USF Libraries
USF Digital Collections

Evolution of a conserved gene regulatory network among echinoderms


Material Information

Evolution of a conserved gene regulatory network among echinoderms a comparison of genes expressed in the skeletogenetic lineage of the ophuroid Ophiocoma wendtii and the echinoid Strongylocentrotus purpuratus
Physical Description:
Ruzek, Mitch James
University of South Florida
Place of Publication:
Tampa, Fla
Publication Date:


Subjects / Keywords:
Brittle star
Dissertations, Academic -- Biology -- Masters -- USF   ( lcsh )
non-fiction   ( marcgt )


ABSTRACT: One of the most fundamental and critical functions of embryological development is the control and regulation of differential genes and gene networks. The study of the gene networks involved in development is a mechanism for understanding the developmental process at its most basic level. An evolutionary change in a morphological feature or features must depend on a reorganization or co-option of one or more developmental gene regulatory network just as retention of an ancestral morphological trait must rely on retention of a common gene regulatory network. Studying two closely related classes in the same phylum with the same essential morphological feature yet with unique developmental characteristics provides insight into the evolution of these evolutionarily resolute gene regulatory networks. We have developed a new model system using brittle stars to further these studies. In this investigation I have identified key genes of the gene regulatory network (GRN) found in embryonic endo-mesoderm development in the sea urchin, responsible for embryonic skeletogenesis, and compared these key genes with homologues in the brittle star. From the examination of two closely related gene regulatory networks found in two related classes of Echinoderms insight can be gained into the foundation of morphological change over time.
Thesis (M.S.)--University of South Florida, 2009.
Includes bibliographical references.
System Details:
Mode of access: World Wide Web.
System Details:
System requirements: World Wide Web browser and PDF reader.
Statement of Responsibility:
by Mitch James Ruzek.
General Note:
Title from PDF of title page.
General Note:
Document formatted into pages; contains 85 pages.

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 002069287
oclc - 608096051
usfldc doi - E14-SFE0003181
usfldc handle - e14.3181
System ID:

This item is only available as the following downloads:

Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam 2200361Ka 4500
controlfield tag 001 002069287
005 20100420155937.0
007 cr bnu|||uuuuu
008 100420s2009 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0003181
QH307.2 (Online)
1 100
Ruzek, Mitch James.
0 245
Evolution of a conserved gene regulatory network among echinoderms :
b a comparison of genes expressed in the skeletogenetic lineage of the ophuroid Ophiocoma wendtii and the echinoid Strongylocentrotus purpuratus
h [electronic resource] /
by Mitch James Ruzek.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 85 pages.
Thesis (M.S.)--University of South Florida, 2009.
Includes bibliographical references.
Text (Electronic thesis) in PDF format.
3 520
ABSTRACT: One of the most fundamental and critical functions of embryological development is the control and regulation of differential genes and gene networks. The study of the gene networks involved in development is a mechanism for understanding the developmental process at its most basic level. An evolutionary change in a morphological feature or features must depend on a reorganization or co-option of one or more developmental gene regulatory network just as retention of an ancestral morphological trait must rely on retention of a common gene regulatory network. Studying two closely related classes in the same phylum with the same essential morphological feature yet with unique developmental characteristics provides insight into the evolution of these evolutionarily resolute gene regulatory networks. We have developed a new model system using brittle stars to further these studies. In this investigation I have identified key genes of the gene regulatory network (GRN) found in embryonic endo-mesoderm development in the sea urchin, responsible for embryonic skeletogenesis, and compared these key genes with homologues in the brittle star. From the examination of two closely related gene regulatory networks found in two related classes of Echinoderms insight can be gained into the foundation of morphological change over time.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: James Garey, Ph.D.
Brittle star
Dissertations, Academic
x Biology
t USF Electronic Theses and Dissertations.
4 856


Evolution of a Conserved Gene Regul atory Network Among Echinoderms: A Comparison of Genes Expressed in the Sk eletogenetic Lineage of the Ophuroid Ophiocoma wendtii and the Echinoid Strongylocentrotus purpuratus by Mitchel James Ruzek A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science Department of Cell Biology, Micr obiology and Molecular Biology College of Arts and Sciences University of South Florida Major Professor: James Garey, Ph.D. Brian Livingston, Ph.D. Jessica Moore, Ph.D. Kristina Schmidt, Ph.D. Date of Approval: September 18, 2009 Keywords: development, skeletogenesis, brittle star Copyright 2009, Mitchel James Ruzek


Acknowledgments I would like to say a special thanks to Dr. Brian Livingston for his wise mentorship and constant support through out this entire process. Without his measured demeanor and friendship this journey would not have been possible from the start and more importantly to the end. I would like to acknowledge the importa nt contribution that the scientists and staff of the Keys Marine La b provide to not just myself and our lab but to the state of Florida and the entire globa l scientific community. The only thing better than having one of the worlds most diverse and rare ecosystems in your backyard is to have a staff and group of great scientists there to help you ma nage and access all that the Florida keys have to offer. Thank you KML and I will continue to rely on you for scientific support, dive advice and tidal updates for research and pleasure. I would like to thank Mary Harmon whose work proved so va luable to me through out my research. Much of the work done with the Alx1 gene was based on initial i nvestigation started by Mary. I would also like to thank Travis Carter and Rachel Vistein who were both undergraduate students in the lab. It was of ten their dedication and enthusiasm that helped to keep things running. In additi on they were eager students with not only inquisitive minds but also a keen understa nding of molecular biology and a valuable source of insight and collaboration through out their time in the lab. Lastly, I would like to thank the great facu lty and staff of the University of South Florida CMMB department


with special consideration to those who so kindly guided my progr ess by serving on my graduate committee; Dr. James Garey, Dr. Je ssica Moore and Dr. Kristina Schmidt.


i Table of Contents List of Figures ............................................................................................................... ..... iii Abstract ..................................................................................................................... ..........v Introduction .................................................................................................................. ........1 The Genetic Basis of Development .........................................................................5 Evolution of Development .......................................................................................6 Sea Urchin Development .........................................................................................8 General Development ..................................................................................8 Skeletal Development ................................................................................13 Hypothesis..............................................................................................................15 Research Design ............................................................................................................... ..17 Animal Husbandry / Spawning ..............................................................................17 Animal Collection ......................................................................................18 Location .........................................................................................18 Time / Date of Dives ......................................................................20 Collection Protocols .......................................................................21 Transport to Tampa ........................................................................24 Storage and Care of Brittle Stars in Tampa ...................................24 Spawning of O. wendtii ..................................................................25 Keys Marine Lab Spawning ..............................................25 USF Tampa O. wendtii Spawning .....................................27 Fixation of Embryos ......................................................................27 Gene Regulatory Network and Molecular .............................................................28 Isolation of a Fragment of O. wendtii Dri cDNA ......................................28 Design of Degenerate Primers ...................................................................29 Alignment of Sea Urchin Dri Amino Acid Sequence With Those From Other Organisms to Determine Conserved Regions ....................................................................................29 Synthesis of O. wendtii cDNA .......................................................31 PCR Including Optimization and Gels of Degenerate Primer PCR ..............................................................................31 Purifying Degenerate Primer PCR Products ..................................32 Crude Mini Plasmid Prepar ations of Degenerate Dri PCR Primer Products ........................................................................33 Purified Plasmid Preparation of Degenerate Primer Dri PCR Products ...........................................................................34


ii Sequencing of Degenerate Dri Primer PCR Products ....................34 Obtaining Full Length cDNA Sequences of Dri and Alx1 ........................35 RLM-R.A.C.E. ...............................................................................35 Non-Degenerate PCR Pr imer Design and Generation ...................37 Empirically Derived Primer Combinations and PCR Protocols ......................................................................40 Ligation and Transformation of PCR Products .............................41 Crude Mini Preparati ons of RACE PCR Products ........................42 Digest of Crude Mini Prep arations of RACE PCR Products .........42 Perfect Plasmid Preparations of RACE PCR Products ..................42 Sequencing of RACE Generated PCR Products ............................42 Determining Spatial and Te mporal Expression Patterns of O. wendtii Skeletal Genes via Megasc ript RNA Probe Generation and Whole Mount In Situ Hybridization .............................................43 Megascript Transcript ion Reactions and RNA Probe Generation ................................................................................43 Southern Blotting ...........................................................................44 Whole Mount In Situ Hybridization ..............................................47 Results ....................................................................................................................... .........51 Animal Husbandry / Spawning ..............................................................................51 Animal Collection ......................................................................................51 O. wendtii Spawning, (Keys Marine Lab, Layton, FL) .............................53 O. wendtii Spawning, (U.S.F. Lab, Tampa, FL) ........................................55 O. wendtii Embryonic Development ..........................................................57 Gene Regulatory Network / Molecular ..................................................................58 cDNA Synthesis and PCR Control ............................................................58 PCR Using Degenerate Dri Primers ..........................................................59 Sequence Analysis of Putative Dri Clones ................................................62 Actin Controls for 5’ RACE PCR .............................................................63 Dri 3’ RACE ..............................................................................................64 Assembling of Dri Sequences Alignments of S. purpuratus .....................65 5’ RACE of Alx1 .......................................................................................67 Alx1 Assembly and Alignment With S. purpuratus Alx1 .........................68 Dig-Labled Probe Preparation a nd Sensitivity Assay via Dot Blot ...........69 Whole Mount In Situ Hybr idization Using Alx1 Probes ...........................71 Temporal Expression During Development Using SemiQuantitative PCR for Alx1...................................................................72 Dri Whole Mount In Situ Hybridization ....................................................74 Phylogenetic Analysis ................................................................................74 Conclusions ................................................................................................................... .....77 References .................................................................................................................... ......81


iii List of Figures Figure 1 Animal Collection Site ......................................................................................18 Figure 2 Debris Removal / Animal Location ...................................................................20 Figure 3 Bongo Device ....................................................................................................21 Figure 4 Ziploc O. wendtii Collection .............................................................................22 Figure 5 Keys Marine Lab Animal Storage .....................................................................23 Figure 6 Dri degenerate primer design ............................................................................30 Figure 7 RACE cDNA confirmation + control procedure ...............................................36 Figure 8 RNA Probe Generation .....................................................................................44 Figure 9 The number of animals obtained pe r diver from a repr esentative series of dives ..............................................................................................................51 Figure 10 Spawning me thod success rates ........................................................................55 Figure 11 Sea urchin and brittle star developmental photos ..............................................57 Figure 12 + control gel of O. wendtii cDNA using actin primers .....................................59 Figure 13 Result of Degenerate primer PCR .....................................................................60 Figure 14 Crude mini plasmid preparat ion of Dri degenerate primer PCR .......................61 Figure 15 Perfect plasmid preparation of Dri degenerate primer PCR .............................62 Figure 16 Dri perfect primer design sequence ...................................................................63 Figure 17 + control for RACE treated and ligated cDNA .................................................64 Figure 18 Results of 3’ RACE extension of Dri sequence ................................................65 Figure 19 Dri 256 Ami no Acid alignment .........................................................................66


iv Figure 20 5’ RACE amplification of Alx1 5’ end .............................................................68 Figure 21 Complete Alx1 gene sequence alignment .........................................................69 Figure 22 Directiona l Plasmid Digest ................................................................................70 Figure 23 Dri Megascript pr obe activity quantification ....................................................71 Figure 24 Whole Mount In Situ Hybridization Alx1 expression .......................................72 Figure 25 Semi-quantitative PCR O. wendtii Alx1 ...........................................................72 Figure 26 Alx1 neighbor joining tree us ing complete Alx1 O.W. sequence .....................74 Figure 27 Dri neighbor joining tree using 1053bp Dri O.W. sequence .............................75 Figure 28 Sea urchin and brittle star endomesodermal development comparison ............78 Figure 29 Sea urchin and brit tle star skeletal gene regul atory network comparison .........78


v Evolution of a Conserved Gene Regulatory Network Among Echinoderms; a Comparison of Genes expressed in the Skeletogenetic Lineage of the Ophuroid Ophiocoma Wendtii and the Echinoid Strongylocentrotus Purpuratus Mitch Ruzek ABSTRACT One of the most fundamental and critic al functions of embryological development is the control and regulation of differentia l genes and gene networks. The study of the gene networks involved in development is a mechanism for understanding the developmental process at its most basic level. An evolutionary change in a morphological feature or features must depend on a reor ganization or co-option of one or more developmental gene regulatory network just as retention of an ancestral morphological trait must rely on retention of a common gene regulatory network. Studying two closely related classes in the same phyl um with the same essential morphological feature yet with unique developmental characteristics provide s insight into the evolution of these evolutionarily resolute gene regulatory networks. We have developed a new model system using brittle stars to further these stud ies. In this investigation I have identified key genes of the gene regulatory networ k (GRN) found in embryonic endo-mesoderm development in the sea urchin, responsible for embryonic skeletogenesis, and compared


vi these key genes with homologues in the brittle star. From th e examination of two closely related gene regulatory networks found in tw o related classes of Echinoderms insight can be gained into the foundation of morphological change over time.


1 Introduction Since the time of Aristotle, Echinoderms have proven fascinating to biologists. While Aristotle’s description of a sea urch in’s bony mouth structure as resembling a lantern will forever link his early curious mind and name to the study of the phylum, this observational knowledge has been added to gr eatly in modern times. Almost a century ago Theodor Boveri used sea urchin embryos to determine that an embryo’s development is controlled by the genes contained in the embryonic cells following fertilization (Boveri, 1918). The process of fe rtilization still progresses as humbly as it did then. One egg surrounded by multiple motile sperm cells in any given volume of seawater. Each male gamete vies for the chance to pass its own version of genetic history to a future generation. Eventually fate in the form of random chance deems one sperm successful and egg and sperm fuse. This union becomes apparent almost immediately as a thin transparent line begins to rise from th e surface of the zygote. This union and the fertilization envelope produced to prevent polyspermy are readily visible with light microscopy. This outward sign of sperm and e gg union is only a sma ll indicator of the myriad of processes that have begun and w ill continue on the inside of what will eventually become a fully functioning multi cellular organism. One single cell, the fertilized egg, beco mes two, two becomes four and so on as mitosis resumes. Eventually, cells begin to specialize and tissues be gin to form. The one fertilized egg cell gives rise to hundreds of cell types that will eventually have many


2 varied cellular functions and will self organi ze to form what will become the adult body plan. For this to happen it must mean that co ntained within the egg cell are both the initial conditions and the program that processes thes e initial conditions ev entually translating them into specification and differentiation. The initial conditions w ithin the egg cell are comprised of particular prot eins or mRNAs. The hardware and software that guide development stems from the genome (Dav idson, 2007). For the sea urchin, this specialization is firs t apparent at the fourth cell division. What was once a ball of inconspicuous visibly identical cells now begins to specialize and develop structures that will play crucial roles as th e organism continues through the process of growth and development. The instruction required to process all the information provided by the maternal mRNA and proteins and transform thei r influence into differential zygotic gene expression is contained in the genome. Becau se the egg, its predecessors and its cellular descendants contain the same genome, we should see the same pattern of gene expression repeated in all cells. However since this is not the case the question arises as to how does a similar static code enable the dynamic differential gene expression, what program regulates this expression and how does the dynamic nature of gene regulation lead to evolution? The regulatory apparatus is com posed of regulatory ge nes and transcription factors that bind to specific sequences of DNA to activate or repress transcription of a gene. The cells ability to control the gene e xpression of thousands of genes is of paramount importance and a fundamental task of the developing sea urchin. It is the systems of regulation that ove rsee this specialization and differentiation of cells and tissues that are specifically intriguing to us. The network of genes responsible for


3 embryonic skeletal formation within the phylum Echinodermata is of particular interest. Studying gene regulatory networks enable s the understanding of the mechanism underlying the developmental process at the most fundamental level (Davidson et. al., 2006). Echinoderms, as a Phylum, form a highly developed and well-defined group of Metazoans with a well-pre served fossil record. Echinoderms, like Hemichordata, Urochordata, Chephalochordata and Verteb rata, are deuterostomes. Echinoderms are subdivided into five classes with at leas t 7,000 extant species. The five classes of Echinoderm are, Crinoidea (sea lilies and f eather stars), Ophiuroi dea (basket stars and brittle stars), Echinoidea (sea urchins, sa nd dollars, and sea biscuits), Asteroidea (starfishes), and Holothuroidea (sea cu cumbers). All members of the phylum Echinodermata contain a skelet on in the adult organism. The class Echinodea is composed of s ea urchins along with sand dollars, sea biscuits and heart urchins. Sea biscuits and heart urchins are grouped as irregular urchins. The more globular and radially symmetrically shap ed sea urchin is referred to as a regular urchin (Hyman 1955). A typical adult echinoid contains 5 gonads arranged radially and attached to the interior side of the test (e ndoskeleton shell). The rema ining majority of an adult echinoid is composed of skeleton a nd coelomic fluid. When in season the gonads swell to a size that almost fills the entire coelom (Nichols 1969, Hyman 1955). The test in echinoderms is composed of 5 skeletal plat es (Hyman 1955). Possibl y the most readily observable skeletal component of ecinoids is a jaw known as Aristotle’s lantern. This lantern creates their masticatory apparatus and contains 5 persistently growing teeth that are used to rasp both plant and animal matter (Nichols 1969, Birkland 1988). Sea urchins


4 can be found throughout the world’s oceans from tidal pools up to depths of about 5000 meters (Hyman 1955). The class Ophiuroidea is one of the la rgest groups of Echinoderms containing approximately 1800 living species. Only the clas s Asteroidea contains more species. The class Ophiuroidea contains the brittle stars (Ophiuroida) and the bask et stars (Euryalida). Brittle stars usually have 5 arms and exhibi t 5-segment pentaradial symmetry. Unlike the Asteroidea (true starfish) brit tle stars have 5 long flexible arms. Where starfish crawl on hundreds of tube feet, brittle stars move fair ly rapidly by movements of their flexible arms. An internal skeleton composed of calci um carbonate plates supports the arms of the brittle star. The basket star is similar st ructurally to the brittle star. The main distinguishing feature is the highly forked and branched arms of the basket star. Ophiuroids can be found in most of the wo rld’s oceans and a few adaptable species can live in brackish waters. Ophiuroids are common in shallow water marine habitats as well as deep ocean beds. Ophiuroids can be scav engers by preying on sm all live animals such as crustaceans and worms. The basket stars in particular are adept filter feeders gathering plankton with their highly branched and web like arm structures. Ophiuroids and Echinoids are the only members of the Phylum Echinodermata to posses a pluteus, bilaterally symmetrical la rvae with a well-developed larval skeleton. The skeleton of Ophiuroidea and Echinoide a like all Echinoderms is composed of calcium carbonate. The embryonic skeleton uni que to Echinoidea and Ophiuroidea and lacking in other Echinoderms appears to be formed by embryonic activation of genes that produce an adult skeleton in the othe r Echinoderms (Killian and Wilt 1996). Two mutually exclusive explanations are possibl e. 1) There could have been one common


5 ancestor to both brittle stars and sea urch ins that contained this change in the skeletogenesis GRN. Or 2) Br ittle stars and sea urchins could have evolved larval skeletons independently from one another. The Genetic Basis of Development Development begins from the time a hapl oid sperm fuses with a haploid egg to produce a diploid zygote at fertilization. From this single fused ce ll until adulthood there is a shared common progression among most an imals. Shortly following fertilization, cleavage begins to occur and cell division resumes. During initial cleavage the rapid succession of cell divisions greatly reduces the cytoplasmic volume of the egg as it divides into numerous individua l blastomeres. The blastomeres form what is collectively known as a blastula by the end of initial cleava ge. As the mitotic divisions begin to slow in frequency a rearrangement of the cells begins called gastrula tion. The embryo moves from blastula to gastrula as the cells rear range into their three separate germ layers, ectoderm, endoderm and mesoderm. Once the cells have sorted themselves out and germ layers are defined organogenesi s begins to proceed. In many organisms it is at this time when specialized germ cells begin to form at this time for future genetic inheritance. The process of meiotic germ cell formation is known as gametogenesis. From a single incomplete sperm and egg cell a genetically un ique and complete organism is about to emerge. The event that specifically demarcates th e progression from fe rtilization to first cleavage is the activatio n of mitosis promoting factor (M PF). MPF is responsible for the


6 resumption of mitotic activity in a fertilized ovum. MPF continues to play a regulatory role by overseeing the biphasic cell cycle of ear ly blastomeres as they pass from M phase to S phase (Gerhart et al., 1984) (Newport a nd Kirschner, 1984). The regulators of MPF are maternal in origin and stored in the cytoplasm of the ovum. As the cytoplasm is depleted during the initial ra pid synchronous rounds of cell division these regulators are also depleted. At this point the embryo enters into the mid-blastula stage and the gap (G1 and G2) phases are added to the cell division s, completing the cell cycle. Then mRNA’s are transcribed that will encode proteins to aid in gastrulation. The embryo must develop what will eventually be its three axis early in development. The blueprint for Anterior – Posterior, Dorsal – Ventral and left – right axis is initiated during gastrulation as the cells of the blastula are rearranged. The gastrula stage gives cells new arrangements from that which they had established during early cleav age. Gastrulation mark s the point where the multilayered body plan of an organism is established. The Ectodermal layer covers what will be the internal structures created from endodermal and mesodermal layers now deep to the ectoderm. It is at this stage of de velopment that a divergence of developmental paths and plans are enacted dictated by the specifics of the orga nism of interest. Evolution of Development As development proceeds the gene networks en sure that an organism is essentially the same as its predecessors. In contrast, evoluti onary change in morphological features of an organism must be driven by a reorganizati on or modification of the very same gene regulatory networks. As genetic change occu rs, essential gene regulatory networks can


7 also change and evolve. Howeve r, when looking at two closely related species with in the same phylum it is expected that there shoul d be some recognizable similarity between homologous gene regulatory networks. The focu s of this work is identifying one of these well-defined gene regulatory networks in s ea urchins and comparing it to that found in the brittle star. I have begun to characterize genes and their proteins that are involved in embryonic skeleton formation using two member s of Echinodermata, an ophiuroid, or brittle star and an echinoid, or sea urchin. The representative ophuroi d I have chosen is Opiocoma wendtii and the representative echinoid is Strongylocentrotus purpuratus The brittle star and sea urchin ha ve been chosen because they represent the only two classes of the phylum Echinodermata that have a well developed larval skeleton and a free floating pluteus larvae. The remaining members of the phylum Echinodermata will possess a skeleton only when they are adult. The sea urchin contains a skeleton compos ed of glycoproteins that have been previously well characterized and genes respon sible for the proteins involved in skeleton formation have been isolated and cloned (reviewed by Killian and Wilt, 2008). The genes involved in skeletal formation comprise a gene regul atory network that includes at least, Dri, Ets1 and Alx1 gene homologues. My hypothe sis is that this gene regulatory network has been conserved over time in both Ophiur oids and Echinoids. The resulting larval skeleton found in both classes of Echinodermat a is the result of activation of this regulatory network. I suggest that embryonic activation of th e gene regulatory networks occurred independently and may include different mechanisms but the “kernel” required for


8 skeletal formation is the same in both ophiuroi ds and echinoids. A “kernel” is defined as, a class of gene regulatory ne twork, which because of their developmental role and their particular internal structure, are most im pervious to change (Davidson et al., 2006). A GRN kernel typically has the following five properties: 1) A kernel is composed of a network of subcircuts that c onsist of regulatory genes. 2) They execute the developmental patterning functions required to specify the spatial domain of an embryo in which a given body part will form. 3) Kernels are dedicated to given developmental functions and are not used elsewhere in development of the or ganism. 4) They have particular form of structure in that the products of multiple regulatory genes of the kernel are required for function of each of the particip ating cis regulatory modules of the kernel. 5) Interference with expression of any one kernel gene will destroy kernel function altogether (Davidson and Douglas, 2006). The resulting developmenta l gene kernel exhibits extraordinary conservation over evolutionary time. I have cloned partial cDNAs encoding proteins similar to the sea urchin skeletal gene kernel from the brittle star. Using whole mount in situ hybridization we have determined the patter n of expression of one of these sea urchin skeletal homologue in brittle stars. Sea Urchin Development General Development The urchin has very little yolk in their zygote. Prior to fertilization the urchin egg’s animal vegetal axis is specified. Cell fate lines are oriented in the unfertilized egg according to animal vegetal axis. The anterior posterior axis is also laid out. The vegetal


9 region of the egg, through an unknown mechan ism, sequesters all of the maternal components responsible for posterior devel opment once a fertiliz ation event occurs (Boveri 1901). The specification of dorsal vent ral axis in sea urchins is not as well understood and appears to happen after fertil ization. The first cleavage furrow generally defines the dorsal ventral axis. The oral (ven tral) pole lies at a 45degree angle clockwise to the first cleavage plane (Cameron et al. 1989). Cleavage begins in these isolecithal (zygot es with little yolk) cells and the furrow extends throughout the entire cell. This form of cleavage is descri bed as holoblastic. The first two divisions of the sea urchin zygot e exhibit meridional di vision passing through the embryos vegetal and animal poles at perp endicular angles to each other. The third cleavage is perpendicular to th e first two and passes through the cells equator. This third cleavage separates the animal and vegetal halv es of the cell. One result of the fourth cleavage is the emergence of the first easily distinguishable cells within the embryo. The four cells of the animal half that separated from their vegetal count erparts during the last cleavage now divide meridional y into eight more blastomeres of equal size and volume. These eight cells are now referred to as me someres. The vegetal portion of the embryo does not divide evenly at the fourth divisi on. An unequal cleavage produces four larger cells called macromeres and four smaller ce lls called micromeres. The micromeres are more vegetaly situated than the macromeres which are oriented toward the animal pole of the embryo (Summers et al., 1993). The em bryo consists of 16 total cells broken down into groups of mesomeres, macromere and micromere cells. The eight mesomeres divide along the equa torial plane to pr oduce two layers of cells, an1 and an2. The macromeres go on to divide me ridionaly producing eight identical


10 cells below the an2 mesomeres. The micromere divisi on proceeds slower than both the mesomeres and macromeres. They produce eight smaller cells below the cluster of newly formed eight macromeres. A regular pattern of division continues for two more rounds. Animal cells that divide meri dionaly and vegetal cells divide equatorially characterize the sixth division of the sea urchin embryo. By the time the sea urchin embryo has reached 60 cells most embryonic cell fates are specified. However, specifica tion is not irreversible at th e 60-cell stage. For the most part, by the 60-cell stage specific blastomeres gi ve rise to specific tissues or organs. Many cells remain pluripotent and can give rise to several other cell types. Normally the animal half of the embryo tends to give rise to the ectoderm. Specifically, animal hemisphere cells give rise to the larval skin and neurons. The vegetal cells give rise to the endoderm and mesoderm and a few ectodermal structures. Specifically, the veg1 cells give rise to ectodermal or endodermal organs. The veg2 cells give rise to the coelom and secondary mesenchyme, as well as, endode rm. Specification of vegetal cells is accomplished in a multi phase manner. Veg2 cells specification is accomplished by waves of specifying signal. First, Bcatenin levels increase in Veg2 cells and specify endoderm. Second, Veg2 cells are exposed to the inductive signaling of ne ighboring micromeres. An early veg2 signal secreted by micromeres amplifies the mesode rm specification provided by B-catenin. The protein Delta next activates the Notch path way in the adjacent Veg2 cells. This notch pathway activation forces the adjacent cel ls to become secondary mesenchyme as opposed to endoderm. If either Notch or B-caten in signals are absent Veg2 cells fail to make secondary mesenchyme or endoderm (R ansick and Davidson 1995). The first tier


11 micromeres of the vegetal hemisphere that gi ve rise to mesenchyme cells that eventually produce the larval skeleton. These vegetal micromer es that are of interest to this thesis. The skeletal first tier micromeres are the only cells of the embryonic sea urchin that are capable of autonomous development and determination. Skeletal forming micromeres give rise to skeletal spicules even when placed by themselves in a Petri dish or test tube and give rise to dislocated skeletal spicules if transplanted elsewhere in the blastula (Ransick and Davidson 1993). Th e second tier (called “small”) micromeres contributes to structures of the coel om and do not give rise to skeleton. The seventh and last regularl y definable division occurs in an opposite orientation to the sixth division. Animal hemisphere cells divide equatorially a nd vegetal cells divide meridionaly. From the seventh division, or 128cell stage, cell divisions become much less regular and cells begin to develop differe ntly. At the 128-cell, blastocoel formation begins. The cells now begin to form a hollo w space surrounding what will be the hollow cavity of the blastocoel. By this time the sl ower cleavage rate of the smaller micromeres and the faster mitotic rate of the larger cell types results in an embryo where all the cells are the same size. Each of the cells is now in contact with both the outer hyaline layer and the inner proteinaceous fluid filled blastoco el. By the 128-cell stage the blastocoel is surrounded by an epithelial sheet of cells connected to each other by tight junctions (DanShokawa and Fujisawa 1980). The blastula rema ins one cell layer thick as cell division progresses. With each new round of cell di vision the cells thin out and expand by adhesion of the individual blastomeres to th e hyaline layer. An influx of water fills and expands the size of the blastocoel (Dan 1960; Wolpert and Gustafson 1961; Ettensohn and Ingersoll 1992). Cell specification does not begin until after the ninth or tenth cell


12 cleavage depending on species. Following the tent h division specialized ciliated cells can begin to rotate the embryo within the fertil ization envelope. Shortly following the tenth division differences can be seen that characterize individual cells. The vegetal plate now also begins to form as cells at the ve getal end on the embryo begin to thicken. The embryo finally becomes free swimming when cells at the animal pole synthesize and secrete a hatching enzyme that dige sts the fertilization envelope. A hollow ball surrounded by a single layer of about one thousand cells characterizes the sea urchin blastula late in its development. The Vegetal end of the blastula is somewhat flattened by this time due to the thickening of the vegetal pole cells. Now each individual group of blastomeres can be distinguished from other groups by size, properties and region of origin. At the center of the vegetal plate the flattened cells begin to extend and contract into long th in processes called filopodia. These cells disassociate from the monolayer blastula and begin to ingress into the hollow cavity that is the blastocoel. These cells are the descenda nts of the micromeres and make up what is called the primary mesenchyme. These primar y mesenchyme cells will make up what will eventually be the larval skeleton. Ther efore theses cells are often referred to as skeletogenic mesenchyme cells. The other cells that make up the blastula retain their affinity for the adjoining cells and for the hya line layer. Micromere descendants lose their affinity for their previous neighbors and gain an affinity for basil lamina and extracellular matrix (P. Oliveri 2002). This change in a ffinity results in the inward migration characteristic of these primary mesenchym e cells. Once inside the blastocoel these primary mesenchyme cells seem to be guided by the extracellular matrix and blastocoel wall. Movement happens as they extend and contract their filopodia in front of them


13 (Galieo and Morrill 1985). Proteins such as fibronectin and su lfated glycoprotein are also attributed to guiding this i ngression of the primary mesenchyme (Wessel et al. 1984). Ultimately these skeletogenic mesenchyme ce lls become localized within what will become the ventrolateral region of the blastocoel. Once at this location the cells fuse into syncytial cables that will serve as the f oundation for the calcium carbonate spicules that are the first indicator of a larval skeleton. Skeletal Development Although the general development of the sea urchin is of interest to us, the focus of this study is on skeletal de velopment. The first tier of micromere cells is responsible for the endomesoderm lineages and ultimately larv al skeleton formation in the sea urchin. The signaling molecule responsible fo r micromere specification, transplanted endodermal induction and normal larval skelet al development is B-Catenin (Ettensohn and Sweet 2000). B-Catenin is a transcripti on factor activated by the Wnt pathway. In canonical Wnt signaling, the Wnt fa mily of factors interact w ith the Frizzled family of transmembrane receptors. Disheveled protein is activated by the Frizzled protein when Wnt binds to Frizzled. The activation of the Disheveled protein inhi bits the activity of glycogen synthase kinase-3 enzyme (GSK -3). When active, GSK-3 prevents the dissociation of B-catenin pr otein from the APC protein. The APC protein targets Bcatenin for degradation. The presence of th e Wnt signal causes inhi bition of GSK-3 and allows B-catenin to dissociate from APC pr otein. This disassociated B-catenin is now free to enter the nucleus. Once nuclear, B-caten in forms a heterodimer with an LEF or TCF DNA-binding protein and becomes a true transcription factor. The B-catenin


14 transcription factor complex now binds to a nd activates the Wnt genes (Behrens et al. 1996, Cadigan and Nusse 1997). A noteworthy item for this study is that in addition to nuclear signaling Wnt also is integral in actin and microtubular cytoskeleton (Shulman et al. 1998). During normal sea urchin development B-ca tenin accumulates in the nuclei of any cell fated to become endoderm and mesode rm. This accumulation is autonomous and can occur even when the cells are removed fr om the embryo (Ransick and Davidson 1999). This indicates that the accumulation of B-caten in in the early blastomeres is independent of a Wnt signal, although the same cytoplasmi c components are involv ed. It appears that this accumulation of B-catenin is what spec ifies mesodermal and endodermal fates of the vegetal cells. Experiments show that inhibi tion of the nuclear accumulation of B-catenin inhibits formation of mesodermal a nd endodermal cells (Logan et al. 1998; Wikramanayake et al. 1998). It is also clear that micromeres from embryos with blocked B-catenin also lose th eir inductive ability wh en transplanted to other embryos (Logan et al. 1998). Once in the nucleus the B-catenin bi nds to TCF transcrip tion factors (Vonica et al. 2000). This binding allows Bcatenin to activate genes. Pmar1 is the gene responsible for mediating micromere sp ecification by B-catenin once it is bound to the TCF transcription factor. The repression of a recently discovered gene by Pmar1 allows expression of several genes required for prim ary mesenchymal cell development (Oliveri et al. 2002). This previously unknown repres sor of the skeletogenic gene regulatory network has been recently isolated and na med HesC (Revilla-i-Domingo et al., 2007). In the sea urchin it is the repression of a repres sor that allows activat ion of the battery of genes responsible for embryonic skeletogenesis At least some of the genes that are


15 repressed in the absence of Pmar1 are known to be Tbr, Ets, Alx1 and Dri. In the presence of Pmar1 these genes benefit from the repression of a repressor that prevents their expression. Each of these genes is responsible for directing a portion of skeletogenesis. Previous attempts in the labor atory have been made to isolate Pmar1 in brittle stars. Although it is present in the genome, it is not expressed during the embryonic development. This suggests that the gene responsible for the initial activation of the skeletogenic gene regulat ory network in sea urchins does not play the same role in brittle stars. Hypothesis The hypothesis tested here is that both sea urchins a nd brittle stars contain a nearly identical regulatory cluster of genes a ssociated with differen tiation of skeletogenic mesoderm, but that the initial genes involved in activation of this network differs between the two groups. When our representative species of O phiuroid and Echinoid patterns of embryonic development are compared, ophiuroids unlike echinoids, do not undergo an asymmetrical fourth division and do not produ ce micromere cells. However, even in the absence of mircomere cells, mesenchymal cel ls appear at approximately the same developmental time point in Ophiuroid deve lopment. This suggests a conservation of essential genes responsible for skelet al formation throughout the phylum. The composition of the sea urchin’s glycoprotein skel etal component has been well defined. Several of the echinoid genes that are responsible for this glycoprotein


16 matrix have been identified (Benson et al., 1987, Livingston et al., 1991, Livingston et al., 1996). Some of the proteins associated with the spicule matrix have also been characterized (Benson et al., 1986, Killian and Wilt, 1996). A voluminous amount of information on these genes as well as others was generated by the Sea Urchin Genome Project (Sea Urchin Genome Consortium, 2006) Genes found to be expressed in the sea urchin mesenchyme are also found in the starfish. A similar understanding of the Ophiuriod skeleton has yet to be attained. By comparing tw o species from of the same phylum it will aid in our understanding of the evolution of these genes and proteins. My hypothesis that there is a similar “ker nel” (Davidson et al., 2006) of a gene regulatory network at work in both classe s, ophiuroid and echinoid, during embryonic skeletal genesis. Conserved genes in this “k ernel” are at least Dri, Alx1 and Ets1. We suggest that this embryonic activation of the gene re gulatory networks occurred independently and may include different mech anisms but that the kernel required for skeletal formation is the same in both classes. My research focuses on the characterization and expression patterns of Dri and Alx1 in the brittle star, O. wendtii


17 Research Design Animal Husbandry / Spawning A particular challenge when working with a non-model organism as is the case with the brittle star, Ophiocoma wendtii is that all procedures involving both molecular as well as animal husbandry have to be genera ted de novo or optimized from procedures used in related organisms. During the course of my project, research procedures used for urchins were often relied on as a starting po int for procedural attempts and technical guidance. Animal collection was aided by our proximity to a site in the Florida Keys where O. wendtii can be found regularly. However, ev ery aspect of animal collection, care and transportation had to be empirically derived. In addition to general collection and care, spawning an animal in which there is very little published data is a process that required much trial and optimization. Molecu lar techniques were initially derived from published echinoderm literature. However, most often variances from species to species or embryo type to embryo type created the n eed for modification to apply to the brittle star system.


18 Animal Collection Location Echinoderm collection was done in the sha llow waters of the Florida Bay region of Long Key in the Florida Keys, USA. O. wendtii specimens were collected from the bay bottom near the Keys Marine Lab (KML) located in the city of Layton Florida (Figure 1). Figure 1. Animal collection site. Mile marker 65 U.S. highway 1, bridge piling 70 – 85, Florida bay, Florida Keys, Layton, Florida, USA The Keys Marine Lab is operated with cooperation between The Florida Institute of Oceanography as well as the State Univer sity System. The KML facility is a 5-hour


19 drive from our lab at the University of Sout h Florida. One of the major difficulties of working with O. wendtii lies in shipping and transport of the collected organism. Our proximity to the collection site and optim ized capture and transportation techniques allows us a relatively high rate of surviv al when transporting collected specimen form Florida Bay to the University Of Sout h Florida in Tampa. The brittle star O. wendtii was chosen due to their la rge size and abundance. O. wendtii are gravid form June to October (Hendler et al., 1995). O. wendtii are relatively abundant in the Florida Keys. O. wendtii possesses a central disk, which varies in size up to 1.5 inches. They have been found to inhabit several areas within a one-hour boat ride from the Keys Marine lab. O.wendtii congregate in debris piles or natural reef structures that provide a shelter from predation and harsh Florida bay tidal currents. Primar y collection occurs under the U.S. highway 1 bridge (bridge pilings #’ 70 – 85) located at mile marker 65, between the islands of Long Key and Conch Key. It has been determined that O. wendtii are more concentrated near or directly at the bridge pilings that support the now unused Flagler.


20 Figure 2. Debris removal / Anim al location. Ocean bottom debris lifting technique used to locate O. wendtii Animals were found by gently raising and loweri ng large concrete scraps or rocky debris segments located at or near th e bridge pilings (Figure 2). Time / Date of Dives It has been reported that sea urchins and other echinoderms are more prone to spawning at or near a full moon. This is also th e case with several other species of marine animals that live near or in a tidal plain. We ha ve also determined that in our specific dive location O. wendtii are gravid during a time period that begins in late spring (April) through early fall (Early Octobe r). Collection dives were plan ned during the spring to fall time that would allow us to dive and collect over a time period that covered one to two


21 days on either side of a full moon cycle to maximize the chances of finding gravid specimen that could be induce to or would naturally spawn in our collection tanks. Dives were completed in 15-25 feet of water depending on the time of the month and high or low tide conditions. Collection Protocols For animal collection, we employed a device termed a bongo (Figure 3). Figure 3. Bongo Device A collection bongo consists of a 24-inch long, 18 inch diameter PVC tube with mesh netting encapsulating either end of the tube A bungee corded lid flap was cut into the tube to allow for secure and safe inserti on and removal of captured specimen. Once an


22 animal is located care must be used when removing it from its environment. To capture a brittle star a hand must be placed under the bulk of its body at least encompassing most of its central disk. With a gentle consistent motion the brittle star can be raised from the sea floor and placed into the hole in the bongo for storage. If grasped by one or more of its appendages, and not the bulk of its body, the brittle star will jettison the threatened limbs as a survival strategy. In addition to this collect ion technique two alternative methods were deployed. In order to determine if gametes were being rel eased in the bongos or in the 5 gallon buckets as a stress response while in tr ansport back to the lab we at tempted to isolate individual O. wendtii from each other immediately following capture. In one technique I would place collected specimen that were return ed to the dive boat form the bongo into individual sealed Tupperware dishes containing seawater. Another attempt to further control or identify the rel ease of gametes from collection to lab involved placing collected animals directly into Ziploc bags while underwater and then placing them into the bongo. This would ensure that we would be ab le to identify if gametes were released any time from capture to lab (Figure 4). Figure 4. Ziploc O.wendtii collection method


23 Collected animals were returned to th e boat in the bongo and placed in a five gallon bucket of sea water and transported to the Keys Marine Lab for short term care and storage. At the dive site several carboys of seawater are collected. This water is used to house the animals once they are back at KML. This seawater is also filtered and used to house spawned developing embryos. Once at the KML animals are placed into large 30 gallon aquariums with heat and aeration (Figure 5). Figure 5. Keys Marine Lab Animal storage tanks Up to 30 animals have been shown to thrive in one 30-gallon tank of seawater. Using these techniques of collection a nd short-term storage we expe rienced a very high survival rate nearing 100%. By using th ese collection techni ques over the past four years we have transported at least 423 healthy spec imens to the Keys Marine Lab.


24 Transport to Tampa Animals that will be transported to U SF for further study and spawning attempts in Tampa are chosen from the specimens that appear the most vibrant and able to withstand the stre ss of transport. O. wendtii Animals chosen for transport are placed into a 1 gallon Tupperware container that contains small slits or holes in it that allow for ventilation and water exchange. O. wendtii The Tupperware container is then placed in a large 75 gallon insulated cooler full of fr eshly collected seawater. Battery operated aerators are also placed in the coolers to allow for sufficient oxygen supply while in transport. Additional carboys of freshly coll ected and air supersaturated sea water are also transported with the collected specimens During the 5 hour driv e to Tampa at least once and usually twice water is drained fo rm the coolers containing the animals and replenished with fresh sea water from th e carboys. By using this collection and transportation technique we have consistently transported O. wendtii from the middle Florida Keys to USF, Tampa. Over the past four years we have transported at least 271 brittle stars to the lab in Tamp a from the Keys Marine Lab. Storage and Care of Brittle Stars in Tampa Once the O. wendtii were transported to USF, Tampa they were checked for signs of stress or early stages of autolysis. Two 30 and two 15-gall on fish tanks were prepared with keys sea water prior to departure to the collection site. Healthy animals were dispersed evenly throughout the three tanks. It has been shown in our lab that the death and autolysis of one animal can trigger the death and autoly sis of the remaining healthy animals in the tank in as early as one da y. The use of a recovery tank for unhealthy


25 individuals eliminates some of the risk of placing an unfit O. wendtii triggering total tank death in an otherwise healthy specimen tank. Th is technique employed over the last three years has greatly increased specimen survival ti mes and almost eliminated entire tank die offs that were once a more frequent occurrence. Animals were feed frozen shrimp on an every third day basis. Tank water was regularly checked for salinity and signs of fouling. Salinity was kept at a specific gravity of 1.028. During the past three years we had several specimens that lived up to a period of one year or more. Spawning of O.wendtii Keys Marine Lab Spawning Once animals are returned to the lab each specimen is examined for the presence of eggs and an attempt at gender identific ation is made. Female s are discerned by the presence of visible swollen, purple gonads at the junction of the flexible arm and central disk. In the absence of this obvious sign of gender females tend to take on a puffed up appearance in the region of their central disk specifically the area s near the arm disk junction. Specimen that are thought to be fema le are separated and stored in individual tanks prior to spawning attempts. 30 gallon ta nks containing freshly collected sea water are used for spawning attempts. Into each ta nk approximately 20 gallons of sea water is placed. Each tank also receives 3 – 5 air st ones, depending on the final number of animals, for aeration prior to receiving animal s. At least three tanks are regularly used with each spawning attempt. One tank is kept as close to current ocean temperature as possible, a second tank is slightly warmer (3-4 degrees C) than ambient ocean temperature and the last tank is kept warmer yet up to 90 de grees F at which temperature


26 the brittle stars start autolysis and die. Sp awning does not seem to occur until the water temperature is at or near 85 degrees F. In addition to varying the water temper ature during the spawning attempts the addition of lab collected sperm also appear s to induce spawning. Sperm can be obtained by .5M – 1 M KCL injection into males. Fema les however are not induced to spawn with the KCL injection method. Males injected at the appendage / central disk junction will regularly release sperm. Sperm generated by this method is collected in a beaker of seawater and this seawater / sperm solution is used to seed th e tanks during spawning attempts. A system of light shock where spec imens are maintained in dark environment and given intermittent exposure to simulated or actual moon light has proven to produce eggs while at the Keys Marine Lab as well. Once collected and sexed animals are placed into the appropriate 30-gallon tanks containing freshly gathered sea water and allowed to acclimate for a period of 1 -5 hours in a dark room. Once the initial period of acclimation is complete and generally at a time after s unset a systematic approach to light shock involving long periods (1 – 2 hrs) of absolute dark and shorter periods (15 -30 min) of light shock are employed. It has been observe d that periodically immediately following light shock or light stimulus both or eith er male and female organisms will release gametes. It has been our experience that a comb ination of moon phase, water temperature, sperm introduction and light shock produce a spawning event while at the Keys Marine Lab. There has been no one stimuli that has proven to definitive ly induce spawning and there has been no definitive combination of the stimuli that consistently produce a spawning event. However over th e course of a spawning attempt if all variables are used


27 and modified regular spawning can be indu ced. It is our experience that a spawning attempt one to two days prior to a full m oon cycle in water that is approximately 85 degrees Celsius with a periodi c light shock event and some freshly collected sperm / sea water solution added to the tanks will consis tently produce a spawning of the brittle star O. wendtii while on site at the Keys Marine Lab. USF Tampa O. wendtii spawning A similar approach of light shock in a temperature-controlled environment during a full moon cycle and sperm solution added to the tanks has not proven successful when specimens are transported back to the lab at USF. Nighttime light shock has been attempted 14 times over the past three years and has yet to produce eggs from lab kept females. A spawning attempt at USF involves the sa me variables of light, temperature, moon phase and sperm solution as are invol ved while attempting to induce spawning while at the Keys Marine Lab. I was able to induce spawning in male animals, but was not able to induce females to spawn. Because of the troubles encountered with on site spawning all lab work has been done using fertilized eggs and RNA gathered from animals that spawned while at the Keys Marine Lab in Long Key. Fixation of Embryos Once a successful spawning is achieved time is noted and the seawater / embryo solution is removed from the tank and watched for development. During the course of the


28 night following a successful spawning, time poi nts are taken at developmentally critical periods and the developing embryos are watche d through out the night for general health and development. Time points are preserved from each developmentally critical time point with special consideration paid to time points critical to embryonic skeletal formation. Following time zero a time points of development are taken at late cleavage, 17 hr. blastula, 26 hour mesenchy me blastula, 39 hr. gastrula 48 hr. late gastrula, and 75 hr + pluteus larvae. Lastly pluteus stage embryos are collected form approximately 100 hours of development and once daily until the embryo solutions stop development and die. For each time point collected 50 ml of developing embryos in filtered seawater solution were placed into a ster ile 50 ml conical vial. The 50 ml conical vial was spun at high speed in a bench top centrifuge for one mi nute. The liquid supernatant was aspirated off and 50 ml of additional embryo / sea wa ter solution was added on top of the embryo pellet and spun again for one minute on high spee d in a bench top centrifuge. This repeat of the collection procedure essentially doubled the size of the collect ed pellet. The liquid is aspirated off again down to approximatel y 4 ml of fluid remaining. 5 ml of 4X paraformaldehyde fix was added to the embr yos. After a minimum of one day in fix embryos were washed into 1X PBS to a volum e of 50 ml and stored at 4 degrees C for later use in either WMIS H or photographic studies. Gene Regulatory Network and Molecular Isolation of a Fragment of O. wendtii Dri cDNA Homologues of the Alx1 and Dri skeletal genes have been isolated in the sea


29 urchin. By comparing the deduced sea urchin protein sequences to the protein sequence of homologous proteins from other organisms, we were able to locate sequences of amino acids that were highly conserved. From th ese regions of homology, degenerate primers were created. The degenerate primers were th en used to amplify the brittle star cDNA prepared from extracted O. wendtii embryonic mRNA. Using this process sequences for both Alx1 and Dri have been iden tified in the brittle star. Design of Degenerate Primers Alignment of Sea Urchin Dri Amino Acid Sequence With Those From Other Organisms to Determine Conserved Regions Degenerate PCR primers were initially created for the O. wendtii dead ringer gene using S. purpuratus sequence information that was aligned with 22 other organisms Dri amino acid sequences and analyzed for areas of homology. The degenerate primers are a convenient way to amplify the same gene from different organisms. In this case the degenerate primers are extremely valuable due to the fact that O. wendtii has very limited genomic information available. The degene rate primers were created by analyzing aligned Echinoderm as well as other Dri se quences found in GenBank. Both forward and reverse degenerate primers were generated as follows. See figure 6 for alignment and primer sequence generation. In order to determine conserved regions within the Dri amino acid sequence 22 Dri amino acid sequences from 22 organisms we re analyzed for regi ons of homology. After analyzing the sequences of 22 orga nisms the list was narrowed to the sea urchin, mouse and ciona. The exclusion of protostomes revealed a higher degree of


30 conservation of sequences among deuterostomes. From the alignment of these three organisms areas of high homology were analyzed for hairpins, self-d imerization potential and melting point. The conclusions of this anal ysis lead to the creation of the initial forward and reverse Dri primers (Dri F1 and Dri R1) (Figure 6). In analyzing areas of homology and degenerate primer design care was taken to ensure that the echinoderm, sea urchin, was well represented in the area of homology. Due to the inter phylum relationship of sea urchin and brittle star it was presumed that there would be shared homology between the sea urchin and brittle star Dri sequences. Figure 6. Dri degenerate primer design


31 Synthesis of O. wendtii cDNA Into a sterile RNAse free tube 0.5micrograms of random primers were added to 1.5 micrograms of combined early gastrula, and gastrula stage O. wendtii RNA. These stages were used due to th e fact that under microscopy embryonic skeleton can be seen forming in these stages. The tube was heated to 70 degrees C for five minutes to melt secondary structures within the template. The tube was cooled on ice for five minutes and briefly spun. The following were added to the tube in order of listing: RNAse free water (to a volume of 25 micro-liters) 11.5 micr o liters, M-MLV 5X buffer 5.0 micro liters, dATP 10 mM 1.25 micro liters, dCTP 10mM 1.25 micro liters, dGTP 10 mM 1.25 micro liters, dTTP 10 mM 1.25 micro liters, rRNAsin (40 units / micro liter) 25 units / .5 micro liters and M-MLV Reverse Transcriptase enzyme (200 units / micro liter) 1.0 micro liter. The tube was mixed gently and incubated fo r 60 minutes at 37 degrees C. A control PCR using Actin forward and reverse primers wa s conducted to confirm cDNA synthesis (Figure 12). Two separate cDNA reactions were used to genera te two separate batches of O. wendtii cDNA (cDNA 1 and cDNA 2). PCR Including Optimization and Gels of Degenerate Primer PCR Degenerate PCR primers were used with the following optimized PCR cycle. 50 micro liter volumes were used for each reaction. Salt concentration was 1.5M. One tube was set up as experimental and contained a cDNA RT reaction and another was set up as a negative control and contained steril e PCR water and no cDNA template. To the experimental reaction tubes 5 micro liters of O. wendtii cDNA was added. To the negative control tube 5 micro liters of wate r was added. To each reaction tube 20.6 micro


32 liters of sterile PCR water, 5.0 micro liters of 10X TAQ buffer, 10 micro liters of 5X TAQ master, 1 micro liter of 0.3 mM dNTP mix, 0.4 micro liters of TAQ, 4 micro liters each of forward and reverse 0.8 mM degene rate primers was added. The following PCR program were completed: 95 degrees C for th ree minutes, [(39 cycl es of) 95 degrees C for 40 seconds, 51 degrees C for 45 seconds, 72 degrees C for 45 seconds], 72 degrees C for eight minutes and hold at 4 degree s C indefinitely. Once PCR was completed visualization was done using a 1% agarose gel electrophoresis. Promising PCR products were visualized as bands of approximate ly 300 bp each for both PCR1 and PCR 2 cDNA templates (Figure 16). Purifying Degenerate Primer PCR Products Degenerate primer PCR products were pur ified using (Millipore, Tempecula Ca.) Montage filtration devices according to the manufacturers instructions. The purified PCR products were the visualized using a 1% agarose gel for confirmation (Figure 15). v. Degenerate Dri primer PCR pr oduct ligation and transformation Purified Dri degenerate PCR products were ligated and transformed using TOPO 4 vector cloning protoc ols and chemically competent E. coli from Invitrogen (Carlsbad, CA) using the manufact urers protocols. Crude Mini Plasmid Prepar ations of Degenerate Dri PCR Primer Products Using sterile technique, 3m l glass culture tubes were prepared with 3ml of LB broth and 12 micro liters of ampicillin. Tube s were labeled and a grid plate of LB-AMP was labeled in kind for a master control pl ate. A single colony from the overnight


33 incubated LB-AMP + TOPO reaction was selected with at autoclaved toothpick. The grid plate was scraped with one side of the toot hpick and the toothpick was then dropped into the LB broth AMP tube that corresponded to the label on the grid plate. Following bacterial colony selection the tubes containing toothpicks were placed into the shaking incubator at 300 RMP and 37 degrees C for overnight incubation (12 – 18 hrs.). Following overnight incubation 2 sets of 1.5 ml eppendorf tubes were labeled that corresponded to the 3 ml culture tubes. Half of the contents of each culture tube was decanted into the corresponding 1.5 ml eppe ndorf tubes. Eppendorf tubes were spun down in a bacterial centrifuge on max speed for one minute and the supernatant was removed. One set of the pair of eppendorf tube s was placed on ice to be used for perfect plasmid preparations once crude plasmid prep aration had identified potential clones of interest. On ice, the second set of eppendorf tube s containing bacterial pellets were resuspended by pipetting 150 micro liters of Qiagen buffer P1 up and down until the entire pellet was re-suspended. To the re-suspended pellet 150 micro liters of Qiagen buffer P2 was added and the tubes were immediately i nverted. Once inverted 150 micro liters of Qiagen buffer P3 was added to the tubes th at were then vortexed thoroughly and placed on ice for 5 minutes. The tubes were then spun for 10 minutes on high at 4 degrees C. Another set of Eppendorf tubes was labeled and 1 ml of 100% EtOH was added to each newly labeled tube. The supern atant from the centrifuged tubes was poured off into the new tubes of corresponding label that cont ained 1 ml of 100% Et OH. The tubes were then incubated at -80 degrees C for one hour. Following -80 degree C incubation the tubes were spun at max speed at 4 degrees C for 15 minutes. The supernatant was poured


34 off and the pellet that remained was washed with cold 75% EtOH. The tube was next spun for 5 minutes on high speed at 4 degr ees C. The pellet wa s air dried and resuspended in 30 micro liters of 1X TE. Crude mini preps were next digested to separate the degenerate Dri PCR product inse rt from the TOPO 4 vector. Digestions were done by labeling a corresponding set of 0.65 micro liter eppendorf tubes and adding to each; 1.0 micro liters of New England Biolabs Buffer H (10X), 0.1 micro liter of 10mg / ml BSA, 0.5 of New England Biolabs EcoR1 enzyme (812 units / micro liter), 3.5 micr o liters of NP water and 5 micro liters of mini prepped DNA. The digests were incubated at 37 degrees C for 3 hours. The results were of the digests were confirmed on a 1% agarose gel. Purified Plasmid Preparation of De generate Primer Dri PCR Products To the second set of tubes that were placed at -20 degrees C from the crude prep the Qiagen procedure for [plasmid purifica tion was applied in order to purify the plasmids with inserts representing the de generate Dri PCR products to prepare for sequencing. The manufacturers protocols were followed. Prod ucts were visualized on a 1% agarose gel to confirm the presence of the expected degenerate Dri primer PCR bands. Sequencing of Degenerate Dri Primer PCR Products Sequencing was carried out using Macr ogen Sequencing services, Seoul South Korea. 10 – 15 micro liters of purified DNA with a concen tration of 50 nano grams per micro liter was prepared for each sample to be sequenced. All samples were sequenced


35 using both T3 and T7 primers provided by M acrogen. Samples were shipped via FedEx according to current state, local and Macr ogen specifications for shipping biological samples. The sequence obtained from initial Dri degenerate primer PCR of O. wendtii cDNA was as follows. Obtaining Full-Length cDNA Sequences of Dri and Alx1 Once initial incomplete O. wendtii sequences were obtained via degenerate Dri primers it became my goal to determine the fu ll-length sequences of both Dri and Alx1. I employed the Ambion RACE (rapid amplifica tion of cDNA ends) technique of cDNA sequence extension in both the 3’ and 5’ directions to extend both the Alx1 partial transcript and Dri partial transcript. The RACE cDNA ex tension approach entailed several rounds of varied primer combinat ions used along with the RACE treated cDNA and RACE specific primers. Several sequence specific primers were generated for each of Alx1 and Dri and used systematically in multiple combinations along with optimized PCR profiles in an attempt to generate comple te transcripts that c ontained the 5’ start sequence and 3’ termination sequence of each respective skeletal gene. Following each RACE PCR attempt, PCR products would be ligated, transformed, plasmid prepped and sequenced to determine if indeed new sequen ce had been generated via our technique. RLM R.A.C.E. First Choice RLM-RACE is a protocol and Kit purchased from the Ambion Corporation (Austin, TX). RACE is an anachronism for R apid A mplification of c DNA


36 E nds and is used to facilitate the cloni ng of full-length cDNA sequences when only a partial cDNA sequence is available. We fo llowed the manufacturers protocol but used our gene specific primers. Temperatures were optimized for each primer set based on their sequence and the manufacturers recommendations. Prior to using the newl y processed and created O. wendtii cDNA with the designed O. wendtii sequence specific primers we generated a positive control mechanism that allowed for the confirmation of the processed O. wendtii RNA This process allowed us to confirm that the RACE adapter had indeed been ligated to the cDNA sequence and therefore could be successf ully used in the RACE procedure. Our method of positive control involved using a co mbination of the Actin reverse primer and the RACE 5’ outer primer. A combination of an Actin reverse primer and a 5’ RACE outer primer could be expected to genera te a PCR product approximately 800 base pairs in size (figure 17). Figure 7. RACE cDNA confirmation + Control using RACE outer and Actin Reverse primers


37 Or, roughly a 300 bp larger PCR product than the Actin forward and reverse primers produced. This result is expect ed due to the known size of 500bp that is produced by the Actin gene when an Actin forward and reve rse primer combination is used and the additional size provided by the RACE adapte r and Dri sequence provided by using the RACE outer primer in combination with the Actin reverse primer. Once RACE outer and inner nested PCR are done, PCR products can be visualized via 1% agarose gel electrophoresis. A positive result is the presence of one to a few bands in the experimental lanes of the agaros e gel. This is due to the variability of the transcription length for each cycle and transcript. Non-Degenerate PCR Primer Design and Generation Primers were designed to produce forw ard and reverse inner and outer nondegenerate primers using O. wendtii sequence obtained from the initial degenerate primer PCR (figure 16). These primers would be us ed in combination with RACE inner and outer primers to extend the ends of known transcripts using the Ambion RACE (Ambion, Austin, TX) described in the RACE section. Primers were created to have the most sequence specificity and highest melting poi nt possible to incr ease primer binding specificity during PCR by allowing for higher annealing temperatures to be used. In addition to an inner and outer reverse RA CE gene specific primer a second inner (2nddririn) was designed to further increase spec ificity effectiveness of the transcript end amplification of the Dri gene. RACE forward and reverse perfect pr imers designed for Dri and Alx1:


38 Dead ringer perfect forw ard 3’ outer primers: Sequence #1 (dripf ) = 5’ GAG GAG CAA TTT AAG CAG CTC TAT GA 3’ Tm = 56. 8 degrees C Sequence #2 (drif41108) = 5’ CAG TG G CGT GAG ATC AC C AAG GGC C 3’ Tm = 65.6 degrees C Dead ringer perfect forw ard 3’ inner primers: Sequence #1 (dripfin) = 5’ TCT AT G AGT TAT CTG ATG ACC CTC A 3’ Tm = 54.5 degrees C Sequence #2 (drifin41108) = 5’ GGG CCT CAA CCT ACC AGC ATC C 3’ Tm = 62.2 degrees C Dead ringer perfect reverse 5’ outer primers: Sequence # 1 (driprev) = 5’ AGG TA C TTC ATA TAT TGG GTA TGA 3’ Tm = 50.7 degrees C Sequence #3 (drir42108) = 5’ GAG CTG CTT AAA TTC CTC CTC 3’ Tm = 53.2 degrees C Dead ringer perfect reverse 5’ inner primers: Sequence #1 (dririn) = 5’ CCC TTG GTG ATC TCA CGC CAC T 3’ Tm = 61.5 degrees C Sequence #2 (2nddririn) = 5’ CCG CTG AGG GTC ATC AGA TAA CT 3’ Tm = 58.3 degrees C


39 Sequence #3 (dririn42408) = 5’ CCT CTG AGG GTC ATC AGA TAA CTC 3’ Tm = 58.8 degrees C Alx1 perfect reverse 5’ outer primers: Sequence #1 (Ow Alx1 Out) = 5’ G AG TTG CTG AAA CCT TTC ACG C 3’ Tm = 57.0 degrees C Sequence #2 (Ow Alx1 Out Spring 2008) = 5’ CTT TGG TCG ACT CAC TGC CTA TCG 3’ Tm = 59.2 degrees C Alx1 perfect forward 5’ inner primers: Sequence #1 (Ow Alx1 in) = 5’ TTC ACG CTT CCG CCA TTT GGC T 3’ Tm = 62.3 degrees C Sequence #2 (Ow Alx1 In Spring 2008) = 5’ CGT TGT GTT GGA TGT TGA AAG CGG 3’ Tm = 59.2 degrees C Alx1 perfect forward 3’ outer primer: Sequence (Alx1 3’ Race Outer) = 5’ GGT GGA GAC AGG AGG ACA AGT AGT 3’ Tm = 59.6 degrees C Alx1 perfect forward 3’ inner primer: Sequence (Alx1 3’ Race Inner) = 5’ GCT CTG CGG TTG AGA GCA AAA GAA 3’ Tm = 59.9 degrees C


40 Race primer combinations and PCR profile optimization was achieved by multiple attempts at PCR using RACE kit pr imers with combinations of Dri and Alx1 gene specific primer combinations. In additi on to varying combinations of primers used and optimized PRC profiles. The following list illustrates the optimized primer combinations and PCR conditions that produced the results that re turned sequences as Dri. All other primer combinations and se quencing attempts did not produce positive results. Empirically Derived Primer Co mbinations and PCR Protocols Dri degenerate primers: Using Dri degenerate primers, Dri forward a nd Dri reverse, with PCR profile; 95 degrees C – 3 min., 39 cycles of [95 degrees C – 40 s ec, 51 degrees C – 45 sec, 72 degrees C – 45 sec], 72 degrees C – 8 minutes. Dri perfect primers 3’ RACE: 3’ RACE using primer combinations drif 40018 (outer 3’) and drif in41108 (inner 3’). Outer PCR profile = 94 degrees C 3 minutes, 35 cycles of [94 degrees C – 30 sec, 65 degrees C – 30 sec, 72 degrees C – 30 sec], 72 degrees C – 30 sec. Alx 1 perfect primers 5’ RACE: 5’ RACE using primer combinations OwAl x1Out (5’ outer) and OwAlx1In (5’ inner), Outer PCR profile = 94 degrees C – 3 minutes, 35 cycles of [94 degrees C – 30 sec, 54


41 degrees C – 30 sec, 72 degrees C – 2 minut es], 72 degrees C – 7 minutes. Inner PCR profile = 94 degrees C – 3 minutes, 40 cycles of [94 degrees C – 30 seconds, 54 degrees C – 30 sec, 72 degrees C – 2 minut es], 72 degrees C – 7 minutes. Alx1 perfect primers 5’ RACE: 5’ RACE using primer combinati ons OwAlx1OutSpring2008 (5’ outer) and OwAlx1InSpring2008 (5’ inner), Outer PCR pr ofile = 94 degrees C – 3 minutes, 40 cycles of [94 degrees C – 30 seconds, 54 de grees C – 30 sec, 72 degrees C – 30 sec], 72 degrees C – 7 minutes. Inner PCR profile = = 94 degrees C – 3 minut es, 40 cycles of [94 degrees C – 30 seconds, 54 degr ees C – 30 sec, 72 degrees C – 30 sec], 72 degrees C – 7 minutes. Alx1 perfect primers 3’ RACE: 3’ RACE using primer combinations Alx13’ RaceOuter (3’ outer) a nd Alx13’RanceInner (3’ inner), Outer PCR profile = = 94 degrees C – 3 minutes, 35 cycles of [94 degrees C – 30 seconds, 58 degrees C – 30 sec, 72 degrees C – 1 minute], 72 degrees C – 7 minutes. Inner PCR profile = 94 degrees C – 3 minutes, 35 cycles of [94 degrees C – 30 seconds, 58 degrees C – 30 sec, 72 degrees C – 1 minute], 72 degrees C – 7 minutes. Ligation and Transformation of RACE PCR Products PCR products of appropriate size and interest were ligated into TOPO 4 vectors as described previously.


42 Crude Mini Preparations of RACE PCR Products These were carried out as described previously. Digest of Crude Mini Prepar ations of RACE PCR Products For the digests of crude mini preps a master mix containing 1 micro liter 10X Buffer H (New England Biolabs, Stockton, New England), .1 mg/ml BSA (New England Biolabs), 0.5 micro liters EcoR1 Enzyme (N ew England Biolabs) and 3.5 micro liters sterile water per each reaction needed was cr eated. 0.65 micro liter eppendorf tubes were labeled to correspond to each of the crude pr epped bacterial samples. 5 micro liters of master mix was added to each 0.65 micro lite r eppendorf tube. 5 micro liters of mini prepped DNA was added to each appropriate ly labeled 0.65-eppendorf tube. Reaction tubes were pulse spun in a micro centrifuge. Reactions were incuba ted at 37 degrees C for 2 – 3 hours. Following a one hour incubatio n at 37 degrees C the digests were run on a 1% agarose gel to id entify positive clones. Perfect Plasmid Preparati ons of RACE PCR Products Perfect preparations were done using Qi agen quicklyse miniprep kits (Qiagen, Foster City, Ca.) as described previously. Sequencing of RACE Generated PCR Products Sequencing was carried out using Macr ogen Sequencing services, Seoul South Korea. See sequencing results se ction for sequencing results.


43 Determining Spatial and Temporal Expressi on Patterns of O. wendtii Skeletal Genes via Megascript RNA Probe Generation an d Whole Mount In Situ Hybridization Megascript Transcription Reac tions and RNA Probe Generation With the sequences provided by Macroge n DIG labeled sense and antisense RNA probes were made using Ambion Mega and Maxiscript transcription reactions and protocols (Ambion, Austin, Texas). On ce generated these RNA probes are 1st tested for activity using Southern Blo tting against linearized plasmid DNA to determine relative activity and strength of the pr obes intended for use in whole mount in situ hybridization. Sequencing is carried out with a T3 and T7 promoter. Once the T3 and T7 sequences are examined for insert orientation. T3 and T7 Megascript reactions are carried out producing both a sense and antisense RNA probe. Sequenced plasmids shown to contain either Alx1 or Dri were first linearized in a digest reaction. Each plasmid was digested in two separate restrictions enzymes in two separa te restriction digests. This produced two versions of the same linearized plasmi d. One restriction digest would produce a linearized plasmid with a T3 promoter si te and the other would produce a linearized plasmid with a T7 promoter site. In one reaction plasmids were cut with NOT1 to produce a linearized plasmid with a T3 promot er site that would be used to produce a anti-sense Megascript probe for Southe rn Blotting and Whole Mount In Situ Hybridization. In the other restriction dige st reaction purified DNA containing plasmids were linearized using restriction enzyme Spe1 to produce a linearized plasmid containing a T7 promoter that would be used to produce a control sense Megascri pt probe (figure 8).


44 Figure 8. RNA probe generation Once restriction digests were complete a gel was run to confirm the linearization reaction that contained one lane of unc ut plasmid as a control and tw o lanes that contained either NOT1 digested plasmid or Spe1 digested pl asmid. The linearized plasmids should appear the same size when visualized with a 1% agarose gel. The uncut plasmid should present as a separate molecular weight when visualized (Figure 22). Southern Blotting In order to determine the activity of ou r Megascript generated DIG labeled probes all probes were first used with linearized plasmid DNA containing the specific sequence of DNA that was used to generate the RNA pr obes. This extra step was done to ensure


45 maximum likelihood that the probe and procedur e could work prior to using valuable and scarce O. wendtii embryos with the RNA probes for w hole mount in situ hybridization. A 1% Agarose gel was used to run two series of varying dilutions of linearized plasmid DNA (1/100, 1/1000 and 1/10,000) as well as D NA ladders for DNA size base pair determination (Figure 23). These series of dilutions are then used to detect the sense and antisense RNA probes. Once the 1% agarose gel was run and photographed the DNA markers were cut away and the gel was denatured and neutraliz ed. The gel was placed into 0.25 M HCL for 10 minutes and then rinsed with nanopure wa ter. The gel was next placed into 1.5 M NaCl / 0.5 M NaOH for 20 minutes and then rins ed with nanopure wate r. Lastly the gel was placed into 1.5 M NaCl / 0.5 M Tris pH 7 for 20 minutes and then rinsed with nanopure water. Following gel denaturing a nd neutralization the DNA was transferred from the agarose gel to a n itro cellulose transf er membrane via cap illary wicking action. Once the DNA was transferred cutting the membrane separated the two dilution series contained on the membrane. One half of the membrane was used for antisense RNA probe detection and the other half was used for sense RNA probe detection. The membranes was first be pre-hybridized in 15 ml of pre-hybridization solution consisting of 7.5 ml formamide, 3.75 ml 20X SSC pH 7, 30 micro liters EDTA 0.5M, 750 micro liters 1 M NaPO4 buffer pH 6.8, 600 micro li ters Denhardt’s solution, 316 micro liters 9.5 mg / ml salmon testes DNA and 2.054 ml nanopure water. The pre-hybridization solution was warmed to 68 degrees C in a warming oven. Each half of the blot was placed into a glass hybridization tube. 7 ml of warmed pre-hybridization was added to the


46 tubes. The tubes with pre-hybridization solu tion and membranes were incubated at 68 degrees while rotating for 30 minutes. Prior to hybridization the probes were diluted to approximately 150 nano grams per micro liter by addition of appropriate pr obe RNA volume to 50 micro liters of DEPC water. The diluted probe was heated to 68 degrees C and immediately placed on ice. Following pre-hybridization and incubation of the membranes, 300 nano grams of RNA probe was added 3 ml of pre-warmed prehybridization solution. A ppropriate sense and antisense probes were incubated at 68 degrees C overnight. A positive control was used when ever available that consisted of a probe proven to work in past experiments along with the corresponding linearized DNA at a conc entration of 1/100. This control served as a reference for the sensitivity of the probes. Membranes were washed in 100 ml of lo w stringency buffer consisting of 20 ml of 20X SSC pH 7, 1 ml 20% SDS and 179 ml nanopure water. Membranes were submerged and shook in low stringency buffer fo r five minutes. This low stringency wash was repeated 5 times. Following low stringenc y buffer washes the membranes were again transferred to glass hybridization tubes. 20 ml of high stringency buffer was added consisting of 200 micro liters of 20 X SSC 200 micro liters 20% SDS and 39.6 ml of nanopure water and the blots were incubated at 68 degrees C for 15 minutes. This high stringency wash was repeated 4 times. Th e membranes were transferred to a tray containing 100 ml of wash buffer consisting of 300 ml maleic acid buffer (M.A.B.) and 900 micro liters of tween 20 a nd incubated at room temperat ure with gentle shaking for two minutes. The wash buffer was discarded and 100 ml lf blocking buffer consisting of


47 12 ml 10X blocking reagent and 108 ml M.A.B. was added to the trays containing the membranes and incubated at room temper ature with gentle shaking overnight. The antibody used to detect the DIGlabeled RNA probes is Anti-DIG-Ab. The anti-DIG antibody was centrifuged at 10,000 RP M for five minutes. 4 micro liters of antibody was carefully pipetted from the surf ace of the antibody vial. The 4 micro liters of antibody was added to 40 ml of blocking solution to create th e antibody solution. 20 ml of antibody solution was added to the blot s. The blots were incubated for 30 minutes in antibody solution. Following antibody solution addition the blots were washed 2 times for 15 minutes each with gentle shaking in 100 ml of wash buffer. Using gloves CDPStar was added to detection buffer in a 1:100 dilution (5.2 micro liters of CDP-star: 514.8 micro liters of detection buffer). For visualiz ation the bag was placed into a developing cassette and exposed to film. Film is devel oped and examined for the expression of DIGlabeled tags associated with the Megascri pt RNA probes. If the sense and antisense probes at a dilution of 1/10,000 times produce vi sible bands (Data / Figure not shown) that correspond with the mo lecular weight of the lin earized plasmid used to complementary base pair with the probes then the probes will be used with O. wendtii embryos in whole mount in situ hybridization. Whole Mount In Situ Hybridization Whole mount in situ hybrid ization was carried out on brittle star embryos of varying developmental time points from fe rtilization to pluteus larva (>75 hours post fertilization). Embryos used for whole m ount in situ hybridiza tion consist of the following time points; late cleavage, 17 hr. blastula, 26 hr. mesenchy me blastula, 39 hr.


48 gastrula, 48 hr. late gastrula and 75 + hr. pl uteus larvae. Embryos used for whole mount in situ hybridization were generated during spawning attempts at the Keys Marine Lab on Long Key, Florida, USA. Embryos were fixed in 4X paraformaldehyde. Following paraformaldehyde fix embryos were washed in to 1X DEPC treated PBS and stored at 4 degrees C until use in whole mount in situ hybr idization. Prior to use in whole mount in situ hybridization embryos will be stepwise washed into 70% EtOH in DEPC treated 1X PBS as has been previously described in echinoderm whole mount in situ hybridization (T. Minokawa et al., 2004). We found that se veral smaller (4 – 5) volume washes work better than two larger washes as described in the Minokawa procedur e. Other than this variation the Minokawa procedure was follo wed for washing embryos into EtOH was followed. Five 1 ml aliquots were prepared of each time point to be used. Each aliquot contains 1 ml 70% EtOH + embryos for that specific time point. Once embryos form each time point are washed into EtOH they can be combined into slide chambers so that each slide cham ber contains several of each developmental time point. For each DIG-labeled RNA probe ther e is a sense (negative control) and anti sense (experimental) probe. Therefore two dishes were set up for use the whole mount procedure for every probe being used. One s lide chamber was used for each sense and antisense DIG-labeled probe being used. In addition if embryos were plentiful enough one dish would be set up for use with the highly reproducible Al x1 labeled probe. This labeled probe would be used as a positive c ontrol. Pre-hybridization fluid was made as follows; for a total volume of 10 ml add; 7 ml formamide 1 ml 1M MOPS pH 7, 1 ml 5 M NaCl, 10 micro liters Tween 20, 1 micro lite r 1 mg / micro liter BSA, 998 micro liter nanopure water. Embryos were washed into pre-hybridization fluid 2 – 4 times or until


49 slide chamber is nearly full. Pre-hybridiza tion was for three hours at 50 degrees C. Pre-hybridization fluid was replaced w ith an equal volume of hybridization solution (pre-hybridizati on solution + 500 ng / ml DI G-labeled RNA probes). Hybridization was for two and a half days at 50 degrees C. Pre-hybridization fluid was added as needed to replace fluid lost via ev aporation over the course of the two and one half day hybridization. Embryos were washed 5 times with MOPS buffer. MOPS = for a 200 ml volume add; 20 ml 1M MOPS, 20 ml 5M NaCl, 200 micro liter Tween 20 and 159.8 ml nanopure water. Blocking was done by incubating in 495.5 micro liters of MOPS buffer with 0.5 micro liters of 10 mg / micro liter BSA at room temperature for 20 minutes followed by incubation in 449.5 micro liters of MOPS buffer with 0.5 micro liters 1 mg / micro liter BSA and 10 % sheep serum (50 micro liters sh eep serum) at 37 degrees C for 30 minutes. Addition of antibody: Remove blocking reagen t from embryos and add 750 micro liters of antibody solution per slide chamber. 750 ul of antibody solution (494.17 micro liters MOPS, 0.5 micro liters 1 mg / micro liter BSA, 5 micro liters sheep serum and 25 units (.33 micro liters) Anti-DIG-Ab) was added And the embryos were incubated overnight at 4 degrees C without shaking. The O. wendtii embryos are very fragile and shaking overnight after the stress and manipulati on involved in the whole mount in situ hybridization procedure result in damaged or destroyed embryos. Antibody solution was removed and the embryos were washed with 400 micro liters of PBT (PBS with1 mg/ml BSA and 0.02% Tween 20) at room temperature. This was repeated 5 times. PBT was removed and replaced with AP buf fer (100mM Tris-HCl, 100mM NaCl, 5mM


50 MgCl2, 0.05% Tween 20, pH 9.5). Incubation was at room temperature for 5 minutes. AP buffer wash and incubation was repe ated a total of three times. 300 micro liters of fresh AP buffer was added to each slide chamber followed by200 micro liters of BM purple. BM purple was added in the dark. Slide chambers were covered with aluminum foil and placed in a desk drawer carefully. Staining has occurred in as little as 15 minutes and taken as long as 4 hours depending on probe strength. Embryos were periodically checked with a dissecting microscope to determine if adequate staining has occurred. When the desired color develops staining was stopped by removing the BM purple with a 200 micro liter pipette. 600 micr o liters of PBT was used to wash the embryos 3x. Staining was visualized with a Nikon Diaphot inverted microscope using a 20X phase contrast objective.


51 Results Animal Husbandry / Spawning Animal Collection We have been 100% successful in obtaini ng animals from the location we identified under the bridge in Long Key, FL. The numbe r of animals obtained varies somewhat, with an average of 18-20 per diver. Figure 9 shows the results per di ver of a series of dives from August 2006. We generally had 2 – 4 divers collecting. As a result, an individual dive would yield 40 to 80 animals. Figure 9. The number of animals obtained/diver from a representative series of dives. In this case, 2 divers were collecting.


52 Our ability to maintain the O. wendtii in the Keys Marine Lab and transport them to USF also approached a 100% success ra te. We have also developed methods to maintain the adults in aquarium s in the laboratory at USF. It has been our experience that with minimal care and maintenance O. wendtii can be kept viable in a laborator y setting for a period of up to and possibly longer than one year. It becomes very difficult to tell apar t different animals from different collection trips and therefore proper iden tification and dating of arriva l times is a challenge. We have experienced cases where one tank of animals remains unchanged and no new O. wendtii are added for the period of one year. A key to successful O. wendtii care and heath is identifying sick and or dying individuals before they foul the entire tank. O. wendtii are particularly sensitive to the death of one organism and res pond quite rapidly to the death of one of their tank mates. It has been observed that following the death of one brit tle star that a tank can foul and all individuals can be lost in as litt le as 2 days if the dead or dying O. wendtii is not removed from the tank. The use of a recovery tank wh ere suspected sick or dying animals can be placed is crucial to the care and longevity of lab kept O. wendtii We utilized a 15-gallon recovery tank that was maintained identical to the tanks that contained healthy individuals. Tank maintenance is very minimal when caring for O. wendtii The salinity of the water is kept very high due to the high salinity of the Florida bay are of the Florida Keys. Salinity can range from 1.026 – 1.030 and is balanced by the addition instant ocean to raise salinity or autoclaved water to lower salinity. The mo st accurate way to keep the water salinity acceptable is to add collected Fl orida Keys seawater to account for lowered


53 water levels due to evaporation. Once the s upply of Florida Keys collected seawater is exhausted, Instant Ocean and nanopure water ar e used to substitute. Animal health and mortality do not seem to be affected by th e change from seawater to Instant Ocean. Tanks were heated to an average temper ature of 80 degrees in the lab. However, during frequent power failures or heat source s expiring water temperatures would drop to ambient air temperature. Never during any of these lower temperature situations was there any observed decrease in O. wendtii health or longevity. In their natural environment O. wendtii are scavengers and o pportunistic feeders. In the lab kept environment we fed them a diet of frozen aquarium shrimp and or standard aquarium fish food. Feeding would occur on average every third to fourth day. Animals would be fed by cutting small segments of shrimp and placing them in the general vicinity of each O. wendtii in the tank. O. wendtii Spawning (Keys Marine Lab, Layton, FL) Animals collected while at the Keys Ma rine Lab are found to routinely spawn under the following circumstances or conditi ons. Spawning attempts took place on a full moon night +/2 days. Outside of this 5-da y period each month there has been little success with spawning of animals. Animals were brought to the lab and assessed for health and viability. Once unhealthy animals ha ve been isolated an attempt to separate animals based on known females suspected females and probable males should be undertaken. This increased the chances of placi ng at least one animal of each gender in each spawning tank as well as decrease the chances of selecting a female for KCL


54 injection, which will kill the animal without producing gametes. Once animals have been placed in appropriate gender ca tegories they can be added to spawning tanks in a gender equal fashion. At least three tanks were pr e set up to hold animals. One tank was at ambient temperature and two tanks were equippe d with heaters. One of the heated tanks should have its temperature elev ated to at lest 85 degrees F. The other heated tank should have its temperature elevated to 85 degrees F or hotter up to a temperature of 90 degrees F. It has been observed that almost all spawni ng occur in water that is at least 85 degrees F. As animals are being placed into tanks at least 2-3 suspected males should be reserved for KCL injection and sperm harvest. A sy ringe containing 0.5 – 1.0 M KCL should be inserted into the central disk / appendage junction of each suspected male. KCL should be slowly injected until there is a visible release of sperm (usually 1 – 3 ml of KCL). Often times it is necessary to spread the injections over several locations on the specimen. Once sperm is visible as a milky white film on the virtual su rface of the animal the brittle star was placed into a beaker containing 50 – 100 ml of water and allowed to shed all sperm before being discarded. This procedure should be repe ated with 1 – 2 more males depending on success and yield of sperm. The 100 ml beaker of sea water / sperm solution can be used directly to seed tanks with or it can be added to 50 ml conical vials and spun down and supernatant pouted off to c oncentrate the sperm fo r transport back to the university. Once animals are segregated into their respective tanks and sperm has been harvested animals should be allowed to acclimate for a period of at least one hour in the dark or until after sunset, which ever is longer, prior to light shock or sperm solution addition. Following a period of acclimation a nd after sunset the animals were light shocked by turning on a single bright light source in the room (bright desk lamp held high


55 over the tanks) for a period of at least 15 minut es. In addition to light shock a volume of 2 – 3 ml of seawater / sperm solution may be added to the tanks. Animals should be observed for signs of spawning ( males release of milky white film that clouds tanks and female release of dark purple eggs that look like small debris on the floor of the tank). If no spawning event takes place the cycle of da rk (45 – 60 minutes) and light shock (15 – 30 minutes) was repeated until dawn. This met hod if attempted over the course of 3 -5 nights routinely produces O. wendtii embryos. Once optimized this procedure proved to always produce fertilized eggs over the cour se of at most two evenings of attempted spawning. During the spawning events that t ook place while housed at the Keys Marine Lab it was impossible to accurately determin e how many males or females were induced to release gametes since multiple animals we re present in the containers. The success rates of various methods, as defined as a spaw ning event in a given container of animals is shown in figure 10. Method % success KCL injection/males 33 KCL injection females 0 Light shock alone 16 Heat alone 2 Light Shock and heat 22 Figure 10. Spawning method success rates O. wendtii spawning (U.S.F. Lab, Tampa, FL) Animal spawning at the U.S.F. lab was optimized to mimic that of the Keys Marine Lab as closely as possible. Anim al spawning was attempted 17 times over the


56 course of the last four year s. Any spawning attempt perfor med while at U.S.F. yielded only sperm. Sperm yielded from spawning attemp ts while in the lab Tampa lab at U.S.F. was derived under the following conditions: 2 – 4 O. wendtii were placed into 500 ml – 1 L beakers of Florida keys collected sea wa ter according to 50% male and 50% female suspected rations on an evening +/2 days surrounding a full moon. The beakers of sea water themselves was heated in a water bath in a dark humid storage closet to simulate the conditions of the Keys Marine Lab. For successful release of male gamete, water temperature was raised up to at point be tween 85 and 90 degrees F for a period of no longer than two hours. Intense light shock was applied every 45 minutes that consisted of placing a desk lamp directly ove r each individual beaker of an imals. 2 – 4 ml of seawater / sperm solution was added during the initial light shock period. Using this procedure male O. wendtii could be reliably induced to releas e sperm. Any attempt at light shock stimulus after a two-hour time point failed to yield any gamete release. Similarly no brittle star was recorded to release any game tes following 2 hours of this procedure. In addition any supplementation of seawater / sperm solution following the initial light shock not only failed to induce spawning but further seemed to increase the rate of specimen death and ill health. Using this method it could be expected th at at least 50% of all animals used would fail to recover from the stimulus and die within 24 hours of a spawning attempt. The average amount of an imals used during these attempts was 12. The most used was 30. The least used wa s 8. The average case of reported males releasing gametes was 2. The most males reco rded to be induced to release sperm was 5. The least cases of males releasing gametes was 0. Never in any attempt was there a recorded case of a female O. wendtii releasing eggs.


57 Figure 11. Ophiocoma wendtii embryonic development. Panels pro ceed from fertilized egg in upper left and are labeled as follows: 88cell, 16= 16 cell, 32 = 32 cell and 64= 64 cell cl eavage stages. B= blastula, EG = early gastrula, MG = mid-gastrula, G= gastrula, P = pluteus. The arrow shows a mid-gastrula stage embryo that was squashed with a cover slip to show the forming skeleton. O. wendtii Embryonic Development Figure 11 shows the development of Ophiocoma wendtii from fertilized egg to pluteus larva. The zygote unde rgoes holoblastic, equal cleavag es, including the fourth cell division (the products are labeled 16 in Figure 11), which is unequal in sea urchins. As a result, unlike sea urchins, no micromer es are formed. Ingression of mesenchyme cells into the blastocoels occurs at a time and stage similar to sea urchins, however a larger number of cells ingress (labeled EG in Figure 11). After invagination of the gut, the mesenchyme cells sort out into two positions in the embryos. One set becomes associated with the tip of the archenteron and will form the coelomic pouches. In sea urchins, three are mesenchyme cells associated with the tip of the archenteron, but they arise from the archenteron itself following inva gination. Their fate is different as well,


58 since most form pigment cells, which are ab sent in brittle stars. The other set of mesenchyme cells gather at th e base and on either side of the archenteron. These are the cells that will make the skeletal spicule. Their position and behavior appear identical to the primary mesenchyme in sea urchins from this point on. Biomineralization begins as these cells gather near the ar chenteron (see the sq uashed mid gastrula (MG) in Figure 11. Gene Regulatory Network / Molecular cDNA Synthesis and PCR Control We tested our synthesis of embryonic c DNA synthesis using actin PCR primers. Amplification of both early ga strula and gastrula cDNAs w ith actin primers generated a product of the expected 300 bp size (Figure 12). The control re action without cDNA added yielded no product. From this we conc luded that we had successfully generated cDNA.


59 Figure 12. Positive control gel of O. wendtii cDNA using actin PCR primers. Lane 1; no cDNA control, Lane 2; early gastrula cDNA, Lane 3; gastrula cDNA, Lanes 5 and 6; DNA size markers. PCR Using Degenerate Dri Primers PCR with degenerate Dri primers us ing early gastrula and gastrula cDNA as target yielded very little product, but wh at was visible using a UV light box was of the expected size of around 300 bp (Figure 13). We purified the reaction an d ligated into the vector as described in methods.


60 1 2 3 4 5 Figure 13. Result of Degenerate Dri primer PCR. Lane 1; early gastrula cDNA Lane 2;gastrula cDNA, Lane 3; blank, Lanes 4 and 5, DNA size markers. After ligation of the PCR products in to vector and transformation into E. coli individual colonies were pi cked and crude minipreps of plasmid DNA were performed on each. EcoR1 digests were performed on each plas mid and the results were separated on a 1% agarose gel (Figure 14). The sizes of the in serts clearly show that the products of the PCR reaction were heterogeneous, indi cating that more than one cDNA had complementary sequences to the degenerate pr imers and that we lik ely had products that were not from the Dri transcript.


61 Figure 14. Crude mini prep Dri degenerate primer PCR. Each lane contains an individual plasmid prep digested with EcoR1 to release th e inserted cDNA. The lanes containi ng DNA size markers are indicated. Plasmids in lanes where inserts are circ led were then purified and sequenced. We chose a number of individual clones w ithin the size range that was expected based on comparative organisms. The plasmids chosen were purified as described so that they could be used for sequencing. The re sulting plasmids are shown in Figure 15.


62 Figure 15. Purified plasmid preparation of Dri degenerate primer PRC products. Numbered lanes indicate individual plasmids that were purified for sequencing. The 100 bp and 1 KB are DNA size ladders. Sequence Analysis of Putative Dri Clones The plasmids were sequenced as descri bed in methods. The sequences obtained were trimmed of vector sequences and then compared to NCBI sequence databases using BLAST. Of the eight clones seque nced, one had Dri sequences from S. purpuratus as the closest match. The nucleotide se quence is shown in figure 16.


63 Figure 16. Dri Nucleotide sequence of the O. wendtii PCR fragment. Sequences where primers were made for RACE PCR reactions are highlighted as indicated on the key. Actin Controls for 5’ RACE PCR Following ligation of the adapter to mRNA and cDNA synthesis, actin 5’ RACE controls were run to ensure those pro cesses were successfully completed. Figure 17 shows the result of two reactions One set of reactions clearly worked, as the appropriate sized actin band was present (Figure 17). The other did not, indicating the either the adaptor ligation or the cDNA synt hesis did not work. As a routin e, we used this control to check the quality of our template for 5’ RA CE prior to proceedi ng with the reactions.


64 Figure 17. + control gel for RACE treated and ligated cDNA Dri 3’ RACE 3’ RACE was carried out for Dri sequences as described in Methods. When the products were separated on a 1% agarose gel, a faint, diffuse band was observed. These products were cloned into the TOPO TA vector as describe d in methods, and transformed into bacteria. Colonies were picked and cr ude digests indicated which had inserts and determined their sizes. Seven of these were then sequenced, and of these, one overlapped


65 with our initial fragment of O. wendtii Dri cDNA. The new sequences and their translated protein products are shown in figure 18. A. NNNNAACGGCTT AGATTCGCCT GGGCCTACCTACCAGCATCC ATCACTAG TGCAGCTTTCACTCTTCGTA CACAATATATGAAGTACCT G TATCCATATG AATGTGAGAAGAAAGGTTTAAGCACA CCGGCTGAATTACAAGCAGCGATA GACGGGAATCGGCGCGAGGGACGACGCCCAATATACCATCATGCAGGTCC GGCAGCTGCGGCATTTAGTCTCCATCATTCACGAGCTTCACCGCCACCTA CCATGATACCCCACCCATCTCGTATACCTCTTGGAACCACATTACCTTCA CCACTGTCATTGTTAACATCAGAGGAAGAACACCTTGCCCTAGCAGCTGC ACGTCAGGAATTCAGTTTCAAGCACACT CTTATGCTGGAGCGAGAACGAG AACGTGAAAGAGAACGTGAAAGAGAGC GTGAACGAGAAAGAGAACGTGAA AGAGAAAGGCTACGTGAGCGAGAG AGGAAAAGAAAGAGAGAGTAAAAAAA AAAAAACCTATAGTGAGTCGTATTAATTCGGATCCGCG AAGGGCGAATTC B. R F A W A Y L P A S I T S A A F T L R T Q Y Met K Y L Y P Y E C E K K G L S T P A E L Q A A I D G N R R E G R R P I Y H H A G P A A A A F S L H H S R A S P P P T Met I P H P S R I P L G T T L P S P L S L L T S E E E H L A L A A A R Q E F S F K H T L Met L E R E R E R E R E R E R E R E R E R E R E R E R L R E R E R K R K R E Stop AAAAAAAA = 3’ drifin4108 Primer AAAAAAAA = Previously known sequence AAAAAAAA = Start of new sequence AAAAAAAA = 3’ inner race primer Figure 18. Results of 3’ RACE extension of Dri sequence. A; nucleotide sequence, B. putative amino acid sequence. The key belo w identifies key sequences. Assembling of Dri Sequences Alignments of S. purpuratus Sequences generated for the O. wendtii Dri gene began by analyzing homologous regions of the Dri gene found in various ot her species and then designing degenerate


66 primers with which brittle star cDNA coul d be screened for a homologous Dri gene. Using degenerate primers, perfect primers and the RACE technique of transcript end amplification a 1053 base pair segment of the O. wendtii Dri gene has been generated. This segment contains what appears to be the 3’ end of the gene and the bulk of its body. The 5’ start has not yet been sequenced (F igure 18). Alignments of the nucleotide sequences with S. purpuratus show some c onservation (Figure 19A). Comparison of the deduced amino acid sequ ence with that of S. purpuratus shows a much higher degree of similarity, as was expected. For the fragment assembled, there is 81% similarity between the two organisms and 60% identity. A. Figure 19. O. wendtii Dri sequence alignment with S. purpuratus A; Nucleotide sequences, B; De duced amino acid sequences.


67 B. Figure 19. Continued. 5’ RACE of Alx1 The initial Alx1 sequence obtained from O. wendtii by Mary Harmon was used to identify primers that could be used for 5’ RACE. The position of the primers is shown in Figure 20. Target cDNA from Gastrula that ha d been verified by actin controls was used as template for 5’ RACE PCR as described in methods. When separated on a 1% agarose gel, the products appeared primarily as a 500 bp band. When cloned and sequenced, 3 out of 6 cloned products were the 5’ end of Al x1. The nucleotide and translated amino acid sequences obtained from these clones are s hown in Figure 21. A potential start codon was


68 observed at the beginning of an extended ope n reading frame, and their were stop codons 5’ to that AUG. A. Q V W F Q N R R A K W R K R E R F Q Q L CAGGTTTGGTTTCARAA TCGTCGAGCCAAATGGCGGAAGCGTGAAAGGTTTCAGCAACTCCAG Q G M R G GGCATGCGCGGA L G P G G G Y E M P I A P R P D A Y A Q ATCGGGCCAGGTGGGGGCTATGAGATGCCTAT CGCCCCTCGTCCCGACGCCTATGCTCAGGTG V S P P G AGTCCCCCTGGTT B. MLGYTSEKGIGLIWLTACADTTPHKEMTPLSTSNTTIGSESTKVDAITAERENGM TSPVSTKVEPDANANTPKTEDNNNKDD DAKSTDGDSKSNGDDSKRKKRRNRTT FTSFQLEEMERVFQKTHYPDVYCREQLALRCDLTEARY Figure 20. 5’ RACE amplification of Alx1 5’ end. A; indicates 5’ end of the original PCR fragment of Alx1. The arrow indicates the sequence and direction of the PCR primer. B; pututative protein product of new sequences obtained. Alx1 Assembly and Alignment With S. purpuratus Alx1 The complete Alx1 coding sequence has b een generated usin g the initial PCR product and the RACE technique. The DNA bi nding domains of Alx1 from the two species are nearly identica l at the amino acid level ( O. wendtii AA# 190250). Outside the DNA binding domain there is less simila rity with the exception of the carboxy terminal of the proteins where ther e is a putative trans-activating domain.


69 Figure 21. Complete Alx1 gene sequence alignment Dig-Labeled Probe Preparation an d Sensitivity Assay via Dot Blot DIG-labeled probe preparation for both Dri and Alx1 were carried out as described in methods. Plasmids were linearized to comple tion as seen on agaros e gels (Figure 23). Probes were synthesized as described in the methods section.


70 Figure 22. Directional Plasmid Digest. Plasmids were digested with either Spe1 (lane 3) or Not1 (lane 4) and products were separated on 1% agarose gels. Undigested plasmid (Lane 5) was used for comparison. Complete digestion was considered to be the absence of supercoiled and nicked circle forms of the plasmid and a corresponding ap pearenceof a linearized plasmid band. Dilutions of the same plasmid used to make the DIG-labled RNA probes were spotted on membranes and cross-linked. Hybr idization and visualization of the bound probe indicated that we had successfully synt hesized probes. The relative sensitivity of the probes compared to probes th at had been empirically tested experimentally was used as an assay for probe synthesis in sufficien t quantity to be useful. Figure 23 shows one example of the assay.


71 Figure 23. Probe Quantification. Dri sense (S) and anti-sense ( ) probes were compared to Alx1 probes used successfu lly for Southern blots (+ control). Whole Mount In Situ Hybridization Using Alx1 probes Alx1 whole mount in situ hybridization has produced results that are reproducible and appear to be exemplary of skeletal forma tion in at least gastrula and late gastrula stage embryos.


72 Figure 24. (BEST photo) Whole Mount In Situ Hybridization Alx1 expression. Spatial and temporal expression of O. wendtii Alx1 during embryogenesis. All views are lateral. Staining follows a similar pattern to that seen in the sea urchin. Temporal Expression During Development Using Semi-Quantitative PCR for Alx1 Figure 25. Semi-quantitative PCR O. wendtii Alx1. Lanes represent stages of development where total RNA was isolated and used as a template for limited-cycle PCR. Products were detected by Southern blotting and chemiluminescent detection. The stages are indicated on the legend.


73 To determine the relative levels of Al x1 gene expression semi-quantitative PCR was carried out. In this modi fied version of quantitative PCR a limited number of PCR cycles is performed. Alx1 levels are semiquantified for developmental time points; egg, 16-32 cell stage, blastula, me senchyme blastula, gastrula, and pluteus larva. The PCR cycle was halted during the exponential growth phase during semi-quantitative PCR of cDNA generated from developmentally critical time points of brittle star development. This visualization and semi-quantification resu lts in a time course of expression for the Alx1 gene in the brittle star. A gel was r un of the resulting PCR products. The gel was transferred to a nitrocellulose membrane and standard southern blotting is performed. The relative amounts of each band re presents the relative levels of Alx1 expression found in that respective stage of development. In the case of the brittle star Alx1 expression is not seen until the blastula stage. Expression re mains constant and level until a marked increase in transcription occurs during th e gastrula phase of development. Once gastrulation is complete a nd the pluteus stage begins O. wendtii Alx1 expression decreases back to blastula / mesenchyme bl astula amounts. This devised method provides us with a semi quantitative measure of the level of expression for each stage of development. From this result we have determ ined that Alx1 is expressed in skeletogenic mesenchyme cells. The expression pattern indi cates that Alx1 is not expressed in all mesenchyme that ingresses indicating that possibly in brittle stars some of the mesenchyme that ingresses ma y give rise to the equivale nt of what is secondary mesenchyme in the sea urchin. However, there are also no pigment cells in the brittle star unlike the sea urchin.


74 Dri Whole Mount In Situ Hybridization DIG-labeled RNA probes designed for Dri ge ne expression in developing brittle star embryos have yet to produce co nsistent or validated results. Phylogenetic Analysis Figures 26 and 27 provide a visualize representation of the evolutionary relationship of O. wendtii Alx1 and Dri proteins to ot her Alx1 and Dri proteins. Sequences attained from brit tle stars have been aligned using Clustal X in order to produce a phylogenetic alignment of Alx1 gene s (figure 26) and separately Dri genes (figure 27). These aligned sequences have then been used to make MEGA neighbor joining trees. Bootstrap values (1000 replications) are given. Figure 26. Alx1 neighbor joining tree using complete Alx1 O. wendtii sequence


75 The results of the phylogenetic anal ysis of the complete sequence of O. wendtii Alx1 skeletogenic gene is represented in th e neighbor-joining tree in found in figure 26. The neighbor joining tree represents co mmonly used distance matrix method of calculating genetic distance from the alignment of the three sea urch in sequences (S.purp, P.lividus and L.lytechinus) as well as the brittle star Alx1 sequence and that found in Mice and humans. By calculating genetic di stance and not directly implying an evolutionary model we can conclude from this tree that the brittle st ar groups outside the clade containing its fellow sequenced echinode rms represented here by the presence of three sea urchins. The results were attained using the fu ll length of each resp ective brittle star sequence aligned against equivalent sequen ces from the varying comparative species. The brittle star O. wendtii Alx1 groups with the three other members of the phylum Echinodermata. The two sea urchins (S. purp and Lytechinus) form a node with each other as well. This suggests th at the Alx1 gene is more sim ilar or more closely conserved within the phylum Echinodermata than when co mpared to the other phylum. This result is consistent with the current consensus on deuterostome phylogeny. Figure 27. Dri neighbor joining tree using 1053bp O. wendtii dead ringer sequence


76 The results of the phylogenetic analysis of the partial (1053 bp) transcript of the O. wendtii Dri dead ringer gene is represented in the neighbor joining tree found in figure 34. In this case the O. wendtii transcript groups more closel y with the starfish, Asteria miniata than with the sea urchin, S. purpurat us. This would lead one to believe that the starfish and brittle star have a closer evolutionary relations hip when compared by this Dri gene sequence. Therefore, it would be a logi cal assumption that the sea urchin and the brittle star may have devel oped the characteristic of the embryonic skeleton in two separate events as apposed to the developm ent of an embryonic skeleton occurring in a shared common ancestor.


77 Conclusions Both Alx1 and Dri sea urchin genes have O. wendtii homologues. The homologous sequences of O. wendtii Alx1 and Dri seem to show an expression pattern similar to those found in the sea urchin. W hole mount in situ hybr idization show the expression of Alx1 consistently at gastrulation in the brittle star. This Alx1 expression is consistent with that seen in the sea urchi n. Semi quantitatively we have shown that Alx1 is expressed in the skeletogeni c mesenchyme of the brittle star (figure 25). However, not all of the skeletogenic mesenchyme that initia lly ingress in the brittle star appears to be expressing Alx1. It is therefor e thought that some of the ingressing mesenchyme of the brittle star may give rise to the equivalent of the secondary mesenchyme found in the sea urchin. From our whole mount in situ hybridi zation and semi quantitative PCR results is was possible to construct a skeletal gene regulatory network comparison (figure 29). In addition to this gene regulatory network comparison an endomesodermal developmental comparison was developed (figure 28) using both photographic and gene expression data of Brittle Stars and Sea Urchins. Figure 28 di splays the lack of micromeres at the 5th cleavage division, the apparent ingression of both primary a nd secondary mesenchyme at the 24th hour of development as well as the lack of pigmented cells found in the brittle star when compared to the development of th e sea urchin. These tra its compose the three distinct and apparent differences found through out our investigation.


78 Figure 28. Sea urchin and brittle star endomesoderm developmental comparison Figure 29. Sea urchin and brittle star skeletal gene regulatory network comparison


79 Dri RNA probes have yet to yield reproduc ible results. However, in the several whole mount in situ attempts using RNA probes created from O. wendtii sequenced DNA there seems to be a consistent increase in stai ning from the gastrula stage embryos on into development. This would suggest a sim ilar expression pattern to that fond in O. wendtii Alx1 (figure 29). Unfortunately, this result is not consiste nt or reproducible enough to make a concrete conclusion. In addition photographic evidence of this observed phenomenon has been elusive. It appears that phylogenetica lly both brittle star Dr i and Alx1 sequences group outside the clade of the sea ur chin. This would lead one to believe that the development of a embryonic skeleton may have developed as a separate trait in each classes as apposed to once in a common ancestor. Based on our sequencing, expression analys is and phylogenetic analysis it is our conclusion that there is a similar kernel of a gene regulatory network at work in both Ophiuroid and Echinoid classe s during embryonic skeletal ge nesis. Conserved genes in this kernel are at least Dr i and Alx1. We speculate that the embryonic activation of the skeletogenic gene network occurred inde pendently and probabl y includes different mechanisms but the kernel required for skelet al formation in brittle stars and sea urchins is the same in both classes. Furthermore, it can be speculated that activation of the gene regulatory network responsible for embryonic skeletogenesis in both the sea urchin and the brittle star would be the result of a separate mechanism. The mechanism in the sea urchin has been shown to be the activation of the repressor Pmar1 and its repre ssion of the repressor HesC


80 (figure 29). This repression of a repressor is responsible for activation of the gene regulatory network that results in the em bryonic skeleton of the sea urchin. Pmar1 expression has yet to be identified in the brittle star. Therefor e the mechanism of activation may be unique in the brittle star. It would be a lo gical next step to try and identify, in brittle stars, expression of the newly char acterized skeletogenic gene regulatory network repressor HesC that has been identified in sea urchins (figure 29). If indeed a unique (non Pmar1) activation occu rs in brittle stars determining if the activation includes repre ssion of HesC would be valuable information. Lastly, this investigation supports the idea of gene regulatory ne twork kernel is a class of gene regulatory network, which becau se of their developmental role and their particular internal structure, are most impe rvious to change (Davidson et al., 2006). As exemplified here the skeletal gene regulatory network of the echinoderms, sea urchin and brittle star, contains conserve d regulatory genes. Where interf erence with one or more of these regulatory genes would likely destroy ke rnel function, activation of these kernels can be a mechanism of evolutionary change in gene regulatory networks over time. The developing animal body plan is controlled by th e action of gene regulatory networks. It is therefore concluded th at evolution must be dependan t on change in gene regulatory networks. This change can come in the form of disruption of or a ch ange in the activation of a gene regulatory network producing cha nges in development, body plan, speciation and adult morphology. In echinoderms it is the brittle star’s and s ea urchin’s embryonic activation of the skeletal gene regulatory network not seen unti l the adult form in Asteroideans, Holothuroideans and Crinoidean s that give evidence for the evolution of the skeletal gene regulato ry network in echinoderms.


81 References AMORE, GABRIELE et al. (2003). Spdeadringe r, a sea urchin embryo gene required separately in skeletogenic and oral ectode rm gene regulatory networks, Developmental Biology, 261, pages 55-81 BENSON, S. et al., (1986) The organic matrix of the skeletal spicule of sea urchin embryos, The Journal of Cell Biology, vol. 102, pages 1878 1886 BENSON, S. et al., (1987) A lineage specific gene encoding a major matrix protein of the sea urchin embryo spicule, De vl. Biol., vol 120, pages 499 506 BEHRENS et al., (1996) Na ture, vol. 382, pages 638 642 BIRKLAND, C. E. (1989) The faustain traits of the crown of thorns starfish, Amer. Sci., no. 77, pages 154-163 BOVERI, T. ZEWI (1901) Theodor Boveri on cytoplasmic organization in the sea urchin egg, Landmarks in Developmental Biology, vol. 14 BOVERI, T. ZWEI, (1918) Zeie Fehler quellen bei Merogoneieversuchen und die Entwicklungsfaheit merogoneisch er und partiellmerogoneischer Seigelbastarde, Organ, 44, pages 417 471 CADIGAN, K.M. and NUSSE, R. (1997) wingl ess signaling in the drosophila eye and embryonic epidermis, Development, vol. 122, Issue 9, pages 2801 2812 CAMERON, ANDREW R. et al. (1989) The oral -aboral axis of a sea urchin embryo is specified by first cleavage, Deve lopment, vol. 106, pages 641 647 CAMERON, ANDREW R. et. al. (2000) A s ea urchin genome project: Sequence scan, virtual map, and additional resour ces, PNAS, vol. 97, no. 17, pages 9514-9518 DAN-SHOKAWA and FUJISAWA ( 1980) Cell dynamics of the blastulation process in the starfish Asterina pectinifera, Devel opmental Biology, vol. 77, issue 2, pages 328 -339 DAVIDSON, E.H., HINMAN, V.F., NGUYE N, A.T., CAMERON, A.R., (2003) Developmental gene regulatory network architecture across 500 million years of echinoderm evolution, PNAS vol. 100, no. 23, pages 13356-13361


82 DAVIDSON, E.H. AND DOUGLAS, E.H. (2006) Gene Regulatory Networks and the Evolution of Animal Body Plans Science, vol. 311, pages 796-800 DAVIDSON, E.H. AND LEVINE, M. (2005) Gene regulatory networks for development, PNAS, vol. 102, no. 14, pages 4939-4942 DAVIDSON, E.H. et al. (2002) A Provisi onal Regulatory Gene Network for Specification of Endomesoderm in the Sea Urchin Embryo, Dev. Bio., 246, pages 162190 DAVIDSON, EH. S. B.-T. d.-L. a. E. H. (2007) Gene Regulation: Gene Control Network in Development. The Annual Review of Biophysics and Biomolecular Structure 36 : 191-212. DAVIDSON, EH. (2006) The regulatory ge nome. Gene regulatory networks in development and evolution. San Diego, CA: Academic. DAVIDSON, E.H. et al. (2002) A Genomic Regulatory Network for Development, Science, vol. 295, pages 1669-1677 ETTENSOHN, CHARLES A., ILLIES, MOCH ELE R., OLIVERI, PAOLA AND DE JONG, DEBORAH L. (2003) Al x1, a member of the Cart 1/Alx3/Alx4 subfamily of Paired – class homeodomain proteins, is a nd essential component of the gene network controlling skeletogenic fate specification in the sea ur chin embryo, Development, 130, 2917-2928 ETTENSOHN, CA and SWEET, HC (2000) Patter ning the early sea urchin embryo, Dev. Biol., vol. 50, pages 1 44 ETTENSOHN, C.A. and INGERSOLL, E.P. (1992) Morphogenesis of the sea urchin. In “Morphogenesis: An Analysis of the Development of Biologi cal Form”, Macrel Dekker, New York, pages 189 262 GAO, FENG AND DAVIDSON, ERIC H. (2008) Tr ansfer of a larg e gene regulatory apparatus to a new developmental addre ss in echinoderm evolution, PNAS, vol. 105, no. 16, pages 6091-6096 GERHART et al., (1984) Cell cycle dynamics of an M-phase specific cytoplasmic factor in Xenopus laevis oocytes and eggs, J. Cell Biol., vol. 98, pages 1247 1255 HENDLER, G. et al., (1995) Sea Stars, Sea urch ins and allies. Smithsonian Institution, p. 90 195 HINMAN, V.F. et al. (2007) Caught in the ev olutionary act: Precise cis regulatory basis of difference in the organization of gene netw orks of sea stars and sea urchins, Dev. Bio., dio:10.1016/j.ydbio.2007.09.006


83 HINAMN, VERONICA F., DAVIDSON, ERIC H. (2003) Expression of a gene encoding a Gata transcription factor during embryogenesi s of the starfish Asteria miniata, Gene expression patterns, 3, 419-422 HINMAN, VERONICA F., DAVIDSON, ERIC H. (2003) Expression of AmKrox, a starfish ortholog of a sea ur chin transcription factor essential for endomesodermal specification, Gene expr ession patterns, 3, 423-426 HYMAN L.H., (1955) The Invertebrates. vol. 4, McGraw-Hill Book Company, pages 589-689 KILIAN C.E., WILT F.H (2008) The dynamics of secreti on during sea urchin embryonic skeleton formation, Exp. Cell Res., no. 314, pages 1744 1752 KILIAN C.E., WILT F.H. (1996) Characterization of the pr oteins comprising the integral matrix of Strongylocentrotus purpuratus embryonic spicules, J Biol. Chem., vol. 271, pages 9150 9159 LIVINGSTON, BRIAN T. AND WILT, FRED H., ( 1990) Range and stability of cell fate determination in isolated sea urchin blastomeres, Development, 108, pages 403-410 LIVINGSTON, B.T. et al., (1991) Matrix and mineral in the sea urchin larval skeleton, Journal of Structural Biology, vol. 126, issue 3, pages 216 226 LIVINGSTON B.T. et al., (1996) Expression of endoderm specific genes are affected differently by inhibitors of two extracellular matrix molecules, Mol. Biol. Of the Cell, vol. 7 LOGAN et al., (1998) Nuclear beta-catenin is required to specify vege tal cell fates in the sea urchin embryo, Development, vol. 126, pages 345 357 MINOKAWA, T. and HIDEWO NAKATA, (200 4) Expression patterns of wnt8 orthologs in two sand dollar species with different developmental modes, Gene Expression Patterns, vol. 9, Issue 3, pages 152 157 NEWPORT, J. and KI RSCHNER, M. (1984) Regulation of the cell cycle during early Xenopus development, Cell, 37, 731 742 NICHOLS, D. (1969) Echinoderms, 4th edition. Hutchinson and Co., London OLIVERI, P., DAVIDSON, E.H. (2004) Gene regulatory network controlling embryonic specification in the sea urchin, Current opi nions in Genetics and Development, 14:351360 OLIVERI, PAOLA, DAVIDSON, E.H. AN D MCCLAY, DAVID R. (2003) Activation


84 of pmar1 controls specificat ion of micromeres in the sea urchin embryo, Developmental Biology, 258, pages 32-43Specification in the Sea Urchin Embryo, 2002, Dev. Bio., 246, pages 209-228 OLIVERI, P., CHARRICK, D.M., DAVIDSON, EH., (2002) A regulatory gene network that directs micromere specification in the sea urchin embryo, Dev. Biol., Vol. 246, pages 209 228 RANSICK A. and DAVIDSON, E. (1999) Specifi cation of cell fate in the sea urchin embryo: summary and some proposed mechanisms, Development, vol. 125, pages 3269 3290 RANSICK, A. AND DAVIDSON, E.H. (1995) Micromeres are required for normal vegetal plate specification in sea urchin embryos, Development, vol. 121, pages 3215 – 3222 RANSICK, A AND DAVIDSON, E.H. (1993) A complete second gut induced by transplanted micromeres in the sea urchin embryo, Science, vol. 259, pages 1134 1138 REVILLA-I-DOMINGO, ROGER AND DAVIDSON E.H. (2003) Developmental gene network analysis, International journal of Developmental Biology, 47, pages 695 – 703 REVILLA-I-DOMINGO et al., (2007) A missi ng link in the sea urchin embryo gene regulatory network: hesC and the double ne gative specifications of micromeres, PNAS, vol. 104, pages 12383 12388 Sea Urchin Genome Consor tium et. al., (2006) The genome of the sea urchin Strongylocentrotus purpuratus, Sc ience, vol. 314, no. 5801, pages 941 952 SHULMAN et al., (1998) Differential recrui tment of disheveled provides signaling specificity in the planar ce ll polarity and wingless signa ling pathways, Genes Dev., vol. 12, pages 2610 2622 SMADAR BEN-TABOU DE -LEON AND DAVIDSON, E.H. (2007) Annual review of Biophysics and Biomolelucular structure, 36, pages 191-212 SMITH, L., THEODORIS, C., DAVIDSON, E.H. (2007) Dynamic Developmental Gene Regulation Science, vol. 318, pages 794-797 SUMMERS, R.G. et al., (1993) A sterometric analysis of ka ryogenesis cytokinesis, and cell arrangements during and following fourth cleavage period in the sea urchin Lytechinus varirgatus, Dev. Grow th Diff., vol. 35, pages 41-58 VONICA, A. et al., (2000) TCF is the nuclear effector of the beta catenin signal that patterns sea urchin anim al vegetal axis, Dev. Biol., vol 217, pages 230 243


85 WESSEL GM et al., (1984) Two embryonic, tissue specific molecules identified by a double label immunofluorescence technique fo r monoclonal antibod ies, Journal of Histochemistry and Cytochemistry, vol. 34, pages 703 706 WIKRAMANAYAKE, A.H. et al., (1998) beta catenin is e ssential for patterning the maternally specified animal vegetal axis in the sea urchin embryo, Natl. Acad. Sci., vol. 95, pages 9343 – 9348 WOLPERT L. and GUSTAFSON, T., (19961) Studies on the ce llular basis of morphogenesis in the sea urchin embryo. Dir ected movements of primary mesenchvme cells in normal and vegetalized larvae, Exp Cell Res., vol. 24, pages 64 – 79