1

Detection of Antibodies to the Feline Leukemia Virus (FeLV) Transmembrane 1

Protein p15E: an Alternative Approach for Serological FeLV Diagnosis Based 2

on Antibodies to p15E 3

4

Eva Boenzli1*, Maik Hadorn2, Sonja Hartnack3, Jon Huder4, Regina Hofmann-5

Lehmann1, and Hans Lutz1# 6

7

8

1 Clinical Laboratory, Vetsuisse Faculty, University of Zurich, Winterthurerstr. 260, 9

8057 Zurich, Switzerland 10

2 Department of Chemistry and Applied Biosciences, Swiss Federal Institute of 11

Technology Zurich, Vladimir-Prelog-Weg 3, 8093 Zurich, Switzerland 12

3 Section of Epidemiology, Vetsuisse Faculty, University of Zurich, Winterthurerstr. 13

260, 8057 Zurich, Switzerland 14

4 Institute for Medical Virology, Swiss National Centre for Retroviruses, University of 15

Zurich, Winterthurerstr. 190, 8057 Zurich, Switzerland 16

17

#Corresponding author. Mailing address: Clinical Laboratory, Vetsuisse Faculty, 18

University of Zurich, Winterthurerstrasse 260, 8057 Zurich, Switzerland. Phone: +41 19

44 635 83 11. Fax: +41 44 635 89 23. E-mail: hlutz@vetclinics.uzh.ch 20

21

*Present affiliation: Eva Boenzli, Department of Chemistry and Applied Biosciences, 22

Swiss Federal Institute of Technology Zurich, Zurich, Switzerland 23

E-mail addresses of remaining authors: 24

JCM Accepts, published online ahead of print on 2 April 2014J. Clin. Microbiol. doi:10.1128/JCM.02584-13Copyright © 2014, American Society for Microbiology. All Rights Reserved.

2

EB: eva.boenzli@org.chem.ethz.ch; MH: maik.hadorn@org.chem.ethz.ch 25

SH: sonja.hartnack@access.uzh.ch; JH: huder.jon@virology.uzh.ch 26

RH: regina@hofmann-lehmann.com. 27

28

Running title: FeLV diagnosis by serology 29

30

3

Abstract 31

It was the aim of this paper to investigate whether diagnosis of FeLV infection by 32

serology might be feasible and useful. Among the various viral proteins, the FeLV 33

env-gene product (SU) and the envelope transmembrane protein p15E were 34

considered promising candidates for serological diagnosis of FeLV infection. 35

Thus, we evaluated p15E and three other FeLV antigens, namely a recombinant 36

env-gene product, whole FeLV, and a short peptide from the FeLV 37

transmembrane protein, for their potential to detect FeLV infection. To evaluate 38

possible exposure of cats to FeLV, we tested by ELISA serum and plasma from 39

experimentally and naturally infected and vaccinated cats for presence of 40

antibodies to these antigens. The serological results were compared with the p27 41

and proviral realtime-PCR results. We found that p15E displayed in 42

experimentally infected cats a diagnostic sensitivity of 95.7% and specificity of 43

100%. In naturally infected cats, p15E showed a diagnostic sensitivity of 77.1% 44

and a specificity of 85.6%. Vaccinated cats displayed minimal antibody levels to 45

p15E suggesting that anti p15E antibodies indicate infection rather than 46

vaccination. The other antigens turned out to be too unspecific. The lower 47

specificity in cats exposed to FeLV under field-conditions may be explained by 48

the fact that some cats become infected and seroconvert in absence of 49

detectable viral nucleic acids in plasma. We conclude that p15E-serology may 50

become a valuable tool to diagnose FeLV infection; it may in some cases replace 51

PCR (< 250 words). 52

Keywords: Feline leukemia virus, infection, diagnosis, serology, transmembrane 53

protein, diagnostic sensitivity, diagnostic specificity. 54

4

55

5

Introduction 56

Infection with the feline leukemia virus (FeLV) (1) is of veterinary relevance (2, 3), 57

although it’s importance differs in most study populations (4, 5). The disease 58

outcome in infected cats is usually defined according to the presence of provirus 59

and viral antigen in the blood (6-8). However, it is highly unpredictable because it 60

is dependent on factors like the virus subtype (9), the age (10) and the general 61

condition of the cat. 62

Diagnosis of FeLV infection is mainly based on the detection of virus or viral 63

antigen in the plasma, serum or whole blood. The most common serological tests 64

detect either the presence of p27 antigen by enzyme-linked immunosorbent 65

assay (ELISA) (11), or FeLV structural antigens in the cytoplasm of infected 66

leukocytes and platelets by immunofluorescence antibody test (IFA) (12, 13). 67

Moreover, western blot analysis detects the presence of specific FeLV antibodies. 68

Alternatively, non-serological diagnoses include virus isolation (14) or PCR to 69

detect proviral (FeLV DNA)- or viral load (FeLV RNA) (15-17). However, due to 70

the laborious and/or cost-intensive character of most of these methods, they are 71

not all suitable for clinical use. 72

It is known that infected cats are able to elicit antibodies against different 73

components of FeLV (18-22). However, until now the detection of antibodies to 74

FeLV had limited significance for several reasons: firstly, there is no evidence 75

how reliable antibody detection can predict FeLV infection; secondly, it is 76

unknown, which antibodies are suitable and thirdly, there is a widespread 77

existence of endogenous FeLV (enFeLV) in cat populations in that every cell in 78

every single cat harbours multiple copies of enFeLV (23, 24). As enFeLV is not 79

6

completely tolerated by the immune system, antibodies are elicited (25) which are 80

undistinguishable from antibodies to exogenous FeLV. Only few techniques, e.g. 81

real-time PCR, are able to distinguish between endogenous and exogenous 82

FeLV (26). Thus, FeLV antibodies were so far not considered to be useful as a 83

diagnostic parameter. Moreover, several studies failed to detect a sufficient 84

antibody response against various epitopes of FeLV. Fontenot and co-workers 85

(27) analyzed the reactivity of a predicted FeLV transmembrane immunodominant 86

domain (Imd-TM peptide), and investigated its potential as diagnostic reagent in 87

serology. It was revealed that this peptide displayed only negligible levels of 88

reactivity using sera from FeLV-infected cats, rendering the Imd-TM peptide as 89

not qualified for FeLV diagnosis. Langhammer and co-workers (25) produced 90

recombinant FeLV p15E and showed that cats infected with FeLV developed 91

antibodies against p15E, although the reactions in ELISA were low. Epitope 92

mapping revealed a variety of epitopes recognized by sera from FeLV-infected 93

animals, including epitopes detected by sera from p15E-immunized cats, but 94

weaker. They concluded that natural FeLV infection results in a weak induction of 95

binding antibodies specific for p15E, and a low induction of neutralizing 96

antibodies. However, Lutz and co-workers (22) qualitatively and quantitatively 97

compared the antibody levels to different FeLV components in naturally infected 98

cats and found that p15E exhibited strong antigenicity. They observed that cats 99

that became immune or viremic after infection displayed elevated levels of 100

antibodies to p15E. They concluded that antibodies to p15E indicate FeLV 101

infection, but may have little involvement in virus neutralization. With these results 102

in mind, we hypothesized that the FeLV TM envelope protein p15E and other viral 103

7

proteins may have the potential to be a useful diagnostic tool in serology. We 104

therefore evaluated p15E, a recombinant FeLV env-gene product (p45), whole 105

virus (FL-74), and a short synthetic peptide (EPK211) derived from the TM unit of 106

the FeLV envelope protein. Using indirect ELISA’s, we systematically screened 107

sera from naturally and experimentally infected and immunized cats. From each 108

sample, the results of provirus-PCR, p27 in plasma, and immunization status 109

were known. 110

111

Materials and Methods 112

Antigens for ELISA 113

The following antigen preparations were evaluated: 114

■ p15E: The whole TM subunit without the membrane spanning helix part of the 115

viral envelope protein of FeLV was cloned and purified in our laboratories. Briefly, 116

the amino acid sequence 446-582 and 626-642 of FeLV-A (GenBank: 117

AAA93093.1) was cloned into the pET-16b expression vector (Novagen, Merck 118

Millipore, Switzerland) and expressed in Escherichia coli BL21-CodonPlus (DE3)-119

RIL cells (Stratagene). The fusion protein containing p15E was purified by affinity 120

chromatography. Afterwards, it was dialyzed against 7.5mM Tris-HCl buffer pH 121

8.0. The protein product was verified with western blot analysis using a primary 122

in-house monoclonal mouse ascites antibody in a dilution of 1:6000 to specifically 123

detect p15E, and a secondary goat α mouse peroxidase (PO)-conjugated 124

antibody (Jackson ImmunoResearch Laboratories, Inc., PA, USA) in a dilution of 125

1:1000. The purity of p15E, i.e. the presence of potential E. coli residuals in p15E 126

8

was determined using three different E. coli sera tested with ELISA (Fig. 1), see 127

Section Enzyme-Linked Immunosorbent Assay. 128

■ p45: The non-glycosylated recombinant variant of gp70 surface unit (vaccine 129

antigen) of the envelope protein (14) of FeLV subtype A (FeLV-A) was purchased 130

from Virbac (Virbac Schweiz AG, Glattbrugg, Switzerland). This antigenic 131

suspension (Leucogen®) is adjuvanted with 0.1ml of a 3% aluminium hydroxide 132

gel and with 10 μg purified extract of Quillaja saponaria. 133

■ Whole virus (FL-74): Sucrose-gradient purified whole FeLV was available in our 134

laboratories. Whole virus originated from FL-74 cells, which is a lymphoblastoid 135

cell line chronically infected with FeLVABC (28). 136

■ EPK211: This 15 amino acids long synthetic peptide (Ac-137

GWFEGWFNRSPWFTT-NH2) mimics a short part of the carboxy-terminal region 138

of the FeLV TM subunit of the viral envelope protein. It was expressed by solid-139

phase peptide synthesis and purified by analytical high-performance liquid 140

chromatography (purity> 85%) in the laboratories of EspiKem Srl (Florence, Italy). 141

142

Sera and plasma samples 143

We used serum and plasma from different laboratory animals as well as from 144

privately owned animals. 145

The samples from the first study (n/group=9) done in 2006 (8) are defined as 146

follows: Group 1 served as unvaccinated control animals. Group 2 was 147

vaccinated with Eurifel® (new name: Purevax®; Merial, Lyon, France). Eurifel is a 148

non-adjuvanted canarypox-vectored live vaccine (ALVAC) containing FeLV-A 149

env, gag and part of pol. Group 3 received Fel-O-Vax® LV-K IV (Fort Dodge, 150

9

Iowa, USA), which is a polyvalent killed whole virus FeLV vaccine. Vaccines were 151

injected twice subcutaneously. Four weeks after the second vaccination, each cat 152

was challenged with FeLV-A/Glasgow-1 (29). 153

The samples from the second vaccination study (n/group=15) (unpublished data) 154

are defined as follows: Group 1 served as unvaccinated control group. Group 2 155

received Fevaxyn® FeLV (Solvay Animal Health, Inc., Minnesota, USA), which 156

consists of inactivated (or killed) antigen. Group 3 was vaccinated with Fel-O-157

Vax® LV-K IV (Fort Dodge), and group 4 received Leucogen® (Virbac). Vaccines 158

were injected twice subcutaneously. Four weeks after the second vaccination, 159

each cat was challenged with FeLV-A/Glasgow-1 (29). In both studies, blood was 160

collected prior to and at the day of challenge exposure. Thereafter, samples were 161

collected weekly until week 15. Furthermore, we used plasma from unvaccinated 162

control groups from week 8 or 10 post challenge from Pfizer (Kent, United 163

Kingdom), Merial (Lyon, France) and Virbac (Glattbrugg, Switzerland) (n=47). For 164

longitudinal effects on antibodies, we used the sera from the unvaccinated but 165

challenged control groups from week 8. Moreover, we used the plasma from 166

naïve, unchallenged and unvaccinated specific pathogen-free (SPF) young cats 167

(n=94) from the different vaccination studies and from old unchallenged SPF cats 168

(n=10) available in our laboratories from an earlier study. To test the reactivity of 169

vaccinated cats, we used all groups from the vaccination studies mentioned 170

above and tested the samples after several vaccine applications but before 171

challenge with FeLV-A. 172

Additionally, sera from privately owned cats (n=294) in Switzerland were used. 173

The samples already existed in our laboratories and were collected from April 174

10

2004 till January 2005 (8). The veterinarians who had submitted these samples 175

to us also provided us with the vaccination record of the cats. For all of the 176

samples we tested, PCR (provirus) and ELISA (p27) results were known. 177

178

Enzyme-linked immunosorbent assay 179

ELISA was based essentially on methods described earlier (30, 31). Anti-FeLV 180

p45 and anti-FeLV whole virus (FL-74) antibodies were measured by ELISA as 181

previously described (22, 32). P45 and whole virus were used at final 182

concentrations of 100ng/well. For EPK211 and p15E, ELISAs were newly 183

established to find out optimal signal to noise-ratios. Before coating, the antigen 184

(100μg/ml) was boiled for 1min at 95°C in a 10% sodium dodecyl sulfate (SDS) 185

(Sigma-Aldrich) solution and then diluted in 0.1M carbonate coating buffer pH9.6 186

(Na2CO3 water-free) (Sigma-Aldrich) to final concentrations of 0.25ng/μl (antigen) 187

and 0.005% (SDS). 100μl of this solution was added to each well of a flat-bottom 188

96-well ELISA plate (TPP, Trasadingen, Switzerland) and incubated for 3h at 189

37°C. Plates were either used after incubation or stored at -20°C. The assay was 190

performed as follows: in order to block remaining empty spaces in the wells to 191

avoid unspecific binding, 1% bovine serum albumin (BSA) (Sigma-Aldrich) in P3x 192

buffer pH7.4 (0.15M sodium chloride, 1mM EDTA, 0.05M Tris-base, 0.1% BSA, 193

0.1% Tween 20) was added and incubated for 1h (37°C). Wells were washed 194

three times with ELISA wash buffer pH7.4 (0.15M sodium chloride, 0.2% Tween 195

20). Sera and plasma samples were diluted 1:200 in P3x and 100μl/well was 196

added to the wells. After incubation of 1h (37°C) and three washing steps, a goat 197

α cat IgG (H+L) PO-conjugated secondary antibody (Milan Analytica AG, 198

11

Rheinfelden, Switzerland) diluted 1:3000 in P3x was added (100μl/well). After 199

another 1h of incubation (37°C) and three washing steps, 100μl of a substrate 200

solution consisting of 0.2M citric acid pH4.0 (98vol%), 2% hydrogen peroxide 201

(1vol%), and 40mM ABTS [2, 2'-azino-di(3-ethylbenzthiazoline-6-sulfonate)] 202

(1vol%), (all chemicals from Sigma-Aldrich) was added. After 10min (room 203

temperature), optical density (OD) values were measured at 415nm with a 204

spectrophotometer (Spectramax plus 384, Molecular Devices, CA, USA). 205

Comparability of assays run on different days was achieved by using 2 control 206

sera as internal standards with every assay; one had a strong reactivity to FeLV, 207

the other was negative. 208

P15E purity was tested with ELISA under the same conditions mentioned above 209

using three different rabbit sera (kindly provided by Prof. M. Wittenbrink) 210

containing antibodies against specific epitopes of E. coli. A secondary goat α 211

rabbit IgG (H+L) PO-conjugated antibody (Milan Analytica) was used to detect 212

any signal. For the negative and positive control, cat sera and affiniPure goat α 213

cat IgG (H+L) PO-conjugated secondary antibody (Milan Analytica) was used. 214

215

PCR 216

Exogenous FeLV Provirus sequences present in blood leukocytes were detected 217

and quantitated by real-time PCR under conditions described earlier (26). This 218

PCR amplifies a 74 bp sequence within the FeLV U3 LTR portion specific for 219

exogenous FeLV. 220

221

Data analysis and Statistics 222

12

ELISA OD values were used as standardized OD values: (OD value (sample)-OD 223

value (negative control)/(OD value (positive control)-OD value (negative control)). 224

Software package NCSS® 2007 Version 07.1.20 (LLC Kaysville, Utah, USA) and 225

PASW® statistics software version 18.0.2 (Polar Engineering and Consulting, 226

Nikiski, USA) were used for the receiver operating characteristic (ROC) analyses, 227

and Win Episcope 2.0® software (Borland International Inc., California, USA) for 228

calculation of the kappa values. To compare the differences in the mean values 229

between the old SPF cats and the young SPF cats, one-way ANOVA and 230

Welch/Brown-Forsythe test was used. 231

232

Results 233

Purity of recombinant p15E. 234

Unlike the other three antigens that were already available for antibody ELISA, 235

the FeLV transmembrane protein p15E had to be specifically synthesized and 236

purified in our laboratories. After cloning in E. coli cells and demonstration of the 237

product’s presence by western blot analysis, p15E was higly purified and tested 238

for possible E. coli contaminants. Three different sera (kindly provided by M. 239

Wittenbrink) from rabbits immunized with E. coli antigens were used to test the 240

levels of antibodies directed to E. coli contaminants of the p15E cloning product. 241

P15E antibody ELISA revealed that OD values were maximally 0.117 (serum 2) 242

which is comparable to the negative control (naïve cat serum, OD= 0.106) (Fig. 243

1). The other two E. coli sera (sera 1 and 3) exhibited even lower OD values (OD 244

< 0.05), which were considered negligible. These results indicated that our p15E-245

13

cloning product was highly pure and could be used for ELISA assay without 246

cross-reactions due to E.coli contaminants. 247

248

Antibody levels in experimentally infected cats 249

Receiver operating curve (ROC) evaluation showing a plot of the true positive 250

rate (sensitivity) against the false positive rate (1-specificity) was used to evaluate 251

the diagnostic utility of the different FeLV antigens and to compare experimentally 252

infected cats with naturally infected cats. 253

In the first ROC analysis for experimentally infected animals, cutoff points 254

resulting in optimal test specificity and sensitivity were determined to discriminate 255

uninfected, provirus negative, and specific pathogen free (SPF) cats from 256

infected, provirus positive (seroconverted) cats. To have defined and comparable 257

sample set of cats of the same age, the old SPF cats (n=10) were excluded from 258

this first study. 259

Evaluation of the antibody levels of experimentally infected cats represented a 260

proof of principle test. Using provirus as a gold standard, the ROC curve for those 261

antibody levels is represented in Figure 2a. As a rule, the cutoff was defined as 262

specificity ≥ 80% and sensitivity x specificity as maximum (Fig. 2b). Antigen p15E 263

exhibited an excellent fit, following the left-hand border and the top border of the 264

ROC space. An optimal cutoff of OD= 0.0495 represented the best tradeoff 265

between sensitivity which was found to be 95.7% and specificity which was found 266

to be 100%. The optimal cutoff (-0.039) for EPK211 displayed a sensitivity of 267

65.7% and specificity of 81.1%, whereas p45 (cutoff 0.016) revealed a sensitivity 268

14

of 80% and specificity of 88.9%. Whole virus displayed a sensitivity of 65.7% and 269

specificity of 81.1% with a cutoff of -0.007. 270

Cohen’s κ (kappa) values, which represent a chance corrected measure of 271

agreement between two sets of categorized data revealed an almost perfect level 272

of agreement (κ = 0.96) between p15E and provirus (Table 1). The agreements 273

between provirus and p45, whole virus (FL-74), and EPK211 showed lower levels 274

ranging between 0.47 and 0.70. Agreement between pairs of antigens showed 275

satisfying to good levels, but lower than agreement between p15E and provirus. 276

Longitudinal effects on antibodies from uninfected SPF cats were evaluated and it 277

was revealed that antibody levels of SPF cats stayed below the cutoff during 22 278

weeks (data not shown). Moreover, comparison of young SPF cats and old SPF 279

animals revealed that 2 of the old animals had values slightly exceeding the cutoff 280

(data not shown). These 2 cats were over 10 years old and had lived for many 281

years in a separate room where they served as blood donors. It can be imagined 282

that during their handling for many years by persons who also had access to 283

clinics, they had become exposed to low doses of FeLV and as a consequence 284

showed seroconversion. Unfortunately, this postulate could not be verified as the 285

cats still serve as blood donors. 286

287

Antibody levels in privately owned naturally infected cats 288

Cats that were naturally infected with FeLV were evaluated with ROC analyses 289

using provirus as gold standard (Fig. 3a). Results revealed a good course for 290

antigen p15E, representing a striking contrast to the other antigens. Because the 291

values of these samples were generally increased, new cutoff points for the 292

15

datasets had to be determined. To simplify, 8 cats were excluded from this study 293

that were, besides being provirus positive also immunized indicating that they 294

might have been immunized after FeLV infection, rendering their immune state 295

highly undefined. The optimal cutoffs were defined as specificity ≥ 80% and 296

sensitivity x specificity as maximum (Fig. 3b) which was calculated to be OD= 297

0.163 for p15E, resulting in a sensitivity of 77.1% and a specificity of 85.6% 298

(specificity ≥ 80%). FL-74 displayed with a cutoff of 0.647 a sensitivity of 42.9% 299

and a specificity of 80.1%, p45 with a cutoff of 0.531 a sensitivity of 40% and 300

specificity of 81.1%, and EPK211 had a sensitivity of 17.1% and a specificity of 301

84.6% with a cutoff of 2.411. 302

Cohen’s κ values (specificity of ≥ 80%) revealed lower levels compared to κ-303

values of experimentally infected cats, which was consistent with the differences 304

between the two sets of data evaluated with ROC analyses. The best agreement 305

level was achieved by p15E plotted against provirus (κ = 0.55) (Table 1). In 306

marked contrast, the other pairs of antigens and pairs of provirus and antigens 307

reached levels of only 0.08 to 0.42 and thus were clearly unsatisfactory. These 308

results indicated that the agreement between p15E and provirus is much better 309

than combinations of p15E with the other antigens. 310

311

Antibody levels in experimentally vaccinated cats 312

To look also at the antibody levels of vaccinated animals, sera from young, 8 313

weeks old SPF cats were tested once before vaccination, and again after they 314

had been vaccinated twice but before challenge infection. Results (Fig. 4) for p45 315

revealed that of all cats (without Placebo), the overall exceed of the cutoff was 316

16

56.92%, but 100% were over the cutoff when immunized with Leucogen®, as 317

expected. Whole FeLV antigen (FL-74) showed an overall increase in antibody 318

responses with the finding that 90.77% exceeded the cutoff, and 100% when 319

Leucogen®-vaccinated. In contrast, EPK211 exhibited that 58.46% were over the 320

cutoff with 40% of the cats vaccinated with Leucogen®. Recombinant p15E 321

revealed that 44.62% exceeded the cutoff, and 13.3% of Leucogen® vaccinated 322

animals. 323

324

Antibody levels in privately owned, vaccinated cats 325

Out of all privately owned provirus negative cats, 50 animals were vaccinated. 8 326

cats were excluded from this study, because they were, besides having received 327

immunization, also provirus positive or even viremic. 98% of all animals 328

considered for the study received Leucogen®, mostly applied in combination with 329

Feligen® (Virbac Schweiz AG, Glattbrugg, Switzerland; against feline 330

panleukopenia and feline rhinotracheitis/cat flu). Results (Fig. 5) revealed that 331

antibody reaction was clearly increased using antigens p45 (70%), whole virus 332

(42%), and EPK211 (24%). Whole TM (p15E) showed a low reactivity to 333

vaccinated cats and exceeded in 16% the cutoff point. 334

335

Discussion 336

This study showed that for serological diagnosis, the FeLV envelope TM protein 337

p15E is a promising candidate. The development of a novel diagnostic test based 338

on a specific FeLV antibody needs to define the critical level the antibody is 339

supposed to exceed to reliably predict any contact with FeLV, or to fall below to 340

17

predict naïvety against FeLV. Thus, this study aimed at screening FeLV-naïve, 341

infected and immunized cats for the presence of antibodies to four different FeLV 342

antigens. Based on our results, we defined a threshold to discriminate cats that 343

had contact with FeLV from naïve cats. Furthermore, we used a well-established 344

gold standard to compare our ELISA with, the real-time PCR. 345

Experimentally infected and vaccinated cats provided first results for a proof-of-346

principle study. We chose laboratory cats, because they have a known infection 347

history and had been living in a defined SPF environment avoiding all external 348

influences; moreover, they are of defined age, gender, and breed. The time 349

points when to challenge them with FeLV, or to apply immunization, can be 350

chosen exactly, as well as the time points of sample collection. The tests revealed 351

that the antigen with the highest diagnostic potential was p15E with a diagnostic 352

sensitivity of 95.7% and a specificity of 100%. The other three antigens EPK211, 353

FL-74 and p45 were not suitable in this context. 354

To evaluate the diagnostic efficiency of the antigens in naturally exposed field 355

cats, the cats status of viremia (p27), provirus (PCR), as well as immunization 356

record were used. In the field cat population, however, many factors were often 357

not controllable- or were even completely unknown due to increased interactions 358

of these animals with their environment. Such factors include the cat’s age, 359

possible multiple immunizations against other viruses than FeLV, and present or 360

past diseases of any nature. Another unclear point is that we did not have any 361

records of how many times these cats were exposed to FeLV and to different 362

FeLV subtypes, especially if these animals were stray cats or originally from a cat 363

shelter. All these factors result in a more undefined outcome of the cat’s immune 364

18

status. For example, we had to exclude five animals that were vaccinated, but 365

provirus positive (p27 negative), and three animals that were vaccinated but p27 366

positive (viremic). Here, it was not clear whether the five animals were immune 367

due to infection or due to vaccination, and whether the three p27 positive animals 368

became viremic because of vaccine failure, or whether they were vaccinated after 369

they had already established viremia. 370

As expected, the threshold value for these samples was elevated; the diagnostic 371

sensitivity and specificity were lower than in the defined laboratory cat 372

populations. 373

Considering the more complex conditions in field cats, p15E revealed to have 374

good diagnostic sensitivity of 77.1% and specificity of 85.6%. The relatively low 375

specificity of 85.6% would probably be much higher if PCR –which had been used 376

as gold standard of infection- was more sensitive: Exposure to FeLV can result in 377

infection and seroconversion in absence of positive p27 and PCR results 378

performed with plasma or serum and whole blood, resp. (33, 34). In these 2 379

studies (33, 34) it was shown that -in contrast to experimental infection by needle- 380

nasal or peroral exposure to FeLV can result in seroconversion and infection of 381

several organs with little involvement of the bone marrow. If PCR results had 382

been available from organs in the privately owned cats of the present study, the 383

specificity had been much higher. Thus, demonstration of antibodies to p 15E 384

appears to be a more sensitive parameter for past contact with FeLV than the 385

most sensitive PCR procedures performed with blood. Our results suggest that 386

detection of antibodies to p 15E may become relevant as a future diagnostic 387

parameter for exposure to FeLV. As was shown for the privately owned cats, the 388

19

cutoffs had to be changed from the conditions of the experimentally infected cats 389

in order to reach an optimal tradeoff between diagnostic sensitivity and specificity. 390

Should an antibody test based on p15E be developed for use in clinical veterinary 391

medicine, larger numbers of privately owned cats would have to be evaluated in 392

order to define an optimal cutoff. Our findings are in line with the results of an 393

earlier publication in which antibodies to FeLV p15E were found to be present at 394

significant levels after FeLV infection (22). On the other hand, the results are in 395

striking contrast to those of Fontenot (27) or Langhammer (25). It is postulated 396

that the discrepancy with the results of Fontenot et al. and Langhammer et al. can 397

be explained by the fact that these authors performed their tests with synthetic 398

peptides which may differ from the threedimensional shape of the entire protein 399

and therefore may not have been completely recognized by the antibodies 400

induced by the viral protein. 401

Cohen's Kappa values were used as a measure of agreement between two 402

antigens and to evaluate the new ELISA compared to a well-established 403

reference test, the real time PCR, which served as a gold standard. A high kappa 404

value shows that there is strong agreement between two tests, however, without 405

a statement about the quality of the tests. The low agreement among the four 406

antigens was in striking contrast to the almost perfect agreement between p15E 407

and provirus in experimentally infected cats, and good agreement between p15E 408

and provirus in the field cats. It is possible that these results may represent a step 409

towards a partial replacement of the more expensive PCR test. 410

We were expecting that in some cats presence of antibodies to enFeLV may 411

interfere with results induced by infection with exogenous FeLV. However, the 412

20

p15E ELISA did not show elevated antibody values in any of the FeLV naïve SPF 413

cats, indicating that the test is not negatively affected by antibodies to enFeLV 414

which is present in multiple copies in every cat cell (24). However, it has to be 415

considered that certain cat breeds might have increased levels of antibodies to 416

enFeLV. 417

We could show that almost all cats seroconverted after they had contact with 418

FeLV, and we were able to demonstrate with p15E positive results that the 419

probability of a cat being infected with FeLV is high also in p27 negative cats. 420

However, p15E was not able to discriminate between viremic and immune cats 421

indicating that p27 is further needed to diagnose viremia. A point-of-care test 422

device in which a p27 antigen test is located on a lateral flow test membrane next 423

to a p 15E test field could readily be imagined. Thus, a cat with negative p27 and 424

p15E antibody results is very likely to have never been in contact with FeLV. In 425

contrast, a negative p27 and positive p15E result suggests that the cat is not 426

viremic/antigenemic but had been in contact with FeLV and may harbour 427

somewhere in its body the virus. Before introduction of such a cat into a multicat-428

household situation the new cat should be further tested for signs of a latent 429

FeLV infection at least by PCR and/or RT-PCR performed on blood. If blood is 430

negative by PCR or RT-PCR, latent virus load may indeed be low. 431

An antibody test not only may detect infected and naïve cats, but also immunized 432

cats. A cat with a p15E value exceeding the cutoff is considered to have been 433

exposed to FeLV (if p27 ELISA is negative). As a positive p15E result may not be 434

identical with immunity to FeLV, cats with positive or negative p15E results should 435

be vaccinated where needed. 436

21

It was the goal of this study to develop a novel serological test that is able to 437

show whether or not a cat had contact to FeLV. To this end, we have evaluated 438

different FeLV antigens. FeLV p15E was found to be useful to differentiate 439

infected from uninfected animals. However, it was found not to be useful to clearly 440

detect vaccination, because most of the vaccinated privately owned cats had 441

antibody values that are lower than the threshold calculated for FeLV naïve cats. 442

Thus, p15E seems to be rather a sign for infection than for vaccination, which is 443

consistent with previous findings (22). Based on our results, we conclude that 444

serological diagnosis based on the p15E antigen may represent a valuable 445

support in evaluating the infection state of a cat and partially replace elaborate or 446

expensive diagnostic techniques like PCR. 447

448

Author Contributions 449

HL initiated the study, EB and HL conceived and designed the experiments. EB 450

carried out the experiments. JH designed the experiments for p15E synthesis. SH 451

performed the statistical analyses. All authors analyzed the data. All authors 452

wrote the paper. All authors read and approved the final manuscript. 453

454

Competing financial interest 455

The authors declare no competing interest. 456

457

Acknowledgements 458

The authors thank M. Wittenbrink for providing the E. coli sera, A. Gomes-Keller, 459

C. Geret, Pfizer (Kent, United Kingdom), Merial (Lyon, France) and Virbac 460

22

(Carros, France) for kindly providing the cat plasma and sera samples. They want 461

to thank S. Wellnitz (RedBiotec) and O. Sonzogni for technical support. They 462

thank T. Meili Prodan for synthesizing p15E, and EspiKem Srl (Florence, Italy) for 463

the synthesis of the peptide EPK211. They are grateful that many veterinarians in 464

Switzerland generously provided informations about the vaccination state of the 465

privately owned cats. 466

This work was supported by a grant obtained from the Promedica Foundation, 467

Chur, Switzerland. The study was performed using the logistics of the Centre for 468

Clinical Studies at the Vetsuisse Faculty, University of Zurich, Switzerland. 469

470

References 471

1. Jarrett WF, Crawford EM, Martin WB, Davie F. 1964. A Virus-Like 472

Particle Associated with Leukemia (Lymphosarcoma). Nature 202:567-473

569. 474

2. Hoover EA, Mullins JI. 1991. Feline leukemia virus infection and 475

diseases. J Am Vet Med Assoc 199:1287-1297. 476

3. Hardy WD, Jr., McClelland AJ. 1977. Feline leukemia virus. Its related 477

diseases and control. The Veterinary clinics of North America 7:93-103. 478

4. Levy JK, Scott HM, Lachtara JL, Crawford PC. 2006. Seroprevalence of 479

feline leukemia virus and feline immunodeficiency virus infection among 480

cats in North America and risk factors for seropositivity. J Am Vet Med 481

Assoc 228:371-376. 482

5. Lutz H, Lehmann R, Winkler G, Kottwitz B, Dittmer A, Wolfensberger 483

C, Arnold P. 1990. [Feline immunodeficiency virus in Switzerland: clinical 484

23

aspects and epidemiology in comparison with feline leukemia virus and 485

coronaviruses]. Schweizer Archiv fur Tierheilkunde 132:217-225. 486

6. Gomes-Keller MA, Gonczi E, Grenacher B, Tandon R, Hofman-487

Lehmann R, Lutz H. 2009. Fecal shedding of infectious feline leukemia 488

virus and its nucleic acids: a transmission potential. Veterinary 489

microbiology 134:208-217. 490

7. Major A, Cattori V, Boenzli E, Riond B, Ossent P, Meli ML, Hofmann-491

Lehmann R, Lutz H. 2010. Exposure of cats to low doses of FeLV: 492

seroconversion as the sole parameter of infection. Veterinary research 493

41:17. 494

8. Hofmann-Lehmann R, Tandon R, Boretti FS, Meli ML, Willi B, Cattori 495

V, Gomes-Keller MA, Ossent P, Golder MC, Flynn JN, Lutz H. 2006. 496

Reassessment of feline leukaemia virus (FeLV) vaccines with novel 497

sensitive molecular assays. Vaccine 24:1087-1094. 498

9. Anderson MM, Lauring AS, Burns CC, Overbaugh J. 2000. 499

Identification of a cellular cofactor required for infection by feline leukemia 500

virus. Science 287:1828-1830. 501

10. Grant CK, Essex M, Gardner MB, Hardy WD, Jr. 1980. Natural feline 502

leukemia virus infection and the immune response of cats of different ages. 503

Cancer research 40:823-829. 504

11. Lutz H, Pedersen NC, Theilen GH. 1983. Course of feline leukemia virus 505

infection and its detection by enzyme-linked immunosorbent assay and 506

monoclonal antibodies. American journal of veterinary research 44:2054-507

2059. 508

24

12. Hardy WD, Zuckerman EE. 1991. Development of the Immunofluorescent 509

Antibody-Test for Detection of Feline Leukemia-Virus Infection in Cats. J. 510

Am. Vet. Med. Assoc. 199:1327-1335. 511

13. Hardy WD, Zuckerman EE. 1991. 10-Year Study Comparing Enzyme-512

Linked-Immunosorbent-Assay with the Immunofluorescent Antibody-Test 513

for Detection of Feline Leukemia-Virus Infection in Cats. J. Am. Vet. Med. 514

Assoc. 199:1365-1373. 515

14. Jarrett O, Ganiere JP. 1996. Comparative studies of the efficacy of a 516

recombinant feline leukaemia virus vaccine. The Veterinary record 138:7-517

11. 518

15. Hofmann-Lehmann R, Huder JB, Gruber S, Boretti F, Sigrist B, Lutz H. 519

2001. Feline leukaemia provirus load during the course of experimental 520

infection and in naturally infected cats. The Journal of general virology 521

82:1589-1596. 522

16. Herring IP, Troy GC, Toth TE, Champagne ES, Pickett JP, Haines DM. 523

2001. Feline leukemia virus detection in corneal tissues of cats by 524

polymerase chain reaction and immunohistochemistry. Vet. Ophtalmol. 525

4:119-126. 526

17. Jackson ML, Haines DM, Taylor SM, Misra V. 1996. Feline leukemia 527

virus detection by ELISA and PCR in peripheral blood from 68 cats with 528

high, moderate, or low suspicion of having FeLV-related disease. J Vet 529

Diagn Invest 8:25-30. 530

25

18. Essex M. 1977. Immunity to leukemia, lymphoma, and fibrosarcoma in 531

cats: a case for immunosurveillance. Contemporary topics in 532

immunobiology 6:71-106. 533

19. Stephenson JR, Khan AS, Sliski AH, Essex M. 1977. Feline 534

Oncornavirus-Associated Cell-Membrane Antigen - Evidence for an 535

Immunologically Cross-Reactive Feline Sarcoma Virus-Coded Protein. P 536

Natl Acad Sci USA 74:5608-5612. 537

20. Denoronha F, Schafer W, Essex M, Bolognesi DP. 1978. Influence of 538

Antisera to Oncornavirus Glycoprotein (Gp71) on Infections of Cats with 539

Feline Leukemia-Virus. Virology 85:617-621. 540

21. Jacquemin PC, Saxinger C, Gallo RC, Hardy WD, Essex M. 1978. 541

Antibody-Response in Cats to Feline Leukemia-Virus Reverse-542

Transcriptase under Natural Conditions of Exposure to the Virus. Virology 543

91:472-476. 544

22. Lutz H, Pedersen N, Higgins J, Hubscher U, Troy FA, Theilen GH. 545

1980. Humoral Immune Reactivity to Feline Leukemia-Virus and 546

Associated Antigens in Cats Naturally Infected with Feline Leukemia-Virus. 547

Cancer research 40:3642-3651. 548

23. Tandon R, Cattori V, Willi B, Lutz H, Hofmann-Lehmann R. 2008. 549

Quantification of endogenous and exogenous feline leukemia virus 550

sequences by real-time PCR assays. Veterinary immunology and 551

immunopathology 123:129-133. 552

24. Tandon R, Cattori V, Willi B, Meli ML, Gomes-Keller MA, Lutz H, 553

Hofmann-Lehmann R. 2007. Copy number polymorphism of endogenous 554

26

feline leukemia virus-like sequences. Molecular and cellular probes 555

21:257-266. 556

25. Langhammer S, Hubner J, Kurth R, Denner J. 2006. Antibodies 557

neutralizing feline leukaemia virus (FeLV) in cats immunized with the 558

transmembrane envelope protein p15E. Immunology 117:229-237. 559

26. Tandon R, Cattori V, Willi B, Lutz H, Hofmann-Lehmann R. 2008. 560

Quantification of endogenous and exogenous feline leukemia virus 561

sequences by real-time PCR assays. Vet Immunol Immunop 123:129-133. 562

27. Fontenot JD, Hoover EA, Elder JH, Montelaro RC. 1992. Evaluation of 563

Feline Immunodeficiency Virus and Feline Leukemia-Virus 564

Transmembrane Peptides for Serological Diagnosis. J. Clin. Microbiol. 565

30:1885-1890. 566

28. Troy FA, Fenyo EM, Klein G. 1977. Moloney Leukemia Virus-Induced 567

Cell-Surface Antigen - Detection and Characterization in Sodium Dodecyl-568

Sulfate Gels. P Natl Acad Sci USA 74:5270-5274. 569

29. Jarrett O, Laird HM, Hay D. 1973. Determinants of Host Range O Feline 570

Leukemia Viruses. The Journal of general virology 20:169-175. 571

30. Engvall E, Perlmann P. 1972. Enzyme-Linked Immunosorbent Assay, 572

Elisa .3. Quantitation of Specific Antibodies by Enzyme-Labeled Anti-573

Immunoglobulin in Antigen-Coated Tubes. J Immunol 109:129-&. 574

31. Voller A, Bidwell DE, Bartlett A. 1976. Enzyme immunoassays in 575

diagnostic medicine. Theory and practice. Bulletin of the World Health 576

Organization 53:55-65. 577

27

32. Lehmann R, Franchini M, Aubert A, Wolfensberger C, Cronier J, Lutz 578

H. 1991. Vaccination of cats experimentally infected with feline 579

immunodeficiency virus, using a recombinant feline leukemia virus 580

vaccine. J. Am. Vet. Med. Assoc. 199:1446-1452. 581

33. Gomes-Keller MA, Gonczi E, Grenacher B, Tandon R, Hofman-582

Lehmann R, Lutz H. 2009. Fecal shedding of infectious feline leukemia 583

virus and its nucleic acids: a transmission potential. Veterinary 584

microbiology 134:208-217. 585

34. Major A, Cattori V, Boenzli E, Riond B, Ossent P, Meli ML, Hofmann-586

Lehmann R, Lutz H. 2010. Exposure of cats to low doses of FeLV: 587

seroconversion as the sole parameter of infection. Veterinary research 588

41:17. 589

590

591

592

593

594

28

Tables and Figures 595

Table 1.Cross Table of Cohen’s κ-values 597

Cross table of κ (kappa)-values showing the agreement levels of antigen pairs 598

and provirus-antigen pairs. The gray background indicates experimentally 599

infected cats (ntot= 157) and the white background indicates naturally infected 600

cats (ntot= 294). The specificity is >80%. 601

602

603

604

605

606

607

608

609

610

611

612

613

614

615

616

617

618

619

SP>80% Provirus p45 FL-74 EPK211 p15E

Provirus 0.18 0.13 0.08 0.55

p45 0.69 0.42 0.16 0.21

FL-74 0.47 0.45 0.24 0.12

EPK211 0.47 0.57 0.42 0.06

p15E 0.96 0.70 0.51 0.48

29

Figure legends 620

621

Figure 1. Purity of recombinant p15E 622

Mean OD values of a positive and a negative cat control serum and three E. coli 623

sera (nr. 1-3) tested with antibody ELISA using the recombinant p15E cloning 624

product as antigen. Compared to the positive and negative control, E. coli sera 1-625

3 reached only low OD levels (< 0.117). 626

627

Figure 2a. Provirus ROC curve for experimentally infected cats 628

Empirical Receiver operating characteristic (ROC) curve of provirus for 629

experimentally infected cats including specific pathogen-free (SPF) cats (n= 94) 630

and provirus positive (p27 positive and p27 negative) cats (n= 70). The true 631

positive rate (sensitivity, y-axis) is plotted against the false positive rate (1-632

specificity, x-axis) at various cutoff points. 633

634

Figure 2b. Determination of the optimal cutoff for experimentally infected 635

cats 636

The sensitivity x specificity (y-axis) is plotted against the cutoff (x-axis). Dark lines 637

represent specificities ≥ 80%. The arrows indicate the cutoff points for each 638

antigen. 639

640

Figure 3a. Provirus ROC curve for naturally infected cats 641

Empirical Receiver operating characteristic (ROC) curve of provirus for naturally 642

infected study cats including uninfected cats (n= 201) and provirus positive (p27 643

30

positive and p27 negative) cats (n= 35). The true positive rate (sensitivity, y-axis) 644

is plotted against the false positive rate (1-specificity, x-axis) at various cutoff 645

points. 646

647

Figure 3b. Determination of the optimal cutoff for naturally infected cats 648

The sensitivity x specificity (y-axis) is plotted against the cutoff (x-axis). Dark lines 649

represent specificities ≥ 80%. The arrows indicate the cutoff points for each 650

antigen. 651

652

Figure 4. Antibody response of experimentally vaccinated cats 653

The y-axes represent the relative OD values; the x-axes represent the vaccines 654

used in the study. All four antigens were tested. The gray bars indicate 8 weeks 655

old specific pathogen-free (SPF) cats (n= 63) before immunization; the white bars 656

indicate the same cats (n= 63) after two immunizations and before challenge. 657

658

Figure 5. Antibody response of privately owned vaccinated cats 659

The y-axis represents the relative OD values, the x-axis the four antigens. The 660

gray bars indicate cats without vaccination (n= 236); the white bars indicate cats 661

that were vaccinated with Leucogen® (98%) (n= 50). The 8 cats found to be p27 662

and/or provirus positive are not included. 663

664

Fig. 1

Fig. 2

Fig. 3a

Fig. 3b

Fig. 3

Fig. 4

Fig. 5