Microsatellite (STRAf) Genotyping Cannot Differentiate between Invasive and Colonizing Aspergillus fumigatus Isolates

TEXT

Diagnosing invasive aspergillosis is challenging and is based on host factors, radiological findings, and mycological criteria (1). Culture of Aspergillus species is widely used, although it is impaired by low specificity in some groups of patients, as shown by the high proportion of nonhematological patients with Aspergillus isolated from lower respiratory tract samples but no invasive disease (2, 3). Therefore, the isolation of Aspergillus fumigatus should be interpreted with caution.

A. fumigatus is a ubiquitous fungus that is able to colonize or infect patients depending on the integrity of the immune system (4); however, different degrees of virulence have been observed in clinical isolates of A. fumigatus (5, 6). Hypothetically, some isolates cause invasive aspergillosis, whereas others only colonize patients. Unfortunately, to our knowledge, the association between specific isolates and the presence of invasive aspergillosis has not been studied.

(This study was presented in part at the 24th European Congress on Clinical Microbiology and Infectious Diseases, Barcelona, Spain, 10 to 13 May 2014.)

A large collection of clinical A. fumigatus sensu stricto isolates was analyzed to determine whether microsatellite (short tandem repeats of Aspergillus fumigatus [STRAf]) genotyping was able to differentiate between isolates from patients with invasive aspergillosis and isolates from colonized patients.

A total of 95 patients admitted to Hospital General Universitario Gregorio Marañón between 2005 and 2012 and diagnosed with proven (n = 16) or probable (n = 79) invasive aspergillosis were studied. Isolates were available from all the patients, who were classified according to the revised criteria of the European Organisation for Research and Treatment of Cancer (1); patients with chronic obstructive pulmonary disease (COPD) fulfilled Bulpa's criteria (7). The isolates were obtained from pulmonary (n = 82), extrapulmonary (wound [n = 3], central nervous system [n = 1], urinary tract [n = 1], and other [n = 2]), and pulmonary and extrapulmonary (central nervous system [n = 4], and wound and central nervous system [n = 1], urinary tract and central nervous system [n = 1]) specimens. Patients with asymptomatic colonization were selected as controls; we recruited a sufficiently large number of control patients (n = 141) to obtain a similar number of samples from the two groups.

A. fumigatus was isolated in 441 samples (n = 217 [49%] from patients with aspergillosis and 224 from colonized patients) that were cultured on bacterial and mycological media; A. fumigatus colonies (n = 1,373; 705 [51%] isolates from patients with invasive aspergillosis) found on the culture plates were prospectively subcultured and stored independently. Most isolates were from the lower respiratory tract (n = 1,227; 89%). All the isolates were identified to the molecular level by sequencing the β-tubulin gene and were genotyped using the STRAf assay (8, 9). Genotypes were considered identical when they showed the same alleles for all 9 markers. A cluster was defined as different isolates with an identical genotype. Genotypic diversity was represented graphically using a minimum spanning tree. Simpson's index of diversity (D) was calculated to study the genotypic diversity of A. fumigatus overall and by groups of patients with and without invasive aspergillosis (10).

This study was approved by the local ethics committee (Comité Ético de Investigación Clínica del Hospital Gregorio Marañón [CEIC-A1]). All patient data were anonymized.

Overall, 395 genotypes were detected in the 236 patients (Fig. 1); 50% of the patients yielded ≥2 genotypes (range, 2 to 7) (Fig. 2). The overall genetic diversity and that found in the groups of the patients with and without invasive aspergillosis was 0.995, 0.988, and 0.992, respectively. Most of the genotypes (n = 345; 87%) were found in 1 patient each, and 153 were from patients with invasive aspergillosis, whereas 192 were from colonized patients. The remaining genotypes (n = 50; 13%) were clusters affecting 95 of the 236 (40%) patients (Fig. 1); the clusters involved 2 patients (37/50 clusters), 3 patients (9/50 clusters), or 4 patients (4/50 clusters). Patients in the clusters were admitted to the hospital on different dates. Some clusters involved only patients with invasive aspergillosis (12/50 genotypes) or colonized patients (14/50 genotypes), whereas the others (24/50 genotypes) involved both groups. The percentages of patients with invasive aspergillosis or colonization in the clusters were 45% and 37%, respectively.

• Open in new tab
• Download powerpoint

FIG 1

Minimum spanning tree (MST) showing the interstrain genetic relatedness of the 395 A. fumigatus sensu stricto genotypes found. Only one representative isolate per genotype and per patient was chosen to construct the MST. Circles represent different genotypes. The size of each circle represents the number of isolates belonging to the same genotype. Colors indicate the sources of the isolates. Connecting lines between the circles show the similarity between profiles; solid and bold lines indicate differences in only 1 marker, a solid line indicates differences in 2 markers, long dashes indicate differences in 3 markers, and short dashes indicate differences in 4 or more markers.

• Open in new tab
• Download powerpoint

FIG 2

Number of patients in whom one or more genotypes were found.

In order to know whether the isolates from patients with a definitive diagnosis of invasive aspergillosis were found exclusively in infected patients, 30 genotypes found in the 16 patients with proven aspergillosis were studied. However, 4 of the 30 genotypes were also found in samples from the colonized patients. The percentages of genotypes found in patients with invasive aspergillosis, in colonized patients, and in both groups were 42%, 52%, and 6%, respectively. The percentages of clusters did not change considerably during the study period, regardless of the number of genotypes found in each year (Fig. 3).

• Open in new tab
• Download powerpoint

FIG 3

Distribution of the number of patients and genotypes per year. The percentage of clusters per year is also shown.

Our study shows that STRAf genotyping did not differentiate between A. fumigatus isolates from patients with invasive aspergillosis and isolates from colonized patients. The significance of the isolation of A. fumigatus in clinical samples is challenging and should be interpreted according to the clinical context of the patient (1). Such an approach is particularly problematic in patients without hematological conditions, in whom colonization by Aspergillus is frequent and the role of biomarker detection and radiology is limited (3, 7). The possibility of having laboratory tools for screening for invasive aspergillosis based exclusively on the characterization of the A. fumigatus isolates is attractive.

A. fumigatus typing has been performed to unravel the source of nosocomial outbreaks of invasive aspergillosis (11), to study the clonal patterns of infection or colonization in specific groups of patients (e.g., patients with cystic fibrosis or aspergilloma) (12), and to study the virulence of A. fumigatus isolates in a Drosophila fly model (6). The presence of variable degrees of virulence between isolates in the fly model suggested the role of virulence in the development of invasive aspergillosis. Unfortunately, we have not been able to find a study in which virulence was associated with the development of invasive aspergillosis in humans. We hypothesized that STRAf may play a role in detecting isolates found exclusively in patients with invasive aspergillosis or in colonized patients. Although we observed high genetic diversity—most STRAf genotypes were found exclusively in patients with invasive aspergillosis or colonized patients—up to 6% of the genotypes were found in both groups. Furthermore, 50% of the clusters involved both groups, and this finding did not change when only the genotypes from patients with proven invasive aspergillosis were included. Our findings suggest that a patient's underlying conditions play a major role in the development of infection. Another explanation is that isolates belonging to a specific clone can show different degrees of virulence, thus confirming the limitation of STRAf for this purpose.

The main limitation of this study is that genes specifically related to virulence that may be better markers than STRAf were not studied. Future studies using procedures with higher resolution, such as whole-genome sequencing, will help us to discern whether there is an association between genotypes and the presence of infection or whether the clusters found by STRAf typing are a consequence of homoplasy or genetic recombination. There is growing evidence of the presence of sexual reproduction in A. fumigatus (13), and this phenomenon may limit the use of STRAf and other classic typing procedures in clinical practice.

We conclude that genotyping cannot discriminate between isolates from patients with invasive aspergillosis and those from colonized patients, as 6% of the genotypes were found in each of the groups.