- Email: [email protected]
Impact of poxvirus lesions on saltwater crocodile (Crocodylus porosus) skins
Veterinary Microbiology 211 (2017) 29–35
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Impact of poxvirus lesions on saltwater crocodile (Crocodylus porosus) skins a
Rhiannon L. Moore , Sally R. Isberg a b c
a,b
c
, Cathy M. Shilton , Natalie L. Milic
a,⁎
MARK
School of Psychology and Clinical Sciences Charles Darwin University, NT 0909, Australia Centre for Crocodile Research, PO Box 329, Noonamah, NT 0837, Australia Berrimah Veterinary Laboratory, Berrimah Farm, Northern Territory, Australia
A R T I C L E I N F O
A B S T R A C T
Keywords: Saltwater crocodile Crocodylidopoxvirus PCR Histology Skin lesions
Cutaneous poxvirus infections are common in several crocodilian species and are of importance in crocodile farming due to their potential impact on the tanned hide. To confirm poxvirus infection and understand the impact on saltwater crocodile (Crocodylus porosus) skin, fourteen animals from different age groups (five hatchlings, five yearlings and four grow-outs) were selected based on a criterion of ten poxvirus-like lesions per animal. One lesion on each animal was extruded for genetic analysis and transmission electron microscopy. Both methods confirmed poxvirus so the remainder of lesions were re-examined every six weeks over a 24 week study period. Each lesion went through four distinct phases: early active, active, expulsion and healing. To understand how these lesions impact on the final skin product, one crocodile from each age group was euthanised and the lesions examined. Using standard skin grading techniques (light-table), the early phase (early active – expulsion) lesions were all translucent and would lead to downgrading of the skin or, at worst, rendering them unsaleable. At the later stages of healing, the translucency reduces. Histological examination of the phases confirm that the basement membrane is not breached by the infection further indicating that poxvirus lesions, given enough time, will eventually have no detrimental effect on skin quality. This is obviously dependent upon no more lesions developing in the interim.
1. Introduction Saltwater crocodiles (Crocodylus porosus) are farmed to produce skins for the international leather industry. However, only skins without defects on the abdominal region are ideally suitable for export, with the remainder being downgraded resulting in a loss of profit for the producer. One reported defect on crocodilian belly skins is caused by poxvirus (Jacobson et al., 1979; Horner, 1988; Pandey et al., 1990; Huchzermeyer et al., 1991, 2009; Penrith et al., 1991; Buenviaje et al., 1992, 1998; Buoro, 1992; Ramos et al., 2002). Most of these studies have confirmed poxvirus using histology and/or transmission electron microscopy but none define how these lesions progress or if they completely heal without commercial implications. Huchzermeyer et al. (2009) is the only study to propose a connection between poxviral lesions and deep lesions that remained as “pitholes” (large depressed foci) on the tanned skin of Nile crocodiles (Crocodylus niloticus). On the live animal, they reported the lesions were cysts infiltrated with intracytoplasmic inclusions, which formed ulcers before healing. Characteristically, these were very different to poxviral lesions previously described. Genetic sequencing of a crocodylidpoxvirus-specific protein of unknown function (ORF19; Afonso et al.,
⁎
2006) revealed seven “atypical” nucleotide substitutions that translated into five amino acid changes. Based on this information, the authors described the virus as an “atypical crocodile poxvirus” (Huchzermeyer et al., 2009). In the absence of similar information for the saltwater crocodile, this study aimed to confirm the presence of poxvirus in the target lesions on saltwater crocodile skins using genetics and transmission electron microscopy, and then define phases of lesion healing. Finally, we aimed to understand the quality implications of poxvirus lesions on the belly skins of saltwater crocodiles, determine if they could indeed heal, and recommend further strategies for industry. 2. Materials and methods 2.1. Animals The research was conducted under approval from Charles Darwin University Animal Ethics Committee (A13017). Five hatchlings (< one year old), five yearlings (one to two years old) and four grow-outs (> two years) from Darwin Crocodile Farm, Northern Territory, Australia were selected based on the criterion of having at least 10
Corresponding author at: School of Psychological and Clinical Sciences, Charles Darwin University, Brinkin, Northern Territory 0909, Australia. E-mail address: [email protected] (N.L. Milic).
http://dx.doi.org/10.1016/j.vetmic.2017.09.019 Received 8 May 2017; Received in revised form 25 September 2017; Accepted 25 September 2017 0378-1135/ Crown Copyright © 2017 Published by Elsevier B.V. All rights reserved.
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
scales to the right or left of the midline. Each lesion was characterised including shape, contour, colour and keratin normality. A digital photograph was taken with a linear scale for digital planimetry using Pictzar (Version 8.0, www.pictzar.com). Lesions were re-documented four times over the 24 week study period. All animals continued to grow both in length and weight throughout the study period and appeared healthy when compared to their pen conspecifics that were not included in the study.
poxvirus-like lesions on their belly region. Throughout the duration of the study the animals, were housed in accordance with the Code of Practice on the Humane Treatment of Wild and Farmed Australian Crocodiles (NRMMC, 2009) in facilities previously described by Miles et al. (2010). 2.2. Genetic confirmation of poxvirus One suspected poxvirus lesion from each animal (14 in total) was selected for sequencing. The pox-like lesion and virus was collected in a sterile 1.5 mL microcentrifuge tube with 200 μL sterile PBS. The DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Australia) as per the manufacturer’s instructions for viral DNA. Two previously published open reading frame (ORF) fragments were identified for amplification based on the complete genome of Nile crocodilepox virus (Accession number NC_008030; Afonso et al., 2006): a crocodile specific poxvirus protein of unknown function, ORF19 (Huchzermeyer et al., 2009) and a RNA polymerase subunit conserved in high GC poxviruses, ORF99 (Li et al., 2010). Published ORF99 primers (forward CATCCCCAAGGAGACCAACGAG, reverse TCCTCGTCGCCGTCGAAGTC; Li et al., 2010) were used to amplify a 548 bp fragment. New primers against ORF19 were designed using primerBLAST (forward AAATGTACGCGCAGGGATCT, reverse GAGACGCTGTACCTTCGCTT, 159 bp; Ye et al., 2012) due to difficulties in replicating the OFR19 PCR reaction detailed by Huchzermeyer et al. (2009). PCR was performed as 25 μL reactions using a Q5 High-Fidelity 2X Master Mix kit (New England BioLabs, USA) following the manufacturer’s instructions. Briefly, 5 μL of DNA was added to 12.5 μL of Q5 High-Fidelity 2X Master Mix, 1.25 μL (20 pmol) forward and reverse primer and 5 μL of double distilled water. PCR was performed using the following cycling parameters; initial denaturation at 98 °C for 2 mins, followed by 30 cycles at 98 °C for 10 s, primer-specific annealing temperature for 20 s (ORF19 and ORF99 were 65 °C and 71 °C, respectively), and 72 °C for 30 s, with a final extension of 72 °C for 2 mins. PCR products were gel extracted and purified (QIAquick Gel Extraction Kit; Qiagen, Victoria, Australia) before sequencing on an ABI3730XL Sanger sequencer at Macrogen (Seoul, Korea). Sequences were aligned using the alignment tool in BioEdit (Version 7.2.5) and consensus sequences screened against GenBank for similarities with other poxviruses using BLAST (http://blast.ncbi.nlm.nih. gov/). The resultant sequences for ORF19 and ORF99 were submitted to GenBank, National Center for Biotechnology Information (http:// www.ncbi.nlm.nih.gov/) and were assigned the accession numbers KY457244 and KY457245, respectively. The neighbour-joining phylogenetic tree was constructed along with other high GC chordopoxviruses noted within Li et al. (2010) using MEGA7 (Kumar et al., 2015) software and tree reliability was tested using 1000 bootstrap replicates.
2.5. Necropsy and histological examination At the end of the trial period, one animal from each age category was selected for necropsy using the above mentioned criteria; at least 10 poxvirus-like lesions on their belly region. The animals were euthanised with 0.5 mL barbiturate/kg diluted with 1:2 sterile saline injected into the occipital venous sinus. A full necropsy was conducted including the macroscopic examination of all internal organs, organ collection for histology (spleen, liver, lung, thymus, pancreas, pharyngeal mucosa, tongue, thyroid gland, intestine, brain and eye) and bacteriology (liver and kidney) using methods previously described in Shilton et al. (2014). The belly skins were examined on a light table (Manolis and Webb, 2011) by observing the translucency associated with the lesion. A series of poxvirus lesions at different healing stages were removed, fixed in 10% buffered formalin for 72 h, processed in standard fashion for histological examination, sectioned at 5 μm, and stained with haematoxylin and eosin (H & E). 3. Results The juxtaposition of this study was to prove poxvirus causation, which requires extraction of the lesions, as well as to observe if and how these lesions heal. One lesion could not meet both purposes. 3.1. Confirmation of poxvirus using molecular characterisation Using PCR, all 14 lesions (one from each crocodile) amplified both poxvirus ORF19 and ORF99, which were identical in sequence across all age groups, confirming poxvirus aetiology. ORF19 primers amplified a 159 bp product that showed 89% sequence similarity to the Nile crocodilepox virus, containing 17 nucleotide substitutions when aligned to the same region from the Nile crocodilepox virus (Afonso et al., 2006; Fig. 1A). Of the seven atypical nucleotides described by Huchzermeyer et al. (2009), our amplified ORF19 region included four of these substitutions, with two of them also present in the C. porosus poxvirus sequence (Fig. 1A). In comparison, the larger, more conserved ORF99 region (548 bp product) showed 96% similarity between the C. porosus poxvirus and Nile crocodilepox virus sequences (Fig. 1B). Phylogenetic analysis with these alignments and other high-GC chordopoxviruses, confirmed the phylogenetic results of Li et al. (2010) with the crocodile poxviruses forming a distinct clade (Fig. 2).
2.3. Transmission electron microscopy (TEM) confirmation One poxvirus-like lesion from one crocodile in each age category was selected for TEM. Briefly, lesions were fixed in 4% chilled glutaraldehyde in 0.1 M (pH 7.2) phosphate buffer for 72 h, before secondary fixing in 1% osmium tetroxide for 2 h. Samples were embedded in Spurr’s resin and ultrathin sections were cut and sections were stained with uranyl acetate and lead citrate. Sections were examined under the Jeol JEM-1200EX at 80 kV.
3.2. Confirmation of poxvirus using TEM TEM on the three representative lesions showed either enlarged epidermal cells completely filled with virions or encased in vacuoles (Fig. 3) within the plasma membrane. Virions were approximately 100 × 300 nm with the stereotypical dumbbell-shaped nucleoid visible within most of the virions suggesting virions had reached maturity.
2.4. Identification and tracking of poxvirus lesions 3.3. Poxvirus lesion development and healing Concurrently to the genetic confirmation, poxvirus-like lesions were grossly characterised. At the initial sampling, lesions were identified and the location on the belly skin was recorded for identification at subsequent samplings. The location of the scale was determined by counting the number of scales down from the collar and the number of
Initially, 152 lesions were identified with an additional 36 lesions developing during the course of the study. The repeated examination of lesions allowed for distinct categorisation of poxvirus lesions into phases: early active, active, expulsion and a healing phase before they 30
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
Fig. 1. Partial nucleotide sequence for (A) open reading frame 19 (ORF19) and (B) open reading frame (ORF99) using the Nile crocodilepox virus genome (Afonso et al., 2006; Accession number NC_008030) aligned against the atypical Nile crocodilepox virus described by Huchzermeyer et al. (2009) and the saltwater crocodile (C. porosus) poxvirus described within this study. The asterisk (*) indicates identical nucleotides between the atypical Nile crocodilepox virus sequence (Huchzermeyer et al., 2009) and the C. porosus poxvirus sequence. The atypical Nile crocodilepox virus sequence was not available for the ORF99 region for comparison.
keratinocytes containing large eosinophilic “Henderson-Patterson-like” inclusion bodies. The overlying keratin was intact, and there was compression of the underlying loose dermis (Fig. 5A). The chromatophore density under the lesion was reduced compared to normal skin rendering 100% of the active lesions clearly visible on the light table. As the lesions progressed to the active stage, there was an increase in surface area (1.03 ± 0.26 mm2) with consistent grey-white colouration (Fig. 4B). Active lesions appeared to have a distinct contour
were barely visible on the skin (Fig. 4). Of these 188 lesions, 55 lesions were examined on the light table from the necropsy animals to understand their impact on the raw skin product. The early active lesions (Fig. 4A) were grey-white foci with normal (64% lesions) to pin-point keratin damage (28% lesions) with an average surface area of 0.85 ± 0.29 mm2. During the early active stage, histology showed a well demarcated focus of marked epidermal hyperplasia and hypertrophy with overlying central plug (CP) of
Fig. 2. Neighbour-joining phylogenetic tree based on the amplified ORF 99 sequence from C. porosus, co mpared to the Nile crocodilepox virus and other high GC content poxviruses with bootstrap values given at the nodes. Genbank accession numbers are given in parentheses.
31
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
Fig. 3. Micrographs of a poxvirus-infected epidermal cell from C. porosus. (A) (5000X). (B) Virion nuclei (174 000X; black arrows) showing characteristic poxvirus dumbbell nucleoid.
Fig. 4. The progression of a typical poxvirus lesion: (A) early active lesion indicated by arrow, (B) active lesion, (C) after expulsion of necrotic plug and (D) healing with abnormal keratin. Ruler increments are 1 mm.
grossly compromised keratin whilst the remaining 28% developed on scales that appeared to have normal keratin. A full history of previous injuries was not available, but it is possible that there may have been damage to the skin which allowed the virus to enter and replicate at the site. An example of this possible pathogenesis is given in Fig. 6, where a scale showing no lesion at the initial sampling (Fig. 6A), acquires a linear abrasion, comprising the keratin (Fig. 6B). The abrasion appears to have healed by the third sampling with the exception of a presumed early active poxvirus lesion that has developed in the bottom left hand corner of the scale (Fig. 6C) and continued to form into an active lesion (Fig. 6D). By the final sampling, the lesion had healed (Fig. 6E). Interestingly, it appeared another early active poxvirus lesion started to develop below the integumentary sensory organ (ISO) by the fourth sampling, and continued to form into an active lesion by the final sampling. Lesion location indicated that 62% of the lesions occurred around the scale margins and there were more lesions on the lower (27%; n = 51) and mid (25%; n = 46) scale margins compared to the upper margins (11%; n = 20). There were eight lesions on the hinge between two scales and 11 on the ISOs with the remaining 52 lesions detected in the central scale region.
with a raised outer and central depressed core with abnormal keratin. Histology revealed that the CP continued to enlarge with keratinocytes containing large “Henderson-Patterson-like” inclusions (Fig. 5B). The keratin above the CP was disrupted, leaving an open hole which was visible macroscopically (Fig. 4B). The pigmented cells were sparse, remaining translucent on the light table. Histology showed mild perivascular dermal lymphocytic infiltration, indicating the lesions evoked a minor immune response (Fig. 5B). From the active stage, the CP of necrotic cells is released, referred to as the expulsion stage. During this stage, the lesion significantly increases in size (2.21 ± 0.29 mm2) and may be visible in one of two states; just prior to expelling the CP or after the expulsion has occurred. When the CP is about to be expelled, it overlies a crust of necrotic inflammatory cells immediately adjacent to the epidermis. The underlying epidermis returns to a relatively normal thickness and cellular morphology (Fig. 5C). After expulsion, the lesions appear grossly as a large, depressed foci with abnormal keratin (Fig. 4C). Commonly, the lesions have an orange/tan colouration. As within the previous stages, all these lesions remain clearly visible on the light table. In the healing phase, the surface area begins to decrease (1.62 ± 0.19 mm2; Fig. 4D). Healing lesions are characterised by undulating epidermal and dermal layers, and the chromatocytes remain sparse compared to the unaffected skin (Fig. 5D) resulting in a slight reduction in visibility on the light-table (87% visible). Macroscopically, the keratin of these lesions appears abnormal with normal skin colouration (Fig. 4D). From here, the lesions continue to heal, slowly becoming less visible with only 41% detected on the light table. On average, these lesions had a surface area of 0.72 mm ± 0.29 mm2. During the tracking of the lesions, an additional 36 lesions were detected on the skin and recorded during successive examinations. Twenty six of the 36 additional lesions (72%) occurred on areas with
4. Discussion Crocodylidpoxvirus, represented only by Nile crocodilepox virus, is a new genus of Chorodopoxvirinae (Afonso et al., 2006; ICTV, 2015). For the first time, the suspected poxvirus lesions on juvenile C. porosus belly skin were confirmed as poxvirus by PCR amplification of two different regions (ORF19 and ORF99; Fig. 1) from all 14 lesions tested. ORF19 is a protein of unknown function but specific to crocodylidpoxviruses (Afonso et al., 2006). However, the inherent variability of this region 32
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
Fig. 5. Histology of poxvirus lesion progression: (A) early Active lesion, (B) active lesion, (C) erupting lesion and (D) healing lesion. Where H is the hinge (inter-scalar) region; CP the central plug of necrotic cells; E the epidermis; De the dermis; arrows show the lymphocytes surrounding a dermal blood vessel; arrow heads indicate the chromatophores.
Horner (1988); 100 × 200 nm) and Buenviaje et al. (1992); 110 × 220 nm). This study is the first to identify and describe a clear progression of poxvirus lesions on the belly skin of the saltwater crocodile. We termed these stages as ‘early active’, ‘active’, ‘expulsion’, and ‘healing’ phases. In general, the lesions within this study conformed to the gross description of poxvirus lesions within Brown caiman (Caiman crocodilus fuscus) (Jacobson et al., 1979), Caiman yacare (Caiman crocodilus yacare) (Ramos et al., 2002), saltwater crocodile and freshwater crocodile (C. johnstoni) (Buenviaje et al., 1998; Ariel, 2011; Shilton et al., 2014). That is, approximately 1–3 mm in diameter foci and mainly grey to white in colour and were most likely descriptions of ‘early active’ to ‘active’ lesions as per our defined healing phases. The ulcerative lesion that breached the deep dermis noted by Buenviaje et al. (1998) was not confirmed in this study even at the ‘expulsion’ stage. In fact, none of the
shown in Huchzermeyer et al. (2009) and this study suggests this may not be a useful region to distinguish poxvirus variants. On the other hand, ORF99 is a highly conserved RNA polymerase subunit (Afonso et al., 2006) and phylogenetic analysis confirmed the C. porosus poxvirus to be closely related to the Nile crocodilepox virus (Fig. 2) and distinctly divergent from other high GC chordopoxviruses. In fact, Li et al. (2010) reported that Nile crocodilepox virus clustered with Molluscum contagiosum virus, but the addition of the C. porosus poxvirus from this study shows crocodylidopoxiruses to be distinctly divergent from other poxviruses. Genome sequencing of the saltwater crocodile isolate will further elucidate if it is indeed a new member of the Crocodylidopoxvirus genus. For completeness, TEM similarly confirmed the presence of stereotypical dumbbell-shaped nuclei from mature virions. The virions were 100 × 300 nm, which were slightly longer than previously reported by
Fig. 6. Poxvirus pathogenesis on an apparently normal belly scale (A). A linear lesion was incurred breaching the protective beta-keratin layer (B) and developed into an early active poxvirus lesion (C). This early active lesion further develops into an active lesion, with a second early active lesion developing below the ISO (arrowhead) (D). The original lesion (E) forms a depressed foci, with the second lesion developing into an active lesion. Arrow highlights the region of interest. Ruler increments are 1 mm. Photos are taken six weeks apart.
33
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
epidemiological studies should also explore possible environmental, vector (e.g. mosquitoes) or husbandry stressors that may dictate poxviral lesion incidence and severity to find more passive methods of control.
lesions histologically observed in this study showed any breach of the basement membrane with all viral bodies confined to the epidermal layers. In general, the Nile crocodile lesions described by Huchzermeyer et al. (1991) were consistent with those described herein. In contrast, the raised nodules (6–8 mm) described by Pandey et al. (1990) and Horner (1988) were not observed in this study, nor was the brown colouration seen by Huchzermeyer et al. (1991) although the yellow colouration noted by Pandey et al. (1990) could be associated to the orange colour we observed. However, none of our histological sections showed the cyst structures described by Huchzermeyer et al. (2009) which would definitively be observed on the tanned skin as they report. The concern for crocodile producers is the impact these lesions have on the final crocodile skin. Our results suggest the stage of the poxvirus lesion at the time of harvest determines whether the lesion will be seen on the skin post-harvest. Harvesting animals with lesions in the early active to expulsion stages has 100% probability of being visible on the light table and hence on the tanned skin. The probability of seeing lesions on the skin decreases during the healing stage (87%), but it is not until the lesions are barely visible on the live animal that the risk of visibility post-harvest begins to diminish (41%). It is not known what the average time for a lesion to completely heal would be, as not all lesions were observed at all stages. Crocodiles from a range of age categories were used in this study (hatchlings, yearlings and grow-outs), being selected due to their high incidence of poxvirus lesions, allowing classification of the different lesion stages. Unfortunately, the largest animals used in this study (grow-outs) were still substantially smaller than the average harvestsize crocodile produced on Australian crocodile farms. Personal observations and farm discussions by the authors suggest early stage poxvirus lesions are less common on harvest size crocodiles, with only the latter stages (erupted and healing phases) visible on these animals. Irrespective of whether the poxvirus lesions eventually heal and can be detected on the harvested skin, poxvirus still invokes a healing and inflammatory response, which is likely partitioning energy away from growth although not noticeably impacting the wellbeing of the crocodiles. The mid region of the scale is comprised of hard thick beta keratin that significantly increases with age from 9.8 μm thick when the crocodile is 0.3 m in total length to 89.9 μm when 1.6 m long (Manolis et al., 2000). This increase in keratin thickness with size might explain why there are fewer lesions on older animals. The hinge and ISOs regions of the scale are comprised of flexible alpha keratin (Richardson et al., 2001), suggesting these regions may be more permeable to injury and pathogens and further explains why 72% (n = 156) of the poxvirus lesions were associated with the periphery, hinge and ISO regions of the scale. Of the lesions that developed during the study period (n = 36), 72% were associated with abnormal keratin. This is consistent with the reported transmission of poxvirus through compromised keratin, allowing penetration of the poxvirus (McKeever et al., 1988; Buller and Palumbo, 1991; Birthistle and Carrington, 1997; Haig and McInnes, 2002; Tyring, 2003; Dlehon et al., 2004; Dohil et al., 2006). It is likely that poxvirus transmission within the saltwater crocodile and other crocodilians could occur as a result of their concentrated environments. This suggests that increasing the hygiene of pens may help control the spread of infection, along with ensuring no mechanical damage occurs to the keratin during husbandry procedures. In the remaining 28% of these lesions, there was no visible damage to the keratin, indicating either microscopic compromises, an alternate transmission route or systemic dissemination of the virus. Previously, latency has also been suggested in relation to crocodile poxvirus (Horner, 1988; Penrith et al., 1991). Horner (1988) reported the utility of using an autogenous vaccine to reduce the longevity of poxviral lesions on Nile crocodiles. To our knowledge, no further work has been done in this area particularly to show if early age vaccination would prevent lesion incursion. Further
Conflict of interest statement This research was conducted with funding from Charles Darwin University and no conflicts of interest are declared. Acknowledgements The authors would like to acknowledge Charles Darwin University and Berrimah Veterinary Laboratory staff, particularly Suresh Benedict for conducting the bacteriology and Sue Aumann for conducting histology on the internal organs. In addition, thank you to Darwin Crocodile Farm staff for access to the crocodiles used in this study. References Afonso, C.L., Tulman, E.R., Delhon, G., Lu, Z., Viljoen, G.J., Wallace, D.B., Kutish, G.F., Rock, D.L., 2006. Genome crocodilepox virus. J. Virol. 80, 4978–4991. Ariel, E., 2011. Viruses in reptiles. Vet. Res. 42, 100. http://dx.doi.org/10.1186/12979716-42-100. Birthistle, K., Carrington, D., 1997. Molluscum contagiosum virus. J. Infect. 34, 21–28. Buenviaje, G.N., Ladds, P.W., Melville, L., 1992. Poxvirus infection in two crocodiles. Aus. Vet. J. 69, 15–16. Buenviaje, G.N., Ladds, P.W., Martin, Y., 1998. Pathology of skin diseases in crocodiles. Aus. Vet. J. 76, 357–363. Buller, R.M., Palumbo, G.J., 1991. Poxvirus pathogenesis. Microbiol. Rev. 55, 80–122. Buoro, I.B.J., 1992. Pox-like virus particles in skin lesions of five Nile crocodiles in Kenya. Discov. Innov. 4, 117–118. Dlehon, G., Tulman, C.L., Afonso, C.L., Lu, Z., de la Concha-Bermejillo, A., Lehmkuhl, H.D., Piccone, M.E., Kutish, G.F., Rock, D.L., 2004. Genomes of the parapoxviruses orf virus and bovine popular stomatitis virus. J. Virol. 78, 168–177. Dohil, M.A., Lin, P., Lee, J., Lucky, A.W., Paller, A.S., Eichenfield, L.F., 2006. The epidemiology of molluscum contagiosum in children. J. Am. Acad. Dermatol. 54, 47–54. Haig, D.M., McInnes, C.J., 2002. Immunity and counter-immunity during infection with the parapoxvirus orf virus. Virus Res. 88, 3–16. Horner, R.F., 1988. Poxvirus in farmed Nile crocodiles. Vet. Rec. 122, 459–462. Huchzermeyer, F.W., Huchzermeyer, K.D.A., Putterill, J.F., 1991. Observations on a field outbreak of pox virus infection in young Nile crocodiles (crocodylus niloticus). J. S. Afr. Vet. Assoc. 62, 27–29. Huchzermeyer, F.W., Wallace, D.B., Putterill, J.F., Gerdes, G.H., 2009. Identification and partial sequencing of a crocodile poxvirus associated with deeply penetrating skin lesions in farmed Nile crocodiles, Crocodylus niloticus. Ondersepoort J. Vet. Res. 76, 311–316. ICTV, 2015. International Committee on Taxonomy of Viruses. (June 15 2015). http:// talk.ictvonline.org/Accessed. Jacobson, E.R., Popp, J.A., Shields, R.P., Gaskin, J.M., 1979. Poxlike skin lesions in captive caimans. J. Am. Vet. Med. Assoc. 175, 937–940. Kumar, S., Stecher, G., Tamura, K., 2015. MEGA7: molecular evolutionary genetics analysis version 7.0. Mol. Bio. Evol. 33, 1870–1874. Li, Y., Meyer, H., Zhao, H., Damon, I.K., 2010. GC Content-based pan-pox universal PCR assays for poxvirus detection. J. Clin. Microbiol. 48, 268–276. Manolis, S.C., Webb, G.J.W., 2011. Tracking Crocodile Skin Defects − from Farm to Product. Rural Industries Research and Development Corporation, Canberra (Publication No. 11/012). Manolis, S.C., Webb, G., Richardson, K., 2000. Improving the Quality of Australian Crocodile Skins. Rural Industries Research and Development Corporation, Canberra (Publication No. 00/21). McKeever, D.J., Jenkinson, D.M., Hutchinson, G., Reid, H.W., 1988. Studies on the pathogenesis of orf virus infection in sheep. J. Comp. Pathol. 99, 317–328. Miles, L., Moran, C., Isberg, S., 2010. Linkage Mapping and Quantitative Trait Loci (QTL) Analysis in Saltwater Crocodiles. Rural Industries Development Corporation, Canberra (Publication No. 10/152). Natural Resource Management of Ministerial Council, 2009. Code of Practice on the Humane Treatment of Wild and Farmed Australian Crocodiles. Australian Government, Canberra. Pandey, G.S., Inque, I., Ohshima, K., Okada, K., Chihaya, Y., Fujimoto, Y., 1990. Poxvirus infection in Nile crocodiles (Crocodylus niloticus). Res. Vet. Sci. 49, 171–176. Penrith, M.L., Nesbit, J.W., Huchzermeyer, F.W., 1991. Pox virus infection in captive juvenile caimans (Caiman crocodilus fucus) in South Africa. J. S. Afr. Vet. Assoc. 62, 137–139. Ramos, M.C.C., Coutinho, S.D., Matushima, E.R., Sinhorini, I.L., 2002. Poxvirus dermatitis outbreak in farmed Brazilian caimans (Caiman crocodilus yacare). Aus.Vet. J. 80, 371–372. Richardson, K.C., Park, Y.J., Webb, G.J.W., Manolis, S.C., 2001. Skin histology of embryonic and hatchling Estuarine and Australian Freshwater Crocodiles. Crocodilian: Biology and Evolution. Surrey Beatty and Sons Pty Ltd., Chipping Norton, pp.
34
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
treatment. Am. J. Obstet. Gynecol. 189, S12–S16. Ye, J., Coulouris, G., Zaretskaya, I., Cutcutache, I., Rozen, S., Madden, T., 2012. PrimerBLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinf. 13, 134.
188–196. Shilton, C., Brown, G.P., Chambers, L., Benedict, S., David, S., Aumann, S., Isberg, S.R., 2014. Pathology of runting in farmed saltwater crocodiles (Crocodylus porosus) in Australia. Vet. Path. 1–13. Tyring, S.K., 2003. Mollusc um contagiosum: the importance of early diagnosis and
35
Contents lists available at ScienceDirect
Veterinary Microbiology journal homepage: www.elsevier.com/locate/vetmic
Impact of poxvirus lesions on saltwater crocodile (Crocodylus porosus) skins a
Rhiannon L. Moore , Sally R. Isberg a b c
a,b
c
, Cathy M. Shilton , Natalie L. Milic
a,⁎
MARK
School of Psychology and Clinical Sciences Charles Darwin University, NT 0909, Australia Centre for Crocodile Research, PO Box 329, Noonamah, NT 0837, Australia Berrimah Veterinary Laboratory, Berrimah Farm, Northern Territory, Australia
A R T I C L E I N F O
A B S T R A C T
Keywords: Saltwater crocodile Crocodylidopoxvirus PCR Histology Skin lesions
Cutaneous poxvirus infections are common in several crocodilian species and are of importance in crocodile farming due to their potential impact on the tanned hide. To confirm poxvirus infection and understand the impact on saltwater crocodile (Crocodylus porosus) skin, fourteen animals from different age groups (five hatchlings, five yearlings and four grow-outs) were selected based on a criterion of ten poxvirus-like lesions per animal. One lesion on each animal was extruded for genetic analysis and transmission electron microscopy. Both methods confirmed poxvirus so the remainder of lesions were re-examined every six weeks over a 24 week study period. Each lesion went through four distinct phases: early active, active, expulsion and healing. To understand how these lesions impact on the final skin product, one crocodile from each age group was euthanised and the lesions examined. Using standard skin grading techniques (light-table), the early phase (early active – expulsion) lesions were all translucent and would lead to downgrading of the skin or, at worst, rendering them unsaleable. At the later stages of healing, the translucency reduces. Histological examination of the phases confirm that the basement membrane is not breached by the infection further indicating that poxvirus lesions, given enough time, will eventually have no detrimental effect on skin quality. This is obviously dependent upon no more lesions developing in the interim.
1. Introduction Saltwater crocodiles (Crocodylus porosus) are farmed to produce skins for the international leather industry. However, only skins without defects on the abdominal region are ideally suitable for export, with the remainder being downgraded resulting in a loss of profit for the producer. One reported defect on crocodilian belly skins is caused by poxvirus (Jacobson et al., 1979; Horner, 1988; Pandey et al., 1990; Huchzermeyer et al., 1991, 2009; Penrith et al., 1991; Buenviaje et al., 1992, 1998; Buoro, 1992; Ramos et al., 2002). Most of these studies have confirmed poxvirus using histology and/or transmission electron microscopy but none define how these lesions progress or if they completely heal without commercial implications. Huchzermeyer et al. (2009) is the only study to propose a connection between poxviral lesions and deep lesions that remained as “pitholes” (large depressed foci) on the tanned skin of Nile crocodiles (Crocodylus niloticus). On the live animal, they reported the lesions were cysts infiltrated with intracytoplasmic inclusions, which formed ulcers before healing. Characteristically, these were very different to poxviral lesions previously described. Genetic sequencing of a crocodylidpoxvirus-specific protein of unknown function (ORF19; Afonso et al.,
⁎
2006) revealed seven “atypical” nucleotide substitutions that translated into five amino acid changes. Based on this information, the authors described the virus as an “atypical crocodile poxvirus” (Huchzermeyer et al., 2009). In the absence of similar information for the saltwater crocodile, this study aimed to confirm the presence of poxvirus in the target lesions on saltwater crocodile skins using genetics and transmission electron microscopy, and then define phases of lesion healing. Finally, we aimed to understand the quality implications of poxvirus lesions on the belly skins of saltwater crocodiles, determine if they could indeed heal, and recommend further strategies for industry. 2. Materials and methods 2.1. Animals The research was conducted under approval from Charles Darwin University Animal Ethics Committee (A13017). Five hatchlings (< one year old), five yearlings (one to two years old) and four grow-outs (> two years) from Darwin Crocodile Farm, Northern Territory, Australia were selected based on the criterion of having at least 10
Corresponding author at: School of Psychological and Clinical Sciences, Charles Darwin University, Brinkin, Northern Territory 0909, Australia. E-mail address: [email protected] (N.L. Milic).
http://dx.doi.org/10.1016/j.vetmic.2017.09.019 Received 8 May 2017; Received in revised form 25 September 2017; Accepted 25 September 2017 0378-1135/ Crown Copyright © 2017 Published by Elsevier B.V. All rights reserved.
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
scales to the right or left of the midline. Each lesion was characterised including shape, contour, colour and keratin normality. A digital photograph was taken with a linear scale for digital planimetry using Pictzar (Version 8.0, www.pictzar.com). Lesions were re-documented four times over the 24 week study period. All animals continued to grow both in length and weight throughout the study period and appeared healthy when compared to their pen conspecifics that were not included in the study.
poxvirus-like lesions on their belly region. Throughout the duration of the study the animals, were housed in accordance with the Code of Practice on the Humane Treatment of Wild and Farmed Australian Crocodiles (NRMMC, 2009) in facilities previously described by Miles et al. (2010). 2.2. Genetic confirmation of poxvirus One suspected poxvirus lesion from each animal (14 in total) was selected for sequencing. The pox-like lesion and virus was collected in a sterile 1.5 mL microcentrifuge tube with 200 μL sterile PBS. The DNA was extracted using the QIAamp DNA Mini Kit (Qiagen, Australia) as per the manufacturer’s instructions for viral DNA. Two previously published open reading frame (ORF) fragments were identified for amplification based on the complete genome of Nile crocodilepox virus (Accession number NC_008030; Afonso et al., 2006): a crocodile specific poxvirus protein of unknown function, ORF19 (Huchzermeyer et al., 2009) and a RNA polymerase subunit conserved in high GC poxviruses, ORF99 (Li et al., 2010). Published ORF99 primers (forward CATCCCCAAGGAGACCAACGAG, reverse TCCTCGTCGCCGTCGAAGTC; Li et al., 2010) were used to amplify a 548 bp fragment. New primers against ORF19 were designed using primerBLAST (forward AAATGTACGCGCAGGGATCT, reverse GAGACGCTGTACCTTCGCTT, 159 bp; Ye et al., 2012) due to difficulties in replicating the OFR19 PCR reaction detailed by Huchzermeyer et al. (2009). PCR was performed as 25 μL reactions using a Q5 High-Fidelity 2X Master Mix kit (New England BioLabs, USA) following the manufacturer’s instructions. Briefly, 5 μL of DNA was added to 12.5 μL of Q5 High-Fidelity 2X Master Mix, 1.25 μL (20 pmol) forward and reverse primer and 5 μL of double distilled water. PCR was performed using the following cycling parameters; initial denaturation at 98 °C for 2 mins, followed by 30 cycles at 98 °C for 10 s, primer-specific annealing temperature for 20 s (ORF19 and ORF99 were 65 °C and 71 °C, respectively), and 72 °C for 30 s, with a final extension of 72 °C for 2 mins. PCR products were gel extracted and purified (QIAquick Gel Extraction Kit; Qiagen, Victoria, Australia) before sequencing on an ABI3730XL Sanger sequencer at Macrogen (Seoul, Korea). Sequences were aligned using the alignment tool in BioEdit (Version 7.2.5) and consensus sequences screened against GenBank for similarities with other poxviruses using BLAST (http://blast.ncbi.nlm.nih. gov/). The resultant sequences for ORF19 and ORF99 were submitted to GenBank, National Center for Biotechnology Information (http:// www.ncbi.nlm.nih.gov/) and were assigned the accession numbers KY457244 and KY457245, respectively. The neighbour-joining phylogenetic tree was constructed along with other high GC chordopoxviruses noted within Li et al. (2010) using MEGA7 (Kumar et al., 2015) software and tree reliability was tested using 1000 bootstrap replicates.
2.5. Necropsy and histological examination At the end of the trial period, one animal from each age category was selected for necropsy using the above mentioned criteria; at least 10 poxvirus-like lesions on their belly region. The animals were euthanised with 0.5 mL barbiturate/kg diluted with 1:2 sterile saline injected into the occipital venous sinus. A full necropsy was conducted including the macroscopic examination of all internal organs, organ collection for histology (spleen, liver, lung, thymus, pancreas, pharyngeal mucosa, tongue, thyroid gland, intestine, brain and eye) and bacteriology (liver and kidney) using methods previously described in Shilton et al. (2014). The belly skins were examined on a light table (Manolis and Webb, 2011) by observing the translucency associated with the lesion. A series of poxvirus lesions at different healing stages were removed, fixed in 10% buffered formalin for 72 h, processed in standard fashion for histological examination, sectioned at 5 μm, and stained with haematoxylin and eosin (H & E). 3. Results The juxtaposition of this study was to prove poxvirus causation, which requires extraction of the lesions, as well as to observe if and how these lesions heal. One lesion could not meet both purposes. 3.1. Confirmation of poxvirus using molecular characterisation Using PCR, all 14 lesions (one from each crocodile) amplified both poxvirus ORF19 and ORF99, which were identical in sequence across all age groups, confirming poxvirus aetiology. ORF19 primers amplified a 159 bp product that showed 89% sequence similarity to the Nile crocodilepox virus, containing 17 nucleotide substitutions when aligned to the same region from the Nile crocodilepox virus (Afonso et al., 2006; Fig. 1A). Of the seven atypical nucleotides described by Huchzermeyer et al. (2009), our amplified ORF19 region included four of these substitutions, with two of them also present in the C. porosus poxvirus sequence (Fig. 1A). In comparison, the larger, more conserved ORF99 region (548 bp product) showed 96% similarity between the C. porosus poxvirus and Nile crocodilepox virus sequences (Fig. 1B). Phylogenetic analysis with these alignments and other high-GC chordopoxviruses, confirmed the phylogenetic results of Li et al. (2010) with the crocodile poxviruses forming a distinct clade (Fig. 2).
2.3. Transmission electron microscopy (TEM) confirmation One poxvirus-like lesion from one crocodile in each age category was selected for TEM. Briefly, lesions were fixed in 4% chilled glutaraldehyde in 0.1 M (pH 7.2) phosphate buffer for 72 h, before secondary fixing in 1% osmium tetroxide for 2 h. Samples were embedded in Spurr’s resin and ultrathin sections were cut and sections were stained with uranyl acetate and lead citrate. Sections were examined under the Jeol JEM-1200EX at 80 kV.
3.2. Confirmation of poxvirus using TEM TEM on the three representative lesions showed either enlarged epidermal cells completely filled with virions or encased in vacuoles (Fig. 3) within the plasma membrane. Virions were approximately 100 × 300 nm with the stereotypical dumbbell-shaped nucleoid visible within most of the virions suggesting virions had reached maturity.
2.4. Identification and tracking of poxvirus lesions 3.3. Poxvirus lesion development and healing Concurrently to the genetic confirmation, poxvirus-like lesions were grossly characterised. At the initial sampling, lesions were identified and the location on the belly skin was recorded for identification at subsequent samplings. The location of the scale was determined by counting the number of scales down from the collar and the number of
Initially, 152 lesions were identified with an additional 36 lesions developing during the course of the study. The repeated examination of lesions allowed for distinct categorisation of poxvirus lesions into phases: early active, active, expulsion and a healing phase before they 30
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
Fig. 1. Partial nucleotide sequence for (A) open reading frame 19 (ORF19) and (B) open reading frame (ORF99) using the Nile crocodilepox virus genome (Afonso et al., 2006; Accession number NC_008030) aligned against the atypical Nile crocodilepox virus described by Huchzermeyer et al. (2009) and the saltwater crocodile (C. porosus) poxvirus described within this study. The asterisk (*) indicates identical nucleotides between the atypical Nile crocodilepox virus sequence (Huchzermeyer et al., 2009) and the C. porosus poxvirus sequence. The atypical Nile crocodilepox virus sequence was not available for the ORF99 region for comparison.
keratinocytes containing large eosinophilic “Henderson-Patterson-like” inclusion bodies. The overlying keratin was intact, and there was compression of the underlying loose dermis (Fig. 5A). The chromatophore density under the lesion was reduced compared to normal skin rendering 100% of the active lesions clearly visible on the light table. As the lesions progressed to the active stage, there was an increase in surface area (1.03 ± 0.26 mm2) with consistent grey-white colouration (Fig. 4B). Active lesions appeared to have a distinct contour
were barely visible on the skin (Fig. 4). Of these 188 lesions, 55 lesions were examined on the light table from the necropsy animals to understand their impact on the raw skin product. The early active lesions (Fig. 4A) were grey-white foci with normal (64% lesions) to pin-point keratin damage (28% lesions) with an average surface area of 0.85 ± 0.29 mm2. During the early active stage, histology showed a well demarcated focus of marked epidermal hyperplasia and hypertrophy with overlying central plug (CP) of
Fig. 2. Neighbour-joining phylogenetic tree based on the amplified ORF 99 sequence from C. porosus, co mpared to the Nile crocodilepox virus and other high GC content poxviruses with bootstrap values given at the nodes. Genbank accession numbers are given in parentheses.
31
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
Fig. 3. Micrographs of a poxvirus-infected epidermal cell from C. porosus. (A) (5000X). (B) Virion nuclei (174 000X; black arrows) showing characteristic poxvirus dumbbell nucleoid.
Fig. 4. The progression of a typical poxvirus lesion: (A) early active lesion indicated by arrow, (B) active lesion, (C) after expulsion of necrotic plug and (D) healing with abnormal keratin. Ruler increments are 1 mm.
grossly compromised keratin whilst the remaining 28% developed on scales that appeared to have normal keratin. A full history of previous injuries was not available, but it is possible that there may have been damage to the skin which allowed the virus to enter and replicate at the site. An example of this possible pathogenesis is given in Fig. 6, where a scale showing no lesion at the initial sampling (Fig. 6A), acquires a linear abrasion, comprising the keratin (Fig. 6B). The abrasion appears to have healed by the third sampling with the exception of a presumed early active poxvirus lesion that has developed in the bottom left hand corner of the scale (Fig. 6C) and continued to form into an active lesion (Fig. 6D). By the final sampling, the lesion had healed (Fig. 6E). Interestingly, it appeared another early active poxvirus lesion started to develop below the integumentary sensory organ (ISO) by the fourth sampling, and continued to form into an active lesion by the final sampling. Lesion location indicated that 62% of the lesions occurred around the scale margins and there were more lesions on the lower (27%; n = 51) and mid (25%; n = 46) scale margins compared to the upper margins (11%; n = 20). There were eight lesions on the hinge between two scales and 11 on the ISOs with the remaining 52 lesions detected in the central scale region.
with a raised outer and central depressed core with abnormal keratin. Histology revealed that the CP continued to enlarge with keratinocytes containing large “Henderson-Patterson-like” inclusions (Fig. 5B). The keratin above the CP was disrupted, leaving an open hole which was visible macroscopically (Fig. 4B). The pigmented cells were sparse, remaining translucent on the light table. Histology showed mild perivascular dermal lymphocytic infiltration, indicating the lesions evoked a minor immune response (Fig. 5B). From the active stage, the CP of necrotic cells is released, referred to as the expulsion stage. During this stage, the lesion significantly increases in size (2.21 ± 0.29 mm2) and may be visible in one of two states; just prior to expelling the CP or after the expulsion has occurred. When the CP is about to be expelled, it overlies a crust of necrotic inflammatory cells immediately adjacent to the epidermis. The underlying epidermis returns to a relatively normal thickness and cellular morphology (Fig. 5C). After expulsion, the lesions appear grossly as a large, depressed foci with abnormal keratin (Fig. 4C). Commonly, the lesions have an orange/tan colouration. As within the previous stages, all these lesions remain clearly visible on the light table. In the healing phase, the surface area begins to decrease (1.62 ± 0.19 mm2; Fig. 4D). Healing lesions are characterised by undulating epidermal and dermal layers, and the chromatocytes remain sparse compared to the unaffected skin (Fig. 5D) resulting in a slight reduction in visibility on the light-table (87% visible). Macroscopically, the keratin of these lesions appears abnormal with normal skin colouration (Fig. 4D). From here, the lesions continue to heal, slowly becoming less visible with only 41% detected on the light table. On average, these lesions had a surface area of 0.72 mm ± 0.29 mm2. During the tracking of the lesions, an additional 36 lesions were detected on the skin and recorded during successive examinations. Twenty six of the 36 additional lesions (72%) occurred on areas with
4. Discussion Crocodylidpoxvirus, represented only by Nile crocodilepox virus, is a new genus of Chorodopoxvirinae (Afonso et al., 2006; ICTV, 2015). For the first time, the suspected poxvirus lesions on juvenile C. porosus belly skin were confirmed as poxvirus by PCR amplification of two different regions (ORF19 and ORF99; Fig. 1) from all 14 lesions tested. ORF19 is a protein of unknown function but specific to crocodylidpoxviruses (Afonso et al., 2006). However, the inherent variability of this region 32
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
Fig. 5. Histology of poxvirus lesion progression: (A) early Active lesion, (B) active lesion, (C) erupting lesion and (D) healing lesion. Where H is the hinge (inter-scalar) region; CP the central plug of necrotic cells; E the epidermis; De the dermis; arrows show the lymphocytes surrounding a dermal blood vessel; arrow heads indicate the chromatophores.
Horner (1988); 100 × 200 nm) and Buenviaje et al. (1992); 110 × 220 nm). This study is the first to identify and describe a clear progression of poxvirus lesions on the belly skin of the saltwater crocodile. We termed these stages as ‘early active’, ‘active’, ‘expulsion’, and ‘healing’ phases. In general, the lesions within this study conformed to the gross description of poxvirus lesions within Brown caiman (Caiman crocodilus fuscus) (Jacobson et al., 1979), Caiman yacare (Caiman crocodilus yacare) (Ramos et al., 2002), saltwater crocodile and freshwater crocodile (C. johnstoni) (Buenviaje et al., 1998; Ariel, 2011; Shilton et al., 2014). That is, approximately 1–3 mm in diameter foci and mainly grey to white in colour and were most likely descriptions of ‘early active’ to ‘active’ lesions as per our defined healing phases. The ulcerative lesion that breached the deep dermis noted by Buenviaje et al. (1998) was not confirmed in this study even at the ‘expulsion’ stage. In fact, none of the
shown in Huchzermeyer et al. (2009) and this study suggests this may not be a useful region to distinguish poxvirus variants. On the other hand, ORF99 is a highly conserved RNA polymerase subunit (Afonso et al., 2006) and phylogenetic analysis confirmed the C. porosus poxvirus to be closely related to the Nile crocodilepox virus (Fig. 2) and distinctly divergent from other high GC chordopoxviruses. In fact, Li et al. (2010) reported that Nile crocodilepox virus clustered with Molluscum contagiosum virus, but the addition of the C. porosus poxvirus from this study shows crocodylidopoxiruses to be distinctly divergent from other poxviruses. Genome sequencing of the saltwater crocodile isolate will further elucidate if it is indeed a new member of the Crocodylidopoxvirus genus. For completeness, TEM similarly confirmed the presence of stereotypical dumbbell-shaped nuclei from mature virions. The virions were 100 × 300 nm, which were slightly longer than previously reported by
Fig. 6. Poxvirus pathogenesis on an apparently normal belly scale (A). A linear lesion was incurred breaching the protective beta-keratin layer (B) and developed into an early active poxvirus lesion (C). This early active lesion further develops into an active lesion, with a second early active lesion developing below the ISO (arrowhead) (D). The original lesion (E) forms a depressed foci, with the second lesion developing into an active lesion. Arrow highlights the region of interest. Ruler increments are 1 mm. Photos are taken six weeks apart.
33
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
epidemiological studies should also explore possible environmental, vector (e.g. mosquitoes) or husbandry stressors that may dictate poxviral lesion incidence and severity to find more passive methods of control.
lesions histologically observed in this study showed any breach of the basement membrane with all viral bodies confined to the epidermal layers. In general, the Nile crocodile lesions described by Huchzermeyer et al. (1991) were consistent with those described herein. In contrast, the raised nodules (6–8 mm) described by Pandey et al. (1990) and Horner (1988) were not observed in this study, nor was the brown colouration seen by Huchzermeyer et al. (1991) although the yellow colouration noted by Pandey et al. (1990) could be associated to the orange colour we observed. However, none of our histological sections showed the cyst structures described by Huchzermeyer et al. (2009) which would definitively be observed on the tanned skin as they report. The concern for crocodile producers is the impact these lesions have on the final crocodile skin. Our results suggest the stage of the poxvirus lesion at the time of harvest determines whether the lesion will be seen on the skin post-harvest. Harvesting animals with lesions in the early active to expulsion stages has 100% probability of being visible on the light table and hence on the tanned skin. The probability of seeing lesions on the skin decreases during the healing stage (87%), but it is not until the lesions are barely visible on the live animal that the risk of visibility post-harvest begins to diminish (41%). It is not known what the average time for a lesion to completely heal would be, as not all lesions were observed at all stages. Crocodiles from a range of age categories were used in this study (hatchlings, yearlings and grow-outs), being selected due to their high incidence of poxvirus lesions, allowing classification of the different lesion stages. Unfortunately, the largest animals used in this study (grow-outs) were still substantially smaller than the average harvestsize crocodile produced on Australian crocodile farms. Personal observations and farm discussions by the authors suggest early stage poxvirus lesions are less common on harvest size crocodiles, with only the latter stages (erupted and healing phases) visible on these animals. Irrespective of whether the poxvirus lesions eventually heal and can be detected on the harvested skin, poxvirus still invokes a healing and inflammatory response, which is likely partitioning energy away from growth although not noticeably impacting the wellbeing of the crocodiles. The mid region of the scale is comprised of hard thick beta keratin that significantly increases with age from 9.8 μm thick when the crocodile is 0.3 m in total length to 89.9 μm when 1.6 m long (Manolis et al., 2000). This increase in keratin thickness with size might explain why there are fewer lesions on older animals. The hinge and ISOs regions of the scale are comprised of flexible alpha keratin (Richardson et al., 2001), suggesting these regions may be more permeable to injury and pathogens and further explains why 72% (n = 156) of the poxvirus lesions were associated with the periphery, hinge and ISO regions of the scale. Of the lesions that developed during the study period (n = 36), 72% were associated with abnormal keratin. This is consistent with the reported transmission of poxvirus through compromised keratin, allowing penetration of the poxvirus (McKeever et al., 1988; Buller and Palumbo, 1991; Birthistle and Carrington, 1997; Haig and McInnes, 2002; Tyring, 2003; Dlehon et al., 2004; Dohil et al., 2006). It is likely that poxvirus transmission within the saltwater crocodile and other crocodilians could occur as a result of their concentrated environments. This suggests that increasing the hygiene of pens may help control the spread of infection, along with ensuring no mechanical damage occurs to the keratin during husbandry procedures. In the remaining 28% of these lesions, there was no visible damage to the keratin, indicating either microscopic compromises, an alternate transmission route or systemic dissemination of the virus. Previously, latency has also been suggested in relation to crocodile poxvirus (Horner, 1988; Penrith et al., 1991). Horner (1988) reported the utility of using an autogenous vaccine to reduce the longevity of poxviral lesions on Nile crocodiles. To our knowledge, no further work has been done in this area particularly to show if early age vaccination would prevent lesion incursion. Further
Conflict of interest statement This research was conducted with funding from Charles Darwin University and no conflicts of interest are declared. Acknowledgements The authors would like to acknowledge Charles Darwin University and Berrimah Veterinary Laboratory staff, particularly Suresh Benedict for conducting the bacteriology and Sue Aumann for conducting histology on the internal organs. In addition, thank you to Darwin Crocodile Farm staff for access to the crocodiles used in this study. References Afonso, C.L., Tulman, E.R., Delhon, G., Lu, Z., Viljoen, G.J., Wallace, D.B., Kutish, G.F., Rock, D.L., 2006. Genome crocodilepox virus. J. Virol. 80, 4978–4991. Ariel, E., 2011. Viruses in reptiles. Vet. Res. 42, 100. http://dx.doi.org/10.1186/12979716-42-100. Birthistle, K., Carrington, D., 1997. Molluscum contagiosum virus. J. Infect. 34, 21–28. Buenviaje, G.N., Ladds, P.W., Melville, L., 1992. Poxvirus infection in two crocodiles. Aus. Vet. J. 69, 15–16. Buenviaje, G.N., Ladds, P.W., Martin, Y., 1998. Pathology of skin diseases in crocodiles. Aus. Vet. J. 76, 357–363. Buller, R.M., Palumbo, G.J., 1991. Poxvirus pathogenesis. Microbiol. Rev. 55, 80–122. Buoro, I.B.J., 1992. Pox-like virus particles in skin lesions of five Nile crocodiles in Kenya. Discov. Innov. 4, 117–118. Dlehon, G., Tulman, C.L., Afonso, C.L., Lu, Z., de la Concha-Bermejillo, A., Lehmkuhl, H.D., Piccone, M.E., Kutish, G.F., Rock, D.L., 2004. Genomes of the parapoxviruses orf virus and bovine popular stomatitis virus. J. Virol. 78, 168–177. Dohil, M.A., Lin, P., Lee, J., Lucky, A.W., Paller, A.S., Eichenfield, L.F., 2006. The epidemiology of molluscum contagiosum in children. J. Am. Acad. Dermatol. 54, 47–54. Haig, D.M., McInnes, C.J., 2002. Immunity and counter-immunity during infection with the parapoxvirus orf virus. Virus Res. 88, 3–16. Horner, R.F., 1988. Poxvirus in farmed Nile crocodiles. Vet. Rec. 122, 459–462. Huchzermeyer, F.W., Huchzermeyer, K.D.A., Putterill, J.F., 1991. Observations on a field outbreak of pox virus infection in young Nile crocodiles (crocodylus niloticus). J. S. Afr. Vet. Assoc. 62, 27–29. Huchzermeyer, F.W., Wallace, D.B., Putterill, J.F., Gerdes, G.H., 2009. Identification and partial sequencing of a crocodile poxvirus associated with deeply penetrating skin lesions in farmed Nile crocodiles, Crocodylus niloticus. Ondersepoort J. Vet. Res. 76, 311–316. ICTV, 2015. International Committee on Taxonomy of Viruses. (June 15 2015). http:// talk.ictvonline.org/Accessed. Jacobson, E.R., Popp, J.A., Shields, R.P., Gaskin, J.M., 1979. Poxlike skin lesions in captive caimans. J. Am. Vet. Med. Assoc. 175, 937–940. Kumar, S., Stecher, G., Tamura, K., 2015. MEGA7: molecular evolutionary genetics analysis version 7.0. Mol. Bio. Evol. 33, 1870–1874. Li, Y., Meyer, H., Zhao, H., Damon, I.K., 2010. GC Content-based pan-pox universal PCR assays for poxvirus detection. J. Clin. Microbiol. 48, 268–276. Manolis, S.C., Webb, G.J.W., 2011. Tracking Crocodile Skin Defects − from Farm to Product. Rural Industries Research and Development Corporation, Canberra (Publication No. 11/012). Manolis, S.C., Webb, G., Richardson, K., 2000. Improving the Quality of Australian Crocodile Skins. Rural Industries Research and Development Corporation, Canberra (Publication No. 00/21). McKeever, D.J., Jenkinson, D.M., Hutchinson, G., Reid, H.W., 1988. Studies on the pathogenesis of orf virus infection in sheep. J. Comp. Pathol. 99, 317–328. Miles, L., Moran, C., Isberg, S., 2010. Linkage Mapping and Quantitative Trait Loci (QTL) Analysis in Saltwater Crocodiles. Rural Industries Development Corporation, Canberra (Publication No. 10/152). Natural Resource Management of Ministerial Council, 2009. Code of Practice on the Humane Treatment of Wild and Farmed Australian Crocodiles. Australian Government, Canberra. Pandey, G.S., Inque, I., Ohshima, K., Okada, K., Chihaya, Y., Fujimoto, Y., 1990. Poxvirus infection in Nile crocodiles (Crocodylus niloticus). Res. Vet. Sci. 49, 171–176. Penrith, M.L., Nesbit, J.W., Huchzermeyer, F.W., 1991. Pox virus infection in captive juvenile caimans (Caiman crocodilus fucus) in South Africa. J. S. Afr. Vet. Assoc. 62, 137–139. Ramos, M.C.C., Coutinho, S.D., Matushima, E.R., Sinhorini, I.L., 2002. Poxvirus dermatitis outbreak in farmed Brazilian caimans (Caiman crocodilus yacare). Aus.Vet. J. 80, 371–372. Richardson, K.C., Park, Y.J., Webb, G.J.W., Manolis, S.C., 2001. Skin histology of embryonic and hatchling Estuarine and Australian Freshwater Crocodiles. Crocodilian: Biology and Evolution. Surrey Beatty and Sons Pty Ltd., Chipping Norton, pp.
34
Veterinary Microbiology 211 (2017) 29–35
R.L. Moore et al.
treatment. Am. J. Obstet. Gynecol. 189, S12–S16. Ye, J., Coulouris, G., Zaretskaya, I., Cutcutache, I., Rozen, S., Madden, T., 2012. PrimerBLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinf. 13, 134.
188–196. Shilton, C., Brown, G.P., Chambers, L., Benedict, S., David, S., Aumann, S., Isberg, S.R., 2014. Pathology of runting in farmed saltwater crocodiles (Crocodylus porosus) in Australia. Vet. Path. 1–13. Tyring, S.K., 2003. Mollusc um contagiosum: the importance of early diagnosis and
35