The Paracoccidioides Genus and Paracoccidioidomycosis
Paracoccidioidomycosis (PCM) is a deep systemic mycosis caused by human fungal pathogens
of the Paracoccidioides genus. The disease is geographically restricted to subtropical
areas of Latin America (from south of Mexico to north of Argentina) with a high prevalence
in Brazil, Colombia, Venezuela, and Argentina [1]. The annual incidence rate in Brazil
is 10–30 infections per million inhabitants, and the mean mortality rate is 1.4 per
million inhabitants per year, making this disease the highest cause of mortality among
systemic mycoses [2]. PCM is endemic in rural populations and mainly affects individuals
engaged in agricultural activities, who inhale aerosols containing fungal material
during manipulation of the soil.
Molecular evolutionary studies place the genus Paracoccidioides in the thermodimorphic
fungal pathogen clade related to the family Ajellomycetaceae (Ascomycetes), which
includes the Blastomyces, Histoplasma, and Emmonsia genera, and with which it shares
a common ancestor, Lacazia loboi. PCM can be caused by two species Paracoccidioides
brasiliensis and P. lutzii
[3]. P. brasiliensis has been considered a single species since its discovery, although
several studies including molecular and morphological data support the split of P.
brasiliensis into two species [3], [4]. P. lutzii is composed of a single monophyletic
and recombining population so far found in central, southwest, and north Brazil and
Ecuador [3]–[5]. On the other hand, P. brasiliensis contains a complex of at least
four different cryptic species (S1, PS2, PS3 and PS4; Figure 1A
[6]). P. brasiliensis S1 represents a monophyletic and recombining population widely
distributed in South America and has been associated with the majority of cases of
PCM detected up until the present time. Strains belonging to P. brasiliensis S1 have
previously been recovered from armadillos, soil, and penguin feces [6]. P. brasiliensis
PS2 is a paraphyletic and recombining population identified so far only in Brazil
and Venezuela [6]. P. brasiliensis PS3 is comprised of a monophyletic and clonal population
that has been recovered in humans and armadillos in endemic regions of Colombia [6].
P. brasiliensis PS4 was recently identified and is composed of a monophyletic population
of clinical isolates from Venezuela [5], [7]. Besides the typical bicorn cocked hat–
and barrel-shaped conidia produced by both species, P. lutzii frequently produces
elongated rod-shaped conidia, a characteristic feature that may be used for species
identification [3]. Because of the difficulties of conidia production in the laboratory
and slight morphological differences among species, molecular identification of Paracoccidioides
species has become the most common tool of choice. Several molecular markers have
already been applied in population studies of the Pararacoccidioides genus, and for
multilocus sequencing typing, gp43, arf, β-tub, and hsp70 loci are the best choices
for species delineation [4], [6].
10.1371/journal.ppat.1004397.g001
Figure 1
(A) Median-joining haplotypic network distribution of the Paracoccidioides genus and
L. loboi based on the gp43 marker. The size of the circumference is proportional to
the haplotype frequency, and colors vary according to the sampling location of each
haplotype. Red dots (median vectors) are hypothetical missing intermediates, and black
dots represent each mutation site. (B) Schematic representation of serum/antigen compatibility
among patients and isolates from Mato Grosso and São Paulo used in serological tests.
The illustration shows the low immunogenic specificity when serum and antigenic pool
from different states are reacted [41].
Phylogeography of the Paracoccidioides Genus
The finding of the cryptic species in the genus Paracoccidioides has led us to explore
the evolutionary mechanisms that were responsible for the current geographic distribution
of its five phylogenetic species (S1, PS2, PS3, PS4, and P. lutzii). Phylogeographic
inferences from three different loci revealed simultaneous geographic expansions of
S1 isolates. This represents a dispersal by distance, leading to a nonsexual population
(PS3) that does not produce any gene flow between other species, and a long-distance
colonization or a fragmentation resulting in the separation of Venezuelan PS4 species
[3]–[5], [7]. The dispersal event that resulted in PS3 has been confirmed, with the
remaining divergence processes due to vicariance events. Despite the great stability
of the Guiana and Brazilian shields, the uplift of the Andes and episodic marine inclusions
61 million years ago, 20 million years ago, and 11.8–7.0 million years ago may have
favored vicariance between Paracoccidioides and Lacazia. This may have occurred by
creating new and empty ecological niches represented by wetlands and/or totally submerged
areas, as well as by the simultaneous emergence of riverine Cetacean mammals. The
recent dispersal of PS3 to Colombia may also be explained by the complete submersion
of Colombian territory by the Pebas/Solimões lake, derived from marine incursions
in the late Miocene era [8] and indicating a recent occupation. As there are no clear
geographic barriers, the most challenging task is to explain the speciation processes
that have given rise to S1, PS2, and P. lutzii in the very stable Brazilian shield.
Indeed, the prevalence of P. lutzii in central-western Brazil and its relatively close
proximity to the S1 and PS2 occurrence areas suggests a parapatric speciation. With
regards to the divergence between S1 and PS2 and the current sympatry observed between
them, probable differences in preferences for substrates and resources in their saprobe
lifestyle that might have triggered a disruptive selection should be considered [5].
However, more studies are required to complete the biogeographical puzzle of the Paracoccidioides
genus. These include mapping of the cryptic species in environmental, rather than
clinical, samples, as well as searching for saprobe differences between sympatric
species.
Ecology of P. brasiliensis and P. lutzii
Onygenalean (Ascomycota) organisms have typically evolved by adapting to two distinct
ecological niches, the first represented by saprobic conditions in soil and the second
by the live tissues of animal hosts. Genomic adaptations such as loss of carbohydrate-degrading
enzymes, gain of proteases, and the ability to produce infective conidia allowing
long association with the mammalian host are more adapted to a biotrophic lifestyle
[9], [10]. Epidemiological evidence indicates that the saprobic forms of P. brasiliensis
and P. lutzii may occur in some restricted and/or protected soil conditions, in places
containing natural and anthropic disturbed vegetation near water sources [11]. Isolation
of P. brasiliensis directly from its saprobic form has proved to be difficult. However,
the fungus has been repeatedly cultured from the armadillo species Dasypus novemcinctus
and Cabassous centralis in endemic PCM areas [12], [13] and, in unique cases, from
dogs and two-toed sloths [14], [15]. Additional evidence of the infection of several
wild and domestic animals has also been provided by intradermal, serological, histopathological,
and molecular tests, revealing a broad distribution and adaptation to mammalian hosts
[16].
Although outbreaks of PCM have not been documented, the geographical distribution
of the disease is heterogeneous and is associated with moderate-to-high precipitation
rates, mild temperatures, and fertile soils. The disease occurs in areas such as the
central-western and northern regions of Brazil where agricultural activities are more
commonly employed. Climatic anomalies, such as those triggered by the 1982–1983 El
Niño event, have been associated with an excess of acute PCM cases when compared with
the number of expected cases for the same period. This indicates the presence of a
temporal cluster of the disease in the state of São Paulo, Brazil, occurring in the
year 1985 [17]. Climatic conditions resulting in an atypical increase in soil water
storage in 1982–1983 and in an increase in the absolute air humidity in 1984 may have
contributed firstly to fungus growth and then to conidial dispersal. This evidently
follows the “grow and blow” model already proposed for coccidioidomycosis outbreaks
[18]. Experimental studies have indicated that the several genetic groups or cryptic
species of P. brasiliensis have different abilities in producing the infective conidia,
and this may in turn produce differential rates of infection. For example, the S1
and PS2 sympatric cryptic species of P. brasiliensis occur at a disproportional rate
of approximately 9∶1 in both patient and armadillo isolates. At the same time, isolates
of the S1 genotype produce many more conidia than PS2 isolates [5].
Reproductive Modes in the Paracoccidioides Species Complex
Paracoccidioides was considered an asexual and clonal microorganism for many years
[1]. The anamorph of Paracoccidioides is characterized by multiple budding yeast cells
that grow at 37°C in mammalian tissues or by mycelia that produce chlamydospores or
conidia at 25°C in the environment. Recently, population genetics and comparative
genomic studies have provided evidence for different breeding strategies in the Paracoccidioides
genus. Recombination events were detected in both P. lutzii and P. brasiliensis (S1
and PS2), and the P. brasiliensis PS3 population was considered clonal [4], [6]. The
mating type locus was identified in the three sequenced genomes, and a single copy
of MAT1-1 or MAT1-2 was found, thus suggesting a bipolar mating system [9]. According
to Torres et al. [19], MAT gene distribution was evaluated with regards to the country
of origin and phylogenetic species, revealing a 1∶1 ratio of MAT1-1 and MAT1-2. Additionally,
Teixeira et al. [20] tested the MAT gene distribution in 98 Paracoccidioides isolates
and revealed a slight (2-fold) prevalence of the MAT1-2 idiomorph. Unexpectedly, both
the MAT1-1 and MAT1-2 genes were identified in 13 of the clinical isolates, suggesting
that homothallism may exist in the Paracoccidioides genus.
Orthologs of mating and meiotic regulators that have been well characterized in a
wide range of fungi were found by comparative genomics to be highly conserved in Paracoccidioides
and other Ajellomycetacean sexual fungi [9], [20]. The biological functionality of
α-pheromone and its receptor was elucidated using heterologous expression of these
Paracoccidioides genes in the corresponding Saccharomyces cerevisiae null mutants
[21]. Features related to sexual reproduction, such as coiled constricted hyphae and
knob-like structures, were observed in Paracoccidioides species, indicating the formation
of young ascocarps. In addition, multiple nuclei were found in coiled constricted
hyphae, possibly as a consequence of nuclear migration during mating. Unfortunately,
no cleistothecium or ascus production has so far been detected. The presence and expression
of sexual machinery as well the ability to produce sex-related structures indicates
that mating may occur in the Paracoccidioides life cycle.
Virulence Factors Associated with Dimorphism and Host Adaptation
The most probable environmental habitat of mycelial-phase Paracoccidioides is the
soil. Once conidia or mycelia fragments are inhaled into the lung alveoli, the fungus
shifts its morphology to a yeast phase because of temperature, hormones, and immune
response. This step is crucial for the survival and maintenance of Paracoccidioides
and other dimorphic fungi in hosts. The dimorphic transition promotes changes in the
cell wall composition and carbohydrate polymer structure. Additionally, the presence
of an outermost layer of α-1,3-glucan in the P. brasiliensis yeast cell wall has been
proposed as a protective shield against host defense [22], [23].
The yeast transcriptional profile indicating the diversion of pyruvate from the glycolytic
pathway into the glyoxylate cycle is consistent with a lower oxygen level in infected
tissues [24]. Stress adaptations that induce genes encoding molecular chaperones,
such as heat shock proteins (HSPs), are a common trait of Paracoccidioides during
dimorphism and exposition to different host niches. These adaptations may be indispensable
for fungal virulence upon infection [25]. In contrast, some HSP genes are down-regulated
when mycelia cells are incubated in vitro with estradiol or human female serum, in
some part explaining the predominance of the disease in adult males [26]. The ability
of fungal cells to adhere to host cells is also vital for the initial steps of the
infection process. Phospholipase B, involved in the fungus-macrophage interaction,
and PbHad32p, a hydrolase involved in adherence to host cells, have both been proposed
to be important for the initial steps of the virulence process [27], [28]. The genes
encoding glyceraldehyde-3-phosphate-dehydrogenase (GAPDH), identified as a cell wall–associated
host-adhesion molecule in Paracoccidioides, and enolase, which binds to fibronectin,
have also been linked to the initial adhesion of the fungus to lung epithelial cells
and alveolar macrophages upon infection [29]. New strategies for gene function studies
using antisense RNA technology have recently emerged, overcoming problems with gene
knockout due to multinucleated cells and unstable transformants [30]. The depletion
of P. brasiliensis–encoding genes Cdc42 [31], AOX [32], Gp43 [33], SCONC [34], HSP90
[35], P27 [36], and Rbt5 [37] contributed significantly to the elucidation of host-pathogen
interaction, pathogen resistance, and virulence in those species.
In years to come, more information on virulence processes will emerge from the accumulated
data from transcriptome and complete genome releases at our disposal. This will pave
the way for the identification of possible mechanisms to control the initial steps
of infection which, when available to clinicians, will benefit patients.
Paracoccidioides Species Complex and Its Impact on Clinical and Serological Aspects
of PCM
Since the discovery of the cryptic speciation in the genus Paracoccidioides, some
important regional features of the disease were discussed regarding its impact on
the current statement of PCM. The acute or subacute ( juvenile PCM) and chronic (adult
PCM) are the two main forms of the disease; however, the presentation and course of
the disease may vary from case to case [16]. The first discrimination between the
PCM pathology related to geographical origin was observed by Barbosa et al. [38],
in which patients from the central region of Brazil had predominantly lymphoabdominal
forms not shared by patients from the south and southeast, a finding later confirmed
by Andrade [39]. Is there a possibility that lymphoabdominal forms are associated
with the pathology caused by P. lutzii and not by P. brasiliensis? Are there different
pathologies caused by different species of Paracoccidioides? This possibility should
not be ruled out and must be fully investigated in order to verify possible associations
of cryptic species and different clinical manifestations of PCM. In addition to clinical
manifestation, issues addressed to treatment have been raised by Hahn et al. [40]
who found that patients infected with P. lutzii had good responses to trimethoprim-sulfamethoxazole
while those infected with P. brasiliensis relapsed with the same drug administration.
Moreover, there are known to be a high number of patients coming from the north-central
region of Brazil with PCM with low or no immunoreactivity [41], [42]. Immunodiffusion
tests with antigens produced by isolates from São Paulo (P. brasiliensis strain 339)
crossed with sera from patients of Mato Grosso have low positivity (Figure 1B). Recently,
serological tests confirm that sera from patients with PCM due to P. lutzii are able
to recognize cell-free antigens from P. lutzii; however, sera from patients with PCM
due to P. brasiliensis could not recognize any P. lutzii antigens [43]. Undoubtedly,
these issues are critical and need an urgent mobilization to improve the methods of
diagnosis and therapy that can specifically detect and effectively combat the Paracoccidioides
species of a given patient with PCM.