- Email: [email protected]
Prevalence of Toxoplasma gondii parasite in captive Mexican jaguars determined by recombinant surface antigens (SAG1) and dense granular antigens (GRA1 and GRA7) in ELISA-based serodiagnosis
Journal Pre-proof Prevalence of Toxoplasma gondii parasite in captive Mexican jaguars determined by recombinant surface antigens (SAG1) and dense granular antigens (GRA1 and GRA7) in ELISA-based serodiagnosis Alejandro Reynoso-Palomar, Dulce Moreno-Gálvez, Abel Villa-Mancera PII:
S0014-4894(19)30381-9
DOI:
https://doi.org/10.1016/j.exppara.2019.107791
Reference:
YEXPR 107791
To appear in:
Experimental Parasitology
Received Date: 29 August 2019 Revised Date:
23 October 2019
Accepted Date: 4 November 2019
Please cite this article as: Reynoso-Palomar, A., Moreno-Gálvez, D., Villa-Mancera, A., Prevalence of Toxoplasma gondii parasite in captive Mexican jaguars determined by recombinant surface antigens (SAG1) and dense granular antigens (GRA1 and GRA7) in ELISA-based serodiagnosis, Experimental Parasitology (2019), doi: https://doi.org/10.1016/j.exppara.2019.107791. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
1
Prevalence of Toxoplasma gondii parasite in captive Mexican jaguars determined by
2
recombinant surface antigens (SAG1) and dense granular antigens (GRA1 and
3
GRA7) in ELISA-based serodiagnosis
4
Alejandro Reynoso-Palomara, Dulce Moreno-Gálveza, Abel Villa-Manceraa*
5 6 7
a
8
Puebla, 4 Sur 304 Col. Centro, C.P. 75482, Tecamachalco Puebla, México.
9
*Corresponding author. E-mail address: [email protected] (A. Villa-Mancera)
Facultad de Medicina Veterinaria y Zootecnia, Benemérita Universidad Autónoma de
10 11
Abstract
12
Toxoplasma gondii is an obligate intracellular protozoan parasite that infects almost
13
all warm-blooded animals, including humans, causing serious public health problems. In
14
this study, the seroprevalence of T. gondii in captive jaguars in 10 Mexican zoos was
15
determined using single and mixtures of recombinant surface antigens (SAG1) and dense
16
granular antigens (GRA1 and GRA7) in immunoglobulin G (IgG) enzyme-linked
17
immunosorbent assays (ELISAs). Their efficacy was compared with the tachyzoite lysate
18
antigen. All recombinant antigens were characterised by high sensitivity (92.5– 97.5%); the
19
specificity of the IgG ELISAs was variable (83.3– 91.6%). Mixtures of the two
20
recombinant proteins were generally more reactive than single antigens. GRA7 + SAG1
21
showed the highest sensitivity (97.5%) and specificity (91.6%), almost perfect agreement
22
(96.2%), and a kappa value of 0.89. An area under the curve value of 0.998 represented a
23
highly accurate test with a cutoff value of 0.8. The seroprevalence of anti-T. gondii IgG
24
antibodies in the single and mixed recombinant antigen ELISAs was 75.0–76.9%. This
25
study shows that GRA7 + SAG1 can be successfully used to diagnose T. gondii infection in
26
jaguars for effective monitoring of prevalence and for devising control methods and
27
prevention strategies against toxoplasmosis.
28 29 30 31
Keywords:
Toxoplasma
gondii;
serodiagnosis; prevalence; zoo; Mexico
jaguars;
ELISA;
recombinant
antigens;
32
1. Introduction
33
Toxoplasmosis is a globally prevalent zoonotic disease caused by the protozoan
34
Toxoplasma gondii (Apicomplexa; Sarcocystidae) which can affect most warm-blooded
35
animals, including humans. Humans and animals usually become infected through
36
ingestion of undercooked or raw meat containing cysts, or by consuming food
37
contaminated with sporulated oocysts (Zhou et al., 2011). Members of the family Felidae
38
are definitive and intermediate hosts of this intracellular parasite, and play a key role in
39
transmission of the infection (Dubey, 2009; Calero-Bernal and Gennari, 2019). Captive
40
jaguars may serve as source of infection for susceptible animal species and human beings
41
sharing the same environment, including zoos (Ullmann et al., 2010).
42
To estimate prevalence more accurately, sensitive and specific diagnostic tests for T.
43
gondii infection need to be performed. Serological diagnosis by enzyme-linked
44
immunosorbent assay (ELISA) for immunoglobulin G (IgG) detection is a useful and
45
convenient tool (Liu et al., 2015). The tachyzoite propagated in vivo or in vitro, has been
46
traditionally used in serological detection of toxoplasmosis, and is characterised by high
47
sensitivity and specificity. However, it has several disadvantages: it is time-consuming and
48
methodologically demanding to produce; inconsistent in quality; subject to contamination
49
with extraparasitic material; and its production may expose laboratory technicians to
50
harmful living parasites (Gatkowska et al., 2006; Ferra et al., 2019). Currently, the lack of
51
continuous availability and abundant quantities of native antigens for diagnostic testing is
52
an obstacle. Therefore, the use of purified recombinant protein is desirable, with the
53
advantage of reducing production costs and the possibility of using more of than one
54
defined antigen for detection of specific antibodies (Rostami et al., 2018; Ferra et al.,
55
2019).
56
In a number of studies, toxoplasmosis in humans and animals has been diagnosed
57
using different recombinant proteins, such as surface proteins (SAGs), microneme proteins
58
(MICs), rhoptry proteins (ROPs), and dense granule proteins (GRAs; Sudan et al., 2013;
59
Wang and Yin, 2014; Rostami et al., 2018). Among the different surface antigen molecules,
60
SAG1 (also named P30) is highly abundant and the most immunodominant and stage-
61
specific antigen on tachyzoites present during the acute and chronic phases of the disease
62
(Bel-Ochi et al., 2013; Wang and Yin, 2014). GRAs are secreted in abundance, and are
63
involved in parasite survival and virulence. They are also a major component of the
64
parasitiphorous vacuole, in which tachyzoites multiply, and the cyst wall surrounding
65
slower growing bradyzoites (Cesbron-Delauw, 1994; Mercier et al., 2005). GRA1 is
66
important for diagnosis of chronic infection, while GRA7 has direct contact with the host
67
immune system, and induces strong antibody responses in the early and late stages of
68
parasite infection (Jacobs et al., 1998, 1999; Wang et al., 2013). In this study, we evaluated
69
the performance of single and different combinations of recombinant antigen proteins
70
(GRA1, GRA7, and SAG1) in ELISA-based serodiagnoses to determine the prevalence of
71
T. gondii infection in jaguars in ten Mexican zoos.
72 73
2. Materials and methods
74 75
2.1. Location and sample collection
76
Serum samples from 52 jaguars (Panthera onca; Felidae), including 28 males and 24
77
females, were collected from 10 Mexican zoos, located in six of Mexico's 32 states
78
(Michoacán, Puebla, Oaxaca, Morelos, Sinaloa, and the State of Mexico). Jaguars were
79
immobilised with an intramuscular injection of tiletamine-zolazepam or ketamine plus
80
xylazine. Blood samples were collected from the jugular, cephalic, or saphenous vein. All
81
serum samples were obtained between January and May 2018, and were centrifuged at
82
2000 g for 15 min at 4 °C, and the supernatant was stored at −20 °C for ELISA analyses.
83
The jaguars’ diet consisted mainly of chicken carcasses, and other animal products (horse,
84
bovine, and rabbit meat) when available. Water was provided ad libitum. This study was
85
approved by the Animal Care and Ethics Committee of Meritorious Autonomous
86
University of Puebla, and all procedures complied with national legislation pertaining to
87
Animal Health Research.
88 89
2.2. Negative and positive serum samples
90
A commercial IgG ELISA (DRG International, USA; EIA-1798) based on tachyzoite
91
lysate antigen (TLA) was modified with horseradish peroxidase (HRP)-conjugated goat
92
anti-cat IgG (Sigma-Aldrich, USA) to replace the conjugated anti-human IgG included in
93
the kit (sensitivity 98.3% and specificity 99.2%). The serum ELISA test was selected as the
94
reference standard to evaluate the performance of the recombinant antigen ELISAs.
95
Confirmatory
96
chemiluminescent immunoassay (MexLab Group, Mexico) following the manufacturer’s
97
instructions using T. gondii antigen and FITC-conjugated goat anti-cat IgG antibody
98
(Sigma-Aldrich, USA) as secondary antibody.
testing
for
negative
and
positive
samples
was
performed
by
99 100
2.3. Detection of antibodies to Toxoplasma gondii by indirect ELISAs using recombinant
101
antigens
102
Optimum dilutions of recombinant coating antigens, jaguar serum samples, and HRP-
103
conjugated goat anti-cat IgG were established by systematic chequerboard titrations. Flat-
104
bottomed 96-well microtiter plates (Costar, Corning, NY, USA) were coated with 100 μl of
105
GRA1: expressed in E. coli, UniProt P13403 (ab43137, Abcam), SAG1 produced in E. coli,
106
containing the GST-tag (amino acids 45-198; ab68110, Abcam), GRA7: expressed in E.
107
coli (amino acids 24-100; ab73748, Abcam), and a mixture of these proteins at a final
108
concentration of 5 μg/ml for each recombinant protein in the coating buffer (15 mM
109
Na2CO3, 35 mM NaHCO3, pH 9.6), and incubated overnight at 4 °C. After three washes
110
with phosphate-buffered saline (PBS) containing 0.05% Tween 20 (PBS-T), nonspecific
111
binding sites were blocked with a reconstituted 3% w/v solution of non-fat milk powder
112
(Amresco, USA) prepared in PBS for 2 h at 37 °C. The plates were then washed again three
113
times, and negative and positive serum samples were used at a dilution of 1:50 in PBS and
114
incubated at 37 °C for 1 h. After another wash, incubation was carried out for 1 h at 37 °C,
115
with the goat anti-cat IgG-HRP diluted at 1:2500 in 3% non-fat milk in PBS. Following
116
incubation and washing, colour was developed by the addition of 100 μl per well of TMB
117
substrate solution (Sigma-Aldrich, USA), and stopped with 50 μl of 4N H2SO4. Finally,
118
optical density (OD) was measured at 450 nm in an ELISA reader (BioTek ELx800,
119
BioTek).
120
Each serum sample was assayed twice, and the results were determined for each
121
sample by calculating the mean OD reading of duplicate wells. The cutoff point for a
122
positive result was defined as any value higher than the average OD reading plus two
123
standard deviations obtained with negative serum samples.
124 125
2.4. Statistical analysis
126
The performance of each recombinant antigen ELISA was compared with the
127
commercial ELISA as a standard test for determination of sensitivity, specificity, and kappa
128
value. We also used McNemar’s test to calculate the relationships between the two assay
129
methods.
130
The optimal cutoff value for each recombinant antigen ELISA was established by
131
receiver operating characteristic (ROC) curve analysis with a 95% confidence interval (CI).
132
The accuracy of area under the curve (AUC) values proposed by Swets (1988) were
133
considered as follows: non-informative (AUC < 0.5); low accuracy (0.5 < AUC < 0.7),
134
moderate accuracy (0.7 < AUC < 0.9); and high accuracy (0.9 < AUC < 1). The relative
135
diagnostic sensitivity and specificity of the ELISAs were evaluated for different cutoffs. All
136
statistical analyses were carried out using IBM SPSS 20 software package (SPSS Inc.,
137
Chicago, USA) for Windows.
138 139
3. Results
140 141
3.1. Evaluation of diagnostic performance of commercial ELISA and ELISAs using
142
recombinant GRA and SAG antigens
143
The commercial ELISA was carried out initially on the 52 jaguar samples for the
144
purpose of comparing it with the ELISAs using single and mixed GRAs and SAGs. Forty
145
samples were positive in the comparison with commercial ELISA (Table 1) and
146
chemiluminescence immunoassay (sensitivity 94% and specificity 100%). As shown in
147
Table 1, there were 39 positive and 13 negative samples for GRA1, SAG1 and GRA1 +
148
SAG1, and 40 positive and 12 negative samples for GRA7, GRA1 + GRA7, and GRA7 +
149
SAG1.
150
The study recorded the highest mean OD value of 0.99 (95% CI: 0.98–1.04) for the
151
antigen mixture GRA7 + SAG1, followed by 0.95 for GRA7 (95% CI: 0.90–0.99), 0.94 for
152
GRA1 + GRA7 (95% CI: 0.93–0.98), 0.88 for GRA1 + SAG1 (95% CI: 0.81–0.94), 0.84
153
for SAG1 (95% CI: 0.77–0.91), and 0.70 for GRA1 (95% CI: 0.64–0.74). Cutoff values
154
ranged from 0.41 to 0.48 (Fig. 1).
155
The highest prevalence of T. gondii infection (76.9%, 40 of the 52 jaguar serum
156
samples) was obtained with GRA7, GRA1 + GRA7, and GRA7 + SAG1, followed by
157
GRA1 (75.0%, 39/52), SAG1, and GRA1 + SAG1 (Table 2). There was no significant
158
difference in the proportion of positive test results when comparing single and mixed
159
recombinant antigen ELISAs with the commercial ELISA (McNemar’s chi-squared test; p
160
> 0.05; Table 2). Based on the statistical analysis, the agreement was high for GRA7,
161
GRA1 + GRA7, and GRA7 + SAG1 (kappa = 0.89). For GRA1 + SAG1, GRA1, and
162
SAG1, kappa values were 0.84, 0.74, and 0.74, respectively (Table 2).
163
For detection of T. gondii antibodies in jaguars, we used the commercial ELISA as a
164
reference test to validate the sensitivity and specificity of the IgG ELISAs conducted with
165
single and mixed recombinant GRAs and SAGs. GRA7, GRA1 + GRA7, and GRA7 +
166
SAG1 had the highest sensitivity (97.5%) and specificity (91.6%), and the lowest false
167
positive and negative rates (Table 2).
168 169
3.2. ROC analysis
170
ROC curve analysis of single and mixed recombinant antigens revealed an AUC of
171
high accuracy ranging between 0.954 (GRA1) and 0.998 (GRA7 + SAG1; Table 3). The
172
sensitivities of IgG ELISAs estimated from all the 40 positive serum samples were different
173
for the single recombinant antigens, with high specificity obtained for GRA1 and SAG1
174
(Table 3). Higher sensitivity (97.5%) was observed in IgG ELISAs with GRA7, GRA1 +
175
SAG1, GRA1 + GRA7, and GRA7 + SAG1. Definitely lower specificity at 83.3% was
176
observed for antigen mixtures GRA1 + SAG1 and GRA1 + GRA7. The cutoff value at 0.38
177
for GRA7 + SAG1 had the highest relative sensitivity (97.5%) and relative specificity
178
(91.7%) among the six ELISAs.
179 180
4. Discussion
181
In this study, we developed IgG ELISA tests based on recombinant T. gondii antigen
182
proteins (composed of three different kinds of proteins and their mixtures), in reference to
183
the TLA. Bacterial recombinant proteins offer many advantages in immunodiagnosis, as
184
they allow easy standardisation of serological tests, and reduce the cost of production and
185
purification (Pietkiewics et al., 2004, Holec-Gasior, 2013). Some key disadvantages of
186
producing recombinant antigens in E. coli are the efficiency of expression, contamination
187
of purified recombinant protein and incorrect folding (Pietkiewicz et al., 2007; Holec-
188
Gasior, 2013).
189
Based on our results, the sensitivity and specificity of the SAG1 antigen varied in IgG
190
ELISA tests with single or mixtures of recombinant antigens, from a high of 97.5% and
191
91.6%, respectively, for GRA7 + SAG1 to a low of 92.5% and 83.3%, respectively, for
192
SAG1. In humans, SAG1 exhibits similar sensitivity and specificity (93.6% and 92.9%,
193
respectively; Khanaliha et al., 2014). In this study, the diagnostic sensitivity (97.5%) and
194
specificity (91.6%) of GRA1 were higher than those for mink (78.9% and 95.5%,
195
respectively; Gu et al., 2015).
196
A limited number of studies in cats have reported the diagnostic performance of T.
197
gondii GRA7 protein, alone or in combination with other recombinant antigens. Our study
198
found the sensitivity and specificity of GRA7 to be 97.5% and 91.6%, respectively, in
199
comparison with the commercially available ELISA. The single recombinant GRA7 protein
200
performed better than other protein antigens used in this study. Similar results for
201
sensitivity and specificity have been reported in cattle (95.5% and 91.1%, respectively),
202
cats (94.9% and 97.9%, respectively), and mink (84.2% and 99.1%, respectively; Wang et
203
al., 2014; Cai et al., 2015; Gu et al., 2015). Lower values for sensitivity and specificity of
204
GRA7 have also been observed in cats (35.1% and 89.86%, respectively; Abdelbaset et al.,
205
2017).
206
In our evaluation of the diagnostic utility of ELISAs with recombinant antigen
207
mixtures, the highest sensitivity and specificity were obtained for GRA7 + SAG1.
208
Abdelbaset et al. (2017) evaluated different combinations of T. gondii recombinant antigens
209
(SAG, GRA, and MIC) in IgG ELISA using cat serum samples. SAG2 + GRA7 showed
210
90.54% sensitivity and 85.51% specificity, while a five-antigen combination that included
211
GRA7 resulted in 89.19% sensitivity and 95.36% specificity. In humans, similar results
212
have been reported (Jacobs et al., 1999; Pietkiewicz et al., 2004). The increase in sensitivity
213
detection of IgG antibodies in sera by ELISA based on a mixture of two or three
214
recombinant antigens may result because several different epitopes of various proteins can
215
be recognised for specific antibodies from acute and/or chronic T. gondii infection (Ferra et
216
al., 2015). In addition, for discriminating between recent and latent infections, IgG avidity
217
assays using chemiluminescence could be used in the serological-screening strategy for
218
jaguars, with several advantages such as reproducibility, fast and cost effectiveness
219
(Rostami et al., 2018). Further research with a larger number of samples is required for
220
standardized detection of IgG antibodies against T. gondii in jaguars.
221
ROC analysis describes the performance of an ELISA test in terms of diagnostic
222
accuracy as defined by a reference assay. Its value can be interpreted as the probability of
223
the test distinguishing between a randomly chosen positive serum and a randomly chosen
224
negative serum (Swets, 1988). Based on this analysis, the AUC of 0.998 for GRA7 + SAG1
225
suggests that it represents a highly accurate test with good discriminatory capacity for
226
infected jaguars; the cutoff point was 0.38, corresponding to a sensitivity of 97.5% and
227
specificity of 91.7%. Similar AUC results have been reported for SAG1 (0.996;
228
Hosseininejad, 2012) and GRA7 (0.969; Cai et al., 2015) in cats.
229
No published reports are available on the use of recombinant antigen proteins for
230
ELISA-based serodetection of anti-T. gondii IgG antibodies in jaguars from Mexico, and
231
there are few reports on the worldwide prevalence of this parasite. Using recombinant
232
antigen ELISAs, this study shows a high seroprevalence of the parasite, ranging from
233
75.0% to 76.9% in 10 Mexican zoos. In Brazil, high prevalence rates in captive jaguars
234
have also been reported by modified agglutination test (MAT; 63.2%; Silva et al., 2001),
235
indirect immunofluorescent assay (IFAT; 84.6%; André et al., 2010), MAT (100%, free-
236
ranging jaguars; Furtado et al., 2015), and indirect hemagglutination and MAT (63.64%;
237
Silva et al., 2016). However, the prevalence of toxoplasmosis in Mexico is different
238
depending on the test used as the criterion for measurement—serum IgG ELISA (100%),
239
IFAT (41.7%), and amplification of a fragment of the SAG1 gene by nested PCR using
240
blood (0%) and stool (11.1%) samples (Gomez-Rios et al., 2019).
241
Antibodies against T. gondii have previously been detected in Mexico in the animals
242
(chicken, cattle, horses, and rabbits) fed to the captive jaguars included in this survey
243
(Hernandez-Cortazar et al., 2015). Silva et al. (2007) concluded that the consumption of
244
raw meat is an important risk factor for T. gondii infection, and the most effective way to
245
reduce the risk of exposure to the parasite is feeding these animals only with meat
246
previously frozen at −12 °C for a period >7 days.
247 248 249 250
Conclusions
251
The results reported in this study show the high serodiagnostic usefulness of IgG
252
ELISA based on a mixture of recombinant GRA7 and SAG1 antigens for detection of T.
253
gondii infection in jaguars, and probably in other domestic animals. This study
254
demonstrated high prevalence of the parasite in captive jaguars in 10 zoos from six
255
Mexican states. Further studies are required to increase our knowledge of risk factors linked
256
to the transmission of T. gondii from animals to humans, and for generating control
257
strategies and preventive measures against toxoplasmosis.
258 259
Competing interest
260
The authors declare that they have no competing interests.
261 262 263 264
Acknowledgments The authors are grateful to veterinary staff and administrative personnel of the zoos.
References
265 266 267
Abdelbaset, A.E., Alhasan, H., Salman, D., Karram, M.H., Ellah Rushdi, M.A.,
268
Xuenan, X., Igarashi, M., 2017. Evaluation of recombinant antigens in combination and
269
single formula for diagnosis of feline toxoplasmosis. Experimental parasitology 172, 1-4.
270
Andre, M.R., Adania, C.H., Teixeira, R.H., Silva, K.F., Jusi, M.M., Machado, S.T.,
271
de Bortolli, C.P., Falcade, M., Sousa, L., Alegretti, S.M., Felippe, P.A., Machado, R.Z.,
272
2010. Antibodies to Toxoplasma gondii and Neospora caninum in captive neotropical and
273
exotic wild canids and felids. The Journal of parasitology 96, 1007-1009. Bel-Ochi, N.C., Bouratbine, A., Mousli, M., 2013. Enzyme-linked immunosorbent
274 275
assay using recombinant SAG1
antigen
to
detect
Toxoplasma
gondii-specific
276
immunoglobulin G antibodies in human sera and saliva. Clin. Vaccine Immunol. 20, 468-
277
473.
278
Cai, Y., Wang, Z., Li, J., Li, N., Wei, F., Liu, Q., 2015. Evaluation of an indirect
279
ELISA using recombinant granule antigen Gra7 for serodiagnosis of Toxoplasma gondii
280
infection in cats. Journal of Parasitology 101, 37-41.
281 282 283 284 285 286
Calero-Bernal, R., Gennari, S.M., 2019. Clinical Toxoplasmosis in Dogs and Cats: An Update. Frontiers in veterinary science 6, 54. Cesbron-Delauw, M.-F., 1994. Dense-granule organelles of Toxoplasma gondii: their role in the host-parasite relationship. Parasitology today 10, 293-296. Dubey, J., 2009. History of the discovery of the life cycle of Toxoplasma gondii. International journal for parasitology 39, 877-882.
287
Ferra, B., Holec-Gasior, L., Kur, J., 2015. Serodiagnosis of Toxoplasma gondii
288
infection in farm animals (horses, swine, and sheep) by enzyme-linked immunosorbent
289
assay using chimeric antigens. Parasitology international 64, 288-294.
290
Ferra, B.T., Holec-Gasior, L., Gatkowska, J., Dziadek, B., Dzitko, K., Grazlewska,
291
W., Lautenbach, D., 2019. The first study on the usefulness of recombinant tetravalent
292
chimeric proteins containing fragments of SAG2, GRA1, ROP1 and AMA1 antigens in the
293
detection of specific anti-Toxoplasma gondii antibodies in mouse and human sera. PloS one
294
14, e0217866.
295
Furtado, M.M., Gennari, S.M., Ikuta, C.Y., de Almeida Jacomo, A.T., de Morais,
296
Z.M., de Jesus Pena, H.F., de Oliveira Porfirio, G.E., Silveira, L., Sollmann, R., de Souza,
297
G.O., 2015. Serosurvey of smooth Brucella, Leptospira spp. and Toxoplasma gondii in
298
free-ranging jaguars (Panthera onca) and domestic animals from Brazil. PloS one 10,
299
e0143816.
300
Gatkowska, J., Hiszczynska-Sawicka, E., Kur, J., Holec, L., Dlugonska, H., 2006.
301
Toxoplasma gondii: an evaluation of diagnostic value of recombinant antigens in a murine
302
model. Experimental parasitology 114, 220-227.
303
Gomez-Rios, A., Ortega-Pacheco, A., Gutierrez-Blanco, E., Acosta-Viana, K.Y.,
304
Guzman-Marin, E., Guiris-Andrade, M.D., Hernandez-Cortazar, I.B., Lopez-Alonso, R.,
305
Cruz-Aldan, E., Jimenez-Coello, M., 2019. Toxoplasma gondii in Captive Wild Felids of
306
Mexico: Its Frequency and Capability to Eliminate Oocysts. Vector borne and zoonotic
307
diseases.
308
Gu, Y., Wang, Z., Cai, Y., Li, X., Wei, F., Shang, L., Li, J., Liu, Q., 2015. A
309
comparative study of Toxoplasma gondii seroprevalence in mink using a modified
310
agglutination test, a Western blot, and enzyme-linked immunosorbent assays. Journal of
311
Veterinary Diagnostic Investigation 27, 616-620.
312
Hernandez-Cortazar, I., Acosta-Viana, K.Y., Ortega-Pacheco, A., Guzman-Marin,
313
E.d.S., Aguilar-Caballero, A.J., Jimenez-Coello, M., 2015. Toxoplasmosis in Mexico:
314
epidemiological situation in humans and animals. Revista do Instituto de Medicina Tropical
315
de Sao Paulo 57, 93-103.
316
Holec-Gasior, L., 2013. Toxoplasma gondii recombinant antigens as tools for
317
serodiagnosis of human toxoplasmosis: current status of studies. Clinical and vaccine
318
immunology : CVI 20, 1343-1351.
319
Jacobs, D., Dubremetz, J.-F., Loyens, A., Bosman, F., Saman, E., 1998. Identification
320
and heterologous expression of a new dense granule protein (GRA7) from Toxoplasma
321
gondii. Molecular and biochemical parasitology 91, 237-249.
322
Jacobs, D., Vercammen, M., Saman, E., 1999. Evaluation of recombinant dense
323
granule antigen 7 (GRA7) of Toxoplasma gondii for detection of immunoglobulin G
324
antibodies and analysis of a major antigenic domain. Clin. Diagn. Lab. Immunol. 6, 24-29.
325 326
Liu, Q., Wang, Z.-D., Huang, S.-Y., Zhu, X.-Q., 2015. Diagnosis of toxoplasmosis and typing of Toxoplasma gondii. Parasites & vectors 8, 292.
327
Mercier, C., Adjogble, K.D., Däubener, W., 2005. Dense granules: are they key
328
organelles to help understand the parasitophorous vacuole of all apicomplexa parasites?
329
International journal for parasitology 35, 829-849.
330
Pietkiewicz, H., Hiszczyńska-Sawicka, E., Kur, J., Petersen, E., Nielsen, H.,
331
Stankiewicz, M., Andrzejewska, I., Myjak, P., 2004. Usefulness of Toxoplasma gondii-
332
specific recombinant antigens in serodiagnosis of human toxoplasmosis. Journal of clinical
333
microbiology 42, 1779-1781.
334
Pietkiewicz, H., Hiszczyńska-Sawicka, E., Kur, J., Petersen, E., Nielsen, H.,
335
Stankiewicz, M., Andrzejewska, I., Myjak, P., 2004. Usefulness of Toxoplasma gondii-
336
specific recombinant antigens in serodiagnosis of human toxoplasmosis. Journal of clinical
337
microbiology 42, 1779-1781. Quigley, H., Foster, R., Petracca, L., Payan, E., Salom, R., Harmsen, B., 2017.
338 339
Panthera onca. The IUCN Red List of Threatened Species 2017: e. T15953A50658693. Rostami, A., Karanis, P., Fallahi, S., 2018. Advances in serological, imaging
340 341
techniques and molecular diagnosis of Toxoplasma gondii infection. Infection 46, 303-315.
342
Silva, C.C., Meneses, A.M.C., Moraes, C.C.G.d., Carmo, E.L.d., Langoni, H., Silva,
343
R.C.d., Pancieri, H.M.H., Martins, B.R., Souza, N.F.d., 2016. soroprevalência e pesquisa de
344
oocistos de Toxoplasma gondii em felídeos selvagens procedentes do estado do pará,
345
Brasil. Veterinária e Zootecnia 23, 400-408.
346
Silva, J.C.R., Marvulo, M.F.V., Dias, R.A., Ferreira, F., Amaku, M., Adania, C.H.,
347
Neto, J.S.F., 2007. Risk factors associated with sero-positivity to Toxoplasma gondii in
348
captive neotropical felids from Brazil. Preventive veterinary medicine 78, 286-295.
349
Silva, J.C.R., Ogassawara, S., Adania, C.H., Ferreira, F., Gennari, S.M., Dubey, J.,
350
Ferreira-Neto, J.S., 2001. Seroprevalence of Toxoplasma gondii in captive neotropical
351
felids from Brazil. Veterinary Parasitology 102, 217-224. Sudan, V., Jaiswal, A.K., Shanker, D., 2013. Recent trends in the diagnosis of
352 353
toxoplasmosis. Clinical Reviews and Opinions 5, 11-17. Swets, J.A., 1988. Measuring the accuracy of diagnostic systems. Science 240, 1285-
354 355
1293.
356
Ullmann, L.S., da Silva, R.C., de Moraes, W., Cubas, Z.S., dos Santos, L.C.,
357
Hoffmann, J.L., Moreira, N., Guimaraes, A.M.S., Montaño, P., Langoni, H., 2010.
358
Serological survey of Toxoplasma gondii in captive Neotropical felids from Southern
359
Brazil. Veterinary parasitology 172, 144-146.
360 361
Wang, Y., Wang, G., Zhang, D., Yin, H., Wang, M., 2013. Identification of novel B cell epitopes within Toxoplasma gondii GRA1. Experimental parasitology 135, 606-610.
362
Wang, Z., Ge, W., Li, J., Song, M., Sun, H., Wei, F., Liu, Q., 2014. Production and
363
evaluation of recombinant granule antigen protein GRA7 for serodiagnosis of Toxoplasma
364
gondii infection in cattle. Foodborne pathogens and disease 11, 734-739.
365 366 367
Zhou, P., Chen, Z., Li, H.-L., Zheng, H., He, S., Lin, R.-Q., Zhu, X.-Q., 2011. Toxoplasma gondii infection in humans in China. Parasites & vectors 4, 165.
368
Table 1. Detection of Toxoplasma gondii immunoglobulin G (IgG) in the serum of captive
369
jaguars, tested by enzyme-linked immunosorbent assays (ELISAs) based on single and
370
mixed recombinant dense granule antigens (GRAs) and surface antigens (SAGs), with a
371
commercial ELISA as the reference method.
372 Commercial ELISA Single/mixture Positive Negative Total of recombinant antigens GRA1 Positive 37 2 39 Negative 3 10 13 Total 40 12 52 SAG1 Positive 37 2 39 Negative 3 10 13 Total 40 12 52 GRA7 Positive 39 1 40 Negative 1 11 12 Total 40 12 52 GRA1 + SAG1 Positive 38 1 39 Negative 2 11 14 Total 40 12 52 GRA1 + GRA7 Positive 39 1 40 Negative 1 11 12 Total 40 12 52 GRA7 + SAG1 Positive 39 1 40 Negative 1 11 12 Total 40 12 52 373 374 375
376 377
Table 2. Comparison of detection of Toxoplasma gondii antibodies in the serum of captive jaguars by enzyme-linked immunosorbent
378
assays (ELISAs) based on recombinant dense granule antigens (GRAs) and surface antigens (SAGs), with a commercial ELISA as the
379
reference method.
380 Single/mixture of
381 382 383
Prevalence
Agreement
Kappa
McNemar’s
Sensitivity
Specificity
False positive
False negative
recombinant antigens
(%)
(%)
value
test
(%)
(%)
rate (%)
rate (%)
GRA1
75.0
90.4
0.74
p > 0.05
92.5
83.3
16.7
7.5
SAG1
75.0
90.4
0.74
p > 0.05
92.5
83.3
16.7
7.5
GRA7
76.9
96.2
0.89
p > 0.05
97.5
91.6
8.3
2.5
GRA1 + SAG1
75.0
94.2
0.84
p > 0.05
95.0
91.6
8.3
5.0
GRA1 + GRA7
76.9
96.2
0.89
p > 0.05
97.5
91.6
8.3
2.5
GRA7 + SAG1
76.9
96.2
0.89
p > 0.05
97.5
91.6
8.3
2.5
384 385
Table 3. Area under the curve (AUC) values and relative sensitivity and specificity at
386
different cutoffs of single and mixed recombinant granule antigens (GRAs) and surface
387
antigens (SAGs) in enzyme-linked immunosorbent assays (ELISAs) for Toxoplasma gondii
388
immunoglobulin G (IgG) in the serum of captive jaguars.
389 Single or mixture of recombinant antigens
AUC (95% CI*)
Sensitivity (%)
Specificity (%)
Cutoff
GRA1
0.954 (0.902–1.00)
87.5
91.7
0.46
SAG1
0.967 (0.923–1.00)
92.5
91.7
0.46
GRA7
0.991 (0.972–1.00)
97.5
83.3
0.37
GRA1 + SAG1
0.992 (0.975–1.00)
97.5
83.3
0.35
GRA1 + GRA7
0.987 (0.961–1.00)
97.5
83.3
0.38
GRA7 + SAG1
0.998 (0.991–1.00)
97.5
91.7
0.38
* CI = confidence interval
390 391
392
Figure legends
393 394
Fig. 1. Comparison of immunoreactivities in the serum of captive jaguars of single
395
and mixed recombinant dense granule antigens (GRAs) and surface antigens (SAGs) in
396
enzyme-linked immunosorbent assays (ELISAs) for Toxoplasma gondii immunoglobulin G
397
(IgG). The horizontal lines represent cutoff values: 0.46 (GRA1); 0.45 (SAG1); 0.47
398
(GRA7); 0.41 (GRA1 + SAG1); 0.48 (GRA1 + GRA7), and 0.41 (GRA7 + SAG1).
399 400
Recombinant GRA7 + SAG1 antigens showed highest sensitivity and specificity Toxoplasma gondii detection using recombinant antigens is suitable for diagnosis First serological study for T. gondii in 52 jaguars from 10 states in Mexico High seroprevalence was observed in captive jaguars using ELISA
S0014-4894(19)30381-9
DOI:
https://doi.org/10.1016/j.exppara.2019.107791
Reference:
YEXPR 107791
To appear in:
Experimental Parasitology
Received Date: 29 August 2019 Revised Date:
23 October 2019
Accepted Date: 4 November 2019
Please cite this article as: Reynoso-Palomar, A., Moreno-Gálvez, D., Villa-Mancera, A., Prevalence of Toxoplasma gondii parasite in captive Mexican jaguars determined by recombinant surface antigens (SAG1) and dense granular antigens (GRA1 and GRA7) in ELISA-based serodiagnosis, Experimental Parasitology (2019), doi: https://doi.org/10.1016/j.exppara.2019.107791. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Inc.
1
Prevalence of Toxoplasma gondii parasite in captive Mexican jaguars determined by
2
recombinant surface antigens (SAG1) and dense granular antigens (GRA1 and
3
GRA7) in ELISA-based serodiagnosis
4
Alejandro Reynoso-Palomara, Dulce Moreno-Gálveza, Abel Villa-Manceraa*
5 6 7
a
8
Puebla, 4 Sur 304 Col. Centro, C.P. 75482, Tecamachalco Puebla, México.
9
*Corresponding author. E-mail address: [email protected] (A. Villa-Mancera)
Facultad de Medicina Veterinaria y Zootecnia, Benemérita Universidad Autónoma de
10 11
Abstract
12
Toxoplasma gondii is an obligate intracellular protozoan parasite that infects almost
13
all warm-blooded animals, including humans, causing serious public health problems. In
14
this study, the seroprevalence of T. gondii in captive jaguars in 10 Mexican zoos was
15
determined using single and mixtures of recombinant surface antigens (SAG1) and dense
16
granular antigens (GRA1 and GRA7) in immunoglobulin G (IgG) enzyme-linked
17
immunosorbent assays (ELISAs). Their efficacy was compared with the tachyzoite lysate
18
antigen. All recombinant antigens were characterised by high sensitivity (92.5– 97.5%); the
19
specificity of the IgG ELISAs was variable (83.3– 91.6%). Mixtures of the two
20
recombinant proteins were generally more reactive than single antigens. GRA7 + SAG1
21
showed the highest sensitivity (97.5%) and specificity (91.6%), almost perfect agreement
22
(96.2%), and a kappa value of 0.89. An area under the curve value of 0.998 represented a
23
highly accurate test with a cutoff value of 0.8. The seroprevalence of anti-T. gondii IgG
24
antibodies in the single and mixed recombinant antigen ELISAs was 75.0–76.9%. This
25
study shows that GRA7 + SAG1 can be successfully used to diagnose T. gondii infection in
26
jaguars for effective monitoring of prevalence and for devising control methods and
27
prevention strategies against toxoplasmosis.
28 29 30 31
Keywords:
Toxoplasma
gondii;
serodiagnosis; prevalence; zoo; Mexico
jaguars;
ELISA;
recombinant
antigens;
32
1. Introduction
33
Toxoplasmosis is a globally prevalent zoonotic disease caused by the protozoan
34
Toxoplasma gondii (Apicomplexa; Sarcocystidae) which can affect most warm-blooded
35
animals, including humans. Humans and animals usually become infected through
36
ingestion of undercooked or raw meat containing cysts, or by consuming food
37
contaminated with sporulated oocysts (Zhou et al., 2011). Members of the family Felidae
38
are definitive and intermediate hosts of this intracellular parasite, and play a key role in
39
transmission of the infection (Dubey, 2009; Calero-Bernal and Gennari, 2019). Captive
40
jaguars may serve as source of infection for susceptible animal species and human beings
41
sharing the same environment, including zoos (Ullmann et al., 2010).
42
To estimate prevalence more accurately, sensitive and specific diagnostic tests for T.
43
gondii infection need to be performed. Serological diagnosis by enzyme-linked
44
immunosorbent assay (ELISA) for immunoglobulin G (IgG) detection is a useful and
45
convenient tool (Liu et al., 2015). The tachyzoite propagated in vivo or in vitro, has been
46
traditionally used in serological detection of toxoplasmosis, and is characterised by high
47
sensitivity and specificity. However, it has several disadvantages: it is time-consuming and
48
methodologically demanding to produce; inconsistent in quality; subject to contamination
49
with extraparasitic material; and its production may expose laboratory technicians to
50
harmful living parasites (Gatkowska et al., 2006; Ferra et al., 2019). Currently, the lack of
51
continuous availability and abundant quantities of native antigens for diagnostic testing is
52
an obstacle. Therefore, the use of purified recombinant protein is desirable, with the
53
advantage of reducing production costs and the possibility of using more of than one
54
defined antigen for detection of specific antibodies (Rostami et al., 2018; Ferra et al.,
55
2019).
56
In a number of studies, toxoplasmosis in humans and animals has been diagnosed
57
using different recombinant proteins, such as surface proteins (SAGs), microneme proteins
58
(MICs), rhoptry proteins (ROPs), and dense granule proteins (GRAs; Sudan et al., 2013;
59
Wang and Yin, 2014; Rostami et al., 2018). Among the different surface antigen molecules,
60
SAG1 (also named P30) is highly abundant and the most immunodominant and stage-
61
specific antigen on tachyzoites present during the acute and chronic phases of the disease
62
(Bel-Ochi et al., 2013; Wang and Yin, 2014). GRAs are secreted in abundance, and are
63
involved in parasite survival and virulence. They are also a major component of the
64
parasitiphorous vacuole, in which tachyzoites multiply, and the cyst wall surrounding
65
slower growing bradyzoites (Cesbron-Delauw, 1994; Mercier et al., 2005). GRA1 is
66
important for diagnosis of chronic infection, while GRA7 has direct contact with the host
67
immune system, and induces strong antibody responses in the early and late stages of
68
parasite infection (Jacobs et al., 1998, 1999; Wang et al., 2013). In this study, we evaluated
69
the performance of single and different combinations of recombinant antigen proteins
70
(GRA1, GRA7, and SAG1) in ELISA-based serodiagnoses to determine the prevalence of
71
T. gondii infection in jaguars in ten Mexican zoos.
72 73
2. Materials and methods
74 75
2.1. Location and sample collection
76
Serum samples from 52 jaguars (Panthera onca; Felidae), including 28 males and 24
77
females, were collected from 10 Mexican zoos, located in six of Mexico's 32 states
78
(Michoacán, Puebla, Oaxaca, Morelos, Sinaloa, and the State of Mexico). Jaguars were
79
immobilised with an intramuscular injection of tiletamine-zolazepam or ketamine plus
80
xylazine. Blood samples were collected from the jugular, cephalic, or saphenous vein. All
81
serum samples were obtained between January and May 2018, and were centrifuged at
82
2000 g for 15 min at 4 °C, and the supernatant was stored at −20 °C for ELISA analyses.
83
The jaguars’ diet consisted mainly of chicken carcasses, and other animal products (horse,
84
bovine, and rabbit meat) when available. Water was provided ad libitum. This study was
85
approved by the Animal Care and Ethics Committee of Meritorious Autonomous
86
University of Puebla, and all procedures complied with national legislation pertaining to
87
Animal Health Research.
88 89
2.2. Negative and positive serum samples
90
A commercial IgG ELISA (DRG International, USA; EIA-1798) based on tachyzoite
91
lysate antigen (TLA) was modified with horseradish peroxidase (HRP)-conjugated goat
92
anti-cat IgG (Sigma-Aldrich, USA) to replace the conjugated anti-human IgG included in
93
the kit (sensitivity 98.3% and specificity 99.2%). The serum ELISA test was selected as the
94
reference standard to evaluate the performance of the recombinant antigen ELISAs.
95
Confirmatory
96
chemiluminescent immunoassay (MexLab Group, Mexico) following the manufacturer’s
97
instructions using T. gondii antigen and FITC-conjugated goat anti-cat IgG antibody
98
(Sigma-Aldrich, USA) as secondary antibody.
testing
for
negative
and
positive
samples
was
performed
by
99 100
2.3. Detection of antibodies to Toxoplasma gondii by indirect ELISAs using recombinant
101
antigens
102
Optimum dilutions of recombinant coating antigens, jaguar serum samples, and HRP-
103
conjugated goat anti-cat IgG were established by systematic chequerboard titrations. Flat-
104
bottomed 96-well microtiter plates (Costar, Corning, NY, USA) were coated with 100 μl of
105
GRA1: expressed in E. coli, UniProt P13403 (ab43137, Abcam), SAG1 produced in E. coli,
106
containing the GST-tag (amino acids 45-198; ab68110, Abcam), GRA7: expressed in E.
107
coli (amino acids 24-100; ab73748, Abcam), and a mixture of these proteins at a final
108
concentration of 5 μg/ml for each recombinant protein in the coating buffer (15 mM
109
Na2CO3, 35 mM NaHCO3, pH 9.6), and incubated overnight at 4 °C. After three washes
110
with phosphate-buffered saline (PBS) containing 0.05% Tween 20 (PBS-T), nonspecific
111
binding sites were blocked with a reconstituted 3% w/v solution of non-fat milk powder
112
(Amresco, USA) prepared in PBS for 2 h at 37 °C. The plates were then washed again three
113
times, and negative and positive serum samples were used at a dilution of 1:50 in PBS and
114
incubated at 37 °C for 1 h. After another wash, incubation was carried out for 1 h at 37 °C,
115
with the goat anti-cat IgG-HRP diluted at 1:2500 in 3% non-fat milk in PBS. Following
116
incubation and washing, colour was developed by the addition of 100 μl per well of TMB
117
substrate solution (Sigma-Aldrich, USA), and stopped with 50 μl of 4N H2SO4. Finally,
118
optical density (OD) was measured at 450 nm in an ELISA reader (BioTek ELx800,
119
BioTek).
120
Each serum sample was assayed twice, and the results were determined for each
121
sample by calculating the mean OD reading of duplicate wells. The cutoff point for a
122
positive result was defined as any value higher than the average OD reading plus two
123
standard deviations obtained with negative serum samples.
124 125
2.4. Statistical analysis
126
The performance of each recombinant antigen ELISA was compared with the
127
commercial ELISA as a standard test for determination of sensitivity, specificity, and kappa
128
value. We also used McNemar’s test to calculate the relationships between the two assay
129
methods.
130
The optimal cutoff value for each recombinant antigen ELISA was established by
131
receiver operating characteristic (ROC) curve analysis with a 95% confidence interval (CI).
132
The accuracy of area under the curve (AUC) values proposed by Swets (1988) were
133
considered as follows: non-informative (AUC < 0.5); low accuracy (0.5 < AUC < 0.7),
134
moderate accuracy (0.7 < AUC < 0.9); and high accuracy (0.9 < AUC < 1). The relative
135
diagnostic sensitivity and specificity of the ELISAs were evaluated for different cutoffs. All
136
statistical analyses were carried out using IBM SPSS 20 software package (SPSS Inc.,
137
Chicago, USA) for Windows.
138 139
3. Results
140 141
3.1. Evaluation of diagnostic performance of commercial ELISA and ELISAs using
142
recombinant GRA and SAG antigens
143
The commercial ELISA was carried out initially on the 52 jaguar samples for the
144
purpose of comparing it with the ELISAs using single and mixed GRAs and SAGs. Forty
145
samples were positive in the comparison with commercial ELISA (Table 1) and
146
chemiluminescence immunoassay (sensitivity 94% and specificity 100%). As shown in
147
Table 1, there were 39 positive and 13 negative samples for GRA1, SAG1 and GRA1 +
148
SAG1, and 40 positive and 12 negative samples for GRA7, GRA1 + GRA7, and GRA7 +
149
SAG1.
150
The study recorded the highest mean OD value of 0.99 (95% CI: 0.98–1.04) for the
151
antigen mixture GRA7 + SAG1, followed by 0.95 for GRA7 (95% CI: 0.90–0.99), 0.94 for
152
GRA1 + GRA7 (95% CI: 0.93–0.98), 0.88 for GRA1 + SAG1 (95% CI: 0.81–0.94), 0.84
153
for SAG1 (95% CI: 0.77–0.91), and 0.70 for GRA1 (95% CI: 0.64–0.74). Cutoff values
154
ranged from 0.41 to 0.48 (Fig. 1).
155
The highest prevalence of T. gondii infection (76.9%, 40 of the 52 jaguar serum
156
samples) was obtained with GRA7, GRA1 + GRA7, and GRA7 + SAG1, followed by
157
GRA1 (75.0%, 39/52), SAG1, and GRA1 + SAG1 (Table 2). There was no significant
158
difference in the proportion of positive test results when comparing single and mixed
159
recombinant antigen ELISAs with the commercial ELISA (McNemar’s chi-squared test; p
160
> 0.05; Table 2). Based on the statistical analysis, the agreement was high for GRA7,
161
GRA1 + GRA7, and GRA7 + SAG1 (kappa = 0.89). For GRA1 + SAG1, GRA1, and
162
SAG1, kappa values were 0.84, 0.74, and 0.74, respectively (Table 2).
163
For detection of T. gondii antibodies in jaguars, we used the commercial ELISA as a
164
reference test to validate the sensitivity and specificity of the IgG ELISAs conducted with
165
single and mixed recombinant GRAs and SAGs. GRA7, GRA1 + GRA7, and GRA7 +
166
SAG1 had the highest sensitivity (97.5%) and specificity (91.6%), and the lowest false
167
positive and negative rates (Table 2).
168 169
3.2. ROC analysis
170
ROC curve analysis of single and mixed recombinant antigens revealed an AUC of
171
high accuracy ranging between 0.954 (GRA1) and 0.998 (GRA7 + SAG1; Table 3). The
172
sensitivities of IgG ELISAs estimated from all the 40 positive serum samples were different
173
for the single recombinant antigens, with high specificity obtained for GRA1 and SAG1
174
(Table 3). Higher sensitivity (97.5%) was observed in IgG ELISAs with GRA7, GRA1 +
175
SAG1, GRA1 + GRA7, and GRA7 + SAG1. Definitely lower specificity at 83.3% was
176
observed for antigen mixtures GRA1 + SAG1 and GRA1 + GRA7. The cutoff value at 0.38
177
for GRA7 + SAG1 had the highest relative sensitivity (97.5%) and relative specificity
178
(91.7%) among the six ELISAs.
179 180
4. Discussion
181
In this study, we developed IgG ELISA tests based on recombinant T. gondii antigen
182
proteins (composed of three different kinds of proteins and their mixtures), in reference to
183
the TLA. Bacterial recombinant proteins offer many advantages in immunodiagnosis, as
184
they allow easy standardisation of serological tests, and reduce the cost of production and
185
purification (Pietkiewics et al., 2004, Holec-Gasior, 2013). Some key disadvantages of
186
producing recombinant antigens in E. coli are the efficiency of expression, contamination
187
of purified recombinant protein and incorrect folding (Pietkiewicz et al., 2007; Holec-
188
Gasior, 2013).
189
Based on our results, the sensitivity and specificity of the SAG1 antigen varied in IgG
190
ELISA tests with single or mixtures of recombinant antigens, from a high of 97.5% and
191
91.6%, respectively, for GRA7 + SAG1 to a low of 92.5% and 83.3%, respectively, for
192
SAG1. In humans, SAG1 exhibits similar sensitivity and specificity (93.6% and 92.9%,
193
respectively; Khanaliha et al., 2014). In this study, the diagnostic sensitivity (97.5%) and
194
specificity (91.6%) of GRA1 were higher than those for mink (78.9% and 95.5%,
195
respectively; Gu et al., 2015).
196
A limited number of studies in cats have reported the diagnostic performance of T.
197
gondii GRA7 protein, alone or in combination with other recombinant antigens. Our study
198
found the sensitivity and specificity of GRA7 to be 97.5% and 91.6%, respectively, in
199
comparison with the commercially available ELISA. The single recombinant GRA7 protein
200
performed better than other protein antigens used in this study. Similar results for
201
sensitivity and specificity have been reported in cattle (95.5% and 91.1%, respectively),
202
cats (94.9% and 97.9%, respectively), and mink (84.2% and 99.1%, respectively; Wang et
203
al., 2014; Cai et al., 2015; Gu et al., 2015). Lower values for sensitivity and specificity of
204
GRA7 have also been observed in cats (35.1% and 89.86%, respectively; Abdelbaset et al.,
205
2017).
206
In our evaluation of the diagnostic utility of ELISAs with recombinant antigen
207
mixtures, the highest sensitivity and specificity were obtained for GRA7 + SAG1.
208
Abdelbaset et al. (2017) evaluated different combinations of T. gondii recombinant antigens
209
(SAG, GRA, and MIC) in IgG ELISA using cat serum samples. SAG2 + GRA7 showed
210
90.54% sensitivity and 85.51% specificity, while a five-antigen combination that included
211
GRA7 resulted in 89.19% sensitivity and 95.36% specificity. In humans, similar results
212
have been reported (Jacobs et al., 1999; Pietkiewicz et al., 2004). The increase in sensitivity
213
detection of IgG antibodies in sera by ELISA based on a mixture of two or three
214
recombinant antigens may result because several different epitopes of various proteins can
215
be recognised for specific antibodies from acute and/or chronic T. gondii infection (Ferra et
216
al., 2015). In addition, for discriminating between recent and latent infections, IgG avidity
217
assays using chemiluminescence could be used in the serological-screening strategy for
218
jaguars, with several advantages such as reproducibility, fast and cost effectiveness
219
(Rostami et al., 2018). Further research with a larger number of samples is required for
220
standardized detection of IgG antibodies against T. gondii in jaguars.
221
ROC analysis describes the performance of an ELISA test in terms of diagnostic
222
accuracy as defined by a reference assay. Its value can be interpreted as the probability of
223
the test distinguishing between a randomly chosen positive serum and a randomly chosen
224
negative serum (Swets, 1988). Based on this analysis, the AUC of 0.998 for GRA7 + SAG1
225
suggests that it represents a highly accurate test with good discriminatory capacity for
226
infected jaguars; the cutoff point was 0.38, corresponding to a sensitivity of 97.5% and
227
specificity of 91.7%. Similar AUC results have been reported for SAG1 (0.996;
228
Hosseininejad, 2012) and GRA7 (0.969; Cai et al., 2015) in cats.
229
No published reports are available on the use of recombinant antigen proteins for
230
ELISA-based serodetection of anti-T. gondii IgG antibodies in jaguars from Mexico, and
231
there are few reports on the worldwide prevalence of this parasite. Using recombinant
232
antigen ELISAs, this study shows a high seroprevalence of the parasite, ranging from
233
75.0% to 76.9% in 10 Mexican zoos. In Brazil, high prevalence rates in captive jaguars
234
have also been reported by modified agglutination test (MAT; 63.2%; Silva et al., 2001),
235
indirect immunofluorescent assay (IFAT; 84.6%; André et al., 2010), MAT (100%, free-
236
ranging jaguars; Furtado et al., 2015), and indirect hemagglutination and MAT (63.64%;
237
Silva et al., 2016). However, the prevalence of toxoplasmosis in Mexico is different
238
depending on the test used as the criterion for measurement—serum IgG ELISA (100%),
239
IFAT (41.7%), and amplification of a fragment of the SAG1 gene by nested PCR using
240
blood (0%) and stool (11.1%) samples (Gomez-Rios et al., 2019).
241
Antibodies against T. gondii have previously been detected in Mexico in the animals
242
(chicken, cattle, horses, and rabbits) fed to the captive jaguars included in this survey
243
(Hernandez-Cortazar et al., 2015). Silva et al. (2007) concluded that the consumption of
244
raw meat is an important risk factor for T. gondii infection, and the most effective way to
245
reduce the risk of exposure to the parasite is feeding these animals only with meat
246
previously frozen at −12 °C for a period >7 days.
247 248 249 250
Conclusions
251
The results reported in this study show the high serodiagnostic usefulness of IgG
252
ELISA based on a mixture of recombinant GRA7 and SAG1 antigens for detection of T.
253
gondii infection in jaguars, and probably in other domestic animals. This study
254
demonstrated high prevalence of the parasite in captive jaguars in 10 zoos from six
255
Mexican states. Further studies are required to increase our knowledge of risk factors linked
256
to the transmission of T. gondii from animals to humans, and for generating control
257
strategies and preventive measures against toxoplasmosis.
258 259
Competing interest
260
The authors declare that they have no competing interests.
261 262 263 264
Acknowledgments The authors are grateful to veterinary staff and administrative personnel of the zoos.
References
265 266 267
Abdelbaset, A.E., Alhasan, H., Salman, D., Karram, M.H., Ellah Rushdi, M.A.,
268
Xuenan, X., Igarashi, M., 2017. Evaluation of recombinant antigens in combination and
269
single formula for diagnosis of feline toxoplasmosis. Experimental parasitology 172, 1-4.
270
Andre, M.R., Adania, C.H., Teixeira, R.H., Silva, K.F., Jusi, M.M., Machado, S.T.,
271
de Bortolli, C.P., Falcade, M., Sousa, L., Alegretti, S.M., Felippe, P.A., Machado, R.Z.,
272
2010. Antibodies to Toxoplasma gondii and Neospora caninum in captive neotropical and
273
exotic wild canids and felids. The Journal of parasitology 96, 1007-1009. Bel-Ochi, N.C., Bouratbine, A., Mousli, M., 2013. Enzyme-linked immunosorbent
274 275
assay using recombinant SAG1
antigen
to
detect
Toxoplasma
gondii-specific
276
immunoglobulin G antibodies in human sera and saliva. Clin. Vaccine Immunol. 20, 468-
277
473.
278
Cai, Y., Wang, Z., Li, J., Li, N., Wei, F., Liu, Q., 2015. Evaluation of an indirect
279
ELISA using recombinant granule antigen Gra7 for serodiagnosis of Toxoplasma gondii
280
infection in cats. Journal of Parasitology 101, 37-41.
281 282 283 284 285 286
Calero-Bernal, R., Gennari, S.M., 2019. Clinical Toxoplasmosis in Dogs and Cats: An Update. Frontiers in veterinary science 6, 54. Cesbron-Delauw, M.-F., 1994. Dense-granule organelles of Toxoplasma gondii: their role in the host-parasite relationship. Parasitology today 10, 293-296. Dubey, J., 2009. History of the discovery of the life cycle of Toxoplasma gondii. International journal for parasitology 39, 877-882.
287
Ferra, B., Holec-Gasior, L., Kur, J., 2015. Serodiagnosis of Toxoplasma gondii
288
infection in farm animals (horses, swine, and sheep) by enzyme-linked immunosorbent
289
assay using chimeric antigens. Parasitology international 64, 288-294.
290
Ferra, B.T., Holec-Gasior, L., Gatkowska, J., Dziadek, B., Dzitko, K., Grazlewska,
291
W., Lautenbach, D., 2019. The first study on the usefulness of recombinant tetravalent
292
chimeric proteins containing fragments of SAG2, GRA1, ROP1 and AMA1 antigens in the
293
detection of specific anti-Toxoplasma gondii antibodies in mouse and human sera. PloS one
294
14, e0217866.
295
Furtado, M.M., Gennari, S.M., Ikuta, C.Y., de Almeida Jacomo, A.T., de Morais,
296
Z.M., de Jesus Pena, H.F., de Oliveira Porfirio, G.E., Silveira, L., Sollmann, R., de Souza,
297
G.O., 2015. Serosurvey of smooth Brucella, Leptospira spp. and Toxoplasma gondii in
298
free-ranging jaguars (Panthera onca) and domestic animals from Brazil. PloS one 10,
299
e0143816.
300
Gatkowska, J., Hiszczynska-Sawicka, E., Kur, J., Holec, L., Dlugonska, H., 2006.
301
Toxoplasma gondii: an evaluation of diagnostic value of recombinant antigens in a murine
302
model. Experimental parasitology 114, 220-227.
303
Gomez-Rios, A., Ortega-Pacheco, A., Gutierrez-Blanco, E., Acosta-Viana, K.Y.,
304
Guzman-Marin, E., Guiris-Andrade, M.D., Hernandez-Cortazar, I.B., Lopez-Alonso, R.,
305
Cruz-Aldan, E., Jimenez-Coello, M., 2019. Toxoplasma gondii in Captive Wild Felids of
306
Mexico: Its Frequency and Capability to Eliminate Oocysts. Vector borne and zoonotic
307
diseases.
308
Gu, Y., Wang, Z., Cai, Y., Li, X., Wei, F., Shang, L., Li, J., Liu, Q., 2015. A
309
comparative study of Toxoplasma gondii seroprevalence in mink using a modified
310
agglutination test, a Western blot, and enzyme-linked immunosorbent assays. Journal of
311
Veterinary Diagnostic Investigation 27, 616-620.
312
Hernandez-Cortazar, I., Acosta-Viana, K.Y., Ortega-Pacheco, A., Guzman-Marin,
313
E.d.S., Aguilar-Caballero, A.J., Jimenez-Coello, M., 2015. Toxoplasmosis in Mexico:
314
epidemiological situation in humans and animals. Revista do Instituto de Medicina Tropical
315
de Sao Paulo 57, 93-103.
316
Holec-Gasior, L., 2013. Toxoplasma gondii recombinant antigens as tools for
317
serodiagnosis of human toxoplasmosis: current status of studies. Clinical and vaccine
318
immunology : CVI 20, 1343-1351.
319
Jacobs, D., Dubremetz, J.-F., Loyens, A., Bosman, F., Saman, E., 1998. Identification
320
and heterologous expression of a new dense granule protein (GRA7) from Toxoplasma
321
gondii. Molecular and biochemical parasitology 91, 237-249.
322
Jacobs, D., Vercammen, M., Saman, E., 1999. Evaluation of recombinant dense
323
granule antigen 7 (GRA7) of Toxoplasma gondii for detection of immunoglobulin G
324
antibodies and analysis of a major antigenic domain. Clin. Diagn. Lab. Immunol. 6, 24-29.
325 326
Liu, Q., Wang, Z.-D., Huang, S.-Y., Zhu, X.-Q., 2015. Diagnosis of toxoplasmosis and typing of Toxoplasma gondii. Parasites & vectors 8, 292.
327
Mercier, C., Adjogble, K.D., Däubener, W., 2005. Dense granules: are they key
328
organelles to help understand the parasitophorous vacuole of all apicomplexa parasites?
329
International journal for parasitology 35, 829-849.
330
Pietkiewicz, H., Hiszczyńska-Sawicka, E., Kur, J., Petersen, E., Nielsen, H.,
331
Stankiewicz, M., Andrzejewska, I., Myjak, P., 2004. Usefulness of Toxoplasma gondii-
332
specific recombinant antigens in serodiagnosis of human toxoplasmosis. Journal of clinical
333
microbiology 42, 1779-1781.
334
Pietkiewicz, H., Hiszczyńska-Sawicka, E., Kur, J., Petersen, E., Nielsen, H.,
335
Stankiewicz, M., Andrzejewska, I., Myjak, P., 2004. Usefulness of Toxoplasma gondii-
336
specific recombinant antigens in serodiagnosis of human toxoplasmosis. Journal of clinical
337
microbiology 42, 1779-1781. Quigley, H., Foster, R., Petracca, L., Payan, E., Salom, R., Harmsen, B., 2017.
338 339
Panthera onca. The IUCN Red List of Threatened Species 2017: e. T15953A50658693. Rostami, A., Karanis, P., Fallahi, S., 2018. Advances in serological, imaging
340 341
techniques and molecular diagnosis of Toxoplasma gondii infection. Infection 46, 303-315.
342
Silva, C.C., Meneses, A.M.C., Moraes, C.C.G.d., Carmo, E.L.d., Langoni, H., Silva,
343
R.C.d., Pancieri, H.M.H., Martins, B.R., Souza, N.F.d., 2016. soroprevalência e pesquisa de
344
oocistos de Toxoplasma gondii em felídeos selvagens procedentes do estado do pará,
345
Brasil. Veterinária e Zootecnia 23, 400-408.
346
Silva, J.C.R., Marvulo, M.F.V., Dias, R.A., Ferreira, F., Amaku, M., Adania, C.H.,
347
Neto, J.S.F., 2007. Risk factors associated with sero-positivity to Toxoplasma gondii in
348
captive neotropical felids from Brazil. Preventive veterinary medicine 78, 286-295.
349
Silva, J.C.R., Ogassawara, S., Adania, C.H., Ferreira, F., Gennari, S.M., Dubey, J.,
350
Ferreira-Neto, J.S., 2001. Seroprevalence of Toxoplasma gondii in captive neotropical
351
felids from Brazil. Veterinary Parasitology 102, 217-224. Sudan, V., Jaiswal, A.K., Shanker, D., 2013. Recent trends in the diagnosis of
352 353
toxoplasmosis. Clinical Reviews and Opinions 5, 11-17. Swets, J.A., 1988. Measuring the accuracy of diagnostic systems. Science 240, 1285-
354 355
1293.
356
Ullmann, L.S., da Silva, R.C., de Moraes, W., Cubas, Z.S., dos Santos, L.C.,
357
Hoffmann, J.L., Moreira, N., Guimaraes, A.M.S., Montaño, P., Langoni, H., 2010.
358
Serological survey of Toxoplasma gondii in captive Neotropical felids from Southern
359
Brazil. Veterinary parasitology 172, 144-146.
360 361
Wang, Y., Wang, G., Zhang, D., Yin, H., Wang, M., 2013. Identification of novel B cell epitopes within Toxoplasma gondii GRA1. Experimental parasitology 135, 606-610.
362
Wang, Z., Ge, W., Li, J., Song, M., Sun, H., Wei, F., Liu, Q., 2014. Production and
363
evaluation of recombinant granule antigen protein GRA7 for serodiagnosis of Toxoplasma
364
gondii infection in cattle. Foodborne pathogens and disease 11, 734-739.
365 366 367
Zhou, P., Chen, Z., Li, H.-L., Zheng, H., He, S., Lin, R.-Q., Zhu, X.-Q., 2011. Toxoplasma gondii infection in humans in China. Parasites & vectors 4, 165.
368
Table 1. Detection of Toxoplasma gondii immunoglobulin G (IgG) in the serum of captive
369
jaguars, tested by enzyme-linked immunosorbent assays (ELISAs) based on single and
370
mixed recombinant dense granule antigens (GRAs) and surface antigens (SAGs), with a
371
commercial ELISA as the reference method.
372 Commercial ELISA Single/mixture Positive Negative Total of recombinant antigens GRA1 Positive 37 2 39 Negative 3 10 13 Total 40 12 52 SAG1 Positive 37 2 39 Negative 3 10 13 Total 40 12 52 GRA7 Positive 39 1 40 Negative 1 11 12 Total 40 12 52 GRA1 + SAG1 Positive 38 1 39 Negative 2 11 14 Total 40 12 52 GRA1 + GRA7 Positive 39 1 40 Negative 1 11 12 Total 40 12 52 GRA7 + SAG1 Positive 39 1 40 Negative 1 11 12 Total 40 12 52 373 374 375
376 377
Table 2. Comparison of detection of Toxoplasma gondii antibodies in the serum of captive jaguars by enzyme-linked immunosorbent
378
assays (ELISAs) based on recombinant dense granule antigens (GRAs) and surface antigens (SAGs), with a commercial ELISA as the
379
reference method.
380 Single/mixture of
381 382 383
Prevalence
Agreement
Kappa
McNemar’s
Sensitivity
Specificity
False positive
False negative
recombinant antigens
(%)
(%)
value
test
(%)
(%)
rate (%)
rate (%)
GRA1
75.0
90.4
0.74
p > 0.05
92.5
83.3
16.7
7.5
SAG1
75.0
90.4
0.74
p > 0.05
92.5
83.3
16.7
7.5
GRA7
76.9
96.2
0.89
p > 0.05
97.5
91.6
8.3
2.5
GRA1 + SAG1
75.0
94.2
0.84
p > 0.05
95.0
91.6
8.3
5.0
GRA1 + GRA7
76.9
96.2
0.89
p > 0.05
97.5
91.6
8.3
2.5
GRA7 + SAG1
76.9
96.2
0.89
p > 0.05
97.5
91.6
8.3
2.5
384 385
Table 3. Area under the curve (AUC) values and relative sensitivity and specificity at
386
different cutoffs of single and mixed recombinant granule antigens (GRAs) and surface
387
antigens (SAGs) in enzyme-linked immunosorbent assays (ELISAs) for Toxoplasma gondii
388
immunoglobulin G (IgG) in the serum of captive jaguars.
389 Single or mixture of recombinant antigens
AUC (95% CI*)
Sensitivity (%)
Specificity (%)
Cutoff
GRA1
0.954 (0.902–1.00)
87.5
91.7
0.46
SAG1
0.967 (0.923–1.00)
92.5
91.7
0.46
GRA7
0.991 (0.972–1.00)
97.5
83.3
0.37
GRA1 + SAG1
0.992 (0.975–1.00)
97.5
83.3
0.35
GRA1 + GRA7
0.987 (0.961–1.00)
97.5
83.3
0.38
GRA7 + SAG1
0.998 (0.991–1.00)
97.5
91.7
0.38
* CI = confidence interval
390 391
392
Figure legends
393 394
Fig. 1. Comparison of immunoreactivities in the serum of captive jaguars of single
395
and mixed recombinant dense granule antigens (GRAs) and surface antigens (SAGs) in
396
enzyme-linked immunosorbent assays (ELISAs) for Toxoplasma gondii immunoglobulin G
397
(IgG). The horizontal lines represent cutoff values: 0.46 (GRA1); 0.45 (SAG1); 0.47
398
(GRA7); 0.41 (GRA1 + SAG1); 0.48 (GRA1 + GRA7), and 0.41 (GRA7 + SAG1).
399 400
Recombinant GRA7 + SAG1 antigens showed highest sensitivity and specificity Toxoplasma gondii detection using recombinant antigens is suitable for diagnosis First serological study for T. gondii in 52 jaguars from 10 states in Mexico High seroprevalence was observed in captive jaguars using ELISA