Taxonomic changes in Bipolaris and Curvularia affect clinical isolates

Over the past 2 decades, mycology has made an important leap towards incorporating molecular phylogeny into taxonomy and classification. Bipolaris and Curvularia species are dematiaceous moulds that have undergone significant taxonomic reclassification on the basis that neither genus was monophyletic and conidial morphology is not a reliable characteristic to differentiate these genera [1–3]. As a result, the majority of clinically important Bipolaris species have been transferred to Curvularia; in particular, B. australiensis, B. hawaiiensis, and B. spicifera are now Curvularia species (C. australiensis, C. hawaiiensis, and C. spicifera) and collectively make up the “spicifera clade” [3].

Bipolaris taxonomic changes present a reporting conundrum for clinical microbiology laboratories

While Bipolaris and Curvularia species have traditionally been differentiated based on conidial characteristics, most clinical microbiology laboratories lack the capacity to identify these isolates to the species level. Thus, without being able to differentiate Curvularia spicifera (formerly Bipolaris spicifera) from Bipolaris cynodontis, for example, based on morphology, it may be unclear how an isolate with the morphological features of a traditional small-conidium Bipolaris species (that is, having straight conidia with pseudoseptae) should be reported to clinicians. That is, should these isolates be reported as Bipolaris species or Curvularia species? And does this have clinical implications? In this Pearl, we address these questions and, importantly, provide reporting guidance for clinical microbiology laboratories.

Curvularia (Bipolaris) “spicifera clade” members dominate published human case reports

To determine those Bipolaris species recovered from human clinical specimens reported in the literature, we accessed case reports through PubMed (https://pubmed.ncbi.nlm.nih.gov/) using the following search terms: “Bipolaris,” “case,” "human,” and “report” (https://pubmed.ncbi.nlm.nih.gov/?term=Bipolaris+case+human+report [accessed 2024 March 6]). The search yielded 96 reports published between 1986 and 2023. Of these, 18 were excluded for various reasons (for example, review articles rather than case reports, not relating to fungal infection, and non-English language reports) leaving 78 for analysis. Of the remaining reports, 79% (62/78) reported the organism to the species level. The overwhelming majority of these case reports (94% [58/62]) identified members of the “spicifera clade” (B. australiensis [7 reports]; B. hawaiiensis [21 reports]; B. spicifera [29 reports]; B. hawaiiensis and B. spicifera [1 case series report]), revealing the predilection of “spicifera clade” members for human infection.

Curvularia (Bipolaris) “spicifera clade” members dominate isolates recovered from clinical specimens

A review of archived human clinical isolates handled in the National Mycology Reference Centre (Australia) that were reported as “Bipolaris species” between 2016 and 2024 revealed 23 isolates. Isolates were recovered from a wide range of specimens, including nasal and sinus specimens, respiratory aspirates, subcutaneous tissue and wound specimens, and corneal tissue. All isolates were recultured and reidentified by sequencing of the internal transcribed spacer (ITS) ribosomal DNA and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene [1], with sequence comparisons to the Mycobank and NCBI databases using the BLASTn algorithm (https://www.mycobank.org/page/Pairwise_alignment; https://blast.ncbi.nlm.nih.gov/Blast.cgi). Of the 23 isolates, 91% (21/23) were identified as Curvularia species (C. hawaiiensis [4 isolates], C. lunata [2 isolates], C. paterae [1 isolate], C. spicifera [13 isolates], and C. tribuli [1 isolate]); 74% (17/23) of these isolates are considered part of the “spicifera clade.” The 2 C. lunata isolates were originally misidentified as Bipolaris species, having conidia that were observed to lack the characteristic enlarged subterminal cell, thus more closely resembling conidia of a traditional Bipolaris. This highlights the fallibility of conidial morphology in differentiating traditional Bipolaris and Curvularia genera. Two isolates were confirmed as Bipolaris species (1 isolate each of B. cynodontis and B. yamadae). The species diversity observed among these Australian clinical isolates is similar to that observed in a study of isolates from the United States [4].

Curvularia (Bipolaris) “spicifera clade” members and other Curvularia species have comparable antifungal susceptibilities

A valid concern related to the reporting of “spicifera clade” species as Curvularia is that important therapeutic considerations could be overlooked, and thus patient management may be compromised by reporting all clinical Bipolaris isolates as “Curvularia species.” To investigate this possibility, we compared antifungal susceptibility testing data for “spicifera clade” species to other Curvularia species (C. brachyspora, C. geniculata, C. lunata, and C. pallescens). Minimum inhibitory concentration (MIC) value distributions for amphotericin B, itraconazole, posaconazole, and voriconazole are tabulated in Table 1. MIC values for Bipolaris (sensu stricto) species were not available for comparison, highlighting the relative paucity of these species in clinical specimens and their involvement in human infection. Most modal MIC values for the “spicifera clade” were the same as those of other clinically important Curvularia species or differed by only 1 doubling dilution. The majority of modal MIC values for the C. spicifera clade versus other Curvularia species are within 1 doubling dilution. Further, the MIC₅₀ and MIC₉₀ values for both groups were identical or within 1 doubling dilution, with the exception of the itraconazole MIC₉₀ value (C. spicifera clade 0.5 μg/mL; other Curvularia species, 8 μg/mL). The greater than 1 doubling dilution difference observed for the itraconazole MIC₉₀ values is a result of 9 (14% [9/65]) non-spicifera clade isolates exhibiting an itraconazole MIC value of 8 μg/mL. Overall, these data indicate antifungal MIC values for C. spicifera clade members are largely comparable to other Curvularia species.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Distribution of minimum inhibitory concentration (MIC) values of C. spicifera clade members and other Curvularia species tested in the National Mycology Reference Centre (Australia) according to the M38M51S Clinical and Laboratory Standards Institute (CLSI) broth microdilution reference method [5].

https://doi.org/10.1371/journal.ppat.1012678.t001

Support for reporting all clinical Bipolaris and Curvularia isolates as “Curvularia species”

An overwhelming majority of published Bipolaris case reports indicate a causative organism of human infection that is now classified within the genus Curvularia, and similarly, reidentification of clinical isolates that were reported as “Bipolaris species” demonstrates that most are now considered to be Curvularia species. Furthermore, there are minimal differences in antifungal MIC distribution values observed between C. spicifera clade species (former Bipolaris species) and other traditional Curvularia species. Taken together these findings provide justification for clinical microbiology laboratories to report all clinical isolates recovered from human specimens that morphologically resemble either Bipolaris or Curvularia simply as “Curvularia species.” Genus-level identification of clinically important moulds, at least in the first instance, is often sufficient for effective clinical management. Of course, isolates can be referred for definitive identification where there is a need, such as in the case of unusual morphology, or as would be required in the setting of an outbreak for epidemiological purposes. Finally, to emphasise its importance and to assist in recalling this important taxonomic and nomenclatural change, we conclude this Pearl with a haiku.

Bipolaris mourns

spicifera is transformed

Curvularia