Rapid PCR–Single-Strand Conformation Polymorphism Method To Differentiate and Estimate Relative Abundance of Pneumocystis carinii Special Forms Infecting Rats

Abstract

A rapid method that uses PCR–single-strand conformation polymorphism analysis of the intron of the nuclear 26S rRNA gene was shown to differentiate the two Pneumocystis carinii special forms that infect rats, P. carinii f. sp. carinii and P. carinii f. sp. ratti. The method also provides a means for estimation of the relative abundance of the two special forms in the case of a coinfected rat. The results suggest that the method described will help to further standardize the immunosuppressed rat model of P. carinii infection and, thus, contribute to a better understanding of P. carinii infection in humans.

Pneumocystis carinii represents a diverse family of atypical fungal organisms that are genetically and antigenically distinct (15) and that exhibit host specificity (17). A system of nomenclature that recognizes these differences was proposed and uses the tripartite “special form” designations (14, 16). In humans, P. carinii f. sp. hominis causes a severe, often fatal pneumonia which is a major opportunistic infection in immunocompromised patients, especially those with advanced human immunodeficiency virus infections (17). Because of the absence of a reliable method of in vitro long-term culture for any member of the genus Pneumocystis, most studies rely on animal models which develop a pneumonia similar to that in humans. The immunosuppressed rat is one of the most common models and harbors two special forms, P. carinii f. sp. carinii, the most prevalent form (2), and P. carinii f. sp. ratti. Analysis of the chromosomes further revealed the existence of 11 different karyotypic forms among P. carinii f. sp. carinii populations, whereas only 1 form was identified for P. carinii f. sp. ratti (2, 3; M. T. Cushion, unpublished data). Coinfections with P. carinii f. sp. carinii and P. carinii f. sp. ratti are not uncommon (4, 9) and constitute a useful model for coinfections, which also frequently occur in humans (3, 11). Because these two special forms are genetically and antigenically distinct, it is important to determine if rats used for experimental studies harbor one or both forms, as well as to estimate the relative abundance of the forms in case of a coinfection. Moreover, it would be useful to be able to differentiate the karyotypic forms of P. carinii f. sp. carinii by the same method. For these purposes, we developed a rapid and sensitive method which consists of the amplification by PCR of a single variable genomic region, the intron of the nuclear 26S rRNA gene, followed by the detection of its polymorphism by the single-strand conformation polymorphism (SSCP) technique. We have previously used the PCR-SSCP technique to type P. carinii f. sp. hominis, the special form that infects humans (5–8, 11). SSCP analysis proved to be highly reproducible, permitting the detection of polymorphisms of a few base pairs (7), and allowed quantification of alleles in a rat with a mixed infection (11). SSCP analysis was expected to differentiate P. carinii f. sp. carinii and P. carinii f. sp. ratti because the genetic divergence at the nucleotide sequence level was reported to be 16% in the 26S rRNA genomic region (10).

DNA samples were prepared from organisms obtained from individual Sprague-Dawley or Long Evans rat lungs (n = 20) and analyzed by contour-clamped homogeneous electric field (CHEF) analysis as described previously (3). CHEF analysis identified the special form(s) present in each rat and provided an estimation of the relative abundance of each special form in case of a coinfection. The DNAs analyzed by PCR-SSCP analysis were extracted from the same low-melting-temperature agarose plugs that were analyzed by CHEF analysis by digestion with agarase (Boehringer Mannheim, Mannheim, Germany) according to the manufacturer's recommendations. To amplify the 26S rRNA genomic region, 1 to 3 μl of extracted DNA was used in 20 μl of a hot-start PCR mixture containing each deoxynucleoside triphosphate at a concentration of 0.2 mM, 3 mM MgCl₂, PCR buffer (Qiagen GmbH), 8 pmol of each primer, and 0.5 U of HotStarTaq DNA polymerase (Qiagen GmbH). The PCR primers used were described previously (7). They are localized in a conserved region of the gene (exons) and can amplify P. carinii f. sp. carinii and P. carinii f. sp. ratti DNAs with equal efficiencies. Forty cycles in which each cycle consisted of 30 s at 94°C, annealing for 1 min at 56°C, and extension for 1 min at 72°C were carried out. Reactions began with 15 min of denaturation at 94°C and ended with a 5-min extension at 72°C. To verify template fidelity, selected PCR products were sequenced directly and bidirectionally. PCR products were purified with a Qiaprep Spin Miniprep kit (Qiagen GmbH) and were sequenced with a Big Dye terminator DNA sequencing kit (Perkin-Elmer Biosystems) and an ABI PRISM 310 automated sequencer (Perkin-Elmer Biosystems). SSCP analysis was carried out as described previously (7). In brief, 20 ng of the 26S rRNA PCR product was heat denatured and then analyzed by SSCP analysis with a 3-h migration at 4°C with precast gels (Amersham Pharmacia Biotech) and a Delect buffer kit (Amersham Pharmacia Biotech). The bands on the gels used for SSCP analysis were visualized by silver staining (Amersham Pharmacia Biotech). To estimate the relative proportion of each special form in case of a coinfection, gels were scanned and each SSCP band was quantified with Image Analysis software (version 1.61; W. Rasband, National Institutes of Health, Bethesda, Md.). The surface of each peak density was measured in arbitrary units by use of the gel plot macro. The relative proportion of each P. carinii special form in a coinfected rat lung was calculated by dividing the sum of the SSCP bands that the special form produced by the sum of the bands that both special forms produced.

The 26S rRNA genomic region was amplified from DNAs of P. carinii f. sp. carinii karyotypic form 1 and P. carnii f. sp. ratti, which were pure according to CHEF analysis (n = 2 for each special form). The PCR products produced different SSCP patterns (Fig. 1, first and last lanes). Each of these patterns consisted of two bands and corresponded to a single allele of the genomic region (each band obtained by SSCP analysis corresponds to one of the two single strands of the PCR product; the efficiency of silver staining varied between the two single strands). The identities of the PCR products from P. carinii f. sp. carinii and P. carinii f. sp. ratti were confirmed by DNA sequencing and comparison to the reference alleles (GenBank accession nos. M86760 and L13614, respectively).

FIG. 1.

PCR-SSCP analysis of the 26S rRNA genomic regions of P. carinii f. sp. carinii (P.c. carinii) karyotypic forms. Each lane corresponds to P. carinii DNA purified from a single rat. The figure was generated with Adobe Photoshop (version 5.5) software. P. c. carinii, P. carinii f. sp. carinii; P. c. ratti, P. carinii f. sp. ratti.

Five samples which were coinfected with P. carinii f. sp. carinii karyotypic form 1 and P. carinii f. sp. ratti according to CHEF analysis were tested by PCR-SSCP analysis of the 26S rRNA genomic region. They all produced SSCP patterns that corresponded to those obtained by superimposition of the patterns produced by the two special forms (see the two example results in the two central lanes of Fig. 1). The PCR-SSCP method clearly detected the relative proportions of the two special forms in each coinfected sample. Visual inspection reveals that these proportions varied in the two coinfected samples and corresponded well to those observed on the CHEF patterns (Fig. 1). Because the SSCP patterns consist of only a few bands, a more precise proportion of each special form could be estimated by quantification of the bands obtained by SSCP analysis. P. carinii f. sp. carinii represented 21, 29, 58, 63, and 85% of the total population in the five samples, respectively. According to the sensitivity of SSCP analysis that we have determined previously (11), a special form present at low levels within a coinfected rat must represent at least 11% of the total population to be detected.

PCR-SSCP analysis of the 26S rRNA genomic region was applied to the 11 different karyotypic forms of P. carinii f. sp. carinii to determine if it could discriminate between them. They produced the same SSCP patterns for all forms except form 2 (Fig. 2; note that the top faint band produced by P. carinii f. sp. carinii DNA by SSCP analysis corresponds to a rare conformation adopted by one of the single strands). To determine if this result was due to a lack of polymorphism or to a low level of polymorphism between the forms, the products obtained by PCR of the 26S rRNA genomic region were sequenced. The sequences of karyotypic forms 3 to 10 were identical and presented a 1-bp polymorphism compared to the sequence of the reference allele present in GenBank (accession no. M86760). The sequences of forms 1, 2, and 11 presented polymorphisms of 2, 10, and 4 bp, respectively. Thus, there is a low level of divergence between P. carinii f. sp. carinii karyotypic forms in the 26S rRNA genomic region, and consistent with this, PCR-SSCP analysis detected only the more divergent form, form 2. These results confirm the low levels of genetic heterogeneity among P. carinii f. sp. carinii karyotypic forms that were reported previously (1, 13, 18).

FIG. 2.

PCR-SSCP analysis of the 26S rRNA genomic region and CHEF analysis of P. carinii special forms infecting rats. Each lane corresponds to P. carinii DNA purified from a single rat. P. carinii f. sp. carinii is karyotypic form 1. The figure was generated with Adobe Photoshop (version 5.5) software. P.c. carinii, P. carinii f. sp. carinii.

PCR-SSCP analysis of the 26S rRNA genomic region is a rapid and simple tool for the differentiation of P. carinii special forms and the estimation of their relative proportions in a given coinfected rat. Although it might be less sensitive than the recently described allele-specific PCR method (12) for the detection of a special form present in small amounts in a coinfected rat, the PCR-SSCP method presents three advantages: (i) it requires a single PCR for quantification of the special forms in a coinfected rat, whereas the PCR method involves several reactions with numerous dilutions for each of the two PCRs; (ii) it allows the precise quantification of the special forms in a coinfected rat, whereas the PCR method allows the identification of only the more abundant form; and (iii) it also detects P. carinii f. sp. carinii karyotypic form 2. This method should help to further standardize the immunosuppressed rat model of P. carinii infection, prevent the misinterpretation of results due to the presence of coexisting divergent organism populations, and, thus, contribute to a better understanding of P. carinii infection in humans.