INTRODUCTION
Neospora caninum is an important cause of abortion in cattle worldwide. It has a wide host range and can cause clinical disease in several species of animals, including dogs (Dubey et al. Reference Dubey, Hemphill, Calero-Bernal and Schares2017). It is morphologically and ancestorily related to Toxoplasma gondii, but canids (dogs, coyote, dingo and wolf) and not felids are its definitive hosts.
Unlike T. gondii, N. caninum is very difficult to isolate from naturally infected animals. For this reason, little is known of the genetic variability among N. caninum isolates from different hosts and most of the isolates were obtained from clinical cases. This limitation is apparent when a comparison is made: for T. gondii 1457 isolates (Shwab et al. Reference Shwab, Zhu, Majumdar, Pena, Gennari, Dubey and Su2014) compared with 108 N. caninum isolates (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Díez-Fuertes, García-Culebras, Moore, González-Warleta, Cuevas, Schares, Katzer, Pedraza-Díaz, Mezo and Ortega-Mora2013) were analysed.
The aims of this study were to isolate viable N. caninum from cattle, to perform biological and molecular characterization and to provide a new N. caninum isolate that may be used as a reference for further studies.
MATERIALS AND METHODS
Naturally infected cattle
Brain and serum samples were collected from nine cattle slaughtered in an abattoir in the municipality of Santa Rita do Passa Quatro, state of São Paulo, Brazil. Serum samples were tested by indirect fluorescent antibody test (IFAT) to detect antibodies against N. caninum using cut-off of 1:100. Tachyzoite cell cultures, derived from the NC-1 strain of N. caninum, were used as antigen (Dubey et al. Reference Dubey, Carpenter, Speer, Topper and Uggla1988). Secondary fluorescein isothiocyanate (FITC)-labelled conjugated IgG anti-bovine antibodies were used (Sigma-Aldrich®, Saint Louis, MO, USA). One adult, male cattle had an IFAT titre of 200 for N. caninum antibodies and was used for the present study.
Approximately 400 g of brain from the seropositive cattle was cut into small pieces and pooled. Around 10 g was separated for trypsin digestion and four aliquots were collected for further PCR analysis. The remainder was used to isolate N. caninum by bioasssay in a dog.
Bioassay in a dog
The brain tissue was mixed with dry commercial dog food and fed over 3-days to a 2-month-old female mixed-breed N. caninum seronegative dog (IFAT titre <50). The dog was donated by the owner to the experiment. This dog had never eaten raw meat and was kept on a diet of dry commercial dog food during the experimental period. Blood samples from the dog were collected weekly via the cephalic vein [0–135 days post-inoculation (PI)] and IFAT tests were performed using FITC-labelled conjugated IgG anti-dog antibodies (KPL® Inc., Gaithersburg, MD, USA) with a cut-off point of 1:50.
Fecal samples were examined daily (0–40 days PI) for detection of Neospora-like oocysts by means of a standard sucrose flotation technique. Daily output of oocysts as described (Pena et al. Reference Pena, Soares, Ragozo, Monteiro, Yai, Nishi and Gennari2007). When oocysts were observed, the fecal samples were mixed with 2% H2SO4 for sporulation at 25 °C and thereafter stored at 4 °C.
To measure the size of the oocysts, images of 100 unsporulated oocysts were captured at the magnification of 400× and analysed using the Image-Pro Plus® software, version 5.1 (Media Cybermetics, Inc., Silver Spring, MD, USA).
Brain inoculation in rodents and cell culture
Around 10 g of the cattle brain tissue were digested with 0·05% trypsin for 1 h at 37 °C (Dubey et al. Reference Dubey, Choudhary, Ferreira, Kwok, Butler, Carstensen, Yu and Su2013). After the digestion, the homogenized suspension was washed three times with HBSS buffer (Hank's Balanced Salt Solution, GIBCO®, Grand Island, NY, USA) by centrifuging it at 1500 × g for 10 min and the resultant pellet was suspended in RPMI 1640 medium (Roswell Park Memorial Institute medium). Two flasks containing a monolayer of VERO cells were seeded with the sediment obtained (500 µL per flask). Two gerbils (Meriones unguiculatus) were intraperitoneally inoculated (500 µL per animal).
Bioassay of oocysts in gerbils and isolation in cell culture
Eight groups of two gerbils each were orally inoculated with 200 oocysts per gerbil. Every week, blood samples were collected from all the animals for serological analysis and one group was randomly euthanized. Tissues were collected from each animal for PCR assay. For isolation in cell culture, nine gerbils, divided into three groups, were orally infected with 101, 102 and 103 oocysts. The brains of the gerbils were aseptically collected, homogenized, digested with 0·05% trypsin or acid–pepsin solution (pepsin, 1·3 g; NaCl, 2·5 g; HCl, 3·5 mL; and distilled water, 250 mL) as described in (Dubey, Reference Dubey1998), and inoculated into a monolayer of VERO cells.
The serum samples were tested by means of IFAT using anti-gerbil IgG FITC-labelled conjugate (Immunology Consultants Laboratory® Inc., Portland, OR, USA). The cut-off was 1:50; the positive samples were diluted until the endpoint and the titres were then attributed.
DNA extraction and PCR assay
DNA from animal tissues was obtained using DNeasy Blood and Tissue Kit® (Qiagen Inc., Hilden, Germany), in accordance with the manufacturer's instructions.
To extract DNA from the oocysts, the oocysts were ruptured by means of six cycles of freezing in liquid nitrogen (−192 °C for 2 min) and thawing at 37 °C. After this pre-treatment, DNA extraction was done using the QIAamp DNA Stool Mini Kit® (Qiagen).
The PCR assays were performed using a pair of primers based on the Nc-5 gene (Np6 plus and Np21 plus) (Müller et al. Reference Müller, Zimmermann, Hentrich and Gottstein1996) for both the animal tissues and oocysts; and based on the ITS-1 region [(JS4) (Šlapeta et al. Reference Šlapeta, Koudela, Votýpka, Modrý, Hořejš and Lukeš2002) and (CT2c, JS4b and CT2b) (Soares et al. Reference Soares, Lopes, Keid, Sercundes, Martins and Richtzenhain2011)] for the oocysts. The amplified DNA was viewed by electrophoresis on 2% agarose gels stained with SYBR® Safe DNA gel stain (Invitrogen™, Carlsbad, CA, USA).
Identity of Neospora-like oocysts
The DNA of the Neospora-like oocysts were tested with an ITS-1 nested PCR–RFLP, in order to confirm the identity of the Neospora-like oocysts and to rule out mixed infection with oocysts of Hammondia heydorni. The primers were designed based on 18S and 5·8SrRNA coding genes and flanked the ITS-1 region, as described by Soares et al. (Reference Soares, Lopes, Keid, Sercundes, Martins and Richtzenhain2011). The amplicons were digested using the restriction endonuclease TaqαI. Also, the oocysts were tested with Np6 plus and Np21 plus primers. The amplified DNA was viewed as described above.
ITS-1 sequencing of the isolate NC-SP1
DNA from tachyzoites was obtained from the cell culture. The ITS1–PCR products based on sense and antisense primers (JS4 and CT2b) were sequenced using the BigDye® Terminator v3·1 cycle sequencing kit and the sequencing products were analysed in the ABI 3730 DNA analyser (Life Technologies; Applied Biosystems, Foster City, CA, USA). The PCR amplicons were sequenced four times in both directions. The sequences were assembled and the contig was formed with the phred base-calling and phrap assembly tools, which are available in the Codoncode aligner suite v.1·5·2. (Codoncode Corp., Dedham, Ma, USA).
Microsatellite (MS) genotyping and eBURST analysis
Neospora caninum multilocus genotyping (MLG) based on nine MS markers (MS4, MS5, MS6A, MS6B, MS7, MS8, MS10, MS12 and MS21) was performed for the NC-SP1 isolate (DNA was derived from tachyzoites) under PCR conditions that had previously been described (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Díez-Fuertes, García-Culebras, Moore, González-Warleta, Cuevas, Schares, Katzer, Pedraza-Díaz, Mezo and Ortega-Mora2013). These multiplex PCRs were performed on approximately 200 ng of DNA. N. caninum (Nc-Spain7) DNA was included in each batch of amplifications as a reference isolate. MS allele assignment was performed in accordance with the sizes determined by capillary electrophoresis and sequencing of the MS5, MS10 and MS7 markers was done as previously described (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Pedraza-Díaz, Gómez-Bautista and Ortega-Mora2006, Reference Regidor-Cerrillo, Díez-Fuertes, García-Culebras, Moore, González-Warleta, Cuevas, Schares, Katzer, Pedraza-Díaz, Mezo and Ortega-Mora2013). The size of the 6-FAM-labelled PCR products for all of the MSs was determined using a 48-capillary 3730 DNA analyser (Applied Biosystems) with Gene Scan-500 (LIZ) size standards (Applied Biosystems) and MS sequencing with the Big Dye Terminator v3·1 cycle sequencing kit (Applied Biosystems) and a 3730 DNA analyser (Applied Biosystems) at the Genomic Unit of the Madrid Science Park, Spain. The sizes of the PCR products and sequences were analysed using the GeneMapper1 v3·5 software (Applied Biosystems) and BioEdit Sequence Alignment Editor v.7·0·1 (Copyright_ 1997–2004 Tom Hall, Ibis Therapeutics, Carlsbad, CA, USA), respectively, as described previously (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Díez-Fuertes, García-Culebras, Moore, González-Warleta, Cuevas, Schares, Katzer, Pedraza-Díaz, Mezo and Ortega-Mora2013).
The eBURST software was used to explore the closest genetically related genotypes of NC-SP1 in a dataset involving 81 N. caninum 9-MLGs from Argentine, Spanish, German and Scottish populations (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Díez-Fuertes, García-Culebras, Moore, González-Warleta, Cuevas, Schares, Katzer, Pedraza-Díaz, Mezo and Ortega-Mora2013). The eBURST software generates networks composed of MLGs represented as dots, linked to their single-locus variants (SLV: 8 shared loci out of 9) and double-loci variants (DVL: 7 shared loci out of 9), line by line (Feil et al. Reference Feil, Li, Aanensen, Hanage and Spratt2004).
Animal ethics
All the animals were handled in accordance with protocols that had been approved by the Animal Ethics Committee of the Faculty of Veterinary Medicine, University of São Paulo, São Paulo, Brazil (CEUA no. 8634270114), following the National Research Council Guide for the Care and Use of Laboratory Animals.
RESULTS
Bioassay in a dog
DNA of N. caninum was detected in one out of four aliquots from the cattle brain.
Neospora-like oocysts were detected in dog feces on day 7 PI. The duration of oocyst shedding was 14 days and the number of oocysts produced during the patent period was 144 500 oocysts. The average number of oocysts per gram of feces was 102 oocysts/g (a range of 2–946) (Supplementary Table S1). The unsporulated oocysts measured 9·85–12·20 (11·19 ± 0·45) μm × 10·28–12·42 (11·46 ± 0·47) μm.
No clinical signs were observed and antibodies against N. caninum (IFAT <50) were not detected in the dog serum during the period analysed (19 weeks).
Inoculation of the cattle brain in rodents and cell culture
The cell culture flask inoculated with the cattle brain tissue was examined for 2 months and no parasites were observed. Also, none of the gerbils seroconverted.
Infection of gerbils with oocysts and isolation in cell culture
All gerbils fed 200 oocysts seroconverted 14 days PI. The range of titres was from 100 to 25 600 (Supplementary Fig. S1). The gerbils remained clinically healthy. The PCR analysis based on the Nc-5 gene showed that the parasite was distributed in the tissues of the 16 gerbils (Supplementary Table S2).
Successful isolation in cell culture occurred with the brain homogenate from a gerbil inoculated with 102 oocysts that was euthanized 78 days PI. The acid–pepsin digestion method had been used for this sample. Tachyzoites were observed 24 days PI in the cell culture, but the isolate had slow growth. Six tissue cysts from fresh brain homogenate of the same gerbil were viewed under a microscope and were found to measure 23·84 ± 4·14 µm × 25·69 ± 5·30 µm.
Identity of Neospora-like oocysts
DNA of N. caninum was detected on 6 of the 14 days of oocyst shedding (days 7, 8, 9, 10, 11 and 13 PI) using nested ITS-1 PCR. In the same fecal samples, using Np6 plus and Np21 plus primers, it was also possible to detect DNA of N. caninum. The results based on ITS-1 nested PCR-RFLP indicated that only oocysts of N. caninum were present.
The final sequence deposited in GenBank®, under accession number KT581980, was 99% identical to homologous sequences of many N. caninum isolates available in the same database.
MS genotyping and genetic relationship
Alleles for each of the nine MS markers analysed were amplified and identified from a DNA sample obtained from NC-SP1 tachyzoites. No new alleles were detected, although comparison of the NC-SP1 MLG with the database of 81 MLGs showed that it has a unique and different genetic profile. In addition, the eBURST analysis detected the absence of SLVs with NC-SP1. Only four DLVs (with variation in two MS markers) were found associated with the NC-SP1 MLG. These genetically closest MLGs involved three MLGs of Spanish and one MLG of Argentine origin (Supplementary Table S3, Figs. S2 and S3). In addition, comparison of NC-SP1 MLG with other MLGs of N. caninum isolates obtained in Brazil, the NC-Bahia (Gondim et al. Reference Gondim, Pinheiro, Santos, Jesus, Ribeiro, Fernandes, Almeida, Freire, Meyer and McAllister2001) and NC-Goiás (García-Melo et al. Reference García-Melo, Regidor-Cerrillo, Ortega-Mora, Collantes-Fernández, de Oliveira, de Oliveira and da Silva2009) isolates, showed variation in three and four out of nine markers, respectively (Table 1).
Microsatellite (MS) genotyping of the Neospora caninum isolate NC-SP1 and comparison with the genetically closest isolates analysed and with other Brazilian isolates
BRA, Brazil; SPA, Spain; ARG, Argentina.
DISCUSSION
Successful isolation of the NC-SP1 isolate of N. caninum from a naturally infected cattle was possible due to the bioassay in a dog. Attempts to isolate the parasite from the primary sample (cattle brain) in cell cultures and rodents were not successful in this study. Dogs are efficient experimental hosts for isolation of N. caninum, because they can ingest large amounts of animal tissues (Gondim et al. Reference Gondim, Gao and McAllister2002). In our study, the dog ate approximately 390 g of brain, compared with 10 g used for cell cultures and rodents. However, not all laboratories have the proper facilities to house and use dogs as experimental models, and therefore standardized protocols to isolate N. caninum need to be improved. Dubey and Schares (Reference Dubey and Schares2011) emphasized in their review that isolation of viable parasites is difficult and that finding DNA is not as valuable as isolating viable parasites.
In this report, the results related to the N. caninum bioassay in a dog corroborate previous findings. The number of oocysts excreted by the dog can be considered to be a medium quantity of oocysts for N. caninum, in comparison with other reports (Lindsay et al. Reference Lindsay, Dubey and Duncan1999; Rodrigues et al. Reference Rodrigues, Gennari, Aguiar, Sreekumar, Hill, Miska, Vianna and Dubey2004). Gondim et al. (Reference Gondim, Gao and McAllister2002) reported variance in the quantity of oocysts shed by the dogs and they pointed out that the dogs shed more oocysts when they were fed with infected cattle tissues than when fed with infected mouse tissues. Furthermore, the dog remained clinically healthy and did not develop antibodies against N. caninum. These facts were also consistent with the findings from other studies (Basso et al. Reference Basso, Venturini, Venturini, Hill, Kwok, Shen and Dubey2001; Gondim et al. Reference Gondim, McAllister and Gao2005; Pena et al. Reference Pena, Soares, Ragozo, Monteiro, Yai, Nishi and Gennari2007).
Gerbils are considered to be susceptible hosts for clinical neosporosis, but results are inconsistent (see Dubey et al. Reference Dubey, Schares and Ortega-Mora2007). However, the stage of the parasite needs to be taken into consideration. In our study, none of the gerbils orally fed 1000 oocysts developed clinical signs, thus suggesting that the new isolate has low pathogenicity in gerbils. The capacity of isolates of N. caninum to induce clinical disease or death in the experimental models and the efficiency of in vitro proliferation are important features to be evaluated in virulence studies (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Gómez-Bautista, Sodupe, Aduriz, Álvarez-García, Del Pozo and Ortega-Mora2011; Dellarupe et al. Reference Dellarupe, Regidor-Cerrillo, Jiménez-Ruiz, Schares, Unzaga, Venturini and Ortega-Mora2014). Nevertheless, the results from virulence studies should be evaluated with caution because the interpretation can change according to the hosts, laboratory conditions and biological characteristics. The NC-SP1 isolate grew slowly in the cell culture and did not have the capacity to destroy the monolayer of VERO cells. More efforts will be made to adapt the new isolate to cell culture.
The sequencing of the N. caninum ITS-1 region confirmed that the new isolate is a member of the species N. caninum. MS genotyping has the power to distinguish between individual organisms and this genetic analysis showed that NC-SP1 has a unique genetic profile, in comparison with the database of 81 N. caninum MLGs. In addition, the four genetically closest MLGs were from isolates that originated from aborted bovine fetuses in Spain and Argentina (Regidor-Cerrillo et al. Reference Regidor-Cerrillo, Díez-Fuertes, García-Culebras, Moore, González-Warleta, Cuevas, Schares, Katzer, Pedraza-Díaz, Mezo and Ortega-Mora2013). Unfortunately, there is only a limited number of Brazilian isolates, and therefore NC-SP1 MLG could only be compared with two other isolates: NC-Bahia (Gondim et al. Reference Gondim, Pinheiro, Santos, Jesus, Ribeiro, Fernandes, Almeida, Freire, Meyer and McAllister2001) and NC-Goiás (García-Melo et al. Reference García-Melo, Regidor-Cerrillo, Ortega-Mora, Collantes-Fernández, de Oliveira, de Oliveira and da Silva2009), isolates from a dog and a bovine, respectively.
In summary, our study presented the biological and genetic characterization of a new isolate of N. caninum, named NC-SP1, from an adult naturally infected cattle. New attempts to isolate N. caninum from domestic and wild animals would be important for improving the understanding of the biological and genetic diversity among N. caninum isolates in Brazil.
SUPPLEMENTARY MATERIAL
The supplementary material for this article can be found at https://doi.org/10.1017/S0031182016002481
FINANCIAL SUPPORT
H. F. J. P. received financial support for this study by Fundação de Amparo à Pesquisa do Estado de São Paulo, FAPESP (2011/10636-4). S. O. is in receipt of scholarship from the Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES). R. M. S. and S. M. G. are in receipt of fellowships from Conselho Nacional de Desenvolvimento Científico e Tecnológico, CNPq.A departmental grant from CAPES/PROEX (2327/2015) also supported this study.
ACKNOWLEDGEMENTS
We thank Dr. Angelo João Stopiglia for kindly giving permission for using the clinics kennel of the Faculty.
