• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of plosgenPLoS GeneticsSubmit to PLoSGet E-mail AlertsContact UsPublic Library of Science (PLoS)View this Article
PLoS Genet. Apr 2008; 4(4): e1000046.
Published online Apr 11, 2008. doi:  10.1371/journal.pgen.1000046
PMCID: PMC2289846

Genomic Islands in the Pathogenic Filamentous Fungus Aspergillus fumigatus

Paul M. Richardson, Editor

Abstract

We present the genome sequences of a new clinical isolate of the important human pathogen, Aspergillus fumigatus, A1163, and two closely related but rarely pathogenic species, Neosartorya fischeri NRRL181 and Aspergillus clavatus NRRL1. Comparative genomic analysis of A1163 with the recently sequenced A. fumigatus isolate Af293 has identified core, variable and up to 2% unique genes in each genome. While the core genes are 99.8% identical at the nucleotide level, identity for variable genes can be as low 40%. The most divergent loci appear to contain heterokaryon incompatibility (het) genes associated with fungal programmed cell death such as developmental regulator rosA. Cross-species comparison has revealed that 8.5%, 13.5% and 12.6%, respectively, of A. fumigatus, N. fischeri and A. clavatus genes are species-specific. These genes are significantly smaller in size than core genes, contain fewer exons and exhibit a subtelomeric bias. Most of them cluster together in 13 chromosomal islands, which are enriched for pseudogenes, transposons and other repetitive elements. At least 20% of A. fumigatus-specific genes appear to be functional and involved in carbohydrate and chitin catabolism, transport, detoxification, secondary metabolism and other functions that may facilitate the adaptation to heterogeneous environments such as soil or a mammalian host. Contrary to what was suggested previously, their origin cannot be attributed to horizontal gene transfer (HGT), but instead is likely to involve duplication, diversification and differential gene loss (DDL). The role of duplication in the origin of lineage-specific genes is further underlined by the discovery of genomic islands that seem to function as designated “gene dumps” and, perhaps, simultaneously, as “gene factories”.

Author Summary

Aspergillus is an extremely diverse genus of filamentous ascomycetous fungi (molds) found ubiquitously in soil and decomposing vegetation. Being supreme opportunists, aspergilli have adapted to overcome various chemical, physical, and biological stresses found in heterogeneous environments. While most species in the genus are saprophytes, a surprising number are able to infect wounded plants and animals. Remarkably, the allergic human host also responds abnormally to the aspergilli with lung and sinus disease. The advent of immunosuppressive agents and other medical advances have created a large worldwide pool of human hosts susceptible to some Aspergillus species, including the world's most harmful mold and the causative agent of invasive aspergillosis, Aspergillus fumigatus. In this study, we have used the power of comparative genomics to gain insight into genetic mechanisms that may contribute to the metabolic versatility and pathogenicity of this important human pathogen. Comparison of the genomes of two A. fumigatus clinical isolates and two closely related, but rarely pathogenic species showed that their genomes contain several large isolate- and species-specific chromosomal islands. The metabolic capabilities encoded by these highly labile regions are likely to contribute to their rapid adaptation to heterogeneous environments such as soil or a living host.

Introduction

Aspergillus fumigatus is exceptional amongst the aspergilli in being both a primary and opportunistic pathogen as well as a major allergen associated with severe asthma and sinusitis [1][3]. It was first reported to cause opportunistic invasive infection about 50 years ago [4]. In immunocompromised patients, mycelial growth can proliferate throughout pulmonary or other tissues causing invasive aspergillosis. For these patients, the incidence of invasive aspergillosis can be as high as 50% and the mortality rate is often 50%, even with antifungal treatment. Since the late 1800's [2], A. fumigatus has been demonstrated to be a primary pathogen of the airways, sinuses, lungs, damaged skin and subcutaneous tissues. For example, it can cause post-operative infection in all human organs [5]. In most cases diagnosis remains problematic and can compromise effective medical treatment.

A. fumigatus is thought to possess particular metabolic capabilities and genetic determinants that allow it to initiate and establish an in vivo infection. This conclusion is supported by the observation that the majority of invasive aspergillosis disease is caused by A. fumigatus, even though its conidia comprise only a small percentage of the total conidia found in air-sampling studies [6]. While the interaction of A. fumigatus spores with the human respiratory mucosa is understood to an extent, the basic biology of the organism has until recently received little attention.

Recently we presented the genomic sequence of A. fumigatus strain Af293 (FGSC A1100) [7] isolated from a neutropenic patient, who died from invasive aspergillosis [8]. Its comparison with the genomes of two distantly related species, Aspergillus nidulans and Aspergillus oryzae, has led to many unexpected discoveries, including the possibility of a hidden sexual cycle in A. fumigatus and A. oryzae, and the detection of remarkable genetic variability of this genus [9],[10]. Although members of the same genus, these three species are approximately as evolutionarily distant from each other at the molecular level as humans and fish (Figures 1 and and2)2) [11]. This significant phylogenetic distance has hindered some aspects of comparative genomic analysis of the aspergilli such as identification of the genetic traits responsible for differences in virulence as well as in sexual and physiological properties.

Figure 1
Molecular Divergence in Molds and Yeasts.
Figure 2
Three Closely Related Aspergilli.

To maximize the resolving power of whole-genome comparative analysis, we selected the environmental type strains of a very closely related sexual species, Neosartorya fischeri NRRL181 (A. fischerianus), and a more distantly related asexual species, A. clavatus NRRL1, for complete sequencing. These three species are referred to here as the Affc lineage for A. fumigatus, N. fischeri, and A. clavatus (Figure 2). In contrast to A. fumigatus, N. fischeri is only rarely identified as a human pathogen [12][15]; while A. clavatus is probably an important allergen and the causative agent of extrinsic allergic alveolitis known as malt worker's lung [16]. A. clavatus also produces a number of mycotoxins and has been associated with neurotoxicosis in sheep and cattle fed infected grain worldwide (e.g. [17]). Our phenotypic characterization (Table S1) has shown that both A. fumigatus and N. fischeri can grow at 42°C, which indicates that A. fumigatus may possess other genetic determinants besides thermotolerance that allow it to establish a successful in vivo infection.

As determined by multilocus sequence comparison, most A. fumigatus isolates, including Af293 and A1163, lie within the main A. fumigatus clade and persist as a single, global phylogenetic population, presumably due to its small spore size [18]. Natural A. fumigatus isolates were described previously as having low genetic diversity in comparison to N. fischeri isolates [19]. However recent studies identified a number of strain-specific [7] and polymorphic [20],[21] genes. To further explore the extent of genetic variation within the A. fumigatus species, we included in this analysis the genome sequence of a second strain, A1163, made available through Merck & Co., Inc., Whitehouse Station, NJ. Our preliminary analysis has shown that Af293 and A1163 isolates vary greatly in their resistance to antifungals (Table S2).

Results/Discussion

A. fumigatus Af293 vs. A. fumigatus A1163

The genome of A. fumigatus strain A1163 was sequenced by the whole genome random sequencing method [22]. Its genome (29.2 Mb) is 1.4% larger than the genome of the first sequenced strain Af293 (28.8 Mb) (Table 1). About 98% of each genome can be aligned with high confidence. Alignment of the A1163 genome against the eight Af293 chromosomes has revealed 17 large syntenic blocks, which correspond roughly to the 16 Af293 chromosomal arms (Figure 3). The syntenic blocks were defined as regions containing at least five syntenic orthologs separated by no more than 20 genes without orthologs.

Figure 3
Alignment of the A1163, N. fischeri, and A. clavatus Assemblies against the Eight Af293 Chromosomes.
Table 1
Genome Statistics

Most translocation events involving A. fumigatus chromosomes appear to have taken place within 300 Kb from the telomeres. The largest exchange involved a ~500 Kb segment between Af293 chromosomes 1 and 6 and A1163, which contain regions aligning with A1163 assembly 1 (syntenic blocks 1.1 and 1.2 in Figure 3). This appears to be a recent event that happened in A293. In addition, Af293 chromosome 1 harbours a 400 Kb subtelomeric region that does not align well with A1163 assemblies. There is evidence of gene conversion between distal subtelomeric sequences encoding RecQ family helicases in A. fumigatus chromosomes 2, 4, and 7.

Consistent with previous reports [19], the identity over the shared regions is very high (99.8% at the nucleotide level). This is higher than 99.3% and 99.5% identity between the two sequenced A. niger isolates (ATCC 1015 and CBS 513.88) [23] and between A. oryzae [10] and A. flavus [8], respectively. Unique regions represent 1.2% and 2.3% (and harbour 143 and 218 genes) in the Af293 and A1163 genomes, respectively. More than half of the Af293-specific genes are also absent in A. fumigatus isolates Af294 and Af71, according to the array-based comparative genome hybridization (aCGH) data [7]. The vast majority of Af293- and A1163-specific genes are clustered together in blocks ranging in size from 10 to 400 Kb, which seem to be the most variable segment of the species genome. A manual examination of these isolate-specific islands revealed that they contain numerous pseudogenes and repeat elements. One of the regions contains a putative secondary metabolism cluster (AFUA_3G02530-AFUA3G02670).

The origin of 20% of Af293-specific genes can be attributed to two segmental duplication events. One of the duplicated regions (AFUA_1G16010- AFUA_1G16170) contains an arsenic detoxification cluster. The other (AFUA_1G00420-AFUA_1G00580) contains genes that may be involved in metabolism of betaine, which is often synthesized under osmotic and heavy metal stress. Interestingly the duplicated regions are also absent in Af294 and Af71 isolates, which suggests that the duplication event took place very recently.

Segmental duplication events are thought to contribute to rapid adaptation of the species by increasing their expression. Since Af293 is a clinical isolate it is possible that these chromosomal aberrations were created due to selective pressures in the host.

Highly Variable Loci in A. fumigatus

Although most Af293 proteins are 100% identical to their A1163 orthologs, we have identified 41 orthologous pairs that share only 37% to 95% identity. To find out if these genes are also divergent in other A. fumigatus isolates, we identified Af293 genes that do not hybridize with DNA extracted from the Af294 and Af71 strains in aCGH experiments [7]. The comparison revealed that 27 out of 41 genes were possibly polymorphic (marked as absent or divergent) with respect to at least one other isolate (Table S3). Further analysis of three polymorphic loci in other A. fumigatus isolates has demonstrated that each of them harbours two or three alleles (Table S4). A PCR survey followed by Southern blot analysis and partial DNA sequencing has shown the presence of at least two alleles at each locus containing nearly identical sequences within each group of alleles (data not shown).

In filamentous fungi, this high level of variability has been previously associated with heterokaryon incompatibility (het) genes involved in a programmed cell death (PCD) pathway triggered by hyphal fusion between two genetically incompatible individuals [24],[25]. So far several het loci have been described in A. nidulans [26], although none have been characterized at the molecular level. Incidentally, our results are consistent with previously identified vegetative incompatibility groups suggesting that some of these polymorphic genes may function in heterokaryon incompatibility in A. fumigatus. Thus, four clinical isolates from the same multi-member incompatibility group (WSA-270, WSA-1195, WSA-449, and WSA-172) contained the same alleles of the polymorphic genes (Table S4).

Furthermore, at least five putative A. fumigatus het genes exhibit a pattern of trans-species (or trans-specific) polymorphism (Table S5), which has been previously associated with somatic and sexual incompatibility in fungi, self-incompatibility in plants, and the major histocompatibility complex (MHC) in vertebrates. These genes are more similar to their orthologs from other Aspergillus species than to those from A1163. We chose one putative het gene, rosA (AFUA_1G15910), and its close relative, nosA (AFUA_4G09710), whose orthologs encode two Zn2C6 transcriptional regulators of sexual development in A. nidulans [27],[28] for phylogenetic analysis (Figure 4). Unexpectedly, Af293 RosA clusters with its A. clavatus ortholog, while A1163 RosA clusters with N. fischeri. This is in contrast with the NosA tree, which perfectly mirrors the species tree (Figure 2), suggesting that these allelic classes may transcend species boundaries in the aspergilli.

Figure 4
The Af293 RosA and NosA Proteins.

This is the first study that shows the diversity of het genes in aspergilli at the molecular level as well as patterns of trans-species polymorphism. These putative het genes are distinct from those identified in Neurospora crassa or Podospora anserina [24],[25], although many of them share the same domains such as the NACHT and NB-ARC domains of the STAND superfamily [29]. Coincidentally four of the A. fumigatus variable genes encoding STAND domain proteins have previously been predicted to function in heterokaryon incompatibility [30]. The discovery of putative het loci in the aspergilli may facilitate identification of downstream components of fungal PCD pathways or other drug targets. These loci may be also used as a basis for classification of natural and clinical isolates into different compatibility groups.

A. fumigatus vs. N. fischeri vs. A. clavatus

The genomes of N. fischeri and A. clavatus were sequenced by the whole genome sequencing method [22]. The N. fischeri genome (32.6 Mb) is 10–15% larger than the A. clavatus and A. fumigatus genomes (Table 1). There are 10,407 protein-coding genes and a large number of transposable elements, which may have contributed to its genome size expansion. The A. clavatus genome (27.9 Mb) is the smallest seen to date among the sequenced aspergilli (Table 1). There are currently 9,125 predicted protein-coding genes. This is consistent with past comparative studies that identified notable (up to 30%) genome size differences between distantly related aspergilli [7],[9],[10].

Despite this significant genome size variability, gene-level comparisons confirmed phylogenetic proximity of A. fumigatus, N. fischeri and A. clavatus (Figures 1 and and2).2). The three genomes also appear to be largely syntenic. Alignment of the N. fischeri and A. clavatus genomes against the eight Af293 chromosomes has revealed 20 and 55 syntenic blocks, respectively (Table 2). There is only one large-scale reciprocal translocation between chromosomes 2 and 5 in N. fischeri (blocks 8927.1, 8927.2, 9292.1 and 9292.2, in Figure 3). The A. clavatus supercontigs align with A. fumigatus chromosomes 2 and 5, suggesting that this was the ancestral topology.

Table 2
Syntenic and Afum-specific Chromosomal Blocks in Af293

Core and Lineage-Specific Genes

Features of Core and Lineage-Specific Genes

Comparative genomic analysis has showed that the three Aspergillus genomes contain a large number of species-specific genes, which is consistent with previous comparative studies [7]. We have identified 7514 orthologous core and 818, 1402 and 1151 species-specific genes in the Af293, N. fischeri and A. clavatus genomes, respectively (Figure 5). Numbers of core- and species-specific genes, however, depend on selection of genomes from which they were derived. Thus, adding new genomes to this comparison resulted in fewer core and specific genes as shown for Af293 in Table S6. The availability of additional sequenced Aspergillus genomes allowed us to explore these patterns in a more systematic manner by comparing A. fumigatus Af293 genes with different lineage specificity (i.e. number of orthologs in other species).

Figure 5
Proteins with Orthologs in the Three Most Closely Related Aspergilli (A. fumigatus, N. fischeri and A. clavatus).

To this end, we have selected four sets of genes based on the presence of orthologs in the six other sequenced aspergilli: N. fischeri, A. clavatus, A. terreus (CH476594), A. oryzae [10], A. nidulans [9] and A. niger CBS 513.88 [23] (Table S6; Figure 5). Genes with orthologs in the three most closely related aspergilli (A. fumigatus, N. fischeri and A. clavatus) constitute the Affc-core group. The genes in the Affc-core can be further divided into two groups, the Aspergillus-core (Asp-core) with orthologs in all six other aspergilli and the Affc-specific group, which is comprised of the remaining Affc-core genes. Finally, the A. fumigatus-specific (Afum-specific) group contains Af293 genes that have orthologs in neither N. fischeri nor A. clavatus.

One of the most striking observations to arise from this comparison was the marked differences in size and number of exons among genes from different lineage-specificity groups (Table 3). For example, Asp-core genes on average are almost twice as large as Afum-specific genes. The latter have on average only 1.35 introns and almost 31% lack introns completely. In contrast, Asp-core genes contain on average 2.16 introns, only 16% of them without introns. Consistent with previous reports of increased evolutionary rates in LS genes (e.g. [31]), Affc- and Afum-specific genes in A. fumigatus exhibit low sequence identity to their orthologs from more distantly related fungi (Table 3).

Table 3
Comparison of Four Af293 Gene Sets with Different Lineage Specificity

These vast differences in gene features between core and specific genes are more likely to be explained by relaxed selective constraints (as discussed below) than by poor annotation quality of LS genes (due to misannotated gene models, gene fragments or random ORFs). We made significant improvements to Af294 gene models by leveraging the comparative genomic data (see Materials and Methods). In addition, all Affc-specific genes have orthologs in N. fischeri and A. clavatus and 43% of them are differentially expressed in various expression studies, which is similar to the A. fumigatus genome average (Table 3). On the other hand, many Afum-specific genes may be non-functional, since only 32% of them are differentially expressed in microarray studies (vs. the 43% genome average) and only 60% of them show sequence similarity to other fungal proteins (Table S7; Figure 6). Nonetheless, at least 20% of Afum-specific genes are supported by combined evidence (homology and expression data) and therefore are likely to be functional. Nonetheless, even these genes are still smaller in size than average Affc- and Asp-core genes.

Figure 6
A. fumigatus-Species Specific Genes Supported by Homology and Expression Data.

Biological Roles and Chromosomal Location of LS Genes

Analysis of Gene Ontology (GO) terms [32] associated with core and lineage-specific groups has demonstrated that certain biological functions are unequally distributed among these groups (Table S8). The Afum-specific group is enriched for genes involved in carbohydrate transport and catabolism, secondary metabolite biosynthesis, and detoxification. In contrast, the invariable Asp-core genome encodes many functions associated with information processing and other cellular processes that contribute to the organism's fitness in most environments. Thus, a significant number of Asp-core genes (15%) are orthologous to yeast essential genes, which represents a two-fold enrichment in comparison to the rest of the proteome.

Although most Af293 genes involved in carbohydrate transport and catabolism are found in the Asp-core group, only 10% of secondary metabolism genes have orthologs in all sequenced aspergilli including siderophore, pigment and Pes1-related clusters. These three conserved clusters are also found in Penicillium species and some more distantly related fungi. Similarly, only 30% of secondary metabolism Af293 genes are shared by N. fischeri and A. clavatus. The three species also vary considerably in the numbers of enzymes that control the first step in secondary metabolite biosynthesis such as nonribosomal peptide synthases (NRPS), polyketide synthases (PKS), and dimethylallyltryptophan synthases (DMATS) (Table S9). Interestingly, N. fischeri genome contains 46 enzymes, which is 35% more than A. clavatus (35) and A. fumigatus (34) genomes.

Likewise, PFAM domains overrepresented among Affc- and Afum-specific genes have been shown to function in efflux or detoxification, secondary metabolite biosynthesis, resistance to antifungals, and other accessory metabolic pathways. They include MSF and ABC transporters, various oxidoreductases, cytochrome P450, glycosyl and alpha/beta fold hydrolases, polyketide synthases, glutathione transferases and methyltransferases (Table S10). On the other hand, core genes often contain AAA-superfamily ATPase, helicase, WD40, and SH3 domains associated with such important functions as cell organization and macromolecule biosynthesis.

Lineage Specific Genomic Islands

In addition to difference in size and function, lineage specific genes display a significant subtelomeric bias. As opposed to telomere-distal Asp- and Affc-core genes, Affc- and Afum-specific genes tend to be located within 300 Kb from chromosome ends (P value>0.01) (Table S11). About 38% of Affc-specific genes are telomere-proximal in comparison to 6% of Asp-core and 9% Affc-core genes (Table 3). Interestingly, 46% of Afum-specific genes with paralogs are telomere-proximal (Table S7), suggesting that they may have been recently duplicated and translocated to these regions. Our findings concur with previous reports of subtelomeric bias in LS genes in A. fumigatus [7], S. cerevisiae [33] and Pichia stipitis [34]. With the exception of one Af293 locus containing four P450 genes, the Aspergillus species do not have large variable subtelomeric arrays arising by a series of tandem duplications found in some protozoan parasites [35].

Almost 50% of the Afum-specific genes can be clustered together in 13 blocks containing more than 10 Afum-specific genes separated by no more than 5 genes outside this category (Table 2). Together these regions, referred to here as Afum-specific genomic islands, show an even more significant telomeric bias (68% of the clustered genes lay within 300 Kb from telomere ends) with larger blocks found almost exclusively at chromosome ends (Figure 3). In addition to non-syntenic genes, species-specific islands harbour a disproportionate number of transposons and other repeat elements in comparison with the syntenic areas of the Af293 genome (Table 2). Notably two A. fumigatus-specific blocks (2.2 and 3.1) contain gene clusters involved in biosynthesis of mycotoxin fumigaclavine and another unknown secondary metabolite [36]. Similar genomic islands have been described in the rice blast fungus Magnaporthe oryzae [37],[38] and in A. oryzae [10] suggesting that they may be shared across all filamentous ascomycota fungi. Unlike variable subtelomeric regions found in other eukaryotes [39],[40], these areas are often quite large (up to 400 Kb) and not always located near chromosome ends.

Evolutionary Origins of Lineage-Specific Genes

Most Affc- and Afum-specific genes have no orthologs in non-Aspergillus fungal species, which suggests that they were created de novo in the Affc lineage. To gain insight into the origin of the LS genes in aspergilli, we have performed phylogenetic analysis of two sets of A. fumigatus- and N. fischeri-specific genes. In Af293 and N. fischeri, Set 1 contains 790 and 1230 genes, respectively, that have an Aspergillus homolog as the best BLASTp hit; Set 2 contains 28 and 178 genes, respectively, that have a non-Aspergillus homolog as the closest relative. There is a significant difference in the numbers of trees including a non-Aspergillus species as the closest relative in N. fischeri and A. fumigatus (P value = 2.6e-08). This is indicative of major differences in retention and/or uptake of new genetic material in these two species, consistent with differences in their reproductive modes.

The four repetitive scenarios identified by phylogenetic analysis are displayed in Figure 7. In both A. fumigatus and N. fischeri, most of the Set 1 genes exhibit topologies that do not strictly follow the Aspergillus species tree (Figure 2), although nested within the Aspergillus clade. Similarly, all 28 A. fumigatus Set 2 genes are nested within the Aspergillus genus. In contrast to the A. fumigatus genes, N. fischeri Set 2 genes sometimes cluster with a non-Aspergillus species with high bootstrap support. As shown in Figure 7B and 7C, both N. fischeri and non-Aspergillus species genes can be nested either in this non-Aspergillus clade or in the Aspergillus clade. At first sight, these repetitive topologies can be interpreted as supportive of a horizontal gene transfer (HGT) from a non-Aspergillus species into N. fischeri or visa versa. Further analysis, however, reveals that most of the conflicts involve sparsely populated trees, long branch attraction artifacts, and other situations, where phylogenetic methods tend to mislead (e.g. [41]). The last repetitive scenario includes genes that are only present in one other distant fungal genome (Figure 7D). The evolutionary origin of genes in this category cannot be resolved at this time.

Figure 7
Four Common Topologies Detected by Phylogenetic Analysis of N. fischeri-Specific Proteins.

Our results are consistent with the well established role of gene duplication and divergence as the principal source of new genes [42][45]. They are however in conflict with previous studies that attributed the origin of LS genes in the aspergilli to gene acquisition through HGT from other fungal species [9],[10],[46]. This assumption was based on circumstantial evidence such as mosaic phyletic distribution, phylogenetic anomalies, and differences in gene content among A. fumigatus, A. nidulans and A. oryzae. Besides the absence of readily apparent HGT examples, the fact that LS genes tend to be smaller in size and have fewer exons is difficult to explain by HGT. These gene features are quite consistent across Aspergillus species, and it is therefore unclear what could be the donor organism for LS genes.

The DDL scenario does not have this weakness, since these size differences can be a direct consequence of relaxed selective constraints operating on duplicate genes. According to the DDL hypothesis, the initial redundancy in gene function allows duplicate genes to quickly accumulate nonsynonymous mutations and even premature stop codons. Notably, over 20% of all Afum-specific genes can be linked to the two very recent segmental duplications events that occurred in Af293 but not in A1163. Both translocated segments are telomere-distal and contain genes that appear to be pseudogenized indicating that translocated gene copies may have evolved under relaxed selective constraints. Similarly in other species, accelerated evolution has been often associated with subtelomeric areas suggesting that the process is dependent on the local chromatin environment (e.g. [47]).

The prevailing role of duplication in the origin of LS genes in the aspergilli is further underlined by their tendency to cluster in genomic islands. These regions may function as designated “gene dumps” and simultaneously as “gene factories”, since some LS genes appear to maintain their functional integrity or at least are differentially expressed in microarray studies as shown above. As shown above, 46% of Afum-specific genes with paralogs are telomere-proximal (Table S7), suggesting that they may have been recently duplicated and translocated to these regions. Evidence for gene duplication and/or transfer to evolutionarily labile regions is found in some protozoan parasites that have large variable subtelomeric arrays arising by a series of tandem duplications [35].

Conservation of Virulence-, Allergy-, and Sex-Associated Genes

Previous studies however have shown a high level of evolutionary conservation and phyletic retention among known A. fumigatus virulence-associated genes [7]. Our analysis confirmed the low rate of protein evolution among these genes in four Aspergillus species (Table S12). Interestingly, four of the virulence-associated genes, pabaA (AFUA_6G04820), fos-1 (AFUA_6G10240), pes1 (AFUA_1G10380) and pksP (AFUA_2G17600), reveal evidence of accelerated evolution in the branch leading to the two A. fumigatus isolates. This pattern can affect only a few amino acid residues (e.g. PksP) or a significant proportion of the protein (e.g. Pes1).

Such a pattern can be due to either relaxation of selection or selection for rapid diversification (positive selection). In the latter case specific amino acid substitutions may decrease susceptibility to specific environmental challenges and thus enhance A. fumigatus virulence. These four genes are involved in oxidative stress or nutrient availability, which is consistent with the positive selection scenario. Indeed, PabaA is involved in biosynthesis of folate, an essential co-factor for DNA synthesis. Since PABA is apparently limited in the mammalian lung, a functional pabaA gene is required for virulence [48]. Fos1, a putative two-component histidine kinase, may play a role in the regulation of cell-wall assembly [49]. Finally, PksP and Pes1 are enzymes, which catalyze the first steps in biosynthesis of the spore pigment and an unknown non-ribosomal peptide, have been shown to mediate resistance to oxidative stress in addition to their role in A. fumigatus virulence [50],[51]. The inclusion of additional taxa in the analyses might clarify the significance of the observed differences.

This overall lack of variability among known virulence-associated factors suggests that yet unknown A. fumigatus-specific genes may contribute to its ability to survive in the human host. A recent microarray study demonstrated that the Affc-specific genes are over-represented among genes that are up-expressed in the neutropenic murine lung (Elaine Bignell submitted for publication). Many of them are found in chromosomal gene clusters associated with macromolecule catabolism and secondary metabolite biosynthesis. Similarly, clustered lineage-specific genes simultaneously induced in infected tissue have been observed in the ubiquitous maize pathogen Ustilago maydis [52] and some other species (for a recent review see [53]). Alternatively A. fumigatus virulence may be a combinatorial process, dependent on a pool of genes, which interact in various combinations in different genetic backgrounds as suggested previously [7]. Similar ‘ready-made’ virulence features have been described in other environmental pathogens such as Pseudomonas aeruginosa [54] and Cryptococcus neoformans [55],[56].

In addition to virulence factors, the A. fumigatus genome encodes 20 allergens (Table S13) and 25 proteins displaying significant sequence similarity to known fungal allergens (Table S14), some of which appear to contribute to its pathogenicity [57]. For example, A. fumigatus Asp f6 (AFUA_1G14550), also known as Mn2+-dependent superoxide dismutase (MnSOD), is specifically recognized by IgE from patients with allergic bronchopulmonary aspergillosis (ABPA) and is differentially expressed during germination [58]. The broad distribution of allergens among fungal taxa (Text S1) suggests that A. fumigatus possesses the same allergen complement as most other aspergilli and that its effect on hypersensitive individuals can be explained mostly by its ubiquity in the environment.

Our analysis has demonstrated that, similar to known virulence-associated genes, most sexual development genes appear to be under negative (purifying) selection in both sexual and asexual Aspergillus species (Text S1 and Table S15). More detailed analysis has revealed four genes in the N. fisheri lineage that may be under positive selection. This suggests that a few amino acid changes may enable sexuality in N. fischeri. The conservation of sex genes in asexual species is due to a latent sexuality, a recent loss of sexuality, pleiotropy, or parasexual recombination following heterokaryon formation as suggested previously [59],[60].

Conclusions

Lineage-specific (LS) genes (i.e. genes with limited phylogenetic distribution of orthologs in related species) have been the focal point of many comparative genomic studies, because of the assumption that they may be responsible for phenotypic differences among species and niche adaptation. Our analyses of the genomes of A. fumigatus and the two closely related species, N. fischeri and A. clavatus, demonstrates that A. fumigatus may possess genetic determinants that allow it to establish a successful in vivo infection. LS genes that have no orthologs in the other two species comprise 8,5% of the A. fumigatus genome and often have accessory functions such as carbohydrate and amino acid metabolism, transport, detoxification, or secondary metabolite biosynthesis. Further analysis showed that these genes have distinct features (e.g. the small gene length and number of introns) and tend to cluster in subtelomeric genomic islands, which may function as “gene dumps/factories”. The phylogenies of LS genes, their subtelomeric bias and size differences are consistent with the DDL hypothesis stating that duplication being the primary genetic mechanism responsible for the origin of species-specific genes. The presence of genomic islands indicates that A. fumigatus and may possess sophisticated genetic mechanisms that facilitate its adaptation to heterogeneous environments such as soil or a living host.

Materials and Methods

Fungal Isolates

A. fumigatus Af293 (FGSC A1100) was isolated from patients with invasive aspergillosis [61]. A. fumigatus A1163 (FGSC A1163) is a derivative of A. fumigatus CEA17 converted to pyrG+ via the ectopic insertion of the A. niger pyrG gene [62],[63]. CEA17 is a uracil auxotroph of A. fumigatus clinical isolate CEA10 (CBS144.89). The type strains of A. clavatus (NRRL 1) and N. fischeri (NRRL 181) were used for sequencing and phenotypic characterization.

Whole Genome Sequencing

A1163, A. clavatus and N. fischeri were sequenced using the whole genome shotgun method as previously described [22]. Random shotgun libraries of 2–3 Kb, 8–12 Kb and 50 Kb were constructed from genomic DNA from each strain, and DNA template was prepared for high-throughput sequencing using Big Dye Terminator chemistry (Applied Biosystems). Sequence data was assembled using Celera Assembler. For A. fumigatus A1163, scaffolds were compared to those of the first sequenced isolate, Af293 [7].

Sequence Identity at the Nucleotide Level

A1163 assemblies larger than 5 Kb were aligned to the Af293 chromosomes using the MUMmer package (http://mummer.sourceforge.net/) [64]. Alignments longer than 100 Kb were used to determine average sequence identity to avoid highly repetitive and duplicated regions. The same approach was used to estimate sequence identity between A. flavus and A. oryzae and between the two sequenced A. niger strains.

Gene Structure Annotation

The JCVI eukaryotic annotation pipeline was applied to the A1163, A. clavatus and N. fischeri assemblies (supercontigs) larger than 2 Kb as described earlier [7]. We used PASA [65] and EvidenceModeler [66] to generate consensus gene models based on predictions from several types of genefinders including GlimmerHMM, Genezilla, SNAP, Genewise and Twinscan. Putative pseudogenes, small species-specific genes (less than 50 amino acids), and gene models overlapping with transposable elements (TE) shown in Table S16 were excluded from the final gene lists.

Repetitive Elements

Identification of repeat elements was performed using RepeatMasker (http://www.repeatmasker.org/), RepeatScout (http://repeatscout.bioprojects.org/), and Tandem Repeats Finder (http://tandem.bu.edu/trf/trf.html). Putative TEs (Table S16) were identified by Transposon-PSI (http://transposonpsi.sourceforge.net), a program that performs tBLASTn searches using a set of position specific scoring matrices (PSSMs) specific for different TE families. TE and repeat densities were calculated as the percentage of nucleotide bases in the regions of interest (i.e., syntenic or non-syntenic blocks) that overlap with a feature of the appropriate type (repeat or TE).

A. fumigatus Annotation Improvements

We leveraged the comparative genomic data to significantly improve annotation quality of the Af293 genome, which was previously annotated with relatively little supporting evidence [7]. The refinement of initial annotation was performed using the Sybil software package (http://sybil.sourceforge.net/), which allows for rapid identification of discrepancies in gene structure among orthologs. The comparison with orthologous N. fischeri and A. clavatus genes resulted in significant changes to the Af293 gene catalogue. Over 1100 gene models were updated and 130 new genes were identified. Initial A. fumigatus A1163 gene models were also improved using the PASA pipeline, initially developed to align expressed sequence tag (EST) data onto genomic sequences [65]. The pipeline was adapted to automatically update A1163 gene models by aligning them against Af293 coding sequences (CDSs).

Functional Annotation

We have performed transitive functional annotation from Af293 proteins to their A1163, N. fischeri and A. clavatus orthologs. Previously GO terms [32] were assigned to Af293 proteins based on sequence similarity to PFAM domains or experimentally characterized S. cerevisiae proteins [7]. Secondary metabolism gene clusters were identified using Secondary Metabolism Region Finder (SMURF) available at http://www.jcvi.org/smurf (Nora Khaldi, unpublished). The complete list of gene clusters can be downloaded at ftp://ftp.jcvi.org/pub/software/smurf/. Gene Ontology (GO) terms [32] were assigned as described in [7]

Ortholog Identification

After extensive computational and manual refinement, the improved protein datasets were used to generate the final set of orthologs. Orthologous groups in Aspergillus genomes were identified using a reciprocal-best-BLAST-hit (RBH) approach with a cut-off of 1e-05. In addition to the A1163, A. clavatus and N. fischeri genomes, the previously sequenced genomes of Af293 [7], A. terreus NIH2624 (http://www.broad.mit.edu), A. oryzae RIB40 [10], A. nidulans FGSC A4 [9] and A. niger CBS 513.55 [23] were included in the comparative analysis. The results of this analysis, as well as synteny visualisation and comparative analysis tools can be also found in the Aspergillus Comparative database at http://www.tigr.org/sybil/asp. Orthologous, unique and divergent genes in Af293 were identified based on alignments of Af293 CDSs against A1163 assemblies using gmap as implemented in PASA [65] using default parameters.

Synteny Analysis

Syntenic blocks for each pair of genomes (Af293 vs. A. clavatus and Af293 vs. N. fischeri) were defined as areas containing a minimum of five matching (orthologous) genes with a maximum of 20 adjacent non-matching genes (having no orthologs) in the reference and target genomes. Since most syntenic regions slightly overlapped, the original blocks were merged to calculate repeat and TE density. Af293 non-syntenic blocks were defined as areas excluded from the syntenic blocks and containing at least ten Af239 non-matching genes.

Statistical Analysis

Genes in four lineage-specificity groups were analyzed by the EASE module [67] in MEV within TM4 (http://TM4.org) [68] to identify overrepresented Gene Ontology (GO) terms, Pfam domains and Chromosomal Regions (telomere-proximal and central). Only categories with Fisher's exact test probabilities above with P>0.05 from the EASE analyses were reported for each gene set.

Selective Constraints

Selective constraints were estimated for sets of orthologous genes from the Af293, A1163, A. clavatus, N. fischeri and A. terreus genomes. The rate of substitution in synonymous (dS) and in non-synonymous (dN) sites, and their ratio (dN/dS) was calculated using the PAML package [69]. If a gene is very well conserved, dN/dS<0.1; if a gene is under weak purifying selection, 0.1<dN/dS<1; if a gene is evolving neutrally (e.g. pseudogenes), dN/dS~ = 1; and if a gene is evolving under diversifying selection, dN/dS>1. The results are reported only for orthologous genes sets having unsaturated dS values, the same number of exons, and sequence alignment coverage >95%. For each gene, the average dN/dS ratio for five pairwise species comparisons was calculated.

Phylogenetic Analyses

We assembled a local database of protein sequences from the 28 publicly available fungal genome projects (Table S17). All phylogenetic analyses in this paper were carried out on protein sequences. The A. niger ATCC 1015, Nectria haematococca, Phanerochaete chrysosporium and Trichoderma reesei genomes projects was completed under the auspices of the US Department of Energy's Office of Science, Biological and Environmental Research Program and the by the University of California, Lawrence Livermore National Laboratory (Contract No. W-7405-Eng-48), Lawrence Berkeley National Laboratory (contract No. DE-AC03-76SF00098) and Los Alamos National Laboratory (contract No. W-7405-ENG-36).

To produce a reference tree of species phylogeny we used the protein sequences of 90 likely orthologs from A. niger, A. nidulans, A. terreus, A. oryzae, A. clavatus, N. fischeri, A. fumigatus and Fusarium graminearum (teleomorph of Gibberella zeae) as an outgroup. To minimize the effect of incorrect or incongruent gene models, these proteins were chosen on the basis of having identical numbers of introns in each species and similar lengths. Sequences were aligned using MUSCLE [70] and columns of low conservation were removed manually. Maximum-likelihood trees were constructed using the PHYLIP package, applying the JTT substitution model with a gamma distribution (alpha = 0.5) of rates over four categories of variable sites.

Phylogenetic analyses of individual Af293, A1163, and N. fischeri proteins were carried out on sets of homologs identified in BLASTP searches against our fungal database. The top 20 hits with E<10−4 were retained for analysis. Sequences were aligned using ClustalW [71]. Poorly aligned regions were removed using Gblocks [72]. Finally, a maximum likelihood tree was drawn using PHYML [73].

Southern Blot Analysis

To detect polymorphisms in the rosA (AFUA_6G07010) gene, several hybridizations were performed using rosA gene as the probe and genomic DNA cleaved with EcoRI, ClaI, BamHI or EcoRV. For comparison, an invariable gene for all species (apg5; AFUA_6G07040) was used as the hybridization probe on genomic DNA digested with HpaI.

Colony Radial Growth Rate Measurement

Colony radial growth rate measurements were performed as described [74]. For each isolate, four (90 mm diameter) Petri dishes containing 25 ml agar medium were inoculated centrally with 2.5 µl of 1×106 spores/ml suspension in PBS/Tween 80. Plates were then incubated at temperatures ranging from 25°C to 50°C and colony edges were marked using a plate microscope. Colonies were marked twice daily for 4–5 days. For each colony, two diameters perpendicular to each other were measured.

Eight replicates were measured for each isolate. The results reported here are the mean of two experiments. At least five time points during the log phase were used to calculate growth rate. The radius of the colonies was plotted against time using least-square regression analysis, and the slope of the regression line, which represents the growth rate, was calculated. Each replicate was analysed separately and the mean of the growth rate was then calculated.

Supporting Information

Text S1

Allergens and sexual development genes.

(0.05 MB DOC)

Table S1

Growth rates of Af293, A1163, N. fischeri, and A. clavatus isolates at various temperatures.

(0.02 MB XLS)

Table S2

Resistance to antifungals among A. fumigatus clinical isolates.

(0.02 MB XLS)

Table S3

Divergent A. fumigatus Af293 genes with respect to Af294, Af71, and A1163.

(0.03 MB XLS)

Table S4

Distribution of polymorphic alleles among A. fumigatus isolates.

(0.02 MB XLS)

Table S5

Five A. fumigatus loci exhibiting trans-species polymorphism.

(0.02 MB XLS)

Table S6

A. fumigatus core and species-specific genes.

(0.02 MB XLS)

Table S7

Features of A. fumigatus-specific genes.

(0.02 MB XLS)

Table S8

Top biological processes overrepresented among four lineage specificity groups.

(0.02 MB XLS)

Table S9

Enzymes that control the first step in secondary metabolite biosynthesis.

(0.02 MB XLS)

Table S10

Top PFAM domains overrepresented among four lineage specificity groups.

(0.02 MB XLS)

Table S11

Lineage specificity and chromosomal location.

(0.02 MB XLS)

Table S12

Selective constraints operating on virulence-associated genes.

(0.02 MB XLS)

Table S13

Known A. fumigatus Af293 allergens.

(0.02 MB XLS)

Table S14

Predicted A. fumigatus Af293 allergens.

(0.02 MB XLS)

Table S15

Selective constraints operating on sex genes.

(0.02 MB XLS)

Table S16

Families of transposable elements identified in the Affc genomes.

(0.01 MB XLS)

Table S17

Fungal genomes used in phylogenetic analyses.

(0.02 MB XLS)

Footnotes

The authors have declared that no competing interests exist.

This work was supported by contract N01-AI30071, U01-AI48830, R21-AI052236 from NIAID and partially funded by Fundação de Amparo a Pesquisa de São Paulo (FAPESP), Brazil to GHG.

References

1. Gugnani HC. Ecology and taxonomy of pathogenic aspergilli. Front Biosci. 2003;8:s346–357. [PubMed]
2. Denning DW. Invasive aspergillosis. Clin Infect Dis. 1998;26:781–803. quiz 804-785. [PubMed]
3. Latge JP. Aspergillus fumigatus and aspergillosis. Clin Microbiol Rev. 1999;12:310–350. [PMC free article] [PubMed]
4. Rankin NE. Disseminated aspergillosis and moniliasis associated with agranulocytosis and antibiotic therapy. Br Med J. 1953;1:918–919. [PMC free article] [PubMed]
5. Pasqualotto AC, Denning DW. Post-operative aspergillosis. Clin Microbiol Infect (in press) 2006 [PubMed]
6. Richardson MD, Rennie S, Marshall I, Morgan MG, Murphy JA, et al. Fungal surveillance of an open haematology ward. J Hosp Infect. 2000;45:288–292. [PubMed]
7. Nierman WC, Pain A, Anderson MJ, Wortman JR, Kim HS, et al. Genomic sequence of the pathogenic and allergenic filamentous fungus Aspergillus fumigatus. Nature. 2005;438:1151–1156. [PubMed]
8. Payne GA, Nierman WC, Wortman JR, Pritchard BL, Brown D, et al. Whole genome comparison of Aspergillus flavus and A. oryzae. Med Mycol. 2006;44(Suppl):9–11.
9. Galagan JE, et al. Sequencing of Aspergillus nidulans and comparative analysis with A. fumigatus and A. oryzae. Nature. 2005;438:1105–1115. [PubMed]
10. Machida M, et al. Genome sequencing and analysis of Aspergillus oryzae. Nature. 2005;438:1157–1161. [PubMed]
11. Taylor JW, Berbee ML. Dating divergences in the Fungal Tree of Life: review and new analyses. Mycologia. 2006;98:838–849. [PubMed]
12. Chim CS, Ho PL, Yuen KY. Simultaneous Aspergillus fischeri and Herpes simplex pneumonia in a patient with multiple myeloma. Scand J Infect Dis. 1998;30:190–191. [PubMed]
13. Lonial S, Williams L, Carrum G, Ostrowski M, McCarthy P., Jr Neosartorya fischeri: an invasive fungal pathogen in an allogeneic bone marrow transplant patient. Bone Marrow Transplant. 1997;19:753–755. [PubMed]
14. Opal SM, Reller LB, Harrington G, Cannady P., Jr Aspergillus clavatus endocarditis involving a normal aortic valve following coronary artery surgery. Rev Infect Dis. 1986;8:781–785. [PubMed]
15. Summerbell RC, de Repentigny L, Chartrand C, St Germain G. Graft-related endocarditis caused by Neosartorya fischeri var. spinosa. J Clin Microbiol. 1992;30:1580–1582. [PMC free article] [PubMed]
16. Blyth W, Grant IW, Blackadder ES, Greenberg M. Fungal antigens as a source of sensitization and respiratory disease in Scottish maltworkers. Clin Allergy. 1977;7:549–562. [PubMed]
17. Kellerman TS, Pienaar JG, van der Westhuizen GC, Anderson GC, Naude TW. A highly fatal tremorgenic mycotoxicosis of cattle caused by Aspergillus clavatus. Onderstepoort J Vet Res. 1976;43:147–154. [PubMed]
18. Pringle A, Baker DM, Platt JL, Wares JP, Latge JP, et al. Cryptic speciation in the cosmopolitan and clonal human pathogenic fungus Aspergillus fumigatus. Evolution Int J Org Evolution. 2005;59:1886–1899. [PubMed]
19. Rydholm C, Szakacs G, Lutzoni F. Low genetic variation and no detectable population structure in aspergillus fumigatus compared to closely related Neosartorya species. Eukaryot Cell. 2006;5:650–657. [PMC free article] [PubMed]
20. Balajee SA, Tay ST, Lasker BA, Hurst SF, Rooney AP. Characterization of a novel gene for strain typing reveals substructuring of Aspergillus fumigatus across North America. Eukaryot Cell. 2007;6:1392–1399. [PMC free article] [PubMed]
21. Levdansky E, Romano J, Shadkchan Y, Sharon H, Verstrepen KJ, et al. Coding tandem repeats generate genetic diversity in Aspergillus fumigatus genes. Eukaryot Cell 2007 [PMC free article] [PubMed]
22. Fleischmann RD, Adams MD, White O, Clayton RA, Kirkness EF, et al. Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science. 1995;269:496–512. [PubMed]
23. Pel HJ, de Winde JH, Archer DB, Dyer PS, Hofmann G, et al. Genome sequencing and analysis of the versatile cell factory Aspergillus niger CBS 513.88. Nat Biotechnol. 2007;25:221–231. [PubMed]
24. Glass NL, Dementhon K. Non-self recognition and programmed cell death in filamentous fungi. Curr Opin Microbiol. 2006;9:553–558. [PubMed]
25. Paoletti M, Saupe SJ, Clave C. Genesis of a fungal non-self recognition repertoire. PLoS ONE. 2007;2:e283. [PMC free article] [PubMed]
26. Anwar MM, Croft JH, Dales RB. Analysis of heterokaryon incompatibility between heterokaryon-compatibility (h-c) groups R and GL provides evidence that at least eight het loci control somatic incompatibility in Aspergillus nidulans. J Gen Microbiol. 1993;139:1599–1603. [PubMed]
27. Vienken K, Fischer R. The Zn(II)2Cys6 putative transcription factor NosA controls fruiting body formation in Aspergillus nidulans. Mol Microbiol. 2006;61:544–554. [PubMed]
28. Vienken K, Scherer M, Fischer R. The Zn(II)2Cys6 putative Aspergillus nidulans transcription factor repressor of sexual development inhibits sexual development under low-carbon conditions and in submersed culture. Genetics. 2005;169:619–630. [PMC free article] [PubMed]
29. Leipe DD, Koonin EV, Aravind L. STAND, a class of P-loop NTPases including animal and plant regulators of programmed cell death: multiple, complex domain architectures, unusual phyletic patterns, and evolution by horizontal gene transfer. J Mol Biol. 2004;343:1–28. [PubMed]
30. Fedorova ND, Badger JH, Robson GD, Wortman JR, Nierman WC. Comparative analysis of programmed cell death pathways in filamentous fungi. BMC Genomics. 2005;6:177. [PMC free article] [PubMed]
31. Cai JJ, Woo PC, Lau SK, Smith DK, Yuen KY. Accelerated evolutionary rate may be responsible for the emergence of lineage-specific genes in ascomycota. J Mol Evol. 2006;63:1–11. [PubMed]
32. Ashburner M, Ball CA, Blake JA, Botstein D, Butler H, et al. Gene ontology: tool for the unification of biology. The Gene Ontology Consortium. Nat Genet. 2000;25:25–29. [PMC free article] [PubMed]
33. Zakian VA. Structure, function, and replication of Saccharomyces cerevisiae telomeres. Annu Rev Genet. 1996;30:141–172. [PubMed]
34. Jeffries TW, Grigoriev IV, Grimwood J, Laplaza JM, Aerts A, et al. Genome sequence of the lignocellulose-bioconverting and xylose-fermenting yeast Pichia stipitis. Nat Biotechnol. 2007;25:319–326. [PubMed]
35. Berriman M, Hall N, Sheader K, Bringaud F, Tiwari B, et al. The architecture of variant surface glycoprotein gene expression sites in Trypanosoma brucei. Mol Biochem Parasitol. 2002;122:131–140. [PubMed]
36. Perrin RM, Fedorova ND, Bok JW, Cramer RA, Wortman JR, et al. Transcriptional Regulation of Chemical Diversity in Aspergillus fumigatus by LaeA. PLoS Pathog. 2007;3:e50. [PMC free article] [PubMed]
37. Rehmeyer C, Li W, Kusaba M, Kim YS, Brown D, et al. Organization of chromosome ends in the rice blast fungus, Magnaporthe oryzae. Nucleic Acids Res. 2006;34:4685–4701. [PMC free article] [PubMed]
38. Thon MR, Pan H, Diener S, Papalas J, Taro A, et al. The role of transposable element clusters in genome evolution and loss of synteny in the rice blast fungus Magnaporthe oryzae. Genome Biol. 2006;7:R16. [PMC free article] [PubMed]
39. Mefford HC, Trask BJ. The complex structure and dynamic evolution of human subtelomeres. Nat Rev Genet. 2002;3:91–102. [PubMed]
40. Pryde FE, Gorham HC, Louis EJ. Chromosome ends: all the same under their caps. Curr Opin Genet Dev. 1997;7:822–828. [PubMed]
41. Rice DW, Palmer JD. An exceptional horizontal gene transfer in plastids: gene replacement by a distant bacterial paralog and evidence that haptophyte and cryptophyte plastids are sisters. BMC Biol. 2006;4:31. [PMC free article] [PubMed]
42. Choi IG, Kim SH. Global extent of horizontal gene transfer. Proc Natl Acad Sci U S A. 2007;104:4489–4494. [PMC free article] [PubMed]
43. Kimura M, Ota T. On some principles governing molecular evolution. Proc Natl Acad Sci U S A. 1974;71:2848–2852. [PMC free article] [PubMed]
44. Kurland CG, Canback B, Berg OG. Horizontal gene transfer: a critical view. Proc Natl Acad Sci U S A. 2003;100:9658–9662. [PMC free article] [PubMed]
45. Ohno S. Evolution by Gene Duplication. New York: Springer-Verlag; 1970.
46. Goffeau A. Genomics: multiple moulds. Nature. 2005;438:1092–1093. [PubMed]
47. Kellis M, Patterson N, Endrizzi M, Birren B, Lander ES. Sequencing and comparison of yeast species to identify genes and regulatory elements. Nature. 2003;423:241–254. [PubMed]
48. Brown JS, Aufauvre-Brown A, Brown J, Jennings JM, Arst H, Jr, et al. Signature-tagged and directed mutagenesis identify PABA synthetase as essential for Aspergillus fumigatus pathogenicity. Mol Microbiol. 2000;36:1371–1380. [PubMed]
49. Clemons KV, Miller TK, Selitrennikoff CP, Stevens DA. fos-1, a putative histidine kinase as a virulence factor for systemic aspergillosis. Med Mycol. 2002;40:259–262. [PubMed]
50. Langfelder K, Jahn B, Gehringer H, Schmidt A, Wanner G, et al. Identification of a polyketide synthase gene (pksP) of Aspergillus fumigatus involved in conidial pigment biosynthesis and virulence. Med Microbiol Immunol (Berl) 1998;187:79–89. [PubMed]
51. Reeves EP, Reiber K, Neville C, Scheibner O, Kavanagh K, et al. A nonribosomal peptide synthetase (Pes1) confers protection against oxidative stress in Aspergillus fumigatus. Febs J. 2006;273:3038–3053. [PubMed]
52. Kamper J, Kahmann R, Bolker M, Ma LJ, Brefort T, et al. Insights from the genome of the biotrophic fungal plant pathogen Ustilago maydis. Nature. 2006;444:97–101. [PubMed]
53. Merrick CJ, Duraisingh MT. Heterochromatin-mediated control of virulence gene expression. Mol Microbiol. 2006;62:612–620. [PubMed]
54. Lee DG, Urbach JM, Wu G, Liberati NT, Feinbaum RL, et al. Genomic analysis reveals that Pseudomonas aeruginosa virulence is combinatorial. Genome Biol. 2006;7:R90. [PMC free article] [PubMed]
55. Casadevall A, Steenbergen JN, Nosanchuk JD. ‘Ready made’ virulence and ‘dual use’ virulence factors in pathogenic environmental fungi–the Cryptococcus neoformans paradigm. Curr Opin Microbiol. 2003;6:332–337. [PubMed]
56. Steenbergen JN, Casadevall A. The origin and maintenance of virulence for the human pathogenic fungus Cryptococcus neoformans. Microbes Infect. 2003;5:667–675. [PubMed]
57. Bowyer P, Denning DW. Genomic analysis of allergen genes in Aspergillus spp.: the relevance of genomics to everyday research. Med Mycol. 2007;45:17–26. [PubMed]
58. Schwienbacher M, Israel L, Heesemann J, Ebel F. Asp f6, an Aspergillus allergen specifically recognized by IgE from patients with allergic bronchopulmonary aspergillosis, is differentially expressed during germination. Allergy. 2005;60:1430–1435. [PubMed]
59. Dyer PS, Paoletti M. Reproduction in Aspergillus fumigatus: sexuality in a supposedly asexual species? Med Mycol. 2005;43(Suppl 1):S7–14. [PubMed]
60. Paoletti M, Rydholm C, Schwier EU, Anderson MJ, Szakacs G, et al. Evidence for sexuality in the opportunistic fungal pathogen Aspergillus fumigatus. Curr Biol. 2005;15:1242–1248. [PubMed]
61. Pain A, Woodward J, Quail MA, Anderson MJ, Clark R, et al. Insight into the genome of Aspergillus fumigatus: analysis of a 922 kb region encompassing the nitrate assimilation gene cluster. Fungal Genet Biol. 2004;41:443–453. [PubMed]
62. d'Enfert C. Selection of multiple disruption events in Aspergillus fumigatus using the orotidine-5′-decarboxylase gene, pyrG, as a unique transformation marker. Curr Genet. 1996;30:76–82. [PubMed]
63. Weidner G, d'Enfert C, Koch A, Mol PC, Brakhage AA. Development of a homologous transformation system for the human pathogenic fungus Aspergillus fumigatus based on the pyrG gene encoding orotidine 5′-monophosphate decarboxylase. Curr Genet. 1998;33:378–385. [PubMed]
64. Delcher AL, Kasif S, Fleischmann RD, Peterson J, White O, et al. Alignment of whole genomes. Nucleic Acids Res. 1999;27:2369–2376. [PMC free article] [PubMed]
65. Haas BJ, Delcher AL, Mount SM, Wortman JR, Smith RK, Jr, et al. Improving the Arabidopsis genome annotation using maximal transcript alignment assemblies. Nucleic Acids Res. 2003;31:5654–5666. [PMC free article] [PubMed]
66. Haas BJ, Salzberg SL, Zhu W, Pertea M, Allen JE, et al. Automated eukaryotic gene structure annotation using EVidenceModeler and the Program to Assemble Spliced Alignments. Genome Biol. 2008;9:R7. [PMC free article] [PubMed]
67. Hosack DA, Dennis G, Jr, Sherman BT, Lane HC, Lempicki RA. Identifying biological themes within lists of genes with EASE. Genome Biol. 2003;4:R70. [PMC free article] [PubMed]
68. Saeed AI, Bhagabati NK, Braisted JC, Liang W, Sharov V, et al. TM4 microarray software suite. Methods Enzymol. 2006;411:134–193. [PubMed]
69. Yang Z. PAML: a program package for phylogenetic analysis by maximum likelihood. Comput Appl Biosci. 1997;13:555–556. [PubMed]
70. Edgar RC. MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 2004;32:1792–1797. [PMC free article] [PubMed]
71. Thompson JD, Higgins DG, Gibson TJ. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 1994;22:4673–4680. [PMC free article] [PubMed]
72. Castresana J. Selection of conserved blocks from multiple alignments for their use in phylogenetic analysis. Mol Biol Evol. 2000;17:540–552. [PubMed]
73. Guindon S, Gascuel O. A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol. 2003;52:696–704. [PubMed]
74. Robson GD, Wiebe MG, Cunliffe B, Trinci AP. Choline- and acetylcholine-induced changes in the morphology of Fusarium graminearum: evidence for the involvement of the choline transport system and acetylcholinesterase. Microbiology. 1995;141( Pt 6):1309–1314. [PubMed]

Articles from PLoS Genetics are provided here courtesy of Public Library of Science
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...