Host-pathogen dynamics of squirrelpox virus infection in red squirrels (Sciurus vulgaris)

C. Fiegnaa, M.P. Dagleisha, L. Coultera, E. Milneb, A. Meredithb, J. Finlaysona, A. Di Nardoc,d & C.J. McInnesa,*

a Moredun Research Institute, Pentlands Science Park, Penicuik, Edinburgh, EH26 0PZ, Scotland, UK

b The Royal (Dick) School of Veterinary Studies and The Roslin Institute, University of Edinburgh, Easter Bush Campus, Roslin, Midlothian, EH25 9RG, Scotland, UK

c The Pirbright Institute, Pirbright, Woking, Surrey, GU24 0NF, UK

d Institute of Biodiversity, Animal Health and Comparative Medicine, College of Medical, Veterinary and Life Sciences, University of Glasgow, Glasgow, G12 8QQ, Scotland, UK

* Corresponding author and contact details: C.J. McInnes, Moredun Research Institute, Pentlands Science Park, Penicuik, Edinburgh, Scotland, UK. E-mail address: Colin.Mcinnes@moredun.ac.uk; Tel: 0131 445 5111

1

1

23

45

67

8

910

111213

14

Abstract

To improve our understanding of squirrelpox virus (SQPV) infection in the susceptible host, three

red squirrels were challenged with wild-type SQPV via scarification of the hind-limb skin. All

squirrels seroconverted to the infection by the end of the experiment (17 days post-challenge).

Challenged animals suffered disease characterised by the development of multiple skin and oral

lesions with rapid progression of skin lesions at the infection site by day 10 post-challenge. No

internal pathological changes were found at post-mortem examination. A novel SQPV Taqman®

Real-time PCR detected viral DNA from multiple organs, with the largest amounts consistently

associated with the primary and secondary skin and oral lesions where viral replication was most

likely occurring. Immunohistochemistry clearly detected viral antigen in the stratified squamous

epithelium of the epidermis, tongue and the oropharyngeal mucosa-associated lymphoid tissue and

was consistently associated with histological changes resulting from viral replication. The lack of

internal pathological changes and the detection of relatively low levels of viral DNA when

compared with primary and secondary skin lesions argue against systemic disease, although

systemic spread of the virus cannot be ruled out. This study allowed a comprehensive investigation

of the clinical manifestation and progression of SQPV infection with a quantitative and qualitative

analysis of virus dissemination and shedding. These findings suggest two separate routes of SQPV

transmission under natural conditions, with both skin and saliva playing key roles in infected red

squirrels.

Keywords:

squirrelpox virus, SQPV, red squirrel, experimental infection, viral shedding, pathogenesis

2

15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

33

34

35

36

37

38

1. Introduction

The native red squirrel (Sciurus vulgaris) populations of the UK and Eire have declined

dramatically over the last 100 years and are now heading towards extinction with just a few isolated

populations remaining (Bosch and Lurz, 2012). In contrast, the grey squirrel (Sciurus carolinensis),

first introduced from North America in 1876, has greatly expanded its range. The reasons for the

decline in the red squirrel population can be attributed to many variables (Bosch and Lurz, 2012),

and although an earlier report suggested disease introduced with the grey squirrel (Middleton,

1930), the connection between epidemic disease in red squirrels and the presence of the imported

grey squirrels was debated throughout the 20th century (Edwards, 1962; Keymer, 1974; Vizoso,

1968). It was only latterly that a viral agent (initially described as parapoxvirus virus) responsible

for disease outbreaks in red squirrels was identified by transmission electron microscopy (TEM) of

the eyelid skin from a diseased red squirrel (Scott et al., 1981). Even then it took several more

years before research established the role of the grey squirrel as a reservoir species for the virus

(Reynolds, 1985; Sainsbury et al., 2000) and emphasised the crucial importance of this epizootic

disease in the wider context of disease-mediated competition with grey squirrels (Rushton et al.,

2006, 2000; Tompkins et al., 2003). Today, conservation strategies for the red squirrel take account

of squirrelpox virus (SQPV) as a major contributing factor in the threat to red squirrels from the

grey squirrel.

Squirrelpox virus is now classified as the sole member of an unclassified genus within the

Poxviridae family (Thomas et al., 2003; McInnes et al., 2006; Darby et al., 2014). The origin of

SQPV is still unknown, although it is thought that it was imported with grey squirrels from North

America. While the virus has never been described in the USA, serological samples collected from

grey squirrels in North America have tested positive for anti-SQPV antibodies (McInnes et al.,

2006). Fatal cases of disease resulting from SQPV infection have been described solely in red

squirrels in the UK and the Ireland (Sainsbury and Ward, 1996; Tompkins et al., 2002; Thomas et

al., 2003; Sainsbury et al., 2008; McInnes et al., 2009; LaRose et al., 2010; McInnes et al., 2013;

3

39

40

41

42

43

44

45

46

47

48

49

50

51

52

53

54

55

56

57

58

59

60

61

62

63

64

Naulty et al., 2013) with the exception of one report of a grey squirrel showing clinical signs of

SQPV disease which was confirmed by TEM (Duff et al., 1996). Otherwise, grey squirrels are

considered to be asymptomatic or sub-clinically affected by the virus (Atkin et al., 2010; Tompkins

et al., 2002). In red squirrels, the infection causes multifocal skin lesions characterised by

erythematous and ulcerative dermatitis which progress to haemorrhagic scabs and the disease

causes significant mortality (Carroll et al., 2009; Chantrey et al., 2014; Duff et al., 2010; LaRose et

al., 2010; McInnes et al., 2013, 2009; Sainsbury and Gurnell, 1995; Sainsbury and Ward, 1996;

Tompkins et al., 2002). To confirm that the disease is clinically asymptomatic in the grey squirrel,

an experimental infection with SQPV was performed (Tompkins et al., 2002). Grey and red

squirrels were inoculated simultaneously via skin scarification and subcutaneous routes. All the red

squirrels developed typical, severe SQPV-associated dermatitis and their health deteriorated rapidly

in contrast to grey squirrels which showed no clinical signs of infection and remained healthy

throughout.

Despite the disease and the potential carrier role of the grey squirrel being recognised for the last 15

years, remarkably little has been discovered about the pathogenesis or the transmission route(s) of

the virus either within red or grey squirrels or between the species. To date, there have been no

detailed studies on the temporal development of SQPV infection in red squirrels focusing on the

incubation period, the occurrence and development of lesions or which tissues or organs support

viral replication and, potentially, shedding. Furthermore, what is known has been derived primarily

from epidemiological studies of field cases. As part of a wider study to investigate the feasibility of

producing a vaccine against SQPV it was necessary first to establish, under experimental

conditions, an infection of red squirrels resembling SQPV infection in the wild. To investigate this

and provide additional information on the pathogenesis of SQPV we developed a specific and

sensitive SQPV Taqman® qPCR assay to compare the viral load from a broad range of tissues from

SQPV positive red squirrels both naturally and experimentally infected. We have correlated these

results with those from a variety of diagnostic techniques such as post mortem examination,

4

65

66

67

68

69

70

71

72

73

74

75

76

77

78

79

80

81

82

83

84

85

86

87

88

89

90

histopathology and a novel SQPV-specific immunohistochemical method (IHC) to determine which

tissues harbour the virus and which are most likely permissive for viral replication and are therefore

likely to be involved in virus transmission and shedding of virus.

2. Materials and methods

2.1 Experimentally-infected animals and related procedures

All experimental protocols involving infection of red squirrels were approved by the

Moredun Research Institute Animal Experiments & Ethical Review Committee and adhered strictly

to the requirements of the UK Animals (Scientific Procedures) Act 1986. Three adult red squirrels

(two male, sq. 01/12 and sq. 06/12; one female, sq. 04/12) were individually housed in 1 m × 0.75

m × 0.75 m cages which were furnished with a nest box with a removable lid. Food (mixed nuts,

sunflower seeds, whole corn and fruit) and water were provided ad libitum. All squirrels were

allowed to adjust to their environment for 14 days before the start of the procedures. All handling

procedures were performed under general anaesthesia. Anaesthesia was induced with gaseous 5%

isoflurane (IsoFlo, Abbott Animal Health, UK) in oxygen administered within an anaesthetic

chamber and maintained by 1.5-4% isoflurane in oxygen via a face mask.

2.2 Preparation of inocula, experimental infection and clinical observations

Dry skin lesions and scabs were collected from dead free-ranging red squirrels in the UK

found with clinical signs typical of SQPV disease and confirmed positive by either SQPV qPCR or

TEM. Scabs were ground with sterile sand in sterile PBS, approximately 6% v/v

penicillin/streptomycin solution (100 units/mL and 100 mg/mL, respectively) was added and the

inoculum clarified by centrifugation at 2,000 x g for 5 min. The inoculum was dispensed into

multiple aliquots and stored at -70 °C. SQPV DNA concentration was quantified by qPCR and

found to be approximately 2.5×1010 virus genome equivalents/mL. The presence of intact SQPV

virion particles was confirmed by TEM (courtesy of David Everest, Animal and Plant Health

Agency, Weybridge, UK). Mock inoculum was prepared using the antibiotic solutions added to

5

91

92

93

94

95

96

97

98

99

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

PBS.

Inocula were applied topically onto a previously shaved and scarified 2 × 2 cm area of skin on the

lateral aspect of the squirrel thighs. Scarification was with the tip of a 16G needle in a cross hatched

pattern with scratches approximately 0.5 cm apart. Each squirrel was challenged with 100 μl of

SQPV inoculum on the right thigh and 100 μl of mock inoculum on the left thigh. Animals were

monitored daily for clinical signs of disease with the skin lesions distant from the challenge sites

being fully assessed and recorded at the time of post-mortem examination (PM). The weight of each

squirrel was measured at the time of the virus challenge and estimated every day thereafter by

weighing the squirrels within their nest boxes. Clinical scores were recorded daily using a

modification of that used previously to assess the impact of squirrelpox disease on red squirrels

(Tompkins et al., 2002). A total clinical score of 6 on three consecutive days was the designated

humane end-point at which an individual animal would be removed from the experiment. All three

animals were euthanised, while under general anaesthetic, by intracardiac injection of

pentobarbitone sodium B.P. (approx. 200 mg/kg) 17 days post challenge (DPC) in line with this

clinical scoring procedure.

2.3 Post-mortem examination and collection of samples from experimental animals

The distribution of SQPV within 34 different tissues, blood, faecal and urine samples was

determined by qPCR analyses. Sterile nylon-flocked swabs (Thermo Scientific, Sterilin, Newport,

UK), pre-wetted or not with sterile PBS as appropriate, were used to collect samples of oral and

ocular secretions, samples from the skin surfaces associated with challenge sites and secondary

lesions and the lids of the nest boxes. Immediately post-euthanasia, 3-5 mL of blood was collected

by cardiac puncture using a 21G hypodermic needle. Approximately 2.5 mL was allowed to clot for

serology to test for the presence of antibodies against SQPV using the enzyme-linked

immunosorbent assay (ELISA) previously described (Sainsbury et al., 2000). The remainder was

placed into paediatric EDTA tubes for extraction of nucleic acids. Individual sterile surgical

instruments were used to harvest each tissue sample to avoid possible SQPV cross-contamination

6

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

between tissues. Scarified skin samples were collected first followed by other representative skin

samples (eyelid, lip, chin, nose, ear, axilla, anterior and posterior digital and mock scarified skin).

Harderian glands, submandibular lymph nodes (SM LN), submandibular salivary glands (SM SG)

and mucosa-associated lymphoid tissue (MALT) present bilaterally between the palatoglossal and

palatopharyngeal arches on the lateral walls of the oropharynx (corresponding to the anatomical

location of palatine tonsils in other mammals) were collected in sequence followed by tongue and

parotid salivary gland samples. Tissue samples were collected in duplicate with one stored at -70 °C

and the other fixed in 10% v/v neutral buffered formalin solution. Due to their small size, both right

and left popliteal lymph nodes were collected for molecular analyses only, whereas MALT samples

from both sides of the oropharyngeal mucosa from squirrel 01/12 were collected for histopathology

only. All major internal body organs were examined for evidence of macroscopic abnormalities

followed, whenever possible, by an additional ~0.5 cm thick tissue sample processed as above. The

brain was removed whole, fixed and processed as above after a small section of the frontal cortex

was collected and frozen for molecular studies. Urine samples were usually collected by

cystocentesis during the PM and faecal samples were collected from the descending colon and

rectum. A full list of tissues sampled appears in Table 1.

2.4 Suspected naturally-infected red squirrels

Wild red squirrel carcasses (n=13) found by red squirrel conservation organizations, ranger

services and members of the public were submitted to the Royal (Dick) School of Veterinary

Studies or to the Moredun Research Institute. Where the condition of the carcasses permitted, a full

diagnostic PM, including histology, was undertaken to establish the cause of death, and sera (or

fluid from the body cavity) tested for the presence of SQPV antibodies (Sainsbury et al., 2000).

SQPV qPCR analyses were performed on a panel of different tissues as indicated in Table 1. Where

no skin lesions suspicious of SQPV disease were present, tissue samples were generally confined to

eyelid, digital and lip skin. A skin lesion or scab sample was also taken from suspected SQPV-

infected animals for confirmation of SQPV by TEM (data not shown).

7

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

2.5 Histology and immunohistochemistry

Tissue samples were processed routinely prior to embedding in paraffin wax. Sections (5 μm

thick) were stained with hematoxylin and eosin (HE) for histology. For SQPV

immunohistochemistry (IHC), sections (5 μm) were mounted on Superfrost™ slides (Menzel-

Gläser, Braunschweig, Germany), dewaxed in xylene and taken through graded alcohols to 95%

prior to quenching endogenous tissue peroxidase activity with 3% hydrogen peroxide in methanol

(v/v) for 20 minutes at room temperature (~18-22 °C). Slides were washed in water for 5 minutes

then transferred to PBS containing 0.05% v/v Tween20 (PBS-T). Antigen retrieval was performed by

immersing sections in a solution of 0.01M Trizma® base, 0.001M EDTA, 0.05% Tween20 at pH 9.0

(all Sigma-Aldrich Co., Dorset, UK) for 10 minutes at 95C. Non-specific antigen binding was

blocked by incubation with 25% normal rabbit serum (NRS), diluted in PBS-T, for 30 minutes at

room temperature. The IgG fraction was purified from pooled sera from grey squirrels previously

testing positive in the SQPV ELISA (Sainsbury et al., 2000). Approximately 0.5 ml of pooled sera

was diluted to 5ml in PBS containing 0.02% sodium azide and applied to a 1ml Protein A/G

cartridge (ThermoFisher Scientific Loughborough UK) prewashed with 10ml PBS.

Chromatography was carried out on an AKTA FPLC (GE Healthcare Life Sciences, Little Chalfont

UK). Bound IgG was washed with 20ml PBS, then eluted with Pierce IgG Elution Buffer 

(ThermoFisher Scientific, UK) and 1ml fractions collected into tubes containing 100µl 1M Tris pH

8.4. The majority of IgG, estimated by monitoring the OD280, was eluted in a 2ml volume which

was dialysed overnight against a 2 L volume of PBS at 4oC. Purified IgG was adjusted to 0.5mg/ml

prior to use. It was then further diluted 1/500 in 25% NRS/PBS-T (final total IgG concentration

24μg/ml), applied to sections and the slides incubated overnight at 4°C prior to being washed in

PBS-T. Primary antibodies were visualised using a rabbit anti-squirrel IgG: horse radish peroxidase

conjugate (kindly provided by David Deane, Moredun Research Institute) diluted 1/200 in 25%

NRS/PBS-T and applied to sections for one hour at room temperature. Sections were washed in

PBS-T, Nova red chromogen (Vector Laboratories, Peterborough, UK) added for 10 minutes,

8

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

washed in tap water and counterstained with hematoxylin Z (Cellpath plc., Newtown, Powys, UK),

blued-up with Scot’s tap water substitute. Negative control slides were prepared by substituting the

primary antibody, at the same concentration, with purified squirrel IgG from the sera of SQPV

antibody negative grey squirrels.

2.6 SQPV Taqman® RealTime PCR

The SQPV Taqman® Real-Time PCR (qPCR) assay was based on the amplification of part

of the SQPV_003 gene (Accession number: HE601899) (Darby et al., 2014; McInnes et al., 2006;

Thomas et al., 2003) which encodes a protein predicted to contain an immunoglobulin-like domain.

Primers: forward 5'-TCCTGCAGTCATCCATCGAA-3', reverse

5'TCGCTGATGTTGTAGATGAAGTTG-3' and probe 5'Fam-CTCCGATCCCCGTCGCAACCT-

3'Tam were designed to amplify a 142 bp fragment of the SQPV gene. Reaction conditions were

established for the ABI PRISM® 7000 Sequence Detection System (Applied Biosystems,

Warrington, UK) in a total reaction volume of 25 μl. The assay mixture contained 12.5 μl of 2 ×

Taqman® Universal PCR Master Mix (Applied Biosystems, UK), 2.5 μl of 3 μM forward primer,

2.5 μl of 9 μM reverse primer and 2.5 μl of 2 μM probe and 5 μl RNase/DNase-free water (for the

no-template control; NTC), 5 μl of DNA template (40ng/ μl) or 5 μl SQPV uncut SQPV cosmid

standard dilutions (40 ng/μl), as appropriate. All samples were tested in quadruplicate. For positive

controls, squirrelpox viral genomic DNA was used. Total DNA extracted from each tissue was

quantified by spectrophotometry at 260/280 nm and adjusted to a final concentration of 40 ng/μl. As

blood, urine and swab samples generally contained lower concentrations of genomic DNA qPCR

reactions were performed with 5 μl of purified DNA without adjusting the concentration. Thermal

reaction conditions consisted of 2 minutes at 50 °C, 10 minutes at 95 °C and 45 cycles each

consisting of 95 °C for 15 seconds and 60 °C for 1 minute. The mean cycle threshold (Ct) value for

each sample was determined for all reactions. When at least one of the quadruplicate NTC gave a

positive reaction, the assay was considered to be invalid and therefore repeated. Unknown samples

with all four replicates negative were considered SQPV-negative. Viral DNA concentrations were

9

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

calculated for all those samples with 4/4 positive replicates and a Ct score ≤36.5. If the standard

deviation of the resultant mean was >1, the sample was retested. Samples with Ct scores ≤36.5 but

with less than 4/4 positive replicates were classified as borderline positive as were those with at

least 1/4 positive replicates and a mean Ct score ≥36.5.

2.7 qPCR validation and relative SQPV quantification in samples

Ten-fold dilutions of a cosmid containing the SQPV_003 gene (McInnes et al., 2006),

ranging from 3 to 3×107 viral genome equivalents/reaction, were used to generate 23 consecutive

and independent standard curves. Standard curves were generated by plotting the resulting Ct values

versus the log10 of the cosmid dilutions . To mimic the concentration and composition of clinical

samples, 200 ng of squirrel genomic DNA (containing no SQPV DNA) were added to each standard

dilution. Values obtained from the 23 standard curves were employed to construct a reference grand

mean calibration curve [i.e. the Grand Mean Standard curve (GMS)] for determining the amount of

SQPV DNA in each clinical sample. Viral load was inferred from viral genome equivalent/μg of

total DNA extracted (V/μg) for all tissue and faecal samples, viral genome equivalent/mL (V/mL)

for urine and blood samples and viral genome equivalent/swab (V/s) for swab samples.

DNA extraction was performed using appropriate commercial kits in accordance with the sample

type and followed as per manufacturer’s instructions. Briefly, DNeasy Blood and Tissue Kit

(Qiagen, Crawley, UK) was used for DNA extraction from 100 μl of EDTA blood, 30 μl SQPV

inoculum and from solid tissues, where 15-25 mg of samples were homogenised in ATL buffer

using Lysing Matrix D beads (MP Biomedical, UK) prior DNA extraction. Homogenates were

incubated overnight at 56 °C with 40 μl proteinase K solution (Qiagen, Crawley, UK). QIAamp

DNA Mini kit (Qiagen, Crawley, UK) was used for swab samples and DNA extractions from 180-

220 mg of frozen faecal pellet was performed using the QIAamp DNA stool Mini Kit (Qiagen,

Crawley, UK). QIAamp Viral RNA Minikit (Qiagen, Crawley, UK) was used for DNA extraction

from 140 μl of urine. To confirm that the latter protocol was suitable for DNA extraction from urine

samples an aliquot of squirrel urine was spiked with two different amounts of SQPV DNA as an

10

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

internal positive control for nucleic acid extraction.

3. Results

3.1 Disease progression in experimentally infected red squirrels and gross pathology at post-

mortem examination

All three SQPV-challenged red squirrels developed extensive lesions typical of SQPV

infection at the inoculation site and one or more skin lesions distal from this. Over the course of the

experiment there was moderate loss of body weight (<20% ). A significant correlation between the

clinical scores (derived from assessing appetite, depression, size of the primary pox lesion and

presence of suspected secondary pox lesions), and the time course of the experiment was observed

(z = 12.2, p = 0.000), with all animals showing a consistent increase in the clinical score from 8

DPC (z = 3.9, p = 0.000).

The site of SQPV challenge was visibly red by 1 DPC similar to the contra-lateral mock-infected

site. However, in contrast to the mock infection site where the initial insult regressed, the skin at the

site of SQPV inoculation became progressively more swollen and markedly erythematous between

2 and 4 DPC, and by 10 DPC the primary skin lesions measured at least 3 mm wide in all three

squirrels. At 8 DPC, mild perifollicular erythema associated with focal alopecia and mild cutaneous

swelling was observed on the upper lip of squirrel 01/12. Similarly, mild bilateral eyelid skin

oedema and focal alopecia was observed in squirrel 06/12 by11 DPC, however, at this time-point no

macroscopic ulcerative pox-like secondary skin lesions were noted in any of the squirrels. By 14

DPC the ill-defined area observed on the upper lip of squirrel 01/12 had a mild sero-sanguineous

exudate. Between 10-17 DPC the progression of the SQPV skin lesions at the challenge site evolved

markedly in all three squirrels and were characterised by irregular, coalescing and centrally

depressed skin ulcers almost entirely coated with dark red/brown sero-sanguineous or sero-purulent

plaque (scab) and surrounded by oedematous and erythematous skin margins (Figure 1).

Post-mortem examinations (17 DPC) confirmed the presence of additional multifocal secondary

skin lesions, mainly confined to the skin of the face and oral mucosa. The lesions at the site of

11

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

SQPV challenge measured approximately 20-25 mm in diameter whereas the scarified skin of the

mock-challenge appeared grossly normal with early hair re-growth. The secondary lesion initially

suspected 8 DPC on the upper lip skin of squirrel 01/12 included most of the muzzle skinand was

covered with a semi-moist red crust. Similarly, squirrel 06/12 had a large secondary skin lesion

affecting part of the muzzle which was severely ulcerated with a thick scab. Multiple additional

secondary cutaneous lesions were also present randomly over the body including one vesicular

lesion on the left pinna of sq. 06/12. Two squirrels had focal umbilicated crusty lesions, one close to

the margin of the muco-cutaneous junction of the lower lip (sq. 04/12), the other on the left eyelid

(sq. 06/12). In addition to the skin lesions, two squirrels (sq. 01/12 and sq.06/12) had an ulcerative

lesion on the dorsal mucosal surface of the tongue. No gross pathological changes were observed in

any internal organs of the three squirrels except mild to moderate enlargement of the sub-

mandibular lymph nodes. All three squirrels had moderate amounts of food in their stomachs and/or

intestines.

3.2 Histopathology and immunohistochemistry of experimentally infected red squirrels

At 17 DPC microscopic skin lesions were similar to those previously reported in naturally-

infected red squirrels (Atkin et al., 2010; Duff, 2012; McInnes et al., 2013, 2009; Naulty et al.,

2013) and changes characteristic of different stages of lesions development were simultaneously

and multifocally present in the same animal. Early infection was characterised by epidermal and

follicular hyperplasia (acanthosis) with associated mild hyperkeratosis and multifocal to focally

extensive ballooning degeneration of epidermal cells. Swollen cells contained the occasional

intracytoplasmic (2-5 µm) palely eosinophilic viral inclusion bodies (IB) and multiple foci of

degenerate cells coalesced resulting in intraepidermal vesiculopustular formations (Figure 2 A). As

the lesions progressed into the chronic stage, a severe ulcerative necrotizing dermatitis/cellulitis,

often associated with numerous intraepidermal pustular lesions, was observed. The lesions were

covered by a thick serocellular crust formed by layers of degenerated and necrotic keratinocytes,

polymorphonuclear leukocytes (PMN) admixed with serous proteinaceous fluid and numerous

12

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

clusters of cocci-shaped bacteria (Figure 2 B). The underlying dermis was expanded by a dense

perivascular to diffuse, mixed inflammatory cell infiltrate composed of variable numbers of viable

and degenerated PMN admixed with macrophages, lymphocytes and plasma cells according to the

degree of epidermal/dermal ulceration, necrosis and the extent of secondary bacterial infection. The

microscopic changes associated with an early stage of SQPV infection were present primarily at the

margins of the gross lesions or in some skin samples (e.g. eyelid skin, mock scarified skin, ear skin)

that lacked typical ulcerative and crusty lesions characteristic of SQPV disease.

In addition to the skin, microscopic changes were also present in the oral mucosa. In 2/3 squirrels

(sq. 01/12 and sq. 06/12) the tongue was affected by multifocal necrotising or ulcerative glossitis

with epithelial hyperplasia and ballooning degeneration of the stratified squamous epithelium which

contained occasional intracytoplasmic IBs. In all three squirrels there was variable disruption of the

epithelium and underlying submucosa of the lateral wall of the oropharyngeal lymphoid tissue, a

component of the mucosal-associated lymphoid tissue (MALT). Morphological changes were

multifocal and similar to those observed in the skin lesions with the presence of numerous

intracytoplasmic IBs (Figure 2 C). The stratified squamous epithelium occasionally had structural

disruption of the underlying lymphoid tissue by infiltration of a large number of viable and

degenerate PMNs that often extended into the surrounding glandular tissue, however there was no

direct evidence of viral infection of lymphocytes.

Submandibular lymph nodes had moderate to marked follicular hyperplasia likely to be as a result

of the severe cutaneous lesions over the face and in the oral mucosa, there were no other changes

suggestive of viral lymphadenitis or peripheral lymphadenopathy. The lungs of the three

experimentally infected squirrels had a small number of alveoli lined by cuboidal cells (presumed

type II pneumocyte hyperplasia) and the epithelia of bronchi and bronchioles were mildly

hyperplastic. No microscopic changes indicative of SQPV infection were detected in any of the

other tissues examined.

13

299

300

301

302

303

304

305

306

307

308

309

310

311

312

313

314

315

316

317

318

319

320

321

322

323

Immunohistochemistry for SQPV demonstrated that viral antigen was restricted to the cytoplasm of

the stratified squamous epithelial cells of the epidermis (skin lesions) and oral mucosa (tongue and

oropharyngeal lymphoid tissues) and it was generally associated with microscopic changes typical

of early SQPV lesions. Stronger and granular antigen labelling was observed within the strata

spinosum and granulosum, associated with hyperplastic and hydropic degenerated cells (Figures 2

D and 2E). Only minor differences were observed between individuals and sites, no labelling was

detected in lung and the negative control preparations were devoid of any labelling.

3.3 Validation of SQPV Real-Time PCR

The amplification of the 10-fold serial dilutions of cosmid containing the SQPV DNA was

linear over 5 orders of magnitude from 3×102 to 3×107 copies per reaction. The amplification

efficiency averaged 93.3±6.2 (95% PI 83.9 to 102.6), with a coefficient of determination (R2) of

0.98±0.01 (95% PI 0.95 to 1.00). There was no significant difference between the individual slopes

in consecutive assays (n= 23; p>0.05). Although the SQPV qPCR assay was able to detect and

quantify as few as 0.13 fg of viral genome equivalent, which represents 3 copies of the cosmid per

reaction (equivalent to 15 V/μg DNA and 600 V/mL), the probability of detection was 15% and

58% for 3 and 30 viral genome equivalents, respectively. At >30 genome equivalents (>150 V/μg

DNA) the detection probability was 100%. The estimated intra-assay coefficient of variation (CV)

ranged between 0.5% and 6.5%, with samples containing lower virus copies consistently producing

higher variability. The inter-assay CV estimated from the linear regressions computed for each of

the 23 standard curves was, on average, 4.3%±1.72 (95% PI 1.9 to 7.9). The resulting GMS curve

constructed from the 23 standard curves is shown in Figure 3. The average efficiency was 95.9%

(95% CI 91.8 to 100.5%) with a slope equivalent to -3.42 (95%CI -3.31 to -3.54) and coefficient of

determination (R2) of 0.95. The variability within the Ct data used for estimating the GMS curve

was expressed by a CV of 5.7%. SQPV DNA load of each tested sample was subsequently

predicted from the GMS curve regression equation as Ct = 41.36 + (-3.42×log10 SQPV DNA copy).

The optimal cut-off for reliable quantification was arbitrarily set at a Ct value of 36.5. No

14

324

325

326

327

328

329

330

331

332

333

334

335

336

337

338

339

340

341

342

343

344

345

346

347

348

349

significant variation in the viral load (p>0.05), expressed as log10 of viral copies/μg DNA, was

reported between the values obtained from each sample run with the relative standard curve and the

average loads derived from the GMS curve, where 97.9% of the values lay within the 95% limits of

agreement.

3.4 Detection of SQPV DNA in experimentally infected red squirrels

Generally similar results were obtained from each of the three experimentally infected red

squirrels with minor individual variations that reflected the size and number of gross lesions. With

the exception of a single axillary skin sample the highest concentrations of SQPV DNA were

detected from tissues of the integumentary system, and in particular the skin (Table1). Highest

virus DNA concentrations of 1.63×106 V/μg DNA to 6.02×107 V/μg DNA were detected from the

SQPV challenge site and in samples collected from, or just proximal to, secondary skin lesions (e.g.

nasal, chin and lip skin), however, concentrations of SQPV DNA of between 5.15×103 V/μg DNA

to 6.84×104 V/μg DNA were also detected in skin samples from the other eyelids, digits, mock-

scarified and ear skin samples, all of which were devoid of gross lesions typical of SQPV infection.

In particular, the average concentration of virus found in the digital skin of the forelimb samples

was 5.55×104 V/μg DNA. Lower average SQPV DNA concentrations (3.86×102 to 1.26×103 V/μg

DNA) or negative results were obtained from the axillary skin, perianal scent gland and Harderian

gland.

Of the lymphoid organs, the highest SQPV DNA concentration was detected in the oropharyngeal

MALT (n=2, average 2.55×107 V/μg DNA). Despite no gross lesions, the viral load was comparable

to those found in secondary skin lesions and scabs (Table1). The amount of SQPV DNA detected in

selected lymph nodes was approximately 3 to 5 orders of magnitude less compared with

oropharyngeal MALT samples and the average viral load in the two right popliteal lymph nodes,

draining the challenge site, was only slightly higher compared to the contra-lateral popliteal lymph

node (1.34×104 and 3.21×102 V/μg DNA, respectively). SQPV concentration in one spleen sample

15

350

351

352

353

354

355

356

357

358

359

360

361

362

363

364

365

366

367

368

369

370

371

372

373

374

(sq. 01/12) was 1.61×102 V/μg DNA, where as the spleens from the other two squirrels were either

negative or borderline positive (as defined above).

In the alimentary tract (Table 1) decreasing amounts of SQPV DNA were detected in the tongue,

faeces, stomach, parotid salivary gland, SI, SM SG, caecum, rectum, LI and liver. The amount of

SQPV DNA detected in faecal samples (average of 1.16×105 V/μg DNA) was approximately

comparable with those in the tongue or stomach tissue samples of squirrel 01/12. Lower

concentrations of viral DNA (≤1.81×102 V/μg DNA) were detected in all other tissues of the

digestive system that were examined. Virus DNA was detected in the liver of all three squirrels but

at very low levels (<150 V/μg DNA). Similarly virus DNA was detected unreliably or at

concentrations never exceeding 4.05×103 V/μg DNA in the lungs, kidneys, gonads, heart and brain,

suggesting that virus replication was unlikely in these tissues (Table 1).

Relatively high amounts of viral DNA were detected in the oral and ocular swabs as well as skin

swabs from the challenge site and the mock-scarified skin (Table 1). The average amount of viral

DNA detected in the oral swabs was higher (5.15×107 V/s) than that obtained from either the tongue

or oropharyngeal MALT and in particular the viral load obtained from the oral swab of the only

squirrel (sq. 04/12) which lacked gross lesions on the tongue was similar to the other two squirrels

which had oral lesions. A swab of the vesiculopustular lesion on the left pinna of squirrel 06/12 also

contained high levels of viral DNA (3.72×107 V/s), confirming the lesion was associated with

SQPV replication. Swabs of the metal surfaces of the nest boxes were also found to contain

relativity high amounts (average of 1.28×105 V/s) of viral DNA. SQPV DNA was detected in the

urine of only one squirrel (4.50×103 V/mL) and that same squirrel was positive for SQPV DNA

(2.83×104 V/mL) in its blood (Table 1).

3.5 Detection of SQPV DNA from naturally infected red squirrel samples

Of the 13 suspected cases of squirrelpox in naturally infected red squirrels, none of the

tissue samples from one of them (R86/07) gave qPCR positive results. It was subsequently regarded

as not having been infected with SQPV and was used as an additional negative control. Among

16

375

376

377

378

379

380

381

382

383

384

385

386

387

388

389

390

391

392

393

394

395

396

397

398

399

400

samples from the other 12 red squirrels, with the exception of a single eyelid sample (squirrel

R05/08), the greatest amounts of SQPV DNA (2.94×107 to 6.78×107 V/μg DNA) were detected

consistently in skin samples(Table 1).

The tissues that contained the next largest amount of viral DNA (7.13×104 to 4.52×105 V/μg DNA)

were stomach, tongue and SM LN, although 48% of these samples were either qPCR negative or

borderline positive. Decreasing viral DNA concentrations ( 4.54×103 to 2.26×104 V/μg DNA) were

present in the rectum, SI, LI, caecum, brain, heart and lung samples, with a large number of samples

negative or borderline positive (Table 1). The overall lowest average viral DNA concentrations

detected, in the order of 4 to 5 magnitudes lower compared to skin samples, were obtained from

liver, spleen and kidney samples many of which were considered negative.

3.6 Serology

Of the naturally-infected red squirrels 6/12 (50%) were positive for anti-SQPV antibodies,

whilst all three (100%) of the experimental animals were positive at 17 DPC.

4. Discussion

This study successfully reproduced SQPV infection in all three experimentally-challenged

red squirrels via epidermal scarification. Clinical signs and PM findings confirmed that secondary

cutaneous lesions were found consistently and yet no virus-associated pathological changes were

present in any of the internal organs. This is in agreement with a single previous description of

SQPV experimental infection in red squirrels (Tompkins et al., 2002) and many reports of naturally

acquired SQPV infection (Atkin et al., 2010; Duff et al., 2010; McInnes et al., 2009; Naulty et al.,

2013; Sainsbury and Ward, 1996). In our study all three squirrels (100%) developed both gross and

histological lesions. Although one animal was less severely affected with regard to the size, number

and severity of skin lesions (sq. 04/12) it was not possible to determine if this animal might have

survived the infection as is found in some 8-10% of naturally-infected animals (Chantrey et al.,

2014; Sainsbury et al., 2008). The incubation period of SQPV in our study, inferred from the

observation of suspected secondary macroscopic skin lesions, was estimated to be ~8-10 days

17

401

402

403

404

405

406

407

408

409

410

411

412

413

414

415

416

417

418

419

420

421

422

423

424

425

426

which agrees with field observations suggesting an incubation time of less than 15 days (Carroll et

al., 2009).

The novel SQPV Taqman® qPCR assay was capable of detecting as few as 3 copies of the virus

gene, as sensitive as other reported poxvirus qPCR assays (Balamurugan et al., 2009; Sofi Ibrahim

et al., 2003). It uses an imported external standard curve (i.e. GMS curve) removing the necessity to

perform one for each run, saving both time and reagents. Overall, it confirmed the relative lack of

virus in the internal organs, but also pointed to a potential source of virus in the saliva.

Virus dissemination within the tissues of the experimentally infected squirrels was similar to the

distribution and relative quantities of viral DNA found in naturally infected animals and, in

agreement with previous reports (Atkin et al., 2010; Collins et al., 2014), the highest average viral

DNA concentration was consistently detected in skin, oral mucosa, scabs and faeces. Although the

amount of viral DNA detected generally correlated with the severity and extent of the skin lesions

significant amounts of viral DNA, with quantities comparable to the primary skin lesions, were

detected in the oropharyngeal mucosa (ie. MALT) of experimentally challenged squirrels despite

the absence of macroscopic lesions. The average amount of viral DNA detected in the oral swabs

was even higher, again even in one squirrel (sq. 04/12) which lacked lesions on the tongue.

Although SQPV-associated lesions were reported previously on the dorsal surface of the tongue

(McInnes et al., 2009) and viral DNA has also been detected previously in saliva (Collins et al.,

2014), this is the first time that significant amounts of SQPV DNA have been associated with viral

replication, as suggested by the IHC, and potential shedding from the stratified squamous

epithelium of the oropharyngeal MALT. Infection of the oropharyngeal epithelial mucosa may have

resulted from persistent ingestion of virus particles during feeding (e.g. direct contact with lip skin

lesions) and self-grooming, followed by the virus entry via oral epithelial micro-abrasions.

Combined IHC, histology and qPCR analyses indicate that oral epithelium and epidermis are highly

permissive for SQPV replication and could suggest that the relatively high viral DNA loads

detected in the gastro-intestinal tract and faeces are as a result of repeated ingestion of virus

18

427

428

429

430

431

432

433

434

435

436

437

438

439

440

441

442

443

444

445

446

447

448

449

450

451

452

particles generated in the oral mucosa. Consequentially, faecal-oral transmission may represent a

potential route of infection as suggested previously (Atkin et al., 2010; Collins et al., 2014). Indeed,

Gledhill (1962) demonstrated that after oral exposure with ectromelia virus (ECTV), infected mice

can sustain infection with virus shed persistently in the faeces of apparently healthy mice. However,

because of the lack of permissive cells for culturing the wild type SQPV used in these studies we do

not know if the viral DNA detected in the faeces represents viable virus or if the gastric acids and

digestive enzymes are able to inactivate the virus.

The results of this study also confirm that cutaneous lesions contain large amounts of virus. These

lesions were most commonly found around the head and face and again are likely to be as a result,

of virus entry through minor skin abrasions. During squirrelpox outbreaks, the majority of diseased

red squirrels are found usually to have more abundant skin lesions compared with our experimental

group. Some of these may be associated with more the numerous traumatic skin abrasions which are

likely to occur in the natural environment.. It is therefore likely that social interactions (e.g.

interactions at feeders, drey sharing and fighting), scent marking (face-wiping behaviour on

branches and bark), self- and allo-grooming may play an important part in the transmission and

dissemination of naturally acquired disease (Rushton et al., 2000; Bruemmer et al., 2010). In

addition, given the very high concentration of virus in skin and scab samples, and also

vesiculopustular lesions, the contribution of mechanical transmission by ectoparasites between

individuals (e.g. fleas in shared dreys) cannot be excluded, as previously suggested (Atkin et al.,

2010; Collins et al., 2014; Rushton et al., 2000), although the epidemiological role of arthropods in

sustaining disease is less clear.

It has been hypothesised that the epidemiological characteristics of SQPV infection resemble those

of acute disease in systemic poxvirus infections, especially ECTV (Atkin et al., 2010; Carroll et al.,

2009). In our study, the detection of SQPV DNA in many organs suggests that systemic spread of

the virus had occurred. However, it seems unlikely that the lymph nodes, thymus, spleen, liver,

heart, brain, lung, kidney or gastro-intestinal tract act as sites of virus replication given that SQPV

19

453

454

455

456

457

458

459

460

461

462

463

464

465

466

467

468

469

470

471

472

473

474

475

476

477

478

DNA was not consistently detected in these tissues and the concentration was relatively low

compared with skin samples. The absence of both gross and histopathological abnormalities in these

organs was also taken as evidence of the lack of systemic disease, although not necessarily lack of

systemic dissemination of the virus via the lymphatic ? haematogenous system. The presence of

SQPV DNA in the blood as an indication of systemic disease, however, remains unclear. Although

we detected SQPV DNA in the blood of experimentally infected squirrels, similar to what had been

found previously in cases of naturally acquired infection (Atkin et al., 2010; Collins et al., 2014),

results were inconsistent between the three animals. This may reflect the fact that we were only able

to sample the blood once, at 17 DPC, and therefore missed the main viraemia. However, similar to

the pathogenesis of vesicular exanthema of swine virus in experimentally infected pigs (Gelberg

and Lewis, 1982), we speculate that SQPV may initially enter the blood via virus-laden infected

leukocytes from the drainage area of the primary lesion and subsequently, attracted by chemotactic

gradients, migrate and be released in distal sites of previously injured epithelial cells (e.g. skin and

oral mucosa).

5. Conclusion

This study has helped describe the possible routes of SQPV transmission and tissue

locations of viral shedding. Firstly, it confirms that viral entry via damaged skin is an effective route

of infection. Secondly, our results suggest that the oral mucosa and consequently saliva may have

important roles in SQPV epidemiology where apparently healthy animals without gross skin lesions

may be a source of infection by contaminating food at feeders, through bite wounds or mutual

grooming. Rapid identification of infectious animals is a key component of surveillance strategies

and the qPCR assay described here could be a useful addition to the diagnostic repertoire as it can

detect low copy number of virus particles at an early stage of infection. In addition to skin samples,

oral and skin swabs are probably useful and easy alternatives for sampling live red squirrels during

SQPV outbreaks and routine screening. Further research should therefore be undertaken to confirm

20

479

480

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495

496

497

498

499

500

501

502

503

whether primary SQPV infection can be contracted via the oral route and through environmental

contamination.

Conflict of interest statement

None

Acknowledgment

The authors would like to thank Ann R. Wood, Jackie Thomson and all MRI BioServices

staff for animal husbandry and monitoring, Kim Willoughby for advice on qPCR assay design,

Janice Gilray and David Dean (MRI) for providing serology results and SQPV antibodies and

David Everest (APHA) for providing TEM. Thanks also to all field workers, conservation

organizations and ranger services for providing red squirrels carcasses. This study was funded by

Wildlife Ark Trust and the Scottish Government Rural and Environment Science and Analytical

Services Division.

References

Atkin, J.W., Radford, A.D., Coyne, K.P., Stavisky, J., Chantrey, J., 2010. Detection of squirrel poxvirus by nested and real-time PCR from red (Sciurus vulgaris) and grey (Sciurus carolinensis) squirrels. BMC Vet. Res. 6, 33.

Balamurugan, V., Jayappa, K.D., Hosamani, M., Bhanuprakash, V., Venkatesan, G., Singh, R.K., 2009. Comparative Efficacy of Conventional and TaqMan Polymerase Chain Reaction Assays in the Detection of Capripoxviruses from Clinical Samples. J. Vet. Diagnostic Investig. 21, 225–231.

Bosch, S., Lurz, P.W.W., 2012. The Eurasian Red Squirrel Sciurus vulgaris. Westarp Wissenschaften Verlagsgesellschaft mbH, Germany.

Bruemmer, C.M, Rushton, S.P., Gurnell, J., Lurz, P.W.W., Nettleton, P., Sainsbury, A.W., Duff, J.P., Gilray,J., McInnes, C.J., 2010. Epidemiology of squirrelpox virus in grey squirrels in the UK. Epidemiol. Infect. 138, 941–950.

Carroll, B., Russell, P., Gurnell, J., Nettleton, P., Sainsbury, A W., 2009. Epidemics of squirrelpox virus disease in red squirrels (Sciurus vulgaris): temporal and serological findings. Epidemiol. Infect. 137, 257–265.

Chantrey, J., Dale, T.D., Read, J.M., White, S., Whitfield, F., Jones, D., McInnes, C.J., Begon, M., 2014. European red squirrel population dynamics driven by squirrelpox at a gray squirrel invasion interface. Ecol Evol 4, 3788–3799.

Collins, L.M., Warnock, N.D., Tosh, D.G., McInnes, C., Everest, D., Montgomery, W.I., Scantlebury, M., Marks, N., Dick, J.T., Reid, N., 2014. Squirrelpox virus: assessing prevalence, transmission and environmental degradation. PLoS One 9, e89521.

21

504

505

506

507

508

509

510

511

512

513

514

515

516

517518519

520521522

523524

525526527

528529530

531532533

534535536

Darby, A.C., McInnes, C.J., Kjaer, K.H., Wood, A.R., Hughes, M., Martensen, P.M., Radford, A.D., Hall, N., Chantrey, J., 2014. Novel host-related virulence factors are encoded by squirrelpox virus, the main causative agent of epidemic disease in red squirrels in the UK. PLoS One 9, e96439.

Duff, J.P., 2012. Squirrelpox. In: Gavier-Widèn, D., Meredith, A., Duff, J.P. (Eds.). Infectious Diseases of Wild Mammals and Birds in Europe. Oxford, UK: Wiley-Blackwell, pp. 196-199.

Duff, J.P., Haley, P., Wood, R., Higgins, R.J., 2010. Causes of red squirrel (Sciurus vulgaris) mortality in England. Vet. Rec. 167, 461.

Duff, J.P., Scott, A., Keymer, I.F., 1996. Parapox virus infection of the grey squirrel. Vet Rec 138, 527.

Edwards, F.B., 1962. Red squirrel disease. Vet. Rec. 74, 739–741.

Gelberg, H.B., Lewis, R.M., 1982. The pathogenesis of vesicular exanthema of swine virus and San Miguel sea lion virus in swine. Vet Pathol 19, 424–443.

Gledhill, A.W., 1962. Latent Ectromelia. Nature 196, 298–298.

Keymer, I.F., 1974. Report of the pathologist, 1971 and 1972. The zoological Society of London, Scientific report 1971-1973. J. Zool. 173, 55.

LaRose, J.P., Meredith, A.L., Everest, D.J., Fiegna, C., McInnes, C.J., Shaw, D.J., Milne, E.M., 2010. Epidemiological and postmortem findings in 262 red squirrels (Sciurus vulgaris) in Scotland, 2005 to 2009. Vet Rec 167, 297–302.

McInnes, C.J., Coulter, L., Dagleish, M.P., Deane, D., Gilray, J., Percival, A., Willoughby, K., Scantlebury, M., Marks, N., Graham, D., Everest, D.J., McGoldrick, M., Rochford, J., McKay, F., Sainsbury, A.W., 2013. The emergence of squirrelpox in Ireland. Anim. Conserv. 16, 51–59.

McInnes, C.J., Coulter, L., Dagleish, M.P., Fiegna, C., Gilray, J., Willoughby, K., Cole, M., Milne, E., Meredith, A., Everest, D.J., Macmaster, A.-M., 2009. First cases of squirrelpox in red squirrels (Sciurus vulgaris) in Scotland. Vet. Rec. 164, 528–531.

McInnes, C.J., Wood, A.R., Thomas, K., Sainsbury, A.W., Gurnell, J., Dein, F.J., Nettleton, P.F., 2006. Genomic characterization of a novel poxvirus contributing to the decline of the red squirrel (Sciurus vulgaris) in the UK. J. Gen. Virol. 87, 2115–2125.

Middleton, A.D., 1930. The Ecology of the American Grey Squirrel (Sciurus carolinensis Gmelin) in the British Isles. Proc. Zool. Soc. London 2, 809–843.

Naulty, F., Everest, D., Warnock, N.D., Phelan, K., Callanan, J.J., 2013. Squirrelpox virus in red squirrels (Sciurus vulgaris) in the Republic of Ireland. J Wildl Dis 49, 1070–1073.

Reynolds, J.C., 1985. Details of the demographic replacement of the red squirrel (Sciurus vulgaris) by the grey squirrel (Sciurus carolinensis) in Eastern England. J. Anim. Ecol. 54, 149–162.

Rushton, S.P., Lurz, P.W.W., Gurnell, J., Fuller, R., 2000. Modelling the spatial dynamics of parapoxvirus disease in red and grey squirrels: a possible cause of the decline in the red squirrel in the UK? J. Appl. Ecol. 37, 997–1012.

Rushton, S.P., Lurz, P.W.W., Gurnell, J., Nettleton, P., Bruemmer, C., Shirley, M.D.F., Sainsbury, A.W., 2006. Disease threats posed by alien species: the role of a poxvirus in the decline of the native red squirrel in Britain. Epidemiol. Infect. 134, 521–533.

22

537538539

540541

542543

544

545

546547

548

549550

551552553

554555556

557558559

560561562

563564

565566

567568

569570571

572573574

Sainsbury, A.W., Deaville, R., Lawson, B., Cooley, W.A., Farelly, S.S., Stack, M.J., Duff, P., McInnes, C.J., Gurnell, J., Russell, P.H., Rushton, S.P., Pfeiffer, D.U., Nettleton, P., Lurz, P.W., 2008. Poxviral disease in red squirrels Sciurus vulgaris in the UK: spatial and temporal trends of an emerging threat. Ecohealth 5, 305–316.

Sainsbury, A.W., Gurnell, J., 1995. An investigation into the health and welfare of red squirrels, Sciurus vulgaris, involved in reintroduction studies. Vet Rec 137, 367–370.

Sainsbury, A.W., Nettleton, P., Gilray, J., Gurnell, J., 2000. Grey squirrels have high seroprevalence to a parapoxvirus associated with deaths in red squirrels. Anim. Conserv. 3, 229–233.

Sainsbury, T., Ward, L., 1996. Parapoxvirus infection in red squirrels. Vet Rec 138, 400.

Scott, A.C., Keymer, I.F., Labram, J., 1981. Parapoxvirus infection of the red squirrel (Sciurus vulgaris). Vet Rec 109, 202.

Sofi Ibrahim, M., Kulesh, D.A., Saleh, S.S., Damon, I.K., Esposito, J.J., Schmaljohn, A.L., Jahrling, P.B., 2003. Real-time PCR assay to detect smallpox virus. J. Clin. Microbiol. 41, 3835–9.

Thomas, K., Tompkins, D.M., Sainsbury, A.W., Wood, A.R., Dalziel, R., Nettleton, P.F., McInnes, C.J., 2003. A novel poxvirus lethal to red squirrels (Sciurus vulgaris). J. Gen. Virol. 84, 3337–3341.

Tompkins, D.M., Sainsbury, A.W., Nettleton, P., Buxton, D., Gurnell, J., 2002. Parapoxvirus causes a deleterious disease in red squirrels associated with UK population declines. Proc. R. Soc. London. Ser. B Biol. Sci. 269, 529–533.

Tompkins, D.M., White, a. R., Boots, M., 2003. Ecological replacement of native red squirrels by invasive greys driven by disease. Ecol. Lett. 6, 189–196.

Vizoso, A.D., 1968. A red squirrel disease. Symp. Zool. Soc. London 24, 29–38.

23

575576577578

579580

581582

583

584585

586587

588589

590591592

593594

595

596

597

Figures and Table captions

Figure 1. Gross appearance of SQPV-scarified skin at 17 days post-challenge

Right lateral thigh - Squirrel 06/12. Skin lesion at inoculation site. Ssevere erythematous, exudative

and ulcerative dermatitis with scab formation.

Figure 2. Histopathology and immunohistochemistry of experimentally infected red squirrels

at 17 days post-challenge with SQPV

(A) Section of nasal skin at the edge of a macroscopic ulcerative lesion - Squirrel 06/12. Marked

epidermal hyperplasia and ballooning degeneration of epidermal and follicular epithelial cells

(asterisks) which leads to intraepidermal pustule formation (arrows). Bar = 200 μm. (B) Section of

SQPV scarified skin at the inoculation site. - Squirrel 06/12. Ulcerative and necrotising dermatitis

covered by scab formation (asterisks) typical of an advanced SQPV skin lesion. Bar = 500 μm. (C)

Section of the mucosal-associated lymphoid tissue (MALT) of the lateral wall of the oropharynx -

Squirrel 04/12. There is marked acanthosis of surface epithelium associated with ballooning

degeneration of the stratum spinosum. Individual swollen cells often have intracytoplasmic, palely

eosinophilic viral inclusion body (arrowheads). The underlying submucosa is expanded by a dense

inflammatory cell infiltrate composed mostly of viable and degenerated neutrophils. Bar = 100 μm.

(D) Lip skin - Squirrel 01/12. Immunohistochemical labelling (brick red colour) of SQPV antigen is

restricted to the epidermal and follicular epithelial cells. Mild to strong, diffuse (arrowhead) to

granular (arrows) labelling is present within the cytoplasm of hyperplastic, spongiotic and

ballooning degenerated epithelial cells . Bar = 500 μm. (E) A mucosal fold (asterisk) of the

oropharyngeal MALT - Squirrel 06/12. Focally extensive, moderate (arrowhead) to strong (arrow)

immunolabelling of intracytoplasmic SQPV antigen which is restricted to the stratified squamous

epithelium and is associated with hyperplastic and hydropic degenerated cells. Bar = 200 μm.

Inserts within D and E represent negative control preparations.

Figure 3. SQPV qPCR Grand Mean Standard calibration curve (GMS)

24

598

599

600

601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

620

621

622

Grand Mean Standard (GMS) calibration curve (red line) obtained by combining the data resulting

from 23 independent and consecutive standard curves (grey palettes) generated by plotting the log10

of the SQPV DNA cosmid serial dilution (x-axis) against the cycle threshold (Ct) values (y-axis).

Coloured areas indicate the extent of the 95% coefficient interval (CI) range, where grey dots

correspond to the actual values of the qPCR reads. Each standard SQPV cosmid dilution was

performed in quadruplicate reactions with concentrations ranging from 0.13 pg/reaction (3 viral

equivalent) to 1.3 ng/reaction (3×107 viral equivalent). The relative slope, y-axis intercept and

correlation coefficient (R2) for the GMS curve are displayed on the bottom left.

Table 1. qPCR detection of SQPV DNA in samples from red squirrels experimentally and naturally infected.

Viral DNA detection, concentrations and relative average from different samples of 3 red squirrels

(01/12, 04/2 and 06/12) experimentally infected with SQPV at 17 DPC and of 12 red squirrels

naturally infected with SQPV (R21/07, R23/07, R52/07, R69/97, R05/08, R06/08, R27/08, R811,

R812, R23/09, R24/09, R31/09). Virus loads were determined from the GMS curve and expressed

as: V/ μg of DNA extracted from lymphoid organs, integumentary system, alimentary tract and

other organs samples; V/S for swab samples; V/mL for blood and urine samples. For squirrel 06/12

the average viral DNA relative to eyelid skin and ocular swabs was obtained using the average of

both samples (right and left). (-) negative qPCR result; (+) borderline positive results; (*) ELISA

positive squirrels; (empty cells) no samples tested.

25

623

624

625

626

627

628

629

630

631632

633

634

635

636

637

638

639

640

641

642