Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

III. Typing methods to approach Pneumocystis carinii genetic heterogeneity

III. Typing methods to approach Pneumocystis carinii genetic heterogeneity The study of the genetic heterogeneity of P. carinii is complicated by the lack of an in vitro culture system, as well as by the likely occurrence of co-infections with several special forms or types in a single host. Karyotyping and multilocus enzyme electrophoresis are useful for studies at the evolutionary level. However, these methods require a large number of cells, which prevents their use for the special form infecting humans. DNA sequence analysis of genomic regions is useful to study P. carinii diversity, both at the evolutionary and epidemiological levels. To type the special form specific to humans, several methods are currently used to detect polymorphism in PCR products of polymorphic regions of the genome : DNA sequencing, type-specific hybridisations, and single-strand conformation polymorphism. All these methods still need evaluation. The frequency of potential co-infections in humans determined by these various methods is different. The differences could be due to methodological problems or to real variations between patient populations, geographical locations and/or prophylaxis regimens. In the future, elucidating the population structure of P. carinii and the frequency of potential co-infections is going to be crucial for a better understanding of its epidemiology, and thus for a better prevention of P. carinii pneumonia in humans. z 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Karyotyping ; DNA sequence analysis ; Type-speci¢c oligonucleotide ; Single-strand conformation polymorphism ; Polymerase chain reaction ; Molecular typing ; Pneumocystis carinii ; Genetic heterogeneity 1. Introduction at the nucleotide level). Thus, organisms recovered from di¡erent hosts have been designated as distinct Organisms recognised as Pneumocystis carinii be- `special forms'. Isolates of each special form, speci¢c long to a family of unusual fungi with a strict host for a given host, show less than 1% divergence and speci¢city to various mammalian species (see [1^3] have been de¢ned as `types'. Moreover, rats and fer- for reviews). Although morphologically similar, these rets were found to harbour two di¡erent special organisms present an important genetic heterogene- forms displaying an intermediate divergence (about ity, as discussed in the preceding chapters. In sum- 5%). There is presently a great interest in studying mary, there is an important divergence between P. carinii genetic heterogeneity in order to better Pneumocystis found in di¡erent hosts (about 20% understand its epidemiology and thus to improve prevention of P. carinii pneumonia (PCP) in hu- mans. This type of study, however, raises important * Corresponding author. issues and problems. Tel. : +41 (21) 314-0268 ; Fax : +41 (21) 340-4060 ; E-mail : [email protected] First, the assessment of P. carinii genetic diversity 0928-8244 / 98 / $19.00 ß 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 928 -82 44( 98) 000 53- 4 Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 28 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 implies di¡erent requirements depending on the culture method has been circumvented by the use scales of time and space. Three levels of time^space of PCR. settings can be considered (see also [4]) : the micro- Fourth, the occurrence of co-infections in a single epidemiological level (based on days to months and host with two or more special forms or types is a hospital to regional geographic area), the macro-epi- supplementary complication which will be addressed demiological level (years to few decades and country below. to continent-wide), and the evolutionary level (hun- In the present chapter, the di¡erent methods that dreds to thousands of years and world-wide). Inves- have been used to investigate P. carinii genetic diver- tigations at the evolutionary level require typing sity are described and their usefulness is discussed. methods allowing the measurement of the genetic distance between organisms, but these methods do not need to be as discriminative as for the other 2. Karyotyping two levels. On the other hand, investigations at the micro-epidemiological level require methods capable The chromosomes of P. carinii can be separated of discriminating micro-organisms with no true epi- by pulse-¢eld gel electrophoresis. The number and demiological link during the time of investigation. It the size of the chromosomes are roughly estimated. also requires that the epidemiological markers used A band pattern is obtained as well as an estimation are stable at least over the time of the investigation. of the genome size. Karyotyping has been applied to Second, the population structure of a micro-or- P. carinii from humans [6,7], rats [6^10], ferrets [11], ganism has important implications in terms of typing and mice [11]. It was the ¢rst technique to reveal the methods. Many micro-organisms have a clonal pop- existence of the host-speci¢c special forms, and of ulation structure, which means that the population is multiple types within a given host. The number of made of lineages of genetically identical cells. On the chromosomes of the di¡erent special forms were other hand, in some populations of micro-organisms, similar, but the size of the chromosomes were di¡er- genetic exchanges were found to occur frequently. In ent. Nine to 15 chromosomes ranging from 300 to this case, the genotypes are not stable because genet- 1000 kbp were observed, suggesting a genome of 7^ ic exchanges regularly alter them. Consequently, typ- 10 Mbp. ing methods are of little use for epidemiological Karyotyping was also the ¢rst method to reveal tracking of these micro-organisms. The impact of that rats harbour two special forms, P. carinii carinii genetic exchanges on the population structure is ap- and P. carinii ratti [6,10]. Intriguingly, the special proached by the study of population genetics [4]. The form ratti was observed only in single rats co-in- population structure of P. carinii is not known. The fected with the two special forms. Co-infections observation of synaptonemal complexes in cysts sug- with two P. carinii carinii types were also observed gests the occurrence of meiosis, and thus of genetic [12]. These co-infections produced patterns which exchanges during sexual reproduction [5]. However, were the superimpositions of the two di¡erent kar- the importance of this process in the multiplication yotypes, and the abundance of each karyotype in the of this organism is unknown. gel allowed to estimate the proportion of each of the Third, the study of P. carinii is complicated by the two co-infecting populations. lack of an in vitro culture system. In animals, organ- Among isolates of P. carinii carinii, eight di¡erent isms can be isolated after a suitable time of immu- karyotypes di¡ering by the size of only some of the nosuppression. A major limitation of this procedure chromosomes could be distinguished [9,12]. For is that the DNA of P. carinii obtained is contami- P. carinii ratti, two di¡erent karyotypes were identi- nated by host cells' DNA to varying extents. This ¢ed [12]. Analysis of infected rats colonies over years precludes the use of some convenient molecular tech- revealed that these karyotypes were stable [9]. They niques of genetic analysis, such as, for example, ran- are likely to correspond to di¡erent types of these dom ampli¢ed polymorphism DNA and restriction special forms, suggesting that types present a poly- endonuclease analysis. For P. carinii hominis (the morphism of chromosomal length potentially useful special form found in humans), the absence of a for epidemiological studies. However, the low level Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 29 of genetic diversity detected by karyotyping limits its 4. DNA sequence analysis value in the epidemiology of the rat-derived Pneumo- cystis. DNA sequence analysis is particularly adapted to Karyotyping presents the advantage of being rela- and has been widely used for the study of micro- tively fast and simple, thus allowing the analysis of organisms at the evolutionary level because nucleo- many samples. Its drawback is the requirement of a tide sequence allows direct measurement of genetic large number of organisms. Consequently, it can distances. Sequencing of the same genomic regions of only be applied to P. carinii special forms that can P. carinii organisms infecting di¡erent hosts, in par- be propagated in animal models but, unfortunately, ticular of the variable region of the mitochondrial its usefulness for the study of P. carinii hominis re- 26S rRNA (mt26S), has revealed the divergence at mains very limited unless an in vitro culture method the nucleotide sequence level between special forms becomes available. of di¡erent hosts and types within the same host. The results are in agreement with those of karyotyp- ing and con¢rm the existence of one P. carinii special 3. Multilocus enzyme electrophoresis form in each of the following host : humans [15], ferrets [16], and mice [17,18], and of two specials Multilocus enzyme electrophoresis (MEE) consists forms in rats [15,19]. Sequence analysis also revealed of the di¡erentiation of isoenzymes by their migra- the existence of other special forms speci¢c to horses tion in a gel. Migration of an enzyme extract is per- [20], pigs [21], shrews [22^24], and rabbits [18]. In formed in a non-denaturating gel and the relevant addition, sequence analysis revealed that two di¡er- enzyme is revealed in situ by its activity. Several ent special forms can be found not only in rats, but di¡erent loci are generally analysed and the di¡erent also in ferrets [16,25]. In agreement with the conclu- patterns are combined. A zymodeme is a group of sion of the studies using MEE [13,14], the results isolates which exhibit the same combination of iso- suggest that the special forms are di¡erent evolutive enzyme patterns. This technique is the gold standard branches. Comparison of gene sequences was used to for the study of phylogeny and population genetics. investigate P. carinii phylogeny, and suggested that MEE has been applied to 31 P. carinii isolates from P. carinii organisms constitute a unique branch be- rats, 17 from mice and 22 from rabbits [13,14]. Only tween the ascomycetes and basidiomycetes in the one zymodeme was observed among isolates of both fungal kingdom. mice and rabbits, whereas three zymodemes were Methods using DNA sequence analysis of PCR observed in rats. Population genetics analysis sug- products carrying variable regions of the genome gested that no genetic exchanges occur between the have been developed to type P. carinii hominis be- special forms. In conjunction with phylogenetic anal- cause of their ability to detect polymorphisms with a yses, this con¢rmed the host-speci¢city of the special high sensitivity. Five of the seven genomic loci that forms. have been investigated for P. carinii hominis were The disadvantage of MEE is its fastidiousness. reported to be polymorphic. Out of these ¢ve loci, Another drawback is its requirement of a large num- mt26S [26^32] and the two internal transcribed ber of cells which prevents its use in the analysis of spacers of the nuclear rRNA genes operon (ITS1 P. carinii hominis. and ITS2) [28,33^35] proved to be the most useful. Table 1 Alleles of variable genomic regions of P. carinii hominis Variable genomic region(s) No. of polymorphic positions No. of alleles identi¢ed ITS1 9 13 ITS2 8 11 ITS1+ITS2 17 30 mt26S 4 14 ITS1+ITS2+mt26S 21 31 Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 30 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 Polymorphisms in these variable regions were iden- sequencing are being developed (C.-H. Lee, personal ti¢ed at the same nucleotide positions by di¡erent communication). researchers from di¡erent continents. The variability The advantage of TSO is its rapidity and the abil- at these positions allowed the de¢nition of di¡erent ity to analyse many samples simultaneously. A dis- alleles (Table 1). advantage is the need to develop new TSO and to DNA sequence analysis is the most discriminative assess conditions of hybridisation each time a new method available. A disadvantage is its fastidious- allele is identi¢ed. ness and cost, which prevents the analysis of large collections of samples. 6. Single-strand conformation polymorphism 5. Type-speci¢c oligonucleotide hybridisation The single-strand conformation polymorphism (SSCP) technique permits the detection of single bp Type-speci¢c oligonucleotides (TSO) hybridising polymorphisms [39]. Polymorphisms modify the con- speci¢cally to di¡erent ITS1 and ITS2 alleles were formation adopted by single-stranded DNA mole- developed and used to type PCR products obtained cules in non-denaturating gels, and thus alter their from P. carinii hominis isolates [36,37]. When the migration. Most polymorphisms are detected if the TSO method was developed, only a few polymorphic DNA fragment is smaller than 400 bp. We recently positions and only two and three alleles of ITS1 and developed a method to type P. carinii hominis using ITS2, respectively, were recognised. Consequently, SSCP [40]. Because of the limited genetic variation of the method is not able to di¡erentiate the alleles each genomic region of P. carinii hominis, we devised which have been identi¢ed after the TSO were devel- a multi-target approach. Four variable regions of the oped (11 for ITS1 and 8 for ITS2). Thus, only six genome were analysed : ITS1, the intron of the nu- possible combinations of ITS1 and ITS2 alleles could clear 26S rRNA gene (26S), mt26S, and the region theoretically be di¡erentiated by the TSO method as surrounding intron 6 of the L-tubulin gene (L-tub). it was initially developed, which is of limited use for For each region, the ampli¢cation of a sample of a epidemiological studies. To date, ¢ve of these combi- given patient resulted either in a simple or in a com- nations were observed [36^38]. To improve the meth- plex SSCP pattern. A simple pattern consisted of two od, new TSO speci¢c to the new alleles identi¢ed by bands. It corresponded to the presence of a single Table 2 P. carinii hominis types identi¢ed in the 25 PCP patients presumably infected by a single type Type no. SSCP pattern No. of patients Country ITS1 26S mt26 L-tub 1 2 2 1 2 2 CH 2 1 3 1 1 2 CH 3 1 1 3 2 4 CH, F 4 1 2 3 2 2 CH, B 5 1 1 1 1 2 CH, F 6 1 2 1 2 1 F 7 1 1 3 1 4 F, CH, G 8 1 1 2 1 1 DK 9 1 1 1 2 3 CH 10 2 2 1 1 1 CH 11 1 2 1 1 1 CH 12 2 2 3 1 1 CH 13 1 4 1 1 1 F B, Belgium ; CH, Switzerland ; DK, Denmark ; F, France ; G, Germany. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 31 Fig. 1. Schematic representation of simple SSCP patterns with two bands identi¢ed for the four variable regions used to type P. carinii hominis. Each lane corresponds to a ¢ctive sample. Each number represents one simple SSCP pattern. For each region, the complex pat- tern 1,2 corresponding to the superimposition of simple patterns 1 and 2 is also represented. allele of the region (each SSCP band corresponds to the prevalence of certain types in the di¡erent Euro- one of the two complementary strands). Among the pean cities was detected. samples from 77 European patients analysed to date The major advantages of the SSCP method are its (56 from Switzerland, 11 from France, four from rapidity and low cost, a¡ording the analysis of many Denmark, three from Belgium, three from Ger- samples. Another advantage is that several loci can many), three to ¢ve di¡erent simple patterns have be analysed. This allows the analysis of population been identi¢ed for each genomic region (Fig. 1). A genetics and thus the study of the population struc- patient with a sample displaying only simple patterns ture of P. carinii. A potential drawback may be the for each of the four regions was probably infected by inability to detect polymorphisms of some DNA se- a single P. carinii hominis type. Thus, each combina- quences. Single bp polymorphisms were detected in tion of four simple patterns allows a type to be de- L-tubulin and mt26S [40], but this might be not the ¢ned. Twenty-¢ve patients were probably infected by case for ITS1. Indeed, SSCP analysis revealed only a single type and 13 di¡erent types were identi¢ed three di¡erent ITS1 patterns in our samples, whereas among them (Table 2). The large diversity of types the existence of 13 types has been described with suggests that this method will be useful for epidemio- DNA sequencing (see Table 1). This discordance logical studies. A complex SSCP pattern consisted of may be due to a low diversity in our samples but it more than two bands. It corresponded to the super- is also possible that the ITS1 single strands adopt imposition of simple patterns (Fig. 1). The observa- conformations which are particularly stable, thus tion of at least one complex pattern for any of the preventing e¤cient detection of polymorphisms. An- four regions in a given sample suggests an infection other disadvantage of the SSCP approach is the with several P. carinii hominis types. Among the 77 amount of work needed to set up the conditions of patients, 43 could have been infected by at least two SSCP, as well as the necessity to validate each new types and 9 by at least three types (because at least simple pattern by cloning and sequencing. one complex pattern made of two and three alleles, respectively, was observed). Among the 77 patients, 36 patients had a unique 7. The problem of potential co-infection with several combination of patterns for the four regions. The P. carinii hominis types remaining 41 patients had the same result as other patients and were distributed as follows : ten pairs, Typing of P. carinii hominis is complicated by the two groups of three and four groups of four. How- fact that co-infections with two or more types prob- ever, further investigation of these cases did not re- ably occur. This phenomenon has been demon- veal obvious epidemiological links. No increase in strated in rats and ferrets. Co-infections in humans Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 32 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 were suggested by the presence of two or sometimes quency of co-infections because of the limited num- three alleles of a genomic region in samples from ber of TSO available (see Section 5). Co-infections single patients. However, there are other possible which involved alleles hybridising to the same oligo- explanations for these ¢ndings : heterozygosity of nucleotide for both ITS1 and ITS2 could not be diploid organisms or presence of two or more copies detected. of the same region per genome, with variation be- With SSCP, the production of one complex pat- tween the copies. In the absence of data, these alter- tern for at least one of the four regions investigated native explanations cannot be ¢rmly excluded [40]. suggests a potential co-infection. A complex pattern Moreover, the three possibilities are not mutually made of two or three alleles suggests the presence of exclusive. at least two or three co-infecting types, respectively. The typing methods presented above used di¡erent Among the 77 samples analysed, samples yielding at strategies to detect these potential co-infections. In least one complex pattern with two or three alleles some studies, PCR products were systematically represented 57 and 12%, respectively. Thus, co-infec- cloned and two or three clones of most of these tion appears to occur in up to 69% of the samples. samples were sequenced [33,34]. When more than The sensitivity of silver staining used in SSCP allows one allele was found, co-infection was postulated. the detection of an allele if it represents at least 5% This was observed in 20^35% of the samples exam- of the molecules (our unpublished data). ined. This means that for these samples, the sequence Our SSCP method detects potential co-infection in of one clone was di¡erent than those of the two a greater (69%) proportion of samples than other other clones analysed. This approach detects an al- approaches (10^35%). There are two possibilities to lele only if this allele represents at least about 33% of explain these di¡erences. First, as discussed above, the molecules. Thus, this method detects a co-infect- the sensitivity of the other approaches seems lower ing type only if present in a signi¢cant proportion of than that of SSCP. Second, the di¡erences might the whole population. re£ect true di¡erences in the proportion of co-infec- Direct sequencing of PCR products has also been tion between di¡erent populations. In favour of the used [26^28]. Co-infections were assessed when dou- latter possibility, a group observed approximately ble bands or peaks were observed. This was found in 85% of potential co-infections among Brazilian 10^30% of the samples analysed. These samples were transplant patients, but only about 5% among further cloned and several clones were sequenced. In AIDS patients from the United States (C.B. Beard, one such study [26], using autoradiography, an allele personal communication). This di¡erence could not was detected if it represented 20% or more of the be explained by methodological problems because molecules. both results were obtained using direct sequencing. In studies using the TSO typing method, co-infec- This suggests that other factors, such as the type of tions were assumed if detected by their hybridisation patient population, the prophylaxis regimen and/or to two oligonucleotides speci¢c for di¡erent alleles of the geographical location may a¡ect the rate of co- same locus (ITS1 or ITS2). In order to type these infection. To resolve this issue, di¡erent methods samples using TSO, the ITS1^ITS2 region of the co- should be applied to the same samples and the re- infecting types were ampli¢ed separately by type-spe- sults compared. ci¢c PCRs [37]. (Type-speci¢c PCR ampli¢es a spe- ci¢c allele by the use of a primer which contains mismatches at its 3P-end for other allele(s).) Co-in- 8. The contribution of typing to P. carinii hominis fections were found in 20^30% of the patients using epidemiology this experimental protocol [36,37]. Control experi- ments suggested that the method was very sensitive Typing methods developed for P. carinii hominis because, in a mixture of alleles, it could detect an have already contributed to the understanding of allele 100 times less abundant than the other allele(s) some epidemiological features of this special form. (C.-H. Lee, personal communication). However, The question of reactivation of latent organisms ver- this approach has certainly underestimated the fre- sus de novo infection in recurrent P. carinii pneumo- Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 33 nia was investigated by DNA sequence analysis of 9. Concluding remarks mt26 or of the ITS1^ITS2 combination [26,27,32, 34,35]. In about 50% of the cases analysed, di¡erent Until a culture method is established, the study of alleles were found in the second episode of the dis- P. carinii genetic heterogeneity will probably rely on ease compared to the ¢rst. Because of the possibility the methods discussed above. Karyotyping, MEE of undetected co-infections, type-speci¢c PCR has and DNA sequencing appear to be appropriate for been used to investigate the possibility of the pres- investigations at the evolutionary level. DNA se- ence of a low amount of another allele during the quencing, TSO and SSCP analyses of PCR products ¢rst PCP episode [32]. This possibility was excluded from variable regions of the genome should be useful in four out of ¢ve cases analysed. In one study [35], for epidemiological studies. However, it is not ex- the alleles of both mt26S and the ITS1^ITS2 combi- cluded that other methods for polymorphism detec- nation were di¡erent in the second episode. This ex- tion, such as heteroduplex analysis or structure-spe- cluded that the alteration of alleles was due to mu- ci¢c endonuclease cleavage [45,46], might also be tation because the probability of concurrent useful, but they have not yet been explored. mutations at independent loci is very low. These re- sults strongly suggested that de novo infections occur frequently in recurrent PCP. Acknowledgments Sequencing of the ITS1^ITS2 region was used to investigate hypothetical transmission within cou- The study of P. carinii epidemiology in our labo- ples [41]. Di¡erent alleles of both ITS1 and ITS2 ratory is partially supported by Grants 94-7213 and were found in the members of each of the three cou- 97-7299 of the Swiss National Programme on AIDS ples analysed, suggesting that transmission did not Research. We thank A.E. Wake¢eld for fruitful dis- occur. cussions and P. Shaw for critical reading of the The TSO method has been used to investigate the manuscript. air of PCP patients' rooms [42,43]. Although only six combinations of ITS1 and ITS2 alleles can be di¡er- References entiated by this method, the results were compatible with dispersion of P. carinii organisms in the envi- [1] Stringer, J.R. and Walzer, P.D. (1996) Molecular biology and ronment because the same ITS1^ITS2 combination epidemiology of Pneumocystis in AIDS. AIDS 10, 561^571. was found both in the air and the clinical sample of [2] Stringer, J.R. (1996) Pneumocystis carinii ^ what is it, exactly ? the patient in 11 cases out of 14. Clin. Microbiol. Rev. 9, 489^498. Although these studies contribute to the under- [3] Wake¢eld, A.E. (1995) Re-examination of epidemiological concepts. In : Pneumocystis carinii, Baillieres Clin. Infect. standing of P. carinii hominis epidemiology, some Dis. 2, 431^448. aspects of the results should be taken with caution [4] Tibayrenc, M. (1995) Population genetics of parasitic proto- because the typing methods used have not been fully zoa and other microorganisms. Adv. Parasitol. 36, 47^115. evaluated according to proposed guidelines [44]. In [5] Matsumoto, Y. and Yoshida, Y. (1984) Sporogony in Pneu- particular, their power of discrimination has not mocystis carinii : synaptonemal complexes and meiotic nuclear divisions observed in precysts. J. Protozool. 31, 420^428. been assessed. A low discriminatory power could [6] Hong, S.T., Steele, P.E., Cushion, M.T., Walzer, P.D., String- have resulted, for example, in an underestimation er, S.L. and Stringer, J.R. (1990) Pneumocystis carinii karyo- of the frequency of de novo infection in recurrent types. J. Clin. Microbiol. 28, 1785^1795. PCP. In addition, the possibility of undetected co- [7] Stringer, J.R., Stringer, S.L., Zhang, J., Baughman, R., Smu- infections has not been considered in the studies of lian, A.G. and Cushion, M.T. (1993) Molecular genetic dis- tinction of Pneumocystis carinii from rats and humans. J. Eu- the couples and air samples mentioned above. For karyotic Microbiol. 40, 733^741. example, if the alleles carried by one partner were [8] Lundgren, B., Cotton, R., Lundgren, J.D., Edman, J.C. and found at low levels in the other partner, the possi- Kovacs, J.A. (1990) Identi¢cation of Pneumocystis carinii bility of transmission within the couple could have chromosomes and mapping of ¢ve genes. Infect. Immun. 58, been missed. 1705^1710. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 34 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 Sorex araneus and Sorex caecutiens. J. Wildlife Dis. 29, 273^ [9] Cushion, M.T., Kaselis, M., Stringer, S.L. and Stringer, J.R. (1993) Genetic stability and diversity of Pneumocystis carinii [25] Shah, J.S., Pieciak, W., Liu, J., Buharin, A. and Lane, D.J. infecting rat colonies. Infect. Immun. 61, 4801^4813. (1996) Diversity of host species and strains of Pneumocystis [10] Cushion, M.T., Zhang, J., Kaselis, M., Giuntoli, D., Stringer, carinii is based on rRNA sequences. Clin. Diagn. Lab. Immu- S.L. and Stringer, J.R. (1993) Evidence for two genetic var- nol. 3, 119^127. iants of Pneumocystis carinii coinfecting laboratory rats. [26] Keely, S.P., Stringer, J.R., Baughman, R.P., Linke, M.J., J. Clin. Microbiol. 31, 1217^1223. Walzer, P.D. and Smulian, A.G. (1995) Genetic variation [11] Weinberg, G.A. and Durant, P.J. (1994) Genetic diversity of among Pneumocystis carinii hominis isolates in recurrent pneu- Pneumocystis carinii derived from infected rats, mice, ferrets, and cell cultures. J. Eukaryotic Microbiol. 41, 223^228. mocystosis. J. Infect. Dis. 172, 595^598. [27] Latouche, S., Poirot, J.L., Bernard, C. and Roux, P. (1997) [12] Cushion, M.T. (1998) Pneumocystis carinii. Topley and Wil- Study of internal transcribed spacer and mitochondrial large- son's Microbiology and Microbial Infections, 9th edn. (Collier subunit genes of Pneumocystis carinii hominis isolated by re- L., Balows A. and Sussman M., Eds.), Edward Arnold, Lon- peated bronchoalvelolar lavage from human immunode¢- don, pp. 645^683. [13] Mazars, E., Guyot, K., Durand, I., Dei-Cas, E., Boucher, S., ciency virus-infected patients during one or several episodes of pneumonia. J. Clin. Microbiol. 35, 1687^1690. Abderrazak, S.E., Banuls, A.L., Tibayrenc, M. and Camus, [28] Latouche, S., Ortona, E., Mazars, E., Margutti, P., Tambur- D. (1997) Isoenzyme diversity in Pneumocystis carinii from rini, E., Siracusano, A., Guyot, K., Nigou, M. and Roux, P. rats, mice, and rabbits. J. Infect. Dis. 175, 655^660. (1997) Biodiversity of Pneumocystis carinii hominis : typing [14] Mazars, E., Odberg-Ferragut, C., Durand, I., Tibayrenc, M., Dei-Cas, E. and Camus, D. (1994) Genomic and isoenzymatic with di¡erent DNA regions. J. Clin. Microbiol. 35, 383^387. [29] Lee, C.H., Lu, J.J., Bartlett, M.S., Durkin, M.M., Liu, T.H., markers of Pneumocystis from di¡erent host species. J. Eu- karyotic Microbiol. 41, 104S. Wang, J., Jiang, B. and Smith, J.W. (1993) Nucleotide se- quence variation in Pneumocystis carinii strains that infect [15] Sinclair, K., Wake¢eld, A.E., Banerji, S. and Hopkin, J.M. humans. J. Clin. Microbiol. 31, 754^757. (1991) Pneumocystis carinii organisms derived from rat and human hosts are genetically distinct. Mol. Biochem. Parasitol. [30] Latouche, S., Roux, P., Poirot, J.L., Lavrard, I., Hermelin, B. and Bertrand, V. (1994) Preliminary results of Pneumocystis 45, 183^184. [16] Banerji, S., Lugli, E.B. and Wake¢eld, A.E. (1994) Identi¢ca- carinii strain di¡erentiation by using molecular biology. tion of two genetically distinct strains of Pneumocystis carinii J. Clin. Microbiol. 32, 3052^3053. [31] Wake¢eld, A.E., Fritscher, C.C., Malin, A.S., Gwanzura, L., in infected ferret lungs. J. Eukaryotic Microbiol. 41, 73S. [17] Kitada, K., Oka, S., Kimura, S., Shimada, K., Serikawa, T., Hughes, W.T. and Miller, R.F. (1994) Genetic diversity in human-derived Pneumocystis carinii isolates from four geo- Yamada, J., Tsunoo, H., Egawa, K. and Nakamura, Y. (1991) Detection of Pneumocystis carinii sequences by polymerase graphical locations shown by analysis of mitochondrial chain reaction : animal models and clinical application to non- rRNA gene sequences. J. Clin. Microbiol. 32, 2959^2961. invasive specimens. J. Clin. Microbiol. 29, 1985^1990. [32] Keely, S.P., Baughman, R.P., Smulian, A.G., Dohn, M.N. [18] Banerji, S., Lugli, E.B., Miller, R.F. and Wake¢eld, A.E. and Stringer, J.R. (1996) Source of Pneumocystis carinii in recurrent episodes of pneumonia in AIDS patients. AIDS (1995) Analysis of genetic diversity at the arom locus in iso- lates of Pneumocystis carinii. J. Eukaryotic Microbiol. 42, 10, 881^888. 675^679. [33] Lu, J.J., Bartlett, M.S., Shaw, M.M., Queener, S.F., Smith, [19] Liu, Y. and Leibowitz, M.J. (1993) Variation and in vitro J.W., Ortiz-Rivera, M., Leibowitz, M.J. and Lee, C.H. (1994) splicing of group I introns in rRNA genes of Pneumocystis Typing of Pneumocystis carinii strains that infect humans based on nucleotide sequence variations of internal tran- carinii. Nucleic Acids Res. 21, 2415^2421. [20] Peters, S.E., Wake¢eld, A.E., Whitwell, K.E. and Hopkin, scribed spacers of rRNA genes. J. Clin. Microbiol. 32, J.M. (1994) Pneumocystis carinii pneumonia in thoroughbred 2904^2912. foals : identi¢cation of a genetically distinct organism by DNA [34] Tsolaki, A.G., Miller, R.F., Underwood, A.P., Banerji, S. and ampli¢cation. J. Clin. Microbiol. 32, 213^216. Wake¢eld, A.E. (1996) Genetic diversity at the internal tran- [21] Christensen, C.B., Settnes, O.P., Bille-Hansen, V., Jorsal, S.E., scribed spacer regions of the rRNA operon among isolates of Henriksen, S.A. and Lundgren, B. (1996) Pneumocystis carinii Pneumocystis carinii from AIDS patients with recurrent pneu- from pigs and humans are antigenically distinct. J. Med. Vet. monia. J. Infect. Dis. 174, 141^156. Mycol. 34, 431^433. [35] Keely, S.P. and Stringer, J.R. (1997) Sequences of Pneumo- [22] Laakkonen, J. and Sukura, A. (1997) Pneumocystis carinii of cystis carinii f. sp. hominis strains asssociated with recurrent the common shrew, Sorex araneus, shows a discrete pheno- pneumonia vary at multiple loci. J. Clin. Microbiol. 35, 2745^ type. J. Eukaryotic Microbiol. 44, 117^121. 2747. [23] Peters, S.E., English, K., Laakkonen, J. and Gurnell, J. (1994) [36] Lu, J.J., Bartlett, M.S., Smith, J.W. and Lee, C.H. (1995) DNA analysis of Pneumocystis carinii infecting Finnish and Typing of Pneumocystis carinii strains with type-speci¢c oligo- English shrews. J. Eukaryotic Microbiol. 41, 108S. nucleotide probes derived from nucleotide sequences of inter- [24] Laakkonen, J., Sukura, A., Haukisalmi, V. and Henttonen, H. nal transcribed spacers of rRNA genes. J. Clin. Microbiol. 33, (1993) Pneumocystis carinii and helminth parasitism in shrews 2973^2977. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 35 and Smith, J.W. (1996) Types of Pneumocystis carinii detected [37] Jiang, B.D., Lu, J.J., Li, B.Z., Tang, X., Bartlett, M.S., Smith, in air samples. J. Eukaryotic Microbiol. 43, 44S. J.W. and Lee, C.H. (1996) Development of type-speci¢c PCR for typing Pneumocystis carinii f. sp. hominis based [43] Bartlett, M.S., Vermund, S.H., Jacobs, R., Durant, P.J., Shaw, M.M., Smith, J.W., Tang, X., Lu, J.J., Li, B., Jin, S. on sequence nucleotide variations of internal transcribed spacer regions of rRNA genes. J. Clin. Microbiol. 34, 3245^ and Lee, C.H. (1997) Detection of Pneumocystis carinii DNA in air samples : likely environmental risk to susceptible per- sons. J. Clin. Microbiol. 35, 2511^2513. [38] Atzori, C., Agostoni, F., Cargnel, A., Perenboom, R. and Beckers, P. (1996) ITSs typing of P. carinii samples from Italy, [44] Struelens, M.J., Members of the European Study Group on Epidemiological Markers (ESGEM) (1996) Consensus guide- The Netherlands and Tanzania. J. Eukaryotic Microbiol. 43, 43S. lines for appropriate use and evaluation of microbial epide- miologic typing systems. Clin. Microbiol. Infect. 2, 2^11. [39] Hayashi, K. and Yandell, D.W. (1993) How sensitive is PCR- [45] Brow, M.A., Oldenburg, M.C., Lyamichev, V., Heisler, L.M., SSCP ? Hum. Mutat. 2, 338^346. Lyamicheva, N., Hall, J.G., Eagan, N.J., Olive, D.M., Smith, [40] Hauser, P.M., Francioli, P., Bille, J., Telenti, A. and Blanc, L.M., Fors, L. and Dahlberg, J.E. (1996) Di¡erentiation of D.S. (1997) Typing of Pneumocystis carinii f. sp. hominis by single-strand conformation polymorphism of four genomic re- bacterial 16S rRNA genes and intergenic regions and Myco- bacterium tuberculosis katG genes by structure-speci¢c endo- gions. J. Clin. Microbiol. 35, 3086^3091. nuclease cleavage. J. Clin. Microbiol. 34, 3129^3137. [41] Latouche, S., Poirot, J.L., Maury, E., Bertrand, V. and Roux, [46] Lyamichev, V., Brow, M.A. and Dahlberg, J.E. (1993) Struc- P. (1997) Pneumocystis carinii hominis sequencing to study ture-speci¢c endonucleolytic cleavage of nucleic acids by eu- hypothetical person-to-person transmission. AIDS 11, 549. [42] Bartlett, M.S., Lu, J.J., Lee, C.H., Durant, P.J., Queener, S.F. bacterial DNA polymerases. Science 260, 778^783. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of the Endocrine Society Oxford University Press

III. Typing methods to approach Pneumocystis carinii genetic heterogeneity

Download PDF
9 pages

Loading next page...
 
/lp/oxford-university-press/iii-typing-methods-to-approach-pneumocystis-carinii-genetic-uLNRWYwm8T

References (48)

Publisher
Oxford University Press
Copyright
© 1998 Federation of European Microbiological Societies.
eISSN
2472-1972
DOI
10.1111/j.1574-695X.1998.tb01184.x
pmid
9792058
Publisher site
See Article on Publisher Site

Abstract

The study of the genetic heterogeneity of P. carinii is complicated by the lack of an in vitro culture system, as well as by the likely occurrence of co-infections with several special forms or types in a single host. Karyotyping and multilocus enzyme electrophoresis are useful for studies at the evolutionary level. However, these methods require a large number of cells, which prevents their use for the special form infecting humans. DNA sequence analysis of genomic regions is useful to study P. carinii diversity, both at the evolutionary and epidemiological levels. To type the special form specific to humans, several methods are currently used to detect polymorphism in PCR products of polymorphic regions of the genome : DNA sequencing, type-specific hybridisations, and single-strand conformation polymorphism. All these methods still need evaluation. The frequency of potential co-infections in humans determined by these various methods is different. The differences could be due to methodological problems or to real variations between patient populations, geographical locations and/or prophylaxis regimens. In the future, elucidating the population structure of P. carinii and the frequency of potential co-infections is going to be crucial for a better understanding of its epidemiology, and thus for a better prevention of P. carinii pneumonia in humans. z 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Karyotyping ; DNA sequence analysis ; Type-speci¢c oligonucleotide ; Single-strand conformation polymorphism ; Polymerase chain reaction ; Molecular typing ; Pneumocystis carinii ; Genetic heterogeneity 1. Introduction at the nucleotide level). Thus, organisms recovered from di¡erent hosts have been designated as distinct Organisms recognised as Pneumocystis carinii be- `special forms'. Isolates of each special form, speci¢c long to a family of unusual fungi with a strict host for a given host, show less than 1% divergence and speci¢city to various mammalian species (see [1^3] have been de¢ned as `types'. Moreover, rats and fer- for reviews). Although morphologically similar, these rets were found to harbour two di¡erent special organisms present an important genetic heterogene- forms displaying an intermediate divergence (about ity, as discussed in the preceding chapters. In sum- 5%). There is presently a great interest in studying mary, there is an important divergence between P. carinii genetic heterogeneity in order to better Pneumocystis found in di¡erent hosts (about 20% understand its epidemiology and thus to improve prevention of P. carinii pneumonia (PCP) in hu- mans. This type of study, however, raises important * Corresponding author. issues and problems. Tel. : +41 (21) 314-0268 ; Fax : +41 (21) 340-4060 ; E-mail : [email protected] First, the assessment of P. carinii genetic diversity 0928-8244 / 98 / $19.00 ß 1998 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 928 -82 44( 98) 000 53- 4 Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 28 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 implies di¡erent requirements depending on the culture method has been circumvented by the use scales of time and space. Three levels of time^space of PCR. settings can be considered (see also [4]) : the micro- Fourth, the occurrence of co-infections in a single epidemiological level (based on days to months and host with two or more special forms or types is a hospital to regional geographic area), the macro-epi- supplementary complication which will be addressed demiological level (years to few decades and country below. to continent-wide), and the evolutionary level (hun- In the present chapter, the di¡erent methods that dreds to thousands of years and world-wide). Inves- have been used to investigate P. carinii genetic diver- tigations at the evolutionary level require typing sity are described and their usefulness is discussed. methods allowing the measurement of the genetic distance between organisms, but these methods do not need to be as discriminative as for the other 2. Karyotyping two levels. On the other hand, investigations at the micro-epidemiological level require methods capable The chromosomes of P. carinii can be separated of discriminating micro-organisms with no true epi- by pulse-¢eld gel electrophoresis. The number and demiological link during the time of investigation. It the size of the chromosomes are roughly estimated. also requires that the epidemiological markers used A band pattern is obtained as well as an estimation are stable at least over the time of the investigation. of the genome size. Karyotyping has been applied to Second, the population structure of a micro-or- P. carinii from humans [6,7], rats [6^10], ferrets [11], ganism has important implications in terms of typing and mice [11]. It was the ¢rst technique to reveal the methods. Many micro-organisms have a clonal pop- existence of the host-speci¢c special forms, and of ulation structure, which means that the population is multiple types within a given host. The number of made of lineages of genetically identical cells. On the chromosomes of the di¡erent special forms were other hand, in some populations of micro-organisms, similar, but the size of the chromosomes were di¡er- genetic exchanges were found to occur frequently. In ent. Nine to 15 chromosomes ranging from 300 to this case, the genotypes are not stable because genet- 1000 kbp were observed, suggesting a genome of 7^ ic exchanges regularly alter them. Consequently, typ- 10 Mbp. ing methods are of little use for epidemiological Karyotyping was also the ¢rst method to reveal tracking of these micro-organisms. The impact of that rats harbour two special forms, P. carinii carinii genetic exchanges on the population structure is ap- and P. carinii ratti [6,10]. Intriguingly, the special proached by the study of population genetics [4]. The form ratti was observed only in single rats co-in- population structure of P. carinii is not known. The fected with the two special forms. Co-infections observation of synaptonemal complexes in cysts sug- with two P. carinii carinii types were also observed gests the occurrence of meiosis, and thus of genetic [12]. These co-infections produced patterns which exchanges during sexual reproduction [5]. However, were the superimpositions of the two di¡erent kar- the importance of this process in the multiplication yotypes, and the abundance of each karyotype in the of this organism is unknown. gel allowed to estimate the proportion of each of the Third, the study of P. carinii is complicated by the two co-infecting populations. lack of an in vitro culture system. In animals, organ- Among isolates of P. carinii carinii, eight di¡erent isms can be isolated after a suitable time of immu- karyotypes di¡ering by the size of only some of the nosuppression. A major limitation of this procedure chromosomes could be distinguished [9,12]. For is that the DNA of P. carinii obtained is contami- P. carinii ratti, two di¡erent karyotypes were identi- nated by host cells' DNA to varying extents. This ¢ed [12]. Analysis of infected rats colonies over years precludes the use of some convenient molecular tech- revealed that these karyotypes were stable [9]. They niques of genetic analysis, such as, for example, ran- are likely to correspond to di¡erent types of these dom ampli¢ed polymorphism DNA and restriction special forms, suggesting that types present a poly- endonuclease analysis. For P. carinii hominis (the morphism of chromosomal length potentially useful special form found in humans), the absence of a for epidemiological studies. However, the low level Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 29 of genetic diversity detected by karyotyping limits its 4. DNA sequence analysis value in the epidemiology of the rat-derived Pneumo- cystis. DNA sequence analysis is particularly adapted to Karyotyping presents the advantage of being rela- and has been widely used for the study of micro- tively fast and simple, thus allowing the analysis of organisms at the evolutionary level because nucleo- many samples. Its drawback is the requirement of a tide sequence allows direct measurement of genetic large number of organisms. Consequently, it can distances. Sequencing of the same genomic regions of only be applied to P. carinii special forms that can P. carinii organisms infecting di¡erent hosts, in par- be propagated in animal models but, unfortunately, ticular of the variable region of the mitochondrial its usefulness for the study of P. carinii hominis re- 26S rRNA (mt26S), has revealed the divergence at mains very limited unless an in vitro culture method the nucleotide sequence level between special forms becomes available. of di¡erent hosts and types within the same host. The results are in agreement with those of karyotyp- ing and con¢rm the existence of one P. carinii special 3. Multilocus enzyme electrophoresis form in each of the following host : humans [15], ferrets [16], and mice [17,18], and of two specials Multilocus enzyme electrophoresis (MEE) consists forms in rats [15,19]. Sequence analysis also revealed of the di¡erentiation of isoenzymes by their migra- the existence of other special forms speci¢c to horses tion in a gel. Migration of an enzyme extract is per- [20], pigs [21], shrews [22^24], and rabbits [18]. In formed in a non-denaturating gel and the relevant addition, sequence analysis revealed that two di¡er- enzyme is revealed in situ by its activity. Several ent special forms can be found not only in rats, but di¡erent loci are generally analysed and the di¡erent also in ferrets [16,25]. In agreement with the conclu- patterns are combined. A zymodeme is a group of sion of the studies using MEE [13,14], the results isolates which exhibit the same combination of iso- suggest that the special forms are di¡erent evolutive enzyme patterns. This technique is the gold standard branches. Comparison of gene sequences was used to for the study of phylogeny and population genetics. investigate P. carinii phylogeny, and suggested that MEE has been applied to 31 P. carinii isolates from P. carinii organisms constitute a unique branch be- rats, 17 from mice and 22 from rabbits [13,14]. Only tween the ascomycetes and basidiomycetes in the one zymodeme was observed among isolates of both fungal kingdom. mice and rabbits, whereas three zymodemes were Methods using DNA sequence analysis of PCR observed in rats. Population genetics analysis sug- products carrying variable regions of the genome gested that no genetic exchanges occur between the have been developed to type P. carinii hominis be- special forms. In conjunction with phylogenetic anal- cause of their ability to detect polymorphisms with a yses, this con¢rmed the host-speci¢city of the special high sensitivity. Five of the seven genomic loci that forms. have been investigated for P. carinii hominis were The disadvantage of MEE is its fastidiousness. reported to be polymorphic. Out of these ¢ve loci, Another drawback is its requirement of a large num- mt26S [26^32] and the two internal transcribed ber of cells which prevents its use in the analysis of spacers of the nuclear rRNA genes operon (ITS1 P. carinii hominis. and ITS2) [28,33^35] proved to be the most useful. Table 1 Alleles of variable genomic regions of P. carinii hominis Variable genomic region(s) No. of polymorphic positions No. of alleles identi¢ed ITS1 9 13 ITS2 8 11 ITS1+ITS2 17 30 mt26S 4 14 ITS1+ITS2+mt26S 21 31 Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 30 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 Polymorphisms in these variable regions were iden- sequencing are being developed (C.-H. Lee, personal ti¢ed at the same nucleotide positions by di¡erent communication). researchers from di¡erent continents. The variability The advantage of TSO is its rapidity and the abil- at these positions allowed the de¢nition of di¡erent ity to analyse many samples simultaneously. A dis- alleles (Table 1). advantage is the need to develop new TSO and to DNA sequence analysis is the most discriminative assess conditions of hybridisation each time a new method available. A disadvantage is its fastidious- allele is identi¢ed. ness and cost, which prevents the analysis of large collections of samples. 6. Single-strand conformation polymorphism 5. Type-speci¢c oligonucleotide hybridisation The single-strand conformation polymorphism (SSCP) technique permits the detection of single bp Type-speci¢c oligonucleotides (TSO) hybridising polymorphisms [39]. Polymorphisms modify the con- speci¢cally to di¡erent ITS1 and ITS2 alleles were formation adopted by single-stranded DNA mole- developed and used to type PCR products obtained cules in non-denaturating gels, and thus alter their from P. carinii hominis isolates [36,37]. When the migration. Most polymorphisms are detected if the TSO method was developed, only a few polymorphic DNA fragment is smaller than 400 bp. We recently positions and only two and three alleles of ITS1 and developed a method to type P. carinii hominis using ITS2, respectively, were recognised. Consequently, SSCP [40]. Because of the limited genetic variation of the method is not able to di¡erentiate the alleles each genomic region of P. carinii hominis, we devised which have been identi¢ed after the TSO were devel- a multi-target approach. Four variable regions of the oped (11 for ITS1 and 8 for ITS2). Thus, only six genome were analysed : ITS1, the intron of the nu- possible combinations of ITS1 and ITS2 alleles could clear 26S rRNA gene (26S), mt26S, and the region theoretically be di¡erentiated by the TSO method as surrounding intron 6 of the L-tubulin gene (L-tub). it was initially developed, which is of limited use for For each region, the ampli¢cation of a sample of a epidemiological studies. To date, ¢ve of these combi- given patient resulted either in a simple or in a com- nations were observed [36^38]. To improve the meth- plex SSCP pattern. A simple pattern consisted of two od, new TSO speci¢c to the new alleles identi¢ed by bands. It corresponded to the presence of a single Table 2 P. carinii hominis types identi¢ed in the 25 PCP patients presumably infected by a single type Type no. SSCP pattern No. of patients Country ITS1 26S mt26 L-tub 1 2 2 1 2 2 CH 2 1 3 1 1 2 CH 3 1 1 3 2 4 CH, F 4 1 2 3 2 2 CH, B 5 1 1 1 1 2 CH, F 6 1 2 1 2 1 F 7 1 1 3 1 4 F, CH, G 8 1 1 2 1 1 DK 9 1 1 1 2 3 CH 10 2 2 1 1 1 CH 11 1 2 1 1 1 CH 12 2 2 3 1 1 CH 13 1 4 1 1 1 F B, Belgium ; CH, Switzerland ; DK, Denmark ; F, France ; G, Germany. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 31 Fig. 1. Schematic representation of simple SSCP patterns with two bands identi¢ed for the four variable regions used to type P. carinii hominis. Each lane corresponds to a ¢ctive sample. Each number represents one simple SSCP pattern. For each region, the complex pat- tern 1,2 corresponding to the superimposition of simple patterns 1 and 2 is also represented. allele of the region (each SSCP band corresponds to the prevalence of certain types in the di¡erent Euro- one of the two complementary strands). Among the pean cities was detected. samples from 77 European patients analysed to date The major advantages of the SSCP method are its (56 from Switzerland, 11 from France, four from rapidity and low cost, a¡ording the analysis of many Denmark, three from Belgium, three from Ger- samples. Another advantage is that several loci can many), three to ¢ve di¡erent simple patterns have be analysed. This allows the analysis of population been identi¢ed for each genomic region (Fig. 1). A genetics and thus the study of the population struc- patient with a sample displaying only simple patterns ture of P. carinii. A potential drawback may be the for each of the four regions was probably infected by inability to detect polymorphisms of some DNA se- a single P. carinii hominis type. Thus, each combina- quences. Single bp polymorphisms were detected in tion of four simple patterns allows a type to be de- L-tubulin and mt26S [40], but this might be not the ¢ned. Twenty-¢ve patients were probably infected by case for ITS1. Indeed, SSCP analysis revealed only a single type and 13 di¡erent types were identi¢ed three di¡erent ITS1 patterns in our samples, whereas among them (Table 2). The large diversity of types the existence of 13 types has been described with suggests that this method will be useful for epidemio- DNA sequencing (see Table 1). This discordance logical studies. A complex SSCP pattern consisted of may be due to a low diversity in our samples but it more than two bands. It corresponded to the super- is also possible that the ITS1 single strands adopt imposition of simple patterns (Fig. 1). The observa- conformations which are particularly stable, thus tion of at least one complex pattern for any of the preventing e¤cient detection of polymorphisms. An- four regions in a given sample suggests an infection other disadvantage of the SSCP approach is the with several P. carinii hominis types. Among the 77 amount of work needed to set up the conditions of patients, 43 could have been infected by at least two SSCP, as well as the necessity to validate each new types and 9 by at least three types (because at least simple pattern by cloning and sequencing. one complex pattern made of two and three alleles, respectively, was observed). Among the 77 patients, 36 patients had a unique 7. The problem of potential co-infection with several combination of patterns for the four regions. The P. carinii hominis types remaining 41 patients had the same result as other patients and were distributed as follows : ten pairs, Typing of P. carinii hominis is complicated by the two groups of three and four groups of four. How- fact that co-infections with two or more types prob- ever, further investigation of these cases did not re- ably occur. This phenomenon has been demon- veal obvious epidemiological links. No increase in strated in rats and ferrets. Co-infections in humans Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 32 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 were suggested by the presence of two or sometimes quency of co-infections because of the limited num- three alleles of a genomic region in samples from ber of TSO available (see Section 5). Co-infections single patients. However, there are other possible which involved alleles hybridising to the same oligo- explanations for these ¢ndings : heterozygosity of nucleotide for both ITS1 and ITS2 could not be diploid organisms or presence of two or more copies detected. of the same region per genome, with variation be- With SSCP, the production of one complex pat- tween the copies. In the absence of data, these alter- tern for at least one of the four regions investigated native explanations cannot be ¢rmly excluded [40]. suggests a potential co-infection. A complex pattern Moreover, the three possibilities are not mutually made of two or three alleles suggests the presence of exclusive. at least two or three co-infecting types, respectively. The typing methods presented above used di¡erent Among the 77 samples analysed, samples yielding at strategies to detect these potential co-infections. In least one complex pattern with two or three alleles some studies, PCR products were systematically represented 57 and 12%, respectively. Thus, co-infec- cloned and two or three clones of most of these tion appears to occur in up to 69% of the samples. samples were sequenced [33,34]. When more than The sensitivity of silver staining used in SSCP allows one allele was found, co-infection was postulated. the detection of an allele if it represents at least 5% This was observed in 20^35% of the samples exam- of the molecules (our unpublished data). ined. This means that for these samples, the sequence Our SSCP method detects potential co-infection in of one clone was di¡erent than those of the two a greater (69%) proportion of samples than other other clones analysed. This approach detects an al- approaches (10^35%). There are two possibilities to lele only if this allele represents at least about 33% of explain these di¡erences. First, as discussed above, the molecules. Thus, this method detects a co-infect- the sensitivity of the other approaches seems lower ing type only if present in a signi¢cant proportion of than that of SSCP. Second, the di¡erences might the whole population. re£ect true di¡erences in the proportion of co-infec- Direct sequencing of PCR products has also been tion between di¡erent populations. In favour of the used [26^28]. Co-infections were assessed when dou- latter possibility, a group observed approximately ble bands or peaks were observed. This was found in 85% of potential co-infections among Brazilian 10^30% of the samples analysed. These samples were transplant patients, but only about 5% among further cloned and several clones were sequenced. In AIDS patients from the United States (C.B. Beard, one such study [26], using autoradiography, an allele personal communication). This di¡erence could not was detected if it represented 20% or more of the be explained by methodological problems because molecules. both results were obtained using direct sequencing. In studies using the TSO typing method, co-infec- This suggests that other factors, such as the type of tions were assumed if detected by their hybridisation patient population, the prophylaxis regimen and/or to two oligonucleotides speci¢c for di¡erent alleles of the geographical location may a¡ect the rate of co- same locus (ITS1 or ITS2). In order to type these infection. To resolve this issue, di¡erent methods samples using TSO, the ITS1^ITS2 region of the co- should be applied to the same samples and the re- infecting types were ampli¢ed separately by type-spe- sults compared. ci¢c PCRs [37]. (Type-speci¢c PCR ampli¢es a spe- ci¢c allele by the use of a primer which contains mismatches at its 3P-end for other allele(s).) Co-in- 8. The contribution of typing to P. carinii hominis fections were found in 20^30% of the patients using epidemiology this experimental protocol [36,37]. Control experi- ments suggested that the method was very sensitive Typing methods developed for P. carinii hominis because, in a mixture of alleles, it could detect an have already contributed to the understanding of allele 100 times less abundant than the other allele(s) some epidemiological features of this special form. (C.-H. Lee, personal communication). However, The question of reactivation of latent organisms ver- this approach has certainly underestimated the fre- sus de novo infection in recurrent P. carinii pneumo- Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 33 nia was investigated by DNA sequence analysis of 9. Concluding remarks mt26 or of the ITS1^ITS2 combination [26,27,32, 34,35]. In about 50% of the cases analysed, di¡erent Until a culture method is established, the study of alleles were found in the second episode of the dis- P. carinii genetic heterogeneity will probably rely on ease compared to the ¢rst. Because of the possibility the methods discussed above. Karyotyping, MEE of undetected co-infections, type-speci¢c PCR has and DNA sequencing appear to be appropriate for been used to investigate the possibility of the pres- investigations at the evolutionary level. DNA se- ence of a low amount of another allele during the quencing, TSO and SSCP analyses of PCR products ¢rst PCP episode [32]. This possibility was excluded from variable regions of the genome should be useful in four out of ¢ve cases analysed. In one study [35], for epidemiological studies. However, it is not ex- the alleles of both mt26S and the ITS1^ITS2 combi- cluded that other methods for polymorphism detec- nation were di¡erent in the second episode. This ex- tion, such as heteroduplex analysis or structure-spe- cluded that the alteration of alleles was due to mu- ci¢c endonuclease cleavage [45,46], might also be tation because the probability of concurrent useful, but they have not yet been explored. mutations at independent loci is very low. These re- sults strongly suggested that de novo infections occur frequently in recurrent PCP. Acknowledgments Sequencing of the ITS1^ITS2 region was used to investigate hypothetical transmission within cou- The study of P. carinii epidemiology in our labo- ples [41]. Di¡erent alleles of both ITS1 and ITS2 ratory is partially supported by Grants 94-7213 and were found in the members of each of the three cou- 97-7299 of the Swiss National Programme on AIDS ples analysed, suggesting that transmission did not Research. We thank A.E. Wake¢eld for fruitful dis- occur. cussions and P. Shaw for critical reading of the The TSO method has been used to investigate the manuscript. air of PCP patients' rooms [42,43]. Although only six combinations of ITS1 and ITS2 alleles can be di¡er- References entiated by this method, the results were compatible with dispersion of P. carinii organisms in the envi- [1] Stringer, J.R. and Walzer, P.D. (1996) Molecular biology and ronment because the same ITS1^ITS2 combination epidemiology of Pneumocystis in AIDS. AIDS 10, 561^571. was found both in the air and the clinical sample of [2] Stringer, J.R. (1996) Pneumocystis carinii ^ what is it, exactly ? the patient in 11 cases out of 14. Clin. Microbiol. Rev. 9, 489^498. Although these studies contribute to the under- [3] Wake¢eld, A.E. (1995) Re-examination of epidemiological concepts. In : Pneumocystis carinii, Baillieres Clin. Infect. standing of P. carinii hominis epidemiology, some Dis. 2, 431^448. aspects of the results should be taken with caution [4] Tibayrenc, M. (1995) Population genetics of parasitic proto- because the typing methods used have not been fully zoa and other microorganisms. Adv. Parasitol. 36, 47^115. evaluated according to proposed guidelines [44]. In [5] Matsumoto, Y. and Yoshida, Y. (1984) Sporogony in Pneu- particular, their power of discrimination has not mocystis carinii : synaptonemal complexes and meiotic nuclear divisions observed in precysts. J. Protozool. 31, 420^428. been assessed. A low discriminatory power could [6] Hong, S.T., Steele, P.E., Cushion, M.T., Walzer, P.D., String- have resulted, for example, in an underestimation er, S.L. and Stringer, J.R. (1990) Pneumocystis carinii karyo- of the frequency of de novo infection in recurrent types. J. Clin. Microbiol. 28, 1785^1795. PCP. In addition, the possibility of undetected co- [7] Stringer, J.R., Stringer, S.L., Zhang, J., Baughman, R., Smu- infections has not been considered in the studies of lian, A.G. and Cushion, M.T. (1993) Molecular genetic dis- tinction of Pneumocystis carinii from rats and humans. J. Eu- the couples and air samples mentioned above. For karyotic Microbiol. 40, 733^741. example, if the alleles carried by one partner were [8] Lundgren, B., Cotton, R., Lundgren, J.D., Edman, J.C. and found at low levels in the other partner, the possi- Kovacs, J.A. (1990) Identi¢cation of Pneumocystis carinii bility of transmission within the couple could have chromosomes and mapping of ¢ve genes. Infect. Immun. 58, been missed. 1705^1710. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 34 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 Sorex araneus and Sorex caecutiens. J. Wildlife Dis. 29, 273^ [9] Cushion, M.T., Kaselis, M., Stringer, S.L. and Stringer, J.R. (1993) Genetic stability and diversity of Pneumocystis carinii [25] Shah, J.S., Pieciak, W., Liu, J., Buharin, A. and Lane, D.J. infecting rat colonies. Infect. Immun. 61, 4801^4813. (1996) Diversity of host species and strains of Pneumocystis [10] Cushion, M.T., Zhang, J., Kaselis, M., Giuntoli, D., Stringer, carinii is based on rRNA sequences. Clin. Diagn. Lab. Immu- S.L. and Stringer, J.R. (1993) Evidence for two genetic var- nol. 3, 119^127. iants of Pneumocystis carinii coinfecting laboratory rats. [26] Keely, S.P., Stringer, J.R., Baughman, R.P., Linke, M.J., J. Clin. Microbiol. 31, 1217^1223. Walzer, P.D. and Smulian, A.G. (1995) Genetic variation [11] Weinberg, G.A. and Durant, P.J. (1994) Genetic diversity of among Pneumocystis carinii hominis isolates in recurrent pneu- Pneumocystis carinii derived from infected rats, mice, ferrets, and cell cultures. J. Eukaryotic Microbiol. 41, 223^228. mocystosis. J. Infect. Dis. 172, 595^598. [27] Latouche, S., Poirot, J.L., Bernard, C. and Roux, P. (1997) [12] Cushion, M.T. (1998) Pneumocystis carinii. Topley and Wil- Study of internal transcribed spacer and mitochondrial large- son's Microbiology and Microbial Infections, 9th edn. (Collier subunit genes of Pneumocystis carinii hominis isolated by re- L., Balows A. and Sussman M., Eds.), Edward Arnold, Lon- peated bronchoalvelolar lavage from human immunode¢- don, pp. 645^683. [13] Mazars, E., Guyot, K., Durand, I., Dei-Cas, E., Boucher, S., ciency virus-infected patients during one or several episodes of pneumonia. J. Clin. Microbiol. 35, 1687^1690. Abderrazak, S.E., Banuls, A.L., Tibayrenc, M. and Camus, [28] Latouche, S., Ortona, E., Mazars, E., Margutti, P., Tambur- D. (1997) Isoenzyme diversity in Pneumocystis carinii from rini, E., Siracusano, A., Guyot, K., Nigou, M. and Roux, P. rats, mice, and rabbits. J. Infect. Dis. 175, 655^660. (1997) Biodiversity of Pneumocystis carinii hominis : typing [14] Mazars, E., Odberg-Ferragut, C., Durand, I., Tibayrenc, M., Dei-Cas, E. and Camus, D. (1994) Genomic and isoenzymatic with di¡erent DNA regions. J. Clin. Microbiol. 35, 383^387. [29] Lee, C.H., Lu, J.J., Bartlett, M.S., Durkin, M.M., Liu, T.H., markers of Pneumocystis from di¡erent host species. J. Eu- karyotic Microbiol. 41, 104S. Wang, J., Jiang, B. and Smith, J.W. (1993) Nucleotide se- quence variation in Pneumocystis carinii strains that infect [15] Sinclair, K., Wake¢eld, A.E., Banerji, S. and Hopkin, J.M. humans. J. Clin. Microbiol. 31, 754^757. (1991) Pneumocystis carinii organisms derived from rat and human hosts are genetically distinct. Mol. Biochem. Parasitol. [30] Latouche, S., Roux, P., Poirot, J.L., Lavrard, I., Hermelin, B. and Bertrand, V. (1994) Preliminary results of Pneumocystis 45, 183^184. [16] Banerji, S., Lugli, E.B. and Wake¢eld, A.E. (1994) Identi¢ca- carinii strain di¡erentiation by using molecular biology. tion of two genetically distinct strains of Pneumocystis carinii J. Clin. Microbiol. 32, 3052^3053. [31] Wake¢eld, A.E., Fritscher, C.C., Malin, A.S., Gwanzura, L., in infected ferret lungs. J. Eukaryotic Microbiol. 41, 73S. [17] Kitada, K., Oka, S., Kimura, S., Shimada, K., Serikawa, T., Hughes, W.T. and Miller, R.F. (1994) Genetic diversity in human-derived Pneumocystis carinii isolates from four geo- Yamada, J., Tsunoo, H., Egawa, K. and Nakamura, Y. (1991) Detection of Pneumocystis carinii sequences by polymerase graphical locations shown by analysis of mitochondrial chain reaction : animal models and clinical application to non- rRNA gene sequences. J. Clin. Microbiol. 32, 2959^2961. invasive specimens. J. Clin. Microbiol. 29, 1985^1990. [32] Keely, S.P., Baughman, R.P., Smulian, A.G., Dohn, M.N. [18] Banerji, S., Lugli, E.B., Miller, R.F. and Wake¢eld, A.E. and Stringer, J.R. (1996) Source of Pneumocystis carinii in recurrent episodes of pneumonia in AIDS patients. AIDS (1995) Analysis of genetic diversity at the arom locus in iso- lates of Pneumocystis carinii. J. Eukaryotic Microbiol. 42, 10, 881^888. 675^679. [33] Lu, J.J., Bartlett, M.S., Shaw, M.M., Queener, S.F., Smith, [19] Liu, Y. and Leibowitz, M.J. (1993) Variation and in vitro J.W., Ortiz-Rivera, M., Leibowitz, M.J. and Lee, C.H. (1994) splicing of group I introns in rRNA genes of Pneumocystis Typing of Pneumocystis carinii strains that infect humans based on nucleotide sequence variations of internal tran- carinii. Nucleic Acids Res. 21, 2415^2421. [20] Peters, S.E., Wake¢eld, A.E., Whitwell, K.E. and Hopkin, scribed spacers of rRNA genes. J. Clin. Microbiol. 32, J.M. (1994) Pneumocystis carinii pneumonia in thoroughbred 2904^2912. foals : identi¢cation of a genetically distinct organism by DNA [34] Tsolaki, A.G., Miller, R.F., Underwood, A.P., Banerji, S. and ampli¢cation. J. Clin. Microbiol. 32, 213^216. Wake¢eld, A.E. (1996) Genetic diversity at the internal tran- [21] Christensen, C.B., Settnes, O.P., Bille-Hansen, V., Jorsal, S.E., scribed spacer regions of the rRNA operon among isolates of Henriksen, S.A. and Lundgren, B. (1996) Pneumocystis carinii Pneumocystis carinii from AIDS patients with recurrent pneu- from pigs and humans are antigenically distinct. J. Med. Vet. monia. J. Infect. Dis. 174, 141^156. Mycol. 34, 431^433. [35] Keely, S.P. and Stringer, J.R. (1997) Sequences of Pneumo- [22] Laakkonen, J. and Sukura, A. (1997) Pneumocystis carinii of cystis carinii f. sp. hominis strains asssociated with recurrent the common shrew, Sorex araneus, shows a discrete pheno- pneumonia vary at multiple loci. J. Clin. Microbiol. 35, 2745^ type. J. Eukaryotic Microbiol. 44, 117^121. 2747. [23] Peters, S.E., English, K., Laakkonen, J. and Gurnell, J. (1994) [36] Lu, J.J., Bartlett, M.S., Smith, J.W. and Lee, C.H. (1995) DNA analysis of Pneumocystis carinii infecting Finnish and Typing of Pneumocystis carinii strains with type-speci¢c oligo- English shrews. J. Eukaryotic Microbiol. 41, 108S. nucleotide probes derived from nucleotide sequences of inter- [24] Laakkonen, J., Sukura, A., Haukisalmi, V. and Henttonen, H. nal transcribed spacers of rRNA genes. J. Clin. Microbiol. 33, (1993) Pneumocystis carinii and helminth parasitism in shrews 2973^2977. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98 P.M. Hauser et al. / FEMS Immunology and Medical Microbiology 22 (1998) 27^35 35 and Smith, J.W. (1996) Types of Pneumocystis carinii detected [37] Jiang, B.D., Lu, J.J., Li, B.Z., Tang, X., Bartlett, M.S., Smith, in air samples. J. Eukaryotic Microbiol. 43, 44S. J.W. and Lee, C.H. (1996) Development of type-speci¢c PCR for typing Pneumocystis carinii f. sp. hominis based [43] Bartlett, M.S., Vermund, S.H., Jacobs, R., Durant, P.J., Shaw, M.M., Smith, J.W., Tang, X., Lu, J.J., Li, B., Jin, S. on sequence nucleotide variations of internal transcribed spacer regions of rRNA genes. J. Clin. Microbiol. 34, 3245^ and Lee, C.H. (1997) Detection of Pneumocystis carinii DNA in air samples : likely environmental risk to susceptible per- sons. J. Clin. Microbiol. 35, 2511^2513. [38] Atzori, C., Agostoni, F., Cargnel, A., Perenboom, R. and Beckers, P. (1996) ITSs typing of P. carinii samples from Italy, [44] Struelens, M.J., Members of the European Study Group on Epidemiological Markers (ESGEM) (1996) Consensus guide- The Netherlands and Tanzania. J. Eukaryotic Microbiol. 43, 43S. lines for appropriate use and evaluation of microbial epide- miologic typing systems. Clin. Microbiol. Infect. 2, 2^11. [39] Hayashi, K. and Yandell, D.W. (1993) How sensitive is PCR- [45] Brow, M.A., Oldenburg, M.C., Lyamichev, V., Heisler, L.M., SSCP ? Hum. Mutat. 2, 338^346. Lyamicheva, N., Hall, J.G., Eagan, N.J., Olive, D.M., Smith, [40] Hauser, P.M., Francioli, P., Bille, J., Telenti, A. and Blanc, L.M., Fors, L. and Dahlberg, J.E. (1996) Di¡erentiation of D.S. (1997) Typing of Pneumocystis carinii f. sp. hominis by single-strand conformation polymorphism of four genomic re- bacterial 16S rRNA genes and intergenic regions and Myco- bacterium tuberculosis katG genes by structure-speci¢c endo- gions. J. Clin. Microbiol. 35, 3086^3091. nuclease cleavage. J. Clin. Microbiol. 34, 3129^3137. [41] Latouche, S., Poirot, J.L., Maury, E., Bertrand, V. and Roux, [46] Lyamichev, V., Brow, M.A. and Dahlberg, J.E. (1993) Struc- P. (1997) Pneumocystis carinii hominis sequencing to study ture-speci¢c endonucleolytic cleavage of nucleic acids by eu- hypothetical person-to-person transmission. AIDS 11, 549. [42] Bartlett, M.S., Lu, J.J., Lee, C.H., Durant, P.J., Queener, S.F. bacterial DNA polymerases. Science 260, 778^783. Downloaded from https://academic.oup.com/femspd/article-abstract/22/1-2/27/462984 by Ed 'DeepDyve' Gillespie user on 27 April 2018 FEMSIM 891 6-10-98

Journal

Journal of the Endocrine Society – Oxford University Press

Published: Sep 1, 1998

There are no references for this article.