Previous Article | Next Article ![]()
Journal of Clinical Microbiology, May 2007, p. 1433-1439, Vol. 45, No. 5
0095-1137/07/$08.00+0 doi:10.1128/JCM.02219-06
Copyright © 2007, American Society for Microbiology. All Rights Reserved.

Centre National de Référence Mycologie et Antifongiques, Unité de Mycologie Moléculaire, CNRS URA3012, Institut Pasteur, 75724 Paris Cedex 15, France,1 Université Paris Descartes, Faculté de Médecine, AP-HP, Hôpital Necker-Enfants-Malades, Service des Maladies Infectieuses et Tropicales, 75743 Paris Cedex 15, France,2 Université Paris Descartes, Faculté de Médecine, AP-HP, Hôpital Européen Georges Pompidou, Unité de Parasitologie-Mycologie, 75015 Paris, France3
Received 30 October 2006/ Returned for modification 12 December 2006/ Accepted 19 February 2007
|
|
|---|
|
|
|---|
The main pathogenic species of the order Mucorales belong to the genera Rhizopus, Absidia, Mucor, and Rhizomucor. Less commonly, other species, such as Apophysomyces elegans, Cunninghamella bertholletiae, or Saksenaea vasiformis, can be responsible for human infections. A few cases due to Cokeromyces recurvatus and Syncephalastrum racemosum have also been reported (22, 23).
Zygomycetes present heterogeneous profiles in their susceptibilities to antifungal drugs like azoles or polyenes (4, 5). Amphotericin B (a polyene) and posaconazole (an azole) are currently used for the treatment of mucormycosis in humans (9, 28, 31, 35). Therefore, identification to the species level might help guide therapy in the future. Identification of agents of mucormycosis to the species level relies mainly on morphological and physiological examinations of pure cultures, which are time-consuming. The different species of Zygomycetes share similar morphological characteristics (6) which complicate identification and often make the expertise of a reference laboratory necessary. Some approaches have used molecular targets within the conserved ribosomal DNA to identify important pathogenic fungi (15). Recently, molecular tools have been evaluated for the identification of Zygomycetes. The 18S (19, 33) and the 28S (33) genes have been used, as have the internal transcribed spacer (ITS) regions (1, 25). Other approaches using the MicroSeq sequencing kit have limited reliability to correctly identify Zygomycetes to the species level (11). A rapid and easy tool for routine species identification in clinical microbiological laboratories is mandatory. Biochemical procedures are commonly used for the identification of bacteria and yeasts (2). Among these procedures, the determination of the assimilation of carbon sources (with an auxanogram) is one of the most discriminating methods for identifying yeasts to the species level. In contrast, for the identification of filamentous fungi, biochemical procedures are rarely used. Nevertheless, biochemical tests have previously been evaluated to identify some Zygomycetes. Indeed, it has been demonstrated that the differentiation of Rhizomucor species is possible based on their different patterns of assimilation of saccharose and some amino acids (32). In addition, it has been shown that the differentiation of certain clusters of Mucor species is possible by specific carbon assimilation profiles (30). Scholer and colleagues suggested that physiological profiles of carbon assimilation could even be useful for identifying some species of Absidia, Mucor, and Rhizomucor (24). However, no study has evaluated carbon assimilation profiles of human-pathogenic Zygomycetes with a large panel of isolates.
The aim of the present study was thus to evaluate carbon assimilation profiles as a tool for species identification within the order Mucorales. For this purpose, the intra- and interspecies variability was assessed using a panel of well-characterized isolates of species of clinical importance.
|
|
|---|
Determination of carbon assimilation profiles. To obtain sufficient sporulation, all strains were cultured for 7 days on Sabouraud agar slants at 28 or 35°C, depending on the optimum growth temperature of each species (6). Stock suspensions were prepared in 1 ml 0.9% NaCl and were counted in a hemacytometer. Two commercial identification kits were used, ID32C (bioMérieux, Marcy l'Etoile, France) and API 50 CH (bioMérieux). With ID32C strips, the assimilation of 29 carbon sources and two additional characteristics (resistance to cycloheximide and hydrolysis of esculin) are evaluated. With API 50 CH strips, assimilation of 48 carbon sources and the ability to hydrolyze esculin are tested. For ID32C strips, an adequate volume of spore stock suspension was transferred to 7 ml of API C medium (bioMérieux) to achieve a final spore concentration of 5 x 105 CFU/ml. The suspension was vigorously vortexed, and 135 µl was distributed into each well of the strips. For API 50 CH strips, a final spore concentration of 5 x 105 CFU/ml was prepared in 20 ml yeast nitrogen base (7.7 g/liter; Difco, Le Pont de Claix, France), containing 0.5 g/liter chloramphenicol (Fluka Chemie GmbH, Buchs, Switzerland) and 0.1% Bacto agar (Difco) and each well of the strips was inoculated with 300 µl of medium. In preliminary experiments, incubation times of 24, 48, and 72 h were tested. It was shown that for some isolates, recording results after 24 or 48 h could lead to erroneous identification because some carbon sources were positive only at 72 h (data not shown). The influence of different incubation temperatures on the assimilation profile has not been tested. The incubation times for ID32C strips and API 50 CH strips were 72 h and 96 h, respectively, at 28°C to obtain sufficient growth for all species. After incubation, the strips were read visually and growth or lack of growth was noted. Weak growth was considered a positive result. Duplicate experiments were carried out for 25% of the tested isolates. Duplicate testing was performed on randomly chosen isolates within each of the seven species for which more than two isolates were available.
Analysis of the results was done using BioloMICS software (Biological Manager for Identification, Classification and Statistics, version 7.2.5; BioAware, Hannut, Belgium). A functional analysis using an agglomerative-clustering method (with the algorithm from the unweighted-pair group method using average linkages) was performed on group isolates and on carbon assimilation results at the same time.
|
|
|---|
-methyl-D-glucoside, L-rhamnose, erythritol, glucuronate, levulinate, inositol, or L-sorbose as a sole source of carbon. Patterns of resistance to actidione and abilities to assimilate D-saccharose, L-arabinose, D-raffinose, potassium 2-ketogluconate, D-sorbitol, D-xylose, D-ribose, glycerol, palatinose, D-melibiose, D-melezitose, potassium gluconate, D-mannitol, D-lactose, and glucosamine were consistently different between species. DL-Lactate and D-maltose were assimilated only by some isolates of A. corymbifera and R. oryzae, respectively.
![]() View larger version (145K): [in a new window] |
FIG. 1. Carbon assimilation profiles for isolates of A. corymbifera (A), R. microsporus (B), and S. racemosum (C) determined with ID32C strips after incubation at 28°C for 72 h. Weak growth was considered positive. For explanations of abbreviations, see Table 1.
|
|
View this table: [in a new window] |
TABLE 1. Carbon assimilation profiles of Zygomycetes obtained with ID32C strips
|
Interspecies variability. Most species presented specific carbon assimilation profiles (Table 1). Nevertheless, R. variabilis, R. variabilis var. regularior, and M. indicus shared very similar profiles. Similar profiles were also found for M. circinelloides and M. rouxii. All other species were distinguishable by their assimilation of at least one nonvariable carbon source. In particular, the R. oryzae profile differed from that of R. microsporus by the use of L-arabinose. R. pusillus isolates differed from R. miehei by the assimilation of D-lactose. An agglomerative clustering of the tested isolates along with their carbon assimilation profiles is presented in Fig. 2A to visualize distances between species and strains. A. corymbifera was closely related to R. pusillus, S. racemosum, and R. miehei but was well separated from all other species. Interestingly, assimilation profiles of R. variabilis and R. variabilis var. regularior were closer to those of Mucor species than to R. pusillus and R. miehei.
![]() View larger version (93K): [in a new window] |
FIG. 2. Grouping of isolates and carbon assimilation results at the same time by functional analysis using an agglomerative clustering method (with the algorithm from the unweighted-pair group method using average linkages) based on results obtained with ID32C strips (A) and API 50 CH strips (B). The vertical tree shows relatedness between isolates, and the horizontal tree shows the groups of carbon sources that are positively correlated. The table between the two trees shows the results of carbon source assimilation for each isolate. Black and gray squares indicate negative and positive tests, respectively. Carbon sources that were not informative (i.e., either negative or positive for all tested isolates) were not included. CBS, Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands; CNRMA, National Reference Center for Mycoses and Antifungals, Institut Pasteur, Paris, France; IP, Pasteur Institute Collection of Fungi, Institut Pasteur, Paris, France.
|
-D-mannopyranoside, methyl-
-D-glucopyranoside, inulin, D-tagatose, D-fucose, and L-fucose). The profiles of assimilation of 19 carbon sources (glycerol, L-arabinose, D-ribose, L-xylose, D-adonitol, D-mannitol, D-sorbitol, D-lactose, D-melibiose, D-saccharose, D-melezitose, D-raffinose, glycogen, xylitol, D-turanose, D-lyxose, L-arabitol, potassium gluconate, and potassium 2-ketogluconate) were different for all species. Amygdalin was assimilated by all species, but incorporation was variable between A. corymbifera isolates. Potassium 5-ketogluconate was assimilated by none of the species except one M. circinelloides isolate. |
View this table: [in a new window] |
TABLE 2. Carbon assimilation profiles of Zygomycetes obtained with API 50 CH strips
|
Interspecies variability. Specific carbon assimilation profiles were found for most Zygomycetes growing on API 50 CH strips (Table 2). Nevertheless, A. corymbifera and R. pusillus shared similar profiles. Identical profiles were also found for R. variabilis and M. indicus and for M. circinelloides and M. rouxii. All other species were distinguishable through their different utilizations of at least one nonvariable carbon source. An agglomerative clustering of the tested isolates, excluding R. oryzae, is presented in Fig. 2B to show the above-mentioned differences between species and strains. The assimilation profile of A. corymbifera was closely related to those of R. pusillus, R. miehei, and S. racemosum. Based on assimilation profiles, R. variabilis and R. variabilis var. regularior were closer to Mucor species than to Rhizomucor species, as already observed with ID32C strips.
Overall, profiles obtained with API 50 CH strips were more discriminating than those obtained with ID32C for Mucor species, although M. circinelloides and M. rouxii were not distinguishable by either system. In contrast, A. corymbifera and R. pusillus gave different profiles with ID32C but not with API 50 CH strips. As already emphasized, it was not possible to identify R. oryzae using API 50 CH strips.
|
|
|---|
The aim of the present study was to describe the carbon assimilation of Zygomycetes on a large panel of isolates, including type strains when available, and to assess the intra- and interspecies variability of profiles determined with ID32C and API 50 CH strips. Intraspecies variability for a given species was very low, as determined with ID32C or API 50 CH strips. Only for R. oryzae was carbon assimilation on API 50 CH strips heterogeneous between isolates, preventing identification of these species. Interspecies variability was determined to be high with ID32C and API 50 CH strips, with a slightly higher number of differences in carbon assimilation profiles determined with API 50 CH strips. All species were distinguishable from the others by at least one nonvariable assimilated carbon source. Exceptions were found for rare species, particularly for M. circinelloides and M. rouxii (with ID32C and API 50 CH strips); for R. variabilis var. regularior, R. variabilis, and M. indicus (with ID32C strips); for R. variabilis and M. indicus (with API 50 CH strips); and for A. corymbifera and R. pusillus (with API 50 CH strips), which shared similar carbon assimilation profiles. Nevertheless, ID32C strips provide a simple, standardized, and reproducible tool for the species identification of Zygomycetes.
Carbon source assimilation has already been evaluated in some studies (24, 30, 32). Nevertheless, medium and growth conditions used in the different studies were variable and different from those used in the present study, which makes comparison of the results difficult. The testing conditions used in our approach are standardized in the commercialized identification systems. As already reported (24, 32), it was possible to differentiate R. pusillus and R. miehei. However, the assimilation of saccharose by R. pusillus was shown with ID32C and API 50 CH strips to be variable between strains under our testing conditions. An assimilation of saccharose by all R. pusillus strains was not found in this study, in contrast to results reported by others (24, 32). Although the intraspecies variability in carbon assimilation was reported to be very high in a previous study (30), the identification of most Mucor species was achieved here using ID32C and API 50 CH strips. For Rhizopus species, it has been shown that intraspecies variability is high and interspecies variability is low (24). Similar results were found here only for R. oryzae, while identification of R. microsporus was still possible with API 50 CH strips. With ID32C strips, the identification of both species was possible using our testing conditions. Overall, our results suggested that the identification of Zygomycetes is possible by determining their carbon assimilation profiles. For some species, a low number of isolates has been evaluated, and results should be confirmed by testing further isolates. In this study, several type strains were tested, and we propose the use of such strains (e.g., Rhizopus microsporus CBS 631.82 or Rhizomucor pusillus CBS 354.68), which are available from international collections, as quality controls to be tested concurrently with unknown isolates.
Interestingly, similarities between species uncovered by carbon assimilation here are in accordance with results obtained by comparison of ITS sequences and by phylogenetic studies based on sequences of the 18S gene, the 28S gene, the actin gene, and the EF-1
elongation factor (25, 33, 34). Clustering of M. circinelloides with M. rouxii has already been described after comparison of ITS sequences, with the two species showing a similarity of 98.5% (25). Generally, the close relationship between Mucor species has been reported in several studies and by several techniques (6, 25, 33, 34). Similarly, a closer relationship of R. variabilis var. regularior and R. variabilis to Mucor spp. than to Rhizomucor spp. was found in this study and was already reported after ITS sequencing (25) and was also reported for R. variabilis after sequencing of several genes (33, 34). The close relationship between A. corymbifera, R. pusillus, and S. racemosum found in this study has already been reported after 28S gene sequencing, and that between A. corymbifera and R. pusillus was also reported after 18S gene, actin gene, and EF-1
elongation factor sequencing (33, 34).
In conclusion, the identification of Zygomycetes to the species level is possible based on their carbon assimilation profiles. Although this method is probably less powerful than DNA sequencing, it is a simple and reliable tool for species identification in routine clinical microbiology laboratories. Evaluation of this method for the identification of more clinical isolates on a routine basis is warranted.
We are grateful to Monique Coutanson and Bernard Papierok from the Pasteur Institute Collection of Fungi for providing reference strains and to Marie-Antoinette Piens from Lyon, France, and Paul Verweij from Nijmegen, The Netherlands, for sharing some clinical isolates. Other clinical isolates were studied as part of the nationwide survey of mucormycosis in France. Members of the French Mycoses Study Group who sent isolates used in this study were as follows (in alphabetical order by city): H. Chardon (Aix en Provence), A. Tottet (Amiens), F. Le Turdu (Argenteuil), C. Duhamel (Caen), S. Bretagne (Créteil), B. Sendid (Lille), S. Ranque, L. Collet (Marseille), O. Morin (Nantes), E. Bingen, G. Buot, M. Develoux, V. Lavarde (Paris), D. Toubas (Reims), and B. Graf (Berlin, Germany).
Published ahead of print on 28 February 2007. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»