Assessing the Pathogenic Cause of Sea Star Wasting Disease in Asterias Forbesi Along the East Coast of the United States

As keystone predators, sea stars serve to maintain biodiversity and distribution through trophic level interactions in intertidal ecosystems. Sea Star Wasting Disease (SSWD) has caused widespread mass mortality of Asterias forbesi in locations along the Northeast Coast of the US in recent years. A similar disease has been described in several sea star species from the West Coast of the US. Recently, a densovirus has been associated with wasting disease in West Coast sea stars and a few limited samples of A. forbesi. The goal of this research is to describe the pathogenesis of SSWD in A. forbesi and other echinoderms in the Northeast Coast of the US and to determine if the densovirus isolated from West Coast sea stars (SSaDV) is associated with the SSWD of A. forbesi on the eastern US coast. Histological examination of A. forbesi tissues affected with SSWD showed cuticle loss, edema, and vacuolation of cells in the epidermis but little to no evidence of pathology caused by bacterial or parasitic agents. Inclusion bodies were noted in two of the stars sampled. Challenge experiments by cohabitation and immersion in infected water suggest that the cause of SSWD is viral in nature, since filtration (0.22 μm) of water from tanks with SSWD does not diminish the transmission and progression of the disease. Death of challenged sea stars occurred 7-10 days after exposure to infected water or sea stars. Of the 48 stars tested by qPCR, 29 (60 %) have tested positive for the SSaDV VP1 gene. These stars represent wild-collected sea stars from all geographical regions (South Carolina to Maine), as well as stars exposed to infected stars or water from affected tanks. However, a clear association SSaDV with SSWD in A. forbesi from the East Coast of the US was not found in this study. Understanding the potential cause of this disease is a first step towards management and prevention of wide spread outbreaks.


LIST OF TABLES
. They may compete directly for space and food with other intertidal invertebrates, or may indirectly affect the competition between organisms (Menge et al. 1994). Over the last four decades, there have been periodic reports of mass mortality events in sea star populations along the Pacific coast of the United States, though no responsible agent was identified (Dungan et al. 1982, Scheibling 2010  Disease" (SSWD) (echinoblog.blogspot.com). Affected stars were flaccid and mucoid on the aboral surface, followed by dropping of the limbs and eventual disintegration of the central disc and body wall. This was often associated with the appearance of large multifocal lesions on the arms of stars, a sign of ulceration, sometimes followed by penetration through the test into the perivisceral coelom (Smolowitz, personal communication). Wasting Disease appears to have caused chronic mortality in the sea star, Asterias forbesi, in locations from Maine to New Jersey (echinoblog.blogspot.com).
The rapid spread of SSWD, as well as the widespread geographical distribution and species affected, makes it of great concern to individuals and organizations interested in the stewardship of biodiversity and the conservation of ocean resources (Hewson et al. 2014). There have been reports of disease outbreaks in echinoderm species from over the world for decades. From 1983From -1984, a mass mortality of the sea urchin Diadema antillarum occurred along their entire range in the Gulf of Mexico, leading to a loss of more than 93% of the existing population at each location (Lessios 1988). Sea Star wasting was first observed in the sunflower star, Heliaster kubiniji along the Gulf of California in June 1978. Stars exhibited white lesion on the aboral surface, while enlarged and led to fragmentation and death (Dungan et al. 1982). In the summer of 1997, more than ten species including Asterina (now Pateria) miniata and Pisaster giganteus were similarly affected at the Channel Islands. Stars exhibited similar signs, including loss of turgor, white lesions on the aboral surface, and finally fragmentation and death (Eckert et al. 1997). In many cases, episodes may be linked to increasing sea surface temperatures, though there is still no clear association (Menge et al. 1994;Scheibling and Lauzon-Guay 2010;Scheibling and Hennigar 1997 (Pinsino et al. 2007). The first barrier of defense sea stars have against pathogens is the cuticle and epithelium. The rigid cuticle helps prevent bacteria and viruses from entering the body wall, and provides support for the epithelial cells just below the surface. The major humoral component of the echinoderm immune system are coelomocytes, which freely circulate in the coloemic cavity, as well as the water vascular system and connective tissues. These cells are responsible for recognition of foreign antigens, and the initiation of immune responses including phagocytosis and release of antimicrobial enzymes (Beck andHabicht 1986, Fuess et al. 2015). in Pacific Coast sea stars were reported as higher in moribund than in healthy-looking individuals, and stars with higher viral loads were more likely to show clinical signs of SSWD. Viral load also increased as disease signs progressed, suggesting a potential relationship between SSaDV and SSWD (Hewson et al. 2014). However, this research did not provide direct proof that SSaDV is the causative agent of SSWD. This is mainly due to the lack of invertebrate cell lines allowing for the isolation and culture of SSaDV, which would facilitate fulfilling Koch's Postulates by providing a source of pure virus to challenge healthy stars and reproduce the disease in controlled conditions (Falkow 1988 (20-23ºC, 29-33ppt). If no clinical signs were observed, stars were used in experimental trials. One tank (3 stars) served as a control group throughout trials, and was not exposed to any diseased individuals.
Stars were fed every two weeks with snails or mussels collected from the GSO Pier. A complete list of all samples collected can be found in Appendix A.

Evaluation of disease range and timeline of epizootics
A survey was designed to obtain information on location and extent of mortality events in the wild and the water conditions associated with the die-offs (Appendices B, C). The survey was distributed to 5 local dive groups (RI, MA) and aquarists at the New England Aquarium (Boston, MA), Mystic Aquarium (Mystic, CT), and the Maritime Aquarium (Norwalk, CT). Samples of sea stars were received from several sites along the reported range (Table 1) and handled as described in the section above, with the exception of sea stars from South Carolina, which were directly placed in tanks (n=5, n=6, respectively) of filtered artificial seawater (FASW) at ambient conditions (19-23ºC, 29-32psu). Photographs, body condition scores, and swab samples were all taken upon arrival. The stars were observed daily for signs of disease onset (Table 2). These signs were documented through photography using an Olympus S2X10, with LG-PS2 illuminator scope and Olympus DP72 camera. Tissue samples were collected from moribund and dead stars and processed as described below for histological examination (all stars) and for microbiological analysis (moribund stars only). Categories of sea star health (turgor pressure, tube feet attachment strength, presence of absence of mucus, bloating or "pinched" look, and/or lesions) were established (Table 2), and each animal was evaluated and scored in each category.

Sample Collection and Processing
Sample collection: For each star (both showing clinical signs and not), two 1.5mL microcentrifuge tubes were filled with 1 mL filtered artificial salt water (FASW, 28 psu), labeled appropriately (swab or tissue), and placed in ice. The animal to be sampled was placed in a sterile disposable Petri dish and rinsed three times with 10 mL FASW to remove surface debris. Photographs were taken as described above to document gross morphology of animal, as well as size, date, water quality from the holding tank, and body condition were recorded. Using a sterile swab, one 1cm 2 area of rays was swabbed gently and the swab was placed into the corresponding microcentrifuge tube of 1 ml of FASW. If stars showed signs of disease, a swab each was taken from lesions and from an area with no visible lesions. Tissue clippings (2-3 mm 3 ) were collected from the epithelium of diseased stars and placed into microcentrifuge tubes containing 1mL FASW and kept in ice. Autotomized limbs and whole bodies were preserved in 10% formalin for fixation for histological examination of tissue. Swabs (after plating for bacteriological isolation, see below) and tissue samples were centrifuged for 10 min at 12,000xg at room temperature. Once the supernatant was decanted, 1 mL of TRIzol fixative was added to each tube and the tube was stored in the -80ºC for future analysis.
Histological examination: Samples of sea stars were removed from the fixative and rinsed with FASW. Cross-sectional pieces (2-3cm in width) were cut through the ray and included the body wall, coelomic cavity, and tube feet, and placed into histology cassettes. Cassettes were decalcified in a 0.5M EDTA-OH solution (pH=8) for 48-96 hours (Howard et al. 2004). The decalcifying solution was changed every 24 hours.
Once decalcified, cassettes containing sea star tissues were rinsed five times for five minutes each with ddH2O and placed in 70% ethanol to be processed by Mass Histology Services (Worcester, MA, USA). For each cassette, one 6-micron thick H&E stained section was received, photographed, and analyzed as described in Howard et al. (2004).
Bacterial culture, DNA isolation, and species identification: Swab samples were mixed using a Vortex (Service and Wardlaw 1985) before preparing serial 1/10 dilutions. An aliquot (20ul) from each of 4 dilutions was plated onto Seawater Tryptone (SWT, prepared with FASW at 30 psu), and Thiosulfate Citrus Bile Salt agar (TCBS) (Pfeffer and Oliver 2003) plates, incubated at room temperature (20 -24˚C), and monitored daily for bacterial colonies. The purpose of plating the samples was to identify culturable species present in lesions. Bacterial colonies in each of the media plates were classified based on morphology (color, shape, and type of growth) at 24 and 96 h after plating, and abundance of each colony type was recorded. Several colonies from bacteria that were present in the highest quantities in diseased animals, but in lower numbers or lacking in healthy individuals, were selected for storage and identification by sequencing of the 16S rDNA (Gauger and Gomez-Chiarri 2002).
Selected colonies of bacteria were lifted using a sterile loop and suspended in 5mL SWT broth and grown overnight for two days at room temperature with shaking. From this culture, 1 mL was placed in a tube, glycerol was added (20% volume) and tubes were stored at -80˚C. Another 1 mL was pipetted into a clean 1.5mL microcentrifuge tube and centrifuged at 1,792 RCF for five minutes. Bacterial pellets were washed two times with 500µL FASW, using a Vortex in between to resuspend samples.

Bacterial challenge experiments
One of the bacterial isolates identified in the bacterial sampling as a potential pathogen (based on abundance and predominance in disease sea stars and species identity) was used in challenge experiments to try to fulfill Koch's postulates.
Postulates to be fulfilled include: 1) The candidate pathogen (bacterial isolate) must be abundant in samples taken from sick stars, and not from samples taken from healthy ones; 2) Exposure of sea stars to the bacterial isolate must lead to SSWD in previously healthy stars; and 3) The candidate pathogen must be reisolated from the experimentally exposed star showing signs of disease. If all of these conditions are met, then the candidate pathogen can be considered a causative agent (Falkow 1988).
Treatments ( . Stars were monitored twice daily for ten days, and signs of morbidity or mortality recorded.

Cohabitation challenge experiments
The purpose of the cohabitation trials was to assess a timeline of disease progression, and to examine modes of transmission. Experiments predominantly assessed transmission between diseased and healthy A. forbesi. One trial involved sick echinoderms collected from the Maine State Aquarium. For all these trials, time to morbidity and mortality were recorded, as well as changes in behavior and physical appearance of stars. Swabs of lesioned areas were collected, as well as tissue clippings (2-3mm) for quantitative real time PCR analysis.

Cohabitation I
The first trial involved stars that showed clinical signs while held in a trough at the Graduate School of Oceanography Aquarium Building (termed "Source"). Four moribund stars were placed into each of 4 tanks containing 38 L of filtered artificial seawater (FASW). Three A. forbesi that had previously passed both stages of acclimation (see animal husbandry section above) and did not develop gross signs of the disease (termed "Challenged") were placed into each of the 4 tanks and allowed direct contact with moribund stars. Stars were then monitored for signs of wasting for 10 days. Once the Source stars died, their bodies were removed from the tank and a 20% water change was performed. Records of water quality, temperature, and body condition of cohabitation stars were taken daily for 10 days or until all Challenged stars were deceased.

Cohabitation II
The second cohabitation trial involved moribund animals received from the Maine State Aquarium (one sea star Asterias rubens, one green sea urchin Stronglyocentrotus droebachiensis, and 2 sea cucumbers Cucumaria frondosa).
Plastic mesh dividers were placed down the middle of the tanks to prevent direct contact of Source animals with Challenge animals. Three stars that had passed the acclimation phase and were negative for gross SSWD lesions were placed into each of the 4 tanks. Stars were then monitored once daily for 3 weeks, or until all cohabitation stars were deceased.

Cohabitation III: Infected Water Accumulation
The third cohabitation trial was named "Infected Water Accumulation". The goal of this experiment was to generate infected water to use in filtration challenge experiments, while also collecting samples for analysis. This trial lasted for 40 days, with consistent turnover of diseased stars. Water was taken from a tank in which SSWD had caused mortality and placed into a clean, sterilized tank with two A.
forbesi. Stars were sampled and monitored daily for signs of disease. When stars began to show signs, samples were taken for microbial analysis, and a new star was added to the tank. Each time a star was found moribund, the star was removed and a 20% water change was performed. Ammonia levels were recorded regularly to ensure proper water quality. A total of 15 stars were exposed in this way.

Filtration challenge experiments
The purpose of the filtration experiments was to separate infected water in different size fractions, in order to determine the size of the pathogenic agent.
Filtration through a 0.22-micron filter should remove any bacteria or larger particles.
The process of UV sterilization should break down any microbes that may have passed through the filter, and thus serves as a negative control.

Filtration I
Water from two tanks in which stars had experienced signs of SSWD was collected in 6-500mL bottles (named "Infected water") immediately after stars started to show signs of disease. These bottles were frozen and stored at -80ºC until challenge trials began. The rest of the water from the 2 infected tanks was separated into two treatment tanks: A) Whole, fresh untreated water from affected tank (infected water); B) 0.22µm filtered fresh infected water. Two A. forbesi that had gone through acclimation and showed no signs of disease were placed into each tank (n=2) and monitored daily. Trial continued for 10 days, or until all stars were dead.

Filtration II
In order to determine if the pathogenesis of SSWD in A. forbesi had a viral component, groups of stars were challenged with previously frozen water collected from tanks containing diseased stars that had been filtered to eliminate particles > 0.22 microns (Omran andEissa 2006, Ottesen 2011). Water was stored in 500 mL bottles in the -80ºC freezer. The bottles were removed from the freezer and placed in an ice bath to thaw (20-30 minutes) before being used in the experiment. As a negative control, a portion of this filtered water was then UV treated to inactivate potential viral particles remaining in the filtered water. A total of 1 L of filtered infected water was split into two 500 mL beakers (depth=15.2 centimeters) and placed inside a hood with UV light (TheSterilGARDHood, VBM400, the Baker Company, Inc.) for 4-5 hours.
Twenty stars without clinical signs were allowed to acclimate for 2-3 weeks while being monitored for signs, before 3 stars were placed into each of 8 tanks containing 38 L of filtered artificial sea water (FASW) in a closed circulation system at ambient conditions (16 -22˚C, 28 -32 psu). Stars were immersed in antibiotics (Enrofloxacin 2.5 mg/kg) for one hour prior to treatment. Infected water was filtered (down to 1 µm) and UV treated (EU25-U, Pentair-Emperor Aquatics) after each pass through the system ( Figure 1). Four different treatments were established: 1) control, FASW, 2) 0.22-micron filtered infected water, 3) 0.22-micron filtered and UV treated infected water, and 4) whole, untreated infected water. Experimental stars were monitored for an additional 3 weeks for signs of SSWD. Swab samples were collected for processing before stars entered treatment, when they started to show clinical signs, and at death.

Filtration III
A third trial was run in order to replicate results of Filtration II, but with some adjustments. Water for this trial was obtained from infected tanks and used immediately for exposure (fresh infected water). Stars in this trial did not receive any antibiotic treatment prior to exposure. The trial included the same treatment groups as Filtration II. Stars were monitored for 3 weeks for signs of SSWD. Samples were collected for processing before stars entered treatment, when they started to show clinical signs, and at death.

Quantification of SSaDV in A. forbesi using quantitative real time PCR
A Taq based assay for quantitative real time PCR for detection of the VP1 These clinical signs were also seen in echinoderms cohabitating with the diseased specimens introduced in the tank. Specimens of cohabitating echinoderms showing signs of disease were sent to URI for analysis (Table 1)   Loss of turgor pressure was defined as a deflated appearance and lack of rigidity.
Bloating was defined by a "pinching" of the body wall along or at the base of the rays.
A mucus coat caused the spines to appear smooth and glossy. Stars from the wild, as well as those that were experimentally exposed to diseased stars, started to show signs and were dead within 7-10 days. The chronic form shows a much slower progression.
Stars may exhibit lethargy and the development of small pinpoint (<3mm) lesions, but these signs may persist for weeks to months before any mortality is noted, if at all. A limited number of animal obtained (n=5) exhibited this form. These stars survived with minor lesions for 115 ± 74 days.

Bacterial challenge experiments Isolation and characterization of bacterial from SSWD A. forbesi
On average, stars that did not show lesions of SSWD had a lower amount of colony forming units (CFU) per mL of samples collected through swabbing, though it

A B
was not statistically significant (p > 0.05). The amount of CFU/mL was similar in areas with and without lesions in diseased stars (Figure 6), suggesting that the bacteria is not contained to visible lesions in infected individuals. Stars (A. forbesi) exposed by immersion to the Roseobacter sp. isolated from lesions of SSWD stars exhibited lethargy, weak tube feet attachment to substrate, and slow righting response within 9 days of being exposed to the bacteria, while control stars not exposed to the bacteria did not show any signs of morbidity or mortality   Healthy-looking stars were exposed to four treatments (n = 6 per treatment): 1) Control; 2) Animals immersed seawater with 10 6 CFU/mL of bacteria for one hour with no other manipulation; 3) Animals immersed in seawater with 10 6 CFU/mL of bacteria after cuticle abrasion 4) Animals injected with 0.1mL of a 10 5 CFU/mL solution of bacteria in seawater through the dorsal epithelium of one ray into the coelomic cavity; 5) Animals injected with 0.1mL FASW water. Mortality was only observed in Immersion no Abrasion, FASW injection, and Immersion Abrasion.

Cohabitation challenge experiments
Preliminary challenge experiments in which Pateria miniata affected by SSWD cohabitating with apparently healthy A. forbesi showed that transmission of the syndrome is independent of direct contact between hosts, and that exposure to water from a diseased animal's tank is enough to cause mortality in an otherwise healthy individual. Transmission also occurred between local species and a Pacific Coast species (P. miniata) in laboratory cohabitation trials, with progression to death occurring within 5 days of the start of the cohabitation (Wessel, personal communication). The first signs of infection included loss of turgor pressure, and an increase of mucus coat on the aboral surface. White lesions resulting from ulceration through the epidermis into the underlying calcified white plate eventually led to ulceration of the gut and internal tissue. Transmission of the disease between A.
forbesi collected from Narragansett Bay, RI, and a naïve Pacific Coast star in cohabitation experiments indicated that the pathogen is not species-specific.

Cohabitation I
Within 3 days of placing a moribund A. forbesi into tanks (n = 4) each containing 3 sea stars without clinical signs, all Source stars with SSWD had died.
Morbidity in cohabitating stars were seen 3 days post exposure, with 77.7% mortality occurring by day 5 post exposure. No mortality was observed in control stars ( Figure   9). Clinical signs of wasting, including loss or turgor, lesion formation, and limb dissociation, were observed in cohabitation stars. T a n k 1 T a n k 2 T a n k 3 T a n k 4 C o n tro l Upon arrival from the MSA, one of the sea cucumbers (Cucumber 1) had a small, pinpoint (<1cm) lesion on the lateral surface. By the next day, the lesion measured over 5cm in length. The lesion continued to increase in size, by three weeks the lesion had ulcerated, and the internal tissue had been eviscerated. No change in physical condition was noted in the sea stars cohabitating with this sea cucumber (no loss of turgor, lesion formation), but 2 of the 3 were found dead 40 days post exposure. The second sea cucumber (Cucumber2) never expressed signs of distress or disease, and neither did any of the stars in cohabitation. Cucumber 2 was found dead in tank at day 45. Differences in clinical signs in sea cucumbers from MSA, as well as differences in time to morbidity and mortality and the inability to transfer the disease to cohabitating sea stars suggest that the mortality seen in sea cucumbers at the Maine Aquarium was not due to SSWD.

Cohabitation III: Infected Water Accumulation
Healthy-looking A. forbesi stars (n = 2) that were placed into the exposure tanks containing a diseased sea star began to show signs of SSWD (morbidity) at 4.4 ± 2.6 days after exposure to a diseased sea star and mortality by 5.3 ± 2.6 days post A B C D exposure. By the end of the trial, 14/15 (93%) stars had presented signs of wasting and died within 5 days of being exposed to infected water and individuals (not shown).
These results are consistent with those seen in Cohabitation I, which was also a species-specific trial.

Filtration I
Stars in the Whole water treatment started to show signs of morbidity within 2 days, and 100% mortality had occurred in these groups by 4 days post exposure.
Morbidity was noted in the Filtered treatments at 3-6 days, with 100% mortality by 15 days post exposure ( Figure 11). No morbidity or mortality was noted in Control tanks during this time.  Figure 11: Effect of incubation in water collected from tanks with diseased sea stars on survival of healthy-looking stars (Filtration I). Healthy-looking acclimated stars from Narragansett Bay were exposed to: freshly collected infected water (whole), n=4, yellow line; 0.22 µm filtered infected water (filtered), n=4, blue line; and no treatment (control), n=3., black line

Filtration II
Stars in tanks receiving 0.22-micron filtered water were the only ones to express morbidity and mortality associated with SSWD in this trial. Clinical signs including loss of turgor, limb curling, and lesion formation were observed at 2-6 days post exposure, with 40% mortality occurring by day 10. Though no signs of disease were noted, one star in the whole water treatment was found dead by 32 days post exposure ( Figure 12). Figure 12: Effect of incubation in water collected from tanks with diseased sea stars on survival of healthy-looking stars (Filtration II). Healthy-looking acclimated stars from Narragansett Bay were treated with antibiotics and then exposed to: previously frozen infected water (whole, n= 6, orange line); 0.22 µm filtered frozen infected water (filtered, n= 6, blue line); 0.22 µm filtered and UV treated frozen infected water (filtered + UV, n= 6, green line); and no treatment (control, n=6, black line).

Filtration III
In this trial, mortality was seen in all treatment groups. In control tanks, all stars were moribund by day 5, and 50% mortality occurred by day 13 post exposure.
No other control stars expressed clinical signs associated with SSWD, but 83% mortality had occurred by day 35 post exposure. In tanks that received 0.22-micron filtered water, 33% mortality occurred within 26 days post exposure. In tanks with  Figure 13: Effect of incubation in water collected from tanks with diseased sea stars on survival of healthy-looking stars (Filtration III). Healthy-looking acclimated stars from Narragansett Bay were treated with antibiotics and then exposed to: previously frozen infected water (whole, n= 6, orange line); 0.22 µm filtered frozen infected water (filtered, n= 6, blue line); 0.22 µm filtered and UV treated frozen infected water (filtered + UV, n= 6, green line); and no treatment (control, n=6, black line).

Summary of challenge trials: comparison of time to morbidity and mortality between experiments
Average time to morbidity and mortality varied between groups. Among the 9 groups analyzed, the means varied significantly for both morbidity and mortality ( Figure 14). In the bacterial challenge experiment, stars showed signs of disease between 8.7 ± 5.2 days, and mortality in 24.7 ± 7.5 days. By contrast, the stars in Cohabitation I showed signs in 3.5 ± 0.5 days, with mortality averaging 4.4 ± 0.5 days.
Time to clinical signs and mortality for sea stars used in the Cohabitation II trials depended on the species of the source animal received from the Maine State Aquarium, ranging from 10 days for the stars exposed to a source sea star and sea urchin to no mortality seen for stars exposed to sea cucumbers. In the cohabitation trials named "Infected water accumulation," morbidity consistently occurred within one week (7 days), and death within 2-8 days post exposure. Between the two most reliable filtration trials (in which no mortality was observed in control animals), morbidity occurred at 3.1 ± 1.5 days (trial I) and 9.6 ± 10.4 days, (trial II) and mortality at 6.4 ± 4.5 (trial I) and 12.6 ± 10.9 (trial II) days post exposure. Differences in time to morbidity and mortality between these 2 trials may have been due to the antibiotic treatment performed in trial II (but not on trial I), differences in environmental conditions or condition of acclimated animals, and/or differences in infective dose in the collected water.

Detection and cloning of SSaDV VP1 in A. forbesi
Samples selected for end-point PCR testing showed amplification using primers designed from the VP1 and VP4 sequences of SSaDV. Of the samples tested, 11/14 (78.6%) elicited DNA bands in the region associated with VP1 and 8/14 (57%) bands in the region associated with VP4 ( Figure 15). Bands matching the expected range of VP1 and VP4 (285 and 492 bp, respectively) were selected for cloning and sequencing. Sequencing of these bands yielded 3 positive identifications of the VP1 gene with 100% identity to the sequence in GenBank (consensus sequence shown in Figure 16). However, no sequences were recovered with any percent identity to the VP4 sequence in GenBank (not shown). Using the sequencing results for VP1 amplified from A. forbesi samples, a ClustalW analysis was used to determine conserved VP1 gene sequences to be used in the design of SSaDV qPCR primers and probe (see highlighted area in Figure 16: FWD and REV=yellow, PRB=blue).

Comparison of VP1 detection between swab and tissue samples
Two types of samples were collected from stars: a skin swab (DNA resuspended in 1000 µL) and a tissue sample (approximate weight = 0.2 mg). Samples obtained from a skin swab averaged 8.92 x 10 19 ± 3.90 x 10 19 copies/µL, while tissue samples averaged 1.25 x 10 20 ± 9.79 10 19 copies/µL (Figure 17). Of the 33 swab samples tested, 22 (33.3%) were positive for the target VP1 sequences. Of the 15 tissue samples tested, 7 (46.6%) were positive for VP1. Unfortunately, we were not able to run a comparison of both sample types collected from the same star. Although the values for tissue samples were slightly higher and showed smaller variation, differences in concentrations between sample types did not differ significantly (p=0.7388). Since no significant difference between swabs and tissue samples were obtained, data from either tissue was included in further analysis. Figure 17: Concentration of SSaDV VP1 (copy number/µL) in swab and tissue samples from challenged stars that were positive for VP1 (Swab samples n=22, Tissue samples, n=7). Box and whisker plot represents minimum and maximum values, quartiles, and mean.

Quantification of SSaDV VP1 in all sea star samples
Of the 48 stars tested, 29 (60 %) tested positive for the VP1 gene through qPCR. These stars represent all geographical regions (South Carolina to Maine), and include stars collected from the wild (n = 6) and stars used in challenge experiments (n = 9) (Appendix A). Of these 29 SSaDV-positive stars, 15 (51%) exhibited gross morphological lesions consistent with SSWD. Atlantic US coast stars positive for SSaDV (both wild and experimentally exposed) showed comparable or higher levels of SSaDV (average Cp of 28.06 ± 9.8, corresponding to a concentration of 3.1 x10 10 ± 6.2 x 10 10 copies/µL based on a standard curve using the cloned VP1 target) as the positive control from the Pacific US coast provided by Ian Hewson (Cp of 33.61 ± 3.79; estimated concentration of 6.85 x 10 5 ± 9.69 x 10 5 copies/µL based on the same

Quantification of SSaDV VP1 in wild-collected samples
Of the 34 wild-collected samples tested, 14 were positive for VP1 (41.2%), with a concentration (mean ± SD) of 9.8 x 10 19 ± 1.9 x 10 20 copies/µL. Concentrations of VP1 in wild-collected samples ranged from 3.91 x 10 10 to 6.08 x 10 20 copies/µL, with the highest viral levels seen in selected samples from Beavertail and the GSO pier in Rhode Island. According to location, 10/13 (76.9%) of the samples collected from Beavertail, RI, tested positive for VP1, with 9/15 (60%) positive samples from the GSO pier (RI), and 6/16 (37.5%) positive samples from South Carolina ( Figure   18). Only one out of 3 echinoderms (the sea star) collected from the Maine State Aquarium tested positive for VP1, with a concentration if 9.7 x 10 10 copies/µL.

Quantification of SSaDV VP1 in sea stars from challenge experiments
Of the 14 sea stars exposed to the disease through experimental challenges, 9 were positive for VP1 (64.3%), with an average concentration of 9.7 x 10 19 ± 2.3 x 10 20 copies/µL ( Figure 19). VP1 concentration was not statistically significant (p=0.462) between wild-collected and experimentally challenged samples that are positive for VP1 (negative samples excluded from this analysis). The 9 (out of 14) experimentally exposed sea stars positive for VP1 included 3 (33.3%) that showed no gross signs of wasting, and 6 (66.6%) that expressed limb curling, bloating, or other SSWD lesions.
Within 10 days of exposure, 14/15 (93.3%) stars had died, with the last one dead by day 15, with gross signs of wasting. Figure 19: Concentration of SSaDV VP1 (copy number/µL) in wild-collected and experimentally exposed sea stars that tested positive for VP1 (Wild samples n=20, Experimental samples, n=9; negative samples were excluded from this analysis). Box and whisker plot represents minimum and maximum values, quartiles, and mean.    Additionally, of the ten samples with the highest VP1 concentration (copy number/µL, range=2.97 x 10 15 -6.92x 10 20 ), only sixty percent showed clinical signs of disease.

Changes in VP1 concentration in sea stars
We also sought to determine if viral load would increase in disease stars as the disease progressed. For stars received from Charleston, South Carolina, we were able to collect samples from the same stars at different stages of the disease for quantification of VP1. On the day of arrival from Charleston, VP1 concentrations in these stars averaged 6.5 x 10 1 ± 10.0 x 10 1 copies/µL (Figure 21). At this stage, stars did not show any signs of SSWD. By the third day after placement of stars in the tanks, these stars experienced 100% mortality showing signs of SSWD.
Concentrations had increased to 2.9 x 10 6 ±3.8 x 10 6 . Although not statistically significant at a 0.05 alpha level (p=0.0907) due to variability, all stars showed lethargy, weak tube feet attachment, limb curling, and lesion formation. Comparative increased water temperatures (Eckert et al. 1997, Scheibling andLauzon-Guay 2010;Scheibling and Hennigar 1997). Interestingly, we observed two major die-off events in our holding tanks, which occurred in both October of 2014 and 2015. This is the time when seawater temperatures decrease rapidly in Narragansett Bay (~0.3 ºC per day) suggesting that this changing temperature conditions may be a trigger for disease epizootics. A more thorough analysis of the timeline of the Atlantic coast of the US outbreak is needed to examine the relationship of mortality events to changes in environmental conditions such as temperature, salinity, and pH, as well as potential relationships with changes in food availability.
Clinical signs and a timeline of disease progression have been defined for A.
forbesi based on gross observations and histological examination. These signs include lethargy, limb-curling, loss of tube feet attachment to substrate, and lesion formation leading to ulceration of internal tissue and death, and are similar to those described for sea stars affected by SSWD in the Pacific coast of the US (Hewson et al. 2014 could be related to an upregulation in expression of genes related to cell adhesion, nervous system, and connective tissue management (Fuess et al. 2015). These three systems are crucial to maintaining structure and shape in sea stars, so an upregulation of gene expression may reflect the gross morphological appearance of infected individuals. Fuess et al. (2015) also report the upregulation of genes associated with immune cell production (coelomocytes, macrophages) in response to injection of viral sized particles from diseased animals. In our study, examination of diseased A. forbesi tissue consistently shows an influx of hemocytes and immune cells, supporting the idea that Atlantic coast sea stars may mount a similar immune response to SSWD exposure.
Results from our cohabitation suggest that the disease is highly transmissible in A. forbesi and A. rubens, leading to rapid and severe morbidity and mortality within 10 days of exposure of a healthy-looking star to a diseased star. Furthermore, filtration trials involving water collected from tanks with sea stars experiencing mortality to SSWD indicate that a viral pathogen is the most likely cause of SSWD in Asterias spp.
in the Atlantic coast of the US. Although several bacterial species have been found to be pathogenic to echinoderms (Becker et al. 2008) and some morbidity and mortality of A. forbesi was observed in stars exposed to bacteria isolated from SSWD stars, the time to morbidity and mortality (significantly longer for stars exposed to bacterial challenge) and the gross and histological signs of disease observed in stars from the bacterial challenge are not consistent with SSWD as observed in both wild and experimentally exposed stars. A viral pathology is also consistent with the conclusions by Hewson et al. (2014) for the SSWD outbreaks in the Pacific coast of the US. Interestingly, in our experiments, high levels of morbidity and mortality were observed in A. forbesi after exposure by immersion of healthy-looking sea stars to fresh or frozen water from infected tanks, while injection challenges with viral sized particles (which should lead to enrichment in the pathogen) were used in experimental challenges of Pacific Coast sea stars (Hewson et al. 2014).
Studies from the SSWD outbreaks in the Pacific coast showed that several species of sea stars are susceptible to SSWD (Hewson et al. 2014 DNA isolated from sea star samples were not processed in the same way as those in Hewson et al. (2014), which prevents us from performing direct comparisons on viral loads between studies. In our study, viral DNA was not isolated from samples, meaning that they reflect microbial as well as sea star DNA, and may result in a dilution of target DNA values. Furthermore, results from the Pacific coast are reported as copies/mg -1 tissue. Samples testing positive for SSaDV from ten Pacific sea star species ranged in concentration from 1.0 x 10 3 to 1.0 x 10 9 copies/mg -1 for the VP4 sequence. From the fifteen tissue samples analyzed from Atlantic coast stars in Hewson et al. 2014, the concentration of VP1 ranged from 2.68 x 10 11 to 3.46 x 10 18 copies/mg -1 , with an average of 3.89 x 10 10 ±8.09 x 10 10 copies/mg -1 . These values fall above the reported range of viral load for Mediaster aequalis, Pisaster giganteus, Pisaster brevispinus, and Patiria miniata (Hewson et al. 2014). Many of the stars tested in our study showed similar or higher viral loads than the positive control obtained from the Hewson laboratory. Future research should focus on analyzing more tissue samples from the east coast in order to provide substantial comparisons to results from the west coast, and to assess the efficacy of tissue as opposed to swab samples. Our research has led to the development of a tool that could be used to screen samples for presence or absence of SSaDV in sea stars in the Atlantic coast of the US.
Our results obtained through quantification of VP1 should also be confirmed through analysis of additional targets in the sequence of SSaDV. Sequencing and characterization of SSaDV from Asterias spp. would allow for the development of other screening tools for SSaDV.
More research is also needed to confirm a viral etiology for this disease and determine if SSaDV is the causative agent of SSWD. Results from our filtration challenge experiments were not always consistent. Factors leading to differences in the outcome of these 3 filtration experiments include whether stars were treated with an antibiotic (Enrofloxacin by injection) prior to challenge, whether water from infected tanks was frozen or not prior to the challenges, and/or differences in viral load between samples of water used in the filtration experiments. Furthermore, there are several major pitfalls for identifying the causative agent of SSWD through challenge experiments, including: 1) the lack of a reliable source of sea stars from an area in the Atlantic coast of the US potentially free of the disease; 2) difficulties in confirming that stars used for the experiments are not already infected with the pathogen causing SSWD due to a lack of a screening tool for the pathogen (qPCR for VP1 was not developed until the final stages of this research); 3) lack of knowledge on the environmental conditions triggering SSWD outbreaks; and 4) the lack of marine invertebrate cell cultures that could be used to isolate and culture candidate viral pathogens.
It should also be noted that there is the potential that some of the stars have developed a resistance to SSWD. Recent reports (Fall 2015) have stated that some populations of A. forbesi around Rhode Island seem to be rebounding. It is unclear whether these stars are the result of spawning post-outbreak, or if some managed to survive the outbreak altogether. Future work should seek to assess any populations that may be untouched by wasting disease. Finding such populations could be beneficial for future work by providing better controls, lending support to any conclusions.
In summary, our research shows that outbreaks of SSWD similar to the ones affecting several species of sea stars in the Pacific Coast of the US have also affected Asterias spp. in the Atlantic Coast of the US. Challenge experiments confirm a viral etiology for the disease. In addition, SSaDV has been detected in Asterias spp. from the Atlantic coast of the US, although a clear association of SSaDV with SSWD has been found in these stars. More research is necessary to characterize the epizootiology of the disease and identify the causative agent.