CHARACTERIZATION OF OVARIAN TISSUE CULTURE AND INVESTIGATION OF EMBRYONIC RESPONSE TO TEMPERATURE STRESS USING A SHP2 PHOSPHATASE INHIBITOR IN CIONA INTESTINALIS

This dissertation is comprised of two parts, the first is cell culture. In this work I describe the successful primary culture of Ciona intestinalis ovary tissue. Improved cell proliferation in 3D Petri Dish system over culture treated plastic is shown, as well as the efficacy of using C. intestinalis hemolymph as a media additive. The seeded ovarian cells assemble into multicellular aggregations, which mimic the morphology of in vivo ovaries. It is also important to understand the molecular pathways in which marine organisms handle stress and how these pathways change in high stress situations. The second is the role of Shp2 in temperature stress. It is projected that ocean temperatures may raise as much as 4C over the next century, and there is very little known about how this increase in temperature will affect both reproduction and embryonic development in marine organisms. In this work I also describe the phenotypic abnormalities which occur during embryogenesis in Ciona intestinalis when Shp2 protein tyrosine phosphatase is inhibited. The phenotypic abnormalities are consistent with the abnormalities observed in C. intestinalis embryos reared at high temperatures stress. This suggests that Shp2 is an important protein in the amelioration of stress in developing C. intestinalis embryogenesis and is required for normal embryogenesis.

viii LIST OF TABLES  To date there are many different vertebrate tissue culture lines available. Among invertebrates, many insect cell culture lines have been established, but there are no established cell culture lines for marine invertebrates. As for primary cultures, a number of studies have been published for various marine invertebrates (Cai and Zhang 2014;Baruch Rinkevich 2011;B Rinkevich 1999). Focusing on ascidians, there has been much interest in tissue culture in colonial ascidians, because of their ability to develop from asexual buds, which has implications for stem cell biology (Duckworth et al. 2004;Kawamura and Fujiwara 1995;Kawamura et al. 2006;B Rinkevich and Rabinowitz 1993;Rabinowitz and Rinkevich 2004). Of the over 3,000 species of tunicates, there are 35 publications on stem cells and 80 publications on cell culture (Rosner et al. 2021), although many of these are in colonial ascidians. Marine invertebrate cell culture provides a promising advancement in the assessment of impacts of ocean acidification, increasing ocean temperatures, and ecotoxicological impacts of pollutants (Rosner et al. 2021).
In solitary ascidians, such as C. intestinalis, there have been very few published papers on cell or tissue culture but it has been reported for nervous system cells (Zanetti et al. 2007), pharynx (Raftos, Stillman, and Cooper 1990;Sawada, Zhang, and Cooper 1994;Arizza, Cooper, and Parrinello 1997), and hemocytes (Sawada, Zhang, and Cooper 1994;Peddie, Richest, and Smith 1995;Raftos, Stillman, and Cooper 1990;Arizza, Cooper, and Parrinello 1997). Ciona hemocytes in culture only showed proliferation for 3 days, after which cell began to die. When co-cultured with mammalian cells, the hemocytes showed cytotoxic activity towards the mammalian cells (Peddie, Richest, and Smith 1995). Ciona nervous system cells were cultured from larval stage embryos and were grown on culture treated glass. In this culture system, cell death began after 7 days and optimal growth of cells was seen for a maximum of 13 days. Electrophysiological analysis of nervous system cells showed that they were indeed neuronal cells (Zanetti et al. 2007).
In previous work in Styela clava, there has been tissue culture of both pharynx explants and hemocytes (Sawada, Zhang, and Cooper 1994;Raftos, Stillman, and Cooper 1990). The hemocytes alone did not proliferate in culture after the initial plating while pharynx explants survived multiple passing attempts for 72 days in culture. More than 1/4 th of the pharynx primary cultures became overrun by protists and were unusable after 24 days. During the 72 days the pharynx explants were in culture, the tissue remained unchanged from its in vivo morphology with the exception that hemocytes had moved from the tissue into the media (Raftos, Stillman, and Cooper 1990). Another attempt at pharynx explant culture in Styela clava showed survival of the explants for 82 days in culture. In this culture system, it was also found that hemocytes has moved from the explants to become free in the culture media (Sawada, Zhang, and Cooper 1994). Analysis of Styela clava hemocytes released from phanyx explant cultures also showed the release of hemagglutinins which can cause agglutination of cells (Arizza, Cooper, and Parrinello 1997).
Traditional tissue culture methods rely on adhesion of cells to culture treated plastic for cell survival. This forces the cells to form unnatural adhesion to a flat plastic surface so that they are not washed away with media changes. This also means that when cells need to be passed and re-seeded or removed for experiments that chemicals must be used to degrade these adhesions. These methods also change the natural histology of the cells, meaning that it is less likely that the cells would behave the same way in vivo. In order to circumvent some of these problems with traditional tissue culture methods I explored the use of a "3D" tissue culture system ). This approach is to grow the cells in non-adherent microwells so that cells may form more natural adhesions to each other and even self-assemble into "organoids" -three-dimensional cellular aggregations that may replicate the function of in vivo tissues (Achilli, Meyer, and Morgan 2012;Desroches et al. 2012; Morgan et al. 1992). There are many benefits to using this 3D Petri Dish system. There is no need for cells to form unnatural adhesions or to add attachment factors, extracellular matrix, or scaffolding. It allows for cell-cell interactions and communication that would occur in normal tissue. It also allows for normal cell density and the cells can be in contact with media on all sides. The 3D molds are also in an ordered array which makes for easy imaging. There is also no need for any enzymatic treatments to remove cells from the wells. These spheroids can be used for a lot of techniques such as RT-PCR, Western blots, transplantation, immunestaining, and histology .

Ciona intestinalis as a Model Organism
Ciona intestinalis is a solitary ascidian and being part of the Urochordata is the closest invertebrate group to the vertebrates phylogenetically (Corbo, Di Gregorio, and Levine 2001). C. intestinalis is an invasive species to the marine environment of Rhode Island, and is readily available for collection between the months of May and December. C. intestinalis was also one of the first animals to have its genome fully sequenced. It has recently been shown that there are two sub-species of C.
intestinalis, species A and species B. Species A, now called Ciona robusta, is found on the west coast of North America and its genome has been extensively annotated.
Species B, now called C. intestinalis, is found on the east coast of North America, and is found in the waters of Rhode Island. This species has a less well annotated genome.
C. intestinalis develop very rapidly, reaching the larval stage within 24 hours of fertilization. The life cycle of C. intestinalis, like most marine invertebrates, begins with fertilization after which the zygote undergoes embryogenesis, hatches from the chorion (eggshell) as a swimming larva, the larva settles and then undergoes metamorphosis into a juvenile which grows and develops into a reproductively competent adult, thus beginning the cycle again. The complete life cycle from egg to mature, reproductively competent adult takes about 3 months. This is ideal for developmental and genetic studies as the life cycle is so fast (Corbo, Di Gregorio, and Levine 2001). C. intestinalis has a relatively small genome size, the haploid genome only contains about 1/20 th of the number of base pairs as the human genome (Simmen et al. 1998). It has been found that many of these genes, especially the ones expressed during the larval stage, have molecular homologies with the genes expressed during development in vertebrates (Takahashi et al. 1999). All together this makes Ciona intestinalis a good model organism.

C. intestinalis Ovarian Development
Ascidians are hermaphroditic animals, meaning they have both male and female gonads. The gonads sit in a loop of the intestine within a sac that also contains the heart and is constantly bathed in hemolymph. Development of the germ cells in solitary ascidians during embryologic development is isolated to several cells in the endodermal strand that can be seen during tailbud stage (Fujimura and Takamura 2000). In juvenile C. intestinalis, the gonad rudiment is connected to the dorsal strand and originates from a portion of the tadpole stage tail that is resorbed during the settling and metamorphosis process (Takamura, Fujimura, and Yamaguchi 2002;Yamamoto and Okada 1999).
During metamorphosis, the larval tail is resorbed and a mass of tissue from the resorbed tail is the origin of the gonad. The gonad rudiment can be seen using light microscopy as early as 2-3 days post metamorphosis in juvenile C. intestinalis.
Electron microscopic examination of the early gonad rudiment shows initiation of germ cell formation around blood cells and phagocytic cells that had engulfed the cellular debris of the resorbed tadpole tail. As the 2-day juvenile develops, the phagocytic cells become more infrequent, and more germ cells are found in the area.
Somatic cells can begin to be found along with the germ cells in the 3-day old juvenile. By day 4-5, germ cells become an independent structure from the phagocytic cells called the gonad rudiment. As the juvenile develops, the number of germ cells and somatic cells continues to increase when on day 7 a cavity begins to develop in the gonad rudiment. The cavity continues to enlarge from day 8 to day 10 post metamorphosis with the majority of germ and somatic cells underneath the empty cavity (Yamamoto and Okada 1999).
The gonad rudiment begins to differentiate into both the ovaries and testes about 11-12 days after metamorphosis. A rounded vesicle of germ cells containing a cavity lined by somatic cells begins to bud off of the gonad rudiment to become the testis and the remainder of the rudiment containing somatic cells associated with each germ cell becomes the ovary. By day 12, the new testicular rudiment completely separates from the ovary rudiment and migrates towards the intestine while the ovary rudiment begins to form lobules and the beginning of the oviduct begins to extend towards the atrial siphon. By day 13-14, the testicular rudiment becomes branched with small follicles at the end of each branch and the ovarian rudiment lobules branched and attached to one another. By day 15-16, the testicular rudiment becomes more club shaped and the sperm duct extends parallel to the oviduct towards the atrial siphon. The ovarian rudiment moves into the ventral loop of the intestine by day 18 (Okada and Yamamoto 1999).
In the ovary, oocytes develop surrounded by an acellular vitelline coat and between the oocyte and the vitelline coat (the perivitelline space) there are test cells.
The vitelline coat is surrounded by follicle cells that produce substances to attract sperm to the oocyte (Noriyuki Satoh 1994). As oocyte maturation continues, the follicle cells extend away from the oocyte within the chorion forming a starburst like shape to the chorion. Once the oocyte moves from the ovary, into the follicle stalk, it moves to the oviduct and continues to mature until it is released from the oviduct into the surrounding water (Okada and Yamamoto 1993).

Methods and Materials
Biological Materials Gravid C. intestinalis adults were collected from Point Judith Marina in South Kingstown, Rhode Island between May and October with a small hiatus during August when the water is at its hottest. Animals were placed in an aquarium under constant light for 36 to 48 hours after collection to allow for the animals to store gametes.

Ovarian Tissue Preparation
The ovary was dissected and rinsed in 75% ethanol for 5 seconds to remove any protists that may have been commensal in the animal. Ovaries were then cut into approximately 1 cubic millimeter pieces. Cell dissociation was performed by placing ovary fragments in a 0.01% trypsin solution in filtered sea water (FSW) for 15-20 minutes. (FSW was made from seawater collected in Narragansett Bay, RI and filtered using 0.22 m bottle filters.) The tissue was triturated every few minutes to completely dissociate the ovarian cells. Cells were transferred to a 2mL tube and spun down at 800 rpm for 3 minutes, the supernatant was discarded, new filtered sea water was added, cells were resuspended, and the washing process was repeated 3 times. After the final wash, cells were resuspended in culture medium and 75µl of the cell media mixture was seeded into each agarose castings and the cells were allowed to settle into the wells for 15 minutes before the additional 1mL of media was added to each well. Media was changed weekly and all tissue preparation and media changes were done under sterile conditions.

3D Petri Dish System
The 3D Petri Dish system (Microtissues, Inc.) was used to create microwells for the tissue culture. 800m diameter x 800m depth well (24-35 large spheroid) and 400m diameter x 800m depth well (24-96 small spheroid) molds were filled with 2% agarose in FSW. Once cooled, the agarose was removed from the molds and transferred to 24 well plates ( Figure 1). Prior to seeding the cells, the agarose microwells were allowed to equilibrate in media solution for 24 hours Dean et al. 2007;Achilli, Meyer, and Morgan 2012).

3D Petri Dish vs Culture Treated Plastic
C. intestinalis ovarian tissue was prepared as described above. Cells were counted on a hemocytometer prior to seeding so as to track the growth from time

Ovarian Tissue Culture Over Time
Ciona intestinalis ovarian tissue culture grown with culture media made of 50% FSW containing 5% hemolymph, 50 U Penicillin/50 g Streptomycin per ml (Gibco 15070-063), Antibiotic-Antimycotic: 10 U penicillin, 10 g streptomycin, .025 g Gibco Amphotericin B per ml (Gibco 15240-062), and 2mM L-Glutamine (Gibco 25030-081). Agarose casting was marked in one corner to ensure that images were taken of the same 9 wells over 7 days in culture. Images were taken on an inverted microscope so as to not contaminate the culture.

Histological Preparation
The entire agarose casting containing C. intestinalis ovarian cell culture were fixed in 10% formalin in PBS overnight. The next day, fixed cultures were rinsed in PBS for 2 hours, and dehydrated in 50% ethanol for 1 hour then 70% ethanol overnight. Fixed cultures were then dehydrated in increasing concentrations of tertbutyl alcohol (20%, 35%, 55%, 75%, 100%, 100%) then imbedded in paraffin wax.
The paraffin wax is then forced into the tissue by placing the molten wax containing cultures into a vacuum oven. The cultures were then added to a mold and covered in fresh paraffin wax and allowed to cool. Once cooled, the cultures were then sliced into 8(m) slices on a microtome (American Optical Company Spencer "820" Microtome). Slices were then mounted onto subbed slides. To stain using hematoxylin and eosin, slides were deparaffinized by submerging them in xylene twice for 3 minutes each, rehydrated in washes of decreasing ethanol concentrations (100%, 95%, 70%, tap water) for 3 minutes per wash, stained with hematoxylin for 10 minutes, washed in tap water to activate hematoxylin for 3 minutes, stained in 25% eosin in 70% ethanol for 10 seconds, dehydrated once more in increasing concentrations of ethanol (70%, 95%, 100%, 100%) for 3 minutes each, deparaffinized by submerging in xylene twice more for 3 minutes each and had a coverslip permanently mounted (Corliss and Humason 1974). To stain using Toludine Blue (Sigma C.I. 52040) according to the manufactures protocol, slides were deparaffinized and rehydrated, stained with 0.01% Toludine Blue in water for 1 minute, washed in distilled water for 1 minute, dehydrated once more and had a coverslip permanently mounted.

Live Dead Assay
Hoechst 33342 and Propidium iodide (PI) were used to stain aggregations to determine the ratio of live to dead cells found within the cultures. As a killed control for cell death, cells were incubated with 10% EtOH for 10 minutes prior to addition of stains. The media was changed in the agarose castings containing C. intestinalis ovarian cell culture 10 minutes prior to staining in order to remove any floating or non-adherent debris. Hoechst 33342 and Propidium iodide (PI) were added to the media, each at a final concentration of 1µg/ml. After 1 hour incubation with fluorescent stains the cultures were washed with fresh media to remove any stain not bound within the cells (Lema, Varela-Ramirez, and Aguilera 2011). Cell cultures were then fixed in 2% paraformaldehyde in Ca/Mg-free seawater for 30 minutes.
Cultures were washed 4 times in phosphate buffered saline (PBS) with 0.1% Tween-20 (P-Tween) and then washed twice in PBS for 15 minutes. Cultures were stored in PBS at 4°C until ready to be imaged. Hoechst 33342 will penetrate the cell membrane of both live and dead cells and fluoresce blue. Propidium iodide will only penetrate the cell membrane of dead cells and produce red fluorescence (Lema, Varela-Ramirez, and Aguilera 2011).

3D Petri Dish vs Culture Treated Plastic
I first compared whether there was more cell growth and proliferation in the agarose microwells as opposed to the traditional culture treated plastic that is commonly used in cell culture techniques ( Figure 2). Day zero (the day the cells were seeded into each system) started with an average of 125 cells/µl on the culture treated plastic and 130 cells/µl in the agarose casting. There was no statistical difference between the number of cells seeded in agarose castings versus culture treated plastic. Cells proliferated significantly faster in the agarose castings than those on culture treated plastic for all 3 subsequent time points. By day 7, cell counts in the agarose castings were about three times higher than on the culture treated plastic, which had declined slightly. By day 14, cell counts in the agarose castings, which had increased slightly, remained about three times higher than on the culture treated plastic, which had also increased slightly. By day 22, cell counts in the agarose castings, which had increased slightly, were about twice that of the culture treated plastic, which had also increased slightly.

Culture Media Trials
I next tested what types of growth additives and at what concentration yielded the most cell proliferation. I found that cells grew significantly better in media made with a base of L-15 media in 50% filtered seawater with a 5% (v/v) growth additive of C. intestinalis hemolymph ( Figure 3). Cells were seeded into the agarose casting wells at a concentration of 22 cells/µl for all media formulations that FSW media containing hemolymph having the highest counts, but there was no statistical difference between any of the media formulations.
By day 13, cell proliferation was fastest in cultures containing L-15 media made with 50% FSW containing 5% Ciona hemolymph. The cell counts in the L-15 media made with 50% FSW containing 5% Ciona hemolymph had doubled from day 5. The cell counts were significantly higher than the counts in cultures containing L-15 media made with 50% FSW media containing 5% FBS (P-value 0.0414), the L-15 media made with 100% FSW media containing 5% FBS (P-value 0.0101), the L-15 media made with 100% FSW media containing 5% hemolymph (P-value 0.0093), and the L-15 media made with 100% FSW media containing 10% hemolymph (P-value 0.0171).
By day 20, cell proliferation remained highest in in cultures containing L-15 media made with 50% FSW containing 5% Ciona hemolymph. The counts were significantly higher than the counts in cultures containing L-15 media made with 50% FSW media containing 5% FBS was 36 cells/µl (P-value 0.0258) as the cell counts in this media had declined. The counts for all media formulations increased between day 13 and day 20 with the exception of the L-15 media made with 50% FSW containing 5% FBS, which declined.
Ovarian Tissue Culture Over Time Figure 4 shows images of the same 9 wells of C. intestinalis tissue culture in an agarose casting over 7 days in culture (800µm diameter x 800µm height). Figure 5 shows images of the same well of C. intestinalis tissue culture on culture treated plastic over 7 days in culture. Day 0 shows cells that are not associated with one another as they had just been seeded into the culture system that day. As time moves on, the cells become more closely associated with one another and the size of the aggregations increases through Day 7.

Morphology of Aggregations
Histology of invertebrate tissues in culture, especially marine invertebrate tissues, is very limited. In order to examine if the aggregations forming in the agarose castings were organizing into ovary-like tissue, I compared the histology of intact C.

Live Dead Assay
I next checked the viability of the cultured cells with a live dead assay. The assay was conducted on Ciona intestinalis ovarian tissue grown in culture for 19 days with L-15 media in 50% FSW containing 5% Ciona hemolymph. To remove aggregation from the agarose castings without disrupting the structure of the aggregations, the 3D dishes were inverted onto a microscope slide that had been silanized (RainX). To ensure the aggregations could fall from the recesses in which they were growing, PBS was added to the indentations until there were no more air bubbles seen. A slight lift in one side of the 3D dish while not allowing air in would allow for the aggregations to fall from the recesses within the 3D dish onto the slide.
The aggregations were held in place on the slide by small pieces of coverslip glass that had been glued to either side of the slide to prevent the movement of the fluid containing the aggregations off the slide. The aggregations, once removed from the wells, were fragile but held together and can be seen in Figure 9,

3D Petri Dish vs Culture Treated Plastic
When cells are grown on culture treated plastic, they do not associate with each other ( Figure 5). The cells will become less 3 dimensional as they flatten out in an attempt to form attachments to the plastic. The agarose casting system allows for the cells to remain in a more natural morphology as they do not form adhesions to the substrate. The agarose casting also potentially allows for better cell-cell interactions because of its 3-dimensionality (Achilli, Meyer, and Morgan 2012;Napolitano et al. 2007;Dean et al. 2007). This increase in cell interactions could allow for the cells to organize properly according to cell type and form an aggregation, as seen in the data in this study. The formation of an aggregation of cells also allows for the removal of the intact aggregation containing multiple cells for use in biochemical assays and physiological experiments.

Culture Media Trials
Optimization of cell culture media was key to optimization of ovarian cell growth in the 3D culture system. Leibovitz's L-15 medium was used as the base media of choice for growth of ovarian tissue culture. Media was purchased in powdered form in order to more closely control the final formulation. Because C.
intestinalis is a marine organism, its internal organs are likely hypotonic with seawater, and proper osmolarity is needed for optimal cell proliferation. To test this, I used two concentrations of seawater to dissolve the L-15 media, one at 100% and one diluted to 50% with deionized water. All media solutions had 2mM L-Glutamine, an antimicrobial, and an antibiotic additive (see Materials and Methods). The antibiotics helped to control any potential contamination and the glutamine helps with cell growth and proliferation.
In traditional mammalian cell culture, fetal bovine serum (FBS) is used as a growth additive. FBS was tested in both forms of media at both a 5% and 10% concentration. C. intestinalis hemolymph was also tested as a media supplement. It was collected and used as a supplement in both media types at both a 5% and 10% concentration. Hemolymph was collected from freshly collected animals via a direct heart puncture with a syringe. Prior to addition of hemolymph to media for use in culture, hemolymph was filter sterilized (0.2m) to ensure that any protists that may be commensal in the animal will not contaminate the culture. The culture media was change weekly under all culture conditions.
Although it was not shown in this study, it seems important to note that during embryonic development the gonad rudiment is seen in close association to blood cells (Yamamoto and Okada 1999). Ciona intestinalis blood contains stem cells (3.7-5.8 µm), amoebocytes (6-15.9 µm), signet ring cells (7.6-8.4 µm), morula cells (8.4-8.9 µm), compartment cells (6.4-10.3 µm), and orange cells (8.0-9.5 µm) (Rowley 1981) but all of these should have been removed from the hemolymph when filter sterilized. Although the blood cells are filtered out of the hemolymph, there are likely proteins found within the hemolymph that assist in growth of Ciona specific cells.
Previous work done to culture Ciona intestinalis hemocytes also used plasma, or what I called hemolymph, as a growth additive to their culture medium (B Rinkevich and Rabinowitz 1993).
The results of this study have shown that the optimal media formulation among those tested for maximal cell proliferation is the l-15 media made with 50% filtered seawater with 5% hemolymph as a growth additive (Figure 4). This is likely that, once all the components of the medium are added, the media is closer to the osmolarity of the ovary in vivo. A media with osmolarity similar to that of the internal organs is better for cell growth and proliferation because, in Ciona intestinalis, the gonads are enclosed in a membrane and bathed in hemolymph constantly. Ciona can also live in the intertidal zone where the salinity can change on a daily basis. This would require that the internal environment of the animal be regulated closely even in the changing environmental conditions.
Hemolymph is also an important growth additive to the culture media because, as previously stated, the ovaries are constantly bathed in hemolymph in vivo. Ciona intestinalis have an innate immune system (Noriyuki Satoh 1994). There is a commensal protist, Cardiosporidium cionae, that lives within the heart chamber of Ciona intestinalis (Hunter, Paight, and Lane 2020). When collecting hemolymph to use as an additive to the culture media, the hemolymph was collected via direct heart puncture. It is likely that these protists were collected along with the hemolymph sample. The hemolymph was filtered prior to the use in the culture media in order to remove all commensal organisms.
It is also important to note that without the addition of hemolymph to the culture media, the tissue cultures became overrun with protists after day 20. These protists are likely commensal to the Ciona and live within the sac that encompasses the gonads and heart. During the initial dissection and dissociation of the ovarian tissue, some of these protists may get transferred to the culture with the cells even though the tissue is washed in ethanol. As there were no living blood cells to act on the protists, it is likely that there is a compound found within the hemolymph that prevents the growth of these contaminating protists.
Ascidians have an innate immune system, and the 3 sites of the immune response are the tunic, the hemocytes, and the digestive system. Phenoloxidase, an enzyme involved in the cytotoxic response, has been found in its inactive form within tunicate hemocytes and becomes activated once released outside of the cell (Franchi and Ballarin 2017). Phenoloxidase activation causes the production of reactive oxygen species which induce oxidative stress on the system. This is how phenoloxidase exerts its cytotoxic activity within ascidians (Ballarin and Cammarata 2016). Research has shown that there are immunity-related genes that are upregulated in the test cell layer and follicle cells of the developing oocytes in Ciona intestinalis ovaries, including Ciona intestinalis phenoloxidase genes (Parrinello et al. 2015;2018). This suggests that some of the ovarian tissue cells themselves could be participating in the control of the protist growth in culture. This also suggests that immune function within Ciona intestinalis is upregulated within the ovaries, making them a good tissue to grow in culture because the tissue would aid in immune regulation of potential contamination.

Ovarian Tissue Culture Over Time
The visible growth of C. intestinalis ovarian tissue in the agarose castings was tracked over a 21 day period to assess if the tissue was forming aggregations and if those aggregations were growing over time. As seen in the inverted microscope images in Figure 4, from Day 0 (the day the tissue was seeded in culture) to Day 1 the cells come together more closely. This is the first indication of aggregation of the tissue, although it is not clear as to how this aggregation is occurring. The tissue could be coming together via cell cell interactions or it could be coming together due to the "egg carton" shape of the wells where the bottom of the well has a smaller diameter than the top or some combination of these two causes. The most promising evidence shown in these images is the increase in density of the cultures by Day 7 in culture ( Figure 4). This shows that there is cell growth and proliferation in this culture system.

Morphology of Aggregations
If the morphology of the aggregations in the ovarian culture system looks similar to the morphology of the whole ovary tissue, this would suggest that the cells can self-organize into similar structures seen in the ovary. One determinant of the ability of the cells to organize around an oocyte could be the maturation state of the oocyte itself. As the cells were not sorted when the tissue was cultured, there were likely oocytes at multiple states of maturation when seeded into the culture. The less mature oocytes would be better equipped to communicate with the supporting cells of the ovary as they rely on them for growth and development. More mature oocytes that are ready for movement from the ovary into the follicle stalk would no longer rely on the surrounding cells for maturation. If ovarian tissue cells were seeded with a less mature oocyte, they would likely be able to aggregate around the oocyte while if the oocyte is mature, they likely will not be able to aggregate properly.
It is important to note that the whole ovarian tissue, as depicted in Figure 6, is not a very compact tissue. In fact, when the ovary was dissected out of the Ciona, more often than not the tissue would almost fall apart before any trypsin was added.
Ovarian tissue as a whole is not expected to be dense as the oocytes require space to develop and grow. Once the oocytes are mature enough, they move through the follicle stalk into the oviduct where they continue to mature until they are released.
As can be seen in Figure 7 and Figure 8, the aggregations of ovarian tissue in culture are also not very dense. This is consistent with the morphology seen in the whole ovary tissue.

Live Dead Assay
Results of the live dead assay showed very few dead cells in the Ciona ovarian tissue aggregations as compared to the ethanol killed aggregations (Figure 9). This suggests that even after 19 days in culture, the aggregations are still living on the culture media. This also verifies again that the agarose casting system is effective for growth of C. intesintalis ovarian tissue culture. Of importance to note was the fragility of the aggregations upon removal from the agarose casting system. Removal using a pipette to suck up the aggregations from the wells of the agarose casting was damaging to the aggregations. This caused the aggregations to dissociate very easily destroying them. In order to remove the aggregations from the wells without dissociation, I had to invert the agarose casting onto a slide to allow for the aggregations to "fall" out of the wells rather than removing them with a pipettor.

Conclusions
In this study, I have shown that tissue culture using Ciona intestinalis ovarian tissue has increased cell proliferation in the 3D intestinalis were reared for 120 days at 18°C or 22°C, after which the ovaries were removed and analyzed for protein expression. There were 62 proteins identified as differentially regulated at 22°C as compared to 18°C. Shp2 (PTPN11) was found to be upregulated 114-fold in ovaries of Ciona that were reared under high temperature stress, the highest upregulated protein identified in this study (Lopez et al. 2017) .
Shp2 was the major protein of interest that emerged from this study because it was the most upregulated protein at high temperatures.

Shp2 Function
Shp2 is required for the sustained activation of the Erk/MAP kinase (MapK) pathway in most cells, acts upstream of Ras and FGF, inhibits the Jnk pathway, and activates or inhibits the P13/Akt pathway (Tajan et al. 2015;Neel, Gu, and Pao 2003;Feng 1999). Shp2 interacts with many receptor protein tyrosine kinases (R-PTKs) and the SH2 domain of Shp2 recognizes the phosphotyrosine site on the PTK. When a ligand is bound to the SH2 domain of Shp2, phosphatase activity is increased. The SH2 domain also facilitates the association of enzymes with their substrates (Feng 1999). Thus, Shp2 acts as both an adaptor protein as well as an enzyme. Shp2 has also been shown to play a role in cytokine signaling, cell adhesion, cell migration, cytoskeletal remodeling, cell proliferation, stem-cell renewal, and cell differentiation (Feng 1999; Hale and den Hertog 2017).
The proposed mechanism of Shp2 function is described as follows: A ligand binds to a R-PTK which recruits Shp2 and Shp2 is activated. Activated Shp2 then transduces the signal to the nucleus through downstream target activation. Activated Shp2 also interacts with other transmembrane proteins to transduce the signal to neighboring cells. This allows for the signal to be received by one cell and propagated to many closely associated surrounding cells. This mechanism during a sensitive process, such as development, must be coordinated precisely (Feng 1999). Incorrect In mice, Shp2 mutations were seen to cause failure of the transition of epithelium to mesenchyme during early development. In chimeric mice with minimal Shp2 mutant cell lineage have hind limb abnormalities caused by deficient cell migration (Feng 1999). In zebrafish, there are two PTPN11 (PTPN11 a&b) genes encoding for two Shp2 (Shp2 a&b) proteins that have similar catalytic activity.
Knockdown of Shp2b led to no phenotypic abnormalities in the zebrafish while mutations in Shp2a led to craniofacial abnormalities, shortened embryos, and death after 5 days. There was decreased Erk activation only in zebrafish with deletion of both Shp2 proteins, not with deletion of a single Shp2 (Hale and den Hertog 2017). In zebrafish, Shp2 activates the MAPK pathway and inhibition of Shp2 inhibits cell growth and proliferation (Hale and den Hertog 2017). In C. elegans, PTP-2, the homologue for Shp2, is required for normal oogenesis (Feng 1999). There is only one Shp2 found in Ciona.

Reactive Oxygen Species
Under stress conditions, reactive oxygen species (ROS) increase within a cell due to an increase in cellular metabolism. In marine species, seasonal temperature changes cause changes in cellular stress seasonally as well which increase ROS production. In a species of blue crab and a species of clam it was seen that during the summer months, where the water temperature is highest and cellular stress is causing increased ROS, there was an increase in phosphorylated MapK (Feidantsis et al. 2020). In oysters, increased ROS stress due to high water temperatures leads to inhibition of the antioxidant defense system (Rahman and Rahman 2020).
ROS can reversibly bind to the catalytic site within the Shp2 PTP domain causing inhibition of the catalytic site (Meng, Fukada, and Tonks 2002). It has been shown that ROS causes oxidative inactivation of SHP2 as part of a feedback loop that leads to downstream platelet aggregation in humans (Jang et al. 2014). In planarians, ROS have been shown to be required for the activation of the Erk/MapK pathway during regeneration events (Jaenen et al. 2021).

Ciona intestinalis as Model Organism
The accumulation of greenhouse gases in the atmosphere is causing a rapid change in marine environments with a projection of a temperature rise in the ocean of as much as 4C by the end of the century ( (Takahashi et al. 1999).

Temperature Effects on Development in Ciona
Embryogenesis is a sensitive process in which cell-cell signaling and cell motility is extremely important. Previous research in the Irvine lab has shown that in

Biological Materials
Gravid C. intestinalis adults were collected from Point Judith Marina in South Kingstown, Rhode Island between May and October with a small hiatus during August when the water is at its hottest. Animals were placed in an aquarium under constant light for 36 to 48 hours after collection to allow for the animals to store gametes.

Oocyte Collection
As Ciona intestinalis are hermaphroditic, eggs and sperm were collected from two separate animals for fertilization. The outer tunic was removed in order to expose the inner tunic which was cut away to expose the muscle layer. Once the muscle layer is exposed, the oviduct and sperm duct are identified, and a small incision is made in the muscle layer to allow access to the oviduct and sperm duct.
Once exposed, a tungsten needle is used to poke a small hole in the oviduct to allow for the eggs to flow out without coming into contact with any sperm. A P1000 pipettor is used to remove the eggs from the hole in the oviduct and place them in a dish containing a mesh screen (to keep the eggs in one location) and filtered seawater. Once all the eggs were collected, they were washed with filtered sea water to ensure no debris was left. Eggs are then aliquoted into the dishes for each experimental condition.

Fertilization
A fresh, gravid C. intestinalis is dissected and the sperm is collected in the same manner that the eggs were. The sperm is diluted (1:100) and activated by mixing with filtered sea water. Each mesh screen dish containing eggs is fertilized using sperm from a different adult. Only a small amount of the diluted sperm (about 100µl) is added to each dish. The eggs and sperm are given 10 minutes to allow for fertilization. After 10 minutes, the eggs are washed with fresh FSW to wash away all remaining sperm.
For dechorionated oocytes, only a small amount of the diluted sperm (about 10 µl) is added to each dish as the chorion has been removed so the chances of polyspermy increases. The eggs and sperm are given 10 minutes to allow for fertilization. After 10 minutes, the eggs are collected at the center of the dish and transferred to a new dish. This ensures that as little sperm is left in the dish while embryogenesis begins.

Fluorescent Staining
Embryos were reared at 18°C either in control sea water or sea water containing 35µM Shp2 inhibitor for 12 hours. Embryos were then fixed in 4% paraformaldehyde in Ca/Mg-free seawater for 30 minutes. Embryos were then washed 4 times with phosphate buffered saline with 1% Tween-20 (P-Tween). to generate a 2X staining solution (2g/ml DAPI and 0.5M Phalloidin). The staining solution is added 1:1 into embryos and embryos were covered and rocked at room temperature for 2 hours. After the staining was complete, embryos were washed once for 5 minutes in phosphate buffered saline containing 0.01% Triton X-100 (PBST1) and then washed twice for 15 minutes in phosphate buffered saline (PBS).

Fluorescent Imaging
Embryos were placed onto a silanized microscope slide. The embryos were held in place on the slide by small pieces of coverslip glass that had been glued to either side of the slide to prevent crushing the embryos. Embryos were imaged in PBS.

Hydrogen Peroxide (H2O2) as Oxidative Stress
H2O2 was used to mimic the increase in oxidative stress in increasing water temperature. Ciona embryos were reared at a control temperature of 18°C at a control of 0 µM H2O2, at 50 µM H2O2 and at 100µM H2O2. Embryos were allowed to develop for 18 hours at which, in normal development, the embryos could hatch out of the chorion as swimming larva. If embryos were not hatched, they were chemically dechorionated to allow for observation of morphological differences.

Dechorionation
Petri dishes were coated with 0.1% gelatin with 0.1% formaldehyde and Eggs are then dispersed into the dishes for each experimental condition to preincubate for 2 hours at 15, 18, or 22°C.

Temperature and Shp2 Inhibitor Chase Experiment
After dechorionation, pre-incubation, and fertilization, embryos were challenged with either temperature stress alone or temperature stress in conjunction with Shp2 inhibition for either the first 3 hours, hours 3-6, or hours 6-12 of embryological development. Embryos were pre-incubated prior to fertilization under the experimental conditions the embryos would spend the first 3 hours of development. Control embryos were kept at 15°C for the entirety of development as this was the control temperature for the environment at the time of year this experiment was conducted. Half of the embryos kept at 15°C for development were exposed to 25 µM Shp2 inhibitor for either 0-3 hours, 3-6 hours, or 6-12 hours of development, after which the inhibitor was washed out. 18°C was used as a mild temperature stress and 22°C was used as an extreme temperature stress for embryological development. A proportion of the embryos were subjected to either 18°C or 22°C for either 0-3 hours, 3-6 hours, or 6-12 hours of embryological development. Half of the embryos subjected to temperature stress were also subjected to 25 µM of Shp2 inhibitor during the same time range of development after which the inhibitor was washed out and the embryos were brought back to the control temperature of 15°C.

RNA Extraction
Total RNA was extracted from embryo samples using the RNA XS kit (Marcherey-Nagel 740902.50) and the total RNA was frozen at -80°C as quickly as possible. RNA was quantified by spectrophotometry.

cDNA Preparation
The preparation of cDNA began with the standardization of the RNA concentration used for each cDNA preparation. 200ng of RNA was used with the New England BioLabs ProtoScript First Strand cDNA Synthesis Kit. According to the manufacturer's instructions the RNA was mixed with SR-Poly-T primer (non-specific primer that will bind to the Poly-A tail on mRNA) and dNTPS and incubated at 65°C for 5 minutes then put on ice. The reaction mixture of reverse transcriptase buffer, MgCl2, DTT, and RNAse inhibitor were then added, and each sample was incubated at 45°C for 2 minutes. 1 µl of reverse transcriptase (RT) enzyme (Protoscript II NEB M0368) was added to each sample, with the exception of the no RT control, and incubated at 45°C for 50 minutes. The reaction was inactivated by incubating at 70°C for 15 minutes and then the samples were put on ice. 1 µl of RNAse H was added to each sample and was incubated at 37°C for 20 minutes. Samples were stored at -20°C until use in RT-PCR reactions.
The thermocycler ran an initial denaturation for 2 minutes at 94°C and 35-40 amplification cycles consisting of a 30 second denature at 94°C, a 30 second anneal at the temperature specified for each primer, and an extension for 1 minute/kb of product at 72°C. RT-PCR product were then run via gel electrophoresis on a 1% agarose gel and imaged.

Gel Analysis of RT-PCR
ImageJ was used to analyze the results of the RT-PCR products run on a 1% agarose gel. The bands were analyzed using the gel analysis tool for density.

Shp2 Inhibitor Concentration
In order to determine the concentration of Shp2 inhibitor that has an effect on embryogenesis in C. intestinalis, a titration experiment was performed, rearing embryos at multiple concentrations of Shp2 inhibitor. Ciona were reared at a control temperature of 18°C with Shp2 inhibitor concentration ranging from 5µM to 45µM ( Figure 10). When embryos were reared with 35 µM of Shp2 inhibitor (Figure 11 B-H) the phenotypes of the embryos mimic the phenotypes seen in embryos reared at high stress temperatures (Irvine et al. 2019) as compared to control embryos not reared with Shp2 inhibitor (Figure 11 A). All four types of embryos (as described in Table 1) were seen in the embryos reared in 35 µM Shp2 inhibitor. There were deformations seen in the tail, typically the milder deformations that happen during embryogenesis, characterized by kinks in the notochord (solid white triangles in D, E, & F) as well as radial pinching of the notochord (white outline triangle in C). There were also more extreme deformations where the notochord cells fail to extend fully, and the morphology of the larval trunk is abnormal (Figure 10 G & H). In all embryos, the although not always in the correct location, meaning late-stage cell type specification is occurring correctly. Figure 12 shows examination of the tail morphology in embryos reared with 35µM Shp2 inhibitor (Figure 12 B-H) as compared to the tail morphology in a control embryo (Figure 12 A). Normal tail morphology, as seen in control embryos (Figure 12 A), has complete extension of the notochord within the tail epidermis so that the tail is straight by larval stage of embryological development. Figure

Hydrogen Peroxide (H2O2) as Oxidative Stress
Ciona intestinalis embryos were reared with differing concentrations of H2O2 to assess if increased concentrations of reactive oxygen species would result in the same phenotypic patterns as seen at increased temperature and Shp2 inhibitor. Figure 14 shows that at 50 µM H2O2 there was about 30% normal development and 70% abnormal development. At 100µM H2O2 there was less than 10% normal development, about 20% abnormal development with the ability to hatch, and over 60% abnormal development with the inability to hatch. At 0 µM H2O2, the control concentration, all four phenotypes described in Table 1 could be seen with a bias towards normal, type I embryos. At 50 µM concentration of H2O2 the percent of normal, type I embryos began to drop while the percent of type II and type III/IV embryos began to increase. At 100 µM H2O2 all four of the phenotypes described in Table 1 could be seen with a bias towards type III and IV phenotypes.

Shp2 Gene Expression in Embryos
Semi-quantitative RT-PCR was used to determine if transcript expression patterns of Shp2 changed with temperature stress and at different stages of embryogenesis. Figure 17, the gel electrophoresis analysis of RT-PCR products, shows there is increased Shp2 transcript expression at gastrula stage embryos as compared to tailbud stage embryos. As the rearing temperature increases, the expression of Shp2 transcripts also increases. Actin serves as a loading control for the RT-PCR while the negative control contained no cDNA. Figure 18 shows densitometry analysis of the Shp2 RT-PCR bands. The densitometry showed that Shp2 transcript expression in gastrula stage embryos is significantly higher than that in tailbud stage embryos (P-value 0.0122). When normalized to Shp2 transcript expression in the ovary, the most upregulation was seen in the 23C gastrula stage embryos. The least upregulation of Shp2 transcript expression was seen in 23C tailbud stage embryos.   Below 25 µM Shp2 inhibitor, there are more than 50% average Type I embryos ( Figure 10) which is consistent with the phenotypic variation seen over the range of reproductively successful temperatures in our lab experiments. At 25 µM Shp2 inhibitor and above, the mean percent of Type I embryos quickly declines, and the abnormal embryo phenotypes (Type II and Type III/IV) begin to become more prevalent. Type II embryos, characterized by kink and bends in the tail, peak at 35 µM Shp2 inhibitor. Type III/IV embryos, characterized by lack of tail extension and/or globular shape, peak at 45 µM Shp2 inhibitor ( Figure 10). Above 15 µM Shp2 inhibitor, the embryos, which normally have the ability to hatch out of their chorion because of the movement of their tail, were no longer able to hatch out of their chorion. If they were able to hatch out of their chorion, the kinks and abnormal tail morphology did not allow for the embryos to swim properly. This data suggests a dose dependent effect of the Shp2 inhibitor on the normal embryogenesis of C. intestinalis.

Shp2 Inhibitor at 35 M
As embryogenesis proceeds in Ciona intestinalis, extension of the tail begins at the tailbud stage and continues until hatching from the chorion at the larval stage.
As extension of the tail occurs the notochord is extending within the surrounding tail epidermis. In normal tail development, the extension of both the notochord and tail epidermis occur at the same rate allowing for, by the time the larva hatches, the tail to be straight so the larva can swim properly. When embryos develop in the presence of Shp2 inhibitor, the tail does not develop properly, there are curves and kinks that would inhibit the ability of the larva to swim properly ( Figure 11).
When examining more closely the tail morphology of the embryos reared in the presence of Shp2 inhibitor (Figure 12), it may be that the extension of the notochord cells is occurring at a faster rate than the extension of the tail epidermis causing the notochord to bend and kink within the tail. It may be that Shp2, as it is responsible for not only signal transduction from external signals but also signal transduction between cells (Tajan et al. 2015;Feng 1999;Neel, Gu, and Pao 2003), is responsible for signaling cells in the tail during extension of the notochord and tail epidermis. Inhibition of Shp2 signaling with the Shp2 inhibitor may not allow for the signaling between cells to occur as it should. This may cause the cells to lose their coordination when it comes to extension which would cause kinks in the tail.
In order for proper attachment to a substrate, larval Ciona intestinalis normally have three palps that extend from the anterior surface of the trunk. The palps serve to inspect substrates, adhere to a suitable substrate, and initiate metamorphosis into the adult form (Zeng et al. 2019). When examining more closely the trunk morphology of embryos reared in the presence of Shp2 inhibitor ( Figure   13), in some cases, the palps were developing improperly. There was failure of complete extension of the trunk epithelium as well as asymmetry in the extensions.
Incomplete extension would lead to palp deformations.

Hydrogen Peroxide (H2O2) as Oxidative Stress
Increased H2O2 concentrations was used to simulate the increase in oxidative stress that the C. intestinalis embryos were under during embryogenesis at high temperatures. Analysis of the different abnormalities seen in embryos reared in 100 µM H2O2 ( Figure 15) suggests that similar, although seemingly not as severe, defects were seen as compared to both temperature stress and Shp2 inhibition. These defects seem less severe because it looks as though the H2O2 affects the tail of the embryos more so than the trunk. Although the trunk seems to be less affected, this could be simply because the concentration of H2O2 was too low to show the same morphological abnormalities as are seen with temperature stress and Shp2 inhibition. Taken together these experiments suggest that deformation in embryogenesis seen at high temperature stress are due to increased oxidative stress.  Figure 16 D3. This addition of the Shp2 inhibitor only affected embryos exposed to mild temperature stress for the first 3 hours of development ( Figure 16 D1-3). When embryos were exposed to Shp2 inhibitor and mild temperature stress for hours 3-6 of development When embryos were exposed, during hours 3-6, to high temperature stress alone ( Figure 16 E4-5) and in conjunction with Shp2 inhibitor (Figure 16 F4-5), it appears that Shp2 inhibition exacerbates the morphological abnormalities seen at high temperature stress alone. Tailbud embryos exposed to the high temperature stress for hours 3-6 show mild morphological abnormalities (Figure 16 E5) while high temperature stress in conjunction with Shp2 inhibitor cause more severe abnormalities (Figure 16 F5). Embryo E5 has a bend in the tail that is more extreme than in a normal embryo and embryo F5 shows multiple kinks in the tail, incomplete extension of the tail, and abnormal trunk morphology. This shows that addition of the Shp2 inhibitor on top of the high temperature stress causes more morphological abnormalities than just temperature stress alone when exposed to stress from hours 3-6 of development. This suggests that Shp2 is required for normal embryogenesis during the 3-6 hour time period in development at high stress temperatures. This again suggests that Shp2 is required to ameliorate the stress caused by high temperatures during development.

Shp2 Inhibition and Temperature Stress
When embryos were exposed, during hours 6-12 of development, to high The beginning of embryogenesis is the most sensitive time in development wherein the embryo needs the most protection from the impacts of stress, both internal and external. During reproduction, the generation of gametes is the most sensitive part of the process. Gametes need the most protection from the impacts of stress to ensure reproductive success. As Shp2 is implicated in transduction of signals, both from external sources of stress as well as internal cellular stress (Tajan to ameliorate this stress.
In a stress situation, such as increased temperature stress, Shp2 may be upregulated in early embryogenesis to protect the developing embryo while cell specification and cell fate are determined and while crucial structures are laid out.
These results are consistent with the results of the Shp2 chase experiment where it appears embryogenesis is most sensitive to inhibition in the first 6 hours of development ( Figure 16). This increased sensitivity to inhibition may be due to the increased concentration of Shp2 proteins at earlier stages of development, although the RT-PCR can only suggest transcript level expression. Tail deformations caused by temperature stress causes the embryos to lose the ability to swim properly, if at all, inhibiting their ability to find a proper substrate to which they would settle and metamorphose into the adult morphology. Trunk abnormalities could negatively affect the ability of the embryo to settle onto a substrate which could in turn negatively affect the ability of the embryo to metamorphose into the adult form. I am proposing that Shp2 is functioning to ameliorate this stress in order to minimize these deformations. Figure 19 shows a model for how the relationship between Shp2 and oxidative stress caused by increased temperature could be twofold: 1) oxidative stress causes ligands to bind to RTKs which activate Shp2 to cause signal transduction through the Erk/MapK pathway as well as signal transduction to neighboring cells (Feng 1999) and 2) ROS can reversibly bind to the catalytic site of Shp2 (Meng, Fukada, and Tonks 2002), rendering it inactive and unable to dephosphorylate proteins. It may be that increased oxidative stress would cause increased Shp2 expression 1) because Erk/MapK signal transduction is upregulated in response to increased stress signaling and 2) because when ROS bind to the Shp2 catalytic site, Shp2 is no longer functional in the Erk/MapK pathway and therefore must be upregulated to compensate for this inactivation. All together this suggests that there is interference between the two interactions between oxidative stress and Shp2 function. Therefore, the inhibition of Shp2 mimics the inactivation of Shp2 by ROS causing the deformation seen in embryos reared at high temperature stress.