with porous skeletons are more susceptible to contamination b y percolating groundwaters than those with dense tightly packed ultrastructures (Curry 1987b 1988). In addition microbial activity will be limited as porosity and permeability tend towards low values (Eglin ton and Logan, 1991 ). Organic matter contained within nacreous and foliate structures is present in thick. continuous sheets; they are more exposed to attack by hydrolyzing or oxidizing solutions (Hudson, 1967) In contrast, intricate ultrastmctures containing very thin organic sheaths of the complex lamellar group are less likely to be attacked (Hudson, 1967). Gastropods. consisting of the complex crossed lamellar structure would be less prone to degradation compared to the prismatic structure found in bivalves. However well-protected macromolecular proteins from most microarchitectures have been shown to escape destruction and part of their original structure can be preserved over long periods. The deposit i ona l environment also plays an important part in shell protein preservation The rapidity of burial and dewatering of the sediment can determine the extent of preservation At a surface environment of 1 0-20 C, hydrolysis is ten times more rapid than in deep-sea sediments where 90% of the peptide bonds are hydrolyzed by 1 Ma at 1-2C (Bada, 1991) It has been suggested that ifthe shells are deposited in an anaerobic environment degradation will be slower and more protein will be preserved 13

PAGE 26

(Robbins et al. 1993b ) However. the deca y proce ss can still be active despite anoxia Ammonium and bicarbonate ions liberated by the decay of animal protein can affect th e local pH/ Eh conditions promoting an alkaline reducing environment (Allison. 1988) Rapid burial and anoxia will reduce the rate of decay but it will not s top it. Protein Role in Biomineralization Formation of mineral s by organisms has increasingl y modified the chemical and physical nature of the biosphere. Biomineralization has radically changed the chemistry of the oceans and has contributed significantly to the sedimentary environment (Lowenstam 1981 ). Analys i s of the organic shell-matrix proteins may allow for insight into the specific proteins that control biomineralization of these molluscs During "or ga nic m atrix -medi ated" s hell formation (Lowen s tam and Weiner 1989). the organism l ays down o r ganic lay er s on w h ic h c alcite crystals s ub se quentl y nucleate. It is th e protein molecule s co mprisin g th ese organic layers that are generally believed to control biomineralization. T h e two broad cla ss e s of proteins found in the mollu scs, a h y drophilic fraction and a h y drophobic fraction are s imilarl y found in benthic and plankti c foraminifera ( Robbin s a nd Brew 1990) as well a s echinoid teeth and tests (Weiner 1986). Weiner and Erez ( 1984) proposed that these two classes of protein s perform different functions durin g te s t fo rmation and that because of their acidic nature are capable of bindin g calcium i ons and are prob ably invo lved in crys tal formation The common amino acid sequence in which a s partic acid i s se parated b y a s in g le amino acid co uld constitute part of thi s bindin g s ite (Weiner and Traub 1984) Two theories of the matrices function 14

PAGE 27

during mineralization are: I) a template which controls crystal growth: and 2) a compartment in which the crystals grow (Weiner and Hood 1975). The matrix .. hypothesis assumes that the organic matrix is deposited as a substrate that nucleates crystals ; the "compartment hypothesis supposes that crystal nucleation occurs within pre existing empty compartments of the organ i c matrix (Crenshaw. 1 980: Lowenstam and Weiner 1989) In addition, the matrix can re gulate the growth of the biomineral in several ways : by initiation of crystal growth determination of the crystal polymorph control of the crystal shape or terminat i on of crystal growth (Wheeler et al., 1 988 ; Albeck et al. 1996). Using electron diffraction, Weiner et al. (1983) demonstrated that gastropods and bivalves nacreous layers have the same spatial relationship between organ i c matrix constituents and aragonite crystallographic axes strongly supporting the notion that aragonite crystals form upon an organic matrix template Degradation Although excellent preservation of organic matter has been observed in rare ancient specimens, proteins will degrade over geologic time Degradation of proteins can take many pathways ; hydrol ysis is one of the most important of these degradation processes (Mitterer, 1993; F i g ur e 8), involving the decomposition of organic compounds in the presence of water. The process can be catalyzed by some metals and by both acids and bases however hydrolysis is accelerated at ext r eme pH values and occurs more rapidly under alkaline condit i ons (Kahne and Still 1988). Hydrolysis breaks the peptide bonds and initially converts the original proteins into a mixture of l arge and small peptides and free amino acids and ultimately to a mixture of only free amino acid s 15

PAGE 28

(Figure 8) A s the process proceeds smaller. more readily leachable molecule s are p ro duced Free amino acids are the smallest and most easily leached amino compound s in foss ils (M itterer 1993) Therefore hydrolysis and the subsequent leaching lead to a los s of free amino acids and a decrease in concentration of the amino acids in the fossil (Mitterer, 1993 ). I ORGANIC DIAGENESIS I Hydrolysis Protein -----. Peptides -----. Free Amino Acids Figure 8 Summary of process of hydrolysis. Water s eeps into the shell from the surrounding sedi ment and shell constituents are leached or modified by the added water. Due to leaching and diffusion the concentration of organic constituents decreases with increasing age of the fossil (Modified after Mitterer, 1993 ). The rate of h y droly sis depends on many factors including: 1) the length of the peptide chains and 2) the nature of the adjacent amino acids. The peptide bonds in larger molecules are more stable than the shorter chains The stability of the peptide bond is also largely a function of the amino acid residues that form the bond (A mbler and Daniel 1991 ) and since the protein composition and amino acid sequences are genetically determined the proteins in organisms of different taxon will have different rates of hydrolysis (Mitterer 1993) The rate of hydrolysis also depends on the amount of water available, the temperature therefore burial history and finally the presence of other biomolecules in the shell such as carbohydrates and metal ions (Vallen t y ne 1964) Despite thi s important de gra dation factor specimens may not be contaminated because 16

PAGE 29

similar amino acids have been found in fossils of the same species in different locations (Abelson 1956). However a possible way to trace the hydrolytic degradation reaction in fossils is through stable isotopic analysis. 813C/815N The diverse isotopic compositions of individual organic molecules in living organisms reflect the complexity of chemical reactions occurring during the growth and metabolism of a cell (Macko et al. 1983) The overall fractionation of stable isotopes between a specific molecule and its ultimate precursors is controlled by a number of factors including key rate-determining steps and major branching points in the flow of carbon and nitrogen (Macko et al., 1983) One of the earliest studies of this type was carried out by Macko et al. (1983) on algae in order to understand the basic fundamentals of isotope biochemistry They found that because amino-nitrogen is transferred from one am i no aci d to another the distr i bution of 1 5N is dependent upon the branching ratios of the components and the number of reactions in which an amino acid is involved The carbon isotope distributions in amino acids are similarly dependent on branching ratios biosynthetic pathways, and inhibition controls. Even from preliminary studies, they concluded that isotopic analysis of both carbon and nitrogen render a valuable tool for tracing specific compounds through biosynthesis and early diagenesis and ultimately into the fos sil record (Macko et al. 1983). Because amino acid constituents of organisms have distinct stable carbon isotopic compositions (Serban et al. 1988) the 81 3C of an amino acid allows evaluation of the 17

PAGE 30

origin and dia ge netic path ways of organic constituents in geologic materials ( Robbin s and Ostrom. 1995) Two main factors control th e o1'C of an amino acid: 1 ) the ty pe of carbon (fo r example, light 12C or heavy 13C) available when the s hell i s created. and 2) the degree of i so tope fractionation at major branchin g points in th e metabolitic pathway. In past s tudi es on molluscs, a lower o13C value wa s seen in fossil amino acids then in modern amino acids (Robbins a nd O s trom 199 5). This lower va lu e could be a result of the reintroduction of i sotop ic ally depleted degradation products. This i s consistent with the lo ss of 1 2C in the amino acid during diagenesis (Robbins and Ostrom 1995) The degree of preservation or degradation of the protein can be measured by the C:N ratio Since this ratio varies in modem shells, it is necessary to compare fos s il s with modern shells of comparab l e s tructure. Fossil s have very low nitro g en. therefore fossil s s how hi g h e r C :N rati os tha n their mod e m co tmt e rpart s (Hudson 1967). Fo ss il s with nacre-prismatic s tructur es show lower C:N ratios than those of the complex-lamellar or foliate g roup s (Hudson, 1967). The res ults e x pressed as C:N ratios do not reveal whether the protein is uniformly d egra ded or whether some of it remain s intact while the rest has deco mp ose d. In recry s talli zed fossils organic matrix remnants h ave no detectabl e nitrogen. It is also pos s ible to remove a large part of the organic carbon and nitro g en from the s hell while l eaving relic s of the original protein in a relatively unaltered condition (Hudson, 1967). 1 8

PAGE 31

Previous Work The first relatively detailed analysis of the biochemistry of the organic matrix of the mollusc shell was completed by Abelson (1954. 1955 ) In this pioneering work. he u sed a Mercenaria shell to examine the amino acid content of the protein-bound. so luble protein/peptide and free amino acid fractions. With increasing age. the protein-bound amino acids decreased ; the soluble protein/peptide amino acids increased from the Recent Pleistocene however disappeared b y the Miocene ; and the free amino acids were greater in older material than in recent specimens (Abelson, 1955) This suggests a degradation cycle from protein-bound to free amino acids through time The conclusion Abelson drew from this data was that initially most of the protein of the shell was present in insoluble layers. This stability was maintained for thousands of years with only moderate changes occurring which did not affect the protein solubility. Once some of the peptide bond s were broken the fragments were more soluble and could be leached from the shell. When all of the bonds were broken, only free amino acids remained ; these could be stable for millions of years The most likely degradation process that Abelson observed was hydrolysis Abelson (1955) and King and Hare (1972b) also showed that thermally unstable amino acids were absent or only present in small amounts in older material and consequently more thermally stable amino acids are still present in older specimens From thi s the y concluded that the amino acids were preserved indigenous organic matter secrete d by the living organism rather than contaminants. An increase in certain amino acids such as alanine in older material has also been noted and is probably due to the generation of this amino acid from the diagenetic breakdown of serine (Vallentyne. 19

PAGE 32

1964). Hare and Abelson (1964. 1965) found that proteins of species closel y related in morphology or shell s tructure have similar amino acid compositions. King and Hare ( 1972a 1972b) expanded on this using foraminifera Even though a similar morphology should show a generally similar amino acid signature a difference can occur in the more sensitive (variable) a mino acids, such as aspartic acid and glycine. The y also found the amino acid compo si tion to be a direct expression of the genetic system and were even abl e to s how differing amino acid compositions from two morphologicall y distinct fossil species ofthe same genus (King and Hare 1972a) Throughout the years, Mercenaria has been extensively studied by many researchers (e.g. Abelson 1954 1955 1 956 ; Hare and Abelson 1964 1965 ; Degens et al., 1967 ; Hare 1969 ; Vallentyne 1969; Akiyama 1971; Crenshaw, 1972 ; Weiner and Ho o d 1975 ; Hare and Hoering 1977; Kennish 1980 ; Andrews et al., 1985 ; Serban et al.. 1988 ; Muyzer et al., 1988 ; Engel et al. 1994 ; Silfer et al., 1994 ; Robbins and Ostrom 1995). Wheeler et al. (1988) showed that numbers and molecular weights of matrix fractions can range from many, low molecular weight components from gastropod shells to a single, hi g h molecular weight glycoprotein from M m erce na r ia While studying the s oluble matri x of aM. m erce naria Crenshaw ( 1972) determined it to be a highly sulfated glycoprotein. About 80% of the weight was protein The amino acids were determined to be 30% aspartic acid, 16% glycine, and 10% serine. Other workers have also shown aspartic acid, glycine and serine to be the most abundant acids in molluscs (Weiner and Hood. 1975 -bivalves ; Weiner et al. 1979-cephalopod s) a nd more specificall y Mercenaria (Hare. 1969 ; Weiner and Hood 1975 ) as well as other mineralizing systems r e pre se ntin g both CaCO and CaP04 biominerals (Robbins and Donachy 1991) 20

PAGE 33

While Polinices has not been as widely studied several investigations do exist (Degens et al. 1967 ; Ghiselin et al. 1967 ; Ostrom et al. 1994; Robbins and Ostrom. 1995) Previous experiments performed on Polinices showed that the protein solution was entirely soluble in water (no IM found). Two major protein fractions within the shell matrix have been identified. They contain individual amino acids with distinct 813C values (Robbins and Ostrom, 1995). Temperature Studies Numerous heating experiments have been performed on molluscs (Abelson. 1956; Hare 1969 ; Totten et al. 1972; Andrews et al. 1979 ; Qian et al., 1995) to observe degradation of organic compounds. In one experiment shell fragments of Mercenaria mercenaria were heated at high temperatures for a different number of hours to represent diagenesis A s expected the immunological reactivity of the Mercenaria proteins decreased with increased heating (Muyzer eta!. 1988) which is analogous to reactivity decreasing with sample age. However, significant positive reactions were still found in the shell fragments even after they were incubated for 80 hours at 140C. Abelson (1956) demonstrated that the rate of degradation of alanine in short laboratory experiments at elevated temperatures can be used to model degradation in fossils that are several million years old. It is possible to draw a curve extending back to predict pathways of degradation at low temperatures (Figure 9). For instance when heated at 450C it takes one second for the concentration of alanine in water to decrease by 63% At 188C, it would take one 21

PAGE 34

month to decrease by the same amount (Abelson 1956). At room temperature. alanine could theoretically last for billions of years. A lth ough alanine this old does not exist. this large extrapolation to room temperature is justified since alanine is observed in ancient fossils (Abelson 1954 ) The most stable amino acids are alanine glutamic acid gly cine. isoleucine proline, and valine (Abelson 1956). Less stable amino acids are arginine. aspartic acid lysine phenylalanine. serine threonine and tyrosine (Abelson 1956). with the last three being particularly unstable at elevated temperatures and in geologic environments (Nagy et al., 1981 ). "' "' 10"'YEARS I'-"' "' 1 YEAR 1'\. 10 DAYS 10'SECONDS """ 1 0:: SECONDS 1SECOND "' '\. SECONDS 20 40 60 80 100 150 200 250 300 400 500 TEMPERATURE (DEGREES C ) Figure 9 Alanine was heated to various temperatures to determine whether it could persist for long periods. The horizontal axis of the small circles indicates the temperature to which each sample of alanine was heated. The vertical axis indicates the length of time it took for 63% of the alanine to break down If a straight line is drawn through the circles, it is shown that at a temperature of 20 C nearl y half of the alanine would remain after 3 billion years This may also be true of other amino acids (After Abelson 1956) 22

PAGE 35

Totten eta!. (1972 ) heated powdered oyster shells at normal atmospheric pressure 130C for two months The aspartic acid and g l yc in e in the protein fraction declined b y 99% i n t he firs t week and 50-80% of the o ther amino acids were lo st. Some a min o acids tran s form e d int o ot h ers when heated and th e amo unt s of some amino acids increased by heating illu s tratin g tha t different amino acids can h ave different diagenetic s tabilities More than h alf of the total amino acid content of the she ll was l ost durin g the entire course of heating The amount of stable proteinaceous material in the heated samples paralleled th a t found in fossils Therefore, the quick protein decomposition b y heat can be shown to roughl y correspond to that by slow aging. 23

PAGE 36

METHODS Sample Co llection and Preparation Fossil Mercenaria mercenaria (MF) and Polinices duplicatus (PF) were collected from the Pinecrest Shell Beds in Sarasota, Florida (3.5-2.0 Ma). The investigation of fossils collected at one locality is preferable because of the fossils exposure to identical thermal histories and percolating groundwaters (Abelson 1954). In addition these shell beds consist of uncemented shell hash eliminating the problem of diagenetic processes that occur with cementation Samples were chosen based on as pristine a shell as possible and only clean fossils separated from the sediment matrix were used Living M mercenaria (MM) were collected from the east side ofBurgess Bay near Cape Coral Florida (MM1 and MM-2) and Fort Desoto Florida (MM -3). Modem P. duplicatus (PM) were collected from Essex, Massachusetts. Modem samples were stored at -70C until processed. The entire gastropod shell and sing le va lve s of the bivalves constituted a sample The shell samples were scrubbed using a brush and rinsed in triple distilled water (3xDH20). Samples were then soaked in a concentrated bleach solution to remove any remaining sediment and contaminants, thoroughly rinsed with 3xDH20 and air dried prior to being cru s hed by mortar and pest l e and subsequently ground to a powder. Samples 24

PAGE 37

were decalcified in I 0% (w/ v) ethylenediaminetetraacetic acid (EDTA, pH 8) containing 0 .1% sodium azide at 4 C. After all the carbonate was dissolved the sample was centrifuged at 12,500 rpm for 25 min to concentrate the insoluble organics The s upernatant solution containing the so luble matrix proteins was then exhaustively dialyzed against 3xDH20 using an Amicon ultrafiltration device (molecular weight cutoff 10 000 Daltons) to remove excess salts and free amino acids. After dialysis, samples were lyophilized and weighed (Appendix 1 ), and stored at -70C until used (Figure 1 0). Permeability The permeability was measured on the two species to determine possible fluid flow pathways that would contribute to diagenetic alteration. A piece of shell material from Mercenaria and Polinices samples was drilled off of the specimen using a Dremel MotoTool. The permeability was then determined for several samples (Appendix I) b y the process of Microflow Permeability (Porous Materials, Inc., Ithaca NY). In this method, the sample is sealed in a chamber of known (calibrated) volume. Pressurized gas is then applied to one side of the sample and the pressure on the other side of the sample is measured. If flow is occurring through the sample the pressure on the post-sample side will increase proportionately From this data, a graph of time versus pressure (post) is plotted The calculations are based on the ASTM Method DI434-82 A linear regression is performed on the pressure versus time data to yield the slope of the line (torrA/sec). 25

PAGE 38

Slope = _ A flow rate. Q is calculated by the equation for both the testin g pressure and t empe ratur e. as well as STP conditions. At STP: Slope(torrA /s ec) x Vol(cc) x T , P (K) Q= ____________________ __ Where: Vol = Volume of chamber (cc) T , P = Temperature at STP (273K) P ,,r = Pres s ure at STP (760 torr) At Test Conditions: Q= Slope(torrA/sec) x Vol(cc) P pre( torr A) P post( torr A) Ppre =Average Pre-Pressure at Test Con ditions (terrA) Pposl = Average Po s t-Pressure at Test Conditions (torrA) T,. , = Temperature of testing gas (K) Thi s value Q is then normalized to the cross-sectional area by dividing by the sample cro s s-section giv ing the Gas Transmission Rate (GTR). Q GTR= 26

PAGE 39

Where: Asample = Cross-sectional area of sample ( cm2 ) The permeance is then calculated by dividing the GTR by the average pre-pressure (pressure on gas side of sample) in torr A. GTR Permeance = ---ppre From the permeance, the permeability can be determined by multiplying the permeance by the sample thickness in meters. Permeability = Permeance x Thickness The ASTM referenced above recommends permeability units of cc / sec / m / torr, as are used here. Heating Experiments Several temperature studies were performed on the modern specimens to simulate diagenesis. By subjecting the shell material to elevated temperatures for specific periods of times the decay process can be accelerated and pathways of diagenesis can be inferred. The amount of degradation depends on time and temperature, however it is more influenced by temperature than time (Abelson, 1954). Therefore, modern samples were heated at I 00C for I week I 00 C was chosen to get a maximum temperature 27

PAGE 40

without boiling. The shell wa s placed i n a beaker w ith 3xDH20 that wa s adjust e d t o a pH o f 7 6 This wa s done to approximate Flor ida groundwater in the Sarasota area (Stewart per s. comm ) where the fossil samples were obtained since pH can be expected to pla y a significant part in the kinetics of amino acid decomposition (Hare 1969). Samples were al s o heated dry as a control. In wet heating distilled water was used for the h y drol y tic re a ctions because higher r a tes of de g radation occur in the pre s ence of water (E g lin ton and Logan 1991 ) This procedure i s slightl y different than has been used previou s ly (Vallentyne 1969 ; Totten et al. 1972 ; Andrews et al. 1979; Qian et al. 1995) In their experiments shells were crushed prior to heating in order to provide more surface area for h y drolysis to occur (En gel, per s comm.) and to assure homogeneit y (Vallentyne 1 969). My experiment s utilized whole shells to more accurately simulate the conditions in natural environments. After heating shell s amples were decalcified according to the procedure s mentioned above and analyzed for protein and amino acid composition as described in detail below SEM A subsample of each species was examined under the Scanning Electron Microscope (SEM). It was assumed that this subsample represented the condition of the entire shell. The shell fragment was mounted on a stub and sputter coated with gold palladium for 3 minutes. It was then obser v ed and photographed on an ISI-DS130 Dual Sta g e SEM located in the Department of Marine Science at the Universit y of South Florida. The photographs were used to ensure that all contaminants were removed and to d e termine the ultrastructural pattern of the shell material. 28

PAGE 41

RPHPLC Analysis and Fraction Collection The organic matrix proteins were se parated into fractions b y rever se phase high performance liquid chromato g r ap h y (RPHPLC) equilibrated with 0.1% trifluoroacetic acid (TF A) in 3xDH20 (Buffer A). Separation was done with a Vydac Protein column (Vydac, Sep ara tion Group Herparia CA) at a flow rate of 5 ml / min. at 25C through an Isco V 4 absorbance detector (28 0 nm ) The proteins were eluted using increasin g co ncentrations of acetonitri l e to 95% acetonitrile contai nin g 0.1% TF A (Buffer B) with a gradient of0-5 min. at 0 % Buffer Band 15-40 min at 100% Buffer B. By using the se volati l e, low pH so lvents and short (C4 ) alkyl chain covalently bonded large pore size (5 1-1m particles 300 A pores ) silica co lumns, the peptides and proteins in a size range from 5 amino acids up to > I 00 000 daltons can be separated in a single run, and more than 90% of the s ample can be recovered (Hunkapiller et al. 1984a) In addition, peptides differing b y as little a s a single amino acid re s idue can be resolved (Hunkapiller eta!.. 1984b ). All fractions for amino acid analysis were manually collected into test tubes and f reeze dried Protein fractions for analysis of 813C and 81 5N were collected into ballasts rotoevaporated, transferred into quartz tubes and subsequently lyophilized (Appendix 2) The column was cleaned between samples by using an acid wash (1 :4 O.IN HN03: isopropano l ; for 40 min.) Bovine Serum Albumin (BSA) was used as a control to determine if isotopic fractionation occurred during chromatographic separations of proteins (Ostrom eta!., 1994). After an initia l practice run to observe where peak s occurred, samples were collected at 12-13 :2 0 min. (pre-peak), 14:30-21 :30 min. (peak), and 23-24:20 min. (post-peak) After cleaning the column a blank was run and column 29

PAGE 42

eluent was collected at exactly the same times. All sample ID's and collection tim es can be found in Appendix 2. 813C/81 5N ofRPHPLC Fractions Quartz tubes with l yoph ili zed samples were sen t to Michigan State U ni vers it y to be analyzed for 813C and 815N I so topic integrity during RPHPLC (Ostrom et aL 1994). and dur i ng rotary evaporation (Feuerstein eta!., 1997) has previously been demonstrated. The ana l ysis of 813C/815N compos ition s of bul k protein material was performed o n a VG Isochrom consisting of a Hewlett-Packard 5890 GC coupled to a VG PRISM (Mi cromass) isotope-ratio ma ss spectrometer via a combustion furnace and water t rap The standard deviation of 8"C va lu es for proteins i s typically 0 5 %o (Os trom et al., 1 994). Nitrogen and carbon isotope ra ti os are ex pressed in per mil as: DNE = [(R samplj R slandard) 1] X 1000 Where : N = the heavy isotope of e l ement E R = the abunda n ce ratio of the hea vy to light isotope T h e standard for carbon is the C hic ago Peedee Belemnite (PDB) and for nitro gen i s atmospheric N2 Amino Acid Analysis Bul k po l ypeptides and fractions containing polypeptides collected from RPHPLC were hydrolyzed with 200 of 6N HCI at 11 0C in vacuo for 24 hour s on a Waters 30

PAGE 43

Pi coTag workstation and redried. H y drol yzed sa mpl es were se nt to the Protein Chemistry Core Laboratory at the University of Florida for amino acid analysis. There. sam ple s were analyzed using a Beckman System 6300 High-P erfo rman ce Amino Aci d Analyzer. Thi s sys tem u ses cation-exchange with postcolumn ninh ydrin derivatization for se paration and quantification of amino-containing components. Percent compositions were calculated for each of the amino acids present in the fractions collected. 31

PAGE 44

Sample Preparation and Analysis Shell sample 1 Sample Heated ( wet and dry) VVashed Crushed to 1 Piece for Permeability 1 Piece for SEM Dec a lcified with EDT A 1 Centrifuged IM SM Dial y zed 1 Lyophilizr and weighed RPHPLC Amino Acid Analysis Figure I 0 Flow chart of major steps in sample preparation and anal yse s 32

PAGE 45

RESULTS SEM Gastropods Observation and comparison of SEM photomicrographs revealed associations of she ll microstructure and its relation to organic material. Photographs of PM can be found in Figures 11-16 and PF in Figures 23-28 The cross-sectional view (Figure 11) ofthe modem gastropod showed the complex-crossed lamellar structure. Two sets of elongated crystals were observed each one ti lt ed at some angle to the shell surface as Lowenstam a nd Weiner (1989) had previously reported. A closer view of th ese two se t s of crystals i s s hown in Figure 16. Also ev id ent in Figure 11 are organic layer s. The s e show up as wavy thin protrusions separating crystal layer s At higher magnification (Figure 12) the alignment of the crystalline packages becomes even more obvious. The outer shell s urface of this sample also exhibits pristine features (Figure 13). A gastropods nacreous layer is formed by stacks of crystals (Wilbur 1 974). Seen here are these nacreous stacks columnar aggrega t es of aragonite crystals The aragonite tablet s making up the stacks grow within multi-layered compartments partitioned by organic sheets and the g rowing tablet s are surrounded by organic envelopes. The regularly s paced organic s heet s are a rranged parallel to the surface, and each aragonite tablet that makes up the stacks grows hori zo ntall y within the compartment (Nakahara et al. 1982) These p yra mid shaped 33

PAGE 46

stacks have been widely observed in gastropods by many other authors (Mutvei. 1970 : Wind and Wise 1974 ; Wilbur 1974 ; Nakahara eta!., 1982) In the Polinices seen in Figure 1 1 a nacreous layer is observed bounded by the two complex cro s sed lamellar layers This nacre comprises about 20% of the thickness of the cross section. Bivalves Photographs of MM can be found in Figures 17-20 and MF in Figures 21 and 22. The outside surface of the modem bivalve shows several p rominent ridge and interridge areas (Figure 17 and 18) These shell ridges are produced when these molluscs grow by adding increments at the shell margin (Wilbur 1974) A biva l ves nacreous layer is normally formed from three lamellae, each one crystal in thickness. and sometimes contains irregular polygonal plates (Mutvei 1970 ; Wilbur 1974). Figure 15 demonstrates that the bivalve sample was dominated by its prismatic layer. with only a small percentage of nacre evident (upper right comer where evidence of polygonal plates can be observed) Secretion of the prismatic or crossed lamellar-like structures is a prerequisite to nacre formation (Wind and Wise, 1974), therefore that may be the reason for the scarcity ofthe nacreous layer. Comparison When comparing the modern t o the fossil gastropod the modern samples crystallites appear pristine (Figure 13 ), and the cross-sectional view shows evidence of layered organics, as mentioned previously. The fossil specimen displays crystal alteration with borings and dissolution evident (Figure 23), and no trace of organic 34

PAGE 47

Figure 11. SEM Photomicrograph of PM cross-sectional view. Can observe possible organic layers enclosed in a nacreous and crossed lamellar structure. Figure 12. SEM Photomicrograph of PM cross-sectional view. Note intact complex crossed lamellar crystallites aligned in packages. 35

PAGE 48

Figure 13. SEM Photomicrograph of PM, outer surface. Nacre stacks show intact, non porous structure. Figure 14. SEM Photomicrograph of PM inside surface. Shows possible active growth s urface of the complexc ro ssed l amellar s tructure. 36

PAGE 49

Figure 15. SEM Photomicrograph of PM, inside surface. Intact linear crysta l s are evident. Figure 16. SEM Photomicrograph of PM. Groups of crystals oriented perpendicular to each other show the complex crossed lamellar structure clearly. 37

PAGE 50

Figure 17. SEM Photomicrograph of MM, outer surface. Well-defined shell ridges and interridges seen. These could be possible paths of fluid flow that would lead to greater dissolution. Figure 18. SEM Photomicrogr aph of MM. View of side of ridge from Figure 17. Pores are visible, but have a defined structure to them. Crystals appear to be in subcircular packages. 38

PAGE 51

Figure 19. SEM Photomicrograph of MM. On right side of photo is the in s ide surface on l eft side is a cross-sectional vie w Figure 20. SEM Photomicrograph o f MM, closer magnification of cross-sectiona l view. Crystal layers of prismatic structure apparent. 39

PAGE 52

Figure 21. SEM Photomicrograph of MF, outside surface. Ridges and interridges no lon ger visible. Cannot distinguish between original porosity and post-mortem diagenesis. Figure 22 SEM Photomicrograph of MF, extreme top of photo is inside surface, rest of view i s a cross-section This cross-sectiona l view does not show significant diagenesis compared to the modern sample in Figure 19. 40

PAGE 53

Figure 23. SEM Photomicrograph of PF, outside surface. Prominent dissolution and boring. Figure 24. SEM Photomicrograph of PF cross-sectional view. Two distinct c r ystal layers are evident with their crystals oriented perpendicular to each other in this low magnification shot. Broken crystal faces indicate some diagenesis. 41

PAGE 54

Figure 25. SEM Photomicrograph of PF cross-sectional view Ridges of crystals in cross-section from lower half of Figure 24. Figure 26. SEM Photomicrograph of PF, c r oss sectio n a l view. Closer magnification of about 2 ridges from Figure 25. 42

PAGE 55

Figure 27. SEM Photomicrograph of PF inside s urface Small round pits with elongate crysta ls across them indicate dissolution. Figure 28. SEM Photomicrograph of PF inside surface. Closer magnification of dissolution pit from Figure 27. 43

PAGE 56

material is left in cross sectional v iew (Figure 24). In addition, the nacre stacks appear broken and less defined (Figure 23) and dis sol ution pits occur on the inside surface (Figure 27 and 28) The most obvious difference between the modem and fossil bivalve i s the disappearance of ridges on the outer s urface. The protein content of the bivalve ridge is nearly double that of the interridge area (Wilbur. 1 974). T heref o re the diagenetic los s of these ridges (Figu r e 21 ), even though they constitute only a small proportion of the entire shell, may adverse l y affect the organic constituents. Abundant pores are also seen, enhancing the original porosity in the modem specimen (Figure 18). Data shown here support the hypothesis that gastropods have a more intac t s hell structure. For example, Polinices lack the original pores seen in the bival ves that could be conduits for pore water s. As pointed out above Polinices also posses s more structure and order to their crystallites. Nakahara et al. ( 1982) observed a central core in the organic s heet s of a gastropod which was not observed in bivalve specimens. Thi s indicates that the sheets of gastropod nacre are more elaborate in structure and possibly more rigid than those in bivalves (Nakahara et al., 1982). In addition. data shown in this study indicated that gastropods have a greater amount of organic material per shell weight than bivalves. RPHPLC A comparison was made of the RPHPLC chromatograms for the modern versus the fossil samples in both Mercenaria merc en aria and Polinic es duplicatus based on the elution times of the fractions and the peak height (absorbance). Data from HPLC 44

PAGE 57

analyses demonstrate that the protein signatures for the species analyzed have overall similar patterns. although the y do show inherent differences. The overall elution profil e shows a lar ge peak at the onset of the run (0% acetonitrile) and three peaks between 5095% acetonitrile. Modern Mercenaria (Figure 29) showed peak retention times of approximately 3:00 minutes for the hydrophilic fraction and 12:30 13:00. and 14:10 minute s for the hydrophobic fraction. The hydrophilic fraction s howed a double peak Fossil Mercenaria (Figure 30) showe d similar results with retention times of 3:00. II :00 1 2:20, and 14:00 minutes; however the hydroph obic fractions indicated a higher absorbance. The heated modern sample (MM-2.1-Figure 32; and MM-2.2-Figure 31) also showed this greater absorbance and single hydrophilic peak The chromatogram for the modern Polinices (Figure 33) showed the same initial injection peak at 3:00 minutes although with a lower absorbance. The h yd rophilic fraction did not indicate any clear peak s. Three small peaks occurred at 11 :30 I2 :40 and 1 3:20 minutes. In the heated samples (PM-3-Figure 36 ; and PM-5 -F igure 35), these 3 peaks were much clearer but still very small. Fossil Polinices (Figure 34) had retention times of3:00, 10 : 00 11 : 40 and 13:30 minutes. These hydrophobic fractions had a greate r absorbance when compared with the modem sample. Amino Acid Analysis Amino acids are compared based on percent composition calculated in nanomolar concentrations. Fractions collected from peaks indicated on Figures 29-36 were used for amino acid analysis because they were the major protein-containing fractions. The r es ults for th ese fractions are s hown in Figures 37-46. The bulk of the proteinaceous 45

PAGE 58

MM-1.2 Protein Fractions MM1200 --Peak 2 -E s:::: 0 CX) Cll 0 s:::: Peak 3 .c ... 0 Ill .c 1 <( I--"' .'-0 10 20 30 40 T i me ( min.) Fi g ure 29. Typical RPHPLC graph of soluble organic fractiOn from modern Mercenaria. Labeled peaks were used for amino acid analysis. Peak 2 appears as only 1 peak because data went out of ran ge. A double peak actually occurs there. MF-2 Protein Fractions M F20 1 Peak 1 E Pe a k 3 s:::: 0 Peak 2 CX) Cll Peak 4 0 s:::: .c ... \\ 0 Ill .c <( I L...--u 0 10 20 30 40 T i me (min .) Figu re 30. T y pical RPHPLC graph of sol u b l e organic fraction from fossi l Mercenaria. Labeled peaks were used for amino acid analysis. 46

PAGE 59

MM-2.2 Heated Dry I MM2200 -Peak 3 Peak 5 P eak 1 Ill -E s::: 0 c:o (I) 0 Peak 2 s::: 111 .c Pea k 6 .... 0 1/) .c <( f--\...-0 10 20 30 40 Time (min) Fig u re 31. RPHPLC graph of so lubl e organic fraction from modem M e rcenaria. This sa mple was h eated dry. Labeled peaks were us ed for a mino acid analysis MM-2 1 Heated Wet MM2100 ----Peak 1 Peak 3 Pea k 4 r--'1 e s::: 0 c:o N (I) P eak 5 0 s::: 111 .c .... 0 1/) Pea k 2 .c <( r-J 0 10 20 30 40 Time (min ) Figure 32 RPHPLC graph of soluble organic fraction from modem Mercenaria This samp le was he ated in water. Labeled peaks were u sed for amino acid analysis. 47

PAGE 60

E t: 0 co C1l C) t: I'U ..0 .... 0 1/) ..0 q: 0 PM-2 Protein Fractions PM200 Peak 2 10 20 r < ) 30 1me mm. 40 Figure 3 3. Typical RPHPLC graph of soluble organic fraction from modem Polinic es. Labeled peaks were used for amino acid analysis. PF-2 Protein Fractions PF200 -Peak 1 -E P eak 3 t: 0 co C1l Peak 2 C) t: I I'U ..0 1\ .... Peak 4 0 1/) ..0 q: f---0 10 20 30 40 Time (min.) Figure 34 Typical RPHPLC graph of soluble orgamc fractiOn from fossil Polmi ces L a beled peaks were used for amino acid analy s is. 48

PAGE 61

PM-5 Heated Dry PM500 Peak 1 -----E c:: 0 IX) C1l () c:: 111 .0 ... Peak 3 0 1/) .0 Peak 2 Peak 4 <{ 0 10 20 30 40 Time (min ) Figure 35. RPHPLC graph of solub l e organic fraction from modem Polinices This samp le was heated dry Labeled peaks were u sed for amino acid analysis PM-3 Heated Wet PM300 -----------------E Peak 1 c:: 0 Pe k 1 IX) I C1l () c:: 111 .0 ... 0 1/) .0 <{ Peak 2 Peak 3 f.-.._.. 0 10 20 30 40 Time (min.) Figure 36 RPHPLC grap h of so lubl e orgamc fraction from modem Po lmz ces. This san1p le was heated in water. Labe l ed peaks were u se d for amino acid a n alysis 49

PAGE 62

material from the SM was found in these peaks because each peak contained most of the individual amino acids. Refer to Appendi x 3 for amino acid abbreviations and Appendix 4 for all amino acid data. The amino acid results for the fractions collected from the modern Mercenaria mercenaria sample are shown in Figure 37 The percent composition of aspartic acid is relatively high in both fractions. Moderate percentages are shown for serine and glycine. Very low percentages (below 5%) are represented by threonine isoleucine. and leucine Proline and alanine only occur in peak 2 while arginine occurs only in peak 3. Peak 3 has generally higher percent compositions of hydrophobic amino acids (e.g., phen y lalanine and lysine), as compared to peak 2. The percent compositions for the fractions collected for the fossil Mercenaria mercenaria sample are shown in Figure 38. High relative concentrations of glycine and phenylalanine occur in most fractions of the fossil sample All fractions show very low percentages of isoleucine leucine, and arginine. A higher percent composition of hydrophobic amino acids (e.g. phenylalanine and histidine ) can be seen in the fossil fractions relative to the modern fractions and of the hydrophilic amino acids (e.g. aspartic acid and proline) in the modern compared to the fossil (Figure 41 ). Valine methionine and tyrosine are completely absent from all fractions of both modern and fossil. Proline is absent from all fractions ofthe fossil and only occurs in a very low percentage in the hydrophilic fraction of the modern An overall general decrease occurs in the number of amino acids present in the fossil fractions relative to the modern fractions. 50

PAGE 63

70 60 c: 0 50 1/) g_ 40 E 0 (.) 30 .... c: Q) 20 Q) a.. 10 Modern Mercenaria Amino Acid Composition asp thr ser glu pro gly ala val rret ile leu tyr phe his lys arg Amino Acids 1!1 Peak 2 Peak3 Figure 37. Modern Mercenaria amino acid compositions Shown are Peaks 2 and 3 from Figure 29 Fossil Mercenaria Amino Acid Composition 70 ----------------------60 c: 0 50 -.. 0 g_ 40 E 0 (.) 30 -c: Q) :: ____._.._..._ . c [b _..I_ .............. t asp thr se r glu pro g ly ala val rret ile leu tyr phe his lys arg Amino Acids i 111 Peak 1 1 Peak2 0 Peak 3 o Peak4 Figure 38. Fossil Mercenaria a mino acid composition s. Shown are Peaks 1-4 from Figure 30 51

PAGE 64

c 0 Ill 0 a. E 0 () ..... c Q) (J .... Q) a.. Modern Polinices Amino Acid Composition 25 ____ 20 15 10 5 0 asx t hr ser glx pro gly ala val rret ile leu tyr phe his lys arg Amino Acids 0 Peak 1 Peak 2 Figure 39. Modern Polinices am ino acid compositions. Shown are Peaks 1 and 2 from Figure 33. c 0 Ill 0 a. E 0 () ..... c Q) (J .... Q) a.. 70 60 50 40 30Fossil Polinices Amino Acid Composition asx thr ser glx pro gly ala val rret ile leu tyr phe his lys arg Amino Aci d s m Peak 1 Peak2 t O Peak 3 I O Peak4 __ .... Figure 40. Fossil Polinices amino ac i d compositions Shown are Peaks 1 4 from Figure 34. 52

PAGE 65

70 60 1:: 0 50 ;e 1/) g_ 40 . E 0 u 30 ... 1:: Modern vs. Fos sil Mercenaria Amino Acid Composition ] 0 -,..._..u..__....._........, ...... L!4 asp th r ser glu gly a l a leu phe his lys Amino Acids 2 M\1 1 Figure 41. Modern vs. fossil Mercenaria amino acid compositions. Only amino acids above 5% composition are shown. Modern vs. Fossil Polinices Amino Acid Composition 70 ---------------------------------------60 as x thr ser glx pro gly ala val ile leu phe lys Amino Acids 1 1 II!J PF 4 Figure 42 Modern vs fossil Polinices amino acid compo s itions. Only am ino acids above 5% compo s ition are shown 53

PAGE 66

Two fractions were collected from a modem Polinices duplicatus RPHPLC run for amino acid analysis (Figure 39). Comparison of the percent compositions in the fractions show a general trend of high percentages of hydrophilic versus hydrophobic amino acids The highest percentages are represented by aspartic acid and glycine Th e amino acid percent composition are shown for four fractions from the fossil Polinices in Figure 40. High percent compositions are seen in most fractions of proline and glycine Once again comparison of the modem and fossil fractions shows a definite decrease in the number of amino acids present in the fossil sample relative to the modern (Figure 42) There are overall lower percentages of aspartic acid proline and alanine and a higher percentage of glycine and glutamic acid in the fossil fractions compared to the modem. Comparing Figures 41 and 42 shows the difference between Mercenaria and Polinices. While proline is essentially absent from Mercenaria it is found in a rather high concentration in Polinices. This is also true of valine. Mercenaria contains a relatively greater amount of phenylalanine histidine and lysine. Heated Mercenaria amino acid compositions are shown in Figures 43 and 44 (heated dry and wet). The two samples show overall similar trends with the exception of less glycine alanine, and histidine and more proline and valine in the sample heated wet. Heated Polinices (Figures 45 and 46) seem to show a greater range of values under dry and wet conditions. The Polinices however, do show the same trend overall with less glycine and histidine more proline and slightly more valine. There is also an increase in aspartic acid and serine. 54

PAGE 67

45 40 c:: 35 0 ;e 30 II) 0 25 0 (.) 20 ..... c:: Q) 15 Q) c. 10 .. asp ser Heated Mercenaria Amino Acid Composition --.....,.,......-------' ----pro gly ala leu his tys Amino Acids El A3ak 1 1 A3ak 2 o A3ak 3 0 A3ak 5 A3ak6 Figure 43. Heated modem Mercenaria (MM-2.2 heated dry) amino acid compositions Shown are peaks 1 2 3 5 and 6 from Figure 31. Only amino acids above 5% composition are shown. Heated Mercenaria Amino Acid Compositions 45 -----___________ ...... ____ 40 c:: 35 . 0 ;e II) 30 : 0 0.. 25 E t 0 (.) 20 1 .... l c:: Q) 15 i 0 Jruill; I .... Q) c. 10 5 0 -JJJ .d asx th r ser glx pro gly ala val leu phe lys Amino Acids l!l A3ak 1 A3ak2 0 A3ak 3 I D A3ak 4 1 A3ak5 Figure 44 Heated modem Mercenana (MM-2 1 heated wet) ammo acid compositions. Shown are peaks 1-5 from Figure 32 Only amino acids above 5% composition are shown. 55

PAGE 68

r:: 0 ;e (/) 0 c. E 0 (.) -r:: Cl) (J ... Cl) a.. 80 70 60 50 40 30 20 10 0 Heated Polinices Amino Acid Composition _____ _. ____ -asp thr ser glu pro g ly ala val me t ile leu tyr phe his lys arg Amino Acids Peak 1 [;]Peak 2 Peak 3 0 Peak 4 Figure 45 Heated modern Polinices (PM-5, heated dry) amino aci d compositions. Shown are peaks 1-4 from Figure 35. r:: 0 ;e (/) 0 45 40 35 30 e 2s. 0 (.) s:: Cl) (J ... Cl) a.. 20 Heated Polinices Amino Acid Composition -----________ ___. ..a I IR:l asx thr ser glx pro g l y ala val rret ile leu tyr phe his l ys arg Amino Acids liE Peak 1 1 Peak2 : o Peak 3 Figure 46. Heated modern Polinices (PM-3, heated wet) amino acid compositions. Shown are peaks 13 from Figure 36. 56

PAGE 69

Stable Carbon and Nitrogen Isotopic Analysis The chromatographic runs u sed for stable carbon and nitrogen i sotope analysi s are s hown in Figures 47-50 At the bulk protein leveL large isotopic differences occurred between hydrophobic and h ydro philic fractions of both modern and fossil shells ; greater than 9 %o for 813C and 3 %o for 815N (Table I). Modern Mercenaria showed the hydrophilic fraction to have an average of -38.0 %o and 0 96 %o for 813C and 81 5N. respectively The hydrophobic fraction had an average of 21.0 %o and 5.0 %o. Fossil Mercenaria showed a hydrophilic average of -37.5 %o and 0 67 %o and a hydrophobic average of -27.5 %o and 3.7 %o for carbon and nitrogen respectively This is a difference of 17 %o for 813C and 4 %o for 81 5N in the modern and 10 %o for 813C and 3 %o for 81 5N in the fossil. Table 1 Averages of isotope dat a for modern and fossi l Polin ices and Mercenaria Polin ices Mercenaria 813C(%o) 81 5N(%o) 813C(%o) 81 5N(%o) Modern Fossil Modern Fossil Modern Fossil Modern Fossil Hydrophilic -37.5 -38 .8 -0.09 1.06 -38 0 -37.5 0.96 0 .67 Hydrophobic -28 .6 -29 7 10. 7 12.6 -21. 0 -27.5 5.0 3.7 Modern Polinices had a h y drophilic average of -37 5 %o and 0.09 %o and a hydrophobic average of 28.6 %o and 10. 7 %o for 81 3C and 815N, respectively. Fossil Polinices showed an average of -38.8 %o and 1 .06 %o for the hydrophilic fraction and 29.7 %o and 12.6 %o for the hydrophobic fraction Samples collected between peaks and those collected during the blank run had a nitrogen content too small to detect. Additionally. samples with fraction averages that are only repre se nt ed by a single data 57

PAGE 70

point either h a d a leak on the g a s line no gas the sample broke or there was a leak in the sample during i s otope analysis In all ca s e s. the h y drophilic fraction was depleted compared to the h y drophobic in both 813C and 815N. In addition when comparin g modern to fossil 813C values. th e f ossil protein s of the hydrophobic fraction are lower than those of the hydrophobic fraction of the mod e rn s ample Thi s i s s hown by a 7 %o difference in carbon and a 1.3 %o difference in nitrogen in the Merc e naria s ample. It is interes ting to note that this difference onl y occurs in the hydrophobic fraction; the hydrophilic fractions have very simi lar averages Permeability Results of the Microflow Permeability analysis can be found in Table 2 and Fi g ure 51. MF-1 was the most pem1eable s ample and MM-1.1 wa s th e mo s t impermeable o f tho s e an a lyzed No s ignificant trend wa s s een between permeabilit y and specie s (Me r c enaria was sli g htl y les s permeabl e than Polinices) however a trend did occur between modern ver s u s fos sil within the genera. As expected in both species the fossil sp e cimens were more permeable than the modem. An aluminum slu g was used for comparison and it is shown to be orders of ma g nitude less permeable than any of the shell samples (Table 2) 58

PAGE 71

IMM11011 ---------------------E c: 0 co Q) u c: ro -e 0 1/l ..c <( ,._ 0 10 20 30 40 lime (min ) I MM11021 E c: 0 co Q) u c: ro ..c ..... 0 1/l ..c <( ,___ 0 10 20 30 4 0 lime (min ) IMM11COOI -E c: 0 co Q) u c: ro ..c 0 1/l 1 I ..c A i <( 0 10 2 0 30 40 lime (min.) F i g u re 47. RPHPLC chromatograms ofMM used for carbon and nit r ogen isotope ana l y sis. Samples were collected from : 30-1:30 (pre P' peak) 2 5:30 ( P peak) 7-8 (post I s peak pre 2nd peak) 10:30 -16 (2"d peak) and 1 8-19 (post 2"d peak). T h e top two graphs are 2 runs of t h e s ample the bottom g r aph is the blank run (See Appen dix 2) 59

PAGE 72

jMF2021 E -c 0 co (I) (.) c ro .0 0 l (/) .0 <( 1-----0 10 20 30 40 lime (min ) IMF2031 E --c 0 co (I) (.) c ro .0 0 l (/) .0 <( 1-0 10 20 30 4 0 lime ( m i n ) IMF2COOI E c 0 co (I) (.) c ro .0 0 (/) .0 <( 0 10 20 30 40 lime (min ) F1g ure 48. RPHPLC chromato g ram s ofMF used for carbon and mtro gen isotope analysi s Samples were collected from :30-1 : 30 (pre 1 s t peak) 25 : 30 ( P peak) 7-8 (po s t I 5 1 peak pre 2"d peak), 10I 6 (2"d peak) and I 8-19 (post 2"d peak) The top two graphs are 2 runs of the sample, the bottom g raph is the blank run (See Appendix 2 ) 60

PAGE 73

I PM201 1 --E' c 0 co Q) (.) c cu .l:l '-0 (Jl .l:l .f"-..r..... A '-------' 0 1 0 20 30 40 Time (min ) IPM2021 E' c 0 co Q) (.) c cu -e 0 (Jl i .l:l .f'.-r... '"-0 10 20 30 4 0 Time (min ) IPM2COOI E' ----c 0 co Q) (.) c cu .l:l 'l i 0 (Jl .l:l 0 10 20 3 0 40 Tim e ( m i n ) Ftgure 49. RPHPLC chromatograms o f PM used for carbon and mtrogen I sotope a n a l ysi s Sample s were collected from :30-1:30 (pre P' p eak) 2:30-5 : 30 (1s t peak ) 7-8 ( po s t 1 s t peak pre 2"0 peak) 11-1 6 (2"d peak) and 18-19 (post 2"d peak) The top two gra ph s a r e 2 run s of t he sa mpl e the bottom graph is the blank run (See Append i x 2). 61

PAGE 74

I PF2011 ------E c 0 co Q) () c ro -e A_ 0 r/l .!) <( f----0 10 20 30 40 lime (min ) I PF2021 E -c 0 co Q) () c ro -e 0 \ r/l .!) <( I--0 10 20 30 40 lime (min ) I PF2BOOI E' c 0 co Q) u c ro .!) .... 0 (/) .!) l l <( 0 10 20 30 40 lime (min. ) Ftgure 50. RPHPLC chromatograms of PF used for carbon and mtrogen isotope analysis Samples were collected from :30-1: 30 (preP' peak) 2 : 30-5:30 (P' peak) 7-8 (post P' peak pre 2nd peak) 9-17 (2"d peak) and 18-19 (post 2"d peak). The top two graphs are 2 run s of the s amp l e the bottom graph is the blank run (See Appendix 2). 62

PAGE 75

Table 2. Permeability results. Sample Permeability ( cc /s ec / m/torr A) Averages MF-I 1.33212e-07 I .07048e-05 MF-2 Overflow* MF-4 2.I2763e-05 MM-I.I I. I 4086eI 0 3.3 8913e-09 MM-1.2 6.66417e-09 PF-1 1.00579e-04 5.02897e-05 PF-2 3.40386e-10 PM-I I. 70959e-08 8. 73 700e-09 PM-2 3.78101e-10 Blank Aluminum Slug 5.62456e-40 Too great a flow, was not able to be mea s ured usm g th1s techn1que even at a very low initial pressure. 63

PAGE 76

I Permeability I 1. 00E-03 1 00E 04 1 .00E-05 1 00E-06 .... E 1 00E-07 0 Gl "' 0 u 1.00E-08 1 00E-09 1 00E-10 1.00E-11 Figure 51. Comparison of permeabilities of modern and fossil bivalves and gastropods. These represent 2 samples. The lines are th e range of the samples with the top of the line representing one sample and the bottom of the line the other. 64

PAGE 77

DISCUSSION Description of Proteins As has been previously reported Mercenaria and Polinices contain an in so luble and soluble shell-matrix fraction. Only the SM was analyzed in this study be cause the so luble fraction has been found to contain almost all the protein present in the shell (Wei ner and Hood 1975 ; Weiner and Erez, 1984). The IM is difficult to work with and has been ignored by most researchers because ana lytical technique s depend upon separating molecules from solution i.e. liquid chromatography (Curry 1988). The SM comprised an average of 0.23% in Mercenaria an d 0.53% in Polinices (Table 3 a lso see Appendix 1 ). While the in so luble portion of all molluscs studied composed a minor fraction of the organic matrix in the bivalves the insoluble fractions constituted a greater portion of the organic matrix than in the gastropods. Several gastropod samp le s even showed an entirely soluble protein solution. Although at this time there is n o exact explanation for specimens demonstrating different characteristics, thi s was also observed by Robbins and Ostrom (1995) for Polinices. One possibility for the difference in the amount of insoluble fraction between these two types of molluscs may be the differences in their biomineralization processe s 65

PAGE 78

Table 3. Weight percent averages These are an average of three samples Percent Sol u ble Matrix MF 0.39% 0.23o/o MM 0.06% PF 0.84% 0.53% PM 0.21% The difference in organic weight percent between the two species (0.23% vs 0.53% overall, 0 39% vs. 0.84% and 0.06% vs. 0 21% for fossil and modem, respectively) indicates that Polinices has a greater potential for preservation of indigenous proteins than does Mercenaria. Presumably, some of the soluble matrix proteins have been leached out of Mercenaria s more permeable shell structure that was seen in the SEM photos Oddly in both specimens the fossil samples demonstrated more organics than the modem. This could be due to the amount of soluble versus insoluble organics since only the solubles are included in the weight percent. As the organism goes through diagenesis, the insolubles convert to soluble organics (Abelson, 1955) perhaps causing older samples to have a higher SM content. It has been shown that the soluble organics increase and insolub l e organics decrease with artificial diagenesis (Totten eta!., 1972). The soluble fraction contains the soluble protein and peptides derived from the insoluble protein by hydrolysis during diagenesis (Akiyama 66

PAGE 79

1971 ). causing the organic matrix to become progressively more soluble with age (Hare. 1969). However. ver y little is known about the insoluble proteins and their ability to solubilize over time (Robbins et al. 1993b). Even though the IM was not part of this study an insoluble fraction was weighed and did decrease from modem to fossil. Another possible explanation could be the age ofthe specimen. Smaller (presumably younger) samples show a higher organic weight percent than larger (presumably older) ones in both species (Appendix 1 ). These younger samples would need the organic layers as a template for mineralization to help them grow. Once the shell reaches maturity and shell precipitation slows these organics are no longer as important. Experiments of heated specimens also showed that Polinices contained more organic material than Mercenaria. Furthermore, heated specimens of Polinices in water had a higher SM content than the sample heated dry. This was the opposite affect than initially expected since the water should have caused the hydrolytic reaction to accelerate and degrade more protein. These specimens follow the same trend in regards to sample age as described above. PM-3 (heated in water) was a smaller specimen than PM-5 (heated dry). When size discrepancy is taken into account these two samples may have had roughly the same organic weight percent. Possibly the sample heated in water had more of the IM broken down by hydrolytic reactions into the SM. The sample was probably not heated long enough to have the SM leach out of the shell. This scenario would account for the higher SM percent in the sample heated wet than in the sample heated dry Contrary to the trend seen in Polinices the two Mercenaria valves (MM-2 1 and MM-2.2) had the same organic content as one another despite differences in heating. These data indicated that wet heating did not have much of an effect on the organic 67

PAGE 80

weig ht percent of the shell with th e time and temperatur e u sed. The we i gh t percents tend to be more s imilar to the unheated m ode rn s than t o the fossils. suggesti n g th at t h e heating experiments may ha ve only s imul ate d the very o n se t of dia g ene s i s. Vallent y n e ( 1 969) found th e tot al amount of amino acids de composed during p yrolysis to be greatest where no wate r was ad ded. The effect of water in d ec r eas in g the extent of decomposi tio n (or h avi n g no effect at all) is most probabl y a resu lt of dilution reducing the frequency of interaction of ami n o acid s both among them se lv es and with other components of the Mercenaria shell (Vallentyne, 1969) Other aspects of these heated s ample s showed that some diag e ne s i s mu st have occurred during the artificial aging. Wherea s the modern sample s protein powder (afte r l yoph ilization) appeare d white and powdery the heated samples showed a more brownish co lor as did the fo ss il s pecimens. Abelson (1956) a l s o observed that pro tein h ad turned brown in fossil sa mple s. Although h ea tin g may not h ave s hown much of an effect on weigh t percents the HPL C chromatograms more closely re sem bled the fossi l samp le s (see Figures 29-36), which was d i sc u sse d prev iously. RPHPLC A c har acteristic chrom a togr a phic profile was obtained for each of the sa mpl es ana l yze d (modem heated unheated and fossil ) ind icatin g similar components b ased on h y drophobicit y in each of the ge nera st udi e d C hromatogram s indicated there were approximately four major prote in/pe ptide compone nts. Each of the se peaks e lut ed a t about the same time and the sa me g r a dient percentage of acetonitrile (Buffer B) The 68

PAGE 81

hydrophilic fractions eluted at 0% Buffer B. and three hydrophobic peaks eluted between 50-95% acetonitrile. Comparison of the modem to fossil and heated chromatogram s showed that th e fossil/heated had more of a hydrophobic than hydrophilic fraction. This was most pronounced in the Polinices sa mples where there are basically no hydrophobic peaks in the modem and three distinct peaks in the fossil. Along the same lines the hydrophilic fractio n decreased with a su b seque nt increase in the hydrophobic fraction, suggest in g that the hydrophilic fraction broke down and was incorporated into the h ydrophobic fraction. Robbins and Ostrom (1995) also found the hydrophobic component of the fossil to be much larger than that of the modem s hell. These shifts are an indication of mild diagenesis and are most likel y due to the preferential loss of hydrophilic organ ic s from leac hin g processes by groundwater (Mitterer 1993) that the fossils were exposed to. This is one of two possibilities that may affect the changes in the characteristics of shell proteins during diagene s is: 1) degradation of the hydrophilic proteins with subsequent incorporation of the diagenetic products into the hydrophobic fraction and 2) the introduction of contaminants during diagenesis (Robbins and Ostrom 1995) Amino Acid Analysis The amino acid compositions of the protein fractions revealed differences between modem a nd fossi l samp les, suggesting diagenetic alteration in the fossils. An abunda n ce of aspartic acid in the modern suggests good preservation. The compositiona l decrease of this acid in the fossil often reflects diagene s i s (Akiyama, 1971 ). The preferential decr ease in the percentage of h y drophilic amino acids and s ub seq uent 69

PAGE 82

increase in the hydrophobic amino acids in the fossil fractions relative to the modern in both species provides evidence for alteration of the total composition by leaching processes (Mitterer 1993) The relative decrease in the overall numbers of amino acids in the fractions could be a function of biochemical reactions which graduall y break down some of the amino acids into other compounds which are not detected by the amino acid analyzer (Mitchell and Curry, 1997). The observed differences in the amino acid composition among the samples may also be related to the differences in the pathway of and extent of diagenesis experienced b y each individual fossil. Those proteins rich in aspartic and glutamic acid are the acidic glycoproteins (Robbins and Brew 1990). The fractions rich in serine, glycine, and alanine are similar to fibrous structural proteins which help promote the structural integrity (Robbins and Brew 1990). If these are indeed structural proteins the y may be less prone to degradation than some other types of proteins because of their composition and fibrous structure (Robbins 1987). While it has been suggested in other studies that the hydrophilic fraction is comprised of glycoproteins and the hydrophobic fraction is comprised of structural proteins (Robbins and Brew 1990) the distinction between these two groups of proteins is unclear in this study. The fractions showed amino acid compositions characteristic of both groups of proteins indicating a heterogeneous mixture. Mercenaria showed a clear increase in glycine and alanine from modern to fossil. The occurrence of glycine and alanine as degradation products is not unusual and has been known to arise from serine (Vallentyne 1964 ; Mitchell and Curry 1997). There was a slightly higher concentration of serine in the modern sample (8.25% vs. 5.4% for modern and fossiL respectivel y ) 70

PAGE 83

Th e m a in differen ce b e tween the specie s was greater am o unt s o f proline g l y c i n e. alanine. a nd valine in Polini ces I s oleucin e and leucine were also s lightl y higher in thi s ga s tr o pod M e r ce naria sh o w e d lar g e amount s of phen y lalanine hi s tidine. and l ys i ne These amino acids are not usuall y found in any g reat quanti ty in s hell protein s. Thi s suggests that these are cont a mination product s that may be found in M e r c enar i a b eca u se of its mor e porous structur e a s ha s been d e m o n s tr a t e d through o ut t hi s s t udy. H owever. comparing the 2 s pecie s (Figure s 41 and 42), s hows overall similar trend s i f y ou disregard those last three amino acids. So, there seems to be a s much if not more of a difference in amino acid compositions between modem and fossil samples of the same genus a s there is between the s pecies (Figure 41 and 42). Despite e arlier studie s th at h a d shown amino acids to be phylogenetically distinct among specie s rec e nt work (Robb i n s and Brew, 1 9 90 ) indicated th at amino acid anal ys i s o f total s h e ll m a t e ri a l repre se n ts a n o vers implification of component s in the org ani c material within the s he ll. Thi s c o uld b e why data on total amino acid compositions demonstrates variabilit y and deviations from what is expected phylogenetically Therefore amino acid anal y si s s e em s t o be a m o re u s eful tool in diagenetic s tudie s than in phylogen e tic determina t i o n Stable Carbon and Nitrogen Isotopic Analysis Although a similar study comparing hydrophilic and hydrophobic fractions of modem and fo ssil shells was p e rformed u s ing th e s ame genera ( R o bbin s a nd O s tr o m. 1995) th e ir f ocu s was on isot o p e data for individu a l amino acids. O ther s tudie s h ave concentrated on amino acid data ( Ostrom et al. 1990 ; Qian e t al. 1 99 5 ), or more 71

PAGE 84

specifically on the o 13C of Dand Lenantiomers (Engel and Macko 1986 ; Serban et al.. 1988; Engel et al. 1994; Ostrom et al., 1994 ; Silfer et al. 1994). In this research. bulk proteins were isotopically analyzed to see if similar trends occurred at this highe r molecular level. Ver y similar trends did occur in all aspects of the data confim1ing that proteins and their constituent amino acids act similarly with respect to o13C and o1;N. Proteins from the hydrophobic fraction of the modem Polinices and Mercenaria demonstrate o13C values that are at least 9 %o higher than the corresponding values in the hydrophilic fraction. Since the o13C of an amino acid is partly controlled by the degree of fractionation during rate-determining steps and at major branching points that occur along degradation pathways (Robbins and Ostrom 1995), this difference could be attributed to variations in the extent of fractionation within pathways of synthesis of these two different types of proteins. The most interesting feature of the o13C data is that the values from the fossil hydrophobic fraction are lower than the corresponding modem hydrophobic fraction as a result of the loss of 13C depleted peptides from the hydrophobic fraction during diagenesis (Robbins and Ostrom, 1995). One explanation for this involves the kinetic isotope effects where the 13C depleted atom is preferentially incorporated into the product (Macko et al., 1986). Another explanation offered for these lower values is the reincorporation of isotopically depleted degradation products from the hydrophilic fraction (Robbins and Ostrom 1995). This does not seem to be the case here because averages of modem and fossil Mercenaria hydrophilic fractions are similar. Additionally, as the amino acid composition changes over time due to diagenesis the isotopic composition of bulk 72

PAGE 85

organic matter ma y also change, re g ardles s of contamination due to lar ge differe n ces in iso topic composition of individual amino acids (E ngel et al., 1994 ) This would b e evidence for diagene s is but not nece ss aril y contamination in the fossil samples Fossil Mercenaria showed a large divergence in nitrogen between the duplicate runs. Se ve ral factors may have caused this and other discrepancies. For one. sa mples had to be tran s f e rred from the roto-evap ball asts into quartz tubes Although great care was taken some l oss may have occurred during the tran s fer. Also samples were heated at 55-60 C for times ran ging from 15-45 minutes during rotary evaporation. Although isotopic integrity during this step has been determined (Feuerstein et al. 1997) heating may have caused molecular alteration to an unknown extent. Clearl y, there are a number of complex proteins with in the shell matrix and these have differin g isotopic compositions Furthermore s hell matrice s contain other organic material s uch as carbohydrates and lipid s, which have n o t yet been eval uated i s otopically It is l ikely that the interactions of s hell matrix proteins and their hydrolytic products wit h the other organic compou nd s, and the effect of the carbonate matrix which ma y protect the protein s from h y drolysis or catalyze the hydrolytic process significantly complicate isotopic fractionation (Qian 1993 ). Because the carbon values were highly depleted compared to amino acid data (Robbins and Ostrom 1995) it was suggested that EDT A may be contributing to these depleted values. Although samples we r e exhaustively dialyzed to remove th e EDT A, it has been implied t h at not all the EDT A can be removed by di a l ys i s ( Weiner et al.. 198 3; Alb eck et al., 1 996). EDTA u s ually elutes in the first peak ; thi s may mean that the actua l 73

PAGE 86

isotope value s of the hydrophilic fraction are higher, resulting in less of a difference between the two fractions. In order to further understand this, more samples wer e prepped using hydrochloric acid (HCl) rather than EDTA to decalcify the shells ( Table 4). Although the actual values are less depleted. the same general trends occur in the data. with less of a difference between fractions. The hydrophilic frac tion is still depleted compared to the hydrophobic although not by as much and the hydrophobic fractions of the fossil are still depleted compared to the modern by a significant amount (7 %o in carbon and greater than 10 %o in nitrogen). Table 4 Isotope results for samples decalcified with HCl. o13C ols N Polinices Modern Fossil Modern Fossil Hydrophilic -22.6 -28.8 2.2 -1.7 Hydrophobic -21 1 -28.2 9.6 -3.1 Furthermore comparison of modern gastropods from different environments (seagrass vs. muddy estuary) and different eating habits (herbivorous and carnivorous) demonstrate that the isotopic composition of the total organic material within the she ll is related to the carbon isotopic composition of the primary producer/consumer that the gastropod eats (Table 5) For example, Polinices from a seagrass bed shows much higher o13C values than that of Polinices from a muddy estuary setting. If this is the case, isotopes would not be as useful as a diagenetic indicator as once previously thought. Despite this several pathways of protein diagenesis could contribute to isotopic differences in o 13C of proteins between and within the organic fractions of modern and fossil shells Simply, fossil proteins may be a diagenetic product of the corresponding modern precursor. Individual fossil proteins could be derived from different fractions of 74

PAGE 87

the modern precursors through degradation and rearrangement of the diagenetic product. To delineate between different diagenetic mechanisms and isotopic variabilit y. further work is needed. Table 5. 81 3C values (in %o) for total shell organic hydrolyzate for Strombus and Polinices. Specimen s came from distinctively different environments: A Florida seagrass bed and a temperate sandy mud estuary in Massachusetts (Robbins et al.. 1998). Sea grass Estuary Polinices (Carnivore) -5 3 -24 5 Strombus (Herbivore) -16 3 Permeability When interpreting permeability data an important trend occurs between modern and fossil specimens, with the fossil specimens demonstrating a higher permeability than the modem samples This higher permeability would contribute to the diagenetic alteration of proteins that is evidenced in other data, such as HPLC chromatograms, amino acid analysis isotopic data and SEM photos Although Mercenaria appears to be slightly less permeable than Polinices, this difference may not be significant. In fossil Mercenaria, sample MF-2 was so permeable it could not be measured (see Table 2) and therefore is not included in the average Also fossil Polinices shows great variability ranging from I 04 to I o-1 0 This variability could have been caused by a leak in the seal during analysis (Gupta, pers. comm.) Finally PM and MM demonstrate permeability of the same order of magnitude and therefore may not be significant. This data has never 75

PAGE 88

been recorded so the difference between modern and fossil is an interesting beginning t o the documentation of permeabilities re l ation to dia ge ne s is. In addition when comparin g with the soluble matri x we i g ht p e rcents ( Figure 5 2). seve r a l intere st in g po i nt s become evi d ent. Referring back to the se weight percents (A ppendi x 1 ). MF-1 h ad a very low organic content ( 0 09% ) compared to other fossil Mercenaria sa mples This may a lso reflect the high permeability of this sample. Other examples of thi s include MM -1.1 versus MM-1.2 where MM-1.1 was les s permeable and had a higher we i gh t percent than MM-1.2 PF-2 is quite impermeable a nd had an unusuall y high organic wei g ht. And finall y, PM-1 i s more permeable than PM-2 and has a lo wer percenta g e of or ga nics Fi g ure 52 has much variability, but the trendline does show an increa se of organics with a decrease in permeability. MF-4 was not included on Figure 52 because part of it's SM was lost during anal ys is and therefore cannot be used in com p ariso n. So although n ot hin g conclusive can be said about whether perm eabi lit y is spec i es determined diagenesis does appear to increase permeability and perhaps there is al so some trend between permeability and organic weight percent on an individual sample basis Future Directions Several other biogeochemical techniques could be performed to further characterize and under s tand protein diagenesis More s pecific conclusions from the molecular dat a require sequencing of the protein or may be obtained through other molecular o r immunological approaches (Robbins et al., 1993b ; Ostrom et al., 1998). Successful seq uencin g of the proteins e n ab le s the re sea rch e r t o determin e the 7 6

PAGE 89

substitution s deletions and insertions of amino acids that have occurred within the polypeptide chain (Robbins and HealyWilliams 1991 ). Therefore sequence data can provide further insight into the role that the protein may play in the biomineralization of the shell and will aid in modeling degradation. ---------Sample Weight vs. Permeability I 1 .ooE-66 ? ... 0 1 00E-07 E () Cl) ..!!! o 1 00E-08 >. .c 1.00E-09 E ... Cl) a.. 1 00E-10 0 0 2 R2 = 0 2136 0.4 0 6 0 8 1 2 1.4 Sample Weight (g) 1 6 Figure 52. Corre lation of sample weight and permeability. Black line represents average trend lin e. DNA sequencing of living and ancient representatives has exciting implications in the establishment of phylogenies and the study of evolution and is a likely research direction for the future (Paabo, 1989) Ribosomal RNA in living organisms can be obtained using the polymerase chain reaction (PCR ) to amplify the sm all amounts of material available Once amplified rRNA can then be directly sequenced These techniques have been successfully applied to living molluscs (Terret eta!., 1996) a nd 77

PAGE 90

have exciting prospect s for being extended to fossilized material. Answer s to phylogenetic biomineralization, and diagenetic questions will likely come from u si n g a co mbination of these techniques Conclusions 1 ) The EDT A-soluble shell matrices of the species studied are composed of het e rogeneou s assemblages of proteins and polypeptide s as indicated by RPHPLC chromatograms and amino acid analysis. 2) SEM confirms that gastropods have a tighter ultrastructure. SEM and permeability experiments demonstrate crystal alteration in the fossils 3) Pyrol ys is on whole specimens demonstrated that shell matrix protein s were o nl y s lightly modified compared to earlier report s of more degradation in powdered samples However some diagenetic alteration during artificial aging was indicated b y HPLC and amino acid analysis The data suggest that temperature studies on whole specimens, rather than shell fragments may be a more accurate way to model in situ degradation 4 ) Amino acids show similarities between ge nera and may not be as u se ful in ph y logenetic studies. Some amino acids ma y be more phylogeneticall y useful than others. However amino acid analy sis can contribute to biomineralization and dia ge netic questions. 78

PAGE 91

5) Stable isotopes show differences between the two different fractions of modern. and also between modern and fossil. Isotope data suggest bulk proteins reflect diagenesis and protein type. Great care must be exercised when extracting stable isotopes from organic components. Man y factors may contr i bute to their depletion and enrichment. and interpretation i s often difficult and signals may be marred by a combination of circumstances. 6) A combination of all the methodologies used in this study, including HPLC amino acid analysis stable isotopic analysis, SEM and permeability can help in understanding diagenetic alteration. 79

PAGE 92

LIST OF REFERENCES Abelson P.H., 1954 Organic constituents of fo ss il s, Carneg i e In s titution ofWashingr on Yearbook vol. 53, pp. 97-101. Abelson P.H., 1 955 Organic constituents of fossils. Carnegie Institution ofWashin gto n Yearbook vol. 54 pp. 107-109. Abelson P.H 1 956, Paleobiochemistry Scient(fic A m erica n vol. 1 95 pp. 83-92 Addadi, L. and Weiner S. 1989 Stereochemical and structural r e lations between macromolecules and crystals in biomineralization in Biomineralization: Chemical and Bioche mical P e r s pective s S Mann, J. Webb, and R.J.P. Williams. eds .. VCH Verlagssgesellschaft Weinheim FRD, pp. 133 -156 Akiyama, M 1971 The amino acid composition of fossil scallop shell proteins and nonprotein s Biominerali z ation Research Report s, vol. 3, pp. 65-70 Albeck, S. Aizenberg J ., Addadi, L. and Wei ner, S ., 1993. Inter actions of various skeleta l intr acrys tall i n e components with calcit e crystals J ournal ofAmerican C hemical Soc i ety vol. 115 pp. 11691 -11697 A l beck, S., Weiner, S. and Addadi, L. 19 96 Polysaccharides of intracrystalline g l yco protein s modul ate calcite crystal growth in vitro, C h ern. Eur J., vol. 2, no 3, pp. 278-284. A llison P .A. 1988 The role of anoxia in the d ecay and mineralization of proteinaceous m acrofossils Paleobiology vo l. 14 no. 2 pp. 139-154. Ambler, R.P. and Dani e l M 1991 P rotei n s and m o lecular paleontology Philo so phical T ra n sac ti o n s of th e Royal Soc i ety of L ondo n B vol. 333 pp. 381-389. Andrews, J.T. Bowen, D.Q. and Kid so n C. 1979 Amino acid ratio s and the correlation of raised beach deposi ts in south-west England and Wales, Na tur e vol. 281 pp. 556-558. Andrews, J T. Miller, G.H. Davie s D.C., and D av i es, K.H. 19 85 Generi c id e ntification of fragmentary Quaternary m o llu scs b y amino acid chromatography : A tool for Quate rnary and paleont o l ogica l researc h Geologica l Journal vol. 20, pp. 1 -2 0. 80

PAGE 93

Anfinsen C. B .. 1959 The Molecular Basis of Evolution John Wiley and Sons. In c .. Ne,.,, York. 228 p. Bada, J .L., 1991, Amino-acid cosmogeochemistry, Philosophical Transactions of the Royal Society of London, Series B-Biological Sciences vol. 333 pp. 349-358. Carter, J.G 1990 Skeletal Biomineralization: Patterns, Processes. and EvolutionarY Trends. Volume I Van Nostrand Reinhold, New York. New York. 832 p. Carter. J.G. and Cla rk. G.R .. 1985 Classification and phylogenetic significance of molluscan shell microstructure. Mollusk-Short Course. Crenshaw, M.A 1972 The solub le matrix from Mercenaria mercenaria shell. Biomineralisation vol. 6 pp. 6-11. Crenshaw M.A., 1980 Mechanisms of shell fonnation and dissolution in Skeletal Growth of Aquatic Organisms: Biological records of environmental change Rhoads, D.C., and Lutz, R.A. eds .. Plenum Press New York, 750 p. Curry G.B. 1987a Molecular paleontology Geology Today pp. 12-16. Curry G.B. 1 987b Molecular paleontology: New life for old molecules Trends in Ecology and Evolution vol. 2, no. G, pp. 161-165. Curry G.B ., 1988 Amino acids and protein s from fossils, in B. Runnegar and J .W. Schopf eds. Molecular Evolution and the Fossil Record : Short Courses in Paleontology Number 1 The Paleontological Society, University ofTennessee, Knoxville. Tennessee, pp. 20-33. Degens E.T Spencer D.W. and Parker R.H. 1967 Paleo biochemistry of molluscan shell proteins Comp Biochem Physiol. vol. 20 pp. 553-579 Eglin ton G. and Logan G .A. 1991, Molecular preservation, Philosophical Transactions ofthe Royal Society a_( London B, vol. 333, pp. 315-328. Engel, M.H., Goodfriend, G.A., Qian Y., and Macko, S.A. 1994 lndigeneity of organic matter in fossils: A test using stable isotope analysis of amino acid enantiomers in Quaternary mollusk shells Proceedings o_fth e National Academy o_(Science USA, vol. 91, pp. 1047 5 -10478 Enge l M.H. and Macko S.A . 1 986. Stable isotope evaluation of the origins of am ino acids in fossils Na tur e vol. 323, pp 531-533. 81

PAGE 94

Feuerstein. T.P., Ostrom, P.H. and Ostrom N.E. 1997. Isotopic biogeochemistry of dissolved organic nitrogen : A new technique and application. Organic Geochemist1y. vol. 27. no. 7 / 8 pp. 363-370. Ghiselin. M.T., Degens, E.T., Spencer. D.W. and Parker. R.H .. 1967 A ph y logenetic survey of molluscan shell matrix proteins. Breviora no. 262 pp. 1-35 Hare P .E. 1969 Organic chemistry of proteins pep tides and amino acids, in Eglinton. M. and Murphy eds. Organic Chemistry, Springer-Verlag New York. 828 p. Hare P.E and Abelson P H., 1964 Proteins in mollusk shells Carnegie Institution of Washington Yearbook vol. 63, pp. 267-270. Hare P.E. and Abelson P.H. 1965, Amino acid composition of some calcified proteins Carnegie Institution a_[ Washington Yearbook vol. 64 pp. 223-232. Hare P.E. and Hoering, T.C ., 1977 The organic constituents of fossil mollusc shells, Carnegie Institution ofWashington Yearbook, vol. 76, pp. 625-631. Hennet, J.C., Holm, G., and Engel M.H., 1992 Abiotic synthesis of amino acids under hydrothermal conditions and the origin of life: Perpetual Phenomenon ? Naturwissenschaf ten vol. 79, pp. 361-365. Herrmann. B. and Hummel S., eds .. Ancie nt DNA : Reco v ery and analysis of genetic material from paleontolo g ical archaeological mu se um medical. and forensic specimens, SpringerVerlag, New York 263 p. Hudson J.D. 1967, The elemental composition of the organic fraction and the water content, of some recent and fossil mollusc shells, Geochimica et Cosmochimica Acta, vol. 31, pp. 2361-2378. Hunkapiller M. Kent S., Caruthers, M., Dreyer W., Firca J. Griffin C. Horva t h S., Hunkapiller T. Tempst P., and Hood, L., 1984a A microchemical facility for the analysis and synthesis of genes and proteins, Nature, vol. 310 pp. 105-111. Hunkapiller M.W., Strickler, J.E., and Wilson, K.J ., 1984b Contemporary methodology for protein structure determination Science vol. 226, pp. 304-311. Kahne, D and Still W.C. 1988 Hydrolysis of a peptide bond in neutral water Journal o.f American C hemical Society, vol. 110 pp. 7529-7534. Kennish M.J. 1980 Shell microgrowth analysis: M e rcenaria mercenaria as a type example for research in population dynamics in Skeletal Growth of Aquatic Organisms: Biological re co rd s of environmental change Rhoad s D.C. a nd Lutz R.A ., eds., Plenum Pre ss, New York, 750 p. 82

PAGE 95

King, K. Jr.. 1977. Amino acid survey of recent c a lcareou s and si liceous deep -sea mi cro f ossils, Micropaleontology vol. 23, no 2 pp 180-193 King, K Jr. and Hare P .E., 1972a Amino acid composition of planktonic foraminifera: A pal eo biochemical approach to e v olution Science vol. 175 pp 1461-1463 Kin g, K Jr. and Hare P.E., 1972b, Amino acid composition of th e test as a ta xo nomic character for li v in g a nd fossil pl a nktonic foraminifera. Mic ropaleontology. vol. 18, pp 285 -29 3. Lowenstam H.A. 1981, Minerals formed b y or ga nism s, Science vol. 211 pp 11261131. Lowenstam, H A. and Weiner S., 1989 On Biomineralization Oxford Univer sity Press New York, 324 p Lowenstein J .M., 1980 Species specific proteins in fossils, Naturwissenschaften vol. 67 pp 343-346. Macko S.A. and Aksu, A.E 1986 Amino acid epimerization in planktonic foraminifera s u gges t s low se dimentation rate s for Alpha Rid g e Arctic Ocean Nature vol. 322. pp 7 30-732. Macko, S .A., Estep M .F., Engel, M.H., and Hare P.E ., 1986, Kinetic fractionation of stable nitro g en i s otopes durin g amino acid tran s amination Ge ochimi ca et Cos m oc himi ca Acta vol. 50, pp. 2143-2146 Macko S .A Este p M.L.F., Hare P.E. and Hoerin g T.C. 1 983, Stable nitrogen and carbon i so topic composition of individual amino acids isolated from cultured microorganism s C arne g ie Institution ofWashington Y ea rbook, vol. 82 pp 404 409 Matter III P., David so n F.D. and Wyckoff R.W.G. 1969, The composition of fossil oyster shell protein s, Proceedin gs of the National Academy ofScience USA, vol. 64 pp 970-972. Mitchell L. and Curry G.B 1997, Diagenesis and s urvival of intracrystalline amino acids in fossil and recent mollu s c s h ells Palaeontology vol. 40 p a rt 3 pp. 855874 Mitterer R M. 1993 The diagenesis of protein s and amino acids in fossil s hell s, in Organic Geochemistry M .H. Engel and S .A. Macko eds ., Plenum New York pp. 739-753. 83

PAGE 96

Moore, R .C., 1969. Treati se on Invertebrate Paleontology (N) Mollusca 6. The Geological Society of America. The University of Kansa s. 460 p. Mutvei, H., 1970 Ultrastructure of the mineral and organic components of molluscan nacre ous layers. in Biomineralization Research Reports. F .K. Schattauer Ver lag. Stuttgart. Germany pp. 49-72. Muyzer, G., We s tbroek P., and Wehmiller J.F., 1988 Phylogenetic implications and diagenetic stability of macromolecules from Pleistocene and recent shells of Mercenaria mercenaria (Mollusca, Bivalvia) Historical Biology vol. 1 pp 135144. Nagy, B., Engel, M.H. Zumberge I.E. Ogino, H. and Chang, S.Y., 1981. Amino acids and hydrocarbons 3 800-Myr old in the Isua Rocks southwestern Greenland. Nature, vol. 289 pp. 53-56. Nakahara, H., Bevelander, G., and Kakei M. 1982, Electron microscopic and amino acid studies on the outer and inner shell layers of Haliotis rufescens Venus vol. 41, no. 1, pp. 33-46. Olsson A.A., Harbison, A., Fargo, W.G. and Pilsbry H.A. 1953 Pliocene Mollusca of Southern Florida Wichersham Printing Co. Lancaster Pennsylvania. 457 p. Ostrom P.H. Ghandi H. Ca hue L. Gage D.A. Shen. T., Huan g, Z., and Hau s chka. P.V. 1998, New perspectives on a ncient proteins: The application ofMALDIMS for characterization of modern and ancient osteocalcin protein sequences Programs and Abstracts: Perspectives in Amino Acid and Protein Geochemistry Washington, D.C. p. 32. Ostrom, P.H. Macko S.A., Engel M.H. Silfer, J.A. and Russel D., 1990 Geochemical characterization of high molecular weight material isolated from Late Cretaceous fossil s, Organic Geochemistry, val. 16, pp 1139-1144. Ostrom, P.H ., Zo nneveld J., and Robbin s, L.L., 1994 Organic geoc hemistry of hard parts: Assessment of isotopic variability and indigeneity, Palaeogeography, Palaeoclimatology, Palaeoecology, vol. 107, pp. 201-212. Paabo, S. 1989 Ancient DNA: Extraction, characterization molecular cloning and enzymatic amplification, Proceedings of the National Academy of Science USA, vol. 86, pp. 1939-1943. Qian Y., 199 3, Kinetic aspects of the diagenesis of organic compound s and the associated kinetic isotope fractionations Ph.D. Dissertation, University of Oklahoma pp. 141-226. 84

PAGE 97

Qian Y., Engel, M.H. Goodfriend G.A. and Macko, S A. 1995, Abundance and stabl e carbon isotope composition of amino acids in molecular weight fractions of fossil and artificially aged mollusk shells, Geochimica et Cosmochimica Acta, vol. 59. no.6, pp.1113-1124. Robbins, L.L. 1987, Morphological variability and protein isolation and characterization of recent planktonic foraminifera. Ph.D. Dissertation. University of Miami. 302 p. Robbins. L.L., Andrews, S., and Ostrom P.H. 1998 Carbon and nitrogen isotopic variability of different protein fractions from modern and fossil shells. Programs and Abstracts: Penpectives in Amino Acid and Protein Geochemistry Washington D.C. p. 60. Robbins L.L. and Brew, K., 1990, Proteins from the organic matrix of core-top and fossil planktonic foraminifera, Geochimica et Cosmochimica Acta, vol. 54, pp. 22852292. Robbins, L.L. and Donachy, J.E. 1991, Mineral regulating proteins from fossil planktonic foraminifera Surface Reactive Peptides and Polymers, American Chemical Society pp.139-148. Robbins L.L. and HealyWilliams, N 1991, Toward a classification of planktonic foraminifera based on biochemical geochemical and morphological criteria. Journal of Foraminiferal Research vol. 21, no. 2 pp. 159-167. Robbins L.L., Muyzer G. and Brew, K. 1993b, Macromolecules from living and fossil biominerals: Implications for the establishment of molecular phylogenies, in Organic Geochemistry M.H. Engel and S.A. Macko, eds. Plenum, New York pp. 799-816. Robbins, L.L. and Ostrom P.H., 1995 Molecular isotopic and biochemical evidence of the origin and diagenesis of shell organic material, Geology, vol. 23, no. 4, pp. 345-348. Robbins, L.L., Toler, S.K. and Donachy J.E. 1993a, Immunological and biochemical analysis of shell matrix proteins in living and fossil foraminifera, Lethaia vol. 26, pp. 269-273. Serban, A., Engel M.H. and Macko S.A. 1988 The distribution, stereochemistry and stable isotopic composition of amino acid constituents of fossil and modern mollusk shells Organic Geochemistry vol. 13, pp. 1123-1129. Silfer J .A., Qian Y., Macko, S.A., and Engel, M H., 1994 Stable carbon isotope composition of individual amino acid enantiomers in mollusc shell by GC/C/IRMS Organic Geochemisfly vol. 21, no. 6/7, pp. 603-609. 85

PAGE 98

Terrett. J.A.. Miles. S . and Thomas R H ., 1996 Co mplete DNA sequence ofthe mitochondrial ge nom e of Cepaea n emo rali s (Gas tropoda : Pulmonata). J o urn a l(?( Molecular Evo l ution. vol. 42. no 2 pp 160-168 Totten. D.K., Da v id so n F.D . and Wyckoff. R.G 1972 Amino acid composition of heated oyster shells. Proceedings of the Natio nal Academy o(Science USA. vo l. 69 pp 784-785 Vallentyne, J R. 1964 Biogeochemistry of organic matter-II : Thermal reaction kinetics and transformation products of amino compounds G eoc himica et Cos mochimi ca Act a, vol. 28, pp I57-I88. Vallentyne, J.R I969, Pyrolysis of amino acids in Plei s tocene Mercenaria shells Geoc himica e t Cos mochimica Acta, vol. 33, pp. 1453-I458. Watabe N., I965 Studies on shell formation: XI. Crystal-matrix relationships in the inner la y ers of mollusk shells Journal of Ultrastructure Re se arch vol. 12 pp. 351-370. Weiner S., 1986 Organization of extracellularl y mineralized tissues: A comparative s tud y of biol og ical crystal growth CRC Crit. R e v Bio c h e m. vol. 20, pp 365-408. Weiner. S an d Erez. J. I984 Organic m a trix of the shell of the foraminifer Heterostegina depressa Journal R esearch, vol. 14. no. 3 pp. 206-212. Weiner S and Hood L. 1975 Soluble proteins of the organic matrix of mollusk shells: A potential template for shell formation Science, vol. 190 pp. 987-988 Weiner, S ., Lowenstam H A. and Hood, L. 1976 Characterization of 80-million-year old mollusk shell proteins Pro cee dings of the National Academy of Sci e nce, USA, vol. 73, no 8 pp. 2541-2545. Weiner S. Lowenstam, H A. Taborek B., and Hood L., I979 Fossil mollu s k s hell organic matrix components preserved for 80 million years Paleobiology vol. 5 no 2 pp. 144-150. Weiner, S. Talmon, Y., and Traub W., 1983 Electron diffraction of mollu sc shell organic matrices and their relationship to the mineral phase Int e rnational J o urnal of Biological Macromolecules, vol. 5 pp 325-328. Weiner S and Traub W., 19 80, X-ra y diffraction s tud y of the in so luble organic matrix of mollu sk s hells. FEBS L ette rs vol. II, n o 2 pp 311-316. 86

PAGE 99

Weiner S. and Traub, W., 1984 Macromolecules in mollusc shells and their function in biominerali z ation Philos o phical Transa c tions of the Royal Society of London. Seri e s B vol. 304 pp. 425-434. Weiner S., Traub W., and Lowenstam H.A. 1983 Organic matrix in calcified exoskeletons in Biomineralization and Biological Metal Accumulation P. Westbroek and E.W de J o ng eds. D. Reidel Publishing C ompan y. H olla nd. pp 205-224. Wheeler A. P Ru s enko K.W., Swift D.M. and Sikes C.S. 1988 Regulation o f in v i tr o and in vivo cr y stallization b y fractions of oyster shell organic matri x Marine Biology vol. 98, pp. 71-80. Wilbur, K M. 1974 Recent studies of invertebrate mineralization, in The Mechanisms of Minerali z ation in the Invertebrates and Plants, N. Watabe and K.M Wilbur eds. Univer s ity of South Carolina Press Columbia South Carolina pp. 79-108. Wilbur K.M. and Manyak D.M. 1984 Biochemical aspect s of molluscan shell mineralization, in Marin e Biodeterioration: An Interdisciplinar y Stud y J.D C o s tlo w and R .C. Tipper eds ., Naval Institute Press Annapoli s Maryland pp. 30-37. Wind F .H. and Wise Jr., S.W. 1974 Organic vs. inorganic processes in archaeogastropod s hell mineralization in The Mechanisms of Mineralization in the Invertebrates and Plants N Watabe and K.M. Wilbur eds University of South Carolina Press Columbia South Carolina pp. 369-387. 87

PAGE 100

APPENDICES 88

PAGE 101

00 "' Appendix 1. Sample weights, tests run and weight percent. Sample Tests Run Original Powder Weight (g) MF-1 p, aa 80.9 MF-2 CIN, p, aa 11.72 MF-3 a a 15.70 MF-4 p,aa 35.88 MM-1 1 CIN, p aa 101.65 MM-1.2 p,aa 105 34 MM-3.1 191.02 PF-1 p,aa 10. 86 PF-2 CIN, p aa 3.27 PF 3 10.86 PM-I p aa 90.56 PM-2 CIN p, aa 28.99 PM-7 2 1 .70 (Appendix 1. Contin ued on next page) Organics Remaining (g) Weight% .07526 .09% .0627 .5% .089 .57% .073 .2%* .09667 095% .02334 .02% .089 05% .0640 6% .0478 1.46% .05055 .47% .10962 .12% I 10159 .35% 03222 .15%

PAGE 102

\{) 0 Appendix 1. (Continued) Heated Samples PM-3 (wet) a a 17.86 PM-4 (dry) 40.83 PM-5 (dry) a a 22.57 MM-2.1 (wet) a a 149.97 MM-2.2 (dry) aa 160.03 Lost part of sample. CIN-Stable carbon and nitro ge n i so tope analysis p Permeability aa Amino Acid analysis .072 .4 % 005 .01% .031 .14 % 112 .07 % 128 .08 %

PAGE 103

'-0 Appendix 2. Sample ID's description co ll ection tim e, data, and other notes for isotope analysis. ID Description Collection time Date run Nts cJ BSA03 #I pre-peak on I st BSA run I2-I3:20 I0/06 too small no sampl e BSA03 #2 peak of I st BSA run I4:30 -2I :30 I0/06 8 6 -14.79 BSA03 #3 post-peak on 1 s t BSA run 23 24:20 10 / 06 too small -23.4 BSA04 #I pre-peak on 2nd BSA run I2-13:20 10 /06 too s mall 29.5 BSA04 #2 peak of 2nd BSA run 14:30-21:30 10 /06 7.9 -14 .2 BSA04 #3 post-peak on 2nd BSA run 23-24:20 10 / 06 too small -27 Column washed with .1 N HN03 :iso (1 :4) for 40 minut es HPLC broke, unable to run Blank right away nothing run on column in between BSAC03 #I BSA blank 12-13 : 20 10 /31 too s mall -28.7 BSAC03 #2 BSA blank 14:30 -21: 30 10 /3 I too small -27.5 BSAC03 #3 BSA blank 23-24:20 I0/3 1 8.83 -27.3 BSA cent. Unfractionated BSA sample that was 10/ 06 rep 1 7.4 -11.4 I cen trifuged rep 2-7.5 -11.2 BSA uncent Unfractionated BSA sample that was not 10 / 06 rep 1 7.8 -11.2 centrifuged rep 2 lost lost BSA03 #2, BSA04 #2, and BSAC03 #2 collected into rotoevap ballasts rotoevaporated for 15-20 min at 5560 degrees until about I-2 ml s remained They were then tran sferred into quartz tubes. 3 ml s of triple distilled wate r was added to eac h ballast in 1 ml increments and then transferred to th e appropriate quartz tube to ensure that all the sample was recovered. All other samples were collected directly into quartz tube s, all quartz tubes were frozen and lyophilized Note: All samp l es run on HPLC are centrifuged prior to injection (Appendix 2. Continued on next page)

PAGE 104

\0 N Appendix 2. (Continued) ID Description Collection time Date run NIS cu MM1101 # 1 pre 1st peak on 1st MM run 30-1:30 11/12 too small 27 MM1101 #2 1 st peak on 1st MM run 2-5:30 11/12 2.231 -38.77 MM110l #3 post 1st peak and pre 2nd peak on 1st MM 7 -8 11112 too small -28.813 run MM1101 #4 2nd peak on 1st MM run 10:30-16 11112 5.046 -20.69 MM1101 #5 post 2nd peak on 1st MM run 18-19 11/12 le ak-gas lin e -23 99 MM1102 # 1 pre 1st peak on 2nd MM run 30-1:30 11112 too small -26.659 MM1102 #2 1st peak on 2nd MM run 2-5:30 11112 .31 -37 277 MM1102 #3 post 1st peak and pre 2nd peak on 2"0 MM 7-8 11112 too small -28.961 run i MM1102 #4 2nd peak on 2nd MM run 10:30-16 11112 no sample no gas I MM1102 # 5 post 2nd peak on 2nd MM run 18-19 11112 too small -26 .669 Column washe d with .IN HN03:iso ( 1 : 4) for 40 minutes MM11COO # 1 MM blank 30-1:30 11112 .947 -38.586 MM11COO #2 MM blank 2-5:30 11/ 12 7.68 -33 .828 MM11COO #3 MM blank 7-8 11112 27.645 MM11COO #4 MM blank 10:30-16 11/ 12 2.579 -34 122 MM11COO #5 MM blank 18-19 11112 -27.125 MMll unfrac Unfractionated MM samp le (this was 11112 1 .2 1 7 -38 586 centrifuged as were all other samp les MM1101 #2, MM1101#4, MM1102 #2, MM1102 #4, MMliCOO #2, and MM11C00#4 were collected in rotoevap ballasts a n d handled exactly as described above. All other samples collected directly into quartz tubes and procedure s mentioned ab ove were followed. (Appendix 2 Continued on nex t page)

PAGE 105

'-0 w Appendix 2. (Continued) ID Description Collection time Date run NIS cu MF202 #1 pre 1st peak on 1st MF run 30-1:30 11/ 24 too small -27.12 MF202 #2 1st peak on 1st MF run 2-5:30 11/ 24 sample broke MF202 #3 post 1st peak and pre 2nd peak on 1st MF 7-8 11/ 24 too s mall 28.96 run MF202 #4 2nd peak on 1st MF run 10-16 11/24 7.43 -28.37 MF202 #5 post 2nd peak on 1st MF run 18-19 11/ 24 too small -15.34 MF203 #1 pre 1st peak on 2nd MF run 30-1:30 11/24 too small -28.94 MF203 # 2 1 st peak on 2nd MF run 2-5:30 11124 .671 37.49 MF203 #3 post 1st peak and pre 2nd peak on 2nd MF 7-8 11124 too small -27. 54 run MF203 # 4 2nd peak on 2nd MF run 10-16 11/2 4 too s mall -26.68 MF203 #5 post 2nd peak on 2nd MF run 18-19 11/ 24 too small -26 .86 Column washed with .IN HN03: iso (1:4) for 40 minutes MF2COO #1 MF blank 30-1:30 11/24 too small 31.66 MF2COO # 2 MF blank 2 -5:30 11124 too small too s mall MF2COO #3 MF blank 7-8 11124 too small 25.31 MF2COO # 4 MF blank 10-16 11/ 24 sa mple leak 27.194 I MF2COO #5 MF blank 18-19 11/24 too s mall -25.02 MF-2 Unfractionated MF sample (centrifuged as 11/ 24 1.502 -40 .57 we re all other samples run on HPLC) --MF202 #2 MF202 #4, MF203 #2, MF203 #4, MF2COO #2, and MF2COO # 4 were collected in rotoevap ballasts and h and l ed exactly as described above. All other samples collected directly into quartz tubes and pr oce dure s m entio n ed above were followed (Appendix 2. Continued on next page)

PAGE 106

'-0 Appendix 2. (Continued) ID Description Collection time Date run NIS cu PM20l # I pre 1st peak on 1 s t PM run 30-1 :30 12/ 08 too small 27 38 PM201 #2 1st peak on 1 s t PM run 2:30 -5: 30 12/ 08 397 37 87 PM20l #3 post 1st peak a nd pre 2nd pea k on l st P M 7-8 12/ 08 3 -28 62 run PM201 #4 possible 2nd peak (?) on 1st PM run 11-16 1 2 / 08 11.8 74 -28 .58 PM201 # 5 post 2nd peak on 1 s t PM run 18-19 12 / 08 t oo s mall 24 3 1 PM202 #I pre 1st peak o n 2nd PM run 30-1:30 1 2 / 08 too small -26.54 PM2 02 #2 1st peak on 2nd PM run 2:30 5:30 12/ 08 2 1 6 -37 1 66 PM202 #3 post 1st peak and pre 2nd peak on 2nd PM 7 8 12/ 08 too small 29 52 run PM202 #4 possible 2nd peak (?) on 2nd PM run 11-16 12/ 08 9.445 28.62 PM202 #5 post 2nd peak o n 2nd PM run 18-19 12/ 08 too small -24 .31 Column washe d with .IN HN03:iso (1:4) for 40 m i nu tes PM2COO # 1 PM blank 30-1:30 12/ 08 NO PM2COO #2 PM blank 2 : 30-5 : 30 12/ 08 NO PM2COO #3 PM blank 7-8 1 2 / 08 N O PM2COO #4 PM bl ank 11-1 6 12/ 08 NO I PM2COO #5 PM blank 18-19 1 2 / 08 NO r::= unfractionated PM s ampl e (centrifuged as 1.25 were all other samp le s run on HPLC) PM201 #2 PM201 #4, PM202 #2, PM202 # 4 PM2COO #2 and PM2COO #4 were collected in rotoe va p balla s ts a nd h and l e d exactly a s described above. All o ther sa mple s collected directly int o quartz tubes a nd procedures mentioned above were foll owed (Ap pendix 2 Continued on next p age)

PAGE 107

\() Vl Append i x 2 (Con t i n ued) ID Description Co ll e ction tim e D a t e run N I S cu PF20 1 # 1 pr e 1 s t p eak on 1st PF run 30 -1: 30 02117 too small 2 6 62 PF201 #2 1 s t p eak on 1st PF run 2 : 30-5:30 02 / 17 1.125 -38.926 PF20 1 # 3 pos t 1 s t peak a n d pre 2nd p eak on 1 s t PF 7 8 02/17 t oo small -28 8 1 run PF20 1 #4 2 nd peak on 1 s t P F ru n 9-17 02 / 1 7 1 3.39 -31.76 PF201 #5 pos t 2 nd p eak on 1st PF ru n 18-19 02 / 1 7 too small 27. 1 9 9 Co l um n washe d w ith I N HN03:iso ( 1 :4 ) fo r 40 m i n u tes PF202 # 1 pre 1st peak on 2nd PF run 301 : 30 02 / 1 7 -0.681 38 507 I I PF202 # 2 1st peak on 2nd PF ru n 2: 30-5 : 30 02 /17 1.00 38 8 1 I PF20 2 # 3 post 1 st peak and pre 2nd peak on 2nd PF 7-8 02/17 too sm a l l -28 30 I I ru n PF20 2 #4 2nd peak on 2 n d P F run 9-17 02117 11.764 -27 7 PF20 2 # 5 pos t 2 nd p eak o n 2nd PF run 1 8-19 02/17 too sma ll 26 334 C o l umn washed wi th .IN HN03: iso ( 1 :4) for 40 m i nu t e s PF2BOO # 1 PF bl ank 30-1:3 0 02117 NO I PF2BOO #2 P F bl ank 2 : 30 5:30 02/17 ND I PF2BOO # 3 PF bl a nk 7 8 02/17 NO PF2B O O # 4 PF b l ank 91 7 02/17 NO PF 2 B O O # 5 PF bl a nk 18-19 02117 NO P F-2 unfractionated PF s amp l e ( centrifug e d as 2.03 4 2 5 7 were a ll o ther s a mpl e s run o n HPLC ) P F201 #2, PF 201 #4, P F202 # 2 PF 2 02 # 4 PF2BOO # 2 a nd PF 2BOO #4 we re c o ll ec ted i n r o to ev ap b allas t s a nd h andl e d e xac t l y as des cribed abo v e All o th e r s a mp l es c olle cted dire c t l y i nt o qu a rt z t u be s and proc e dur es m entio n e d a b ove w e r e f ollowe d *N ote: Vacuu m p ump on r o t oeva p wa s brok e n u s ed s ink v a c uum T h i s inc r e ase d tim e o f eva p o rati o n t o 40-4 5 minute s.

PAGE 108

Appendix 3. Amino Acid Abbreviations asp aspartic acid thr threonine ser-senne glu g lutamic acid pro proline gly-glycine ala alanine val v aline met methionin e il e isoleucine leu leucine t y r -t y rosine phe phen y lalanine hi s histidine lys-lysine arg-argmme 96

PAGE 109

\0 -.....J Appendix 4 All amino acid data in nanomoles Pl\1-101 Ill Pl\1101 112 Pl\1500 Ill Amino Acid Cone. in nmo l Cone. in nmol Cone in nmol asp 0 569497 0.2694 thr 0 5872 0 .129099 0 1 073 scr 1 0542 0.225068 0.4994 glu 0 8517 0.319589 0.3856 pro 0 505875 0 0651 gly 1.4058 0 .631136 1.9297 ala 0 .376 0 1 538 1 9 0 0 128 val 0 197029 0 0603 met ilc 0.151231 0 0138 leu 0 134 0 207861 0 0783 tyr 0 0165 phc 0 3968 0 212767 0 0268 his 7 6872 11.9691 lys 0 3312 0 326887 0 6033 arg 0 1213 56 0.0339 Total 12. 8241 3 .751 214 1 6 0713 (Appendix 4. Continued on nex t page) Pl\1500 112 P l\ 1500 113 Pl\1500 114 PM200 Ill Pl\1200 112 Cone in nmol Cone. in nmol Cone. in nmo l Cone. i n nmol Cone. innmol 0 0728 0 0169 0 791 0.2974 1 40. 97 7941 0 03 48. 760359 13.684425 0 0268 0 016 0.23 67 130727 15.200 874 0 1223 0 2785 60 205762 22 642832 0 085 35 585238 2.3386 0 1 23 6 0 77 127 .396211 30. 930588 0 0 1 67 0 6955 73 6 1 2141 20 .73595 I 0 022 74 66218 23 768 96 3 0 0152 0 017 5 0.0776 17 389258 2.44449 1 0 02 0 0 1 25 0 0194 48 664883 12. 624332 0 0136 0 3323 200 1 42922 1 6 42 50 1 6 23 616523 2.54 1 674 0 1686 0 .145 0 2084 58.663746 14.49735 5 0 2 1 77 0 .2 1 9 7 0 2548 9 .110392 3 .876907 0 5949 0 3118 0 4257 53 793359 20 .82 9002 0 0154 0 0 1 53 37. 277039 11.3 7967 3 6292 0 87 8 4 4 2135 900. 7229 1 2 288 145259

PAGE 110

1.0 00 Appendix 4 (Con tinued) I Pl\1102 # I Pl\1-102 #2 PM-1 02 #3 !Amino acid Cone in nmo l Co ne. in nmo l Cone i n nmo l as p 0 70 3 2 0833 0 5482 thr 0 .1193 0.5917 0 0597 scr 0 5409 1 6324 0.4804 glu 0 3445 1.1795 0.1347 pro 0.0793 g l y 1 3082 2.6253 1 0233 a la 0 3757 0.6328 0 328 \ 8 1 m e t 0 .0 1 47 ile 0.448 0 0529 leu 0 0884 0 7111 0 1 603 t y r phe 0.4292 0.44 1 9 0 3785 his 3 .911 2 1.8183 1.44 1 6 lys 0 1326 0.1 476 arg 0.0546 Tot a l 7 953 1 2.38 1 2 4 6869 -(Appendi x 4 C on tinu ed on next page) Pl\1102 #4 Pl\1300 # I Pl\1300 # 2 Pl\1300#3 Cone in nmol Co n e i n nmo l Cone in nmo l Co n e in nmo l 1.1569 0 1168 1 100084 3.225163 0.4 1 24 0 05605 0 36024 3 0.097976 0.4803 0.137843 1.2725 5 1 0.73477 0.6407 0 205977 0.608966 0.40445 0 .141 666 4 3 876 7 1 0 36096 1 0.9578 0.436224 4.4 5 8751 2 263143 0.3626 0 03 3 868 2 802726 0.12779 0 0559 1 3 1 .40940 3 0.115 48 5 0 0678 0 036338 0 1 2 5721 0.096425 0.088 1 0.058474 0.26087 0 226324 0 281999 0.3966 0.065443 0 098 1 72 0 .2211 0.726642 0 .0578 0.036782 0.29 4924 0.1 932 14 0.5708 0 1 962 2 7 0 065523 5.4 1 29 1.59 7 934 18. 0 70 2 22 8 009396 ---

PAGE 111

\0 \0 Appendix 4. (Continued) PF200 # I PF200 #2 PF200 # 3 Amino Acid Cone i n nmol Cone. in nmol Co ne in nmol a s x 0.203935 0 1 05178 0 064 1 53 thr 0.7 1 3846 0.056229 0 121336 ser 1 030924 0 1 51037 0 086697 gb: 0.67253 0.196126 0 1 7422 pro 4 58257 1 0 .53 9008 0 13773 g l y 4 458263 0.554366 1.196572 a l a 3.608472 0.047379 0.009 4 5 val 2 501663 0 .038282 0 042815 m et ile 0.25 1 64 0.035142 leu 0 36256 0.055238 0.04204 1 t yr phe 0.20058 0.028168 his 0 675257 lys 0 3 1 6551 0 028731 0 019973 arg 0 264885 Total 19. 843677 1 .834884 1.894987 (Appendix 4 Continu ed on n ex t page) PF200 #4 l'FIOO # I PFIOO #2 PFIOO #3 PFIOO #4 Cone in nmo l Co n e in nmol Co ne in nmol Cone. in nmo l Cone in nrnol 0.643251 0.382357 0 052158 0.160787 0 239123 0.22345 1 0 22 1 214 0 032657 0.1137 1 3 0 1 94089 0.324109 0 356593 0.057812 0.2979 1 3 0.454579 1.171399 0.498126 0.075775 0.367074 0.305055 0.57595 1 0 334629 0.057061 4.615205 0.281149 1.126406 0 835766 0.17151 2.138785 1.06 1 179 0.283 1 08 0 285446 0.01400 1 0.837019 0 1 9 1 645 0 .349764 0.295647 0.019555 0 5 1 0 1 49 0 187811 0.226045 0.1 5 1 231 0.055398 0 1 35025 0.430987 0.275405 0.05111 0.078309 0 245215 0.15539 0 .148516 0.016336 0.117363 0 239 1 66 0.388027 0 2 12089 0.10673 5 0.220712 0 02474 0.094042 0 150683 0 07699 0.082615 0 0 1 4592 0 06119 5.693586 4 327423 0.587307 9.65642 1 3 836195

PAGE 112

0 0 Appendix 4. (Continued) 1\ll\1-11 00 1\11\11100 1\11\1-1100 Ill 112 113 #4 Amino Co n e. in Con e. in Cone. i n Con e in acid nmol nmol nm o l nmol ASp 1 9.3026 2 1431 thr 1 4.4507 1.3594 16 .300 8 0 4717 ser 6 0928 52.9262 0 9381 g lu 21.5653 2 0136 26.8202 0.6779 pro 27. 1 976 3 0004 28 .306 0.5621 g 1 y 88. 7172 9.2103 64.469 1.1707 ala 16. 1684 1.6 15.862 0 5465 val 1 1.7584 0.6418 14.6837 met 1.2971 i l e 1 2 5273 1 .0656 14. 7443 0.2667 l eu 1 7 2701 1 .6 085 14. 8385 0.3525 t y r 0 6079 0 2261 phe 7 6513 0.4643 9 1975 0.4027 h is 3 2394 0.6086 2 782 1 0 2865 l ys 20 80 48 2.7859 22.4374 0 0188 arg 1 0 3968 0.6 481 11.4 3 1 6 0 0399 T o ta l 252.3552 5 0.4 019 296.3225 7.8772 (Append i x 4 Continued on next page) 1\11\11200 # I Con e in nmol 11. 7053 1 2908 4.2403 2 .1456 1.69 1 2 1 92 4 9 1 877 0 6267 0 8483 1.1338 0 6 4 85 3.4686 1.9021 0.431 53.9339 1\11\12100 1\11\12100 113 114 l\li\121o o 1 #2 # 3 # I 112 115 Co n e in Cone. i n Cone. in Cone in Cone in nmol Cone. in Co ne in nmo l nmol nmol nmol nmol nmol 2 .616 0.4796 2 934207 91.195906 55.433895 0 .52 9558 0 4052 0 0235 0.523072 88.392992 23.94038 1 0 29 1 9 1 .0848 0 158 5 1 .607143 0 56976 1 67 194742 13.5 7444 0.17 1 256 0.67 0.0841 1 .24685 0.319 4 52 1 84.657234 26.54985 0.27026 0 .1187 3 718426 0.253224 1 41.639313 39.075637 0.42624 1 .7 426 0.2299 4 .638758 2 134956 129 554484 32. 299227 0.464505 0 6568 0.852985 0 202062 95 471438 21.912373 0.242359 0.152834 0 .111 565 122.313781 25.025219 0.180482 19.7 82953 3 .5 75889 0 1 811 0.0361 0 110444 0 .0 805 1 2 60.76607 12. 60256 0 129303 0 .3257 0.042 1 0.250368 0 21139 76 053062 16. 522166 0 201201 35.349133 1.94269 0.7242 0 6596 0 115551 89.324852 17. 222729 0.234215 1.80 1 6 0.22 93 0.528039 1 0 2 4 8903 2 699469 0 09 36 0.6753 0.275483 0.179263 57.4 1 743 19.216924 0.346018 0.052 1 0 4723 2.618 13.904402 7.111 943 1 269.362293 311 .593449 3.487297 ----

PAGE 113

Appendix 4. (Cont inued) 1\11\12200 #I 1\11\12200 #2 1\11\12200 #3 1\11\12200 # 4 1\11\12200 #5 Ml\12200 #6 A m ino Acid Cone in nmol Cone in nmol Co n e in nmol Cone. in nmo l Co n e in nmol Co n e in nmol asp 3.1346 20.2354 94.4556 0.4849 19. 5 3 03 4 .361 3 thr 0.0976 0.7303 7 5446 0 .0436 4 06 1 2 0 6806 ser 1.1861 2.3175 16. 702 1 0 1 211 4 9332 1.2516 glu 0 6636 0 5 154 8.249 0.0841 3 6473 0 7896 pro 1.71 58 48 9 1 37 1 7 1825 2.503 g l y 9 7268 1 6 7954 84 5472 0.294 1 8 90 1 6 3.0738 ala 0.5189 1 0 1 696 65 116 7 0.1 933 17. 8042 2 2915 \'31 8.5775 3 8549 m e t 0 0773 0 2337 2.9953 1.4553 0.1605 ile 0.036 1 0.5006 7.5738 2 7366 0.4696 0 leu 0.1603 3.4686 1 7 0925 0.0507 5.5975 0.8632 tyr 1.389 19. 7576 4 .918 0.5665 phe 0 154 1 0 323 15. 7802 0.2145 0.2098 his 5 866 1 0.0645 1 953 0.2121 1 .4038 0.0966 lys 0.33 4.6584 1 7 676 0 382 6 284 1 4572 arg 0 0455 0.043 1 0 2014 4 .04 74 0 0214 Total 21.9969 63 8695 427 1 362 2.0803 1 1 6 3 5 78 18. 7962 (Appendix 4. Continued on next page)

PAGE 114

0 N Appendi x 4 (Continued) I\IF201 # I I\IF201 112 I\1F201 113 A min o Acid Cone. in nmol C o ne. in nmo l Co n e i n nmol a s p 0 307 thr 0.3348 0 1 853 ser 0 983 0.2709 glu 0 764 1 0.2349 p ro g l y 2 3288 0 2024 0.4945 ala 0 5 2 3 1 0 .2 44 \'a I met ile 0 0949 0 0497 leu 0 1 764 0 1651 tyr p h c 0 .896 6 0 .7 433 1 0263 his 3 5586 0.2218 0 1 685 lys 0 0629 0 03 1 7 arg 0 0408 To t a l 9 .7 6 4 1.1675 3 1 779 (Appendi x 4. Continued on next page) I\1F201 # 4 Co n e in nmo l 0 1 06 1 0.0618 0.0841 0. 1 7 1 5 0 0375 0 .73 1 0.206 1 0 .575 4 0.0234 1.9969 -1\IFIOO # I 1\IFIOO 112 1\I FIOOIIJ 1\I FIOO 114 1\IFIOO 115 1\IFIOO#G Cone. in nmo l Co n e in nmol Co n e in nmo l Cone in nmo l Cone innm o l Co n e. in nmol 0.244 1 0 050 1 0.055 8 0 444 0 3303 0 0647 0.0856 0.1 697 0.24 1 0 3013 0 0323 0 09 1 2 0.2883 0 5035 0 0826 0 4837 0.0889 0 .076 6 0 1 8 1 5 0.3323 0 6046 I 1.8911 0 1 534 0 2 1 72 0 3 1 29 0.6385 4 1 782 0 .9 4 24 0 .0386 0 3 1 3 1 0.1094 0 1073 0 0226 0 0345 0 0672 0.2398 0 0624 0.7184 0.6605 1 0874 0 8 594 0.9617 0 7095 1.944 1 0.1 048 0.1938 0 .1854 0.1 96 0 182 1 0 693 0 0144 0.5995 0.5757 0 .65 8 0 04 1 9 0 0 16 9 0 0 2 29 0.020 4 7.411 1.127 2.3729 3. 1 227 4 5465 5.9464 L__ __ -----------------

PAGE 115

0 VJ Appe ndi x 4 (Con tinued) 1\IFJOO # I 1\IFJOO # 2 A m i n o Acid Cone. i n nmo l Cone. in nmol a s p 1.6386 thr 0.6 439 ser 1.1731 g l u 1 .3936 pr o 0 4 2 49 g l y 3.3 4 87 ala 1.1167 v a l 0.4225 ... ,. m e t i l e 0.263 leu 0 .53 68 tyr phe 0.5998 0 4 5 6 3 h i s 5.4832 0 1588 lys 0.5385 0.503 1 a rg T o t a l 1 7 .5 833 1.1 1 82 I\IFJOO # 3 Cone. in nmol 0 6995 0 1 33 7 0.56 1.3932 --I\IFJOO # 4 I\I F 4 0 3 # I I\IF403 #2 I\I F403 # 3 I\I F403 # 4 1\IF-10 3 # 5 Cone. i n nmol Cone. in n n10l Co ne. i n n m o l Cone. in nn10l Cone. i n nmol Cone. in n m o l 0 .5 402 0.037605 0.58964 0.1 46343 0 .3303 0.816289 0.037882 0.084574 0.095606 0 .4235 1.0155 46 0 1 44 824 0.061829 0.302458 0.1 57423 0.5076 0.207389 0.1 24843 1.175786 0.238414 3.527084 0.6669 1 7 0 1 4338 1 0.448685 0.206 1 95 0.5864 4 581383 0.4 54576 0.9 1 7849 0.909632 0.380283 0.5024 1.08078 0.096136 0.078098 0.382094 0.148334 0 0895 0.299295 0.3 17981 0.007526 0.06 1 1 0.1 9 33 67 0 201668 0.040358 0 2457 0.3 1 7703 0.059823 0 0387 1 8 0.42888 0.080728 0 6 1 92 0.6 1 9797 0.071575 0.211 321 0.0 1 203 1 0 2296 0 .082 834 0.4541 0.273379 0.04 1 4 0.1 1 5 1 87 0.055 1 42 4.5896 1 3.05245 1 1.573 133 1.449292 5. 1 78938 1.472777 ---



PAGE 1

COMPARATIVE BIOGEOCHEMISTRY OF MODERN. FOSSIL. AND ARTIFICIALLY AGED MOLLUSCS USING PROTEIN AMINO ACID. STABLE ISOTOPIC AND ULTRASTRUCTURAL METHODS by SAMANTHA DEBRA ANDREWS A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology University of South Florida August 1998 Major Profe s sor: Lisa L. Robbins Ph.D

PAGE 2

Graduate School University of South Florida Tampa, Florida CERTIFICATE OF APPROVAL Master's Thesis This is to certify that the Master's Thesis of SAMANTHA DEBRA ANDREWS with a major in Geology has been approved by the Examining Committee on May 18, 1998 as satisfactory for the thesis requirement for the Master of Science degree Examining Committee: ( < I Lisa L. Robbins Ph.D. Membe/ Peter J. Harries, Ph.D. Member: Carl I. Ste/fel, Ph.D.

PAGE 3

ACKNOWLEDGEMENTS I would like to express sincere thanks to Dr. Lisa Robbins my major advisor and mentor for her unyielding support ( both financially and emotionally) inspiration. a nd g uidance on this project. I would also like to thank her for her grea t patience during HPLC emergencies and other technical difficulties. I also wish to thank Dr. Peter Harries and Dr. Carl Steefel for their valuable contributions as members of m y committee. Dr. Pegg y Ostrom Michigan State University, kindly ran carbon and nitrogen isotope analyses. I am also grateful to the Tampa Bay Fossil Club for their generous grant in aid of research. This work would not have been possible without any of them I'd like to thank Matthew Schmidt for everything, sociall y and academically Kristin Yale for years of memories both in Connecticut and Florida Bradley Raiche for computer usage and formatting support pa s t and present students of the Molecular Paleontology and Biomineralization Laboratory for k ee ping me sane in times of distres s and most importantly, all of the above for their friendship. Finally my family deserves great thanks for always showing enormous support, interest, and love. My parent s were always there through the trials and tribulations of this th esis.

PAGE 4

TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES ABSTRACT INTRODUCTION Purpose ofThis Study/Hypothesis Significance BACKGROUND Shell Ultrastructure Bivalves Mercenaria Gastropods Polinices Conditions that Enhance Shell Matrix Protein Preservation Protein Role in Biomineralization Degradation 813C /81 5N Previous Work Temperature Studie s METHODS Sample Collection and Preparation Permeability Heating Experiments SEM RPHPLC Anal y sis and Fraction Collection 81 3C / 815N ofRPHPLC Fractions Amino Acid Analysis RESULTS SEM Gastropods 111 IV Vll I 2 4 6 7 7 8 8 9 10 14 15 17 19 21 24 24 25 27 28 29 30 31 33 33 33

PAGE 5

Bivalves Comparison RPHPLC Amino Acid Analysis Stable Carbon and Nitrogen Isotopic Analysis Permeabilit y DISCUSSION Description of Proteins RPHPLC Amino Acid Analysis Stable Carbon and Nitrogen Isotopic Analysis Permeability Future Directions Conclusions LIST OF REFERENCES APPENDICES APPENDIX 1. APPENDIX2. APPENDIX 3. APPENDIX4. Sample weights, tests run and weight percent. Sample ID's description collection time data and other notes for isotope analysis. Amino Acid Abbreviations. All amino acid data in nanomoles. II 34 34 44 45 5 7 58 65 65 68 69 71 75 76 78 80 88 89 91 96 97

PAGE 6

Table 1. Table 2. Table 3. Table 4. Table 5. LIST OF TABLES Averages of isotope data for modern and fossil Polin ices and Mercenaria. Permeability results. Weight percent averages. Isotope results for samples decalcified with HCl. 813C values (in %o) for total shell organic hydrolyzate for Strombus and Polinices. iii 57 63 66 74 75

PAGE 7

LIST OF FIGURES Figure 1. Preserved organic composition of fossil molluscs. 6 Figure 2. The general structure of a bivalve shell. 7 Figure 3. The simple prismatic structure of a bivalve shell. 8 Figure 4. The nacreous layer of a bivalve shell. 8 Figure 5. The crossed lamellar structure of a gastropod shell. 9 Figure 6 The nacreous structure of a gastropod shell. 9 Figure 7. Schematic representation of a composite section of one individual matrix sheet bounded on both sides by mineral. 11 Figure 8. Summary of process of hydrolysis. 16 Figure 9. Alanine was heated to various temperatures to determine whether it could persist for long periods 22 Figure 10. Flow chart of major steps in sample preparation and analysis. 32 Figure 11. SEM Photomicrograph of PM, cross-sectional view 35 Figure 12. SEM Photomicrograph of PM, cross-sectional view. 35 Figure 13. SEM Photomicrograph of PM outer surface. 36 Figure 14. SEM Photomicrograph of PM, inside surface. 36 Figure 15. SEM Photomicrograph ofPM, inside surface. 37 Figure 16. SEM Photomicrograph of PM. 37 Figure 17. SEM Photomicrograph of MM, outer surface. 38 Figure 18. SEM Photomicro g raph of MM 38 iv

PAGE 8

Figure 19. SEM Photomicrograph of MM. 39 Figure 20. SEM Photomicrograph of MM closer magnification of cross-sectional view. 39 Figure 21. SEM Photomicrograph of MF, outside surface. 40 Figure 22. SEM Photomicrograph of MF extreme top of photo is insid e s urfac e rest of v iew i s a cross-section. 40 F i gure 23. SEM Photomicrograph of PF outside surface. 41 F i g ur e 24. SEM Photomicrograph ofPF, cross-sectional view. 41 Figure 25. SEM Photomicrograph ofPF, cross-sectional view. 42 Figure 26. SEM Photomicrograph ofPF, cross-sectional view. 42 Figure 27 SEM Photomicrograph of PF, inside s urface. 43 Figure 28. SEM Photomicrograph ofPF, inside surface. 43 Figure 29. Typical RPHPLC graph of soluble organic fraction from modern Mercenaria. 46 Figure 30. T y pical RPHPLC graph of so luble organic fraction from fossil Mercenaria. 46 Figure 31. RPHPLC g raph of soluble organic fraction from modern Mercenaria. 47 Figure 32. RPHPLC graph of soluble organic fraction from modem Mercenaria. 47 Figure 33. Typica l RPHPLC g raph of soluble organic fraction from modern Polinices. 48 Figure 34 Typical RPHPLC g raph of soluble organic fraction from fossil Polinices. 48 Figure 35. RPHPLC graph of so lubl e organic fraction from modern Polini ces 49 Figure 36 RPHPLC graph of soluble organic fraction from modern Polinices. 49 v

PAGE 9

Figure 37. Modem Mercenaria amino acid compositions. 51 Figure 38. Fossil J\.1ercenaria amino acid compositions. 51 Figure 39. Modem Polinic es amino acid compositions. )_ Figure 40. Fossil Polinices amino acid compositions. 52 Figure 41. Modem vs. fossil Mercenaria amino acid compositions. 53 Figure 42. Modern vs. fossil Polinices amino acid compositions 53 Figure 43. Heated modem Mercena r ia (MM-2.2 heated dry) amino acid compositions. 55 Figure 44. Heated modem Mercenaria (MM -2.1 heated wet) amino acid compositions 55 Figure 45 Heated modem Polinices (PM-5, heated dry) amino acid compositions. 56 Figure 46. Heated modem Polinices (PM-3 heated wet) amino acid compositions. 56 Figure 47. RPHPLC chromatogram of MM used for carbon and nitrogen isotope analysis 59 Figure 48. RPHPLC chromatogram of MF used for carbon and nitrogen isotope analysis 60 Figure 49. RPHPLC chromatogram of PM used for carbon and nitrogen isotope analysis. 61 Figure 50. RPHPLC chromatogram of PF used for carbon and nitrogen isotope analysis. 62 Figure 51. Comparison of permeabilities of modem and fossil bivalves and gastropods. 64 Figure 52. Correlation of sample weight and permeability. 77 vi

PAGE 10

COMPARATIVE BIOGEOCHEMISTRY OF MODERN, FOSSIL. AND ARTIFICIALLY AGED MOLLUSCS USING PROTEI N. AMINO ACID. STABLE ISOTOPIC AND ULTRASTRUCTURAL METHODS by SAMANTHA DEBRA ANDREWS An Abstract Of a thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Geology University of South Florida August 1998 Major Professor: Lisa L. Robbins Ph D VII

PAGE 11

ABSTRACT Data obtained from organic separations of ancient and modem Polinices duplicatu s and Mercenaria mercenaria indicate that the ethylenediaminetetraacetic acid (EDT A) soluble organic shell-matrix proteins are composed of heterogeneous assemblages of proteins and polypeptides. Samples from specimens of Polinices and Mercenaria illustrated overall similar reverse phase high performance liquid chromatography (RPHPLC) chromatograms with hydrophilic and hydrophobic fractions as well as overall similar amino acid compositions although in different concentrations of the major protein-containing fractions At the bulk protein level large isotopic differences were observed between hydrophobic and hydrophilic proteins of both modern and fossil shells indicating some diagenesis had occurred, and possibly reflecting differences in diet and metabolism Heated modem samples showed that some diagenetic alteration occurred during artificial aging as indicated by HPLC and amino acid analysis suggesting that temperature studies may be useful in modeling degradation. Ultrastructurally, many interesting similarities and differences were observed between specimens The present study was able to illustrate that the gastropod's ultrastructure is more intact than the bivalve because of its structurally intact closely packed crystallites, and its lack of original pores in the modern specimen. The bivalve on the other hand shows wider packages of crystals and has porosity within its shell structure The viii

PAGE 12

gastropod also showed evidence of a thicker nacreous la ye r when compared to th e bivalve although both predominantly consisted of either a crossed lamellar (gastropod) or prismatic (bivalve) structure. In both genera, fossil specimens demonstrated significant crystal alteration suggesting that ancient intracrystallin e organics ma y be adversely affected. Mercenaria demonstrated a slightly lower permeability than Po!inices. and fossil specimens showed a higher permeability than modern specimens. C haracteri za tion of shell proteins and their associated calcium carbonate will aid in studies of biomineralization and crystal formation paleoenvironmental conditions and diagenesis. Abstract Approved_:_ ."""" --r= -=.::;...-...,;..-=:;.__ .;......;;..____;;. ________ Major Professor : Lisa L. Robbins Ph. D. Associate Professor Department of Geolo gy i x

PAGE 13

INTRODUCTION Genetic (Paabo 1989) biochemical (Robbins and Brew 1990) immunological (Lowenstein 1980 ; Muyzer et al., 1988) and other geochemical methods such as stable isotope analyses (Ostrom et al. 1994) are some of the approaches which can be used to investigate organic matter in shell and bone. While genetic techniques based on DNA or RNA are considered more direct biogeochemical approaches based on organic matter encapsulated in the calcium carbonate of the shell have the potential for use on fossil as well as living species. Although the use of ancient DNA in molecular paleontology is becoming increasingly advanced (Herrmann and Hummel, 1994), "classical methods of biogeochemistry are still better developed for use on very ancient material. Only in exceptionally well preserved cases can significant amounts of DNA be retrieved. Molecular data from shell matrix proteins provides new possibilities for describing diagenetic pathways from modern to fossil shells. The preservation of amino acid sequences from shell glycoproteins can offer the unique opportunity to compare these fossils with their modem counterparts. Whereas in the past only shell structure was available to provide information on evolutionary trends today fossil shell proteins may contribute to our understanding of molecular evolution. Because of the enormous amount of information in an amino acid sequence proteins give substantially more phylogenetic information than can be obtained from morphology alone (Robbins et al., 1993b). By

PAGE 14

comparin g the structure of modem and fossil molluscs with their protein and amino acid si g natur es. evolutionary questions may be addres s ed It ma y be possible to compar e protein s at v arious stages of their evolution; and because shell protein s pla y a major rol e in s hell morphology, it also ma y be possible to compare the evolution of shell morphology with that of shell proteins (Weiner et al., 1976). Because gastropod s h ave a more complex st ructure than bivalves (Nakahara et al. 1982 ; Andrew s et al.. 1985). these difference s in morpholog y s hould be reflected b y their protein signatures. By isolating and characterizing fossil shell protein matter the steps in how individual proteins degrade over geologic time can be better understood (Robbins et al. 1993b). Owing to the complexity of diagenetic reactions, the mechanisms involved in the diagene sis of proteins have been largely speculative To accurately interpret the results of paleo biochemical investigations a better comprehension of the dia ge netic his tor y of prot e in aceo u s material in carbonate shells i s esse ntial. A thorou g h understandin g of the various components of the shells will provide insight into the preservational pot e ntial of the sys tem" as a whole. Therefore I initiated a multipronged investigation to analy ze s tructural and biochemical components of s hell s. Purpose of This Study/Hypothesis Shells of a modem and fossil gastropod Polinices duplicatus and bivalve Merc e naria m erce naria were analyzed and compared for ultrastructural differences s tructural permeabilit y, protein amino acid, and stable carbon and nitro gen i so topic compo s ition s to la y th e foundation for under s tanding dia g enetic r es i s tance and pro te in indi gene it y in f o ssils. Th ese t wo clas ses of mollusc s provide useful comparison s 2

PAGE 15

because gastropods have been shown to have a mechanically stronger structure than bivalves (Andrews et al., 1985), yet there are few studies which analyze various components (structural and biochemical) to evaluate pathways of degradation. A subsample of each shell was measured for permeability to help evaluate the potential transport of molecule s through the shell. The hypothesis that shells with higher permeability have a higher chance for in situ degradation was tested Although such a hypothesis is simplistic there are no previous studies to document this trend. Furthermore, the amount of organic material and extent of permeability were compared to determine the effect on the preservation of the proteins and their potential indigeneity Additionally samples were heated to simulate diagenetic alteration and elucidate pathways of diagenesis. Isotopic analysis of individual proteins and polypeptides has been suggested as a way to provide an additional criterion in assessing indigeneity (Ostrom et al. 1994). As tracers of protein degradation o13C and o1 5N were measured on individual fractions of modem and fossil material. Finally scanning electron microscopy (SEM) was used to document shell ultrastructural differences between: 1) the classes, and 2) a -2.5 million year old fossil and a modem shell. The goals for this study are as follows: 1) to characterize the soluble matrix proteins of these two types of molluscs; 2) to test the hypothesis that greater shell permeability will result in altered organic matter; 3

PAGE 16

3) to determine if the soluble matrix proteins and their amino acid s are sufficiently distinct as to justify clas s -level distinction in the se two species ; and 4) to analyze the o13C and o1; N of bulk proteins to document difference s between classe s of proteins and to determine the extent of diagenesis. Significance Many researchers have assumed that increased shell permeability will show a decrease in organic material (Andrews et al. 1985 ; Curry, 1987b, 1988), but no baseline data exist s to support thi s assumption. For example Andrews et al. (1985) noted the impermeable nature ofthe molluscan shell matrix both to losses of primary amino acids and incorporation of secondary amino acids. Permeability data sets may provide ways of comparing intergeneric dia ge nesi s The a mino acid compos ition of single proteins from foss il shell matrice s can pro v ide information on the genetic relatednes s (Robbins and HealyWilliam s, 1991 ). Proteins reflect the genetic makeup of an organism because the amino acid sequence is encoded b y th e DNA (Robbins and HealyWilliams 1991 ). Becau s e the amino acid signature i s so di s tinct s m all fragments may be assigned to a genus and possibly even a spec ie s, ba se d on amino acid chromatography (Andrews et al. 1985) Phylogenetic tree s have be
PAGE 17

to geochronology biomineralization and diagenesi s (King 1977) Evolutionar y lineage s can now be traced by studying molecular and biochemical as well as morphological. variations through geologic time (King and Hare 1972a) I s otopic analysis of bulk organic matter can provide an under s tanding of paleoecolo g ical, taxonomic and most importantly diagenetic relationships. In addition to as s essing i s otopic variabilit y among different organic fractions from modem organism s, isotope studies can be used for assessing the indigeneity of ancient organic fractions that have experienced alteration (Ostrom eta!., 1994). The ability to document geochemical changes in individual proteins as a function of time will provide insight into mechanism s that occur during diagenesis. 5

PAGE 18

BACKGROUND Shell proteins have the potential to show exceptional preservation in ultrastructurally well-preserved specimens as far back as the Jurassic (Figure 1. Robbin s et a!., 1993 b). B y combining the study of shell u l trastructure and organic matrix preservation, a better understanding of protein and i n organic diagenesis can be reached (Hudson, 1967). A 500 E 400 0> Qj 0. E 300 E "' '0 i3 200 ., 0 c 100 E <: 0 B 500 400 E 300 0> a; 0. E 200 100 0 q \ a. 'q -o- Amino acids 0 o surviving P leist Plio Mio Olig Cret Jur Increasing Geological Age P leist Ill lnsolubles 0 Soluble peplides Free am1no acids Plio Mio Otig Cret Jur Increasing Geological Age 30 25 'U m m 20 "' c 15 < :;:: !E.. 10 :s 3 n; 5 0 0 Figure 1 Preserved organic composition of fossil molluscs. A) Time-dependent decrease of amino acids from fossil molluscs. B) Relative proportions of free amino acids, so luble peptides and insoluble res idues in P l e i stocene P l iocene Miocene Oligocene, Cretaceous, and Jurassic molluscs (After Curry, 1988). 6

PAGE 19

Shell Ultrastructure Bivalve s In genera l the s hell ultrastructure of bivalves consists of an inner prismati c layer overlain by a nacreou s layer. w hich i s s ub seque ntl y overlain b y an outer prismatic la yer (Carter 1990 ; Figure 2). Th e inner layer consists of a homogeneou s structure (crossed acicular and /or a fine complex crossed lamellar st ructure). The middle la yer is homogeneou s or crossed lamellar. The outer layer is crossed lamellar or composite prismatic s tructure. The s imple prismatic layer (Figure 3) contains parallel prisms oriented vertically with the long axes perpendicular to the shell surface. These could be either aragonite or calcite The aragonitic pri s matic crystals are elongate with a fanlike form. The nacreous layer (Figure 4) contains lamellar or sheet like structures that form one layer at a time and it i s a lw ays com po sed of aragonite (Carter 1990) I I; I i .II: I i I !OUTER PRISMATIC LAYER 'I ,l i NACRE ... ,.\ ,\\.t t'l: ,'I I J_t L L r =--"_.........-'' PRISMATIC LAYER / ,. I'; t F i g ure 2. T h e ge neral structure of a bivalve shell (After Lowenstam and Weiner 1989). 7

PAGE 20

Figure 3. The s imple pri s m atic structure of a bivalve shell (After Carter and Clark, 1985) Mercenaria Sheet F i gu re 4. The nacre o u s la yer of a bivalve shell (After Carter and Clark, 1985) Mercenaria (superfamily Veneracea) co nt a in s an outer and inner aragonitic pri s matic st ru cture bounding a l am inar nacreous l a y er (Carter 1 990) T h e middl e shell laye r consist s o nl y of a homogeneous struct ur e ear l y in g rowth and a crossed lam ellar s tructure i s sec re te d temporarily in later growt h s ta ges near the boundary between th e middl e and outer shell layer s (Carter 1990) The crossed lamellar structure doe s not form a di stinc t la ye r. This c h ange from hom ogeneo u s t o crossed lam ellar s tructu re i s a grad u a l change. Gastropods Gast r opo d ultr as tru c ture s include a crosse d lamellar s tructur e (which i s homologou s t o t h e pr i s matic structure ofth e bivalve ; Figure 5) composed oftwo differen t alternat in g layers of lam ellae (Carte r 1 990) whic h are mineralogically di sti nct. T h e first-8

PAGE 21

order lamellae are rod-shaped and oriented perpendicular to the s hell surface. The second-order lamellae are differentiated by dip angle away from the s hell margins. Higher dipping angles are composed of aragonite ; whereas lower dipping angles are calcite. The structure of the nacreous layer (Figure 6) is composed of stacks of columnar structures that form vertically rather than laterally as it does in bivalves. Columns form a stacking pattern first and then join laterally into tablet forms (Carter 1990). Simple crossed lamellar Figure 5. The crossed lam ellar structure of a gastropod shell (After Carter and C lark 1985). Polin ices C. Columnar F i gure 6. The nacreous structure of a gastropod shell (After Carter and Clark 1985). Polinices (family Naticidae) has a subovate-to egg shape, and a low elevated spire of about four whorls (Olsson et al., 1953). The shell is composed of aragonite (Robbins and Ostrom, 1995) and is non-nacreous. It is necessary to have an understanding of shell structure because even if two different species have the same mineralogy different crysta l structures can subject the 9

PAGE 22

organ i c matter to different diagenetic resistances. different amounts of crystal di sso luti ons and th e refore contamination ( Robbin s and Ostrom. 1 995). Conditions that Enhance Shell Matrix Protein Preservation Numerous factors includin g she ll material are known to increase the preservation of s hell matrix proteins. Fo r example if the protein is completely enclosed i n the calcite l att ic e dissolution and hydroly s is will b e limited Proteins in shell are naturall y protected by the close relationship of the protein with the mineral matrix retarding decay (Ambler and Daniel 1991 ) The crys tals comprising the shell provide an intracrystalline micro environment where organic material i s protected from contamination (Curry 1987 a) Onl y if the s hell is broken or recrystalli z ed will th e proteins be exposed to water and un dergo h y d ro l ys i s (Robbins eta!., 1 993b). The protein structure and organic materia l can a l so play a role in preservation of the protein Factors such as the primary seco ndary and tertiary structure of the protein w ill affect the pathway of its degradation (Robbins et al. 1993a ). The term primary structure refers to the fixed amino aci d se quence ofthe poly peptide chain (o r chains) m aki n g up the backbone of the molecule. The seco ndary structure i s based on the helical co iling of proteins stabilized by hydrogen bond s Stabilization of the helical coiling require s the presence of di s ulfide brid ges or the tertiary structure (Anfinsen 1959 ). The t e rtiary s tructure of the protein i s likely t o pla y a major role in determining th e res istanc e to modific a tion dur ing fossiliz ation and diagenesis (Eg lin ton and Logan 1991 ). 10

PAGE 23

The location of organics within the shell structure is important, and may play a role in the preservation of these proteins These locations vary between taxa. but u suall y the organics are located between and within the carbonate crystals of the shell. forming layered structures (Robbins eta!.. 1993b). Most molluscs have la yered organic and inorganic structures with intraand extra-crystalline organics (Crenshaw, 1 972: Wheeler eta!., 1988). The majority of crystals are enclosed in an organic matrix (usually comprised of glycoproteins) (We iner and Hood, 1975; Weiner and Traub, 1980 1984 ; Figure 7). This organic matrix forms prior to mineralization, and it is involved in the formation of the shell (Weiner and Hood, 1975; Weiner and Traub, 1980). After most of the shell formation is complete the entombed organic material will contribute to the mechanical properties of the shell (Weiner and Traub, 1984) -D m ineral acid macromolecules D silk-fibroin-like proteins Figure 7 Schematic representation of a composite section of one individual organic matrix sheet bounded on both sides by mineral. The total thickness of individual matrix sheets may vary between 30 nm and 300 nm (After Weiner and Traub, 1984). 11

PAGE 24

The organic matrix of a molluscan shell appears to be very complex with re s p ec t to the number and compo s ition of components within a single species, between species. and even between different crystal layer s of a s in g le bivalve shell ( Wilbur and Manya k. 1984). An intimate relation s hip exi s ts between the cry s tals and thi s organic matri x (Watabe 1965). The majority of skeletal organic molecules are intercrystalline and occur between individual crystal s in the skeleton (Curry 1988). Within th e skeletal elem e nt s trapped bet wee n the cry s tal are intracrystalline protein s, which make up 0 .01 to 0 1 weight percent of the mineral (Addadi and Weiner 1989; Albeck et al., 1993 ) In fossil shells, the concentration of the organic matter initially decreases rapidly as the intercrystalline material is lost. The slowly decomposing intracrystalline organic matter preserved within a protected micro-environment within the mineralized phase shows a s lo wer rate of decline an d th ere fore ha s a greater s urviv a l pot e ntial (C urr y. 1988) The organic matri x consists mainl y of so luble and insoluble protein fraction s. The in so luble matrix (IM) and soluble matri x (SM) are distinct entitie s. The IM i s the structural framework of the matrix and the SM interacts with the crystal (Wheeler et al. 1988 ) The SM is thought to b e intracrystalline, while the IM is intercry s talline (Wilbur a nd Man yak 1984). The so luble protein of the organic matri x o f mollusc s consi s t s of sulfated g lycoprotein s rich in a cidic amino acids (Weiner and Traub 1984) and i s therefore usuall y composed of repeating sequences of aspartic acid separated by either g lycine or se rine (a lternatin g aspartic acid residues). The regularl y spaced negatively ch a r ge d a s p a rtic acid could function as a templ a t e on which the minerali za tion occurs. T hi s i s acco mpli s h e d by binding the po s itiv e l y char ge d calcium t o the ne g ativel y charged aspa rti c acid. 1 2

PAGE 25

Partially degraded organics may be preserved as coherent proteinaceous sheets or they can exist as free amino acids derived from the breakdown of the original protein (Hudson. 1967) An organic matrix may be observed even in recrystallized shells. but they will not necessarily retain their original organic composition Furthermore. porou s sa mple s in warm oxidizing environments are poor preservers of organic matter (Hudson. 1967). Fossils with porous skeletons are more susceptible to contamination b y percolating groundwaters than those with dense tightly packed ultrastructures (Curry 1987b 1988). In addition microbial activity will be limited as porosity and permeability tend towards low values (Eglin ton and Logan, 1991 ). Organic matter contained within nacreous and foliate structures is present in thick. continuous sheets; they are more exposed to attack by hydrolyzing or oxidizing solutions (Hudson, 1967) In contrast, intricate ultrastmctures containing very thin organic sheaths of the complex lamellar group are less likely to be attacked (Hudson, 1967). Gastropods. consisting of the complex crossed lamellar structure would be less prone to degradation compared to the prismatic structure found in bivalves. However well-protected macromolecular proteins from most microarchitectures have been shown to escape destruction and part of their original structure can be preserved over long periods. The deposit i ona l environment also plays an important part in shell protein preservation The rapidity of burial and dewatering of the sediment can determine the extent of preservation At a surface environment of 1 0-20 C, hydrolysis is ten times more rapid than in deep-sea sediments where 90% of the peptide bonds are hydrolyzed by 1 Ma at 1-2C (Bada, 1991) It has been suggested that ifthe shells are deposited in an anaerobic environment degradation will be slower and more protein will be preserved 13

PAGE 26

(Robbins et al. 1993b ) However. the deca y proce ss can still be active despite anoxia Ammonium and bicarbonate ions liberated by the decay of animal protein can affect th e local pH/ Eh conditions promoting an alkaline reducing environment (Allison. 1988) Rapid burial and anoxia will reduce the rate of decay but it will not s top it. Protein Role in Biomineralization Formation of mineral s by organisms has increasingl y modified the chemical and physical nature of the biosphere. Biomineralization has radically changed the chemistry of the oceans and has contributed significantly to the sedimentary environment (Lowenstam 1981 ). Analys i s of the organic shell-matrix proteins may allow for insight into the specific proteins that control biomineralization of these molluscs During "or ga nic m atrix -medi ated" s hell formation (Lowen s tam and Weiner 1989). the organism l ays down o r ganic lay er s on w h ic h c alcite crystals s ub se quentl y nucleate. It is th e protein molecule s co mprisin g th ese organic layers that are generally believed to control biomineralization. T h e two broad cla ss e s of proteins found in the mollu scs, a h y drophilic fraction and a h y drophobic fraction are s imilarl y found in benthic and plankti c foraminifera ( Robbin s a nd Brew 1990) as well a s echinoid teeth and tests (Weiner 1986). Weiner and Erez ( 1984) proposed that these two classes of protein s perform different functions durin g te s t fo rmation and that because of their acidic nature are capable of bindin g calcium i ons and are prob ably invo lved in crys tal formation The common amino acid sequence in which a s partic acid i s se parated b y a s in g le amino acid co uld constitute part of thi s bindin g s ite (Weiner and Traub 1984) Two theories of the matrices function 14

PAGE 27

during mineralization are: I) a template which controls crystal growth: and 2) a compartment in which the crystals grow (Weiner and Hood 1975). The matrix .. hypothesis assumes that the organic matrix is deposited as a substrate that nucleates crystals ; the "compartment hypothesis supposes that crystal nucleation occurs within pre existing empty compartments of the organ i c matrix (Crenshaw. 1 980: Lowenstam and Weiner 1989) In addition, the matrix can re gulate the growth of the biomineral in several ways : by initiation of crystal growth determination of the crystal polymorph control of the crystal shape or terminat i on of crystal growth (Wheeler et al., 1 988 ; Albeck et al. 1996). Using electron diffraction, Weiner et al. (1983) demonstrated that gastropods and bivalves nacreous layers have the same spatial relationship between organ i c matrix constituents and aragonite crystallographic axes strongly supporting the notion that aragonite crystals form upon an organic matrix template Degradation Although excellent preservation of organic matter has been observed in rare ancient specimens, proteins will degrade over geologic time Degradation of proteins can take many pathways ; hydrol ysis is one of the most important of these degradation processes (Mitterer, 1993; F i g ur e 8), involving the decomposition of organic compounds in the presence of water. The process can be catalyzed by some metals and by both acids and bases however hydrolysis is accelerated at ext r eme pH values and occurs more rapidly under alkaline condit i ons (Kahne and Still 1988). Hydrolysis breaks the peptide bonds and initially converts the original proteins into a mixture of l arge and small peptides and free amino acids and ultimately to a mixture of only free amino acid s 15

PAGE 28

(Figure 8) A s the process proceeds smaller. more readily leachable molecule s are p ro duced Free amino acids are the smallest and most easily leached amino compound s in foss ils (M itterer 1993) Therefore hydrolysis and the subsequent leaching lead to a los s of free amino acids and a decrease in concentration of the amino acids in the fossil (Mitterer, 1993 ). I ORGANIC DIAGENESIS I Hydrolysis Protein -----. Peptides -----. Free Amino Acids Figure 8 Summary of process of hydrolysis. Water s eeps into the shell from the surrounding sedi ment and shell constituents are leached or modified by the added water. Due to leaching and diffusion the concentration of organic constituents decreases with increasing age of the fossil (Modified after Mitterer, 1993 ). The rate of h y droly sis depends on many factors including: 1) the length of the peptide chains and 2) the nature of the adjacent amino acids. The peptide bonds in larger molecules are more stable than the shorter chains The stability of the peptide bond is also largely a function of the amino acid residues that form the bond (A mbler and Daniel 1991 ) and since the protein composition and amino acid sequences are genetically determined the proteins in organisms of different taxon will have different rates of hydrolysis (Mitterer 1993) The rate of hydrolysis also depends on the amount of water available, the temperature therefore burial history and finally the presence of other biomolecules in the shell such as carbohydrates and metal ions (Vallen t y ne 1964) Despite thi s important de gra dation factor specimens may not be contaminated because 16

PAGE 29

similar amino acids have been found in fossils of the same species in different locations (Abelson 1956). However a possible way to trace the hydrolytic degradation reaction in fossils is through stable isotopic analysis. 813C/815N The diverse isotopic compositions of individual organic molecules in living organisms reflect the complexity of chemical reactions occurring during the growth and metabolism of a cell (Macko et al. 1983) The overall fractionation of stable isotopes between a specific molecule and its ultimate precursors is controlled by a number of factors including key rate-determining steps and major branching points in the flow of carbon and nitrogen (Macko et al., 1983) One of the earliest studies of this type was carried out by Macko et al. (1983) on algae in order to understand the basic fundamentals of isotope biochemistry They found that because amino-nitrogen is transferred from one am i no aci d to another the distr i bution of 1 5N is dependent upon the branching ratios of the components and the number of reactions in which an amino acid is involved The carbon isotope distributions in amino acids are similarly dependent on branching ratios biosynthetic pathways, and inhibition controls. Even from preliminary studies, they concluded that isotopic analysis of both carbon and nitrogen render a valuable tool for tracing specific compounds through biosynthesis and early diagenesis and ultimately into the fos sil record (Macko et al. 1983). Because amino acid constituents of organisms have distinct stable carbon isotopic compositions (Serban et al. 1988) the 81 3C of an amino acid allows evaluation of the 17

PAGE 30

origin and dia ge netic path ways of organic constituents in geologic materials ( Robbin s and Ostrom. 1995) Two main factors control th e o1'C of an amino acid: 1 ) the ty pe of carbon (fo r example, light 12C or heavy 13C) available when the s hell i s created. and 2) the degree of i so tope fractionation at major branchin g points in th e metabolitic pathway. In past s tudi es on molluscs, a lower o13C value wa s seen in fossil amino acids then in modern amino acids (Robbins a nd O s trom 199 5). This lower va lu e could be a result of the reintroduction of i sotop ic ally depleted degradation products. This i s consistent with the lo ss of 1 2C in the amino acid during diagenesis (Robbins and Ostrom 1995) The degree of preservation or degradation of the protein can be measured by the C:N ratio Since this ratio varies in modem shells, it is necessary to compare fos s il s with modern shells of comparab l e s tructure. Fossil s have very low nitro g en. therefore fossil s s how hi g h e r C :N rati os tha n their mod e m co tmt e rpart s (Hudson 1967). Fo ss il s with nacre-prismatic s tructur es show lower C:N ratios than those of the complex-lamellar or foliate g roup s (Hudson, 1967). The res ults e x pressed as C:N ratios do not reveal whether the protein is uniformly d egra ded or whether some of it remain s intact while the rest has deco mp ose d. In recry s talli zed fossils organic matrix remnants h ave no detectabl e nitrogen. It is also pos s ible to remove a large part of the organic carbon and nitro g en from the s hell while l eaving relic s of the original protein in a relatively unaltered condition (Hudson, 1967). 1 8

PAGE 31

Previous Work The first relatively detailed analysis of the biochemistry of the organic matrix of the mollusc shell was completed by Abelson (1954. 1955 ) In this pioneering work. he u sed a Mercenaria shell to examine the amino acid content of the protein-bound. so luble protein/peptide and free amino acid fractions. With increasing age. the protein-bound amino acids decreased ; the soluble protein/peptide amino acids increased from the Recent Pleistocene however disappeared b y the Miocene ; and the free amino acids were greater in older material than in recent specimens (Abelson, 1955) This suggests a degradation cycle from protein-bound to free amino acids through time The conclusion Abelson drew from this data was that initially most of the protein of the shell was present in insoluble layers. This stability was maintained for thousands of years with only moderate changes occurring which did not affect the protein solubility. Once some of the peptide bond s were broken the fragments were more soluble and could be leached from the shell. When all of the bonds were broken, only free amino acids remained ; these could be stable for millions of years The most likely degradation process that Abelson observed was hydrolysis Abelson (1955) and King and Hare (1972b) also showed that thermally unstable amino acids were absent or only present in small amounts in older material and consequently more thermally stable amino acids are still present in older specimens From thi s the y concluded that the amino acids were preserved indigenous organic matter secrete d by the living organism rather than contaminants. An increase in certain amino acids such as alanine in older material has also been noted and is probably due to the generation of this amino acid from the diagenetic breakdown of serine (Vallentyne. 19

PAGE 32

1964). Hare and Abelson (1964. 1965) found that proteins of species closel y related in morphology or shell s tructure have similar amino acid compositions. King and Hare ( 1972a 1972b) expanded on this using foraminifera Even though a similar morphology should show a generally similar amino acid signature a difference can occur in the more sensitive (variable) a mino acids, such as aspartic acid and glycine. The y also found the amino acid compo si tion to be a direct expression of the genetic system and were even abl e to s how differing amino acid compositions from two morphologicall y distinct fossil species ofthe same genus (King and Hare 1972a) Throughout the years, Mercenaria has been extensively studied by many researchers (e.g. Abelson 1954 1955 1 956 ; Hare and Abelson 1964 1965 ; Degens et al., 1967 ; Hare 1969 ; Vallentyne 1969; Akiyama 1971; Crenshaw, 1972 ; Weiner and Ho o d 1975 ; Hare and Hoering 1977; Kennish 1980 ; Andrews et al., 1985 ; Serban et al.. 1988 ; Muyzer et al., 1988 ; Engel et al. 1994 ; Silfer et al., 1994 ; Robbins and Ostrom 1995). Wheeler et al. (1988) showed that numbers and molecular weights of matrix fractions can range from many, low molecular weight components from gastropod shells to a single, hi g h molecular weight glycoprotein from M m erce na r ia While studying the s oluble matri x of aM. m erce naria Crenshaw ( 1972) determined it to be a highly sulfated glycoprotein. About 80% of the weight was protein The amino acids were determined to be 30% aspartic acid, 16% glycine, and 10% serine. Other workers have also shown aspartic acid, glycine and serine to be the most abundant acids in molluscs (Weiner and Hood. 1975 -bivalves ; Weiner et al. 1979-cephalopod s) a nd more specificall y Mercenaria (Hare. 1969 ; Weiner and Hood 1975 ) as well as other mineralizing systems r e pre se ntin g both CaCO and CaP04 biominerals (Robbins and Donachy 1991) 20

PAGE 33

While Polinices has not been as widely studied several investigations do exist (Degens et al. 1967 ; Ghiselin et al. 1967 ; Ostrom et al. 1994; Robbins and Ostrom. 1995) Previous experiments performed on Polinices showed that the protein solution was entirely soluble in water (no IM found). Two major protein fractions within the shell matrix have been identified. They contain individual amino acids with distinct 813C values (Robbins and Ostrom, 1995). Temperature Studies Numerous heating experiments have been performed on molluscs (Abelson. 1956; Hare 1969 ; Totten et al. 1972; Andrews et al. 1979 ; Qian et al., 1995) to observe degradation of organic compounds. In one experiment shell fragments of Mercenaria mercenaria were heated at high temperatures for a different number of hours to represent diagenesis A s expected the immunological reactivity of the Mercenaria proteins decreased with increased heating (Muyzer eta!. 1988) which is analogous to reactivity decreasing with sample age. However, significant positive reactions were still found in the shell fragments even after they were incubated for 80 hours at 140C. Abelson (1956) demonstrated that the rate of degradation of alanine in short laboratory experiments at elevated temperatures can be used to model degradation in fossils that are several million years old. It is possible to draw a curve extending back to predict pathways of degradation at low temperatures (Figure 9). For instance when heated at 450C it takes one second for the concentration of alanine in water to decrease by 63% At 188C, it would take one 21

PAGE 34

month to decrease by the same amount (Abelson 1956). At room temperature. alanine could theoretically last for billions of years. A lth ough alanine this old does not exist. this large extrapolation to room temperature is justified since alanine is observed in ancient fossils (Abelson 1954 ) The most stable amino acids are alanine glutamic acid gly cine. isoleucine proline, and valine (Abelson 1956). Less stable amino acids are arginine. aspartic acid lysine phenylalanine. serine threonine and tyrosine (Abelson 1956). with the last three being particularly unstable at elevated temperatures and in geologic environments (Nagy et al., 1981 ). "' "' 10"'YEARS I'-"' "' 1 YEAR 1'\. 10 DAYS 10'SECONDS """ 1 0:: SECONDS 1SECOND "' '\. SECONDS 20 40 60 80 100 150 200 250 300 400 500 TEMPERATURE (DEGREES C ) Figure 9 Alanine was heated to various temperatures to determine whether it could persist for long periods. The horizontal axis of the small circles indicates the temperature to which each sample of alanine was heated. The vertical axis indicates the length of time it took for 63% of the alanine to break down If a straight line is drawn through the circles, it is shown that at a temperature of 20 C nearl y half of the alanine would remain after 3 billion years This may also be true of other amino acids (After Abelson 1956) 22

PAGE 35

Totten eta!. (1972 ) heated powdered oyster shells at normal atmospheric pressure 130C for two months The aspartic acid and g l yc in e in the protein fraction declined b y 99% i n t he firs t week and 50-80% of the o ther amino acids were lo st. Some a min o acids tran s form e d int o ot h ers when heated and th e amo unt s of some amino acids increased by heating illu s tratin g tha t different amino acids can h ave different diagenetic s tabilities More than h alf of the total amino acid content of the she ll was l ost durin g the entire course of heating The amount of stable proteinaceous material in the heated samples paralleled th a t found in fossils Therefore, the quick protein decomposition b y heat can be shown to roughl y correspond to that by slow aging. 23

PAGE 36

METHODS Sample Co llection and Preparation Fossil Mercenaria mercenaria (MF) and Polinices duplicatus (PF) were collected from the Pinecrest Shell Beds in Sarasota, Florida (3.5-2.0 Ma). The investigation of fossils collected at one locality is preferable because of the fossils exposure to identical thermal histories and percolating groundwaters (Abelson 1954). In addition these shell beds consist of uncemented shell hash eliminating the problem of diagenetic processes that occur with cementation Samples were chosen based on as pristine a shell as possible and only clean fossils separated from the sediment matrix were used Living M mercenaria (MM) were collected from the east side ofBurgess Bay near Cape Coral Florida (MM1 and MM-2) and Fort Desoto Florida (MM -3). Modem P. duplicatus (PM) were collected from Essex, Massachusetts. Modem samples were stored at -70C until processed. The entire gastropod shell and sing le va lve s of the bivalves constituted a sample The shell samples were scrubbed using a brush and rinsed in triple distilled water (3xDH20). Samples were then soaked in a concentrated bleach solution to remove any remaining sediment and contaminants, thoroughly rinsed with 3xDH20 and air dried prior to being cru s hed by mortar and pest l e and subsequently ground to a powder. Samples 24

PAGE 37

were decalcified in I 0% (w/ v) ethylenediaminetetraacetic acid (EDTA, pH 8) containing 0 .1% sodium azide at 4 C. After all the carbonate was dissolved the sample was centrifuged at 12,500 rpm for 25 min to concentrate the insoluble organics The s upernatant solution containing the so luble matrix proteins was then exhaustively dialyzed against 3xDH20 using an Amicon ultrafiltration device (molecular weight cutoff 10 000 Daltons) to remove excess salts and free amino acids. After dialysis, samples were lyophilized and weighed (Appendix 1 ), and stored at -70C until used (Figure 1 0). Permeability The permeability was measured on the two species to determine possible fluid flow pathways that would contribute to diagenetic alteration. A piece of shell material from Mercenaria and Polinices samples was drilled off of the specimen using a Dremel MotoTool. The permeability was then determined for several samples (Appendix I) b y the process of Microflow Permeability (Porous Materials, Inc., Ithaca NY). In this method, the sample is sealed in a chamber of known (calibrated) volume. Pressurized gas is then applied to one side of the sample and the pressure on the other side of the sample is measured. If flow is occurring through the sample the pressure on the post-sample side will increase proportionately From this data, a graph of time versus pressure (post) is plotted The calculations are based on the ASTM Method DI434-82 A linear regression is performed on the pressure versus time data to yield the slope of the line (torrA/sec). 25

PAGE 38

Slope = A flow rate. Q is calculated by the equation for both the testin g pressure and t empe ratur e. as well as STP conditions. At STP: Slope(torrA /s ec) x Vol(cc) x T P (K) Q= ____________________ __ Where: Vol = Volume of chamber (cc) T P = Temperature at STP (273K) P ,,r = Pres s ure at STP (760 torr) At Test Conditions: Q= Slope(torrA/sec) x Vol(cc) P pre( torr A) P post( torr A) Ppre =Average Pre-Pressure at Test Con ditions (terrA) Pposl = Average Po s t-Pressure at Test Conditions (torrA) T,. = Temperature of testing gas (K) Thi s value Q is then normalized to the cross-sectional area by dividing by the sample cro s s-section giv ing the Gas Transmission Rate (GTR). Q GTR= 26

PAGE 39

Where: Asample = Cross-sectional area of sample ( cm2 ) The permeance is then calculated by dividing the GTR by the average pre-pressure (pressure on gas side of sample) in torr A. GTR Permeance = ---ppre From the permeance, the permeability can be determined by multiplying the permeance by the sample thickness in meters. Permeability = Permeance x Thickness The ASTM referenced above recommends permeability units of cc / sec / m / torr, as are used here. Heating Experiments Several temperature studies were performed on the modern specimens to simulate diagenesis. By subjecting the shell material to elevated temperatures for specific periods of times the decay process can be accelerated and pathways of diagenesis can be inferred. The amount of degradation depends on time and temperature, however it is more influenced by temperature than time (Abelson, 1954). Therefore, modern samples were heated at I 00C for I week I 00 C was chosen to get a maximum temperature 27

PAGE 40

without boiling. The shell wa s placed i n a beaker w ith 3xDH20 that wa s adjust e d t o a pH o f 7 6 This wa s done to approximate Flor ida groundwater in the Sarasota area (Stewart per s. comm ) where the fossil samples were obtained since pH can be expected to pla y a significant part in the kinetics of amino acid decomposition (Hare 1969). Samples were al s o heated dry as a control. In wet heating distilled water was used for the h y drol y tic re a ctions because higher r a tes of de g radation occur in the pre s ence of water (E g lin ton and Logan 1991 ) This procedure i s slightl y different than has been used previou s ly (Vallentyne 1969 ; Totten et al. 1972 ; Andrews et al. 1979; Qian et al. 1995) In their experiments shells were crushed prior to heating in order to provide more surface area for h y drolysis to occur (En gel, per s comm.) and to assure homogeneit y (Vallentyne 1 969). My experiment s utilized whole shells to more accurately simulate the conditions in natural environments. After heating shell s amples were decalcified according to the procedure s mentioned above and analyzed for protein and amino acid composition as described in detail below SEM A subsample of each species was examined under the Scanning Electron Microscope (SEM). It was assumed that this subsample represented the condition of the entire shell. The shell fragment was mounted on a stub and sputter coated with gold palladium for 3 minutes. It was then obser v ed and photographed on an ISI-DS130 Dual Sta g e SEM located in the Department of Marine Science at the Universit y of South Florida. The photographs were used to ensure that all contaminants were removed and to d e termine the ultrastructural pattern of the shell material. 28

PAGE 41

RPHPLC Analysis and Fraction Collection The organic matrix proteins were se parated into fractions b y rever se phase high performance liquid chromato g r ap h y (RPHPLC) equilibrated with 0.1% trifluoroacetic acid (TF A) in 3xDH20 (Buffer A). Separation was done with a Vydac Protein column (Vydac, Sep ara tion Group Herparia CA) at a flow rate of 5 ml / min. at 25C through an Isco V 4 absorbance detector (28 0 nm ) The proteins were eluted using increasin g co ncentrations of acetonitri l e to 95% acetonitrile contai nin g 0.1% TF A (Buffer B) with a gradient of0-5 min. at 0 % Buffer Band 15-40 min at 100% Buffer B. By using the se volati l e, low pH so lvents and short (C4 ) alkyl chain covalently bonded large pore size (5 1-1m particles 300 A pores ) silica co lumns, the peptides and proteins in a size range from 5 amino acids up to > I 00 000 daltons can be separated in a single run, and more than 90% of the s ample can be recovered (Hunkapiller et al. 1984a) In addition, peptides differing b y as little a s a single amino acid re s idue can be resolved (Hunkapiller eta!.. 1984b ). All fractions for amino acid analysis were manually collected into test tubes and f reeze dried Protein fractions for analysis of 813C and 81 5N were collected into ballasts rotoevaporated, transferred into quartz tubes and subsequently lyophilized (Appendix 2) The column was cleaned between samples by using an acid wash (1 :4 O.IN HN03: isopropano l ; for 40 min.) Bovine Serum Albumin (BSA) was used as a control to determine if isotopic fractionation occurred during chromatographic separations of proteins (Ostrom eta!., 1994). After an initia l practice run to observe where peak s occurred, samples were collected at 12-13 :2 0 min. (pre-peak), 14:30-21 :30 min. (peak), and 23-24:20 min. (post-peak) After cleaning the column a blank was run and column 29

PAGE 42

eluent was collected at exactly the same times. All sample ID's and collection tim es can be found in Appendix 2. 813C/81 5N ofRPHPLC Fractions Quartz tubes with l yoph ili zed samples were sen t to Michigan State U ni vers it y to be analyzed for 813C and 815N I so topic integrity during RPHPLC (Ostrom et aL 1994). and dur i ng rotary evaporation (Feuerstein eta!., 1997) has previously been demonstrated. The ana l ysis of 813C/815N compos ition s of bul k protein material was performed o n a VG Isochrom consisting of a Hewlett-Packard 5890 GC coupled to a VG PRISM (Mi cromass) isotope-ratio ma ss spectrometer via a combustion furnace and water t rap The standard deviation of 8"C va lu es for proteins i s typically 0 5 %o (Os trom et al., 1 994). Nitrogen and carbon isotope ra ti os are ex pressed in per mil as: DNE = [(R samplj R slandard) 1] X 1000 Where : N = the heavy isotope of e l ement E R = the abunda n ce ratio of the hea vy to light isotope T h e standard for carbon is the C hic ago Peedee Belemnite (PDB) and for nitro gen i s atmospheric N2 Amino Acid Analysis Bul k po l ypeptides and fractions containing polypeptides collected from RPHPLC were hydrolyzed with 200 of 6N HCI at 11 0C in vacuo for 24 hour s on a Waters 30

PAGE 43

Pi coTag workstation and redried. H y drol yzed sa mpl es were se nt to the Protein Chemistry Core Laboratory at the University of Florida for amino acid analysis. There. sam ple s were analyzed using a Beckman System 6300 High-P erfo rman ce Amino Aci d Analyzer. Thi s sys tem u ses cation-exchange with postcolumn ninh ydrin derivatization for se paration and quantification of amino-containing components. Percent compositions were calculated for each of the amino acids present in the fractions collected. 31

PAGE 44

Sample Preparation and Analysis Shell sample 1 Sample Heated ( wet and dry) VVashed Crushed to 1 Piece for Permeability 1 Piece for SEM Dec a lcified with EDT A 1 Centrifuged IM SM Dial y zed 1 Lyophilizr and weighed RPHPLC Amino Acid Analysis Figure I 0 Flow chart of major steps in sample preparation and anal yse s 32

PAGE 45

RESULTS SEM Gastropods Observation and comparison of SEM photomicrographs revealed associations of she ll microstructure and its relation to organic material. Photographs of PM can be found in Figures 11-16 and PF in Figures 23-28 The cross-sectional view (Figure 11) ofthe modem gastropod showed the complex-crossed lamellar structure. Two sets of elongated crystals were observed each one ti lt ed at some angle to the shell surface as Lowenstam a nd Weiner (1989) had previously reported. A closer view of th ese two se t s of crystals i s s hown in Figure 16. Also ev id ent in Figure 11 are organic layer s. The s e show up as wavy thin protrusions separating crystal layer s At higher magnification (Figure 12) the alignment of the crystalline packages becomes even more obvious. The outer shell s urface of this sample also exhibits pristine features (Figure 13). A gastropods nacreous layer is formed by stacks of crystals (Wilbur 1 974). Seen here are these nacreous stacks columnar aggrega t es of aragonite crystals The aragonite tablet s making up the stacks grow within multi-layered compartments partitioned by organic sheets and the g rowing tablet s are surrounded by organic envelopes. The regularly s paced organic s heet s are a rranged parallel to the surface, and each aragonite tablet that makes up the stacks grows hori zo ntall y within the compartment (Nakahara et al. 1982) These p yra mid shaped 33

PAGE 46

stacks have been widely observed in gastropods by many other authors (Mutvei. 1970 : Wind and Wise 1974 ; Wilbur 1974 ; Nakahara eta!., 1982) In the Polinices seen in Figure 1 1 a nacreous layer is observed bounded by the two complex cro s sed lamellar layers This nacre comprises about 20% of the thickness of the cross section. Bivalves Photographs of MM can be found in Figures 17-20 and MF in Figures 21 and 22. The outside surface of the modem bivalve shows several p rominent ridge and interridge areas (Figure 17 and 18) These shell ridges are produced when these molluscs grow by adding increments at the shell margin (Wilbur 1974) A biva l ves nacreous layer is normally formed from three lamellae, each one crystal in thickness. and sometimes contains irregular polygonal plates (Mutvei 1970 ; Wilbur 1974). Figure 15 demonstrates that the bivalve sample was dominated by its prismatic layer. with only a small percentage of nacre evident (upper right comer where evidence of polygonal plates can be observed) Secretion of the prismatic or crossed lamellar-like structures is a prerequisite to nacre formation (Wind and Wise, 1974), therefore that may be the reason for the scarcity ofthe nacreous layer. Comparison When comparing the modern t o the fossil gastropod the modern samples crystallites appear pristine (Figure 13 ), and the cross-sectional view shows evidence of layered organics, as mentioned previously. The fossil specimen displays crystal alteration with borings and dissolution evident (Figure 23), and no trace of organic 34

PAGE 47

Figure 11. SEM Photomicrograph of PM cross-sectional view. Can observe possible organic layers enclosed in a nacreous and crossed lamellar structure. Figure 12. SEM Photomicrograph of PM cross-sectional view. Note intact complex crossed lamellar crystallites aligned in packages. 35

PAGE 48

Figure 13. SEM Photomicrograph of PM, outer surface. Nacre stacks show intact, non porous structure. Figure 14. SEM Photomicrograph of PM inside surface. Shows possible active growth s urface of the complexc ro ssed l amellar s tructure. 36

PAGE 49

Figure 15. SEM Photomicrograph of PM, inside surface. Intact linear crysta l s are evident. Figure 16. SEM Photomicrograph of PM. Groups of crystals oriented perpendicular to each other show the complex crossed lamellar structure clearly. 37

PAGE 50

Figure 17. SEM Photomicrograph of MM, outer surface. Well-defined shell ridges and interridges seen. These could be possible paths of fluid flow that would lead to greater dissolution. Figure 18. SEM Photomicrogr aph of MM. View of side of ridge from Figure 17. Pores are visible, but have a defined structure to them. Crystals appear to be in subcircular packages. 38

PAGE 51

Figure 19. SEM Photomicrograph of MM. On right side of photo is the in s ide surface on l eft side is a cross-sectional vie w Figure 20. SEM Photomicrograph o f MM, closer magnification of cross-sectiona l view. Crystal layers of prismatic structure apparent. 39

PAGE 52

Figure 21. SEM Photomicrograph of MF, outside surface. Ridges and interridges no lon ger visible. Cannot distinguish between original porosity and post-mortem diagenesis. Figure 22 SEM Photomicrograph of MF, extreme top of photo is inside surface, rest of view i s a cross-section This cross-sectiona l view does not show significant diagenesis compared to the modern sample in Figure 19. 40

PAGE 53

Figure 23. SEM Photomicrograph of PF, outside surface. Prominent dissolution and boring. Figure 24. SEM Photomicrograph of PF cross-sectional view. Two distinct c r ystal layers are evident with their crystals oriented perpendicular to each other in this low magnification shot. Broken crystal faces indicate some diagenesis. 41

PAGE 54

Figure 25. SEM Photomicrograph of PF cross-sectional view Ridges of crystals in cross-section from lower half of Figure 24. Figure 26. SEM Photomicrograph of PF, c r oss sectio n a l view. Closer magnification of about 2 ridges from Figure 25. 42

PAGE 55

Figure 27. SEM Photomicrograph of PF inside s urface Small round pits with elongate crysta ls across them indicate dissolution. Figure 28. SEM Photomicrograph of PF inside surface. Closer magnification of dissolution pit from Figure 27. 43

PAGE 56

material is left in cross sectional v iew (Figure 24). In addition, the nacre stacks appear broken and less defined (Figure 23) and dis sol ution pits occur on the inside surface (Figure 27 and 28) The most obvious difference between the modem and fossil bivalve i s the disappearance of ridges on the outer s urface. The protein content of the bivalve ridge is nearly double that of the interridge area (Wilbur. 1 974). T heref o re the diagenetic los s of these ridges (Figu r e 21 ), even though they constitute only a small proportion of the entire shell, may adverse l y affect the organic constituents. Abundant pores are also seen, enhancing the original porosity in the modem specimen (Figure 18). Data shown here support the hypothesis that gastropods have a more intac t s hell structure. For example, Polinices lack the original pores seen in the bival ves that could be conduits for pore water s. As pointed out above Polinices also posses s more structure and order to their crystallites. Nakahara et al. ( 1982) observed a central core in the organic s heet s of a gastropod which was not observed in bivalve specimens. Thi s indicates that the sheets of gastropod nacre are more elaborate in structure and possibly more rigid than those in bivalves (Nakahara et al., 1982). In addition. data shown in this study indicated that gastropods have a greater amount of organic material per shell weight than bivalves. RPHPLC A comparison was made of the RPHPLC chromatograms for the modern versus the fossil samples in both Mercenaria merc en aria and Polinic es duplicatus based on the elution times of the fractions and the peak height (absorbance). Data from HPLC 44

PAGE 57

analyses demonstrate that the protein signatures for the species analyzed have overall similar patterns. although the y do show inherent differences. The overall elution profil e shows a lar ge peak at the onset of the run (0% acetonitrile) and three peaks between 5095% acetonitrile. Modern Mercenaria (Figure 29) showed peak retention times of approximately 3:00 minutes for the hydrophilic fraction and 12:30 13:00. and 14:10 minute s for the hydrophobic fraction. The hydrophilic fraction s howed a double peak Fossil Mercenaria (Figure 30) showe d similar results with retention times of 3:00. II :00 1 2:20, and 14:00 minutes; however the hydroph obic fractions indicated a higher absorbance. The heated modern sample (MM-2.1-Figure 32; and MM-2.2-Figure 31) also showed this greater absorbance and single hydrophilic peak The chromatogram for the modern Polinices (Figure 33) showed the same initial injection peak at 3:00 minutes although with a lower absorbance. The h yd rophilic fraction did not indicate any clear peak s. Three small peaks occurred at 11 :30 I2 :40 and 1 3:20 minutes. In the heated samples (PM-3-Figure 36 ; and PM-5 -F igure 35), these 3 peaks were much clearer but still very small. Fossil Polinices (Figure 34) had retention times of3:00, 10 : 00 11 : 40 and 13:30 minutes. These hydrophobic fractions had a greate r absorbance when compared with the modem sample. Amino Acid Analysis Amino acids are compared based on percent composition calculated in nanomolar concentrations. Fractions collected from peaks indicated on Figures 29-36 were used for amino acid analysis because they were the major protein-containing fractions. The r es ults for th ese fractions are s hown in Figures 37-46. The bulk of the proteinaceous 45

PAGE 58

MM-1.2 Protein Fractions MM1200 --Peak 2 -E s:::: 0 CX) Cll 0 s:::: Peak 3 .c ... 0 Ill .c 1 <( I--"' .'-0 10 20 30 40 T i me ( min.) Fi g ure 29. Typical RPHPLC graph of soluble organic fractiOn from modern Mercenaria. Labeled peaks were used for amino acid analysis. Peak 2 appears as only 1 peak because data went out of ran ge. A double peak actually occurs there. MF-2 Protein Fractions M F20 1 Peak 1 E Pe a k 3 s:::: 0 Peak 2 CX) Cll Peak 4 0 s:::: .c ... \\ 0 Ill .c <( I L...--u 0 10 20 30 40 T i me (min .) Figu re 30. T y pical RPHPLC graph of sol u b l e organic fraction from fossi l Mercenaria. Labeled peaks were used for amino acid analysis. 46

PAGE 59

MM-2.2 Heated Dry I MM2200 -Peak 3 Peak 5 P eak 1 Ill -E s::: 0 c:o (I) 0 Peak 2 s::: 111 .c Pea k 6 .... 0 1/) .c <( f--\...-0 10 20 30 40 Time (min) Fig u re 31. RPHPLC graph of so lubl e organic fraction from modem M e rcenaria. This sa mple was h eated dry. Labeled peaks were us ed for a mino acid analysis MM-2 1 Heated Wet MM2100 ----Peak 1 Peak 3 Pea k 4 r--'1 e s::: 0 c:o N (I) P eak 5 0 s::: 111 .c .... 0 1/) Pea k 2 .c <( r-J 0 10 20 30 40 Time (min ) Figure 32 RPHPLC graph of soluble organic fraction from modem Mercenaria This samp le was he ated in water. Labeled peaks were u sed for amino acid analysis. 47

PAGE 60

E t: 0 co C1l C) t: I'U ..0 .... 0 1/) ..0 q: 0 PM-2 Protein Fractions PM200 Peak 2 10 20 r < ) 30 1me mm. 40 Figure 3 3. Typical RPHPLC graph of soluble organic fraction from modem Polinic es. Labeled peaks were used for amino acid analysis. PF-2 Protein Fractions PF200 -Peak 1 -E P eak 3 t: 0 co C1l Peak 2 C) t: I I'U ..0 1\ .... Peak 4 0 1/) ..0 q: f---0 10 20 30 40 Time (min.) Figure 34 Typical RPHPLC graph of soluble orgamc fractiOn from fossil Polmi ces L a beled peaks were used for amino acid analy s is. 48

PAGE 61

PM-5 Heated Dry PM500 Peak 1 -----E c:: 0 IX) C1l () c:: 111 .0 ... Peak 3 0 1/) .0 Peak 2 Peak 4 <{ 0 10 20 30 40 Time (min ) Figure 35. RPHPLC graph of solub l e organic fraction from modem Polinices This samp le was heated dry Labeled peaks were u sed for amino acid analysis PM-3 Heated Wet PM300 -----------------E Peak 1 c:: 0 Pe k 1 IX) I C1l () c:: 111 .0 ... 0 1/) .0 <{ Peak 2 Peak 3 f.-.._.. 0 10 20 30 40 Time (min.) Figure 36 RPHPLC grap h of so lubl e orgamc fraction from modem Po lmz ces. This san1p le was heated in water. Labe l ed peaks were u se d for amino acid a n alysis 49

PAGE 62

material from the SM was found in these peaks because each peak contained most of the individual amino acids. Refer to Appendi x 3 for amino acid abbreviations and Appendix 4 for all amino acid data. The amino acid results for the fractions collected from the modern Mercenaria mercenaria sample are shown in Figure 37 The percent composition of aspartic acid is relatively high in both fractions. Moderate percentages are shown for serine and glycine. Very low percentages (below 5%) are represented by threonine isoleucine. and leucine Proline and alanine only occur in peak 2 while arginine occurs only in peak 3. Peak 3 has generally higher percent compositions of hydrophobic amino acids (e.g., phen y lalanine and lysine), as compared to peak 2. The percent compositions for the fractions collected for the fossil Mercenaria mercenaria sample are shown in Figure 38. High relative concentrations of glycine and phenylalanine occur in most fractions of the fossil sample All fractions show very low percentages of isoleucine leucine, and arginine. A higher percent composition of hydrophobic amino acids (e.g. phenylalanine and histidine ) can be seen in the fossil fractions relative to the modern fractions and of the hydrophilic amino acids (e.g. aspartic acid and proline) in the modern compared to the fossil (Figure 41 ). Valine methionine and tyrosine are completely absent from all fractions of both modern and fossil. Proline is absent from all fractions ofthe fossil and only occurs in a very low percentage in the hydrophilic fraction of the modern An overall general decrease occurs in the number of amino acids present in the fossil fractions relative to the modern fractions. 50

PAGE 63

70 60 c: 0 50 1/) g_ 40 E 0 (.) 30 .... c: Q) 20 Q) a.. 10 Modern Mercenaria Amino Acid Composition asp thr ser glu pro gly ala val rret ile leu tyr phe his lys arg Amino Acids 1!1 Peak 2 Peak3 Figure 37. Modern Mercenaria amino acid compositions Shown are Peaks 2 and 3 from Figure 29 Fossil Mercenaria Amino Acid Composition 70 ----------------------60 c: 0 50 -.. 0 g_ 40 E 0 (.) 30 -c: Q) :: ____._.._..._ c [b _..I_ .............. t asp thr se r glu pro g ly ala val rret ile leu tyr phe his lys arg Amino Acids i 111 Peak 1 1 Peak2 0 Peak 3 o Peak4 Figure 38. Fossil Mercenaria a mino acid composition s. Shown are Peaks 1-4 from Figure 30 51

PAGE 64

c 0 Ill 0 a. E 0 () ..... c Q) (J .... Q) a.. Modern Polinices Amino Acid Composition 25 ____ 20 15 10 5 0 asx t hr ser glx pro gly ala val rret ile leu tyr phe his lys arg Amino Acids 0 Peak 1 Peak 2 Figure 39. Modern Polinices am ino acid compositions. Shown are Peaks 1 and 2 from Figure 33. c 0 Ill 0 a. E 0 () ..... c Q) (J .... Q) a.. 70 60 50 40 30Fossil Polinices Amino Acid Composition asx thr ser glx pro gly ala val rret ile leu tyr phe his lys arg Amino Aci d s m Peak 1 Peak2 t O Peak 3 I O Peak4 __ .... Figure 40. Fossil Polinices amino ac i d compositions Shown are Peaks 1 4 from Figure 34. 52

PAGE 65

70 60 1:: 0 50 ;e 1/) g_ 40 E 0 u 30 ... 1:: Modern vs. Fos sil Mercenaria Amino Acid Composition ] 0 -,..._..u..__....._........, ...... L!4 asp th r ser glu gly a l a leu phe his lys Amino Acids 2 M\1 1 Figure 41. Modern vs. fossil Mercenaria amino acid compositions. Only amino acids above 5% composition are shown. Modern vs. Fossil Polinices Amino Acid Composition 70 ---------------------------------------60 as x thr ser glx pro gly ala val ile leu phe lys Amino Acids 1 1 II!J PF 4 Figure 42 Modern vs fossil Polinices amino acid compo s itions. Only am ino acids above 5% compo s ition are shown 53

PAGE 66

Two fractions were collected from a modem Polinices duplicatus RPHPLC run for amino acid analysis (Figure 39). Comparison of the percent compositions in the fractions show a general trend of high percentages of hydrophilic versus hydrophobic amino acids The highest percentages are represented by aspartic acid and glycine Th e amino acid percent composition are shown for four fractions from the fossil Polinices in Figure 40. High percent compositions are seen in most fractions of proline and glycine Once again comparison of the modem and fossil fractions shows a definite decrease in the number of amino acids present in the fossil sample relative to the modern (Figure 42) There are overall lower percentages of aspartic acid proline and alanine and a higher percentage of glycine and glutamic acid in the fossil fractions compared to the modem. Comparing Figures 41 and 42 shows the difference between Mercenaria and Polinices. While proline is essentially absent from Mercenaria it is found in a rather high concentration in Polinices. This is also true of valine. Mercenaria contains a relatively greater amount of phenylalanine histidine and lysine. Heated Mercenaria amino acid compositions are shown in Figures 43 and 44 (heated dry and wet). The two samples show overall similar trends with the exception of less glycine alanine, and histidine and more proline and valine in the sample heated wet. Heated Polinices (Figures 45 and 46) seem to show a greater range of values under dry and wet conditions. The Polinices however, do show the same trend overall with less glycine and histidine more proline and slightly more valine. There is also an increase in aspartic acid and serine. 54

PAGE 67

45 40 c:: 35 0 ;e 30 II) 0 25 0 (.) 20 ..... c:: Q) 15 Q) c. 10 .. asp ser Heated Mercenaria Amino Acid Composition --.....,.,......-------' ----pro gly ala leu his tys Amino Acids El A3ak 1 1 A3ak 2 o A3ak 3 0 A3ak 5 A3ak6 Figure 43. Heated modem Mercenaria (MM-2.2 heated dry) amino acid compositions Shown are peaks 1 2 3 5 and 6 from Figure 31. Only amino acids above 5% composition are shown. Heated Mercenaria Amino Acid Compositions 45 -----___________ ...... ____ 40 c:: 35 0 ;e II) 30 : 0 0.. 25 E t 0 (.) 20 1 .... l c:: Q) 15 i 0 Jruill; I .... Q) c. 10 5 0 -JJJ .d asx th r ser glx pro gly ala val leu phe lys Amino Acids l!l A3ak 1 A3ak2 0 A3ak 3 I D A3ak 4 1 A3ak5 Figure 44 Heated modem Mercenana (MM-2 1 heated wet) ammo acid compositions. Shown are peaks 1-5 from Figure 32 Only amino acids above 5% composition are shown. 55

PAGE 68

r:: 0 ;e (/) 0 c. E 0 (.) -r:: Cl) (J ... Cl) a.. 80 70 60 50 40 30 20 10 0 Heated Polinices Amino Acid Composition _____ _. ____ -asp thr ser glu pro g ly ala val me t ile leu tyr phe his lys arg Amino Acids Peak 1 [;]Peak 2 Peak 3 0 Peak 4 Figure 45 Heated modern Polinices (PM-5, heated dry) amino aci d compositions. Shown are peaks 1-4 from Figure 35. r:: 0 ;e (/) 0 45 40 35 30 e 2s. 0 (.) s:: Cl) (J ... Cl) a.. 20 Heated Polinices Amino Acid Composition -----________ ___. ..a I IR:l asx thr ser glx pro g l y ala val rret ile leu tyr phe his l ys arg Amino Acids liE Peak 1 1 Peak2 : o Peak 3 Figure 46. Heated modern Polinices (PM-3, heated wet) amino acid compositions. Shown are peaks 13 from Figure 36. 56

PAGE 69

Stable Carbon and Nitrogen Isotopic Analysis The chromatographic runs u sed for stable carbon and nitrogen i sotope analysi s are s hown in Figures 47-50 At the bulk protein leveL large isotopic differences occurred between hydrophobic and h ydro philic fractions of both modern and fossil shells ; greater than 9 %o for 813C and 3 %o for 815N (Table I). Modern Mercenaria showed the hydrophilic fraction to have an average of -38.0 %o and 0 96 %o for 813C and 81 5N. respectively The hydrophobic fraction had an average of 21.0 %o and 5.0 %o. Fossil Mercenaria showed a hydrophilic average of -37.5 %o and 0 67 %o and a hydrophobic average of -27.5 %o and 3.7 %o for carbon and nitrogen respectively This is a difference of 17 %o for 813C and 4 %o for 81 5N in the modern and 10 %o for 813C and 3 %o for 81 5N in the fossil. Table 1 Averages of isotope dat a for modern and fossi l Polin ices and Mercenaria Polin ices Mercenaria 813C(%o) 81 5N(%o) 813C(%o) 81 5N(%o) Modern Fossil Modern Fossil Modern Fossil Modern Fossil Hydrophilic -37.5 -38 .8 -0.09 1.06 -38 0 -37.5 0.96 0 .67 Hydrophobic -28 .6 -29 7 10. 7 12.6 -21. 0 -27.5 5.0 3.7 Modern Polinices had a h y drophilic average of -37 5 %o and 0.09 %o and a hydrophobic average of 28.6 %o and 10. 7 %o for 81 3C and 815N, respectively. Fossil Polinices showed an average of -38.8 %o and 1 .06 %o for the hydrophilic fraction and 29.7 %o and 12.6 %o for the hydrophobic fraction Samples collected between peaks and those collected during the blank run had a nitrogen content too small to detect. Additionally. samples with fraction averages that are only repre se nt ed by a single data 57

PAGE 70

point either h a d a leak on the g a s line no gas the sample broke or there was a leak in the sample during i s otope analysis In all ca s e s. the h y drophilic fraction was depleted compared to the h y drophobic in both 813C and 815N. In addition when comparin g modern to fossil 813C values. th e f ossil protein s of the hydrophobic fraction are lower than those of the hydrophobic fraction of the mod e rn s ample Thi s i s s hown by a 7 %o difference in carbon and a 1.3 %o difference in nitrogen in the Merc e naria s ample. It is interes ting to note that this difference onl y occurs in the hydrophobic fraction; the hydrophilic fractions have very simi lar averages Permeability Results of the Microflow Permeability analysis can be found in Table 2 and Fi g ure 51. MF-1 was the most pem1eable s ample and MM-1.1 wa s th e mo s t impermeable o f tho s e an a lyzed No s ignificant trend wa s s een between permeabilit y and specie s (Me r c enaria was sli g htl y les s permeabl e than Polinices) however a trend did occur between modern ver s u s fos sil within the genera. As expected in both species the fossil sp e cimens were more permeable than the modem. An aluminum slu g was used for comparison and it is shown to be orders of ma g nitude less permeable than any of the shell samples (Table 2) 58

PAGE 71

IMM11011 ---------------------E c: 0 co Q) u c: ro -e 0 1/l ..c <( ,._ 0 10 20 30 40 lime (min ) I MM11021 E c: 0 co Q) u c: ro ..c ..... 0 1/l ..c <( ,___ 0 10 20 30 4 0 lime (min ) IMM11COOI -E c: 0 co Q) u c: ro ..c 0 1/l 1 I ..c A i <( 0 10 2 0 30 40 lime (min.) F i g u re 47. RPHPLC chromatograms ofMM used for carbon and nit r ogen isotope ana l y sis. Samples were collected from : 30-1:30 (pre P' peak) 2 5:30 ( P peak) 7-8 (post I s peak pre 2nd peak) 10:30 -16 (2"d peak) and 1 8-19 (post 2"d peak). T h e top two graphs are 2 runs of t h e s ample the bottom g r aph is the blank run (See Appen dix 2) 59

PAGE 72

jMF2021 E -c 0 co (I) (.) c ro .0 0 l (/) .0 <( 1-----0 10 20 30 40 lime (min ) IMF2031 E --c 0 co (I) (.) c ro .0 0 l (/) .0 <( 1-0 10 20 30 4 0 lime ( m i n ) IMF2COOI E c 0 co (I) (.) c ro .0 0 (/) .0 <( 0 10 20 30 40 lime (min ) F1g ure 48. RPHPLC chromato g ram s ofMF used for carbon and mtro gen isotope analysi s Samples were collected from :30-1 : 30 (pre 1 s t peak) 25 : 30 ( P peak) 7-8 (po s t I 5 1 peak pre 2"d peak), 10I 6 (2"d peak) and I 8-19 (post 2"d peak) The top two graphs are 2 runs of the sample, the bottom g raph is the blank run (See Appendix 2 ) 60

PAGE 73

I PM201 1 --E' c 0 co Q) (.) c cu .l:l '-0 (Jl .l:l .f"-..r..... A '-------' 0 1 0 20 30 40 Time (min ) IPM2021 E' c 0 co Q) (.) c cu -e 0 (Jl i .l:l .f'.-r... '"-0 10 20 30 4 0 Time (min ) IPM2COOI E' ----c 0 co Q) (.) c cu .l:l 'l i 0 (Jl .l:l 0 10 20 3 0 40 Tim e ( m i n ) Ftgure 49. RPHPLC chromatograms o f PM used for carbon and mtrogen I sotope a n a l ysi s Sample s were collected from :30-1:30 (pre P' p eak) 2:30-5 : 30 (1s t peak ) 7-8 ( po s t 1 s t peak pre 2"0 peak) 11-1 6 (2"d peak) and 18-19 (post 2"d peak) The top two gra ph s a r e 2 run s of t he sa mpl e the bottom graph is the blank run (See Append i x 2). 61

PAGE 74

I PF2011 ------E c 0 co Q) () c ro -e A_ 0 r/l .!) <( f----0 10 20 30 40 lime (min ) I PF2021 E -c 0 co Q) () c ro -e 0 \ r/l .!) <( I--0 10 20 30 40 lime (min ) I PF2BOOI E' c 0 co Q) u c ro .!) .... 0 (/) .!) l l <( 0 10 20 30 40 lime (min. ) Ftgure 50. RPHPLC chromatograms of PF used for carbon and mtrogen isotope analysis Samples were collected from :30-1: 30 (preP' peak) 2 : 30-5:30 (P' peak) 7-8 (post P' peak pre 2nd peak) 9-17 (2"d peak) and 18-19 (post 2"d peak). The top two graphs are 2 run s of the s amp l e the bottom graph is the blank run (See Appendix 2). 62

PAGE 75

Table 2. Permeability results. Sample Permeability ( cc /s ec / m/torr A) Averages MF-I 1.33212e-07 I .07048e-05 MF-2 Overflow* MF-4 2.I2763e-05 MM-I.I I. I 4086eI 0 3.3 8913e-09 MM-1.2 6.66417e-09 PF-1 1.00579e-04 5.02897e-05 PF-2 3.40386e-10 PM-I I. 70959e-08 8. 73 700e-09 PM-2 3.78101e-10 Blank Aluminum Slug 5.62456e-40 Too great a flow, was not able to be mea s ured usm g th1s techn1que even at a very low initial pressure. 63

PAGE 76

I Permeability I 1. 00E-03 1 00E 04 1 .00E-05 1 00E-06 .... E 1 00E-07 0 Gl "' 0 u 1.00E-08 1 00E-09 1 00E-10 1.00E-11 Figure 51. Comparison of permeabilities of modern and fossil bivalves and gastropods. These represent 2 samples. The lines are th e range of the samples with the top of the line representing one sample and the bottom of the line the other. 64

PAGE 77

DISCUSSION Description of Proteins As has been previously reported Mercenaria and Polinices contain an in so luble and soluble shell-matrix fraction. Only the SM was analyzed in this study be cause the so luble fraction has been found to contain almost all the protein present in the shell (Wei ner and Hood 1975 ; Weiner and Erez, 1984). The IM is difficult to work with and has been ignored by most researchers because ana lytical technique s depend upon separating molecules from solution i.e. liquid chromatography (Curry 1988). The SM comprised an average of 0.23% in Mercenaria an d 0.53% in Polinices (Table 3 a lso see Appendix 1 ). While the in so luble portion of all molluscs studied composed a minor fraction of the organic matrix in the bivalves the insoluble fractions constituted a greater portion of the organic matrix than in the gastropods. Several gastropod samp le s even showed an entirely soluble protein solution. Although at this time there is n o exact explanation for specimens demonstrating different characteristics, thi s was also observed by Robbins and Ostrom (1995) for Polinices. One possibility for the difference in the amount of insoluble fraction between these two types of molluscs may be the differences in their biomineralization processe s 65

PAGE 78

Table 3. Weight percent averages These are an average of three samples Percent Sol u ble Matrix MF 0.39% 0.23o/o MM 0.06% PF 0.84% 0.53% PM 0.21% The difference in organic weight percent between the two species (0.23% vs 0.53% overall, 0 39% vs. 0.84% and 0.06% vs. 0 21% for fossil and modem, respectively) indicates that Polinices has a greater potential for preservation of indigenous proteins than does Mercenaria. Presumably, some of the soluble matrix proteins have been leached out of Mercenaria s more permeable shell structure that was seen in the SEM photos Oddly in both specimens the fossil samples demonstrated more organics than the modem. This could be due to the amount of soluble versus insoluble organics since only the solubles are included in the weight percent. As the organism goes through diagenesis, the insolubles convert to soluble organics (Abelson, 1955) perhaps causing older samples to have a higher SM content. It has been shown that the soluble organics increase and insolub l e organics decrease with artificial diagenesis (Totten eta!., 1972). The soluble fraction contains the soluble protein and peptides derived from the insoluble protein by hydrolysis during diagenesis (Akiyama 66

PAGE 79

1971 ). causing the organic matrix to become progressively more soluble with age (Hare. 1969). However. ver y little is known about the insoluble proteins and their ability to solubilize over time (Robbins et al. 1993b). Even though the IM was not part of this study an insoluble fraction was weighed and did decrease from modem to fossil. Another possible explanation could be the age ofthe specimen. Smaller (presumably younger) samples show a higher organic weight percent than larger (presumably older) ones in both species (Appendix 1 ). These younger samples would need the organic layers as a template for mineralization to help them grow. Once the shell reaches maturity and shell precipitation slows these organics are no longer as important. Experiments of heated specimens also showed that Polinices contained more organic material than Mercenaria. Furthermore, heated specimens of Polinices in water had a higher SM content than the sample heated dry. This was the opposite affect than initially expected since the water should have caused the hydrolytic reaction to accelerate and degrade more protein. These specimens follow the same trend in regards to sample age as described above. PM-3 (heated in water) was a smaller specimen than PM-5 (heated dry). When size discrepancy is taken into account these two samples may have had roughly the same organic weight percent. Possibly the sample heated in water had more of the IM broken down by hydrolytic reactions into the SM. The sample was probably not heated long enough to have the SM leach out of the shell. This scenario would account for the higher SM percent in the sample heated wet than in the sample heated dry Contrary to the trend seen in Polinices the two Mercenaria valves (MM-2 1 and MM-2.2) had the same organic content as one another despite differences in heating. These data indicated that wet heating did not have much of an effect on the organic 67

PAGE 80

weig ht percent of the shell with th e time and temperatur e u sed. The we i gh t percents tend to be more s imilar to the unheated m ode rn s than t o the fossils. suggesti n g th at t h e heating experiments may ha ve only s imul ate d the very o n se t of dia g ene s i s. Vallent y n e ( 1 969) found th e tot al amount of amino acids de composed during p yrolysis to be greatest where no wate r was ad ded. The effect of water in d ec r eas in g the extent of decomposi tio n (or h avi n g no effect at all) is most probabl y a resu lt of dilution reducing the frequency of interaction of ami n o acid s both among them se lv es and with other components of the Mercenaria shell (Vallentyne, 1969) Other aspects of these heated s ample s showed that some diag e ne s i s mu st have occurred during the artificial aging. Wherea s the modern sample s protein powder (afte r l yoph ilization) appeare d white and powdery the heated samples showed a more brownish co lor as did the fo ss il s pecimens. Abelson (1956) a l s o observed that pro tein h ad turned brown in fossil sa mple s. Although h ea tin g may not h ave s hown much of an effect on weigh t percents the HPL C chromatograms more closely re sem bled the fossi l samp le s (see Figures 29-36), which was d i sc u sse d prev iously. RPHPLC A c har acteristic chrom a togr a phic profile was obtained for each of the sa mpl es ana l yze d (modem heated unheated and fossil ) ind icatin g similar components b ased on h y drophobicit y in each of the ge nera st udi e d C hromatogram s indicated there were approximately four major prote in/pe ptide compone nts. Each of the se peaks e lut ed a t about the same time and the sa me g r a dient percentage of acetonitrile (Buffer B) The 68

PAGE 81

hydrophilic fractions eluted at 0% Buffer B. and three hydrophobic peaks eluted between 50-95% acetonitrile. Comparison of the modem to fossil and heated chromatogram s showed that th e fossil/heated had more of a hydrophobic than hydrophilic fraction. This was most pronounced in the Polinices sa mples where there are basically no hydrophobic peaks in the modem and three distinct peaks in the fossil. Along the same lines the hydrophilic fractio n decreased with a su b seque nt increase in the hydrophobic fraction, suggest in g that the hydrophilic fraction broke down and was incorporated into the h ydrophobic fraction. Robbins and Ostrom (1995) also found the hydrophobic component of the fossil to be much larger than that of the modem s hell. These shifts are an indication of mild diagenesis and are most likel y due to the preferential loss of hydrophilic organ ic s from leac hin g processes by groundwater (Mitterer 1993) that the fossils were exposed to. This is one of two possibilities that may affect the changes in the characteristics of shell proteins during diagene s is: 1) degradation of the hydrophilic proteins with subsequent incorporation of the diagenetic products into the hydrophobic fraction and 2) the introduction of contaminants during diagenesis (Robbins and Ostrom 1995) Amino Acid Analysis The amino acid compositions of the protein fractions revealed differences between modem a nd fossi l samp les, suggesting diagenetic alteration in the fossils. An abunda n ce of aspartic acid in the modern suggests good preservation. The compositiona l decrease of this acid in the fossil often reflects diagene s i s (Akiyama, 1971 ). The preferential decr ease in the percentage of h y drophilic amino acids and s ub seq uent 69

PAGE 82

increase in the hydrophobic amino acids in the fossil fractions relative to the modern in both species provides evidence for alteration of the total composition by leaching processes (Mitterer 1993) The relative decrease in the overall numbers of amino acids in the fractions could be a function of biochemical reactions which graduall y break down some of the amino acids into other compounds which are not detected by the amino acid analyzer (Mitchell and Curry, 1997). The observed differences in the amino acid composition among the samples may also be related to the differences in the pathway of and extent of diagenesis experienced b y each individual fossil. Those proteins rich in aspartic and glutamic acid are the acidic glycoproteins (Robbins and Brew 1990). The fractions rich in serine, glycine, and alanine are similar to fibrous structural proteins which help promote the structural integrity (Robbins and Brew 1990). If these are indeed structural proteins the y may be less prone to degradation than some other types of proteins because of their composition and fibrous structure (Robbins 1987). While it has been suggested in other studies that the hydrophilic fraction is comprised of glycoproteins and the hydrophobic fraction is comprised of structural proteins (Robbins and Brew 1990) the distinction between these two groups of proteins is unclear in this study. The fractions showed amino acid compositions characteristic of both groups of proteins indicating a heterogeneous mixture. Mercenaria showed a clear increase in glycine and alanine from modern to fossil. The occurrence of glycine and alanine as degradation products is not unusual and has been known to arise from serine (Vallentyne 1964 ; Mitchell and Curry 1997). There was a slightly higher concentration of serine in the modern sample (8.25% vs. 5.4% for modern and fossiL respectivel y ) 70

PAGE 83

Th e m a in differen ce b e tween the specie s was greater am o unt s o f proline g l y c i n e. alanine. a nd valine in Polini ces I s oleucin e and leucine were also s lightl y higher in thi s ga s tr o pod M e r ce naria sh o w e d lar g e amount s of phen y lalanine hi s tidine. and l ys i ne These amino acids are not usuall y found in any g reat quanti ty in s hell protein s. Thi s suggests that these are cont a mination product s that may be found in M e r c enar i a b eca u se of its mor e porous structur e a s ha s been d e m o n s tr a t e d through o ut t hi s s t udy. H owever. comparing the 2 s pecie s (Figure s 41 and 42), s hows overall similar trend s i f y ou disregard those last three amino acids. So, there seems to be a s much if not more of a difference in amino acid compositions between modem and fossil samples of the same genus a s there is between the s pecies (Figure 41 and 42). Despite e arlier studie s th at h a d shown amino acids to be phylogenetically distinct among specie s rec e nt work (Robb i n s and Brew, 1 9 90 ) indicated th at amino acid anal ys i s o f total s h e ll m a t e ri a l repre se n ts a n o vers implification of component s in the org ani c material within the s he ll. Thi s c o uld b e why data on total amino acid compositions demonstrates variabilit y and deviations from what is expected phylogenetically Therefore amino acid anal y si s s e em s t o be a m o re u s eful tool in diagenetic s tudie s than in phylogen e tic determina t i o n Stable Carbon and Nitrogen Isotopic Analysis Although a similar study comparing hydrophilic and hydrophobic fractions of modem and fo ssil shells was p e rformed u s ing th e s ame genera ( R o bbin s a nd O s tr o m. 1995) th e ir f ocu s was on isot o p e data for individu a l amino acids. O ther s tudie s h ave concentrated on amino acid data ( Ostrom et al. 1990 ; Qian e t al. 1 99 5 ), or more 71

PAGE 84

specifically on the o 13C of Dand Lenantiomers (Engel and Macko 1986 ; Serban et al.. 1988; Engel et al. 1994; Ostrom et al., 1994 ; Silfer et al. 1994). In this research. bulk proteins were isotopically analyzed to see if similar trends occurred at this highe r molecular level. Ver y similar trends did occur in all aspects of the data confim1ing that proteins and their constituent amino acids act similarly with respect to o13C and o1;N. Proteins from the hydrophobic fraction of the modem Polinices and Mercenaria demonstrate o13C values that are at least 9 %o higher than the corresponding values in the hydrophilic fraction. Since the o13C of an amino acid is partly controlled by the degree of fractionation during rate-determining steps and at major branching points that occur along degradation pathways (Robbins and Ostrom 1995), this difference could be attributed to variations in the extent of fractionation within pathways of synthesis of these two different types of proteins. The most interesting feature of the o13C data is that the values from the fossil hydrophobic fraction are lower than the corresponding modem hydrophobic fraction as a result of the loss of 13C depleted peptides from the hydrophobic fraction during diagenesis (Robbins and Ostrom, 1995). One explanation for this involves the kinetic isotope effects where the 13C depleted atom is preferentially incorporated into the product (Macko et al., 1986). Another explanation offered for these lower values is the reincorporation of isotopically depleted degradation products from the hydrophilic fraction (Robbins and Ostrom 1995). This does not seem to be the case here because averages of modem and fossil Mercenaria hydrophilic fractions are similar. Additionally, as the amino acid composition changes over time due to diagenesis the isotopic composition of bulk 72

PAGE 85

organic matter ma y also change, re g ardles s of contamination due to lar ge differe n ces in iso topic composition of individual amino acids (E ngel et al., 1994 ) This would b e evidence for diagene s is but not nece ss aril y contamination in the fossil samples Fossil Mercenaria showed a large divergence in nitrogen between the duplicate runs. Se ve ral factors may have caused this and other discrepancies. For one. sa mples had to be tran s f e rred from the roto-evap ball asts into quartz tubes Although great care was taken some l oss may have occurred during the tran s fer. Also samples were heated at 55-60 C for times ran ging from 15-45 minutes during rotary evaporation. Although isotopic integrity during this step has been determined (Feuerstein et al. 1997) heating may have caused molecular alteration to an unknown extent. Clearl y, there are a number of complex proteins with in the shell matrix and these have differin g isotopic compositions Furthermore s hell matrice s contain other organic material s uch as carbohydrates and lipid s, which have n o t yet been eval uated i s otopically It is l ikely that the interactions of s hell matrix proteins and their hydrolytic products wit h the other organic compou nd s, and the effect of the carbonate matrix which ma y protect the protein s from h y drolysis or catalyze the hydrolytic process significantly complicate isotopic fractionation (Qian 1993 ). Because the carbon values were highly depleted compared to amino acid data (Robbins and Ostrom 1995) it was suggested that EDT A may be contributing to these depleted values. Although samples we r e exhaustively dialyzed to remove th e EDT A, it has been implied t h at not all the EDT A can be removed by di a l ys i s ( Weiner et al.. 198 3; Alb eck et al., 1 996). EDTA u s ually elutes in the first peak ; thi s may mean that the actua l 73

PAGE 86

isotope value s of the hydrophilic fraction are higher, resulting in less of a difference between the two fractions. In order to further understand this, more samples wer e prepped using hydrochloric acid (HCl) rather than EDTA to decalcify the shells ( Table 4). Although the actual values are less depleted. the same general trends occur in the data. with less of a difference between fractions. The hydrophilic frac tion is still depleted compared to the hydrophobic although not by as much and the hydrophobic fractions of the fossil are still depleted compared to the modern by a significant amount (7 %o in carbon and greater than 10 %o in nitrogen). Table 4 Isotope results for samples decalcified with HCl. o13C ols N Polinices Modern Fossil Modern Fossil Hydrophilic -22.6 -28.8 2.2 -1.7 Hydrophobic -21 1 -28.2 9.6 -3.1 Furthermore comparison of modern gastropods from different environments (seagrass vs. muddy estuary) and different eating habits (herbivorous and carnivorous) demonstrate that the isotopic composition of the total organic material within the she ll is related to the carbon isotopic composition of the primary producer/consumer that the gastropod eats (Table 5) For example, Polinices from a seagrass bed shows much higher o13C values than that of Polinices from a muddy estuary setting. If this is the case, isotopes would not be as useful as a diagenetic indicator as once previously thought. Despite this several pathways of protein diagenesis could contribute to isotopic differences in o 13C of proteins between and within the organic fractions of modern and fossil shells Simply, fossil proteins may be a diagenetic product of the corresponding modern precursor. Individual fossil proteins could be derived from different fractions of 74

PAGE 87

the modern precursors through degradation and rearrangement of the diagenetic product. To delineate between different diagenetic mechanisms and isotopic variabilit y. further work is needed. Table 5. 81 3C values (in %o) for total shell organic hydrolyzate for Strombus and Polinices. Specimen s came from distinctively different environments: A Florida seagrass bed and a temperate sandy mud estuary in Massachusetts (Robbins et al.. 1998). Sea grass Estuary Polinices (Carnivore) -5 3 -24 5 Strombus (Herbivore) -16 3 Permeability When interpreting permeability data an important trend occurs between modern and fossil specimens, with the fossil specimens demonstrating a higher permeability than the modem samples This higher permeability would contribute to the diagenetic alteration of proteins that is evidenced in other data, such as HPLC chromatograms, amino acid analysis isotopic data and SEM photos Although Mercenaria appears to be slightly less permeable than Polinices, this difference may not be significant. In fossil Mercenaria, sample MF-2 was so permeable it could not be measured (see Table 2) and therefore is not included in the average Also fossil Polinices shows great variability ranging from I 04 to I o-1 0 This variability could have been caused by a leak in the seal during analysis (Gupta, pers. comm.) Finally PM and MM demonstrate permeability of the same order of magnitude and therefore may not be significant. This data has never 75

PAGE 88

been recorded so the difference between modern and fossil is an interesting beginning t o the documentation of permeabilities re l ation to dia ge ne s is. In addition when comparin g with the soluble matri x we i g ht p e rcents ( Figure 5 2). seve r a l intere st in g po i nt s become evi d ent. Referring back to the se weight percents (A ppendi x 1 ). MF-1 h ad a very low organic content ( 0 09% ) compared to other fossil Mercenaria sa mples This may a lso reflect the high permeability of this sample. Other examples of thi s include MM -1.1 versus MM-1.2 where MM-1.1 was les s permeable and had a higher we i gh t percent than MM-1.2 PF-2 is quite impermeable a nd had an unusuall y high organic wei g ht. And finall y, PM-1 i s more permeable than PM-2 and has a lo wer percenta g e of or ga nics Fi g ure 52 has much variability, but the trendline does show an increa se of organics with a decrease in permeability. MF-4 was not included on Figure 52 because part of it's SM was lost during anal ys is and therefore cannot be used in com p ariso n. So although n ot hin g conclusive can be said about whether perm eabi lit y is spec i es determined diagenesis does appear to increase permeability and perhaps there is al so some trend between permeability and organic weight percent on an individual sample basis Future Directions Several other biogeochemical techniques could be performed to further characterize and under s tand protein diagenesis More s pecific conclusions from the molecular dat a require sequencing of the protein or may be obtained through other molecular o r immunological approaches (Robbins et al., 1993b ; Ostrom et al., 1998). Successful seq uencin g of the proteins e n ab le s the re sea rch e r t o determin e the 7 6

PAGE 89

substitution s deletions and insertions of amino acids that have occurred within the polypeptide chain (Robbins and HealyWilliams 1991 ). Therefore sequence data can provide further insight into the role that the protein may play in the biomineralization of the shell and will aid in modeling degradation. ---------Sample Weight vs. Permeability I 1 .ooE-66 ? ... 0 1 00E-07 E () Cl) ..!!! o 1 00E-08 >. .c 1.00E-09 E ... Cl) a.. 1 00E-10 0 0 2 R2 = 0 2136 0.4 0 6 0 8 1 2 1.4 Sample Weight (g) 1 6 Figure 52. Corre lation of sample weight and permeability. Black line represents average trend lin e. DNA sequencing of living and ancient representatives has exciting implications in the establishment of phylogenies and the study of evolution and is a likely research direction for the future (Paabo, 1989) Ribosomal RNA in living organisms can be obtained using the polymerase chain reaction (PCR ) to amplify the sm all amounts of material available Once amplified rRNA can then be directly sequenced These techniques have been successfully applied to living molluscs (Terret eta!., 1996) a nd 77

PAGE 90

have exciting prospect s for being extended to fossilized material. Answer s to phylogenetic biomineralization, and diagenetic questions will likely come from u si n g a co mbination of these techniques Conclusions 1 ) The EDT A-soluble shell matrices of the species studied are composed of het e rogeneou s assemblages of proteins and polypeptide s as indicated by RPHPLC chromatograms and amino acid analysis. 2) SEM confirms that gastropods have a tighter ultrastructure. SEM and permeability experiments demonstrate crystal alteration in the fossils 3) Pyrol ys is on whole specimens demonstrated that shell matrix protein s were o nl y s lightly modified compared to earlier report s of more degradation in powdered samples However some diagenetic alteration during artificial aging was indicated b y HPLC and amino acid analysis The data suggest that temperature studies on whole specimens, rather than shell fragments may be a more accurate way to model in situ degradation 4 ) Amino acids show similarities between ge nera and may not be as u se ful in ph y logenetic studies. Some amino acids ma y be more phylogeneticall y useful than others. However amino acid analy sis can contribute to biomineralization and dia ge netic questions. 78

PAGE 91

5) Stable isotopes show differences between the two different fractions of modern. and also between modern and fossil. Isotope data suggest bulk proteins reflect diagenesis and protein type. Great care must be exercised when extracting stable isotopes from organic components. Man y factors may contr i bute to their depletion and enrichment. and interpretation i s often difficult and signals may be marred by a combination of circumstances. 6) A combination of all the methodologies used in this study, including HPLC amino acid analysis stable isotopic analysis, SEM and permeability can help in understanding diagenetic alteration. 79

PAGE 92

LIST OF REFERENCES Abelson P.H., 1954 Organic constituents of fo ss il s, Carneg i e In s titution ofWashingr on Yearbook vol. 53, pp. 97-101. Abelson P.H., 1 955 Organic constituents of fossils. Carnegie Institution ofWashin gto n Yearbook vol. 54 pp. 107-109. Abelson P.H 1 956, Paleobiochemistry Scient(fic A m erica n vol. 1 95 pp. 83-92 Addadi, L. and Weiner S. 1989 Stereochemical and structural r e lations between macromolecules and crystals in biomineralization in Biomineralization: Chemical and Bioche mical P e r s pective s S Mann, J. Webb, and R.J.P. Williams. eds .. VCH Verlagssgesellschaft Weinheim FRD, pp. 133 -156 Akiyama, M 1971 The amino acid composition of fossil scallop shell proteins and nonprotein s Biominerali z ation Research Report s, vol. 3, pp. 65-70 Albeck, S. Aizenberg J ., Addadi, L. and Wei ner, S ., 1993. Inter actions of various skeleta l intr acrys tall i n e components with calcit e crystals J ournal ofAmerican C hemical Soc i ety vol. 115 pp. 11691 -11697 A l beck, S., Weiner, S. and Addadi, L. 19 96 Polysaccharides of intracrystalline g l yco protein s modul ate calcite crystal growth in vitro, C h ern. Eur J., vol. 2, no 3, pp. 278-284. A llison P .A. 1988 The role of anoxia in the d ecay and mineralization of proteinaceous m acrofossils Paleobiology vo l. 14 no. 2 pp. 139-154. Ambler, R.P. and Dani e l M 1991 P rotei n s and m o lecular paleontology Philo so phical T ra n sac ti o n s of th e Royal Soc i ety of L ondo n B vol. 333 pp. 381-389. Andrews, J.T. Bowen, D.Q. and Kid so n C. 1979 Amino acid ratio s and the correlation of raised beach deposi ts in south-west England and Wales, Na tur e vol. 281 pp. 556-558. Andrews, J T. Miller, G.H. Davie s D.C., and D av i es, K.H. 19 85 Generi c id e ntification of fragmentary Quaternary m o llu scs b y amino acid chromatography : A tool for Quate rnary and paleont o l ogica l researc h Geologica l Journal vol. 20, pp. 1 -2 0. 80

PAGE 93

Anfinsen C. B .. 1959 The Molecular Basis of Evolution John Wiley and Sons. In c .. Ne,.,, York. 228 p. Bada, J .L., 1991, Amino-acid cosmogeochemistry, Philosophical Transactions of the Royal Society of London, Series B-Biological Sciences vol. 333 pp. 349-358. Carter, J.G 1990 Skeletal Biomineralization: Patterns, Processes. and EvolutionarY Trends. Volume I Van Nostrand Reinhold, New York. New York. 832 p. Carter. J.G. and Cla rk. G.R .. 1985 Classification and phylogenetic significance of molluscan shell microstructure. Mollusk-Short Course. Crenshaw, M.A 1972 The solub le matrix from Mercenaria mercenaria shell. Biomineralisation vol. 6 pp. 6-11. Crenshaw M.A., 1980 Mechanisms of shell fonnation and dissolution in Skeletal Growth of Aquatic Organisms: Biological records of environmental change Rhoads, D.C., and Lutz, R.A. eds .. Plenum Press New York, 750 p. Curry G.B. 1987a Molecular paleontology Geology Today pp. 12-16. Curry G.B. 1 987b Molecular paleontology: New life for old molecules Trends in Ecology and Evolution vol. 2, no. G, pp. 161-165. Curry G.B ., 1988 Amino acids and protein s from fossils, in B. Runnegar and J .W. Schopf eds. Molecular Evolution and the Fossil Record : Short Courses in Paleontology Number 1 The Paleontological Society, University ofTennessee, Knoxville. Tennessee, pp. 20-33. Degens E.T Spencer D.W. and Parker R.H. 1967 Paleo biochemistry of molluscan shell proteins Comp Biochem Physiol. vol. 20 pp. 553-579 Eglin ton G. and Logan G .A. 1991, Molecular preservation, Philosophical Transactions ofthe Royal Society a_( London B, vol. 333, pp. 315-328. Engel, M.H., Goodfriend, G.A., Qian Y., and Macko, S.A. 1994 lndigeneity of organic matter in fossils: A test using stable isotope analysis of amino acid enantiomers in Quaternary mollusk shells Proceedings o_fth e National Academy o_(Science USA, vol. 91, pp. 1047 5 -10478 Enge l M.H. and Macko S.A 1 986. Stable isotope evaluation of the origins of am ino acids in fossils Na tur e vol. 323, pp 531-533. 81

PAGE 94

Feuerstein. T.P., Ostrom, P.H. and Ostrom N.E. 1997. Isotopic biogeochemistry of dissolved organic nitrogen : A new technique and application. Organic Geochemist1y. vol. 27. no. 7 / 8 pp. 363-370. Ghiselin. M.T., Degens, E.T., Spencer. D.W. and Parker. R.H .. 1967 A ph y logenetic survey of molluscan shell matrix proteins. Breviora no. 262 pp. 1-35 Hare P .E. 1969 Organic chemistry of proteins pep tides and amino acids, in Eglinton. M. and Murphy eds. Organic Chemistry, Springer-Verlag New York. 828 p. Hare P.E and Abelson P H., 1964 Proteins in mollusk shells Carnegie Institution of Washington Yearbook vol. 63, pp. 267-270. Hare P.E. and Abelson P.H. 1965, Amino acid composition of some calcified proteins Carnegie Institution a_[ Washington Yearbook vol. 64 pp. 223-232. Hare P.E. and Hoering, T.C ., 1977 The organic constituents of fossil mollusc shells, Carnegie Institution ofWashington Yearbook, vol. 76, pp. 625-631. Hennet, J.C., Holm, G., and Engel M.H., 1992 Abiotic synthesis of amino acids under hydrothermal conditions and the origin of life: Perpetual Phenomenon ? Naturwissenschaf ten vol. 79, pp. 361-365. Herrmann. B. and Hummel S., eds .. Ancie nt DNA : Reco v ery and analysis of genetic material from paleontolo g ical archaeological mu se um medical. and forensic specimens, SpringerVerlag, New York 263 p. Hudson J.D. 1967, The elemental composition of the organic fraction and the water content, of some recent and fossil mollusc shells, Geochimica et Cosmochimica Acta, vol. 31, pp. 2361-2378. Hunkapiller M. Kent S., Caruthers, M., Dreyer W., Firca J. Griffin C. Horva t h S., Hunkapiller T. Tempst P., and Hood, L., 1984a A microchemical facility for the analysis and synthesis of genes and proteins, Nature, vol. 310 pp. 105-111. Hunkapiller M.W., Strickler, J.E., and Wilson, K.J ., 1984b Contemporary methodology for protein structure determination Science vol. 226, pp. 304-311. Kahne, D and Still W.C. 1988 Hydrolysis of a peptide bond in neutral water Journal o.f American C hemical Society, vol. 110 pp. 7529-7534. Kennish M.J. 1980 Shell microgrowth analysis: M e rcenaria mercenaria as a type example for research in population dynamics in Skeletal Growth of Aquatic Organisms: Biological re co rd s of environmental change Rhoad s D.C. a nd Lutz R.A ., eds., Plenum Pre ss, New York, 750 p. 82

PAGE 95

King, K. Jr.. 1977. Amino acid survey of recent c a lcareou s and si liceous deep -sea mi cro f ossils, Micropaleontology vol. 23, no 2 pp 180-193 King, K Jr. and Hare P .E., 1972a Amino acid composition of planktonic foraminifera: A pal eo biochemical approach to e v olution Science vol. 175 pp 1461-1463 Kin g, K Jr. and Hare P.E., 1972b, Amino acid composition of th e test as a ta xo nomic character for li v in g a nd fossil pl a nktonic foraminifera. Mic ropaleontology. vol. 18, pp 285 -29 3. Lowenstam H.A. 1981, Minerals formed b y or ga nism s, Science vol. 211 pp 11261131. Lowenstam, H A. and Weiner S., 1989 On Biomineralization Oxford Univer sity Press New York, 324 p Lowenstein J .M., 1980 Species specific proteins in fossils, Naturwissenschaften vol. 67 pp 343-346. Macko S.A. and Aksu, A.E 1986 Amino acid epimerization in planktonic foraminifera s u gges t s low se dimentation rate s for Alpha Rid g e Arctic Ocean Nature vol. 322. pp 7 30-732. Macko, S .A., Estep M .F., Engel, M.H., and Hare P.E ., 1986, Kinetic fractionation of stable nitro g en i s otopes durin g amino acid tran s amination Ge ochimi ca et Cos m oc himi ca Acta vol. 50, pp. 2143-2146 Macko S .A Este p M.L.F., Hare P.E. and Hoerin g T.C. 1 983, Stable nitrogen and carbon i so topic composition of individual amino acids isolated from cultured microorganism s C arne g ie Institution ofWashington Y ea rbook, vol. 82 pp 404 409 Matter III P., David so n F.D. and Wyckoff R.W.G. 1969, The composition of fossil oyster shell protein s, Proceedin gs of the National Academy ofScience USA, vol. 64 pp 970-972. Mitchell L. and Curry G.B 1997, Diagenesis and s urvival of intracrystalline amino acids in fossil and recent mollu s c s h ells Palaeontology vol. 40 p a rt 3 pp. 855874 Mitterer R M. 1993 The diagenesis of protein s and amino acids in fossil s hell s, in Organic Geochemistry M .H. Engel and S .A. Macko eds ., Plenum New York pp. 739-753. 83

PAGE 96

Moore, R .C., 1969. Treati se on Invertebrate Paleontology (N) Mollusca 6. The Geological Society of America. The University of Kansa s. 460 p. Mutvei, H., 1970 Ultrastructure of the mineral and organic components of molluscan nacre ous layers. in Biomineralization Research Reports. F .K. Schattauer Ver lag. Stuttgart. Germany pp. 49-72. Muyzer, G., We s tbroek P., and Wehmiller J.F., 1988 Phylogenetic implications and diagenetic stability of macromolecules from Pleistocene and recent shells of Mercenaria mercenaria (Mollusca, Bivalvia) Historical Biology vol. 1 pp 135144. Nagy, B., Engel, M.H. Zumberge I.E. Ogino, H. and Chang, S.Y., 1981. Amino acids and hydrocarbons 3 800-Myr old in the Isua Rocks southwestern Greenland. Nature, vol. 289 pp. 53-56. Nakahara, H., Bevelander, G., and Kakei M. 1982, Electron microscopic and amino acid studies on the outer and inner shell layers of Haliotis rufescens Venus vol. 41, no. 1, pp. 33-46. Olsson A.A., Harbison, A., Fargo, W.G. and Pilsbry H.A. 1953 Pliocene Mollusca of Southern Florida Wichersham Printing Co. Lancaster Pennsylvania. 457 p. Ostrom P.H. Ghandi H. Ca hue L. Gage D.A. Shen. T., Huan g, Z., and Hau s chka. P.V. 1998, New perspectives on a ncient proteins: The application ofMALDIMS for characterization of modern and ancient osteocalcin protein sequences Programs and Abstracts: Perspectives in Amino Acid and Protein Geochemistry Washington, D.C. p. 32. Ostrom, P.H. Macko S.A., Engel M.H. Silfer, J.A. and Russel D., 1990 Geochemical characterization of high molecular weight material isolated from Late Cretaceous fossil s, Organic Geochemistry, val. 16, pp 1139-1144. Ostrom, P.H ., Zo nneveld J., and Robbin s, L.L., 1994 Organic geoc hemistry of hard parts: Assessment of isotopic variability and indigeneity, Palaeogeography, Palaeoclimatology, Palaeoecology, vol. 107, pp. 201-212. Paabo, S. 1989 Ancient DNA: Extraction, characterization molecular cloning and enzymatic amplification, Proceedings of the National Academy of Science USA, vol. 86, pp. 1939-1943. Qian Y., 199 3, Kinetic aspects of the diagenesis of organic compound s and the associated kinetic isotope fractionations Ph.D. Dissertation, University of Oklahoma pp. 141-226. 84

PAGE 97

Qian Y., Engel, M.H. Goodfriend G.A. and Macko, S A. 1995, Abundance and stabl e carbon isotope composition of amino acids in molecular weight fractions of fossil and artificially aged mollusk shells, Geochimica et Cosmochimica Acta, vol. 59. no.6, pp.1113-1124. Robbins, L.L. 1987, Morphological variability and protein isolation and characterization of recent planktonic foraminifera. Ph.D. Dissertation. University of Miami. 302 p. Robbins. L.L., Andrews, S., and Ostrom P.H. 1998 Carbon and nitrogen isotopic variability of different protein fractions from modern and fossil shells. Programs and Abstracts: Penpectives in Amino Acid and Protein Geochemistry Washington D.C. p. 60. Robbins L.L. and Brew, K., 1990, Proteins from the organic matrix of core-top and fossil planktonic foraminifera, Geochimica et Cosmochimica Acta, vol. 54, pp. 22852292. Robbins, L.L. and Donachy, J.E. 1991, Mineral regulating proteins from fossil planktonic foraminifera Surface Reactive Peptides and Polymers, American Chemical Society pp.139-148. Robbins L.L. and HealyWilliams, N 1991, Toward a classification of planktonic foraminifera based on biochemical geochemical and morphological criteria. Journal of Foraminiferal Research vol. 21, no. 2 pp. 159-167. Robbins L.L., Muyzer G. and Brew, K. 1993b, Macromolecules from living and fossil biominerals: Implications for the establishment of molecular phylogenies, in Organic Geochemistry M.H. Engel and S.A. Macko, eds. Plenum, New York pp. 799-816. Robbins, L.L. and Ostrom P.H., 1995 Molecular isotopic and biochemical evidence of the origin and diagenesis of shell organic material, Geology, vol. 23, no. 4, pp. 345-348. Robbins, L.L., Toler, S.K. and Donachy J.E. 1993a, Immunological and biochemical analysis of shell matrix proteins in living and fossil foraminifera, Lethaia vol. 26, pp. 269-273. Serban, A., Engel M.H. and Macko S.A. 1988 The distribution, stereochemistry and stable isotopic composition of amino acid constituents of fossil and modern mollusk shells Organic Geochemistry vol. 13, pp. 1123-1129. Silfer J .A., Qian Y., Macko, S.A., and Engel, M H., 1994 Stable carbon isotope composition of individual amino acid enantiomers in mollusc shell by GC/C/IRMS Organic Geochemisfly vol. 21, no. 6/7, pp. 603-609. 85

PAGE 98

Terrett. J.A.. Miles. S and Thomas R H ., 1996 Co mplete DNA sequence ofthe mitochondrial ge nom e of Cepaea n emo rali s (Gas tropoda : Pulmonata). J o urn a l(?( Molecular Evo l ution. vol. 42. no 2 pp 160-168 Totten. D.K., Da v id so n F.D and Wyckoff. R.G 1972 Amino acid composition of heated oyster shells. Proceedings of the Natio nal Academy o(Science USA. vo l. 69 pp 784-785 Vallentyne, J R. 1964 Biogeochemistry of organic matter-II : Thermal reaction kinetics and transformation products of amino compounds G eoc himica et Cos mochimi ca Act a, vol. 28, pp I57-I88. Vallentyne, J.R I969, Pyrolysis of amino acids in Plei s tocene Mercenaria shells Geoc himica e t Cos mochimica Acta, vol. 33, pp. 1453-I458. Watabe N., I965 Studies on shell formation: XI. Crystal-matrix relationships in the inner la y ers of mollusk shells Journal of Ultrastructure Re se arch vol. 12 pp. 351-370. Weiner S., 1986 Organization of extracellularl y mineralized tissues: A comparative s tud y of biol og ical crystal growth CRC Crit. R e v Bio c h e m. vol. 20, pp 365-408. Weiner. S an d Erez. J. I984 Organic m a trix of the shell of the foraminifer Heterostegina depressa Journal R esearch, vol. 14. no. 3 pp. 206-212. Weiner S and Hood L. 1975 Soluble proteins of the organic matrix of mollusk shells: A potential template for shell formation Science, vol. 190 pp. 987-988 Weiner, S ., Lowenstam H A. and Hood, L. 1976 Characterization of 80-million-year old mollusk shell proteins Pro cee dings of the National Academy of Sci e nce, USA, vol. 73, no 8 pp. 2541-2545. Weiner S. Lowenstam, H A. Taborek B., and Hood L., I979 Fossil mollu s k s hell organic matrix components preserved for 80 million years Paleobiology vol. 5 no 2 pp. 144-150. Weiner, S. Talmon, Y., and Traub W., 1983 Electron diffraction of mollu sc shell organic matrices and their relationship to the mineral phase Int e rnational J o urnal of Biological Macromolecules, vol. 5 pp 325-328. Weiner S and Traub W., 19 80, X-ra y diffraction s tud y of the in so luble organic matrix of mollu sk s hells. FEBS L ette rs vol. II, n o 2 pp 311-316. 86

PAGE 99

Weiner S. and Traub, W., 1984 Macromolecules in mollusc shells and their function in biominerali z ation Philos o phical Transa c tions of the Royal Society of London. Seri e s B vol. 304 pp. 425-434. Weiner S., Traub W., and Lowenstam H.A. 1983 Organic matrix in calcified exoskeletons in Biomineralization and Biological Metal Accumulation P. Westbroek and E.W de J o ng eds. D. Reidel Publishing C ompan y. H olla nd. pp 205-224. Wheeler A. P Ru s enko K.W., Swift D.M. and Sikes C.S. 1988 Regulation o f in v i tr o and in vivo cr y stallization b y fractions of oyster shell organic matri x Marine Biology vol. 98, pp. 71-80. Wilbur, K M. 1974 Recent studies of invertebrate mineralization, in The Mechanisms of Minerali z ation in the Invertebrates and Plants, N. Watabe and K.M Wilbur eds. Univer s ity of South Carolina Press Columbia South Carolina pp. 79-108. Wilbur K.M. and Manyak D.M. 1984 Biochemical aspect s of molluscan shell mineralization, in Marin e Biodeterioration: An Interdisciplinar y Stud y J.D C o s tlo w and R .C. Tipper eds ., Naval Institute Press Annapoli s Maryland pp. 30-37. Wind F .H. and Wise Jr., S.W. 1974 Organic vs. inorganic processes in archaeogastropod s hell mineralization in The Mechanisms of Mineralization in the Invertebrates and Plants N Watabe and K.M. Wilbur eds University of South Carolina Press Columbia South Carolina pp. 369-387. 87

PAGE 100

APPENDICES 88

PAGE 101

00 "' Appendix 1. Sample weights, tests run and weight percent. Sample Tests Run Original Powder Weight (g) MF-1 p, aa 80.9 MF-2 CIN, p, aa 11.72 MF-3 a a 15.70 MF-4 p,aa 35.88 MM-1 1 CIN, p aa 101.65 MM-1.2 p,aa 105 34 MM-3.1 191.02 PF-1 p,aa 10. 86 PF-2 CIN, p aa 3.27 PF 3 10.86 PM-I p aa 90.56 PM-2 CIN p, aa 28.99 PM-7 2 1 .70 (Appendix 1. Contin ued on next page) Organics Remaining (g) Weight% .07526 .09% .0627 .5% .089 .57% .073 .2%* .09667 095% .02334 .02% .089 05% .0640 6% .0478 1.46% .05055 .47% .10962 .12% I 10159 .35% 03222 .15%

PAGE 102

\{) 0 Appendix 1. (Continued) Heated Samples PM-3 (wet) a a 17.86 PM-4 (dry) 40.83 PM-5 (dry) a a 22.57 MM-2.1 (wet) a a 149.97 MM-2.2 (dry) aa 160.03 Lost part of sample. CIN-Stable carbon and nitro ge n i so tope analysis p Permeability aa Amino Acid analysis .072 .4 % 005 .01% .031 .14 % 112 .07 % 128 .08 %

PAGE 103

'-0 Appendix 2. Sample ID's description co ll ection tim e, data, and other notes for isotope analysis. ID Description Collection time Date run Nts cJ BSA03 #I pre-peak on I st BSA run I2-I3:20 I0/06 too small no sampl e BSA03 #2 peak of I st BSA run I4:30 -2I :30 I0/06 8 6 -14.79 BSA03 #3 post-peak on 1 s t BSA run 23 24:20 10 / 06 too small -23.4 BSA04 #I pre-peak on 2nd BSA run I2-13:20 10 /06 too s mall 29.5 BSA04 #2 peak of 2nd BSA run 14:30-21:30 10 /06 7.9 -14 .2 BSA04 #3 post-peak on 2nd BSA run 23-24:20 10 / 06 too small -27 Column washed with .1 N HN03 :iso (1 :4) for 40 minut es HPLC broke, unable to run Blank right away nothing run on column in between BSAC03 #I BSA blank 12-13 : 20 10 /31 too s mall -28.7 BSAC03 #2 BSA blank 14:30 -21: 30 10 /3 I too small -27.5 BSAC03 #3 BSA blank 23-24:20 I0/3 1 8.83 -27.3 BSA cent. Unfractionated BSA sample that was 10/ 06 rep 1 7.4 -11.4 I cen trifuged rep 2-7.5 -11.2 BSA uncent Unfractionated BSA sample that was not 10 / 06 rep 1 7.8 -11.2 centrifuged rep 2 lost lost BSA03 #2, BSA04 #2, and BSAC03 #2 collected into rotoevap ballasts rotoevaporated for 15-20 min at 5560 degrees until about I-2 ml s remained They were then tran sferred into quartz tubes. 3 ml s of triple distilled wate r was added to eac h ballast in 1 ml increments and then transferred to th e appropriate quartz tube to ensure that all the sample was recovered. All other samples were collected directly into quartz tube s, all quartz tubes were frozen and lyophilized Note: All samp l es run on HPLC are centrifuged prior to injection (Appendix 2. Continued on next page)

PAGE 104

\0 N Appendix 2. (Continued) ID Description Collection time Date run NIS cu MM1101 # 1 pre 1st peak on 1st MM run 30-1:30 11/12 too small 27 MM1101 #2 1 st peak on 1st MM run 2-5:30 11/12 2.231 -38.77 MM110l #3 post 1st peak and pre 2nd peak on 1st MM 7 -8 11112 too small -28.813 run MM1101 #4 2nd peak on 1st MM run 10:30-16 11112 5.046 -20.69 MM1101 #5 post 2nd peak on 1st MM run 18-19 11/12 le ak-gas lin e -23 99 MM1102 # 1 pre 1st peak on 2nd MM run 30-1:30 11112 too small -26.659 MM1102 #2 1st peak on 2nd MM run 2-5:30 11112 .31 -37 277 MM1102 #3 post 1st peak and pre 2nd peak on 2"0 MM 7-8 11112 too small -28.961 run i MM1102 #4 2nd peak on 2nd MM run 10:30-16 11112 no sample no gas I MM1102 # 5 post 2nd peak on 2nd MM run 18-19 11112 too small -26 .669 Column washe d with .IN HN03:iso ( 1 : 4) for 40 minutes MM11COO # 1 MM blank 30-1:30 11112 .947 -38.586 MM11COO #2 MM blank 2-5:30 11/ 12 7.68 -33 .828 MM11COO #3 MM blank 7-8 11112 27.645 MM11COO #4 MM blank 10:30-16 11/ 12 2.579 -34 122 MM11COO #5 MM blank 18-19 11112 -27.125 MMll unfrac Unfractionated MM samp le (this was 11112 1 .2 1 7 -38 586 centrifuged as were all other samp les MM1101 #2, MM1101#4, MM1102 #2, MM1102 #4, MMliCOO #2, and MM11C00#4 were collected in rotoevap ballasts a n d handled exactly as described above. All other samples collected directly into quartz tubes and procedure s mentioned ab ove were followed. (Appendix 2 Continued on nex t page)

PAGE 105

'-0 w Appendix 2. (Continued) ID Description Collection time Date run NIS cu MF202 #1 pre 1st peak on 1st MF run 30-1:30 11/ 24 too small -27.12 MF202 #2 1st peak on 1st MF run 2-5:30 11/ 24 sample broke MF202 #3 post 1st peak and pre 2nd peak on 1st MF 7-8 11/ 24 too s mall 28.96 run MF202 #4 2nd peak on 1st MF run 10-16 11/24 7.43 -28.37 MF202 #5 post 2nd peak on 1st MF run 18-19 11/ 24 too small -15.34 MF203 #1 pre 1st peak on 2nd MF run 30-1:30 11/24 too small -28.94 MF203 # 2 1 st peak on 2nd MF run 2-5:30 11124 .671 37.49 MF203 #3 post 1st peak and pre 2nd peak on 2nd MF 7-8 11124 too small -27. 54 run MF203 # 4 2nd peak on 2nd MF run 10-16 11/2 4 too s mall -26.68 MF203 #5 post 2nd peak on 2nd MF run 18-19 11/ 24 too small -26 .86 Column washed with .IN HN03: iso (1:4) for 40 minutes MF2COO #1 MF blank 30-1:30 11/24 too small 31.66 MF2COO # 2 MF blank 2 -5:30 11124 too small too s mall MF2COO #3 MF blank 7-8 11124 too small 25.31 MF2COO # 4 MF blank 10-16 11/ 24 sa mple leak 27.194 I MF2COO #5 MF blank 18-19 11/24 too s mall -25.02 MF-2 Unfractionated MF sample (centrifuged as 11/ 24 1.502 -40 .57 we re all other samples run on HPLC) --MF202 #2 MF202 #4, MF203 #2, MF203 #4, MF2COO #2, and MF2COO # 4 were collected in rotoevap ballasts and h and l ed exactly as described above. All other samples collected directly into quartz tubes and pr oce dure s m entio n ed above were followed (Appendix 2. Continued on next page)

PAGE 106

'-0 Appendix 2. (Continued) ID Description Collection time Date run NIS cu PM20l # I pre 1st peak on 1 s t PM run 30-1 :30 12/ 08 too small 27 38 PM201 #2 1st peak on 1 s t PM run 2:30 -5: 30 12/ 08 397 37 87 PM20l #3 post 1st peak a nd pre 2nd pea k on l st P M 7-8 12/ 08 3 -28 62 run PM201 #4 possible 2nd peak (?) on 1st PM run 11-16 1 2 / 08 11.8 74 -28 .58 PM201 # 5 post 2nd peak on 1 s t PM run 18-19 12 / 08 t oo s mall 24 3 1 PM202 #I pre 1st peak o n 2nd PM run 30-1:30 1 2 / 08 too small -26.54 PM2 02 #2 1st peak on 2nd PM run 2:30 5:30 12/ 08 2 1 6 -37 1 66 PM202 #3 post 1st peak and pre 2nd peak on 2nd PM 7 8 12/ 08 too small 29 52 run PM202 #4 possible 2nd peak (?) on 2nd PM run 11-16 12/ 08 9.445 28.62 PM202 #5 post 2nd peak o n 2nd PM run 18-19 12/ 08 too small -24 .31 Column washe d with .IN HN03:iso (1:4) for 40 m i nu tes PM2COO # 1 PM blank 30-1:30 12/ 08 NO PM2COO #2 PM blank 2 : 30-5 : 30 12/ 08 NO PM2COO #3 PM blank 7-8 1 2 / 08 N O PM2COO #4 PM bl ank 11-1 6 12/ 08 NO I PM2COO #5 PM blank 18-19 1 2 / 08 NO r::= unfractionated PM s ampl e (centrifuged as 1.25 were all other samp le s run on HPLC) PM201 #2 PM201 #4, PM202 #2, PM202 # 4 PM2COO #2 and PM2COO #4 were collected in rotoe va p balla s ts a nd h and l e d exactly a s described above. All o ther sa mple s collected directly int o quartz tubes a nd procedures mentioned above were foll owed (Ap pendix 2 Continued on next p age)

PAGE 107

\() Vl Append i x 2 (Con t i n ued) ID Description Co ll e ction tim e D a t e run N I S cu PF20 1 # 1 pr e 1 s t p eak on 1st PF run 30 -1: 30 02117 too small 2 6 62 PF201 #2 1 s t p eak on 1st PF run 2 : 30-5:30 02 / 17 1.125 -38.926 PF20 1 # 3 pos t 1 s t peak a n d pre 2nd p eak on 1 s t PF 7 8 02/17 t oo small -28 8 1 run PF20 1 #4 2 nd peak on 1 s t P F ru n 9-17 02 / 1 7 1 3.39 -31.76 PF201 #5 pos t 2 nd p eak on 1st PF ru n 18-19 02 / 1 7 too small 27. 1 9 9 Co l um n washe d w ith I N HN03:iso ( 1 :4 ) fo r 40 m i n u tes PF202 # 1 pre 1st peak on 2nd PF run 301 : 30 02 / 1 7 -0.681 38 507 I I PF202 # 2 1st peak on 2nd PF ru n 2: 30-5 : 30 02 /17 1.00 38 8 1 I PF20 2 # 3 post 1 st peak and pre 2nd peak on 2nd PF 7-8 02/17 too sm a l l -28 30 I I ru n PF20 2 #4 2nd peak on 2 n d P F run 9-17 02117 11.764 -27 7 PF20 2 # 5 pos t 2 nd p eak o n 2nd PF run 1 8-19 02/17 too sma ll 26 334 C o l umn washed wi th .IN HN03: iso ( 1 :4) for 40 m i nu t e s PF2BOO # 1 PF bl ank 30-1:3 0 02117 NO I PF2BOO #2 P F bl ank 2 : 30 5:30 02/17 ND I PF2BOO # 3 PF bl a nk 7 8 02/17 NO PF2B O O # 4 PF b l ank 91 7 02/17 NO PF 2 B O O # 5 PF bl a nk 18-19 02117 NO P F-2 unfractionated PF s amp l e ( centrifug e d as 2.03 4 2 5 7 were a ll o ther s a mpl e s run o n HPLC ) P F201 #2, PF 201 #4, P F202 # 2 PF 2 02 # 4 PF2BOO # 2 a nd PF 2BOO #4 we re c o ll ec ted i n r o to ev ap b allas t s a nd h andl e d e xac t l y as des cribed abo v e All o th e r s a mp l es c olle cted dire c t l y i nt o qu a rt z t u be s and proc e dur es m entio n e d a b ove w e r e f ollowe d *N ote: Vacuu m p ump on r o t oeva p wa s brok e n u s ed s ink v a c uum T h i s inc r e ase d tim e o f eva p o rati o n t o 40-4 5 minute s.

PAGE 108

Appendix 3. Amino Acid Abbreviations asp aspartic acid thr threonine ser-senne glu g lutamic acid pro proline gly-glycine ala alanine val v aline met methionin e il e isoleucine leu leucine t y r -t y rosine phe phen y lalanine hi s histidine lys-lysine arg-argmme 96

PAGE 109

\0 -.....J Appendix 4 All amino acid data in nanomoles Pl\1-101 Ill Pl\1101 112 Pl\1500 Ill Amino Acid Cone. in nmo l Cone. in nmol Cone in nmol asp 0 569497 0.2694 thr 0 5872 0 .129099 0 1 073 scr 1 0542 0.225068 0.4994 glu 0 8517 0.319589 0.3856 pro 0 505875 0 0651 gly 1.4058 0 .631136 1.9297 ala 0 .376 0 1 538 1 9 0 0 128 val 0 197029 0 0603 met ilc 0.151231 0 0138 leu 0 134 0 207861 0 0783 tyr 0 0165 phc 0 3968 0 212767 0 0268 his 7 6872 11.9691 lys 0 3312 0 326887 0 6033 arg 0 1213 56 0.0339 Total 12. 8241 3 .751 214 1 6 0713 (Appendix 4. Continued on nex t page) Pl\1500 112 P l\ 1500 113 Pl\1500 114 PM200 Ill Pl\1200 112 Cone in nmol Cone. in nmol Cone. in nmo l Cone. i n nmol Cone. innmol 0 0728 0 0169 0 791 0.2974 1 40. 97 7941 0 03 48. 760359 13.684425 0 0268 0 016 0.23 67 130727 15.200 874 0 1223 0 2785 60 205762 22 642832 0 085 35 585238 2.3386 0 1 23 6 0 77 127 .396211 30. 930588 0 0 1 67 0 6955 73 6 1 2141 20 .73595 I 0 022 74 66218 23 768 96 3 0 0152 0 017 5 0.0776 17 389258 2.44449 1 0 02 0 0 1 25 0 0194 48 664883 12. 624332 0 0136 0 3323 200 1 42922 1 6 42 50 1 6 23 616523 2.54 1 674 0 1686 0 .145 0 2084 58.663746 14.49735 5 0 2 1 77 0 .2 1 9 7 0 2548 9 .110392 3 .876907 0 5949 0 3118 0 4257 53 793359 20 .82 9002 0 0154 0 0 1 53 37. 277039 11.3 7967 3 6292 0 87 8 4 4 2135 900. 7229 1 2 288 145259

PAGE 110

1.0 00 Appendix 4 (Con tinued) I Pl\1102 # I Pl\1-102 #2 PM-1 02 #3 !Amino acid Cone in nmo l Co ne. in nmo l Cone i n nmo l as p 0 70 3 2 0833 0 5482 thr 0 .1193 0.5917 0 0597 scr 0 5409 1 6324 0.4804 glu 0 3445 1.1795 0.1347 pro 0.0793 g l y 1 3082 2.6253 1 0233 a la 0 3757 0.6328 0 328 \ 8 1 m e t 0 .0 1 47 ile 0.448 0 0529 leu 0 0884 0 7111 0 1 603 t y r phe 0.4292 0.44 1 9 0 3785 his 3 .911 2 1.8183 1.44 1 6 lys 0 1326 0.1 476 arg 0.0546 Tot a l 7 953 1 2.38 1 2 4 6869 -(Appendi x 4 C on tinu ed on next page) Pl\1102 #4 Pl\1300 # I Pl\1300 # 2 Pl\1300#3 Cone in nmol Co n e i n nmo l Cone in nmo l Co n e in nmo l 1.1569 0 1168 1 100084 3.225163 0.4 1 24 0 05605 0 36024 3 0.097976 0.4803 0.137843 1.2725 5 1 0.73477 0.6407 0 205977 0.608966 0.40445 0 .141 666 4 3 876 7 1 0 36096 1 0.9578 0.436224 4.4 5 8751 2 263143 0.3626 0 03 3 868 2 802726 0.12779 0 0559 1 3 1 .40940 3 0.115 48 5 0 0678 0 036338 0 1 2 5721 0.096425 0.088 1 0.058474 0.26087 0 226324 0 281999 0.3966 0.065443 0 098 1 72 0 .2211 0.726642 0 .0578 0.036782 0.29 4924 0.1 932 14 0.5708 0 1 962 2 7 0 065523 5.4 1 29 1.59 7 934 18. 0 70 2 22 8 009396 ---

PAGE 111

\0 \0 Appendix 4. (Continued) PF200 # I PF200 #2 PF200 # 3 Amino Acid Cone i n nmol Cone. in nmol Co ne in nmol a s x 0.203935 0 1 05178 0 064 1 53 thr 0.7 1 3846 0.056229 0 121336 ser 1 030924 0 1 51037 0 086697 gb: 0.67253 0.196126 0 1 7422 pro 4 58257 1 0 .53 9008 0 13773 g l y 4 458263 0.554366 1.196572 a l a 3.608472 0.047379 0.009 4 5 val 2 501663 0 .038282 0 042815 m et ile 0.25 1 64 0.035142 leu 0 36256 0.055238 0.04204 1 t yr phe 0.20058 0.028168 his 0 675257 lys 0 3 1 6551 0 028731 0 019973 arg 0 264885 Total 19. 843677 1 .834884 1.894987 (Appendix 4 Continu ed on n ex t page) PF200 #4 l'FIOO # I PFIOO #2 PFIOO #3 PFIOO #4 Cone in nmo l Co n e in nmol Co ne in nmol Cone. in nmo l Cone in nrnol 0.643251 0.382357 0 052158 0.160787 0 239123 0.22345 1 0 22 1 214 0 032657 0.1137 1 3 0 1 94089 0.324109 0 356593 0.057812 0.2979 1 3 0.454579 1.171399 0.498126 0.075775 0.367074 0.305055 0.57595 1 0 334629 0.057061 4.615205 0.281149 1.126406 0 835766 0.17151 2.138785 1.06 1 179 0.283 1 08 0 285446 0.01400 1 0.837019 0 1 9 1 645 0 .349764 0.295647 0.019555 0 5 1 0 1 49 0 187811 0.226045 0.1 5 1 231 0.055398 0 1 35025 0.430987 0.275405 0.05111 0.078309 0 245215 0.15539 0 .148516 0.016336 0.117363 0 239 1 66 0.388027 0 2 12089 0.10673 5 0.220712 0 02474 0.094042 0 150683 0 07699 0.082615 0 0 1 4592 0 06119 5.693586 4 327423 0.587307 9.65642 1 3 836195

PAGE 112

0 0 Appendix 4. (Continued) 1\ll\1-11 00 1\11\11100 1\11\1-1100 Ill 112 113 #4 Amino Co n e. in Con e. in Cone. i n Con e in acid nmol nmol nm o l nmol ASp 1 9.3026 2 1431 thr 1 4.4507 1.3594 16 .300 8 0 4717 ser 6 0928 52.9262 0 9381 g lu 21.5653 2 0136 26.8202 0.6779 pro 27. 1 976 3 0004 28 .306 0.5621 g 1 y 88. 7172 9.2103 64.469 1.1707 ala 16. 1684 1.6 15.862 0 5465 val 1 1.7584 0.6418 14.6837 met 1.2971 i l e 1 2 5273 1 .0656 14. 7443 0.2667 l eu 1 7 2701 1 .6 085 14. 8385 0.3525 t y r 0 6079 0 2261 phe 7 6513 0.4643 9 1975 0.4027 h is 3 2394 0.6086 2 782 1 0 2865 l ys 20 80 48 2.7859 22.4374 0 0188 arg 1 0 3968 0.6 481 11.4 3 1 6 0 0399 T o ta l 252.3552 5 0.4 019 296.3225 7.8772 (Append i x 4 Continued on next page) 1\11\11200 # I Con e in nmol 11. 7053 1 2908 4.2403 2 .1456 1.69 1 2 1 92 4 9 1 877 0 6267 0 8483 1.1338 0 6 4 85 3.4686 1.9021 0.431 53.9339 1\11\12100 1\11\12100 113 114 l\li\121o o 1 #2 # 3 # I 112 115 Co n e in Cone. i n Cone. in Cone in Cone in nmol Cone. in Co ne in nmo l nmol nmol nmol nmol nmol 2 .616 0.4796 2 934207 91.195906 55.433895 0 .52 9558 0 4052 0 0235 0.523072 88.392992 23.94038 1 0 29 1 9 1 .0848 0 158 5 1 .607143 0 56976 1 67 194742 13.5 7444 0.17 1 256 0.67 0.0841 1 .24685 0.319 4 52 1 84.657234 26.54985 0.27026 0 .1187 3 718426 0.253224 1 41.639313 39.075637 0.42624 1 .7 426 0.2299 4 .638758 2 134956 129 554484 32. 299227 0.464505 0 6568 0.852985 0 202062 95 471438 21.912373 0.242359 0.152834 0 .111 565 122.313781 25.025219 0.180482 19.7 82953 3 .5 75889 0 1 811 0.0361 0 110444 0 .0 805 1 2 60.76607 12. 60256 0 129303 0 .3257 0.042 1 0.250368 0 21139 76 053062 16. 522166 0 201201 35.349133 1.94269 0.7242 0 6596 0 115551 89.324852 17. 222729 0.234215 1.80 1 6 0.22 93 0.528039 1 0 2 4 8903 2 699469 0 09 36 0.6753 0.275483 0.179263 57.4 1 743 19.216924 0.346018 0.052 1 0 4723 2.618 13.904402 7.111 943 1 269.362293 311 .593449 3.487297 ----

PAGE 113

Appendix 4. (Cont inued) 1\11\12200 #I 1\11\12200 #2 1\11\12200 #3 1\11\12200 # 4 1\11\12200 #5 Ml\12200 #6 A m ino Acid Cone in nmol Cone in nmol Co n e in nmol Cone. in nmo l Co n e in nmol Co n e in nmol asp 3.1346 20.2354 94.4556 0.4849 19. 5 3 03 4 .361 3 thr 0.0976 0.7303 7 5446 0 .0436 4 06 1 2 0 6806 ser 1.1861 2.3175 16. 702 1 0 1 211 4 9332 1.2516 glu 0 6636 0 5 154 8.249 0.0841 3 6473 0 7896 pro 1.71 58 48 9 1 37 1 7 1825 2.503 g l y 9 7268 1 6 7954 84 5472 0.294 1 8 90 1 6 3.0738 ala 0.5189 1 0 1 696 65 116 7 0.1 933 17. 8042 2 2915 \'31 8.5775 3 8549 m e t 0 0773 0 2337 2.9953 1.4553 0.1605 ile 0.036 1 0.5006 7.5738 2 7366 0.4696 0 leu 0.1603 3.4686 1 7 0925 0.0507 5.5975 0.8632 tyr 1.389 19. 7576 4 .918 0.5665 phe 0 154 1 0 323 15. 7802 0.2145 0.2098 his 5 866 1 0.0645 1 953 0.2121 1 .4038 0.0966 lys 0.33 4.6584 1 7 676 0 382 6 284 1 4572 arg 0 0455 0.043 1 0 2014 4 .04 74 0 0214 Total 21.9969 63 8695 427 1 362 2.0803 1 1 6 3 5 78 18. 7962 (Appendix 4. Continued on next page)

PAGE 114

0 N Appendi x 4 (Continued) I\IF201 # I I\IF201 112 I\1F201 113 A min o Acid Cone. in nmol C o ne. in nmo l Co n e i n nmol a s p 0 307 thr 0.3348 0 1 853 ser 0 983 0.2709 glu 0 764 1 0.2349 p ro g l y 2 3288 0 2024 0.4945 ala 0 5 2 3 1 0 .2 44 \'a I met ile 0 0949 0 0497 leu 0 1 764 0 1651 tyr p h c 0 .896 6 0 .7 433 1 0263 his 3 5586 0.2218 0 1 685 lys 0 0629 0 03 1 7 arg 0 0408 To t a l 9 .7 6 4 1.1675 3 1 779 (Appendi x 4. Continued on next page) I\1F201 # 4 Co n e in nmo l 0 1 06 1 0.0618 0.0841 0. 1 7 1 5 0 0375 0 .73 1 0.206 1 0 .575 4 0.0234 1.9969 -1\IFIOO # I 1\IFIOO 112 1\I FIOOIIJ 1\I FIOO 114 1\IFIOO 115 1\IFIOO#G Cone. in nmo l Co n e in nmol Co n e in nmo l Cone in nmo l Cone innm o l Co n e. in nmol 0.244 1 0 050 1 0.055 8 0 444 0 3303 0 0647 0.0856 0.1 697 0.24 1 0 3013 0 0323 0 09 1 2 0.2883 0 5035 0 0826 0 4837 0.0889 0 .076 6 0 1 8 1 5 0.3323 0 6046 I 1.8911 0 1 534 0 2 1 72 0 3 1 29 0.6385 4 1 782 0 .9 4 24 0 .0386 0 3 1 3 1 0.1094 0 1073 0 0226 0 0345 0 0672 0.2398 0 0624 0.7184 0.6605 1 0874 0 8 594 0.9617 0 7095 1.944 1 0.1 048 0.1938 0 .1854 0.1 96 0 182 1 0 693 0 0144 0.5995 0.5757 0 .65 8 0 04 1 9 0 0 16 9 0 0 2 29 0.020 4 7.411 1.127 2.3729 3. 1 227 4 5465 5.9464 L__ __ -----------------

PAGE 115

0 VJ Appe ndi x 4 (Con tinued) 1\IFJOO # I 1\IFJOO # 2 A m i n o Acid Cone. i n nmo l Cone. in nmol a s p 1.6386 thr 0.6 439 ser 1.1731 g l u 1 .3936 pr o 0 4 2 49 g l y 3.3 4 87 ala 1.1167 v a l 0.4225 ... ,. m e t i l e 0.263 leu 0 .53 68 tyr phe 0.5998 0 4 5 6 3 h i s 5.4832 0 1588 lys 0.5385 0.503 1 a rg T o t a l 1 7 .5 833 1.1 1 82 I\IFJOO # 3 Cone. in nmol 0 6995 0 1 33 7 0.56 1.3932 --I\IFJOO # 4 I\I F 4 0 3 # I I\IF403 #2 I\I F403 # 3 I\I F403 # 4 1\IF-10 3 # 5 Cone. i n nmol Cone. in n n10l Co ne. i n n m o l Cone. in nn10l Cone. i n nmol Cone. in n m o l 0 .5 402 0.037605 0.58964 0.1 46343 0 .3303 0.816289 0.037882 0.084574 0.095606 0 .4235 1.0155 46 0 1 44 824 0.061829 0.302458 0.1 57423 0.5076 0.207389 0.1 24843 1.175786 0.238414 3.527084 0.6669 1 7 0 1 4338 1 0.448685 0.206 1 95 0.5864 4 581383 0.4 54576 0.9 1 7849 0.909632 0.380283 0.5024 1.08078 0.096136 0.078098 0.382094 0.148334 0 0895 0.299295 0.3 17981 0.007526 0.06 1 1 0.1 9 33 67 0 201668 0.040358 0 2457 0.3 1 7703 0.059823 0 0387 1 8 0.42888 0.080728 0 6 1 92 0.6 1 9797 0.071575 0.211 321 0.0 1 203 1 0 2296 0 .082 834 0.4541 0.273379 0.04 1 4 0.1 1 5 1 87 0.055 1 42 4.5896 1 3.05245 1 1.573 133 1.449292 5. 1 78938 1.472777 ---