Genomic Insights into the Fungal Pathogens of the Genus Pneumocystis: Obligate Biotrophs of Humans and Other Mammals
Citation: Hauser PM (2014) Genomic Insights into the Fungal Pathogens of the
Genus Pneumocystis: Obligate Biotrophs of Humans and Other Mammals. PLoS
Pathog 10(11): e1004425. doi:10.1371/journal.ppat.1004425
Genomic Insights into the Fungal Pathogens of the Genus Pneumocystis : Obligate Biotrophs of Humans and Other Mammals
Philippe M. Hauser 0
Joseph Heitman, Duke University Medical Center, United States of
0 Institute of Microbiology, Centre Hospitalier Universitaire Vaudois and University of Lausanne , Lausanne , Switzerland
Pneumocystis organisms were first believed to be a single protozoan species able to colonize the lungs of all mammals. Subsequently, genetic analyses revealed their affiliation to the fungal Taphrinomycotina subphylum of the Ascomycota, a clade which encompasses plant pathogens and free-living yeasts. It also appeared that, despite their similar morphological appearance, these fungi constitute a family of relatively divergent species, each with a strict specificity for a unique mammalian species . The species colonizing human lungs, Pneumocystis jirovecii, can turn into an opportunistic pathogen that causes Pneumocystis pneumonia (PCP) in immunocompromised individuals, a disease which may be fatal. Although the incidence of PCP diminished in the 1990s thanks to prophylaxis and antiretroviral therapy, PCP is nowadays the second-most-frequent, life-threatening, invasive fungal infection worldwide, with an estimated number of cases per year above 400,000 . Despite this clinical importance, studies of P. jirovecii progressed slowly, at least in part because of the lack of a continuous culture system. Nevertheless, recent genomic findings provided insights into the lifestyle of these fungi.
Despite the absence of culture, the nuclear genome sequences of
three Pneumocystis spp. have been recently released to the public.
That of P. carinii, the species infecting rats, was obtained from
resected lungs of an infected rat and purification of P. carinii cells
(pgp.cchmc.org). This provided sufficient amounts of relatively
pure DNA for conventional cloning and sequencing of
chromosomes separated on gel, as well as for high throughput DNA
sequencings (HTS). The nuclear genome sequence of P. murina,
the species infecting mice, was obtained from resected lungs
followed by HTS (Pneumocystis murina Sequencing Project,
Broad Institute of Harvard and the Massachusetts Institute of
Technology, http://www.broadinstitute.org/). Thanks to cell
immunoprecipitation and whole genome amplification, the P.
jirovecii nuclear genome sequence was obtained using HTS from a
single bronchoalveolar lavage fluid sample of a single patient with
PCP . Because of the low DNA purity, the assembly
necessitated an innovative approach using iterative identification
of the fungal homologies among the reads from the host and lung
microbiota. The genome sequences of the mitochondria were also
obtained from the same sequencings, as well as by PCR [4,5].
The nuclear genome assembly of P. carinii is still highly
fragmented, whereas that of P. murina is made of 17 contigs likely
to correspond to the 17 chromosomes composing this genome
(Table 1). The P. jirovecii nuclear assembly presents an
intermediary fragmentation, which results from the difficulty of assembling
it out of a mixture of reads. Because of their repetitive nature,
telomeres were not assembled. The nuclear genomes of the three
Pneumocystis spp. are approximately 8 Mb.
The mitochondrial genomes of P. murina and P. carinii are
approximately 25 kb, whereas that of P. jirovecii is around 35 kb
(Table 1). This size difference is due to a supplementary region that is
non-coding and highly variable in size and sequence among isolates
. The circular structure of the P. jirovecii mitochondrial genome
differs from the linear one present in the two other Pneumocystis spp.
The biological significance of this difference remains unknown.
Using gene models specifically developed, the three
Pneumocystis spp. nuclear genomes were predicted to encode
approximately 3,600 protein coding genes (Table 1). Mapping of the
genes onto the chromosomes is achieved for P. murina because
contigs correspond to chromosomes, as well as for P. carinii
because isolated chromosomes were sequenced, but not for P.
jirovecii. Functional annotation was optimized by the use of
transcription data as well as of carefully chosen fungal proteomes
as intermediary data for mapping onto the Kyoto Encyclopedia of
Genes and Genomes (KEGG) atlas of biochemical pathways [3,6].
About 30%40% of the genes were reported to encode
hypothetical proteins without significant homology with the
databases, but this proportion is decreasing as new fungal genome
sequences are released. A maximum likelihood phylogeny from
the alignment of 458 concatenated orthologs revealed Taphrina
deformans and the members of the Schizosaccharomyces genus as
the closest relatives of Pneumocystis spp. [3,7]. The genome
content of the three Pneumocystis spp. covered most of the
biochemical pathways corresponding to the basal metabolism and
standard cellular processes [6,8]. Specific features included the
presence of a single operon encoding the ribosomal RNA
(Table 1), such as T. deformans , which contrasts with the tens
or hundreds in other fungi, and the lack of common fungal
Assembly size [Mb]b
Mean GC content [%]
Protein coding genesc
Coding regions [%]
Assembly size [kb]
Protein coding genes
Orfs (unknown function)
virulence factors such as the glyoxylate cycle and polyketide
synthase clusters [3,6,8]. The mitochondrial genomes of the three
Pneumocystis spp. encode 15 to 17 proteins (Table 1).
Genomic Insights by Comparative Genomics
Relevant features of the P. carinii genome were compared to
those of the free-living yeast S. cerevisiae and of the extreme fungal
obligate parasite Encephalitozoon cuniculi . Intermediate values
of gene number, genome size, and mean intergenic space
suggested that P. carinii might be in the process of becoming
dependent on its host. The use of Schizosaccharomyces pombe
genome sequence as a control for genomic annotation revealed the
absence of most of the enzymes specifically dedicated to the
synthesis of amino acids in Pneumocystis spp. [3,6]. Whole genome
comparison to other representatives of the Taphrinomycotina
subphylum revealed several supplementary gene losses in P.
carinii and P. jirovecii (P. murina could not be analyzed because
its genome sequence was released prior to publication under
specific terms). The hypothetical gene repertoires of ancestors were
reconstructed using maximum parsimony and the irreversible
Dollo model of evolution . The approach identified
approximately 2,000 genes presumably lost by the common ancestor of
Pneumocystis and Taphrina genera during its evolution towards
the Pneumocystis genus. Analysis of these genes revealed losses of
genes that impair the biosynthesis of thiamine, the assimilation of
inorganic nitrogen and sulfur, and the catabolism of purines. In
addition, lytic proteases, which are believed to be crucial to fungal
virulence, were underrepresented. The absence of the genes was
ascertained by extensive gene searches in partially overlapping
data sets expected to cover 100% of the genomes, and by the fact
that it was observed in both P. carinii and P. jirovecii.
Nevertheless, it is important to keep in mind that we cannot
firmly exclude the presence of genes of a previously unknown
origin that have not been observed in other organisms so far, and
thus would be undetectable because they are absent from the
databases. These gene losses constitute an important signature of
the lifestyle of Pneumocystis spp.
The loss of biosynthetic pathways of essential molecules is a
hallmark of obligate parasitism in both eukaryotic and prokaryotic
organisms [11,12]. These losses are believed to be allowed by the
availability of the end products of the pathways within the host
environment. Similarly, the absence of substrate may explain the
loss of assimilation pathways. The loss of amino acids and thiamine
biosyntheses in Pneumocystis spp. strongly suggests that they are
obligate parasites. Thus, their entire cycle probably takes place
within the host lungs, and no free-living form of these fungi would
exist. In addition to amino acids and thiamine, Pneumocystis spp.
are believed to scavenge cholesterol from their host to build their
own membranes . In bacteria, gene losses associated with
obligate parasitism imply a reduction of the genome size linked to a
reduction of the guanine-cytosine (GC) content . Likewise,
Pneumocystis spp. genomes have smaller genome size and GC
content than their free-living and facultative parasite relatives,
respectively Schizosaccharomyces spp. and T. deformans (Table 2).
Two categories of parasites are recognized: biotrophs, which
secrete low amounts of lytic enzymes and obtain food from living
amino acids synthesis
amino acids synthesis
nitrogen+sulfur assimilation 10
amino acids synthesis
amino acids synthesis
amino acids synthesis
host cells, and necrotrophs, which secrete many degrading
enzymes and toxins and obtain food from dead host cells .
The missing pathways of Pneumocystis spp. are shown in Table 2
together with those of selected microorganisms with various
lifestyles. The requirement in thiamine and the absence of
inorganic nitrogen and sulfur assimilation are hallmarks of obligate
plant biotrophs . Several other biological characteristics of
Pneumocystis spp. are also hallmarks of obligate biotrophs: (i) the
absence of destruction of host cells during colonization as well as
during pathogenic infection, (ii) the lack of known virulence
factors, (iii) a sex life cycle occurring within the host, and (iv) the
difficulty to be cultured in vitro so far [8,16,17]. On the other
hand, the loss of the catabolism of purines and of the amino acids
syntheses revealed in Pneumocystis spp. has not been observed in
fungal biotrophs so far. The first loss might be specific to
Pneumocystis spp., but the second is a hallmark of organisms
feeding on other animals, such as obligate parasites of humans 
(Table 2). This latter loss might be related to the adaptation to
animal hosts and suggests that there are more proteins available in
animal than plant hosts. Thus, the genomic and biological
characteristics of Pneumocystis spp. suggest the working hypothesis
that they are obligate biotrophic parasites of mammals. This
hypothesis has been already proposed previously on the basis of
the analysis of P. carinii transcriptome , as well as of the
biological characteristics of Pneumocystis spp. . Experiments are
required in order to test this hypothesis. In particular, the
computational predictions supporting obligate biotrophy need to
be verified. Pneumocystis spp. would be the first fungal biotrophs
of animals recognized so far. Of note, the lifestyle of P. jirovecii
differs from the other fungal obligate parasites of humans,
Malassezia and Candida spp., which are obligate commensals
and opportunistic necrotrophs.
The reduction of genome size in Pneumocystis spp. contrasts
with the increase of this parameter in some fungal obligate plant
biotrophs (Table 2). This latter increase corresponds to a
proliferation of retrotransposons, which may create genetic
variability and diversity including panels of effectors involved in
virulence . As in Ustilago maydis (Table 2), the genome size
reduction in Pneumocystis spp. might be compensated by the
genetic diversity generated by sexuality, suggesting that they might
be obligate sexual organisms. This hypothesis would fit the fact
that the asci issued from the sexual cycle might be the unique
airborne particles responsible for the transmission of the fungus
Obligate parasitism of P. jirovecii has important implications in
the management of immunocompromised patients susceptible to
PCP. Indeed, the fungus is most probably restricted to the lungs of
humans so that the only sources of the infection are patients
with PCP and colonized humans, such as infants experiencing
primo-infection and transitory carriers, as well as, possibly,
pregnant women and elderly people .
I thank Michel Monod, Joseph Kovacs, Liang Ma, Ousmane Cisse, and
Melanie Cushion for critical reading of the manuscript and fruitful
1. Cushion MT , Stringer JR ( 2010 ) Stealth and opportunism: alternative lifestyles of species in the fungal genus Pneumocystis . Annu Rev Microbiol 64 : 431 - 452 .
2. Brown GD , Denning DW , Gow NAR , Levitz SM , Netea MG , et al. ( 2012 ) Hidden killers: Human fungal infections . Sci Transl Med 4 : 165rv13 .
3. Cisse OH, Pagni M , Hauser PM ( 2012 ) De novo assembly of the Pneumocystis jirovecii genome from a single bronchoalveolar lavage fluid specimen from a patient . MBio 4 : e00428 - 00412 .
4. Sesterhenn TM , Slaven BE , Keely SP , Smulian AG , Lang BF , et al. ( 2010 ) Sequence and structure of the linear mitochondrial genome of Pneumocystis carinii . Mol Genet Genomics 283 : 63 - 72 .
5. Ma L , Huang DW , Cuomo CA , Sykes S , Fantoni G , et al. ( 2013 ) Sequencing and characterization of the complete mitochondrial genomes of three Pneumocystis species provide new insights into divergence between human and rodent Pneumocystis . FASEB J 27 : 1962 - 1972 .
6. Hauser PM , Burdet FX , Cisse OH , Keller L , Taffe P , et al. ( 2010 ) Comparative genomics suggests that the fungal pathogen Pneumocystis is an obligate parasite scavenging amino acids from its host's lungs . PLoS ONE 5 : e15152 .
7. Cisse OH, Almeida JM , Fonseca A , Kumar AA , Salojarvi J , et al. ( 2013 ) Genome sequencing of the plant pathogen Taphrina deformans, the causal agent of peach leaf curl . MBio 4 : e00055 - 00013 .
8. Cushion MT , Smulian AG , Slaven BE , Sesterhenn T , Arnold J , et al. ( 2007 ) Transcriptome of Pneumocystis carinii during fulminate infection: carbohydrate metabolism and the concept of a compatible parasite . PLoS ONE 2 : e423 .
9. Cushion MT ( 2004 ) Comparative genomics of Pneumocystis carinii with other protists: implications for life style . J Eukaryot Microbiol 51 : 30 - 37 .
10. Cisse OH , Pagni M , Hauser PM ( 2014 ) Comparative genomics suggests that the human pathogenic fungus Pneumocystis jirovecii acquired obligate biotrophy through gene loss . Genome Biol Evol 6 . In Press.
11. Andersson JO , Andersson SG ( 1999 ) Insights into the evolutionary process of genome degradation . Curr Opin Genet Dev 9 : 664 - 671 .
12. Sakharkar KR , Dhar PK , Chow VT ( 2004 ) Genome reduction in prokaryotic obligatory intracellular parasites of humans: a comparative analysis . Int J Syst Evol Microbiol 54 : 1937 - 1941 .
13. Joffrion TM , Cushion MT ( 2010 ) Sterol biosynthesis and sterol uptake in the fungal pathogen Pneumocystis carinii . FEMS Microbiol Lett 311 : 1 - 9 .
14. Merhej V , Royer-Carenzi M , Pontarotti P , Raoult D ( 2009 ) Massive comparative genomic analysis reveals convergent evolution of specialized bacteria . Biol Direct 4 : 13 .
15. Kemen E , Jones JD ( 2012 ) Obligate biotroph parasitism: can we link genomes to lifestyles? Trends Plant Sci 17 : 448 - 457 .
16. Spanu PD ( 2012 ) The genomics of obligate (and nonobligate) biotrophs . Annu Rev Phytopathol 50 : 91 - 109 .
17. O'Connell RJ , Panstruga R ( 2006 ) Tete a` tete inside a plant cell: establishing compatibility between plants and biotrophic fungi and oomycetes . New Phytol 171 : 699 - 718 .
18. Payne SH , Loomis WF ( 2006 ) Retention and loss of amino acid biosynthetic pathways based on analysis of whole-genome sequences . Eukaryot Cell 5 : 272 - 276 .
19. Cushion MT , Linke MJ , Ashbaugh A , et al. ( 2010 ) Echinocandin treatment of Pneumocystis pneumonia in rodent models depletes cysts leaving trophic burdens that cannot transmit the infection . PLoS ONE 5 : e8524 .
20. Morris A , Norris KA ( 2012 ) Colonization by Pneumocystis jirovecii and its role in disease . Clin Microbiol Rev 25 : 297 - 317 .