Previous Article | Next Article ![]()
Journal of Clinical Microbiology, June 2004, p. 2450-2454, Vol. 42, No. 6
0095-1137/04/$08.00+0 DOI: 10.1128/JCM.42.6.2450-2454.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.
Unitat de Microbiologia, Facultat de Medicina i Ciències de la Salut, Universitat Rovira i Virgili, Reus, Spain
Received 18 June 2003/ Returned for modification 21 August 2003/ Accepted 25 February 2004
|
|
|---|
|
|
|---|
The genus Colletotrichum is one of the most important genera of plant pathogens. It has a worldwide distribution but is found mainly in subtropical and tropical regions (1). Colletotrichum species cause economically significant diseases of plants (generally referred to as anthracnoses) that affect cereals and grasses, legumes, vegetables, and perennial crops, including fruit trees (1). The key morphological features which identify the genus are its acervular conidiomata, often with setae (dark-pigmented, unbranched, thick-walled sterile hyphae usually pointed at the tip), producing elongated slimy conidia, and the presence of appressoria (thick-walled swellings at the end of a hypha or germ tube useful for attaching the fungus to the host surface before penetration of the tissue). The genus Colletotrichum encompasses numerous species, and the key criterion for their identification is based mainly on determining the plant host. In the most recent review of the genus, Sutton (23) accepted 39 species but indicated that the separation or recognition of some of these species is unclear.
Five species of Colletotrichum have been reported to cause infections in humans. They are Colletotrichum coccodes, C. crassipes, C. dematium, C. gloeosporioides, and C. graminicola (3, 4). The predominant infection is keratitis following traumatic implantation (5, 27), but subcutaneous and systemic infections among immunosuppressed patients have also been reported recently (3, 7, 14). A recent article described three cases of phaeohyphomycosis caused by Colletotrichum, involving patients who were undergoing chemotherapy for hematological malignancies (18). A sixth species, C. acutatum, recently caused a disseminated infection in a sea turtle (13).
When members of the genus Colletotrichum grow in culture, they do not usually produce their typical conidiomata, but they can be recognized easily by the presence of appressoria. However, in clinical strains, even such structures can be absent, making their identification difficult. Prompt diagnosis of the species involved in a Colletotrichum infection may be necessary to assess the correct treatment, because antifungal susceptibility depends on the species, e.g., while itraconazole was not active in vitro against C. coccodes and C. dematium, it was active against some isolates of C. gloeosporioides (7). We report a diagnostic molecular test that is suitable for quick and unambiguous recognition of any fungal strain belonging to Colletotrichum. We also review the most suitable morphological characteristics and compare the sequences of a specific fragment of the internal transcribed spacer 1 (ITS1) region of reference strains of the clinically important Colletotrichum species in order to determine whether sufficient variability exists for identification to species level. We also tested the D1-D2 domain of the large-subunit (LSU) ribosomal DNA (rDNA) gene to prove the consistency of the clades obtained by analyzing the ITS1 region.
|
|
|---|
|
View this table: [in a new window] |
TABLE 1. Strains of Colletotrichum studied genetically and/or morphologically
|
PCR assays. DNA was extracted and purified directly from fungal colonies using the Fast DNA kit (Bio101, Vista, Calif.). Fungal suspensions were vortexed with a FastPrep FP120 instrument (Thermo Savant, Holbrook, N.Y.) to disrupt the fungal cells. The reaction mixture for the PCR comprised 0.1 to 10 ng of DNA template in a total volume of 50 µl, with final concentrations of 10 mM Tris-HCl (pH 8.3), 50 mM KCl, and 2.5 mM MgCl2 (10x Perkin-Elmer buffer II plus MgCl2 solution [Roche Molecular Systems, Branchburg, N.J.]), a 1 µM concentration of each primer, 1.5 U of AmpliTaq DNA polymerase (Roche), and a 100 µM concentration of each deoxynucleoside triphospate (Promega, Madison, Wis.). The PCR conditions were denaturation for 5 min at 94°C followed by a touchdown procedure consisting of 30 s at 95°C and then annealing for 1 min at decreasing temperatures from 60 to 55°C during the first seven cycles (with 2°C decremental steps in cycles 2 to 6 and a 1°C decremental step in cycle 7). The final step was an extension at 72°C for 7 min. A total of 36 cycles were performed. Next, 5 µl of the PCR product was electrophoresed in a 2% agarose gel in the presence of ethidium bromide and visualized under UV light. A 100-bp DNA ladder marker (Invitrogen, Barcelona, Spain) was used as the size standard. The products were purified using the GFX PCR DNA purification kit (Pharmacia Biotech, Cerdanyola, Spain) and stored at 20°C until they were used in sequencing.
D1-D2 rDNA amplification. The D1-D2 regions of the LSU rDNAs from the strains in Table 1 were amplified using the NL1 and NL4 primers (17). The PCR conditions were predenaturation at 94°C for 5 min; 30 cycles at 95°C for 30 s, 51°C for 1 min, and 72°C for 1 min; and a final incubation at 72°C for 7 min. The final products were electrophoresed and purified as described above.
DNA sequencing and phylogenetic analysis. The protocol for sequencing was that of the Taq DyeDeoxy Terminator Cycle Sequencing kit (Applied Biosystems, Gouda, The Netherlands). The reactions were performed using the newly designed primers (Col-F and Col-R) and the NL1 and NL4 primers, and they were run on a model 310 DNA sequencer (Applied Biosystems). The consensus sequences were obtained using the Autoassembler program (Applied Biosystems). The sequences were aligned and adjusted as described above.
Phylogenetic analyses of the Col fragments and the D1-D2 domains were performed by the neighbor-joining method (19) with the MEGA version 2.1 computer program (11). The trees were constructed using the Kimura two-parameter distance model (10) with the pairwise deletion of gaps option. The robustness of branches was assessed by bootstrap analysis with 1,000 replicates.
|
|
|---|
![]() View larger version (35K): [in a new window] |
FIG. 1. Conidia and appressoria of Colletotrichum species of clinical interest. (A) C. coccodes; (B) C. crassipes; (C) C. dematium; (D) C. gloeosporioides; (E) C. graminicola.
|
|
View this table: [in a new window] |
TABLE 2. Most important morphological features for separating five Colletotrichum species of clinical interest
|
130 to 157 bp in length. The specificities of these primers were confirmed by the unsuccessful amplification of DNAs from the variety of human pathogenic fungi listed above and exposed under the previously described conditions. Figure 2 shows a distance tree of the region corresponding to the Col products of 4 of our strains and 11 strains from GenBank. Two main clades were observed. One of these comprised the sequences of C. coccodes (Z32930, Z32931, and Z32933), the sequence of C. crassipes (AJ565918), and two sequences of C. dematium (Z32941 and AJ565917). The other comprised the sequences of C. gloeosporioides (Z32945, Z32971, AF090855, and AJ565915) and C. graminicola (Z32972 and Z32973) and the other two sequences of C. dematium (Z32938 and AJ565916). All strains of each species tested, except those of C. dematium, formed a cluster that received high statistical support. The four sequences of C. dematium were placed in two clusters that were far away from each other. Each of them comprised one of the two sequences from GenBank and others sequenced by us. The results were similar in the phylogenetic tree obtained by analyzing the sequences of the moderately variable D1-D2 domain of the LSU rDNA gene region (Fig. 3). In that case, the species tested were also clearly separated.
![]() View larger version (26K): [in a new window] |
FIG. 2. Unrooted neighbor-joining tree of 15 sequences of Col products of rDNA from Colletotrichum species. The branch lengths are proportional to genetic distance, which is indicated by the bar. Bootstrap values (percentages of 1,000 replications) are indicated at the internodes.
|
![]() View larger version (18K): [in a new window] |
FIG. 3. Unrooted neighbor-joining tree of 10 sequences of D1-D2 regions of LSU rDNAs from Colletotrichum species. The branch lengths are proportional to genetic distance, which is indicated by the bar. Bootstrap values (percentages of 1,000 replications) are indicated at the internodes.
|
|
|
|---|
900 species have been included in the genus (23). For C. gloeosporioides alone, the most common species of the genus,
600 synonyms have been cited (26). When Colletotrichum species parasitize plants in vivo, they form intra- or subepidermal, acervular, or occasionally stromatic conidiomata with a wall of isodiametric or elongate cells generally covered or surrounded by stiff pigmented setae. In culture, as mentioned, these structures are not produced, and for an unskilled laboratorian a clinical isolate of Colletotrichum can be confused with an undetermined hyphomycete because the conidia of Colletotrichum are very different from those of any known opportunistic or pathogenic species. Fertile structures (perithecia, asci, and ascospores) of the sexual state (Glomerella) of some species of Colletotrichum, such as C. gloeosporioides and C. graminicola, can sometimes be found in vitro, but this only occurs in very old cultures and under suitable conditions (16). These structures are therefore not helpful in identification of Colletotrichum species. The most important features are those detailed in Table 2. Because of the difficulties involved in recognizing isolates of Colletotrichum, especially somewhat atypical strains, and in identifying them to the species level, the simple molecular technique that we have developed offers an alternative method for confirming and identifying clinical strains. To quality control the molecular method, we tested several strains belonging to different opportunistic or pathogenic fungal species and some representatives of common laboratory contaminants, all of them hyphomycetes. The differences shown between the two C. dematium ITS1 sequences (Z32938 and Z32941) provided by GenBank (Fig. 2) have been reported by Sreenivasaprasad et al. (21). These authors considered that the sequence Z32941 belonged to C. truncatum. This species was considered valid by Sutton (23), but only as a form within C. dematium by von Arx (26). Our study demonstrated that one of the two strains that we received from the Centraalbureau voor Schimmelcultures (CBS) as C. dematium (CBS 351.73) also belonged to C. truncatum. The morphological differences between these taxa are very subtle. According to Sutton (23), they differ in the size of the conidia, but we were not able to differentiate them morphologically. The only clinical isolate of C. dematium so far reported (12) is not available, and from the description of the case, it is very difficult to ascertain to which of the two species, C. dematium or C. truncatum, it actually belongs.
For diagnostic purposes, the sequencing method developed here showed its objective value by confirming that the strain CBS 109355, recently isolated from a subcutaneous infection in Brazil and identified according to its morphological features as C. crassipes (3), does indeed belong to that species. This strain had previously been misidentified as C. gloeosporioides (L. G. M. Castro, E. Miki Ito, R. S. Nunes, E. M. Heins-Vaccari, C. da Silva Lacaz, and J. Guarro, Proc. 14th Int. Soc. Hum. Anim. Mycol., p. 264, 2000). In the D1-D2 phylogenetic tree (Fig. 3), the sequence of this strain was in the same branch as the other two reference strains of C. crassipes (CBS 159.75 and IMI 302450) and received high bootstrap support (96%). The identification of a previously described clinical strain from Brazil as C. gloeosporioides was also confirmed here (Fig. 2 and 3). The sequence of that strain (CBS 102275) nested with the other sequences of the species obtained from GenBank. In the two trees, the clades received high bootstrap support (92 and 99%, respectively).
In summary, this study shows that molecular techniques are very useful and important tools. They can be used to complement identification based only on morphological criteria, which in the diagnosis of mycoses is still the most common technique.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»