- Email: [email protected]
Preference, adaptation and survival of Mycoplasma pneumoniae subtypes in an animal model
ARTICLE IN PRESS
International Journal of Medical Microbiology 294 (2004) 149–155 www.elsevier.de/ijmm
REVIEW
Preference, adaptation and survival of Mycoplasma pneumoniae subtypes in an animal model Roger Dumkea,, Ina Catreinb, Richard Herrmannb, Enno Jacobsa a
Institut fu¨r Medizinische Mikrobiologie und Hygiene, Technische Universita¨t, Medizinische Fakulta¨t Carl Gustav Carus, Fetscherstr. 74, D-01307 Dresden, Germany
b
Zentrum fu¨r Molekulare Biologie, Universita¨t Heidelberg, Im Neuenheimer Feld 282, D-69120 Heidelberg, Germany
Abstract The interaction between Mycoplasma pneumoniae and its natural host, humans, cannot be studied directly for obvious reasons. Therefore, we used guinea pigs instead, which had been recently introduced as an acceptable alternative host organism. The following experimental approaches were taken to study the pathogen–host relationship: characterization and subtyping of M. pneumoniae strains isolated from human patients, infection of guinea pigs with selected M. pneumoniae strains, and analysis of adaptation, preference and survival of individual strains in guinea pigs under competitive conditions. The results of our study indicated that the species M. pneumoniae is genetically very homogeneous. From 115 independently isolated strains two subtypes and one variant were found. The subtypes differed significantly in the amino acid composition of the P1 protein, the main adhesin of M. pneumoniae, while the variant showed only minor amino acid exchanges. Infection of guinea pigs indicated differences between the subtypes and the variant in their ability to colonize and survive in the animal. Preinfection of the host with a certain subtype or variant caused a subtype-specific immunity and had a strong influence on the type of surviving bacteria in superinfection experiments. The results of these studies explain the shift of subtypes frequently observed in epidemic outbreaks of M. pneumoniae infection appearing in intervals of 3–7 years. r 2004 Elsevier GmbH. All rights reserved. Keywords: Mycoplasma pneumoniae; Epidemiology; Infection; Adaptation
Content Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 Epidemiology of M. pneumoniae diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 The animal model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Adaptation of M. pneumoniae to its host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Subtype-specific host immune response as important factor for outbreaks due to M. pneumoniae? . . . . . . . . . . . . . . . . . . . . 153 Corresponding author. Tel.: +49-351-458-6577; fax: +49-351-458-6310.
E-mail address: [email protected] (R. Dumke). 1438-4221/$ - see front matter r 2004 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijmm.2004.06.020
ARTICLE IN PRESS 150
R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Introduction Patients suffering from Mycoplasma pneumoniae infection exhibit mild respiratory symptoms which are disproportionate to the massive infiltrations seen on chest roentgenograms (Jacobs, 2002). The failure to culture an agent from respiratory material of such patients stimulated Eaton et al. (1944) to search for an organism which was not detectable by Gram staining (for a historical commentary, see Marmion, 1990). Finally, M. pneumoniae was identified as the causative agent of this respiratory infection. It is one of the smallest known bacteria (Morowitz and Tourtellotte, 1962) with a genome size of 816 kbp (Himmelreich et al., 1996) and a proposed coding capacity for 688 open reading frames (ORF) (Dandekar et al., 2000). The pleiomorphic M. pneumoniae is characterized by the complete lack of a cell wall. It is only surrounded by a cytoplasmic membrane, which contains – unusual for eubacteria – cholesterol. A cytoskeleton-like proteinaceous structure provides probably stability against osmotic and mechanical stress (Krause, 1996). M. pneumoniae depends in nature on humans as the only host because during its reductive evolution from more complex bacteria it has lost many metabolic and anabolic pathways. To interact efficiently with the host, M. pneumoniae developed a specific attachment organelle, also named tip structure, which is the prerequisite for adherence of the bacteria to the surface of epithelia
Fig. 1. Organization of the P1 operon of M. pneumoniae with ORF5 which is encoding the P1 adhesin. The ORF5 gene contains two copies each (striped areas) of the 10 different RepMP2/3 and the 8 RepMP4 elements which are scattered all over the genome of M. pneumoniae M129 and are found as active or silent expressed genome regions in both subtypes. Further minor recombined regions distinguish the known variants and subtypes (black bars).
cells in the respiratory tract and functions in cell division and gliding motility (Razin and Jacobs, 1992; Krause and Balish, 2001). The attachment organelle appears as an asymmetric extension of the cell body. To function properly, certain proteins, for instance the main adhesin, the P1 protein, have to be correctly inserted into this organelle. The characterization of various cytadherencenegative mutants resulted in a further understanding of the unique architecture and the probable step-wise assembly of the attachment organelle (Krause and Balish, 2004). The P1 protein serves also as important marker for subtyping. So far two significantly different P1 sequences led to the classification of only two subtypes (Su et al., 1990). The sequence differences are caused by two repetitive DNA sequences called RepMP2/3 and RepMP4 (Fig. 1). These sequences are between 1.1 and 1.8 kbp long and appear 8–10 times dispersed throughout the genome (Ruland et al., 1990; Himmelreich et al., 1996). They are characterized by a variable middle region, which is flanked by two conserved regions. The P1 gene of each subtype contains a specific set of RepMP2/3 and RepMP4. Exchange of these copies is supposed, but not experimentally proven, to cause a subtype shift by gene conversion. The P1 gene (ORF5) is organized in the P1 operon together with two other genes, denoted ORF4 and ORF6. The products of the ORF6 gene are the proteins P40 and P90, which arose from a larger precursor by cleavage. A defect in the ORF6 gene causes an adherence-negative phenotype (Franzoso et al., 1993). Data from crosslinking experiments suggest, that P40 and P90 form a complex with P1 and direct the correct insertion into the attachment organelle (Layh-Schmitt and Herrmann, 1994). Similar to the P1 gene, the ORF6 gene contains also repetitive DNA sequences (RepMP5), which are 1.8–2.0 kbp long and exist in 8 copies. Prototype for subtype 1 is M. pneumoniae M129 (ATCC 29343) and for subtype 2 M. pneumoniae FH (ATCC 15531). The reduced genome size with the relative small number of genes, the knowledge of the complete genome sequence (Himmelreich et al., 1996), detailed studies on the expression of these genes (Ueberle et al., 2002; Weiner et al., 2003), the parasitic life style and the possibility to use guinea pigs as a host organism (Jacobs et al., 1988) make M. pneumoniae to a promising model organism for studying a parasite-host relationship.
ARTICLE IN PRESS R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Epidemiology of M. pneumoniae diseases Common etiological agents of non-nosocomial acquired infectious diseases of the respiratory tract are very often observed following an accumulation of patient cases in the wet and cold season. For M. pneumoniae, as for other bacterial and viral agents the epidemiological course depends on close contacts to humans. Fast replicating agents, e.g. the influenza or SARS virus, usually within a short time induce hundreds of patient cases which should lead to fast reports in health information systems and to worldwide attention regarding a special outbreak. Quite different is the experience with agents characterized by a fairly long incubation period as seen with M. pneumoniae-caused diseases. In many field studies, the knowledge of an accumulation of respiratory diseases due to M. pneumoniae was more or less a by-product of national programs which were established to provide an overview about the seasonal outbreaks of most common infectious respiratory diseases, especially influenza. Analysing former outbreaks, M. pneumoniae subtypespecific infections were registered in the Northern Hemisphere with one or more years delay between Japan, Scandinavia and Central Europe (Ursi et al., 1994; Jacobs et al., 1996; Sasaki et al., 1996; Cousin-Allery et al., 2000; Dorigo-Zetsma et al., 2000). In the years following a peaking outbreak, we and others found that the predominant specific subtype had changed. However, the characterization of M. pneumoniae strains isolated during the endemic periods proved a further circulation of both of the M. pneumoniae subtypes in the human population at a low level and may be the reservoir for subsequent outbreaks which were reported in time intervals of 3–7 and even more years later. Recently, we published a study on subtyping M. pneumoniae isolates of a strain collection containing 91 patient isolates from Germany, 14 isolates from USA and 10 from France (Dumke et al., 2003). Genotypic approaches including multilocus sequence typing of eleven housekeeping genes, six different structural genes involved in the adherence process and five repetitive elements of unknown function, indicated that the major variations of the different M. pneumoniae strains were located in the P1 and ORF6 gene. Screening these different patient isolates, all strains could be divided into two major subtypes and one further variant with close relationship to subtype 2 (Dumke et al., 2003). A subtype is defined by the specific copies of RepMP2/3 and RepMP4 in the P1 gene and the specific copy of RepMP5 in the ORF6 gene, while a variant is characterized by the exchange of a small number of amino acids in either one of these repetitive DNA sequences. Worldwide, the same two subtypes were isolated most frequently, but the variant 2 of subtype 2 was also only rarely detected in Japan and in Europe
151
(Jacobs et al., 1996; Sasaki et al., 1996; Kenri et al., 1999). To our knowledge, only one strain of the variant 1 was isolated in Denmark recently (Dorigo-Zetsma et al., 2001a). So far, differences between the subtypes and variants with regard to severity of disease, colonization of the respiratory tract of patients or resistance to any antibiotics are not reported. Information on reported outbreaks due to M. pneumoniae goes back to the year 1946. Lind et al. (1997) tested sera of patients collected at the Statens Serum Institute in Copenhagen, Denmark, and found elevated titers of anti-M. pneumoniae antibodies in a regular pattern of epidemics which showed an accumulation every 4.5 years (mean), which were followed by hypoendemic periods. A large epidemic was registered in the winter of 1991 lasting until 1993. It was characterized by an unusual high number of pneumonia cases due to M. pneumoniae infections. Two national influenza surveillance studies included also testing of patients for M. pneumoniae as agent of a respiratory tract disease. The results showed unexpectedly high numbers of M. pneumoniae infections. The overall calculated incidences were 1234 per 100,000 persons in France in the winter 1992/3 decreasing to 190 per 100,000 persons in 1995/6 (Layani-Milon et al., 1999). The data for the outbreak in the Netherlands were in a similar range of 587 per 100,000 for a patient population with respiratory diseases consulting outpatient general practitioners (Dorigo-Zetsma et al., 2001b). After this long lasting major outbreak and the following hypoendemic period a new increase starting in 1995/6 and peaking in 1998 was registered. Mostly because of the lack of experience in cultivating and conserving M. pneumoniae patient isolates, only a few laboratories worldwide were able to follow the outbreaks and to subtype the isolated strains. Analysing 250 isolates collected in Japan, Sasaki et al. (1996) found that the most common subtype in 1976 was subtype 1, but changed in 1979 and 1980 to subtype 2. From 1985 to 1991 almost all isolates were of subtype 1, and as from 1992 subtype 2 was isolated again with increasing tendency. Analysis of our collection of 91 isolates from Germany revealed that the first strains from 1989 were of subtype 2 (Dumke et al., 2003). The outbreak strains in the winter 1992/3 were of subtype 1, followed by the reappearence of subtype 2 in the year 1998 indicating that switching of subtypes is following a global pattern.
The animal model From these epidemiological data one could hypothesize that outbreaks with subtype-specific M. pneumoniae led to an immune response generating protecting
ARTICLE IN PRESS 152
R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Fig. 2. Schematic diagram of the in vivo experiments using guinea pigs as host model and patient isolates of M. pneumoniae belonging to subtype 1, subtype 2 and the variant 2. The combination of strains characterized as subtype 1, subtype 2 or variant 2 were used in a three-component infection mixture and applied to the animals via an intranasal infection to test the efficiency of the three different strains to colonize and propagate in the guinea pig respiratory tract within seven to ten days. To evaluate the percentage of each of the subtypes and the variant 2, hundred colonies of the agarplated infection mixture and of each of the recovered bronchoalveolar lavages were picked and subtyped using PCR. In 13 experiments with different M. pneumoniae threecomponent combinations the percentage of colonies belonging to the subtypes and the variant 2 in the infection mixture and in the corresponding bronchoalveolar lavage were calculated (see Fig. 3). In a further experiment guinea pigs were preinfected intranasally three times within 38 days with a variant 2 strain alone to establish variant 2 immunocompetent animals. After another 10 days the competent animals were infected with the three-component infection mixture and the recovered colonies from the bronchoalveolar lavages were subtyped (see Fig. 4).
antibodies within the human population against the dominating subtype of the epidemic peak. This should give the alternative subtype an advantage over the strains causing the last outbreak. To test the hypothesis, we infected guinea pigs intranasally and analysed whether the different subtypes or the variant 2 were able to colonize the respiratory tract with the same efficiency. In addition, we wanted to learn whether a preinfection with a defined subtype or variant alters its ability to colonize the respiratory tract for a second time and promotes the preferential colonization by other subtypes or variants. For the animal experiments, the patient isolates and the isolates recovered from bronchial fluids of the guinea
Fig. 3. Mean proportions of subtype 1, subtype 2 and variant 2 M. pneumoniae cells in the three-component infection mixture and in the corresponding bronchoalveolar lavages of naı¨ ve guinea pigs. To characterize the efficacy of each strain to colonize and to propagate in the respiratory tract of guinea pigs within seven to ten days after intranasal infection, different three-component infection mixtures with various strain combinations were tested (n ¼ 13).
pigs were typed by various parameters including Western blotting with mono-specific antibodies, PCR and DNA sequencing (for a description of the typing methods, see Dumke et al., 2003). The established procedure for infecting guinea pigs with M. pneumoniae (Jacobs et al., 1988) was modified by intranasal application of a mean dose of 6 107 colony-forming units of M. pneumoniae cells. To show subtype- and variant-specific differences in host colonization and propagation, we infected the animals with a mixture of three M. pneumoniae patient isolates which belonged to the subtype 1, the subtype 2 and the variant 2 of M. pneumoniae (Fig. 2). Each of the 13 animals was infected with a mixture composed of different patient strains, isolated at different time periods, to exclude the dominance of a single strain within the three tested groups. As a reference, parallel to the infection of the guinea pigs, the mixture of M. pneumoniae strains was cultured on agar plates and the colonies were counted and picked for typing with PCR methods (for a description of the typing methods see Dumke et al., 2003). Unexpectedly, under the applied conditions the variant 2 showed the highest survival rate, which, however, was statistically not significantly different from subtype 1, whereas subtype 2 was less fit for survival and reproduction in these animals (Fig. 3). Obviously, the in vivo experiments led to a selection of variant 2- and of subtype 1-specific strains. Since the number of in vitro passages (4–10) was different for each strain used, we had to exclude that the survival or the ability to adhere was influenced by the in vitro passages. Therefore, animals were infected with subtype 2, and in
ARTICLE IN PRESS R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
153
a second step another set of animals were infected with the freshly recovered subtype 2 isolates. This approach did not result in a higher recovery of subtype 2, indicating that the reduced survival rate of subtype 2 in guinea pigs is not correlated with differences in the number of in vitro passages of subtype 2 isolates prior to infection (results not shown). The use of other wild-type (e.g. variant 1) or genetically modified M. pneumoniae strains in these animal model experiments in the future may reveal differences in virulence or host colonization pattern between the strains.
Adaptation of M. pneumoniae to its host Comparing the results from the above-described experiments with guinea pigs with the data from the natural human host, the major difference concerns the dominance of variant 2 in the animal model. This is in contrast to its rare isolation from the human respiratory tract. In humans, subtype 1 and also subtype 2 were detected most frequently in the respiratory material. According to our subtyping results only the subtypes 1 and 2 were responsible for outbreaks of M. pneumoniae infections. The variant 2 was isolated only sporadically and did not seem to be a relevant cause of epidemics. The preference for subtype 1 and 2 in the human respiratory tract and on the other hand the preference for variant 2 and subtype 1 in guinea pigs might be correlated with selection and adaption of M. pneumoniae strains to a specific host. During evolution, M. pneumoniae may have tested host cells with different receptor specificities. The present, constant environment of these bacteria, i.e. the human respiratory tract, is probably optimal for the growth of subtype 1 and 2. This could explain why we and others almost exclusively find subtype 1 and 2 with their conserved P1 and ORF6 genes. Nevertheless, despite the reduced genome size, M. pneumoniae keeps the repetitive DNA sequences (RepMP2/3, RepMP4, RepMP5), which make up to about 8% of the genome and which could be used, at least in theory, by M. pneumoniae to assemble new P1 and ORF6 genes by homologous recombination. We have here a clear contradiction between the results of the subtyping experiments and the genetic potential of M. pneumoniae. This discrepancy could be explained, if we assume that either new P1 genes are inferior to the subtypes 1 or 2 existing ones, and/or that homologous recombination is a very rare event and does not contribute significantly to the assembly of new P1 genes. The sporadic finding of variants without relevance for epidemic outbreaks and the reported inability to transform M. pneumoniae by inserting DNA in its genome via homologous recombination (Dhandayuthapani et al., 1999) support these hypotheses, which can be tested now experimentally.
Fig. 4. Three guinea pigs were preinfected with variant 2 alone (three times during 38 days) to establish a strong immunocompetence versus variant 2. After further ten days these three animals and further three naive animals with no prior M. pneumoniae contact were infected with a three-component infection mixture which consisted of the variant 2 strain (which was used also for preinfection) and a patient isolate of subtype 1 and of subtype 2 each. The proportions of the three strains were calculated for the infection mixture and for the bronchoalveolar lavages of the three variant 2-preinfected and the three naive animals.
Subtype-specific host immune response as important factor for outbreaks due to M. pneumoniae? In in vitro experiments using erythrocytes as host cells and M. pneumoniae reference strains of subtypes 1 and 2, Jacobs et al. (1996) were able to inhibit the adherence of bacteria by adding sera of patients suffering from M. pneumoniae diseases. In these adherence inhibition tests the sera selectively blocked adherence of subtype 1 or 2. Only a few sera inhibited both subtypes. The authors speculated that the differences in the amino acid sequences of the P1-protein were important for optimal adherence to the host but also for a subtype-specific protective immunity. The study by Jacobs et al. (1996) was hampered by the lack of patient strains and therefore patient sera had to be tested with reference strains. Nonetheless, sera taken during one of the outbreaks were able to inhibit mostly the outbreakspecific subtype. To prove the prevalence switches from one to the other subtype/variant in specifically prestimulated animals, we preinfected three guinea pigs with variant 2 and reinfected these animals with a infection mixture consisting of 30% subtype 1, 30% subtype 2 and 40% of the variant 2 (Fig. 4). In parallel to the prestimulated
ARTICLE IN PRESS 154
R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
animals, also three unstimulated guinea pigs were intranasally infected as controls. All six animals were kept separately in ‘‘filtered air box systems’’. The bronchoalveolar lavages of the six animals were collected after ten days and plated for M. pneumoniae growth. In all cases 100 colonies per animal were picked and typed by PCR (for a description of the typing methods, see Dumke et al., 2003). The bronchial fluids of the three unstimulated control animals were dominated by M. pneumoniae variant 2 with about 83% of the recovered colony-forming bacteria. The other colonies were typed as subtype 1 in almost 20% of the colonies, whereas less than 1% of the reisolates belonged to the subtype 2. In animals preinfected with variant 2 we found a major reduction of variant 2 colonies to 5%, whereas subtype 1 made up to 95% of the reisolates, and subtype 2 was again below 1%. The results of the in vivo experiments presented here show for the first time that a variant 2-specific prestimulated animal was able to develop a protective immunity to variant 2. On the other hand the protection was not complete, as under our applied experimental conditions variant 2 was not eradicated from the respiratory tract of guinea pigs. Interestingly, also not all patient sera were able to inhibit the adherence of M. pneumoniae (Jacobs et al., 1996) indicating that the P1 adhesin might be part of a much more complex biological structure with different major and minor cell adherence-associated proteins (Krause, 1996). Because of the low recovery of subtype 2 in the animal model, the closely related variant 2 was used to test the effect of prestimulating the guinea pigs and to provoke a possible reduction of variant 2 in a second infection. The experiment shown in Fig. 4 clearly revealed that the number of variant 2 colonies was drastically reduced and subtype 1 became the predominant strain. Our results indicate that we were able to mimic in the animal model the phenomenon observed in humans, i.e. a switching of the predominant subtype to the other subtype in a following epidemic outbreak. Our results may also explain why – between epidemics – both subtypes continue to circulate in the human population with a much lower number of registered patient cases probably due to the effect of a limited subtype-specific immunity.
Conclusions The knowledge about the interaction of M. pneumoniae and its host is still insufficient despite the improvement in cultivating these fastidiously growing bacteria, diagnosing them in patient respiratory material and the large body of data derived from the annotation of genome sequences. One of the important questions is
the influence of the host on the generation of new variants or subtypes; for instance, how do M. pneumoniae escape the immune defense of the host? The results of all analyses aiming at the identification and characterization of M. pneumoniae patient isolates led to the same conclusion: M. pneumoniae is genetically very homogeneous. There are in principle only two subtypes and a few variants. The latter ones do not play any role in epidemic outbreaks so far. These observations are in sharp contrast to the results of the annotation of the genome sequence of M. pneumoniae M129 (Himmelreich et al., 1996) and M. pneumoniae FH (Reiser et al., unpublished), which clearly predict a reservoir of repetitive DNA sequences, which could be used by the bacterium to assemble a large number of modified P1 or ORF6 genes by recombination processes creating many different subtypes. There are two possible explanations for this phenomenon, either new P1 and ORF6 genes are synthesized, but they are not advantageous in the hostprovided environment and disappear rather quickly, or the required recombinations do not take place. These two hypotheses can now be investigated experimentally by constructing new P1 and ORF6 genes based on the variation of the nucleotide sequences of the repetitive DNA sequence and testing them in guinea pigs for survival, adaptation and preference. To prove the principle of our approach, new ORF6 genes were constructed keeping the P1 gene constant. M. pneumoniae clones carrying the new P1/ORF6 gene combinations survived in vivo passages in guinea pigs (unpublished results). This shows that combining in vitro construction of new M. pneumoniae subtypes with testing them in an animal model is a promising strategy for evaluating the epidemic potential of new subtypes.
Acknowledgement We thank E. Pirkl for excellent technical assistance. The project was supported by the grants Ja 399/-2 and He 399-2 from the Deutsche Forschungsgemeinschaft.
References Cousin-Allery, A., Charron, A., de Barbeyrac, B., Fremy, G., Jensen, J.S., Renaudin, H., Bebear, C., 2000. Molecular typing of Mycoplasma pneumoniae strains by PCR-based methods and pulsed-field gel electrophoresis. Application to French and Danish isolates. Epidemiol. Infect. 124, 103–111. Dandekar, T., Huynen, M., Regula, J.T., Ueberle, B., Zimmermann, C.U., Andrade, M., Doerks, T., SanchezPulido, L., Snel, B., Suyama, M., Yuan, Y.P., Herrmann, R., Bork, P., 2000. Re-annotating the Mycoplasma pneumoniae genome sequence: adding value, function and reading frames. Nucleic Acids Res. 28, 3278–3288.
ARTICLE IN PRESS R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Dhandayuthapani, S., Rasmussen, W.G., Baseman, J.B., 1999. Disruption of gene mg218 of Mycoplasma genitalium through homologous recombination leads to an adherence-deficient phenotype. Proc. Natl. Acad. Sci. USA 96, 5227–5232. Dorigo-Zetsma, J.W., Dankert, J., Zaat, S.A.J., 2000. Genotyping of Mycoplasma pneumoniae clinical isolates reveals eight P1 subtypes within two genomic groups. J. Clin. Microbiol. 38, 965–970. Dorigo-Zetsma, J.W., Wilbrink, B., Dankert, J., Zaat, S.A.J., 2001a. Mycoplasma pneumoniae P1 type 1- and type 2specific sequences within the P1 cytadhesin gene of individual strains. Infect. Immun. 69, 5612–5618. Dorigo-Zetsma, J.W., Wilbrink, B., van der Nat, H., Bartelds, A.I.M., Heijnen, M.L.A., Dankert, J., 2001b. Results of molecular detection of Mycoplasma pneumoniae among patients with acute respiratory infection and in their household contacts reveals children as human reservoirs. J. Infect. Dis. 183, 675–678. Dumke, R., Catrein, I., Pirkl, E., Herrmann, R., Jacobs, E., 2003. Subtyping of Mycoplasma pneumoniae isolates based on extended genome sequencing and on expression profiles. Int. J. Med. Microbiol. 292, 513–525. Eaton, M.D., Meiklejohn, G., VanHerick, W., 1944. Studies on the etiology of primary atypical pneumonia: a filterable agent transmissible to cotton rats, hamsters, and chick embryos. J. Exp. Med. 79, 649–667. Franzoso, G., Hu, P.C., Meloni, G.A., Barile, M.F., 1993. The immunodominant 90-kilodalton protein is localized on the terminal tip structure of Mycoplasma pneumoniae. Infect. Immun. 61, 1523–1530. Himmelreich, R., Hilbert, H., Plagens, H., Pirkl, E., Li, B.C., Herrmann, R., 1996. Complete sequence analysis of the genome of the bacterium Mycoplasma pneumoniae. Nucleic Acids Res. 24, 4420–4449. Jacobs, E., 2002. Mycoplasma pneumoniae disease manifestations and epidemiology. In: Razin, S., Herrmann, R. (Eds.), Molecular Biology and Pathogenicity of Mycoplasmas. Kluwer Academic/Plenum Publishers, New York, pp. 519–530. Jacobs, E., Drews, M., Stuhlert, A., Bu¨ttner, C., Klein, P.J., Kist, M., Bredt, W., 1988. Immunological reactions of guinea pigs following intranasal Mycoplasma pneumoniae infection and immunization with the 168 kDa aherence protein. J. Gen. Microbiol. 134, 473–479. Jacobs, E., Vonski, M., Oberle, K., Opitz, O., Pietsch, K., 1996. Are outbreaks and sporadic respiratory infections by Mycoplasma pneumoniae due to two distinct subtypes? Eur. J. Clin. Microbiol. Infect. Dis. 15, 38–44. Kenri, T., Taniguchi, R., Sasaki, Y., Okazaki, N., Narita, M., Izumikawa, K., Umetsu, M., Sasaki, T., 1999. Identification of a new variable sequence in the P1 cytadhesin gene of Mycoplasma pneumoniae: evidence for the generation of antigenic variation by DNA recombination between repetitive sequences. Infect. Immun. 67, 4557–4562.
155
Krause, D.C., 1996. Mycoplasma pneumoniae cytadherence: unravelling the tie that binds. Mol. Microbiol. 20, 247–253. Krause, D.C., Balish, M.F., 2001. Structure, function, and assembly of the terminal organelle of Mycoplasma pneumoniae. FEMS Microbiol. Lett. 198, 1–7. Krause, D.C., Balish, M.F., 2004. Cellular engineering in a minimal microbe: structure and assembly of the terminal organelle of Mycoplasma pneumoniae. Mol. Microbiol. 51, 917–924. Layani-Milon, M.P., Gras, I., Valette, M., Luciani, J., Stagnara, J., Aymard, M., Lina, B., 1999. Incidence of upper respiratory tract Mycoplasma pneumoniae infections among outpatients in Rhone-Alpes, France, during five successive winter periods. J. Clin. Microbiol. 37, 1721–1726. Layh-Schmitt, G., Herrmann, R., 1994. Spatial arrangement of gene products of the P1 operon in the membrane of Mycoplasma pneumoniae. Infect. Immun. 62, 974–979. Lind, K., Benzon, M.W., Jensen, J.S., Clyde Jr., W.A., 1997. A seroepidemiological study of Mycoplasma pneumoniae infections in Denmark over the 50-year period 1946–1995. Eur. J. Epidemiol. 13, 581–586. Marmion, B.P., 1990. Eaton agent—science and scientific acceptance: a historical commentary. Rev. Infect. Dis. 12, 338–353. Morowitz, H.J., Tourtellotte, M.E., 1962. The smallest living cells. Sci. Am. 206, 117–125. Razin, S., Jacobs, E., 1992. Mycoplasma adhesion. J. Gen. Microbiol. 138, 407–422. Ruland, K., Wenzel, R., Herrmann, R., 1990. Analysis of three different repeated DNA elements present in the P1 operon of Mycoplasma pneumoniae: size, number and distribution on the genome. Nucleic Acids Res. 18, 6311–6317. Sasaki, T., Kenri, T., Okazaki, N., Iseki, M., Yamashita, R., Shintani, M., Sasaki, Y., Yayoshi, M., 1996. Epidemiological study of Mycoplasma pneumoniae infections in Japan based on PCR-restriction fragment length polymorphism of the P1 cytadhesin gene. J. Clin. Microbiol. 34, 447–449. Su, C.J., Chavoya, A., Dallo, S.F., Baseman, J.B., 1990. Sequence divergency of the cytadhesin gene of Mycoplasma pneumoniae. Infect. Immun. 58, 2669–2674. Ueberle, B., Frank, R., Herrmann, R., 2002. The proteome of the bacterium Mycoplasma pneumoniae: comparing predicted open reading frames to identified gene products. Proteomics 2, 754–764. Ursi, D., Heven, M., van Bever, H., Quint, W., Niesters, H.G.M., Goossens, H., 1994. Typing of Mycoplasma pneumoniae by PCR-mediated DNA fingerprinting. J. Clin. Microbiol. 32, 2873–2875. Weiner III, J., Zimmermann, C.U., Go¨hlmann, H.W., Herrmann, R., 2003. Transcription profiles of the bacterium Mycoplasma pneumoniae grown at different temperatures. Nucleic Acids Res. 31, 6306–6320.
International Journal of Medical Microbiology 294 (2004) 149–155 www.elsevier.de/ijmm
REVIEW
Preference, adaptation and survival of Mycoplasma pneumoniae subtypes in an animal model Roger Dumkea,, Ina Catreinb, Richard Herrmannb, Enno Jacobsa a
Institut fu¨r Medizinische Mikrobiologie und Hygiene, Technische Universita¨t, Medizinische Fakulta¨t Carl Gustav Carus, Fetscherstr. 74, D-01307 Dresden, Germany
b
Zentrum fu¨r Molekulare Biologie, Universita¨t Heidelberg, Im Neuenheimer Feld 282, D-69120 Heidelberg, Germany
Abstract The interaction between Mycoplasma pneumoniae and its natural host, humans, cannot be studied directly for obvious reasons. Therefore, we used guinea pigs instead, which had been recently introduced as an acceptable alternative host organism. The following experimental approaches were taken to study the pathogen–host relationship: characterization and subtyping of M. pneumoniae strains isolated from human patients, infection of guinea pigs with selected M. pneumoniae strains, and analysis of adaptation, preference and survival of individual strains in guinea pigs under competitive conditions. The results of our study indicated that the species M. pneumoniae is genetically very homogeneous. From 115 independently isolated strains two subtypes and one variant were found. The subtypes differed significantly in the amino acid composition of the P1 protein, the main adhesin of M. pneumoniae, while the variant showed only minor amino acid exchanges. Infection of guinea pigs indicated differences between the subtypes and the variant in their ability to colonize and survive in the animal. Preinfection of the host with a certain subtype or variant caused a subtype-specific immunity and had a strong influence on the type of surviving bacteria in superinfection experiments. The results of these studies explain the shift of subtypes frequently observed in epidemic outbreaks of M. pneumoniae infection appearing in intervals of 3–7 years. r 2004 Elsevier GmbH. All rights reserved. Keywords: Mycoplasma pneumoniae; Epidemiology; Infection; Adaptation
Content Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150 Epidemiology of M. pneumoniae diseases . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 The animal model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151 Adaptation of M. pneumoniae to its host . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153 Subtype-specific host immune response as important factor for outbreaks due to M. pneumoniae? . . . . . . . . . . . . . . . . . . . . 153 Corresponding author. Tel.: +49-351-458-6577; fax: +49-351-458-6310.
E-mail address: [email protected] (R. Dumke). 1438-4221/$ - see front matter r 2004 Elsevier GmbH. All rights reserved. doi:10.1016/j.ijmm.2004.06.020
ARTICLE IN PRESS 150
R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 Acknowledgement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
Introduction Patients suffering from Mycoplasma pneumoniae infection exhibit mild respiratory symptoms which are disproportionate to the massive infiltrations seen on chest roentgenograms (Jacobs, 2002). The failure to culture an agent from respiratory material of such patients stimulated Eaton et al. (1944) to search for an organism which was not detectable by Gram staining (for a historical commentary, see Marmion, 1990). Finally, M. pneumoniae was identified as the causative agent of this respiratory infection. It is one of the smallest known bacteria (Morowitz and Tourtellotte, 1962) with a genome size of 816 kbp (Himmelreich et al., 1996) and a proposed coding capacity for 688 open reading frames (ORF) (Dandekar et al., 2000). The pleiomorphic M. pneumoniae is characterized by the complete lack of a cell wall. It is only surrounded by a cytoplasmic membrane, which contains – unusual for eubacteria – cholesterol. A cytoskeleton-like proteinaceous structure provides probably stability against osmotic and mechanical stress (Krause, 1996). M. pneumoniae depends in nature on humans as the only host because during its reductive evolution from more complex bacteria it has lost many metabolic and anabolic pathways. To interact efficiently with the host, M. pneumoniae developed a specific attachment organelle, also named tip structure, which is the prerequisite for adherence of the bacteria to the surface of epithelia
Fig. 1. Organization of the P1 operon of M. pneumoniae with ORF5 which is encoding the P1 adhesin. The ORF5 gene contains two copies each (striped areas) of the 10 different RepMP2/3 and the 8 RepMP4 elements which are scattered all over the genome of M. pneumoniae M129 and are found as active or silent expressed genome regions in both subtypes. Further minor recombined regions distinguish the known variants and subtypes (black bars).
cells in the respiratory tract and functions in cell division and gliding motility (Razin and Jacobs, 1992; Krause and Balish, 2001). The attachment organelle appears as an asymmetric extension of the cell body. To function properly, certain proteins, for instance the main adhesin, the P1 protein, have to be correctly inserted into this organelle. The characterization of various cytadherencenegative mutants resulted in a further understanding of the unique architecture and the probable step-wise assembly of the attachment organelle (Krause and Balish, 2004). The P1 protein serves also as important marker for subtyping. So far two significantly different P1 sequences led to the classification of only two subtypes (Su et al., 1990). The sequence differences are caused by two repetitive DNA sequences called RepMP2/3 and RepMP4 (Fig. 1). These sequences are between 1.1 and 1.8 kbp long and appear 8–10 times dispersed throughout the genome (Ruland et al., 1990; Himmelreich et al., 1996). They are characterized by a variable middle region, which is flanked by two conserved regions. The P1 gene of each subtype contains a specific set of RepMP2/3 and RepMP4. Exchange of these copies is supposed, but not experimentally proven, to cause a subtype shift by gene conversion. The P1 gene (ORF5) is organized in the P1 operon together with two other genes, denoted ORF4 and ORF6. The products of the ORF6 gene are the proteins P40 and P90, which arose from a larger precursor by cleavage. A defect in the ORF6 gene causes an adherence-negative phenotype (Franzoso et al., 1993). Data from crosslinking experiments suggest, that P40 and P90 form a complex with P1 and direct the correct insertion into the attachment organelle (Layh-Schmitt and Herrmann, 1994). Similar to the P1 gene, the ORF6 gene contains also repetitive DNA sequences (RepMP5), which are 1.8–2.0 kbp long and exist in 8 copies. Prototype for subtype 1 is M. pneumoniae M129 (ATCC 29343) and for subtype 2 M. pneumoniae FH (ATCC 15531). The reduced genome size with the relative small number of genes, the knowledge of the complete genome sequence (Himmelreich et al., 1996), detailed studies on the expression of these genes (Ueberle et al., 2002; Weiner et al., 2003), the parasitic life style and the possibility to use guinea pigs as a host organism (Jacobs et al., 1988) make M. pneumoniae to a promising model organism for studying a parasite-host relationship.
ARTICLE IN PRESS R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Epidemiology of M. pneumoniae diseases Common etiological agents of non-nosocomial acquired infectious diseases of the respiratory tract are very often observed following an accumulation of patient cases in the wet and cold season. For M. pneumoniae, as for other bacterial and viral agents the epidemiological course depends on close contacts to humans. Fast replicating agents, e.g. the influenza or SARS virus, usually within a short time induce hundreds of patient cases which should lead to fast reports in health information systems and to worldwide attention regarding a special outbreak. Quite different is the experience with agents characterized by a fairly long incubation period as seen with M. pneumoniae-caused diseases. In many field studies, the knowledge of an accumulation of respiratory diseases due to M. pneumoniae was more or less a by-product of national programs which were established to provide an overview about the seasonal outbreaks of most common infectious respiratory diseases, especially influenza. Analysing former outbreaks, M. pneumoniae subtypespecific infections were registered in the Northern Hemisphere with one or more years delay between Japan, Scandinavia and Central Europe (Ursi et al., 1994; Jacobs et al., 1996; Sasaki et al., 1996; Cousin-Allery et al., 2000; Dorigo-Zetsma et al., 2000). In the years following a peaking outbreak, we and others found that the predominant specific subtype had changed. However, the characterization of M. pneumoniae strains isolated during the endemic periods proved a further circulation of both of the M. pneumoniae subtypes in the human population at a low level and may be the reservoir for subsequent outbreaks which were reported in time intervals of 3–7 and even more years later. Recently, we published a study on subtyping M. pneumoniae isolates of a strain collection containing 91 patient isolates from Germany, 14 isolates from USA and 10 from France (Dumke et al., 2003). Genotypic approaches including multilocus sequence typing of eleven housekeeping genes, six different structural genes involved in the adherence process and five repetitive elements of unknown function, indicated that the major variations of the different M. pneumoniae strains were located in the P1 and ORF6 gene. Screening these different patient isolates, all strains could be divided into two major subtypes and one further variant with close relationship to subtype 2 (Dumke et al., 2003). A subtype is defined by the specific copies of RepMP2/3 and RepMP4 in the P1 gene and the specific copy of RepMP5 in the ORF6 gene, while a variant is characterized by the exchange of a small number of amino acids in either one of these repetitive DNA sequences. Worldwide, the same two subtypes were isolated most frequently, but the variant 2 of subtype 2 was also only rarely detected in Japan and in Europe
151
(Jacobs et al., 1996; Sasaki et al., 1996; Kenri et al., 1999). To our knowledge, only one strain of the variant 1 was isolated in Denmark recently (Dorigo-Zetsma et al., 2001a). So far, differences between the subtypes and variants with regard to severity of disease, colonization of the respiratory tract of patients or resistance to any antibiotics are not reported. Information on reported outbreaks due to M. pneumoniae goes back to the year 1946. Lind et al. (1997) tested sera of patients collected at the Statens Serum Institute in Copenhagen, Denmark, and found elevated titers of anti-M. pneumoniae antibodies in a regular pattern of epidemics which showed an accumulation every 4.5 years (mean), which were followed by hypoendemic periods. A large epidemic was registered in the winter of 1991 lasting until 1993. It was characterized by an unusual high number of pneumonia cases due to M. pneumoniae infections. Two national influenza surveillance studies included also testing of patients for M. pneumoniae as agent of a respiratory tract disease. The results showed unexpectedly high numbers of M. pneumoniae infections. The overall calculated incidences were 1234 per 100,000 persons in France in the winter 1992/3 decreasing to 190 per 100,000 persons in 1995/6 (Layani-Milon et al., 1999). The data for the outbreak in the Netherlands were in a similar range of 587 per 100,000 for a patient population with respiratory diseases consulting outpatient general practitioners (Dorigo-Zetsma et al., 2001b). After this long lasting major outbreak and the following hypoendemic period a new increase starting in 1995/6 and peaking in 1998 was registered. Mostly because of the lack of experience in cultivating and conserving M. pneumoniae patient isolates, only a few laboratories worldwide were able to follow the outbreaks and to subtype the isolated strains. Analysing 250 isolates collected in Japan, Sasaki et al. (1996) found that the most common subtype in 1976 was subtype 1, but changed in 1979 and 1980 to subtype 2. From 1985 to 1991 almost all isolates were of subtype 1, and as from 1992 subtype 2 was isolated again with increasing tendency. Analysis of our collection of 91 isolates from Germany revealed that the first strains from 1989 were of subtype 2 (Dumke et al., 2003). The outbreak strains in the winter 1992/3 were of subtype 1, followed by the reappearence of subtype 2 in the year 1998 indicating that switching of subtypes is following a global pattern.
The animal model From these epidemiological data one could hypothesize that outbreaks with subtype-specific M. pneumoniae led to an immune response generating protecting
ARTICLE IN PRESS 152
R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Fig. 2. Schematic diagram of the in vivo experiments using guinea pigs as host model and patient isolates of M. pneumoniae belonging to subtype 1, subtype 2 and the variant 2. The combination of strains characterized as subtype 1, subtype 2 or variant 2 were used in a three-component infection mixture and applied to the animals via an intranasal infection to test the efficiency of the three different strains to colonize and propagate in the guinea pig respiratory tract within seven to ten days. To evaluate the percentage of each of the subtypes and the variant 2, hundred colonies of the agarplated infection mixture and of each of the recovered bronchoalveolar lavages were picked and subtyped using PCR. In 13 experiments with different M. pneumoniae threecomponent combinations the percentage of colonies belonging to the subtypes and the variant 2 in the infection mixture and in the corresponding bronchoalveolar lavage were calculated (see Fig. 3). In a further experiment guinea pigs were preinfected intranasally three times within 38 days with a variant 2 strain alone to establish variant 2 immunocompetent animals. After another 10 days the competent animals were infected with the three-component infection mixture and the recovered colonies from the bronchoalveolar lavages were subtyped (see Fig. 4).
antibodies within the human population against the dominating subtype of the epidemic peak. This should give the alternative subtype an advantage over the strains causing the last outbreak. To test the hypothesis, we infected guinea pigs intranasally and analysed whether the different subtypes or the variant 2 were able to colonize the respiratory tract with the same efficiency. In addition, we wanted to learn whether a preinfection with a defined subtype or variant alters its ability to colonize the respiratory tract for a second time and promotes the preferential colonization by other subtypes or variants. For the animal experiments, the patient isolates and the isolates recovered from bronchial fluids of the guinea
Fig. 3. Mean proportions of subtype 1, subtype 2 and variant 2 M. pneumoniae cells in the three-component infection mixture and in the corresponding bronchoalveolar lavages of naı¨ ve guinea pigs. To characterize the efficacy of each strain to colonize and to propagate in the respiratory tract of guinea pigs within seven to ten days after intranasal infection, different three-component infection mixtures with various strain combinations were tested (n ¼ 13).
pigs were typed by various parameters including Western blotting with mono-specific antibodies, PCR and DNA sequencing (for a description of the typing methods, see Dumke et al., 2003). The established procedure for infecting guinea pigs with M. pneumoniae (Jacobs et al., 1988) was modified by intranasal application of a mean dose of 6 107 colony-forming units of M. pneumoniae cells. To show subtype- and variant-specific differences in host colonization and propagation, we infected the animals with a mixture of three M. pneumoniae patient isolates which belonged to the subtype 1, the subtype 2 and the variant 2 of M. pneumoniae (Fig. 2). Each of the 13 animals was infected with a mixture composed of different patient strains, isolated at different time periods, to exclude the dominance of a single strain within the three tested groups. As a reference, parallel to the infection of the guinea pigs, the mixture of M. pneumoniae strains was cultured on agar plates and the colonies were counted and picked for typing with PCR methods (for a description of the typing methods see Dumke et al., 2003). Unexpectedly, under the applied conditions the variant 2 showed the highest survival rate, which, however, was statistically not significantly different from subtype 1, whereas subtype 2 was less fit for survival and reproduction in these animals (Fig. 3). Obviously, the in vivo experiments led to a selection of variant 2- and of subtype 1-specific strains. Since the number of in vitro passages (4–10) was different for each strain used, we had to exclude that the survival or the ability to adhere was influenced by the in vitro passages. Therefore, animals were infected with subtype 2, and in
ARTICLE IN PRESS R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
153
a second step another set of animals were infected with the freshly recovered subtype 2 isolates. This approach did not result in a higher recovery of subtype 2, indicating that the reduced survival rate of subtype 2 in guinea pigs is not correlated with differences in the number of in vitro passages of subtype 2 isolates prior to infection (results not shown). The use of other wild-type (e.g. variant 1) or genetically modified M. pneumoniae strains in these animal model experiments in the future may reveal differences in virulence or host colonization pattern between the strains.
Adaptation of M. pneumoniae to its host Comparing the results from the above-described experiments with guinea pigs with the data from the natural human host, the major difference concerns the dominance of variant 2 in the animal model. This is in contrast to its rare isolation from the human respiratory tract. In humans, subtype 1 and also subtype 2 were detected most frequently in the respiratory material. According to our subtyping results only the subtypes 1 and 2 were responsible for outbreaks of M. pneumoniae infections. The variant 2 was isolated only sporadically and did not seem to be a relevant cause of epidemics. The preference for subtype 1 and 2 in the human respiratory tract and on the other hand the preference for variant 2 and subtype 1 in guinea pigs might be correlated with selection and adaption of M. pneumoniae strains to a specific host. During evolution, M. pneumoniae may have tested host cells with different receptor specificities. The present, constant environment of these bacteria, i.e. the human respiratory tract, is probably optimal for the growth of subtype 1 and 2. This could explain why we and others almost exclusively find subtype 1 and 2 with their conserved P1 and ORF6 genes. Nevertheless, despite the reduced genome size, M. pneumoniae keeps the repetitive DNA sequences (RepMP2/3, RepMP4, RepMP5), which make up to about 8% of the genome and which could be used, at least in theory, by M. pneumoniae to assemble new P1 and ORF6 genes by homologous recombination. We have here a clear contradiction between the results of the subtyping experiments and the genetic potential of M. pneumoniae. This discrepancy could be explained, if we assume that either new P1 genes are inferior to the subtypes 1 or 2 existing ones, and/or that homologous recombination is a very rare event and does not contribute significantly to the assembly of new P1 genes. The sporadic finding of variants without relevance for epidemic outbreaks and the reported inability to transform M. pneumoniae by inserting DNA in its genome via homologous recombination (Dhandayuthapani et al., 1999) support these hypotheses, which can be tested now experimentally.
Fig. 4. Three guinea pigs were preinfected with variant 2 alone (three times during 38 days) to establish a strong immunocompetence versus variant 2. After further ten days these three animals and further three naive animals with no prior M. pneumoniae contact were infected with a three-component infection mixture which consisted of the variant 2 strain (which was used also for preinfection) and a patient isolate of subtype 1 and of subtype 2 each. The proportions of the three strains were calculated for the infection mixture and for the bronchoalveolar lavages of the three variant 2-preinfected and the three naive animals.
Subtype-specific host immune response as important factor for outbreaks due to M. pneumoniae? In in vitro experiments using erythrocytes as host cells and M. pneumoniae reference strains of subtypes 1 and 2, Jacobs et al. (1996) were able to inhibit the adherence of bacteria by adding sera of patients suffering from M. pneumoniae diseases. In these adherence inhibition tests the sera selectively blocked adherence of subtype 1 or 2. Only a few sera inhibited both subtypes. The authors speculated that the differences in the amino acid sequences of the P1-protein were important for optimal adherence to the host but also for a subtype-specific protective immunity. The study by Jacobs et al. (1996) was hampered by the lack of patient strains and therefore patient sera had to be tested with reference strains. Nonetheless, sera taken during one of the outbreaks were able to inhibit mostly the outbreakspecific subtype. To prove the prevalence switches from one to the other subtype/variant in specifically prestimulated animals, we preinfected three guinea pigs with variant 2 and reinfected these animals with a infection mixture consisting of 30% subtype 1, 30% subtype 2 and 40% of the variant 2 (Fig. 4). In parallel to the prestimulated
ARTICLE IN PRESS 154
R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
animals, also three unstimulated guinea pigs were intranasally infected as controls. All six animals were kept separately in ‘‘filtered air box systems’’. The bronchoalveolar lavages of the six animals were collected after ten days and plated for M. pneumoniae growth. In all cases 100 colonies per animal were picked and typed by PCR (for a description of the typing methods, see Dumke et al., 2003). The bronchial fluids of the three unstimulated control animals were dominated by M. pneumoniae variant 2 with about 83% of the recovered colony-forming bacteria. The other colonies were typed as subtype 1 in almost 20% of the colonies, whereas less than 1% of the reisolates belonged to the subtype 2. In animals preinfected with variant 2 we found a major reduction of variant 2 colonies to 5%, whereas subtype 1 made up to 95% of the reisolates, and subtype 2 was again below 1%. The results of the in vivo experiments presented here show for the first time that a variant 2-specific prestimulated animal was able to develop a protective immunity to variant 2. On the other hand the protection was not complete, as under our applied experimental conditions variant 2 was not eradicated from the respiratory tract of guinea pigs. Interestingly, also not all patient sera were able to inhibit the adherence of M. pneumoniae (Jacobs et al., 1996) indicating that the P1 adhesin might be part of a much more complex biological structure with different major and minor cell adherence-associated proteins (Krause, 1996). Because of the low recovery of subtype 2 in the animal model, the closely related variant 2 was used to test the effect of prestimulating the guinea pigs and to provoke a possible reduction of variant 2 in a second infection. The experiment shown in Fig. 4 clearly revealed that the number of variant 2 colonies was drastically reduced and subtype 1 became the predominant strain. Our results indicate that we were able to mimic in the animal model the phenomenon observed in humans, i.e. a switching of the predominant subtype to the other subtype in a following epidemic outbreak. Our results may also explain why – between epidemics – both subtypes continue to circulate in the human population with a much lower number of registered patient cases probably due to the effect of a limited subtype-specific immunity.
Conclusions The knowledge about the interaction of M. pneumoniae and its host is still insufficient despite the improvement in cultivating these fastidiously growing bacteria, diagnosing them in patient respiratory material and the large body of data derived from the annotation of genome sequences. One of the important questions is
the influence of the host on the generation of new variants or subtypes; for instance, how do M. pneumoniae escape the immune defense of the host? The results of all analyses aiming at the identification and characterization of M. pneumoniae patient isolates led to the same conclusion: M. pneumoniae is genetically very homogeneous. There are in principle only two subtypes and a few variants. The latter ones do not play any role in epidemic outbreaks so far. These observations are in sharp contrast to the results of the annotation of the genome sequence of M. pneumoniae M129 (Himmelreich et al., 1996) and M. pneumoniae FH (Reiser et al., unpublished), which clearly predict a reservoir of repetitive DNA sequences, which could be used by the bacterium to assemble a large number of modified P1 or ORF6 genes by recombination processes creating many different subtypes. There are two possible explanations for this phenomenon, either new P1 and ORF6 genes are synthesized, but they are not advantageous in the hostprovided environment and disappear rather quickly, or the required recombinations do not take place. These two hypotheses can now be investigated experimentally by constructing new P1 and ORF6 genes based on the variation of the nucleotide sequences of the repetitive DNA sequence and testing them in guinea pigs for survival, adaptation and preference. To prove the principle of our approach, new ORF6 genes were constructed keeping the P1 gene constant. M. pneumoniae clones carrying the new P1/ORF6 gene combinations survived in vivo passages in guinea pigs (unpublished results). This shows that combining in vitro construction of new M. pneumoniae subtypes with testing them in an animal model is a promising strategy for evaluating the epidemic potential of new subtypes.
Acknowledgement We thank E. Pirkl for excellent technical assistance. The project was supported by the grants Ja 399/-2 and He 399-2 from the Deutsche Forschungsgemeinschaft.
References Cousin-Allery, A., Charron, A., de Barbeyrac, B., Fremy, G., Jensen, J.S., Renaudin, H., Bebear, C., 2000. Molecular typing of Mycoplasma pneumoniae strains by PCR-based methods and pulsed-field gel electrophoresis. Application to French and Danish isolates. Epidemiol. Infect. 124, 103–111. Dandekar, T., Huynen, M., Regula, J.T., Ueberle, B., Zimmermann, C.U., Andrade, M., Doerks, T., SanchezPulido, L., Snel, B., Suyama, M., Yuan, Y.P., Herrmann, R., Bork, P., 2000. Re-annotating the Mycoplasma pneumoniae genome sequence: adding value, function and reading frames. Nucleic Acids Res. 28, 3278–3288.
ARTICLE IN PRESS R. Dumke et al. / International Journal of Medical Microbiology 294 (2004) 149–155
Dhandayuthapani, S., Rasmussen, W.G., Baseman, J.B., 1999. Disruption of gene mg218 of Mycoplasma genitalium through homologous recombination leads to an adherence-deficient phenotype. Proc. Natl. Acad. Sci. USA 96, 5227–5232. Dorigo-Zetsma, J.W., Dankert, J., Zaat, S.A.J., 2000. Genotyping of Mycoplasma pneumoniae clinical isolates reveals eight P1 subtypes within two genomic groups. J. Clin. Microbiol. 38, 965–970. Dorigo-Zetsma, J.W., Wilbrink, B., Dankert, J., Zaat, S.A.J., 2001a. Mycoplasma pneumoniae P1 type 1- and type 2specific sequences within the P1 cytadhesin gene of individual strains. Infect. Immun. 69, 5612–5618. Dorigo-Zetsma, J.W., Wilbrink, B., van der Nat, H., Bartelds, A.I.M., Heijnen, M.L.A., Dankert, J., 2001b. Results of molecular detection of Mycoplasma pneumoniae among patients with acute respiratory infection and in their household contacts reveals children as human reservoirs. J. Infect. Dis. 183, 675–678. Dumke, R., Catrein, I., Pirkl, E., Herrmann, R., Jacobs, E., 2003. Subtyping of Mycoplasma pneumoniae isolates based on extended genome sequencing and on expression profiles. Int. J. Med. Microbiol. 292, 513–525. Eaton, M.D., Meiklejohn, G., VanHerick, W., 1944. Studies on the etiology of primary atypical pneumonia: a filterable agent transmissible to cotton rats, hamsters, and chick embryos. J. Exp. Med. 79, 649–667. Franzoso, G., Hu, P.C., Meloni, G.A., Barile, M.F., 1993. The immunodominant 90-kilodalton protein is localized on the terminal tip structure of Mycoplasma pneumoniae. Infect. Immun. 61, 1523–1530. Himmelreich, R., Hilbert, H., Plagens, H., Pirkl, E., Li, B.C., Herrmann, R., 1996. Complete sequence analysis of the genome of the bacterium Mycoplasma pneumoniae. Nucleic Acids Res. 24, 4420–4449. Jacobs, E., 2002. Mycoplasma pneumoniae disease manifestations and epidemiology. In: Razin, S., Herrmann, R. (Eds.), Molecular Biology and Pathogenicity of Mycoplasmas. Kluwer Academic/Plenum Publishers, New York, pp. 519–530. Jacobs, E., Drews, M., Stuhlert, A., Bu¨ttner, C., Klein, P.J., Kist, M., Bredt, W., 1988. Immunological reactions of guinea pigs following intranasal Mycoplasma pneumoniae infection and immunization with the 168 kDa aherence protein. J. Gen. Microbiol. 134, 473–479. Jacobs, E., Vonski, M., Oberle, K., Opitz, O., Pietsch, K., 1996. Are outbreaks and sporadic respiratory infections by Mycoplasma pneumoniae due to two distinct subtypes? Eur. J. Clin. Microbiol. Infect. Dis. 15, 38–44. Kenri, T., Taniguchi, R., Sasaki, Y., Okazaki, N., Narita, M., Izumikawa, K., Umetsu, M., Sasaki, T., 1999. Identification of a new variable sequence in the P1 cytadhesin gene of Mycoplasma pneumoniae: evidence for the generation of antigenic variation by DNA recombination between repetitive sequences. Infect. Immun. 67, 4557–4562.
155
Krause, D.C., 1996. Mycoplasma pneumoniae cytadherence: unravelling the tie that binds. Mol. Microbiol. 20, 247–253. Krause, D.C., Balish, M.F., 2001. Structure, function, and assembly of the terminal organelle of Mycoplasma pneumoniae. FEMS Microbiol. Lett. 198, 1–7. Krause, D.C., Balish, M.F., 2004. Cellular engineering in a minimal microbe: structure and assembly of the terminal organelle of Mycoplasma pneumoniae. Mol. Microbiol. 51, 917–924. Layani-Milon, M.P., Gras, I., Valette, M., Luciani, J., Stagnara, J., Aymard, M., Lina, B., 1999. Incidence of upper respiratory tract Mycoplasma pneumoniae infections among outpatients in Rhone-Alpes, France, during five successive winter periods. J. Clin. Microbiol. 37, 1721–1726. Layh-Schmitt, G., Herrmann, R., 1994. Spatial arrangement of gene products of the P1 operon in the membrane of Mycoplasma pneumoniae. Infect. Immun. 62, 974–979. Lind, K., Benzon, M.W., Jensen, J.S., Clyde Jr., W.A., 1997. A seroepidemiological study of Mycoplasma pneumoniae infections in Denmark over the 50-year period 1946–1995. Eur. J. Epidemiol. 13, 581–586. Marmion, B.P., 1990. Eaton agent—science and scientific acceptance: a historical commentary. Rev. Infect. Dis. 12, 338–353. Morowitz, H.J., Tourtellotte, M.E., 1962. The smallest living cells. Sci. Am. 206, 117–125. Razin, S., Jacobs, E., 1992. Mycoplasma adhesion. J. Gen. Microbiol. 138, 407–422. Ruland, K., Wenzel, R., Herrmann, R., 1990. Analysis of three different repeated DNA elements present in the P1 operon of Mycoplasma pneumoniae: size, number and distribution on the genome. Nucleic Acids Res. 18, 6311–6317. Sasaki, T., Kenri, T., Okazaki, N., Iseki, M., Yamashita, R., Shintani, M., Sasaki, Y., Yayoshi, M., 1996. Epidemiological study of Mycoplasma pneumoniae infections in Japan based on PCR-restriction fragment length polymorphism of the P1 cytadhesin gene. J. Clin. Microbiol. 34, 447–449. Su, C.J., Chavoya, A., Dallo, S.F., Baseman, J.B., 1990. Sequence divergency of the cytadhesin gene of Mycoplasma pneumoniae. Infect. Immun. 58, 2669–2674. Ueberle, B., Frank, R., Herrmann, R., 2002. The proteome of the bacterium Mycoplasma pneumoniae: comparing predicted open reading frames to identified gene products. Proteomics 2, 754–764. Ursi, D., Heven, M., van Bever, H., Quint, W., Niesters, H.G.M., Goossens, H., 1994. Typing of Mycoplasma pneumoniae by PCR-mediated DNA fingerprinting. J. Clin. Microbiol. 32, 2873–2875. Weiner III, J., Zimmermann, C.U., Go¨hlmann, H.W., Herrmann, R., 2003. Transcription profiles of the bacterium Mycoplasma pneumoniae grown at different temperatures. Nucleic Acids Res. 31, 6306–6320.