Restriction Fragment Length Polymorphism-based Genotyping of Toxoplasma gondii from Autopsy-Proven Cases of Acquired Immunodeficiency Syndrome-associated Cerebral Toxoplasmosis.

CONTEXT: Published data on genetic characterization of Toxoplasma gondii (T.gondii) from clinical cases of toxoplasmosis from India is lacking. AIMS: The present study was aimed at identifying genetic types of T. gondii in fatal cases of cerebral toxoplasmosis (CT) associated with HIV, from India. SETTINGS AND
DESIGN: Archived tissues of CT were obtained postmortem from 25 acquired immunodeficiency syndrome patients between 2000 and 2014. SUBJECTS AND METHODS: Direct amplification of eight different loci, namely, SAG1, 5'-3'SAG2, Alt. SAG2, SAG3, BTUB, GRA6, C22-8, and L358 followed by restriction fragment length polymorphism was used to genotype the parasite.
RESULTS: The canonical Types I, II, or III were not found in our study. More than 96% of the cases harbored atypical genotypes-likely recombinants of the canonical types; one case closely corresponded to Type II genotype.
CONCLUSIONS: Thus, a majority of T. gondii causing CT in South India belonged to a noncanonical lineage. These nonarchetypal genotypes differed from the conventional Types I, II, and III and caused devastating severity in patients with CT in the background of HIV. These results are a step further to deciphering the population genetics of this important zoonotic parasitic infection in Indian patients, information that has thus far been lacking.

INTRODUCTION

Toxoplasma gondii (T.gondii) is a unicellular eukaryotic protozoan parasite which infects one-third of the global human population.[1] In general, T. gondii infections in humans are mild but can cause fatal disease in congenitally infected children[2] and in immunocompromised individuals including patients with acquired immunodeficiency syndrome (AIDS).[34] In the developed world, cerebral toxoplasmosis (CT) was one of the most common opportunistic infections in AIDS patients seropositive for T. gondii, in the prehighly activated antiretroviral therapy (HAART) era, with a decreasing trend in later years.[5] In developing countries such as India, however, CT still continues to be an important cause of focal brain lesions.[6]

There are three main genotypes of T. gondii, namely, I, II, and III, and Genotype IV has been recognized recently in North America.[7] Canonical Types II and III of the parasite are commonly found to cause infection in immunocompromised patients in Europe and North America.[891011] Hence, the parasite was assumed to be clonal, globally and virulence was thought to be associated with particular genotypes.

T. gondii in South America is found to be genetically and biologically diverse.[12131415] In Central and Eastern Asia, the genetic makeup of the parasite has been extensively studied only in China: Overall, 13 ToxoDB genotypes and five atypical genotypes were identified across China: ToxoDB#9 was the most common genotype found in 65.5% of the cases, followed by ToxoDB#10 (Type I) in 18.3%. Clonal Types II and III and the genotype #205 were less frequently detected.[16] Among South Asian countries, Sri Lanka has reported mainly a single atypical genotype in 46% of the cases (not referenced in ToxoDB), followed by ToxoDB#20 in 38%, clonal Type III in 8%, and ToxoDB#9 in 4%.[16] Indian studies on genetic characterization of the parasite have been restricted to locus-specific genotyping of animal strains.[1718] In an earlier study, we carried out limited sequencing of four selected genetic loci (BTUB, GRA6, Alt. SAG2, SAG3) of T. gondii in 25 fatal cases of CT in HIV/AIDS patients. This revealed recombinants of genotypes II/III, with clonal Type III in only one case.[19] To unravel the degree of variation across the parasite genome, we undertook characterization of T. gondii DNA derived from the lesions of CT, by eight loci restriction fragment length polymorphism (RFLP). We found majority of samples to contain nonarchetypal strains. We conclude that polymerase chain reaction (PCR)-RFLP – a technique initially designed to identify SNPs that are biallelic among archetypal strains may not be entirely useful to genotype nonarchetypal strains of T. gondii observed across the world.

SUBJECTS AND METHODS

Materials

This study plan was approved by the Institutional Ethics Committee of NIMHANS (No. 59-IEC/2008 dt. 21-03-2008).

The study group comprised of postmortem brain specimens of 25 AIDS patients who had succumbed to severe CT during the years 2000–2014. Written informed consent to utilize samples of postmortem material for research and teaching was obtained from patients' first-degree relatives at the time of autopsy.

Appropriate methods for collecting the samples and individual protection was used in all procedures for the collection and handling of the biological samples. Diagnosis of CT was based on clinical, imaging, cerebrospinal fluid (CSF) findings, and confirmed by histopathology and immunohistochemistry and included the following criteria: (i) progressive neurological deficits; (ii) contrast-enhancing mass lesion(s) on computed tomography; and (iii) CSF seropositivity for T. gondii IgG on undiluted samples by Latex Agglutination slide test (Toxogen, Tulip Diagnostics Pvt. Ltd. Goa, India) (iv) histopathological confirmation of T. gondii infection.

The brains retrieved at autopsy were coronally sliced and examined. Lesions on gross examination were multiple involving basal ganglia, thalamus, brain stem and cerebellum, and appeared hemorrhagic and necrotic [Figure 1a]. Sections from these lesions revealed characteristic findings of CT with zones of hemorrhagic necrosis walled in by histiocytes and lymphocytes. Prominent perivascular inflammation and vascular thickening were present [Figure 1b].

Figure 1

Coronal slice of brain shows a large necrotic lesion in left basal ganglia and smaller hemorrhagic lesions in left frontal and right putamen (arrows, a). Microscopy from the lesions shows necrotizing inflammation with thrombosis of vessels (arrow, b). Immunohistochemistry demonstrates numerous small ruptured tachyzoites of in the lesion (c). [b: H and Ex Obj. 10, c: Immunohistochemistry with p30 antigen of x Obj. 20]

Immunohistochemistry using polyclonal antibodies to p30 antigen of T gondii (ready to use, Biogenex, USA) revealed numerous tachyzoite forms of T gondii in the inflamed wall of the lesions confirming the diagnosis of CT [Figure 1c]. Formalin-fixed paraffin-embedded (FFPE) tissues taken from histologically confirmed lesions were utilized for molecular genetic studies. All required precautions were undertaken to avoid contamination during molecular work including a three-room set up which exists in the Department of Microbiology for the molecular diagnosis of opportunistic infections.

Extraction of DNA and Toxoplasma gondii detection by B1 polymerase chain reaction

Areas of brain tissue representing higher concentrations of T. gondii antigen were marked onto the FFPE blocks by the pathologist (AM). Excess wax and tissue in the unmarked area of the blocks were trimmed off. Tissue sections were then captured from these blocks and DNA extracted by Guanidinium thiocyanate method.[19] Briefly, 100 μm sections from the marked areas of FFPE blocks representative of toxoplasma lesions were transferred into sterile Eppendorf tubes. As a precaution, new blades were used for each case to prevent carryover contamination. Tissue sections were deparaffinized in Xylene at 56°C overnight. Subsequently, the sections were digested overnight using Proteinase K and Guanidinium thiocyanate. This was followed by extraction of genomic DNA by the Phenol: Chloroform method.

Total genomic DNA extracted from FFPE tissues were subjected to nested polymerase chain reaction (nPCR) for the amplification of B1 gene of T. gondii using primers and conditions described earlier.[20] The RH strain of T. gondii was used as a positive control (kindly provided by Dr. M. L. Dubey). Beta (β)-globin gene amplification served as internal control for DNA extraction and amplification from all the samples.[21]

Multilocus polymerase chain reaction-restriction fragment length polymorphism genotyping

DNA from tissue blocks in which B1 gene of the parasite was amplifiable was taken up for genotyping.[22] A multilocus PCR was used to amplify T. gondii at eight-independent loci as detailed earlier, namely, SAG1, (5'+3') SAG2, Alt. SAG2, SAG 3, BTUB, GRA6, L358, and C22-8.[22] GT1 (ToxDB#10, Type I), PTG (ToxoDB #1, Type II), CTG (ToxoDB #2, Type III), TgCgCa1 (ToxoDB #66), MAS (ToxoDB #17), TgCtBr5 (ToxoDB #19), TgCtBr64 (ToxoDB #111), and TgRsCr1 (ToxoDB #52) served as reference strains (kindly provided by Dr. Su, Associate Professor, Department of Microbiology, The University of Tennessee, USA).[22] The multiplex PCR reaction was carried out in 25-μl volume containing 1× PCR buffer, 2 mM MgCl2, 200 μM each of deoxynucleoside triphosphates, 0.5 μM each of the forward and reverse primers, 0.03units/μl of Q5 high fidelity hot start Taq DNA polymerase (cat. No. M0493S, New England Biolabs), and 1 μl of extracted genomic DNA. The reaction mixture was first treated at 98°C for 30 s, followed by 35 cycles of 98°C for 10 s, 55°C for 30 s, 72°C for 30 s, and final extension at 72°C for 2 min. The amplified products were further subjected to a second round of amplification (nPCR) with internal primers for each marker separately. Restriction enzyme digestion was carried out with 5 μl of nPCR products as described previously.[22] All products were analyzed by 2.5% agarose gel electrophoresis containing 0.3 μg/ml ethidium bromide and viewed using a transilluminator. The restriction enzyme digests of all the reference strains was processed in parallel and run on every gel, so that all test samples could be directly compared on the same gel as the controls. The obtained pattern of digestion was compared with the pattern of genotypes deposited in ToxoDB (http://toxodb.org/toxo/).

RESULTS

Multilocus polymerase chain reaction-restriction fragment length polymorphism and genotypes of Toxoplasma gondii

The characteristics of 25 patients of CT are provided in Table 1. All the patients had shown common symptoms such as headache, neck stiffness, altered sensorium, left hemiplegia, ataxia, and weight loss before death. FFPE tissue blocks of brains were sourced from autopsied brains, DNA was extracted from the areas of the brain showing maximum presence of T. gondii P30 antigen. The parasite genome was genotyped by eight loci PCR-RFLP. Representative gel pictures of eight loci PCR-RFLP on standard strains and 10 samples are provided [Supplementary Figure 1–8]. We wish to state that due to limited DNA available, typing for all samples at all loci could not be accomplished. Of the 25 CT cases, T. gondii from 17 brain specimens could be genotyped at seven loci and eight could be typed at six loci. The obtained pattern of digestion was compared with the toxoplasma database (http://toxodb.org/toxo/). The segregation of alleles between the eight loci was compared to the recognized genotypes in the database and arbitrarily labeled based on its closest similarity in the database. Accordingly, 7 new types were identified and labeled as #232 to #238. These have been submitted to ToxoDB but since their ID is still awaited, these have been labeled as “New.” The RFLP pattern of only one sample (A39/06) was closely corresponding to Genotype II at all loci tested, except at c22-8 locus for which data could not be obtained in spite of repeated attempts. None of the specimens harbored clonal lineages of Types I, II, or III [Table 2].

Table 1

History, clinical signs, and demographical details of acquired immunodeficiency syndrome patients’ with cerebral toxoplasmosis

Table 2

Multilocus polymerase chain reaction-restriction fragment length polymorphism genotyping of T. gondii from cases of cerebral toxoplasmosis in acquired immunodeficiency syndrome

A representative gel picture of SAG1 locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of SAG3 locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of (5’+3’SAG2) locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of BTUB locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of GRA6 locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of L358 locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of Alt. SAG2 locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11; Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

A representative gel picture of C22-8 locus polymerase chain reaction-restriction fragment length polymorphism of standard strains of Toxoplasma gondii and tissue samples of cerebral toxoplasmosis. The enzyme-digested polymerase chain reaction products were electrophoresed on a 2.5% agarose gel. Lane 1: LMW Ladder; Lane 2: A13/02; Lane 3: A89/01; Lane 4: A24/07; Lane 5: A23/07; Lane 6: A23/08; Lane 7: A27/01; Lane 8: A12/10; Lane 9: A45/07; Lane 10: A42/05; Lane 11: M3/11 (no amplification); Lane 12: GT1; Lane 13: PTG; Lane 14: CTG; Lane 15: TgCgCa1; Lane 16: MAS; Lane 17: TgCatBr5; Lane 18: TgCatBr64; Lane 19: TgRsCr1; Lane 20: LMW Ladder

Click here for additional data file.

DISCUSSION

Based on virulence studies in mice, initial T. gondii genotyping studies from France and USA documented three main canonical Types (I, II, and III).[23] Strikingly Genotype II is widely prevalent in France, reflected by its presence in most patients with congenital and ocular toxoplasmosis.[8] In Germany, Type II was the major prevailing genotype irrespective of the clinical presentation of toxoplasmosis.[24]

T. gondii strains infecting South American patients and animals were found to be highly divergent.[15] A recent study examined archival tissues from fatal disseminated toxoplasmosis in Brazilian AIDS patients using multilocus PCR-RFLP genotyping, which yielded six new genotypes including ToxoDB#11,[25] a type previously identified in domestic animals.[13] In-depth DNA sequencing of some of the strains classified as “atypical” by RFLP revealed several new variants.[2627] Accordingly, genotypes that differed from the main clonal types were designated as atypical, recombinant, exotic, or nonarchetypal and were chiefly found in other continents.[23]

Genetic characterization of T. gondii isolates from Asian countries, other than China, is sparse. Chinese studies reveal limited diversity of the parasite. Besides the classical lineages of I, II, and III, Chinese I (ToxDB#9) genotype has been reported as the most common genotype in both animals and patients from mainland China.[152829] One of these studies employed PCR-RFLP to type the parasite at 10 different genetic loci.[28] Interestingly, this genotype has also been reported from other Asian countries, South America, and the USA.[29]

Based on four loci sequencing results, we had earlier concluded that majority of the CT patients in India were infected with recombinants of Type II/III.[19] Using partly the same set of clinical samples and PCR-RFLP on a wider span of genetic loci of the parasite, we now found that the strains were much more diverse than our previous results [Table 2]. Hence, we deduced that these were atypical types exhibiting a high level of genetic diversity, namely, recombination of the three clonal types at different loci. This diversity is better appreciated by comparing the results obtained in the present study using PCR-RFLP on the four loci, namely, Alt. SAG2, SAG3, BTUB, and GRA6 [Supplementary Table 1] versus our earlier results obtained by PCR sequencing on the same four loci, on 21 common samples.[19] The present results segregated the parasite into five different groups, whereas our sequencing results (earlier study) on the same loci differentiated the corresponding isolates into 14 different groups.[19] This was not surprising, considering that new variants and lineages have been discovered by in-depth DNA sequencing of some of the previously RFLP classified atypical strains.[2627] In support of our results is a study, wherein sequencing-based genotyping of T. gondii has been shown to be more accurate in Brazil.[30]

Analysis of four loci polymerase chain reaction-restriction fragment length polymorphism results presented in Table 2

Click here for additional data file.

Hence, our results indicated that the majority of T. gondii causing CT in South India belonged to a nonarchetypal lineage with strains that are recombinants of the mainstream pedigrees I, II, or III at different loci. This being a retrospective postmortem study on formalin-fixed material, culture confirmation of T. gondii from clinical samples, was not possible – hence this is a limitation of the study. Given that the sensitivity of culturing the parasite from clinical samples of CT is extremely low, an alternative approach to determine the true depth of genetic diversity in T. gondii would be whole-genome sequencing of the parasite. Nevertheless, the findings in this study could be considered a forerunner for subsequent studies crucial for mapping the epidemiology of human clinical toxoplasmosis, both in India and globally. Extensive genotyping at many more loci including microsatellites and other housekeeping genes, and on a larger sample size collected from across India, may be required to determine the extent of genetic diversity in our country. A point to note is that, in the present study, we genotyped T. gondii using multilocus PCR RFLP – a technique standardized on isolates seen in North America and Europe. These markers may not be sufficient to type all strains prevalent globally. Hence, new markers may be required to genotype nonarchetypal strains observed across the world. Identifying the genetic diversity may provide an insight into population genetics of T. gondii infection throughout the world.

CONCLUSION

Cases of cerebral toxoplasmosis in AIDS patients in India are found to be associated with atypical genotypes of T. gondii.

Financial support and sponsorship

This study was financially supported by the Indian Council of Medical Research.

Conflicts of interest

There are no conflicts of interest.