Aspects of the pathology induced by Spinitectus petterae Boomker, 1993 in the stomach of Clarias gariepinus (Burchell, 1822) using light and scanning electron microscopy

Abstract Spinitectus spp. (Rhabdochonidae) are enteric nematodes characterized by annular spines. At the anterior end, these spines assist attachment and aid penetration into the host tissue. During parasitological surveys of the Vaal River system from three localities, below the Vaal River Barrage in the Vaal Dam surrounding UJ island and below the Grootdraai Dam, Spinitectus specimens were collected from the stomach lining of the sharptooth catfish, Clarias gariepinus (Burchell, 1822). Histopathological effects induced by Spinitectus petterae Boomker, 1993 on C. gariepinus has not been studied. Thus, the aim of this study was to describe the histopathology induced by S. petterae. For light microscopic examination, tissue samples with attached S. petterae were sectioned and stained with haematoxylin and eosin. Additionally, attached nematodes were also studied using scanning electron microscopy. Leukocytes were counted with the Disector principle. Standard infection parameters (prevalence, mean intensity and abundance) were calculated and compared to host parameters. Prevalence varied greatly (11.77% to 100%) between localities. Histopathology induced by S. petterae to C. gariepinus stomach (cardiac region) consisted of significant leukocyte infiltration, acute ulcerations and chronic granuloma formation. This was similar to the pathology of other Spinitectus occurring in host anterior intestine and stomach, but granuloma formation had not been previously reported and this suggests chronic infection in wild caught fish.


| Sample collection
Fish were collected with the use of gill nets and rod and reel. Live fish were kept in shaded, and aerated live-wells until death. Lengths (cm) and weight (g) of each fish were recorded and thereafter, they were killed by severing the spinal cord according to the standard protocols set by the South African National Animal Ethics Guidelines.
The intestine of each fish, from the oesophagus to the rectum, was removed, transferred to 0.9% saline, opened with two #5 forceps and assessed on site with brightfield and darkfield illumination for the presence of nematodes and other intestinal parasites using, a Zeiss DV4 Stereomicroscope (Carl Zeiss). Infections were recorded and tissue from the cardiac and pyloric regions with attached nematodes were removed and fixed in 10% neutral buffered formalin (NBF).
For scanning electron microscopy (SEM), ten tissue samples from four different fish (4 from the Vaal Dam and 6 from below the Grootdraai Dam) were washed under a gentle stream of tap water for 24 h, dehydrated in a stepwise increasing concentrations (30%-100%) of ethanol followed by increasing concentrations of hexamethyldisilazane (Merck) in absolute ethanol (30%-100%) (Nation, 1983). They were then transferred to a Sanpla Dry Keeper desiccator cabinet (Kita-Ku, Osaka, Japan) for at least a week until completely dry. Samples were sputter coated with gold using an Emscope SC500 sputter coater (Quorum Technologies) and photomicrographs obtained using a VEGA 3 LMH scanning electron microscope (Tescan), set at 3-6 kV.
For histopathological examination, 7 NBF fixed tissue samples from 4 different fish (3 from Vaal Dam and 4 from below the Grootdraai Dam) were prepared by rinsing under a gentle stream of tap water for 24 h. Samples were dehydrated in a gradual series of ethanol to 70% ethanol in water, then further dehydrated in a gradual series of acetone (from 70% to 100%) in water. Samples were infiltrated with resin (TAAB ® Laboratories Ltd) under vacuum and cured at 60°C. Blocks were sectioned using a manual rotary microtome (Anglia Scientific) at 5-7 µm with a glass knife. The resin was removed with saturated NaOH in acetone (King, 1983), stained with haematoxylin and eosin and mounted with Entellan ® (Merck).
Photomicrographs were obtained with the aid of a Zeiss Axioplan 2 Imaging Light Microscope (Carl Zeiss) with Axiovision 4.7.2 software (Carl Zeiss). To characterize inflammation, leukocytes were counted according to the Disector principle (Kaplan et al., 2012), at the gastric gland and lamina muscularis mucosa regions. Leukocyte counts were taken from close to the infection site and at a 1150µm distance from infection site.
Abundance, mean intensity and prevalence were calculated as set out by Bush et al. (1997).

| Light and scanning electron microscopy
in the pyloric region appeared loosely attached to the mucosa. When samples from this region were collected, nematodes were dislodged and the attachment sites could not be macroscopically identified and were, therefore, not assessed. However, specimens in the cardiac region were deeply attached and remained so during processing.
Specimens were covered by mucus ( Figure 2b); however, excessive mucus was not seen, and no distinct colour change was observed.
After mucus removal (Figure 2c), nematodes were observed to be at- More leukocytes were recorded in infected (at infection site) than in uninfected tissue (far from infected tissue), while neutrophils and monocytes displayed the most significant increase at the infiltration site. This trend was followed by lymphocytes and eosinophils. Basophil counts increased insignificantly at 90% confidence levels in infected tissue. As such, all leukocyte counts differed significantly between uninfected and infected tissue, except for basophils (Table 2).

| Leukocyte counts
There was a statistically significant difference between the number of eosinophils, neutrophils, lymphocytes and monocytes (Table 2) between the uninfected and infected tissue. No significant difference was observed for basophil counts (p-value .17).

| DISCUSS ION
Spinitectus induced pathology varies greatly depending on the host and parasite species, with S. micracanthus, not impacting L. macrochirus notably (Keppner, 1975), while S. carolini in L. cyanellus caused damage to the mucosa and submucosa and caused the formation of ulcers alongside yellow mucus (Meguid & Eure, 1996).
Spinitectus petterae occurred in the stomach, attached to the cardiac and pyloric regions, similar to S. jamundensis in P. lineatus (Ramallo et al., 2000). However, S. jamundensis penetrated the muscular layer in the cardiac region, while S. petterae only penetrated to the sub- TA B L E 2 Analysis of leukocyte numbers in the lamina muscularis mucosa, comparing normal tissue and tissue infected by Spinitectus petterae observations by Ramallo et al. (2000) who reported only small numbers of larval Spinitectus sp. attached to the pyloric region of the stomach of P. lineatus. Jilek & Crites (1982a) and Keppner (1975) reported that S. carolini and S. micracanthus larvae migrate, but these do not cause severe pathology. Other species occurred in the anterior intestine, pyloric cecae and rectum of their hosts, with the 3rd stage larvae occasionally encysting in the mesenteries (Jilek & Crites, 1982a, 1982b, none of which was observed in the present study. In the present study, both acute and chronic pathology was reported, localized to the attachment site with the surrounding areas not affected. Ulceration, with no infiltration of leukocytes or fibrosis at the ulceration site, was interpreted as acute pathology caused by S. petterae. It is speculated that S. petterae tunnels through the columnar epithelium, lamina propria and gastric glands. These tissues are mechanically disrupted by the movements and insertion of annular spines into the tissue. The ulcer, which presents as an elongated tunnel, is similar to the 'migration tracts' observed by Jilek & Crites (1982a) for S. carolini.
Ulcers that occurred adjacent to the granulomas and associated with embedded S. petterae, were most likely produced by migrating or not fully attached specimens. Erythrocyte and leukocyte infiltration was absent in the area surrounding ulcerations and could indicate an early-stage infection (Goodman & Fuller, 2009 (Gaugler & Bilgrami, 2003). Bacteria were not present between the destructed cells but were observed in the vestibule and on the cephalic structures of S. petterae, together with leukocytes. These bacteria could be from the C. gariepinus microbiome, a part of S. petterae microbiome or from the ingested material. Therefore, it is speculated that S. petterae either actively feeds on shredded granuloma tissue and/ or on bacteria.
Foreign body granuloma formation occurs due to chronic infection and gastric excretion can be reduced by the presence of a granuloma or multiple merged granulomas, influencing digestion and causing malnutrition in the host (Bjørgen et al., 2020). Furthermore, goblet cells undergo hyperplasia, to increase secretions and to form a barrier to separate and eventually eliminate the chronic nematode infection in fish, which is considered an important response to infections (Grencis et al., 2014). This elimination of nematodes is also often associated with the production of yellow mucus (Grencis et al., 2014). However, chronic infection of S. petterae in C. gariepinus displayed no goblet cell hyperplasia nor thick yellow mucus production, instead the mucus was opaque and white. Similarly, most other Spinitectus species do not display goblet cell hyperplasia or yellow mucus (Hoffman, 1975;Jilek & Crites, 1982a;Keppner, 1975;Ramallo et al., 2000), but S. carolini from L. cyanellus does (Meguid & Eure, 1996).
The average mean intensity of S. petterae in the present study was always above five, excluding the 216 individuals recorded from a single host. Higher parasite intensities cause more severe pathology, and infection above five was considered severe in S. carolini in L. cyanellus (Meguid & Eure, 1996). Similar to other nematodes, a possible aggregation was observed. Infection parameters may also be influenced by seasonality, host sex, host size, sample size and water quality.
In conclusion, the pathology caused by S. petterae to the cardiac stomach region of C. gariepinus is similar to that of S. carolini, described from the anterior intestine of L. macrochirus by Jilek and TA B L E 3 Summary of the collection data for Spinitectus petterae from the Summer of 2018 to the Summer of 2019 from three localities