Hook‐shaped enterolith and secondary cachexia in a free‐living grey nurse shark (Carcharias taurus, Rafinesque 1810)

Abstract The carcass of a critically endangered, juvenile female grey nurse shark (Carcharias taurus, Rafinesque 1810) was recovered from a south‐eastern Australian beach and subjected to necropsy. The 1.98‐m‐long shark exhibited advanced cachexia with its total weight (19.0 kg) and liver weight (0.37 kg) reduced by 60% and 89%, respectively, compared with a healthy individual of the same length. Marked tissue decomposition was evident preventing histopathology and identification of a definitive cause of death. At necropsy, the abdominal organs were abnormally displaced and showed marked reductions in size compared with a healthy individual of the same size. Importantly, a hook‐shaped enterolith (HSE), with a rough surface and cream in colour, was found within the spiral valve of the intestine and is to the authors’ knowledge, the first description of such in any marine animal. X‐ray diffractometry showed that the HSE comprised the minerals monohydrocalcite (Ca[CO₃].H₂O; ~70 wt%) and struvite (Mg [NH4] [PO4]. [H2O]6; ~30 wt%). A CT scan showed concentric lamellate concretions around a 7/o offset J‐hook that formed the nidus of the HSE. Nylon fishing line attached to the hook exited the HSE and was evident in the abdominal cavity through a perforation in the intestinal wall where the posterior intestinal artery merges. The most parsimonious reconstruction of events leading to enterolithiasis and secondary cachexia in this shark was the consumption of a hooked fish and subsequent hook migration causing perforations of the cardiac stomach wall followed by the thin, muscular wall of the apposed, sub‐adjacent intestine.

Observations of enteroliths and/or uroliths are relatively rare in marine fauna, particularly in free-living animals and only a few, notable examples have been documented. A 100-mm spherical, calcium hydroxyapatite enterolith with an octopus beak nidus was surgically removed from the intestine of an adult female sand tiger shark held in London Aquarium (Montreal-Pawlowsky, Thornton, Stidworthy, & Hale, 2016;Thornton, Monreal-Pawlowsky, Stidworthy, & Hale, 2012). A carbonate-apatite/struvite enterolith with a stingray spine nidus was recovered, at necropsy, from a stranded Atlantic bottlenose dolphin, Tursiops truncatus, (Burdett & Osborne, 2010). Two struvite/apatite uroliths were found, at necropsy, in the urogenital sinus of a 70-kg captive, female C. taurus from Sea World, Florida USA (Walsh & Murru, 1987). Finally, two 10 mm diameter, enteroliths of unknown composition were also found at necropsy in the distal intestine of a captive green moray eel, Gymnothorax funebris, (Boylan et al., 2016). The scant information on enterolithiasis in wild marine animals is especially evident in relation to its occurrence among species, enterolith composition and the causes and clinical consequences. This is likely due to few dedicated programmes undertaking necropsies of marine fauna and the difficulties in finding and accessing carcasses other than cetaceans.
This case report documents necropsy results for a stranded, female C. taurus recovered from One Mile Beach, Port Stephens, NSW, Australia (32° 46.755'S, 152° 7.010'E) on 5 March 2009. The stranding incident was reported at 1,300 hr by a local council lifeguard and, when recovered an hour later, the shark was extremely emaciated, and decomposition was evident. At necropsy, a hook-shaped enterolith (HSE) was found in the intestine. We use results from necropsies of healthy C. taurus to assess the anatomical changes associated with enterolithiasis and secondary cachexia in this shark. We also describe the structure and composition of the enterolith and the likely events leading to its formation.

| Necropsies
Necropsies of C. taurus commenced by recording total weight (TW, nearest 0.5 kg) using a chain winch with suspended dial-weighing scale (Salter, Model 235 10X, max 200 kg) rigged on a weighing gantry. While each shark was suspended, a hand-held metal detector (Ranger Security Detectors Inc., EL Paso, Texas, USA) was used to locate possible retained fishing hooks. Thereafter, carcass lividity, eye condition, colour and presence/absence of gill filaments, tooth colour and their gingival retention, and any spinal deformities were recorded to assess the ante-mortem health and degree of decomposition using criteria modified from Otway (2015). The external examination concluded by measuring 50 morphometric lengths to the nearest mm (e.g. Bass et al., 1975), but are reported to the nearest two decimal places. Included were: total length (TL) with the caudal fin in a depressed position (Francis, 2006); precaudal length (PCL); snout to pectoral fin origin length (SPecO); snout to pelvic fin origin length (SPelO); abdominal cavity length (ACL, via SPelO-SPecO); pectoral-pelvic space (PPS) and five girths at standardized long-axis locations (Table 1).
An internal examination with gross inspection of the organs also included measurement (length and diameter to nearest mm) of the liver, spleen and GI tract organs comprising the descending cardiac stomach, ascending pyloric stomach and intestine with ring/columnar spiral valve (Holmgrem & Nilsson, 1999;Leigh, Papastamatiou, & German, 2017). Tissue samples were taken and preserved in neutral buffered formalin for histopathology. Reproductive status was assessed following standard methods (Gilmore, Dodrill, & Linley, 1983;Walker, 2005) with liver weight (LW) and gonad weight recorded (nearest 0.1 kg). Sexual maturity was evaluated using the TL, PPS, reproductive tract development and presence/absence of uterine hymen. Bile from the gall bladder in the left liver lobe (Figure 1b) was aspirated using a 10 ml disposable syringe and 18G × 38 mm needle (Terumo, Tokyo, Japan). Fluids from the abdominal cavity and organs of the GI tract were aspirated using disposable syringes of varying volumes (10-50 ml). Finally, age was estimated using thin sections of vertebral centra sampled from under the cranial (first) dorsal fin (Goldman et al., 2006).

| Laboratory analysis of the HSE
The HSE was dried to constant weight in an air-tight container with silica gel at 21°C and then weighed (nearest 0.5 g) on an electronic balance.
A sub-sample was pulverized, and elemental composition determined using an Olympus Delta Pro X-ray fluorescence spectrometer with tantalum anode tube, which can quantify the concentrations of elements from aluminium to uranium. In this study, operating conditions of 50 kV (strontium-lanthanum), 40 kV (iron-zinc, arsenic-rubidium, tantalumbismuth) and 15 kV (phosphorus-manganese) were used, with a total measurement time of 90 s. Limits of quantification vary for each element and sample, but are typically around 0.001 wt% for elements from rows 4 to 6 in the periodic table, and 1-0.1 wt% for elements in row 3.
Measurements have typical accuracies of better than 20% where concentrations are >0.1 wt%. Mineralogy was determined using X-ray diffractometry. A sample (~5 mg) was hand milled under ethanol, dried and mounted as a powder on a silicon crystal low background sample holder.
TA B L E 1 Morphometric relationships derived from healthy free-living grey nurse sharks (Carcharias taurus) subjected to necropsy following capture in the coastal waters off south-eastern Australia Note: Regression for liver weight (LW) on total length (TL) was determined for juvenile females.
Abbreviations: ACL, abdominal cavity length; all lengths and girths in metres; FL, fork length; G1DO, girth at the first dorsal fin origin; GPCP, girth at the cranial edge of the precaudal pit; GPecI, pectoral fin insertion; GPecO, girth at the pectoral fin origin; GPelO, girth at the pelvic fin origin; LW, liver weight in kilogrammes; PCL, precaudal length; PPS, pectoral-pelvic space; S1DO, length of snout to first (cranial) dorsal fin origin; S2DO, length of snout to second dorsal (caudal) fin origin; SAL, snout to anal fin length; SPecO, length of snout to pectoral fin origin; SPelO, length of snout to pelvic fin origin; TL, total length; TW, total weight in kilogrammes.
Diffractograms were collected from 5 to 90° 2θ using a PANalytical

| Statistical analyses
To date, 195 C. taurus have been subjected to necropsy following their incidental capture along the south-eastern Australian coastline and the associated morphometric data from individuals classified as healthy ante-mortem were used to generate least-squares linear regression relationships. Least-squares linear regressions of TW and LW on TL were also generated following log 10 transformation.
Differences between the TW, LW, morphometric lengths and girths of healthy C. taurus and the stranded female were assessed using 2-tailed asymmetrical t-tests (Sokal & Rohlf, 1969) with predicted values for a healthy 1.98 m TL individual obtained from each respective linear regression. The dimensions (length, diameter to nearest mm) of the liver lobes, cardiac stomach, pyloric stomach, intestine and spleen were expressed as a proportion of the ACL as this did not change significantly with increasing TL (range, 0.20-0.29).
Differences in the proportional lengths of the liver, GI tract organs and spleen between healthy C. taurus and the stranded female were assessed using two-tailed asymmetrical t-tests. To this end, data from a subset of 62 healthy individuals (both sexes) that had not fed for at least 24-hr ante-mortem were used to control for cardiac stomach expansion that occurs with feeding.

| Healthy C. taurus
One hundred and thirteen C. taurus (53 males, 60 females) were classified as healthy ante-mortem and had normal copper-brown skin and no pallor, eyes normal in appearance, bright red gill filaments and white teeth that were firmly anchored in the cream-coloured gingiva. None of these sharks had skin lesions, spinal deformities, ingested hooks or other fishing-related injuries. Histopathological analyses on a subset of these sharks were also unremarkable and confirmed the ante-mortem health assessment. Their TW, LW and morphometric data resulted in significant linear regression relationships accounting for 58%-99% of the variation in the respective dependent variables (Table 1).

| Stranded female C. taurus with HSE
When recovered from the beach, the stranded female C. taurus had minimal lividity remaining, the skin was pale fawn and the normal red-brown spots had faded substantially. The gill arches and pectoral and pelvic girdles were prominent under the skin because of skeletal muscle loss. The gill filaments were totally absent from the five gill slits on both sides of the shark and its eyes were sunken into the orbits and dehydrated. The orobranchial cavity was abnormally bright white in colour and while the white teeth were anchored within the abnormally white gingiva, they could be removed easily in contrast to recently deceased individuals. There were no spinal deformities or external fishing related injuries, but examination with the metal detector indicated that a likely fishing hook had been retained within the abdominal cavity along the ventral mid-line, inferior to the cranial (first) dorsal fin origin and cranial to the pelvic fin origins.
Finally, tissue samples for histopathology were not collected from the stranded female owing to the degree of post-mortem autolysis.
The results of comparisons between the healthy (n = 113) and stranded female C. taurus using 2-tailed asymmetrical t-tests are summarized in Internally, the abdominal cavity contained ~200 ml of serous fluid and the organs of the GI tract were markedly reduced in size (Table 3 for proportional dimensions) and this altered their normal 'S-shaped' position within the abdominal cavity (Figure 1a). The liver was bluish-black in colour, greatly reduced in size (Figure 1a and b) and the 0.37 kg LW was significantly less than the predicted 3.27 kg (Table 2, p < .001) and represented an ~89% reduction. The liver lobes were abnormally different in length and the left lobe, with gall bladder, was ~22% shorter than the right lobe ( Figure 1b). The right liver lobe proportional dimensions were significantly less than the respective means in healthy sharks (Table 3,  The stranded female's cardiac stomach was greatly reduced in size leading to its cranial retraction within the abdominal cavity ( Figure 1a). The proportional length and diameter of the cardiac stomach were significantly less than the respective means in healthy C. taurus (both p < .001) and represented reductions of ~32% and ~38%, respectively. The pyloric stomach was also markedly reduced (~52% and ~32%, respectively) and the proportional length and diameter were significantly less than their respective means in healthy C. taurus (both p < .001).
The intestine had an irregularly coiled length of orange nylon fishing line (120 mm long × 1.20 mm diameter) protruding through a perforation in the midline of the thin, dorsal wall immediately proximal to the junction of the inflamed, posterior intestinal artery (Figure 1c).

TA B L E 2
Comparisons between the cachectic, female grey nurse shark (Carcharias taurus) with hook-shaped enterolith and normal healthy individuals using two-tailed asymmetrical t-tests based on the morphometric regression relationships in Table 1  Note: Lengths and girths in metres; total weight in kilograms; subscript denotes the independent variable in each respective linear regression in Table 1.
Intestinal dimensions were also markedly reduced (~46% and ~26%, respectively) and the proportional length and diameter were significantly less than their respective means in healthy C. taurus (both p < .001).
The spleen was predominantly pink-red and this contrasted with its dark blue to red-purple colour in healthy C. taurus and had dimensions similar to the pyloric stomach. The spleen's proportional length and diameter were significantly less than the respective means in healthy C. taurus both p < .001) and represented reductions of ~53% and ~67%, respectively. Finally, the pancreas was an abnormally pale cream colour, appeared markedly reduced in size, but was not measured or weighed for quantitative comparison.

| Weight, composition, mineralogy and structure of the HSE
The dry weight of HSE and attached nylon line was 20.5 g. Elemental analyses revealed the major elements calcium (~26 wt%) and phosphorus (20 wt%), with minor elements sulphur, chlorine and potassium  Figure 2).
The CT scan clearly defined the three-dimensional, external and internal features of the enterolith which were centred on the hook and led to its particular shape (Figure 3a and b).

| D ISCUSS I ON
While this account is not the first description of an enterolith or urolith in C. taurus, it is to our knowledge, the first description of a hookshaped enterolith in any marine animal. Given the apparent GI tract stasis, it is very likely that this shark became anorexic and the minimal energy assimilation led to the 60% and 89% reductions in TW and LW, respectively. The advanced cachectic state of the shark also suggested that no demonstrable skeletal growth would have occurred after the initial hooking and enterolithiasis. This outcome is similar to the captive 2.70 m TL female C. taurus in London Aquarium that exhibited severe weight loss over 18 months (Monreal-Pawlowsky et al., 2017;Thornton et al., 2012). At surgery, a 100 mm diameter spherical enterolith, causing partial obstruction of the spiral valve, was removed from the intestine. The enterolith had concentric lamellate layers and the nidus was an undigested octopus beak. Unfortunately, the shark died 7 hr after surgery without recovering from anaesthesia. The subsequent necropsy also noted a markedly reduced liver mass and suggested that this likely contributed to the animal's demise. Importantly, the cachectic changes present in both studies are similar to those in a blue shark with a retained fishing hook (Borucinska et al., 2001), and in humans and other wild/domesticated animals (Freeman, 2018).

| Enterolithiasis
The CT scan of the HSE in the stranded, female C. taurus showed primary, concentric lamellae (concretions) around the hook nidus and these provided an obvious history of growth, but the period of time over which the enterolith was deposited could not be estimated. The can, via biochemical processes, form the mineral components of fish otoliths (Campana, 1999) and guinea pig uroliths (Hawkins, Ruby, Drazenovich, & Westropp, 2009). Interestingly, when sharks have been experimentally fasted, they continue to drink small quantities of seawater (Anderson, Taylor, Good, Hazon, & Grosell, 2007;Wood, Kajimura, Bucking, & Walsh, 2009)  collected over 13 months (2006)(2007) from waters off the New South Wales mid-north coast in the shark's last perceived habitat consistently produced magnesium/calcium ratios exceeding 4.2 . Additionally, an analysis of the gastrointestinal handling of water and solutes has shown that the bamboo shark, Chiloscyllium plagiosum, also maintains a magnesium/calcium ratio exceeding 10.0 in the chime of its intestine (Anderson et al., 2010).
The absence of food items, GI tract stasis and the occurrence of monohydrocalcite in the HSE suggests a history of seawater drinking and anorexia in the stranded, female C. taurus. Moreover, the conspicuous absence of the mineral hydroxyapatite from the HSE in the stranded, female C. taurus can be argued as further evidence of anorexia in this shark.
Precipitation of struvite in gastrointestinal and urinary calculi is associated with an alkaline pH, the presence of phosphate ions, a super-saturation of magnesium ions and the presence of ammonium ions that are often generated via bacterial hydrolysis of urea (Hassel et al., 2001;Rouff, Lager, Arrue, & Jaynes, 2018). Struvite is commonly found in enteroliths of equids (e.g. Hassel et al., 2001) but has also been found in marine mammals (Burdett & Osborne, 2010) and in the uroliths of various animals (e.g. Domingo-Neumann, Ruby, Ling, Schiffman, & Johnson, 1996;Houston, Moore, Favrin, & Hoff, 2004). Canine and feline uroliths are often associated with infections caused by urease positive bacteria with the precipitation of struvite occurring together with apatite (Le Geros & Le Geros, 1984).
Moreover, the predominant mineral deposited is ultimately determined by the urinary magnesium/calcium ratio with struvite favoured by greater magnesium concentrations (Buffington, 1994;Le Geros & Le Geros, 1984).
With this in mind, it was not surprising to find struvite present in the HSE from the stranded, female C. taurus given that sharks support a range of urease positive bacterial genera (e.g. Vibrio, Pseudomonas, Aeromonas) in their tissues including the GI tract, liver and kidneys, (Grimes, Brayton, Colwell, & Gruber, 1985;Knight, Grimes, & Colwell, 1988;Smith, 1992). Additionally, phosphate ions have been demonstrated to increase in concentration with fasting in the dogfish, Squalus acanthias, (Martini, 1978

| Hook migration
There are at least two possible hook migration routes from the cardiac stomach to the intestine. First, the hook could have passed through the pyloric stomach and then entered the intestine without becoming embedded in the mucosa. Second and more likely, the hook perforated the cardiac stomach wall and this was followed soon after by perforation of the intestinal wall. This parsimonious sequence of events would have been facilitated by the initial (normal) apposition of the cardiac stomach and intestine. The gut contents of healthy C. taurus provided further evidence with many individuals having fish with embedded hooks attached to fishing line (100-200 mm long) in the cardiac stomach lumen. As sharks use the lateral line and electroreception to locate and capture prey (Kalmijn, 1971;Maruska, 2001), it is very likely that the fish were consumed fol-  (Borucinska et al., 2001(Borucinska et al., , 2002(Borucinska et al., , 2003.

| SUMMARY AND CON CLUS I ON
The To this end, the aetiology appeared similar to previously published accounts of enterolithiasis and urolithiasis in C. taurus. Irrespective of this, however, retention of the hook in the GI tract ultimately led to the shark's mortality, an outcome that was also consistent with previous studies.

ACK N OWLED G EM ENTS
The authors are grateful to B. Louden and J. Gilligan for recovering the shark's carcass and their assistance with the necropsy. R. Field is thanked for enterolith sample preparation and the X-ray diffractometry measurements. The staff members from Macquarie Medical Imaging assisted with the CT imaging and their expertise enabled enhanced interpretation. We thank the reviewers for providing constructive comments on the manuscript. This paper is dedicated to the memory of Russell Field.

CO N FLI C T S O F I NTE R E S T
The authors declare that they have no conflicts of interest.

E TH I C A L S TATEM ENT
The study was carried out under an Animal Research Authority The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines page, have been adhered to.

PE E R R E V I E W
The peer review history for this article is available at https://publo ns.com/publo n/10.1002/vms3.333.