- Top of page
- Materials and Methods
- Transparency Declaration
Pneumocystis jirovecii is recognized as a worldwide pathogenic fungus that parasitizes the human alveolar epithelium causing life-threatening interstitial pneumonia, known as Pneumocystis pneumonia, or pneumocystosis (PcP). This opportunistic infectious disease inflicts severe morbidity and mortality in immunocompromised patients, especially in those with AIDS and is therefore of growing importance among non-HIV-infected subjects who are undergoing immunosuppressive treatments related to malignancies, connective tissue diseases or organ transplantation [1–6].
In the absence of a culture system to isolate and maintain live organisms, efforts to type and characterize P. jirovecii have relied on PCR-based approaches [7–10]. Single-nucleotide polymorphisms (SNPs) are increasingly recognized as the markers of choice to study the population genetics, geographical distribution, modes of transmission and drug susceptibility or resistance of specific P. jirovecii genotypes [7,9,11–23]. Multiplex amplification of genomic DNA associated with single base extension (SBE) is a suitable high-throughput methodology for large-scale SNP screening [16,24,25]. Current genotyping surveys in a large number of samples are still time consuming methods. Alternatively, DNA pooling is a reliable genotyping method, in which equal amounts of DNA from a large number of individual samples are pooled and the SNP allele frequencies are estimated [26–29].
Dihydrofolate reductase (DHFR), superoxide dismutase (SOD) and mitochondrial large-subunit rRNA (mtLSU rRNA) are important genomic regions involved in basic metabolic mechanisms (folic acid synthesis and drug resistance, protection against free oxygen radicals and translation) that carry SNPs reported previously to be associated with clinical data and outcome of PcP [10,13,14,30–32].
The goals of the present study were to: (i) develop robust high-throughput methodologies for large-scale P. jirovecii genetic characterization; and (ii) correlate clinical parameters of infection such as parasite burden, clinical diagnosis of PcP and follow-up with genotyping results, using four P. jirovecii candidate SNPs (DHFR312, mt85, SOD110 and SOD215).
- Top of page
- Materials and Methods
- Transparency Declaration
Correlating multiple genetic differences in P. jirovecii samples and clinical data enables the achievement of robust information that may be used for preventing PcP [15,16,30]. In this study, P. jirovecii DNA pools were genetically characterized by MPCR/SBE. The potential relationships between candidate P. jirovecii SNPs and parasite burden, clinical diagnosis and follow-up of infection were evaluated. To our knowledge, the present report is the first study of P. jirovecii in which the DNA pooling survey was applied.
Data analysis demonstrated statistically significant differences between the average Cq values across pools constituted by pulmonary specimens collected from HIV-positive patients with AIDS-related PcP presenting different levels of parasite burden. The qPCR results corroborate the data obtained by the semi-quantitative method (IF), showing that qPCR is a sensitive and objective quantification technique, which could be used for quantification of P. jirovecii in pulmonary specimens [8,34,38]. Samples from pools B and C demonstrated low average DNA concentrations (see Table 2), which corroborates the fact that these pulmonary specimens, especially those from pool C (HIV-negative patients), tend to demonstrate lower parasite burdens . These low concentrations influenced the final amount of DNA contribution from each sample. Hence, whereas in P. jirovecii DNA pools A1, A2, A3, D and E each sample was diluted to contribute with 1 × 10−4 ng DNA, pools B and C were prepared with 1 × 10−5 ng DNA. Samples from pool D (positive follow-up) had a higher average DNA concentration than samples from pool E (negative follow-up). This could be caused by small sample size combined with inadequate sampling, but also it may suggest that, in some cases, unfavourable outcome from PcP episodes may be a consequence of insidious P. jirovecii infections difficult to diagnose because of low quantities of parasite cells in the host lungs or even because of the presence of more virulent P. jirovecii genotypes that inflict a higher severity of infection albeit at low concentrations. In further studies, serial pulmonary specimens should be examined in this group of patients to assess the evolution of the parasite burden levels during the course of infection, and correlate it with the PcP clinical progress.
The allelic frequency distribution analysis revealed that DHFR312T was the most prevalent DHFR allele, as has been found in recent studies [11,14,16]. DHFR312T was associated previously with PcP cases presenting moderate to high parasite burden . However, in the present work, no statistically significant differences were observed in the allelic frequency distribution of this SNP among the pools studied.
The allelic frequency distribution of the SNP mt85 revealed that mt85C was more frequent among pools A1 and A2, suggesting that this allele is probably related to PcP episodes presenting low to moderate parasite burden. Also, previous studies on Portuguese subjects showed that low to moderate parasite burden was more frequently observed among pulmonary specimens with mt85C and that this allele was associated with undiagnosed or atypical PcP cases and favourable follow-up [16,30]. Nevertheless, in the present study mt85C was statistically more common in pool E and less frequent in pool D, which suggests that mt85C is probably associated with unfavourable follow-up. Considering the relationship between lower levels of parasite burden and negative follow-up discussed above, mt85C may be related to P. jirovecii infections characterized by a difficult to diagnose insidious clinical presentation, which could contribute to bad clinical outcome of PcP episodes. Further studies involving a larger sample size and pulmonary specimens from different geographical regions should be conducted to clarify the relationship between mt85C and the follow-up of the infection. The allele mt85A was more common among pools A2, A3 and D, suggesting that mt85A is probably associated with PcP cases presenting moderate to high parasite burden and positive follow-up. Despite being associated with higher parasite burdens, mt85A proves to be quite common among PcP episodes with favourable clinical outcome. These data suggest that mt85A may be associated with P. jirovecii organisms with relatively high replication rates but that inflict low severity of infection, or even that this allele may be associated with P. jirovecii infections characterized by moderate to high parasite burden fulfilling the typical clinical parameters for PcP and easier to diagnose. The allele mt85T was more frequent in pools prepared with samples presenting low or high parasite burden (A1 and A3, respectively). The MPCR/SBE-DNA pooling demonstrated that mt85T was associated with PcP episodes presenting negative follow-up (E), suggesting that this allele may be associated with P. jirovecii organisms that inflict a higher severity of infection, regardless of parasite burden levels. Nucleotide variation of the mtLSUrRNA gene at base 85 may influence secondary structures of the mitochondrial rRNA, thereby having potential impact on ribosomal conformation and activity [16,30,39].
The incidence of mt85C was lower than reported in previous studies. The frequency of mt85C ranged between 6% and 23% in the present study, whereas in other studies on Portuguese patients the frequencies of this allele ranged between 46% and 51%. Other reports from different geographical regions have also revealed an mt85C frequency higher than that of the present report (e.g. 43% in Great Britain, 44% in USA and 50–58% in Spain) [7,9,11,30–32]. This may be a consequence of different sampling, which despite random selection may have different characteristics between pulmonary specimens and/or the fact that experimental conditions inherent to the multiplex-SBE methodology used may interfere with signal capture of the cytosine fluorochrome (ddCTP-D2) at the mt85C site during reaction. As a consequence of the limited sample volume, we were not able to repeat the MPCR/SBE DNA-pooling assay. Yet, this possible interference does not have an impact on the allelic frequency distribution variation analysis between the pools studied. This probable occurrence echoed proportionally in all SBE reactions, which validates the results obtained.
With respect to the allelic frequency distribution of P. jirovecii SOD SNPs, the present results revealed that SOD215C and SOD110T were more common in pools A1 and A2, whereas SOD215T and SOD110C were more frequent in pool A3. These results suggested a potential association between SOD215C/SOD110T and low to moderate parasite burden, as well as between SOD215T/SOD110C and high parasite burden. The differences observed could be because these SNP combinations may integrate distinct P. jirovecii organisms with different replication rates as a result of association with haplotypes involving other functional non-synonymous polymorphisms, or as a consequence of allele-specific differences in mRNA folding that could influence the splicing process, or translational control and regulation of the protein [15,30,40,41]. The present data support the fact that SNPs at bases 110 and 215 of the SOD locus may be in linkage disequilibrium, as suggested previously [11,13,16,30].
The present study showed that the amplification of P. jirovecii genomic DNA in a multiplex format associated with SBE genotyping and DNA pooling is a suitable high-throughput methodology for large-scale P. jirovecii clinically relevant SNP screening. The SNPs associated with distinct parasite burden levels and/or severity of PcP episodes (e.g. mt85, SOD215 and SOD110) were shown to be promising candidate markers to be used in MPCR/SBE protocols designed for molecular characterization of P. jirovecii organisms. Further studies involving a larger sample size and pulmonary specimens from different geographical regions are required to confirm the clinical relevance of these genetic variations and to evaluate the utility of these molecular techniques for prognostic classification of P. jirovecii infection to help clinical decision-making in the management of PcP cases.