Limits...
The evolution of drug resistance in clinical isolates of Candida albicans.

Ford CB, Funt JM, Abbey D, Issi L, Guiducci C, Martinez DA, Delorey T, Li BY, White TC, Cuomo C, Rao RP, Berman J, Thompson DA, Regev A - Elife (2015)

Bottom Line: Studies in clinical isolates have implicated multiple mechanisms in resistance, but have focused on large-scale aberrations or candidate genes, and do not comprehensively chart the genetic basis of adaptation.LOH events were commonly associated with acquired resistance, and SNPs in 240 genes may be related to host adaptation.Conversely, most aneuploidies were transient and did not correlate with drug resistance.

View Article: PubMed Central - PubMed

Affiliation: Department of Biology, Broad Institute of MIT and Harvard, Cambridge, United States.

ABSTRACT
Candida albicans is both a member of the healthy human microbiome and a major pathogen in immunocompromised individuals. Infections are typically treated with azole inhibitors of ergosterol biosynthesis often leading to drug resistance. Studies in clinical isolates have implicated multiple mechanisms in resistance, but have focused on large-scale aberrations or candidate genes, and do not comprehensively chart the genetic basis of adaptation. Here, we leveraged next-generation sequencing to analyze 43 isolates from 11 oral candidiasis patients. We detected newly selected mutations, including single-nucleotide polymorphisms (SNPs), copy-number variations and loss-of-heterozygosity (LOH) events. LOH events were commonly associated with acquired resistance, and SNPs in 240 genes may be related to host adaptation. Conversely, most aneuploidies were transient and did not correlate with drug resistance. Our analysis also shows that isolates also varied in adherence, filamentation, and virulence. Our work reveals new molecular mechanisms underlying the evolution of drug resistance and host adaptation.

Show MeSH

Related in: MedlinePlus

Co-occurrence of nonsynonymous SNPs occurring in conjunction with ashift in MIC.(A) For each of the 166 recurrently mutated genes associatedwith a change in MIC, we constructed a patient-by-gene binary vector. Weclustered the resulting patient by gene matrix using NMF clustering toreveal 5 coherent clusters (correlation matrix of the clusters left; red:positive correlation; blue: negative correlation; white: no correlation).(B) Co-occurrence clusters. For the genes in each cluster(rows), shown are their mutated occurrences in each patient (columns);green: gene is persistently mutated in patient, white: no persistentmutation. Functional enrichment of clusters was revealed using geneontology, and genes matching the enriched cluster function arebolded.DOI:http://dx.doi.org/10.7554/eLife.00662.015
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4383195&req=5

fig5s1: Co-occurrence of nonsynonymous SNPs occurring in conjunction with ashift in MIC.(A) For each of the 166 recurrently mutated genes associatedwith a change in MIC, we constructed a patient-by-gene binary vector. Weclustered the resulting patient by gene matrix using NMF clustering toreveal 5 coherent clusters (correlation matrix of the clusters left; red:positive correlation; blue: negative correlation; white: no correlation).(B) Co-occurrence clusters. For the genes in each cluster(rows), shown are their mutated occurrences in each patient (columns);green: gene is persistently mutated in patient, white: no persistentmutation. Functional enrichment of clusters was revealed using geneontology, and genes matching the enriched cluster function arebolded.DOI:http://dx.doi.org/10.7554/eLife.00662.015

Mentions: We identified persistent nonsynonymous coding SNPs within 1470 genes outside LOHtracts, 167 of them harboring 336 driver-like polymorphisms (Figure 5—source data1B). These again include ERG11 in patients 9, 14, 30, and59 and TAC1 in patients 1, 7, 14, 15, 30 and 43 (Figure 5). Applying the recurrence filter (i.e.,persistent nonsynonymous SNPs that appeared in the same ORFs in three or more patientseries), we identified 240 polymorphic genes that are more likely to have contributedto adaptation (Figure5—source data 1A). This number of genes is higher than expected bychance (empirical p < 10−4 based on a Poisson model ofbackground mutation, ‘Materials and methods’). Though the codingsequence for these 240 recurrent genes is longer than average (2.21 ± 1.53 kb vs1.83 ± 1.29 kb for non-recurrent persistent genes, p < 3.68 ×10−5, t-test), and thus a larger target formutation, our simulation accounts for gene length. Notably, 17 persistent recurrentlypolymorphic genes also had driver-like polymorphisms, eight of which were alsohomozygosed in an LOH tract in at least one patient series (Figure 5, Figure 5—source data 1A,B,C). Finally, polymorphisms in 166 of the240 genes appeared together with an increase in MIC and are thus stronger candidatesfor making a significant functional contribution to resistance (Figure 5—figure supplement 1, Figure 5—source data1D, empirical p < 10−5 based a binomial model,‘Materials and methods’).


The evolution of drug resistance in clinical isolates of Candida albicans.

Ford CB, Funt JM, Abbey D, Issi L, Guiducci C, Martinez DA, Delorey T, Li BY, White TC, Cuomo C, Rao RP, Berman J, Thompson DA, Regev A - Elife (2015)

Co-occurrence of nonsynonymous SNPs occurring in conjunction with ashift in MIC.(A) For each of the 166 recurrently mutated genes associatedwith a change in MIC, we constructed a patient-by-gene binary vector. Weclustered the resulting patient by gene matrix using NMF clustering toreveal 5 coherent clusters (correlation matrix of the clusters left; red:positive correlation; blue: negative correlation; white: no correlation).(B) Co-occurrence clusters. For the genes in each cluster(rows), shown are their mutated occurrences in each patient (columns);green: gene is persistently mutated in patient, white: no persistentmutation. Functional enrichment of clusters was revealed using geneontology, and genes matching the enriched cluster function arebolded.DOI:http://dx.doi.org/10.7554/eLife.00662.015
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4383195&req=5

fig5s1: Co-occurrence of nonsynonymous SNPs occurring in conjunction with ashift in MIC.(A) For each of the 166 recurrently mutated genes associatedwith a change in MIC, we constructed a patient-by-gene binary vector. Weclustered the resulting patient by gene matrix using NMF clustering toreveal 5 coherent clusters (correlation matrix of the clusters left; red:positive correlation; blue: negative correlation; white: no correlation).(B) Co-occurrence clusters. For the genes in each cluster(rows), shown are their mutated occurrences in each patient (columns);green: gene is persistently mutated in patient, white: no persistentmutation. Functional enrichment of clusters was revealed using geneontology, and genes matching the enriched cluster function arebolded.DOI:http://dx.doi.org/10.7554/eLife.00662.015
Mentions: We identified persistent nonsynonymous coding SNPs within 1470 genes outside LOHtracts, 167 of them harboring 336 driver-like polymorphisms (Figure 5—source data1B). These again include ERG11 in patients 9, 14, 30, and59 and TAC1 in patients 1, 7, 14, 15, 30 and 43 (Figure 5). Applying the recurrence filter (i.e.,persistent nonsynonymous SNPs that appeared in the same ORFs in three or more patientseries), we identified 240 polymorphic genes that are more likely to have contributedto adaptation (Figure5—source data 1A). This number of genes is higher than expected bychance (empirical p < 10−4 based on a Poisson model ofbackground mutation, ‘Materials and methods’). Though the codingsequence for these 240 recurrent genes is longer than average (2.21 ± 1.53 kb vs1.83 ± 1.29 kb for non-recurrent persistent genes, p < 3.68 ×10−5, t-test), and thus a larger target formutation, our simulation accounts for gene length. Notably, 17 persistent recurrentlypolymorphic genes also had driver-like polymorphisms, eight of which were alsohomozygosed in an LOH tract in at least one patient series (Figure 5, Figure 5—source data 1A,B,C). Finally, polymorphisms in 166 of the240 genes appeared together with an increase in MIC and are thus stronger candidatesfor making a significant functional contribution to resistance (Figure 5—figure supplement 1, Figure 5—source data1D, empirical p < 10−5 based a binomial model,‘Materials and methods’).

Bottom Line: Studies in clinical isolates have implicated multiple mechanisms in resistance, but have focused on large-scale aberrations or candidate genes, and do not comprehensively chart the genetic basis of adaptation.LOH events were commonly associated with acquired resistance, and SNPs in 240 genes may be related to host adaptation.Conversely, most aneuploidies were transient and did not correlate with drug resistance.

View Article: PubMed Central - PubMed

Affiliation: Department of Biology, Broad Institute of MIT and Harvard, Cambridge, United States.

ABSTRACT
Candida albicans is both a member of the healthy human microbiome and a major pathogen in immunocompromised individuals. Infections are typically treated with azole inhibitors of ergosterol biosynthesis often leading to drug resistance. Studies in clinical isolates have implicated multiple mechanisms in resistance, but have focused on large-scale aberrations or candidate genes, and do not comprehensively chart the genetic basis of adaptation. Here, we leveraged next-generation sequencing to analyze 43 isolates from 11 oral candidiasis patients. We detected newly selected mutations, including single-nucleotide polymorphisms (SNPs), copy-number variations and loss-of-heterozygosity (LOH) events. LOH events were commonly associated with acquired resistance, and SNPs in 240 genes may be related to host adaptation. Conversely, most aneuploidies were transient and did not correlate with drug resistance. Our analysis also shows that isolates also varied in adherence, filamentation, and virulence. Our work reveals new molecular mechanisms underlying the evolution of drug resistance and host adaptation.

Show MeSH
Related in: MedlinePlus