TEXT

Top Abstract Text References

Encephalitozoon cuniculi, Encephalitozoon hellem, and Encephalitozoon intestinalis are strict intracellular protozoan parasites responsible for emerging microsporidioses in human immunodeficiency virus-infected patients (14). To date, they have been isolated from only a few clinical samples by using cultured cell lines (2, 14). Screening of drugs with potential anti-Encephalitozoon activity has also been based on cell culture systems (1, 6). Cell cultures can be contaminated by mycoplasmas (5), and a microsporidial strain can be contaminated from the first stage of its isolation or during its propagation, as has been found in our experience. No protocol for mycoplasma decontamination of microsporidial cultures has been published. We herein present such a protocol based on exploiting the differential cell targeting of mycoplasmas and microsporidia after inoculation in mice.

E. hellem- and E. cuniculi-inoculated Vero cell CCL-81 (American Type Culture Collection, Rockville, Md.) cultures were determined to be mycoplasma positive by using a mycoplasma detection kit (Boehringer, Mannheim, Germany). Twelve 18- to 20-g BALB/c female mice (Charles River, Saint-Aubin lès Elbeuf, France) distributed into three groups were inoculated intraperitonally with 0.5 ml of a 10⁹ E. cuniculi spore suspension (group 1), 0.5 ml of a 10⁹ E. hellem spore suspension (group 2), or 0.5 ml of cell culture medium as a negative control group (group 3); each animal received in addition no less than 10⁶ CFU of mycoplasma cells per ml. The brains and spleens of the animals were collected aseptically 3 h postinoculation (p.i.) (two mice per group) or 24 h p.i. (two mice per group) for direct examination after Gram-Weber staining (8) and centrifugation at 3,500 rpm (Avanti 30 compact centrifuge; Beckman Instruments, Gagny, France) for 1 h at room temperature on mycoplasma-free Vero cell cultures in shell vials. The shell vials were incubated at 37°C under a 5% CO₂ atmosphere, and culture medium was changed weekly. Identification of microsporidia in tissues and Vero cell cultures was performed by direct sequencing of PCR-amplified microsporidial 16S rRNA genes after QIAmp tissue kit (QIAGEN SA, Courtaboeuf, France) nucleic acid extraction. Consensus primers M1-22 (5'-CACCAGGTTGATTCTGCCTGAC-3') and M331-315 (5'-CAAGTTTCGCGCCTGCT-3') were designed to amplify a 331-bp region in E. cuniculi (GenBank accession no. L17072) (15), E. hellem (accession no. L19070), and E. intestinalis (GenBank accession no. U09929). PCR was performed with a final volume of 50 µl in a Perkin-Elmer 9600 thermocycler, by using the GeneAmp kit (Perkin-Elmer Cetus, Norwalk, Conn.). An initial denaturation step at 94°C for 10 min was followed by three cycles consisting of 94°C for 30 s, 40°C for 30 s, and 60°C for 30 s; 35 cycles consisting of 94°C for 30 s, 60°C for 30 s, and 72°C for 30 s; and a final hold at 72°C for 10 min.

Positive and negative controls consisted of E. hellem DNA and sterile, distilled water, respectively. Amplicons were sequenced in both directions by using primers M1-22 and M331-315 and the ABI Prism dRhodamine Dye Terminator Cycle sequencing ready reaction kit (Perkin-Elmer) after 25 cycles consisting of 95°C for 20 s, 60°C for 10 s, and 60°C for 4 min. Sequencing products resolved by electrophoresis in a 0.2-mm, 6% polyacrylamide denaturating gel were recorded with an ABI Prism 377 DNA sequencer (Applied Biosystems) as described in the protocol of the supplier. Data processed by Sequence Analysis software (Applied Biosystems) were compared to GenBank sequences by using PC software (Intelligenetics).

No clinical difference was noted between negative control and microsporidium-inoculated mice. Microsporidia were observed in brain and spleen tissues of inoculated mice 24 but not 3 h p.i. (Fig. 1), and no microsporidia were detected in the tissue of negative control mice (Table 1). Microsporidia were cultured 6 days p.i. Amplification of a 331-bp fragment of the microsporidial 16S rRNA gene was achieved for brain and spleen tissues collected 3 and 24 h p.i. for eight of eight microsporidium-inoculated mice but zero of four negative controls (Fig. 2). Sequences of amplified 16S rRNA fragments exhibited 100% similarity with that deposited in GenBank for the homologous Encephalitozoon species. Mycoplasmas could not be detected in Vero cell cultures inoculated with the brains or spleens of mice over a 6-month period.

View larger version (97K): [in this window] [in a new window]   FIG. 1.   Gram-Weber staining of the brain (top panel) and the spleen (bottom panel) collected from a mouse inoculated intraperitoneally with E. cuniculi for 24 h. Arrows indicate microsporidia.

View larger version (33K): [in this window] [in a new window]   FIG. 2.   Ethidium bromide-stained agarose gel of a PCR-amplified microsporidial 16S rRNA gene fragment. Lanes A and J, molecular size marker V (Boehringer); lane B, sterile distilled water; lane C, noninoculated mouse brain tissue; lane D, noninoculated mouse spleen tissue; lane E, E. hellem DNA as positive control; lane F, E. cuniculi-inoculated mouse brain tissue; lane G, E. cuniculi-inoculated mouse spleen tissue; lane H, E. hellem-inoculated mouse brain tissue; lane I, E. hellem-inoculated mouse spleen tissue. The molecular size in base pairs is indicated by an arrow to the left of the figure.

Mycoplasma contamination represents a serious threat to any cell culture. Its elimination is greatly impeded when coculture represents the only method of growing an intracellular microorganism. No procedure was available for the decontamination of microsporidium-inoculated cell culture, and filtration systems were ineffective in mycoplasma elimination (9). Treatment of contaminated cell lines with a mycoplasma- removing agent (11), merocyanine 540, Hoechst 33257 (12), or 5-bromouracil (7) is limited by their potential activity against Encephalitozoon species. Tetracyclines and fluoroquinolones are effective against mycoplasmas (9) but also inhibit E. cuniculi. Effective decontamination of Rickettsia-inoculated cell cultures has been reported after mouse inoculation based on differential target tissue specificity (3). We hypothesized that such tissue specificity could be applied to the purification of contaminated cultures of Encephalitozoon species. Mycoplasmas are epicellular bacteria associated with epithelium-covered organs (9). In contrast, Encephalitozoon species are strict intracellular parasites (13); data from experimental and natural E. cuniculi infections in mice (4, 10) indicated perivascular cerebral and splenic granulomas. Clinical observations and observations at the time of autopsy of AIDS patients with disseminated E. cuniculi infection produced similar data (14). By using our protocol, Encephalitozoon strains were detected and cultured in brain and spleen tissues from inoculated mice 24 h p.i., confirming systemic infection and intravascular dissemination after intraperitoneal inoculation. Since laboratory mice may be naturally infected with E. cuniculi (4), we carefully checked the absence of microsporidia in negative control mice and confirmed the identity of microsporidia recovered after mouse inoculation by molecular characterization.

The centrifugation shell vial technique and direct sequencing of PCR-amplified 16S rRNA gene proved to be effective and innovative tools in the recovery and identification of Encephalitozoon strains. Although a small number of animals were used in this study, clear-cut results suggest that the use of laboratory animals during the primary isolation of microsporidia from environmental and clinical specimens can prevent contamination. We therefore propose that this protocol be used in cases of mycoplasma contamination of microsporidium-inoculated cell cultures.