Towards an Understanding of the Epigenetic Regulation of Hydra Regeneration

Towards an understanding of the epigenetic regulation of Hydra regeneration

Abstract

 Hydra is a Cnidarian polyp that has been used as a model organism in a wide range of settings for over 200 years. This small freshwater organism has a remarkable ability to regenerate, which can be attributed to a large pool of stem cells that reside in its body column. Classically, Hydra has been shown to regenerate via morphallaxis following decapitation. A large number of genes that are expressed during head regeneration have been isolated. Recent evidence points to a role for epimorphic regeneration following mid-gastric bisection. In order to elucidate the epigenetic mechanisms behind Hydra regeneration, a screen of 80 epigenetic modulators will be used following decapitation and mid-gastric bisection of Hydra. Specifically, any difference between decapitation and mid-gastric bisection will be investigated in the pursuit of understanding the roles of epigenetic modulation in morphallactic and epimorphic regeneration of Hydra.

1. Review of Literature

1.1 Hydra: A Model Organism for Over 200 Years

 The freshwater Cnidarian polyp known as Hydra is a relatively simple organism (Carter et al., 2016). Nevertheless, its heuristic value as a model organism means that it has been studied as far back as the mid 17^th century. Particularly interesting to pioneering efforts in science at this time, was the classification of plants and animals into distinct phyla (Galliot, 2012). Despite this, Hydra remained as an elusive species to philosophers and naturalists of the time due to its apparent capacity to display both animal and vegetal attributes, such as asexual reproduction (Galliot, 2012). In 1744, Abraham Trembley was determined to put this ambiguity to rest and set forth trying to solve the issue by testing its regenerative ability (Galliot, 2012). Upon careful examination of the organism over several days, Trembley observed complete regeneration (Trembley, 1744; translated by Lenhoff and Lenhoff, 1986) suggesting that this organism should fall into the vegetal kingdom (Galliot, 2012). However, Trembley also observed several behaviours that did not fit this suggestion including locomotion and an active feeding response. In all, Trembley decided upon placing the organism into the kingdom Animalia. Through the active investigation and manipulationof his study, Trembley has been credited with advancing marine biology from an entirely observational science to an experimental one (Galliot, 2012). Nearly 270 years ago, Hydra served as a fruitful model organism for scientific inquiry and curious minds alike.

 Ever since Abraham Trembley set about studying Hydra, this small freshwater organism has been used as a model system in a wide range of settings. Apart from regeneration, Hydra has been used to study senescence, neural circuits, developmental genetics, evolution, and stem cell biology, among others (Galliot, 2012).

1.2 Anatomy

 Hydra has a simple body plan containing a head, body column and foot along a single axis with radial symmetry (Bode, 2011; Bode 2009). At the oral (head) region, tentacles emerge below a dome-shaped hypostome containing the mouth (Carter et al., 2016). The mouth is not a permanent opening – when closed it is a continuous epithelial sheet containing septate junctions (Carter et al., 2016). Consequently, Hydra must tear a hole through this epithelial sheet on each occasion of opening its mouth (Carter et al., 2016). This process occurs via cell shape rearrangement and radial contractile activity of myonemes (Carter et al., 2016). Interestingly, the mouth is capable of opening wider than the body column, enabling Hydra to consume organisms larger than itself (Carter et al., 2009). The tentacles of Hydra contain nematocysts (stinging cells) used for capture of prey and defense (Balasubramanian et al., 2012). A mechanosensory apparatus triggers the discharge of these projectile organelles, which contain proteins with venomous, lytic, adhesive and fibrous properties (Balasubramanian et al., 2012). The peduncle and basal disk reside at the opposite end of Hydra (aboral). The basal disk allows the organism to adhere to a surface (Bode, 2009). The body column lies between the oral and aboral regions and contains the gastric region and budding region. The budding region serves as the site of asexual reproduction (Bode, 2009).

1.2.1 Cell Types

 Hydra contains only two cell layers, the endoderm and ectoderm, which are separated by an extracellular matrix (mesoglea). The ectodermal layer is thinner and the cells are more columnar (Carter et al., 2009). There are three adult stem cell (ASC) populations. The epidermal and gastrodermal epithelial stem cells (ESC) are unipotent and reside in the body column (Buzgariu et al., 2018). The interstitial cells (ISC) are multipotent and are scattered throughout the body in spaces among epithelial cells. The ISC lineage differentiates into neurons, secretory cells, gametes and nematocytes (Buzgariu et al., 2018; Bode, 2003). ESCs divide every three to four days, while ISCs divide every 24-30 hours. The endodermal and ectodermal cells of the body column remain in the mitotic cycle and continuously displace tissue towards the extremities, where it is shed into the environment (David and Campbell, 1972). In contrast, the cells of the tentacles and foot are arrested in the G₂ phase of mitosis. Hence, the body column of Hydra is rich in dividing stem cells, while the extremities primarily consist of terminally differentiated cells (Buzgariu et al., 2018).

1.2.2 Control of Body Shape

 Without a rigid skeletal system, Hydra must rely on its hydrostatic skeleton for support and movement. In this way, force is transmitted by internal pressure and this allows for the diameter and length of the body column to be controlled (Kier, 2012). In order to control this process, ectodermal myonemes are oriented longitudinally and endodermal myonemes are oriented circumferentially around the body column (Carter et al., 2009). Contraction of the ectodermal myonemes is established by electrical impulses from the nervous system and results in contraction of the Hydra body column (Josephson, 1967).

1.2.3 Nervous System

 Cnidarians including Hydra were among the earliest animals in evolution to have nervous systems. Depending on the size of Hydra, their nervous system consists of a few hundred to a few thousand neurons (Dupre and Yuste, 2017). There are two types of neurons in Hydra: sensory cells that are exposed to the external environment and ganglion cells that form a lattice called the nerve net. These neurons form networks that are non-overlapping both in structure and function. The networks are associated with specific behaviours, such as longitudinal contraction, elongation and radial contraction (Dupre and Yuste, 2017). Interestingly, neurons are continuously made in the body column and migrate towards the extremities where they are sloughed. This requires that the phenotype of any neuron be plastic as it moves through the body (Koizumi and Bode, 1991).

1.3 Reproduction

 Hydra are capable of both sexual and asexual reproduction. Generally, Hydra reproduce asexually, via budding, at warm temperatures (18-22°C). Bud formation beings with evagination of the body column wall that continues to elongate into a cylindrical protrusion. A head forms at the apical end and a foot at the basal end of the protrusion. Subsequently, the bud detaches from the adult and continues its growth until reaching the adult stage (Lengfeld et al., 2009). Sexual reproduction involves gamete differentiation and often occurs at lower temperatures (10-12°C). The difference in reproductive state can be understood by examination of cells of the sperm lineage (Lengfeld et al., 2009). It was discovered that under conditions where sperm do not usually differentiate (18-22°C), cells continue to enter the sperm pathway but their progression is arrested prematurely. This occurs due to a temperature sensitivity of the cell that becomes increasingly prominent as the cell moves down the sperm pathway. This lethal effect is eliminated at lower temperatures, corresponding to the increase in sexual differentiation observed at temperatures ranging from 10°C to 12°C (Lengfeld et al., 2009).

1.4 An Overview of Regeneration

 Ever since Abraham Trembley took a pair of scissors to Hydra, this freshwater polyp has remained a fruitful model organism for the study of regeneration (Galliot, 2012). A curious observation scientists made was that Hydra is only capable of regeneration from the gastric column and not from tissue of the extremities (Buzgariu et al., 2018). Since the extremities of Hydra are composed of differentiated cells and the body column is rich in dividing stem cells, this suggests a role of stem cells in regeneration (Buzgariu et al., 2018). The classical view is that regeneration of Hydra primarily occurs via morphallaxis, or regeneration in the absence of cellular proliferation (Bosch, 2007; Buzgariu et al., 2018). This route is favoured following decapitation because cells near the oral region are already committed to differentiate, therefore hindering cellular proliferation. Morphallaxis in Hydra likely represents an adaptation process (Buzgariu et al., 2018). Epimorphosis requires active cellular proliferation and occurs following mid-gastric bisection of Hydra (Buzgariu et al., 2018).

 Through the selective elimination of various cell types in Hydra, it was discovered that regeneration can be accomplished by epithelial cells only (Marcum and Campbell, 1978). When chimeric Hydra were produced using a normal strain and mutant reg16 strain, defects in head and tentacle regeneration were identified. These defects were located in the ectodermal and endodermal epithelial cell lineages (Wanek et al., 1986). These results suggest that the epithelial cells contain all factors and genes required for regeneration (Bosch, 2007). In addition to this, regeneration of Hydra requires an intact extracellular matrix (ECM). As noted in section 2.2.1, the endoderm and ectoderm of Hydra are separated by and ECM of mesoglea (Carter et al., 2009). The mesoglea contains collagens, laminins, proteoglycans and fibronectin-like molecules (Bosch, 2007). Regeneration begins with retraction and re-fusion of the mesoglea (Shimizu et al., 2002) and is blocked by drugs affecting collagen structure (Sarras et al., 1993). The role of mesoglea in Hydra regeneration is further demonstrated by apoptosis of Hydra epithelial cells that have been detached from the ECM (Kuznetsov et al., 2002). A third requirement for Hydra regeneration is a minimum tissue size. Hydra is only capable of regenerating from a pool of 300 cells or more – less than that consistently results in disintegration (Shimizu et al., 1993).

1.4.1 Signal Transduction Pathways

 In order to achieve successful and complete regeneration, cell-cell communication must be established. In Hydra, this occurs through various signal transduction pathways including, but not limited to, those of anklet, RTKs, MAPK, and PI(3)K.

 The anklet geneis the first one expressed in the regenerating basal end of the differentiated basal disk cells. Anklet protein is restricted to the lowest part of the peduncle. Amimoto et al. (2006) used RNA-mediated interference (RNAi) to supress the level of anklet during foot regeneration, which led to a smaller foot and a decrease in the size of the basal disk, along with a delay in basal disk regeneration. This result demonstrates that anklet is involved in basal disk formation during regeneration of Hydra. In addition to this, anklet contains a perforin domain that may have cytotoxic functions and may be responsible for a phenotypic difference between cells of the peduncle and basal disk regions (Amimoto et al., 2006).

 Receptor tyrosine kinases (RTKs) reside in the plasma membrane and are essential to signal transduction pathways. Typically, they are activated by dimerization following ligand binding (Hubbard and Miller, 2007). In Hydra, the src-type receptor tyrosine kinase (STK) plays an important role in signal transduction and head regeneration where it is upregulated following decapitation. Inhibition of STK activity using RNAi does not affect foot regeneration, but creates Hydra that cannot regenerate their head. Therefore, it is speculated that the initial commitment of cells to form head structures originates from STK activity (Cardenas and Salgado, 2003). Manuel et al. (2006) found that STK regulates cell differentiation and pattern, but not cell proliferation.

 The mitogen activated protein kinase (MAPK) pathway was identified in Hydra more than a decade ago and is differentially expressed during head regeneration (Bosch, 2007). Following mid-gastric bisection of Hydra, the interstitial cells undergo apoptosis and induce the surrounding progenitor cells to proliferate. This process is required for head regeneration and is initiated by the MAPK pathway. Knockdown of this pathway prevents apoptosis and head regeneration (Chera et al., 2011).

 The PI(3)K pathway acts early in head regeneration to affect the balance between proliferation and apoptosis. Inhibition of the PI(3)K pathway blocks the regeneration of the head in Hydra (Manuel et al., 2006).

 In addition to the pathways presented above, a large number of genes that are expressed during head regeneration and bud formation have been isolated. These include HyBra1, HyWnt, HyTcf, and Budhead that are expressed in the hypostome. Ks1, HyAlx and Cnox-3 are expressed in the tentacles and basal region of the head and are likely involved in tentacle formation. The majority of these genes are transcription factors (Manuel et al., 2006).

1.4.2 Axial Patterning of Hydra

  In order to establish axial pattern formation, morphogenic gradients acting early in development are critical. However, unlike most organisms that display these gradients only in the early stages of embryogenesis, Hydra maintain these patterning events into adulthood (Bode, 2009). There are three primary patterning processes that occur in the adult organism: the head activation gradient, head inhibition gradient and head organizer (Bode, 2009).

 The head activator (HA) was the first signalling molecule identified in Hydra and was observed to accelerate regeneration of the head (Schaller, 1973). HA is produced in the hypostome and displays a morphogenic gradient, being highest in the head and declining towards the foot of the organism. Schaller (1976) demonstrated that Hydra treated with HA have increased epithelial and interstitial cell division. When the level of HA is sufficient, this allows for the formation of the head organizer (Bode, 2009). HA has also been shown to stimulate budding and the formation of new nerves during head regeneration (Schaller, 1976). The head inhibiter (HI) is produced by the head organizer in the hypostome and prevents head formation along the body column. The HI gradient decreases in concentration towards the head (Bode, 2011).

 The head organizer is present in the hypostome of a Hydra polyp and serves as an embryonic organizer (Bode, 2011). Transplantation of the organizer onto the body column of another Hydra results in a second axis. The canonical Wnt pathway plays a major role in establishing the head organizer (Lengfeld et al., 2009). This pathway stabilizes beta-catenin and leads to its interaction with T-cell factor (TCF) in the nucleus. In turn, this leads to the synthesis of several genes involved in formation of the organizer (Lengfeld et al., 2009).

1.5 An Overview of Epigenetics

 The genome of Hydra contains ~20,000 protein-coding genes and each one can be regulated by various mechanisms (Chapman et al., 2010). The concept of epigenetics pertains to the manner in which genes expression is regulated via modifications that do not include those made to the DNA sequence itself (Deans and Maggert, 2015). This includes modifications made to the histone proteins around which DNA is wrapped. This structure, chromatin, is regulated by post-translational modifications (PTMs) made to the N-terminal tails of histones (Gillette and Hill, 2015). There are three factors that recognize PTMs: histone readers, writers and erasers. Readers contain recognition sites for specific PTMs and are able to positively or negatively affect transcription. Writers, such as acetyltransferases or methyltransferases, are enzymes that add PTMs to histone tails, while erasers remove PTMs (Gillette and Hill, 2015).

2. Purpose and Objectives

 Little is known about the epigenetic modulation of morphallaxis and epimorphosis in Hydra. The purpose of this study is to identify the difference in decapitation and mid-gastric bisection of Hydra. Secondly, the epigenetic mechanisms that control morphallaxis and epimorphosis will be examined.

2.1 Goals

 This study will utilize 80 epigenetic modulators from the Toriscreen Epigenetics Toolbox. Since each compound is provided as a 10mM DMSO solution, it will be necessary to identify a DMSO toxicity range, as it is vital that DMSO does not affect Hydra viability. Furthermore, it is necessary to identify a suitable concentration range to use for the Epigenetic Toolbox. Once these goals have been met, the effects of the epigenetic modulators on Hydra regeneration will be examined on those that have undergone decapitation and mid-gastric bisection.

2.2 Hypotheses

• There will be a difference in regeneration dynamics of Hydra that have undergone decapitation and mid-gastric bisection.
• In a screen of epigenetic modulators, a difference between morphallaxis and epimorphosis will be observed.

3. Methodology

3.1 Hydra Maintenance

 Hydra cultures are kept in Pyrex plates filled with Hydra medium and stored in an incubator at ~22°C. Each Hydra is fed approximately 4-6 brine shrimpevery second day. Hydra are cleaned every second day by dislodging any uneaten shrimp from the Pyrex plate and replacing old Hydra medium with new medium. On a weekly basis, Hydra cultures are transferred to new Pyrex plates whilst old plates are cleaned with 75% ethanol.

3.1.1 Preparation of Hydra Medium

 Hydra medium consists of an equal volume (100ml) of three stock solutions, diluted in 50L of Ultrapure water. Stock solution A consists of 0.5M CaCl₂, 0.5M NaCl and 0.05M KCl, stock solution B consists of 0.5M Trizma base and stock solution C is comprised of 0.05M MgSO₄. All stock solutions are autoclaved on liquid cycle for 20 minutes following preparation. Hydra medium is stored in a 50L Nalgene carboy.

3.1.2 Preparation of Brine Shrimp Hatchery

 To prepare the solution in which brine shrimp cysts will hatch, 2.5 tbsp. of Instant Ocean salt and 1/8 tsp. of sodium bicarbonate is dissolved in 400ml of distilled water. One-quarter tsp. of Artemia cysts are soaked in distilled water for 30 minutes before being placed in the dissolved salt solution in the hatchery. An air bubbler in the hatchery is necessary to ensure proper hatching of the cysts. Brine shrimp to be used for feeding of Hydra are collected 48 hours following preparation of the hatchery.

3.2 Viability Assay

 Since each compound in the Toriscreen Epigenetic Toolbox is provided as a 10mM DMSO solution, it is important that Hydra viability is not affected by exposure to DMSO. Hence, a viability assay on 24-hours starved Hydra was carried out to assess toxic concentrations of DMSO. Three concentrations of DMSO were tested: 0.05% (n=20), 0.1% (n=20) and 0.2% (n=20). These concentrations were made by serial dilution. The positive control (n=60) was 1% DMSO, while the negative control (n=60) was Hydra medium. In each of the treatments, Hydra were equally divided into a cut and uncut group. Subsequently, they were transferred to three, 12-well plates containing 5 Hydra per well (Figure 1). The Hydra were washed twice with their respective concentrations of DMSO, or with Hydra medium for the negative control. Their morphology was scored at time 0, 12, 24, 36, 48, 60, 72, 84 and 96 hours according to the regeneration scale shown in Appendix 1 (Figure 2). Hydra are imaged on darkfield using a dissecting microscope.

3.3 Regeneration Assay

 The regeneration assay is used to assess variation in regeneration time between mid-gastric bisection and decapitation of Hydra. Five Hydra are placed in each well of a 12-well plate, following bisection or decapitation. Control Hydra are not cut and are used for comparison during statistical analyses, including post hoc tests. Hydra regeneration is scored according to Figure 2 at 12-hour intervals. Hydra are imaged on darkfield using a dissecting microscope.

3.4 Cell Maceration

 Hydra are 24-hours starved and the gastric region is isolated. Hydra are placed in microfuge tubes and any remaining liquid is aspirated. Maceration solution is added to the microfuge tube, corresponding to time 0 hours. Following 10 minutes of immersion, 5ul of 10,000nM Mitotracker is added to the maceration mixture. The tube is inverted several times to ensure proper mixing and stored in the dark. Dissociated cells are fixed using 100ul of 8% paraformaldehyde. After 90 minutes, 25ul of 10% Tween and 5ul of 109uM DAPI are added to the microfuge tube, which is gently inverted after each step. Following 15 minutes of incubation, an equal amount of maceration mixture is pipetted onto a polylysine slide. Slides are imaged with fluorescent microscopy.

3.5 Toluidine Blue Staining

 Hydra are placed in a microfuge tube containing 2% urethane and allowed to elongate. Hydra are fixed using 100% ethanol and subsequently washed with distilled water three times. The solution is aspirated and Hydra are washed with Tris-Cl twice. Following aspiration of Tris, 200ul of 0.05% toluidine blue stain is added to the microfuge tube. Hydra are washed three times with distilled water and then dehydrated with five grades of ethanol in the order of: 50%, 75%, 95% and twice at 100% ethanol. Lastly, Hydra are transferred to polylysine slides and mounted in clear nail polish.

3.6 Microscopy and Image Analysis

 For viability and regeneration assays, Hydra are imaged at 1X magnification using a dissecting microscope on darkfield. Hydra stained with toluidine blue are imaged using phase contrast microscopy. Stenotele density is determined via the 5-box method of counting using ImageJ software. Cell macerates are imaged using fluorescent microscopy at 20X magnification.

Literature Cited

• Amimoto, Y., Kodama, R., Kobayakawa, Y. (2006) Foot formation in hydra: a novel gene,               anklet, is involved in basal disk formation. Mech. Dev. 123: 352-361.
• Balasuramanian, P., Beckmann, A., Warnken, U., Schnolzer, M., Schuler, A., Bornberg-Bauer,               E., Holstein, T., Ozbek, S. (2012). Proteome of hydra nematocyst. J. Biol. Chem. 287(13): 9672-9681.
• Bode, H. (2003). Head regeneration in hydra. Dev. Dynamics. 226(2).
• Bode, H. (2009). Axial patterning in hydra. Cold Spring Harb. Perspect. Biol.
• Bode, H. (2011). Axis formation in hydra. Annu. Rev. Genet. 45: 105-17.
• Bosch, T. (2007). Why polyps regenerate and we don’t: towards a cellular and molecular framework for hydra regeneration. Dev. Biol. 303: 421-433.
• Buzgariu, W., Wenger, Y., Tcaciuc, N., Catunda-Lemos, AP., Galliot, B. (2018). Impact of cycling cells and cell cycle regulation on hydra regeneration. Dev. Biol. 433: 240-253.
• Cardenas, M., Salgado, L. (2003). STK, the src homologue, is responsible for the initial commitment to develop head structures in hydra. Dev. Biol. 264: 495-505.
• Carter, J., Hyland, C., Steele, R., Collins, E. (2016). Dynamics of mouth opening in hydra.               Biophys. J. 110(5): 1191-1201.
• Chapman, J., Kirkness, E., Simakov, O., Hampson, S., Mitros, T., Weinmaier, T., Rattei, T., Balasubramanian, P., Borman, J., Busam, D., et al. (2010). The dynamic genome of hydra. Nat. 464: 592-596.
• Chera, S., Ghila, L., Wenger, Y., Galliot, B. (2011). Injury-induced activation of the MAPK/CREB pathway triggers apoptosis-induced compensatory proliferation in hydra head regeneration. Dev. Growth and Differ. 53(2).
• David, C., Campbell, R. (1972). Cell cycle kinetics and development of hydra attenuata. J. Cell               Sci. 11: 557-568.
• Deans, C., Maggert, K. (2015). What do you mean, “epigenetic”. Gen. 199(4): 887-896.
• Dupre, C., Yuste, R. (2017). Non-overlapping neural networks in hydra vulgaris. Curr. Biol. 27: 1085-1097.
• Galliot, B. (2012). Hydra, a fruitful model system for 270 years. Int. J. Dev. Biol. 56: 411-423.
• Gillette, T., Hill, J. (2015). Readers, writers, and erasers. Circ. Res. 1245-1253.
• Hubbard, S., Miller, W. (2007). Receptor tyrosine kinases: mechanisms of activation and signalling. Curr. Opin. Cell Biol. 19(2): 117-123.
• Josephson, R. (1967). Conduction and contraction in the column of hydra. J. Exp. Biol. 47: 179-190.
• Kier, W. (2012). The diversity of hydrostatic skeletons. J. Exp. Biol. 215: 1247-1257.
• Koizumi, O., Bode, H. (1991). Plasticity in the nervous system of adult hydra. III. Conversion of               neurons to expression of a vasopressin-like immunoreactivity depends on axial location.               J. Neuro. 11(7): 2011-2020.
• Kuznetsov, S., Anton-Erxlleben, F., Bosch, T. (2002). Epithelial interactions in hydra: apoptosis in interspecies grafts is induced by detachment from the extracellular matrix. J. Exp. Biol. 205: 3809-3817.
• Lengfeld, T., Watanabe, H., Simakov, O., Lindgens, D., Gee, L., Law, L., Schmidt, H., Ozbek, S., Bode, H., Holstein, T. (2009). Multiple wnts are involved in hydra organizer formation and regeneration. Dev. Biol. 330: 186-199.
• Manuel, G., Reynoso, R., Gee, L., Salgado, L., Bode, H. (2006). PI3K and ERK 1-2 regulate early stages during head regeneration in hydra. Dev. Growth and Differ. 48(2).
• Marcum, B., Campbell R. (1978). Developmental roles of epithelial and interstitial cell lineages in hydra: analysis of chimeras. J. Cell Sci. 32: 233-247.
• Sarras, MP Jr., Zhang, X., Huff, JK., Accavitti, MA., St John, PL., Abrahamson, DR. (1993). Extracellular matrix (mesoglea) of hydra vulgaris III. Formation and function during               morphogenesis of hydra cell aggregates. 157(2): 383-98.
• Schaller, HC. (1976). Action of the head activator as a growth hormone in hydra. Cell. Differ.               5(1): 1-11.
• Shimizu, H., Sawada, Y., Sugiyama, T. (1993). Minimum tissue size required for hydra regeneration. Dev. Biol. 155(2): 287-296.
• Shimizu, H., Zhang, X., Zhang, J., Leontovich, A., Fie, K., Yan, L., Sarras, MP Jr. (2002). Epithelial morphogenesis in hydra requires do novo expression of extracellular matrix               components and matrix metalloproteinases. Dev. 129(6): 1521-32.
• Wanek, N., Nishimiya, C., Achermann, J., Sugiyama, T. (1986). Genetic analysis of developmental mechanisms in hydra: XIII. Identification of the cell lineages responsible for the reduced regenerative capacity in a mutant strain, reg-16. Dev. Biol. 115(2): 459-468.

Appendix 1