FOLIA PARASITOLOGICA 45: 140-148, 1998.

Polymerase chain reaction for diagnosis and species differentiation of

Caspar Franzen1, Andreas Müller1, Pia Hartmann1, Petra Hegener1, Matthias Schrappe2, Volker Diehl1, Gerd Fätkenheuer1 and Bernd Salzberger1

1Department of Internal Medicine I, University of Cologne, D-50924 Cologne, Germany; 2Quality Management, University of Cologne, D-50924 Cologne, Germany

Key words: microsporidia, Enterocytozoon bieneusi, Encephalitozoon, polymerase chain reaction

Abstract. Polymerase chain reaction (PCR) techniques have been developed for the detection of microsporidian DNA in different biological samples. We used sequence data of the rRNA for the identification of Enterocytozoon bieneusi, Encephalitozoon intestinalis, E. cuniculi, and E. hellem in different biological samples of HIV-infected patients by PCR, Southern blot hybridization, restriction endonuclease digestion analysis, cloning, and comparative genetic sequencing. One primer pair was used for amplification of the entire small subunit (SSU)-rRNA gene of E. bieneusi, E. intestinalis, and E. hellem from samples with electron microscopy confirmed infection. The amplified 1.2 kb SSU-rRNA gene fragments were ligated into a pMOSBlue T-vector, transfected into pMOSBlue competent cells, and were used as positive controls. Several primer pairs and hybridization probes were used to amplify and identify microsporidian DNA from different samples. Light microscopical examination of samples was performed in all patients and electron microscopy was done on a subset of patient samples. DNA products were obtained from all samples with confirmed microsporidial infections. The identity of the DNA fragments was determined by Southern blot hybridization or by restriction endonuclease digestion analysis or by DNA sequencing. The results show that PCR is a reliable and sensitive indicator for the presence of microsporidian DNA in different biological samples of HIV-infected patients. PCR can be used further for species differentiation of microsporidia, even between species which cannot be differentiated by light and/or electron microscopy.

Microsporidia are obligate intracellular protozoan smears, bronchoalveolar lavage fluid, sputum, nasal parasites that infect a broad range of vertebrates and discharge, and biopsy tissues. invertebrates. These parasites were first recognized as Although the diagnosis and identification of of silkworms (Naegeli 1857), and had been microsporidia by light microscopy has greatly improved recognized as a cause of disease in many non-human during the last years, species differentiation is often hosts including insects, , and fish (Canning impossible using these techniques. This is critical and Lom 1986) long before they were described as because is effective against human pathogens. Only ten well documented human Encephalitozoon spp. but not against E. bieneusi (Owen microsporidial infections had been reported up to 1985, 1997). Immunfluorescent staining techniques have been when a new species Enterocytozoon bieneusi has been developed for species differentiation of microsporidia, described in an HIV-infected patient from France but used in these procedures are not yet (Desportes et al. 1985, Modigliani et al. 1985). Since available outside research laboratories (Schwartz et al. then many cases of human have been 1994). A variety of serological tests have been reported from all over the world, and microsporidia are developed to detect antibodies to microsporidia but the now recognized as one of the most common pathogens sensitivity and specifity of these tests are unknown and in HIV-infected patients (Weber et al. 1994). they are not suitable methods to diagnose infections in For most infectious diseases, techniques involving immunosuppressed persons (Hollister et al. 1991, Van microbiological isolation and growth characterization Gool et al. 1997). Cell culture systems are used for in offer the most rapid and specific determination of the vitro cultivation of microsporidia but this is not a ethiologic agent. This is not a suitable procedure for suitable technique for routine use (Schwartz et al. 1994, microsporidia, being obligate intracellular parasites that Van Gool et al. 1994a). require cell culture systems for growth. Therefore, During the last 10 years, the diagnosis of agents of diagnosis of human microsporidiosis is dependent on infectious diseases has begun to include the use of the identification of spores in clinical samples, and nucleic acid-based technologies. Diagnosis of parasitic microsporidia have now been described from virtually organisms is the last field of clinical microbiology to every tissue and body fluid in humans including stool incorporate these techniques and ribosomal RNA specimes, duodenal/bile juice, urine, conjunctival (rRNA) data are now increasingly used for diagnosis,

Address for correspondence: C. Franzen, Department of Internal Medicine I, University of Cologne, Joseph-Stelzmann Str. 9, D-50924 Cologne, Germany. Phone: ++ 49 221 4784433; Fax: ++ 49 221 4786456; E-mail: [email protected]

140 Franzen et al.: PCR for diagnosis of microsporidia species differentiation, and phylogenetic analysis of Species-specific primers. The primer pair V1 and EB450 microsporidia (Schuitema et al. 1993, Vossbrink et al. (5’-ACTCAGGTGTTATACTCACGTC-3’) was used for 1993, Zhu et al. 1993, 1994, Weiss et al 1994, Baker et species-specific amplification of a 348-bp DNA fragment of al. 1995, Fedorko et al. 1995, Franzen et al. 1995, the SSU-rRNA gene of E. bieneusi at an annealing temperature of 48°C (Zhu et al. 1993) and the primer pair V1 1996a, 1996b, Da Silva et al. 1996, David et al. 1996, and SI500 (5’-CTCGCTCCTTTACACTCG-3’) was used for Didier et al. 1996a, Katzwinkel-Wladarsch et al. 1996). species-specific amplification of a 370-bp DNA fragment of This paper reviews data generated in our laboratory on the SSU-rRNA gene of E. intestinalis at an annealing the use of molecular-based techniques for diagnosis and temperature of 58°C (Weiss et al. 1994). Two primer pairs species differentiation of microsporidia. EHEL-F/EHEL-R (5’-TGAGAAGTAAGATGTTTAGCA-3’/ 5’-GTAAAAACACTCTCACACTCA-3’) and ECUN-F/ ECUN-R (5’-ATGAGAAGTGATGTGTGTGTGCG-3’/5’- MATERIALS AND METHODS TGCCATGCACTCACAGGCATC-3’) were used for species- Microsporidia and DNA isolation specific amplification of DNA from the SSU-rRNA gene of E. hellem and E. cuniculi, respectively (Visvesvara et al. 1994). Encephalitozoon intestinalis spores were obtained from PCR. Amplifications were done in 50 µl reactions under urine of a patient with transmission electron microscopy the following conditiones: 25 pmol of each primer, 200 µM of confirmed infection and grown as described (Visvesvara et al. each dNTP, 10 mM Tris-HCl, pH 9.0, 50 mM KCl, 1.5 mM 1994, 1995, Franzen et al. 1996c) in MRC-5 human lung MgCl2, and 2.5 U Taq DNA polymerase (Perkin Elmer). One fibroblasts in MEM containing 10% FBS (Sigma). Cells were tenth to 1.0 µg genomic DNA (10 µl) and 40 µl mineral oil fed every three days and the supernatant which contains the were used. Reactions were run in a Perkin-Elmer thermocycler spores was harvested and stored at –20°C until processing. using a step cycle programm. After initial denaturation of the Intestinal biopsies from HIV-infected patient were obtained DNA at 94°C for 3 min, 35 cycles were run: 94°C for 1 min, from the distal duodenum by flexible fibreglass endoscopy. 40-60°C for 2 min, and 72°C for 3 min with a 10 min 72°C Samples were snap frozen in liquid nitrogen and stored at extention after the 35 cycles. A 10 µl aliquot from each -80°C until processing. Stool samples, urine, nasal discharge, reaction was run on a 3% NuSieve 3 : 1 electrophoresis-grade sputum, bronchoalveolar lavage fluid, duodenal/bile juice, and agarose gel (FMC) in 1 × TAE buffer (0.04 M Tris, 0.001 M blood from HIV-infected patients were stored at –20°C until EDTA) with ethidium bromide (0.5 mg/ml) to visualize the amplified PCR-products under UV-illumination (Franzen et al. processing. 1995, 1996a) Biopsies were incubated in digestion buffer with 400 µg Because of the fact that PCR testing is susceptible to cross proteinase K (Qiagen) at 55°C for two hours and DNA was contaminations, procedures for avoiding contamination were prepared using QIAamp spin columns (Qiagen) in an strictly followed. Pre- and post-PCR handling were physically Eppendorf microcentrifuge. Stool samples were concentrated seperated and performed in different rooms and positive by a water-ether sedimentation technique (Van Gool et al. displacement tips were used for all manipulations. Negative 1994b) and other body fluids by centrifugation for 10 minutes controls containing reaction mixtures without DNA were at 2,000 g. Sediments were incubated in digestion buffer with always run. 400 µg proteinase K and 0.4 U chitinase (Sigma) at 55°C for two hours with additional mechanical disruption of the spores Southern Blot Hybridization by glass beads (425-600 µm, Sigma). DNA was prepared After gel electrophoresis PCR-products were denaturated using QIAamp spin columns (Qiagen) in an Eppendorf and neutralized by soaking the gel in denaturation buffer (0.4 microcentrifuge. N NaOH, 0.6 M NaCl) and neutralization buffer (1.5 M NaCl, 0.5 M Tris, pH 7.5), for 30 min each. The separated amplified Polymerase Chain Reaction (PCR) DNA fragments were transferred to positively charged nylon General microsporidian primers. Two primers V1 (5’- membranes (Hybond-N+, Amersham) using a vacuum blotter CACCAGGTTGATTCTGCCTGAC-3’) and 1492 (5’- (Appligene) at 55 mbar for 1 hour in 20 × SSPE (3 M NaCl, GGTTACCTTGTTACGACTT-3’) were used for amplifi- 0.2 M NaH2PO4 · H2O, 1 M EDTA). The DNA was then cation of the entire small subunit (SSU)-rRNA gene of coupled to the membranes by using an UV cross-linker Enterocytozoon bieneusi, Encephalitozoon intestinalis, and E. (Appligene) with 1.2 J/qcm. Membranes were washed in 2 x hellem at an annealing temperature of 50°C. This primer pair SSC (300 mM NaCl, 30 mM Na3Citrate) and prehybridized was designed to amplify a DNA fragment of approximately for 1 hour at 55°C with hybridization buffer (Amersham). 1.2 kb from E. bieneusi, E. intestinalis, and E. hellem infected Internal probes EB150 (5’-TGTTGCGGTAATTTGGTCT tissues (Zhu et al. 1993). The primer pair int 530f (5’- CTGTGTGTAAA-3’) (Zhu et al. 1993) and SI60 (5’- TGCAGTTAAAATGTCCGTAGT-3’) and int 580r (5’- TGTTGATGAACCTTGTGG-3’) (Weiss et al. 1994) were 3’- TTTCACTCGCCGCTACTCAG-3’) was used for amplifi- labeled with fluorescein-11-dUTP using terminal cation of a large portion of the 3’end of the SSU-rRNA gene, deoxynucleotidyl transferase (Amersham), and 15 pmol of the the entire intergenic spacer region, and a small portion of the labeled probes were added to the hybridization buffer and 5’end of the LSU-rRNA gene of E. hellem, E. cuniculi, and E. allowed to hybridize overnight at 55°C. The blots were then intestinalis at an anealing temperature of 40°C (Vossbrinck et washed twice in 2 × SSC with 0.1% SDS, two times in 1 × Friedman 1989). The primer pair V1 and PMP2 (5’- SSC with 0.1% SDS, and in 0.1 × SSC with 0.1% SDS at CCTCTCCGGAACCAAACCTG-3’) was used for 55°C. amplification of DNA from the four microsporidian pathogens Bound fluorescein-11-dUTP labeled oligonucleotides were E. bieneusi, E. hellem, E. cuniculi, and E. intestinalis at an visualised using ECL detection kit (Amersham). Blots were annealing temperature of 60°C (Fedorko et al. 1995). washed in buffer 1 (0.15 M NaCl, 0.1 M Tris, pH 7.5),

141 blocked with buffer 1 containing 0.5% blocking agent by PstI into two fragments. However, reliable (Amersham), incubated with 1 : 1000 diluted anti-fluorescein- differentiation between E. intestinalis and E. hellem horseradish peroxidase conjugate in buffer 2 (0.4 M NaCl, 0.1 could not be accomplished using this primer pair. M Tris, 0.5% bovine serum albumine, pH 7.5) for 30 min, Species-specific sequences could be detected by PCR washed thoroughly in buffer 2 and then incubated with detection solution containing luminol for 1 min. Where the amplification with either V1 and EB450 using plasmids probe bound there is peroxidase-catalyzed oxidation of containing the 1.2 kb DNA fragment from E. bieneusi, luminol to 3-aminophthalate and subsequent enhanced V1 and SI500 using plasmids containing the 1.2 kb chemiluminescence (428 nm) which was autoradiographed on DNA fragment from E. intestinalis, and EHEL-F and Hyperfilm ECL (Amersham) for 1 hour (Franzen et al. 1995, EHEL-R using plasmids containing the 1.2 kb DNA 1996a). fragment from E. hellem as template. No amplification Purification with V1/SI500, EHEL-F/EHEL-R, or ECUN-F/ECUN- The desired bands were excised under UV illumination R was seen when plasmids containing the 1.2 kb DNA from ethidium bromide stained 1.5% agarose gels and DNA fragment from E. bieneusi were used as template. was eluted from the gels using a QIAquick Gel extraction Kit Likewise no bands of the correct size were detected (Qiagen) following the manufacturer’s instructions. using the primers V1/EB450, EHEL-F/EHEL-R, or Restriction Digestion ECUN-F/ECUN-R and plasmids containing the 1.2 kb PCR pruducts amplified using primers V1 and PMP 2 were DNA fragment from E. intestinalis as template, or by digested with with 20 U of PstI (Sigma) or 10 U of HaeIII (Sigma) in a final volume of 15 µl, and products were detected using the primer pairs V1/EB450, V1/SI500, or ECUN- by agarose gel electrophoresis and ethidium bromide staining F/ECUN-R and and plasmids containing the 1.2 kb (Fedorko et al. 1995). DNA fragment from E. hellem as template (Fig. 2). Cloning Using DNA prepared from infected cell cultures, The purified 1.2 kb SSU-rRNA gene fragments of E. biopsy tissues, and other samples with transmission bieneusi, E. hellem, and E. intestinalis were ligated into a electron microscopy confirmed E. bieneusi, E. pMOSBlue T vector (Amersham) in the presence of T4 DNA intestinalis, or E. hellem infection as template, the ligase (Amersham) at 16°C overnight. The plasmids were primer pair V1 and PMP2 produced the expected 250 transfected in pMOSBlue competent cells and plated on bp, 270 bp, and 279 bp DNA fragments. No tetracycline/amplicillin/IPTG/X-gal agar plates. White amplification was seen with target DNA from other colonies were selected and screened by PCR for those parasites like Cryptosporidium or Giardia lamblia, or containing correct inserts (Franzen et al 1996a). other pathogens like Mycobacterium avium and Sequencing . Also no amplification of DNA was Purified PCR products were directly sequenced using 32P seen using template DNA prepared from Toxoplasma labeled primers and a ds cycle sequencing system (Life gondii. E. bieneusi could be differentiated from Technologies) or an automated sequencer (DNA Sequencer 377, Applied Biosystems). Sequences of the DNA fragments Encephalitozoon spp. by the smaller size of the were aligned with other microsporidial rRNA sequences amplified DNA fragment (250 bp vs. 268 to 279 bp) and present in the GenBank data base by using the DNASIS by restriction digestion analysis with PstI but program, version 3.0. differentiation between Encephalitozoon spp. was very uncertain. RESULTS Using DNA prepared from biopsy tissues and stool PCR amplification with the two primers V1 and 1492 samples with transmission electron microscopy produced the expected 1.2 kb DNA fragments on confirmed E. bieneusi infection, amplification with the ethidium bromide stained gels when DNA from samples primers V1 and EB450 produced the expected 348 bp infected with Enterocytozoon bieneusi, Encephalitozoon DNA fragment on ethidium bromide-stained gels. The intestinalis, or E. hellem was used as template. After two primers did not amplify DNA from samples ligation of the DNA fragments into the pMOSBlue T- infected with other microsporidia (E. intestinalis or E. vector, the primer pair V1 and PMP2 amplified the cuniculi) or from tissues infected with other parasites expected 250 bp, 270 bp, or 279 bp DNA fragments like Cryptosporidium or G. lamblia, or other pathogens from all plasmids containing DNA of E. bieneusi, E. like M. avium and C. albicans. Also no amplification of intestinalis, or E. hellem, respectively. Differentiation of DNA was seen using template DNA prepared from T. E. bieneusi from Encephalitozoon spp. was possible by gondii. The nature of the PCR-products was confirmed the different size of the amplified DNA products (250 by Southern blot hybridization with the internal probe bp vs. 270 to 279 bp) and by restriction digestion EB150. The probe hybridized with all 348 bp fragments analysis with PstI and HaeIII. E. bieneusi does not have amplified from E. bieneusi infected samples, indicating a PstI restriction site in the amplified area whereas the that the amplified 348 bp fragments were derived from DNA fragments of the Encephalitozoon spp. were cut the SSU-rRNA gene of E. bieneusi.

142 Franzen et al.: PCR for diagnosis of microsporidia

Table 1. Primer pairs and hybridization probes. Primer Probe de- Amplicon size Primer pair Hybridization probe Species Target gene References designation signation in base pairs 5´-CACCAGGTTGATTCTGCCTGAC-3´ V11492 Enterocytozoon bieneusi SSU-rRNA About 1,200 Zhu et al. 1993 5'-GGTTACCTTGTTACGACTT-3' Encephalitozoon cuniculi Vossbrinck et al. Encephalitozoon intestinalis 1989 Encephalitozoon hellem 5´-CACCAGGTTGATTCTGCCTGAC-3´ PMP1 (V1) Enterocytozoon bieneusi SSU-rRNA 250 Fedorko et al. 1995 5´-CCTCTCCGGAACCAAACCTG-3´ PMP2 Encephalitozoon cuniculi 268 Encephalitozoon intestinalis 270 Encephalitozoon hellem 279 5´-CACCAGGTTGATTCTGCCTGAC-3 V1 5´-TGTTGCGGTAATTTGGTCTCT EB150 Enterocytozoon bieneusi SSU-rRNA 348 Zhu et al. 1993 5´-ACTCAGGTGTTATACTCACGTC-3´ EB450 GTGTGTAAA-3´ 5´-CACCAGGTTGATTCTGCCTGAC-3´ V1 5´-TGTTGATGAACCTTGTGG-3´ SI60 Encephalitozoon intestinalis SSU-rRNA 370 Weiss et al. 1994 5´-CTCGCTCCTTTACACTCGAA-3´ SI500 5´-TGAGAAGTAAGATGTTTAGCA-3´ EHEL-F Encephalitozoon hellem SSU-rRNA 547 Visvesvara et al. 5´-GTAAAAACACTCTCACACTCA-3´ EHEL-R 1994 5´-ATGAGAAGTGATGTGTGTGTGCG-3´ ECUN-F Encephalitozoon cuniculi SSU-rRNA 549 Visvesvara et al. 5´-TGCCATGCACTCACAGGCATC-3´ ECUN-R 1994

5´-TGCAGTTAAAATGTCCGTAGT-3´ Int530f 5´-TAGCGGCTGACGAAGCTGC-3´ Encephalitozoon cuniculi SSU-rRNA About 1,000 Schuitema et al. 1993 5´-TTTCACTCGCCGCTACTCAG-3´ Int580r 5´-TGAGTGTGAGAGTGTTTTTACAT-3´ Encephalitozoon hellem ISR Schuitema et al. 1993 5´-GGGCAGGAGAACGAGGACGG-3´ Encephalitozoon intestinalis LSU-rRNA Didier et al. 1996a SSU – small subunit; LSU – large subunit; ISR – intergenic spacer region

Fig. 1. Diagram of the small and large subunit rDNA with the intergenic spacer region. The positions of the used primers are indicated.

143 Table 2. Number of 5‘-GTTT-3‘ repeats in the intergenic from five biopsy specimens, too. A DNA fragment of spacer region of different Encephalitozoon cuniculi isolates correct size (370 bp) could be amplified also by using from different hosts and geographical areas. DNA prepared from a biopsy with transmission electron No. of No. of microscopy and confirmed E. cuniculi Geographic area Reference repeats isolates infection. After Southern blotting and hybridization 2 4 - blue fox Norway Mathis et al. 1997 with the probe SI60 all amplified 370 bp DNA 2 - mouse Czech. Rep., UK Didier et al. 1995 fragments hybridized with the probe. Partial sequencing 3 11 - rabbit Swizerland Mathis et al. 1997 of these 370 bp DNA fragments which were amplified 3 - rabbit USA Didier et al. 1995 from tissue infected with E. bieneusi and E. cuniculi 1 - mouse USA Didier et al. 1995 obtained 215 bases for comparison (GenBank accession 6 - human Switzerland Deplazes et al. 1996a, No. U39297). Alignment of these partial sequences 1996b against sequences in the GenBank database showed 4 3 - human Mexico, USA Deplazes et al. 1996b, high genetic homology (100%) with previously Katiyar et al. 1995, published sequences of E. intestinalis. Other matches Didier et al. 1996b against E. hellem, E. cuniculi, and E. bieneusi showed 2 - dog USA Didier et al. 1995 only limited homology.

Species differentiation between E. hellem and E. Using DNA prepared from cell cultures infected with cuniculi was done using two specific primer pairs. E. intestinalis or from biopsy tissues and other samples EHEL-F/EHEL-R amplifies DNA from different including blood of patients with confirmed E. samples (urine, nasal discharge, sputum, stool) with E. intestinalis infection, strong bands of correct size (370 hellem infection but not from samples with E. bp) could be amplified with the primer pair V1 and intestinalis, E. cuniculi or E. bieneusi infection. On the SI500. Using this primer pair with DNA prepared from other hand ECUN-F/ECUN-R did not amplify DNA biopsy tissues with transmission electron microscopy from samples with E. hellem, E. intestinalis, or E. and PCR confirmed E. bieneusi infection, DNA bieneusi infection but from samples with E. cuniculi fragments of correct size (370 bp) could be amplified infection. For further confirmation of a case with E.

Fig. 2. Ethidium bromide stained gel of samples amplified with three different primer sets. Lane 1 – 100 bp DNA ladder; lane 2 – biopsy with Enterocytozoon bieneusi infection with V1 and 1492; lane 3 – biopsy with Encephalitozoon intestinalis infection with V1 and 1492; lane 4 – pMOSBlue containing E. bieneusi SSU-rRNA with V1 and SI500; lane 5 – pMOSBlue containing E. bieneusi SSU-rRNA with V1 and EB450; lane 6 – pMOSBlue containing E. intestinalis SSU-rRNA with V1 and SI500; lane 7 – pMOSBlue containing E. intestinalis SSU-rRNA with V1 and EB450; lanes 8-11 – biopsies with E. intestinalis infection with V1 and SI500; lane 12 – biopsy with E. cuniculi infection with V1 and SI500; lanes 13-18 – biopsies with E. bieneusi infection with V1 and SI500; lane 19 – 100 bp DNA ladder.

144 Franzen et al.: PCR for diagnosis of microsporidia hellem infection the 3’end of the SSU-rRNA gene, the DNA from E. bieneusi infected tissues, but also E. intergenic spacer region, and the 5’end of the LSU- hellem DNA from cell cultures. Specificity was tested rRNA gene were amplified using primers int530f and with DNA prepared from Trypanosoma cruzi, E. coli, int580r. Sequencing of the amplified 1.0 kb DNA Saccharomyces cerevisiae, E. intestinalis, V. necatrix, products obtained 679 bases for comparison. Alignment E. cuniculi, G. stephani, Nosema locustae, N. bombycis, of these sequences against sequences in the GenBank Pleistophora sp., and E. hellem and only weak database showed high genetic homology (99.8%) with amplification of DNA prepared from E. hellem was seen previously published sequences of E. hellem. Other (Zhu et al. 1993). In other studies the primer set did not matches against E. cuniculi, E. intestinalis, and E. amplify DNA from E. hellem (Coyle et al. 1996) and the bieneusi showed only limited homology. SSU-rRNA sequence of E. hellem does not suggest that this would occur. It is likely that the amplification DISCUSSION observed by Zhu et al. (1993) was due either to The most frequent intestinal microsporidian species contamination of the original E. hellem culture with Enterocytozoon bieneusi occurs in about 10 to 30% of spores of E. bieneusi (which were being used in the HIV-infected patients with diarrhea but infections with same incubator) or to contamination of the culture with Encephalitozoon species are increasingly recognized as a plasmid containing the cloned E. bieneusi SSU-rRNA pathogens in HIV-infected patients (Weber et al. 1994). gene (Coyle et al. 1996). We have shown that the Therefore, pan-microsporidian primer pairs are very V1/EB450 primer pair reliably amplifies E. bieneusi useful for screening of clinical samples for DNA from gastrointestinal biopsy specimens with light microsporidia. The primer pair PMP1 (the nucleotide and electron microscopy confirmed E. bieneusi sequence of this primer is identical to that of primer V1) infection and from stool specimens with intestinal and PMP2 that amplifies four microsporidian pathogens microsporidiosis, as confirmed by light and/or electron that infect humans (E. bieneusi, Encephalitozoon microscopy (Franzen et al. 1995). We further tested hellem, E. cuniculi, and E. intestinalis) and also tissues infected with E. cuniculi and E. intestinalis and Vittaforma corneae from culture has been reported by with other parasites and bacteria such as Fedorko et al. (1995). They used DNA extracted from Cryptosporidium sp, Giardia lamblia, Candida albicans cultured organisms and stool samples but we have and Mycobacterium avium, and DNA isolated from shown that this primer pair is also useful for the Toxoplasma gondii, and no amplification was seen. This detection of microsporidian DNA in different other suggests that the primer pair V1 and EB450 only clinical specimens (nasal discharge, urine, amplifies DNA from the microsporidian species E. bronchoalveolar lavage, sputum, stool samples, bieneusi. Similar results with this primer pair have been intestinal biopsy tissue). To confirm the identity of PCR reported by others (Ombrouck et al. 1997) but it has amplicons restriction enzyme digestion with PstI and been also reported that this primer pair gave negative HaeIII was used. The amplicons differ in size and E. results with one bile sample that was confirmed for the bieneusi does not have a PstI restriction site in the presence of E. bieneusi by electron microscopy amplified region but PstI cuts the Encephalitozoon evaluation and with E. bieneusi derived from a short species amplicons into two fragments. This enabled term culture (Da Silva et al. 1996). Resequencing of the ready differentiation of E. bieneusi and E. hellem from SSU-rRNA region of different E. bieneusi isolates E. intestinalis and E. cuniculi but did not allow the showed a 2.3% dissimilarity with the sequence reported differentiation of E. intestinalis from E. cuniculi, by Zhu et al. (1993), resulting in a base mismatch of the although differentiation between the three primer EB450 (position 9 of the EB 450 primer has a G Encephalitozoon species can be difficult, thus limitating in place of a C) (Da Silva et al. 1996). As mentioned the use of this procedure for species differentiation. above the foreward primer V1 is not specific for E. The data presented here show that the primer pairs bieneusi but is directed against a conserved sequence V1/EB450, V1/SI500, EHEL-F/EHEL-R, and ECUN- similar to many microsporidian species (Weiss et al. F/ECUN-R can be used for species differentiation of the 1994). In combination with the base mismatch of primer four major human microsporidian species. The foreward EB450 this may lead to negative results with some primer V1 based on the conserved sequence of the SSU- samples (Da Silva et al. 1996). rRNA gene of Vairimorpha necatrix at the 5’ end We have shown that the primer pair V1 and SI500 (Vossbrinck et al. 1987) but the reverse primers EB450 described by Weiss et al. (1994) amplifies cloned E. and SI500 are located in highly variable areas at intestinalis SSU-rRNA sequences and DNA from E. position 450 and 500 of the SSU-rRNA gene based on intestinalis infected tissues, body fluids, stool samples, the alignment of E. bieneusi, E. intestinalis and blood, and cell cultures, but did not amplify cloned E. Eschirichia coli (Zhu et al. 1993, Weiss et al. 1994). bieneusi or E. hellem SSU-rRNA sequences. Specificity PCR diagnosis of E. bieneusi was first reported by of this primer pair was further tested with DNA Zhu et al. (1993). The primer pair V1 and EB450 prepared from E. hellem, E. cuniculi, V. necatrix, G. amplified cloned E. bieneusi SSU-rRNA sequences and stephani, N. locustae, N. bombycis, Pleistophora sp., S.

145 cerevisiae, and E. coli and no amplification was includes a large portion of the SSU-rRNA gene, the observed (Coyle et al. 1996). Surprisingly we found that entire intergenic spacer, and a small portion of the LSU- the primer pair V1 and SI500 amplified also DNA rRNA gene of E. hellem, E. cuniculi, and E. intestinalis, fragments of correct size from tissues with transmission but did not amplify DNA from culture derived V. electron microscopy and PCR confirmed E. bieneusi corneae or intestinal biopsy derived E. bieneusi (Didier and E. cuniculi infection. As mentioned above the et al. 1996a). Species differentiation can be done either foreward primer V1 is not specific for E. intestinalis but by Southern blot hybridization using hybridization is directed against a conserved sequence similar to many probes specific for E. cuniculi, E. hellem, or E. microsporidian species (Weiss et al. 1994). Several intestinalis (Didier et al. 1996a) or as we described by confirmation techniques were used to determine the DNA sequencing. specificity of the primer pair V1 and SI500 and to The primer pair, int530f and int580r, has been also exclude the possibility of an amplification of the DNA used for identification of three different subtypes of E. fragments as a result of unspecific annealing of the cuniculi. Didier et al. (1995) first demonstrated isolate primer SI500. On one side hybridization with the E. differences in the intergenic spacer region of eight E. intestinalis specific internal probe SI60, which is cuniculi isolates from mice, rabbits, and dogs. After located at position 60 of the SSU-rRNA gene of E. DNA sequencing of the intergenic spacer region the intestinalis based on the alignment of E. intestinalis and isolates were shown to differ from each other by a small E. coli (Weiss et al. 1994), and on the other side partial repetitive sequence of 5’-GTTT-3’. This sequence was sequencing of the DNA fragments showed high repeated twice in two isolates from mice, three times in homology with published E. intestinalis sequences and three isolates from rabbits and in one isolate from a confirmed that the amplified PCR-product really mouse, and four times in the two isolates from domestic derived from the SSU-rRNA gene of E. intestinalis. dogs. These differences were used to characterize a Furthermore, we have shown that the primer pair rabbit strain (strain I), a mouse strain (strain II), and a V1/SI500 did not amplify cloned DNA from E. bieneusi dog strain (strain III) (Didier et al. 1995). Several or E. hellem. Double infections with two different types different isolates of E. cuniculi have now been of microsporidia had been reported previously examined (Didier et al. 1995, 1996b, Deplazes 1996a, (Blanshard et al. 1992), but subclinical infection with E. Hollister et al. 1996, Mathis et al. 1997) and all isolates intestinalis was not expected until similar results were from humans correspond to the dog strain III with four obtained by Van Gool et al. (1994a) who cultivated E. tetranucleotide repeats or to the rabbit strain I with three intestinalis from stool of five HIV-infected patients with tetranucleotide repeats (table 2). This might suggest that transmission electron microscopy confirmed E. bieneusi dogs or rabbits are the source of infection to man, but it infection. Similar to our group they were not able to is difficult to understand how host related differences visualize the parasites by light or electron microscopy. can be maintained when the strains can be transmitted The results presented here show that the primer pair V1 between hosts, at least experimentally. Too few isolates and SI500 and the hybridization probe SI60 could be of E. cuniculi have been obtained and analyzed so far to able to detect very light infections with E. intestinalis. It assess host specificity or to determine if humans are at seems that E. intestinalis occurs more frequently in risk for becoming infected by exposure to dogs or form of latent infections and suggests that the true rabbits. More data on strains from many different hosts prevalence of the parasites may be much higher than the including man are needed before the epidemiology of E. previously reported 2-4%. cuniculi is clarified and to determine that encephalito- The two primer pairs ECUN-F/ECUN-R, on the basis is a zoonotic disease. of positions 344 to 364 and 872 to 892 of the E. cuniculi The results generated in our laboratory show that SSU-rRNA sequence, and EHEL-F/EHEL-R, on the PCR is a reliable and sensitive indicator for the presence basis of positions 358 to 378 and 884 to 904 of the E. of microsporidian DNA in different biological samples hellem SSU-rRNA sequence, were first reported by of HIV-infected patients. PCR can be used further for Visvesvara et al. (1994). They used the primer pairs for species differentiation of microsporidia, even between species differentiation of cultured organisms species which cannot be differentiated by light and/or (Visvesvara et al. 1994, De Groote et al. 1995), but we electron microscopy. Furthermore, application of PCR have shown that these primers are also useful for to stool samples offers an approach that can be adapted species differentiation of microsporidia in various for processing large numbers of specimens to define the clinical samples (nasal discharge, urine, bronchoal- epidemiology extent of microsporidial infection in veolar lavage, sputum, stool). human populations (Owen 1997). The pan-Encephalitozoon primers, int530f and int580r, selected from sequences conserved between E. Acknowledgement. This work was supported by the Köln cuniculi and E. hellem (Vossbrinck et al. 1993), amplify Fortune programe/Faculty of Medicine, University of a product of approximately 1,000 base pairs that Cologne.

146 Franzen et al.: PCR for diagnosis of microsporidia

REFERENCES BAKER M.D., VOSSBRINCK C.R., DIDIER E.S., DIDIER E.S., ROGERS L.B., BRUSH A.D., WONG S., MADDOX J.V.,SHADDUCK J.A. 1995: Small subunit TRAINA-DORGE V., BERTUCCI D. 1996a: Diagnosis ribosomal DNA phylogeny of various microsporidia with of disseminated microsporidian Encephalitozoon hellem emphasis on AIDS related forms. J. Euk. Microbiol. 42: infection by PCR-Southern analysis and successful 564-570. treatment with albendazole and fumagillin. J. Clin. BLANSHARD C., HOLLISTER W.S., PEACOCK C.S., Microbiol. 34: 947-952. TOVEY D.G., ELLIS D.S., CANNING E.U., GAZZARD DIDIER E.S., VISVESVARA G.S., BAKER M.D., ROGERS B.G. 1992: Simultaneous infection with two types of L.B., BERTUCCI D.C., DE GROOTE M.A., intestinal microsporidia in a patient with AIDS. Gut 33: VOSSBRINK C.R. 1996b: A microsporidian isolated from an AIDS patient corresponds to Encephalitozoon cuniculi 418-420. III, originally isolated from domestic dogs. J. Clin. CANNING E.U., LOM J. 1986: The Microsporidia of Microbiol. 34: 2835-2837. Vertebrates. Academic Press, Inc., New York. FEDORKO D.P., NELSON N.A., CARTWRIGHT C.P. 1995: COYLE C.M., WITTNER M., KOTLER D.P., NOYER C., Identification of microsporidia in stool specimens by using ORENSTEIN J.M., TANOWITZ H.B., WEISS L.M. PCR and restriction endonucleases. J. Clin. Microbiol. 33: 1996: Prevalence of microsporidiosis due to 1739-1741. Enterocytozoon bieneusi and Encephalitozoon (Septata) FRANZEN C., MÜLLER A., HEGENER P., SALZBERGER intestinalis among patients with AIDS-related diarrhea: B., HARTMANN P., FÄTKENHEUER G., DIEHL V., determination by polymerase chain reaction to the SCHRAPPE M. 1995: Detection of microsporidia microsporidian small-subunit rRNA gene. Clin. Infect. (Enterocytozoon bieneusi) in intestinal biopsy specimens Dis. 23: 1002-1006. from human immunodeficiency virus-infected patients by DA SILVA A.J., SCHWARTZ D.A., VISVESVARA G.S., PCR. J. Clin. Microbiol. 33: 2294-2296. DE MOURA H., SLEMENDA S.B., PIENIAZEK N.J. FRANZEN C., KÜPPERS R., MÜLLER A., SALZBERGER 1996: Sensitive PCR diagnosis of infections by B., FÄTKENHEUER G., VETTEN B., DIEHL V., Enterocytozoon bieneusi (Microsporidia) using primers SCHRAPPE M. 1996a: Genetic evidence for latent based on the region coding for small-subunit rRNA. J. Septata intestinalis infection in human immunodeficiency Clin. Microbiol. 34: 986-987. virus-infected patients with intestinal microsporidiosis. J. DAVID F., SCHUITEMA A.R.J., SARFATI C., LIGUORY Infect. Dis. 173: 1038-1040. O., HARTSKEERL R.A., DEROUIN F., MOLINA J. FRANZEN C., MÜLLER A., HEGENER P., HARTMANN 1996: Detection and species identification of intestinal P., SALZBERGER B., FRANZEN B., DIEHL V., FÄTKENHEUER G. 1996b: Polymerase chain reaction microsporidia by polymerase chain reaction in duodenal for microsporidian DNA in gastrointestinal biopsy biopsies from human immunodeficiency virus-infected specimens of HIV-infected patients. AIDS 10: F23-27. patients. J. Infect. Dis. 174: 874-877. FRANZEN C., MÜLLER A., FÄTKENHEUER G., DE GROOTE M.A., VISVESVARA G., WILSON M.L., SALZBERGER B., MAHRLE G., DIEHL V., PIENIAZEK N.J., SLEMENDA S.B., DA SILVA A.J., SCHRAPPE M. 1996c: In vitro Kultur von LEITCH G.J., BRYAN R.T., REVES R. 1995: Encephalitozoon intestinalis isoliert von einem HIV- Polymerase chain reaction and culture confirmation of infizierten Patienten mit disseminierter Microsporidiose. disseminated Encephalitozoon cuniculi in a patient with 17. Tagung der Deutschen Gesellschaft für Parasitologie, AIDS: successful therapy with albendazole. J. Infect. Dis. München, 27-29 March 1996. 171: 1375-1378. HOLLISTER W.S., CANNING E.U., WILLCOX A. 1991: DEPLAZES P., MATHIS A., BAUMGARTNER R., Evidence for widespread occurrence of antibodies to TANNER I., WEBER R. 1996a: Immunologic and Encephalitozoon cuniculi (Microspora) in man provided molecular characteristics of Encephalitozoon-like by ELISA and other serological tests. Parasitology 102: microsporidia isolated from humans and rabbits indicate 33-43. that Encephalitozoon cuniculi is a zoonotic parasite. Clin. KATZWINKEL-WLADARSCH S., LIEB M., HELSE W., Infect. Dis. 22: 557-559. LÖSCHER T., RINDER H. 1996: Direct amplification DEPLAZES P., MATHIS A., MÜLLER C., WEBER R. and species determination of microsporidian DNA from 1996b: Molecular epidemiology of Encephalitozoon stool specimens. Trop. Med. Int. Hlth. 1: 373-378. cuniculi and first detection of Enterocytozoon bieneusi in MODIGLIANI R., BORIES C., LE CHARPENTIER Y., faecal samples of pigs. J. Euk. Microbiol. 43: 93S. SALMERON M., MESSING B., GALIAN A., RAMBAUD J.C., LAVERGNE A., COCHAND DESPORTES I., LE CHARPENTIER Y., GALIAN A., PRIOLLET B., DESPORTES I. 1985: Diarrhoea and BERNARD F., COCHAND PRIOLLET B., LAVERGNE malabsorption in acquired immune deficiency syndrome: a A., RAVISSE P, MODIGLIANI R. 1985: Occurrence of a study of four cases with special emphasis on opportunistic new microsporidan: Enterocytozoon bieneusi n. g., n. sp., protozoan infestations. Gut 26: 179-187. in the enterocytes of a human patient with AIDS. J. NAEGELI K. W. 1857: Über die neue Krankheit der Protozool. 32: 250-254. Seidenraupe und verwandte Organismen. Bot. Z. 15: 760- DIDIER E. ., VOSSBRINK C.R., BAKER M.D., ROGERS 761. L.B., BERTUCCI D.C., SHADDUCK. J.A. 1995: OMBROUCK C., CICERON L., BILIGUI S., BROWN S., Identification and characterization of three MARECHAL P., VAN GOOL T., DATRY A., DANIS Encephalitozoon cuniculi strains. Parasitology 111: 411- M., DESPORTES-LIVAGE I. 1997: Specific PCR assay 421. for direct detection of intestinal microsporidia

147 Enterocytozoon bieneusi and Encephalitozoon intestinalis monoclonal antibody and PCR-amplified small-subunit in fecal specimens from human immunodeficiency virus- rRNA identification of a microsporidian, Encephalitozoon infected patients. J. Clin. Microbiol 35: 652-655. hellem, isolated from an AIDS patient with disseminated OWEN R.L. 1997: Polymerase chain reaction of stool: a infection. J. Clin. Microbiol. 32: 2760-2768. poweful tool for specific diagnosis and epidemiologic VOSSBRINCK C.R., BAKER M.D., DIDIER E.S., investigation of enteric microsporidia infections. AIDS 11: DEBRUNNER VOSSBRINCK B.A., SHADDUCK J.A. 817-818. 1993: Ribosomal DNA sequences of Encephalitozoon SCHUITEMA A.R. J., HARTSKEERL R.A., VAN GOOL T., hellem and Encephalitozoon cuniculi: species LAXMINARAYAN R., TERPSTRA W.J. 1993: identification and phylogenetic construction. J. Eukaryot. Application of the polymerase chain reaction for the Microbiol. 40: 354-362. diagnosis of microsporidiosis. AIDS 7 (Suppl. 3): S62- VOSSBRINCK C.R., FRIEDMAN S. 1989: A 28S ribosomal S63. RNA phylogeny of certain cyclorrhapous Diptera based on SCHWARTZ D.A., BRYAN R.T., WEBER R., a hypervariable region. Syst. Entomol. 14: 417-431. VISVESVARA G.S. 1994: Microsporidiosis in HIV WEBER R., BRYAN R.T., SCHWARTZ D.A., OWEN R.L. positive patients: current methods for diagnosis using 1994: Human microsporidial infections. Clin. Microbiol. biopsy, cytologic, ultrastructural, immunological, and Rev. 7: 426-461. tissue culture techniques. Folia Parasitol. 41: 101-109. WEISS L.M., ZHU X., CALI A., TANOWITZ H.B., VAN GOOL T., CANNING E.U., GILIS H., VAN DEN WITTNER M. 1994: Utility of microsporidian rRNA in BERGH WEERMAN M.A., EEFTINCK diagnosis and phylogeny: a review. Folia Parasitol. 41: 81- SCHATTENKERK J.K., DANKERT J. 1994a: Septata 90. intestinalis frequently isolated from stool of AIDS patients ZHU X., WITTNER M., TANOWITZ H.B., KOTLER D., with a new cultivation method. Parasitology 109: 281-289. CALI A., WEISS L.M. 1993: Small subunit rRNA VAN GOOL T., CANNING E.U., DANKERT J. 1994b: An sequence of Enterocytozoon bieneusi and its potential improved practical and sensitive technique for the diagnostic role with use of the polymerase chain reaction. detection of microsporidian spores in stool samples. Trans. J. Infect. Dis. 168: 1570-1575. R. Soc. Trop. Med. Hyg. 88: 189-190. ZHU X., WITTNER M., TANOWITZ H.B., CALI A., WEISS VISVESVARA G.S., LEITCH G.J., DA SILVA A.J., L.M. 1994: Ribosomal RNA sequences of Enterocytozoon CROPPO G.P., MOURA H., WALLACE S., bieneusi, Septata intestinalis and Ameson michaelis: SLEMENDA S.B., SCHWARTZ D.A., MOSS D., phylogenetic construction and structural correspondence. BRYAN R.T., PIENIAZEK N.J. 1994: Polyclonal and J. Euk. Microbiol. 41: 204-209.

Received 16 July 1997 Accepted 30 October 1997

148