Medical Mycology Advance Access published November 27, 2014 Medical Mycology, 2014, 00, 1–8 doi: 10.1093/mmy/myu079 Advance Access Publication Date: 0 2014 Original Paper
Original Paper
DNA barcoding of clinically relevant Cunninghamella species Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Jin Yu1,2 , G. Walther3,4 , A. D. Van Diepeningen2 , A. H. G. Gerrits Van Den Ende2 , Ruo-Yu Li1 , T. A. A. Moussa5,6 , O. A. Almaghrabi5 and G. S. De Hoog1,2,5,7,∗ 1
Research Center for Medical Mycology, Peking University Health Science Center, Department of Dermatology and Venereology, Peking University First Hospital, Beijing, China, 2 CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands, 3 Institute of Microbiology, Department of Microbiology and Molecular Biology, University of Jena, Jena, Germany, 4 Leibniz-Institute for Natural Product Research and Infection ¨ Biology – Hans-Knoll-Institute, Jena Microbial Resource Collection, Jena, Germany, 5 Biological Sciences Department, Faculty of Science (North Jeddah), King Abdulaziz University, Jeddah, Saudi Arabia, 6 Botany and Microbiology Department, Faculty of Science, Giza, Egypt and 7 Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, The Netherlands, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China, Shanghai Institute of Medical Mycology, Changzheng Hospital, Second Military Medical University, Shanghai, China, Basic Pathology Department, Federal University of Parana´ State, Curitiba, Brazil *To whom the correspondence should be addressed: G. S. De Hoog, Centraalbureau voor Schimmelcultures, KNAW Fungal Biodiversity Centre, P. O. Box 85167, 3508 AD Utrecht, The Netherlands; E-mail: [email protected] Received 11 May 2014; Revised 30 September 2014; Accepted 18 October 2014
Abstract Mucormycosis caused, in part, by representatives of the genus Cunninghamella is a severe infection with high mortality in patients with impaired immunity. Several species have been described in the literature as etiologic agents. A DNA barcoding study using ITS rDNA and tef-1α provided concordance of molecular data with conventional characters. The currently accepted Cunninghamella species were well supported in phylogenetic trees of both markers except for C. septata with ITS that clustered in the C. echinulata clade. Sequence variability was distinctly higher for the ITS than for tef-1α. Intraspecific ITS variability of some of the species exceeded that between some closely related species, but the marker remained applicable for species identification. The most variable species for both markers was C. echinulata. Cunninghamella bertholletiae is the main pathogenic species; infections by C. blakesleeana, C. echinulata, and C. elegans are highly exceptional. Key words: Cunninghamella, ITS, tef-1α, mucormycosis.
Introduction Cunninghamella is a genus in the order Mucorales covering species with distinctive morphological features. The
erect sporophores with balloon-shaped branches producing synchronous single-spored sporangiola are unique within the fungi, and as a result, Cunninghamella was formerly
C The Author 2014. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology.
All rights reserved. For permissions, please e-mail: [email protected]
1
2
development of molecular diagnostics. In Cunninghamella, only one molecular phylogenetic tree based on the rDNA Internal Transcribed Spacer (ITS) of 20 strains has been published so far [15]. The study nicely shows the suitability of the ITS for species identification in Cunninghamella, but the intraspecific variability of many species cannot be assessed because of the low number of isolates per species that have been investigated. A recent DNA barcoding study of the Mucorales with focus on the Mucoraceae [3] provided numerous ITS sequences of Cunninghamella species without a phylogenetic analysis. The sequence data of this study revealed a wide range of intraspecific ITS sequence variability within the genus Cunninghamella ranging from 0.9% in C. blakesleeana to 13.3% in C. echinulata, suggesting the involvement of more than one species in the latter. The present article therefore addresses the questions: Are the species limits especially of the clinically relevant taxa well defined in ITS phylogenies including the entire known diversity of the species? Is the topology of the ITS trees supported by a second phylogenetic marker, the partial Translation Elongation Factor-1α (tef-1α)? And does the tree topology of tef-1α contradict the single species concept of species with a high intraspecific variability such as C. echinulata? We focus on the position of clinical strains in phylogenetic trees and discuss their ecology and hypothetical infective potential.
Materials and methods Strains Thirty-three isolates of Cunninghamella were studied, comprising eight species: C. bertholletiae, C. binariae, C. blakesleeana, C. clavata, C. echinulata, C. elegans, C. homothallica, and C. vesiculosa originating from the reference collection of the Centraalbureau voor Schimmelcultures (Utrecht, The Netherlands) (Table 1); the set included nine ex-type strains. Five isolates were from clinical origins, 12 were environmental, and 16 were from unknown sources. Lyophilized strains were transferred to 5% malt extract agar (MEA; Oxoid, Basingstoke, UK) in 8 cm culture plates and incubated at 30◦ C for 3 days.
Morphology One or two abundantly sporulating strains from each species were selected to examine macro- and microscopic morphology. Slides were made using cotton blue as a mounting fluid. Sporulation was determined in cultures on Potato Dextrose Agar (PDA; Oxoid) derived from inocula
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
classified as the only member of a separate family, the Cunninghamellaceae [1]. However, DNA sequence clearly demonstrates that Cunninghamella is monophyletic and the genera Absidia, Halteromyces, and Chlamydoabsidia are the closest relatives in the extended Cunninghamellaceae [2,3]. Cunninghamella species are saprobes that are isolated from soil, decaying fruit, or rotten wood. In humans, severe disseminated infections have occasionally been observed. But during the last decades there has been a significant increase in the number of mucoralean that is correlated with growing patient populations with impaired immunity and uncontrolled diabetes. Judging from literature data, Cunninghamella predominantly infects hematological patients. Recently Gomes et al. [4] published the following data relative to the underlying clinical conditions of patients with C. bertholletiae infections, in other words, leukemia 51%, diabetes mellitus 19%, nonmalignant hematological diseases 16%, deferoxamine-based therapy 12%, solid organ transplantation 9%, and other diseases 13%. In two cases Cunninghamella infections were reported from otherwise apparently healthy patients [5,6]. Since 1990, more than 40 cases of infection by Cunninghamella species have been published [7]. In an epidemiology study of mucormycoses in Europe, Cunninghamella proved to be a relatively rare etiologic agent of the disease accounting only for about 5% of the cases [22]. However, among opportunistic Mucorales, infections caused by C. bertholletiae have the poorest prognosis [4], possibly due to limited treatment options. The antifungals of choice against mucoralean infections, in other words, amphotericin B and posaconazole, showed reduced activity against Cunninghamella strains [8]. In vitro, C. echinulata and C. blakesleeana showed higher MIC values than C. bertholletiae against AmB, posaconazole and itraconazole [9,12]. Given the severity and high mortality of these infections, and species specific differences in the antifungal susceptibility [9], the distinction of species of the genus Cunninghamella is of prime importance. Currently 12 species are accepted in Cunninghamella [10,11], which are all taken to be saprobic. Four of these have been reported to cause human infection, with C. bertholletiae being the most frequent recovered, accounting for >90% of cases [7]. Cunninghamella elegans was isolated from a lung biopsy specimen [3], and recently, C. blakesleeana [12] and C. echinulata [13] were described causing infections in two patients with neutropenia; infections by the latter species had been reported earlier by Lemmert et al. [14]. The knowledge of molecular variability within the species and species limits is of prime importance to the
Medical Mycology, 2014, Vol. 00, No. 0
Yu et al.
3
Table 1. Data and GenBank accession numbers of Cunninghamella strains studied in this paper (T – ex-type strain, NT – ex-neotype strain). Name and status
Geography
Source
ITS
tef-1α
CBS 151.80
C. bertholletiae
USA
JN205875
KJ156472
CBS 182.84 CBS 190.84
C. bertholletiae C. bertholletiae
USA USA
JN205877 JN205878
KJ156481 KJ156485
CBS 191.84 CBS 372.95 CBS 373.95 CBS 779.68 CBS 186.84
C. bertholletiae C. bertholletiae C. bertholletiae C. bertholletiae C. bertholletiae
USA China China
Human, lung, patient with leukemia Human Human, heart, autopsy with lymphosarcoma (case 3 from Hutter) Human, right tibia Forest soil Rotten log
JN205879 JN205872 JN205873 JN205874 JN205876
KJ156489 KJ156488 KJ156497 KJ156471 KJ156499
CBS 693.68
C. bertholletiae (NT of C. polymorpha) C. blakesleeana C. blakesleeana C. blakesleeana C. blakesleeana (T)
JN205871, AF254931
KJ156490
JN205868 JN205867 JN205869 NR103567, JN205865, AF254932 JN205891 JN205890 KJ183115 AF254936, JN205895, JN942997 JN205893, AF254938
KJ156483 KJ156482 KJ156478 KJ156479
CBS 177.36 CBS 433.84 CBS 782.68 CBS 133.27
USA
Yugoslavia
Human, lung, autopsy with underlying chronic myelogenous leukemia Soil
Kuwait Switzerland
CBS 362.95 CBS 100178 CBS 148.29 CBS 156.28
C. clavata C. clavata (T) C. echinulata C. echinulata var. echinulata
China
Soil
CBS 545.75
C. echinulata var. antarctica (T) C. echinulata var. verticillata C. echinulata C. echinulata
Chile
Soil
C. echinulata C. echinulata C. echinulata C. elegans C. elegans (NT) C. elegans C. elegans C. elegans C. elegans C. elegans C. homothallica (T) C. vesiculosa (T)
India
CBS 595.68 CBS 596.68 CBS 656.85 CBS 694.68 CBS 766.68 CBS 770.68 CBS 158.28 CBS 160.28 CBS 167.53 CBS 318.78 CBS 481.66 CBS 773.68 CBS 781.68 CBS 168.53 CBS 989.96
AF254937 USA Egypt
Flower, Hibiscus (Malvaceae) Soil
JN205896, JN942996
Canada Turkey Brazil Czechoslovakia
Linum usitatissimum, seed Soil under shrub Soil
Japan
Soil
JN205894 KJ183114 JN205888 AF254928 JN205882 JN205886 JN205889 JN205887 JN205885 JN205863
India
Soil of Shorea robusta forest
JN205897
KJ156473 KJ156477 KJ156476 KJ156500
KJ156492
KJ156475 KJ156480 KJ156491 KJ156493 KJ156496 KJ156484 KJ156486 KJ156470 KJ156494 KJ156502 KJ156495 KJ156487 KJ156501 KJ156498 KJ156474
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Number
4
Medical Mycology, 2014, Vol. 00, No. 0
as mycelial fragments of growing colonies as well as from fresh spores from 3-day-old cultures incubated at 24◦ C for 3 days.
reactions were performed with a BigDye Terminator Cycle Sequence Ready Reaction Kit (Applied Biosystems) and analyzed on a Prism 3730XL Sequencer (Applied Biosystems). Newly obtained sequences have been deposited in GenBank (nos KJ156470 to KJ156502) and are boldfaced in Table 1.
DNA extraction, amplification, and sequencing GenBank search, alignment, and phylogenetic reconstruction All ITS sequences of Cunninghamella species that were available in GenBank were checked for their length, quality, and uniqueness. In total, 63 ITS sequences were retained including 40 sequences of the DNA barcoding study on Mucorales [3] that have not been analyzed phylogenetically. For 11 clinical Cunninghamella strains species identification was confirmed by our sequence analyses (Table 2). For the clinical strain UAMH 11661 of C. echinulata only a LSU sequence was available. In this case species identification was confirmed by alignment using a set of LSU sequences (data not shown). ITS and tef-1α sequence data were edited in SeqMan (DNAStar-Lasergene, Madison, WI, USA) and aligned using the server version of the MAFFT program (www.ebi.ac.uk/Tools/mafft) with manual correction using the program Se-Al v. 2.0a11 [17]. Halteromyces radiatus was chosen as outgroup because it was shown to belong together with Absidia and Chlamydoabsidia to the sister clade of Cunninghamella [2, 3]. The server version of Gblocks [18] was used to exclude positions of the ITS alignment that were too heterogeneous for reliable phylogenetic analyses. Molecular phylogenies were estimated using the maximum
Table 2. Clinical strains of Cunninghamella included in the phylogenetic analyses. Strain number
Genbank number
Species
Country
Source
CBS 151.80
JN205875
C. bertholletiae
USA
CBS 182.84 CBS 186.84
JN205877 JN205876
C. bertholletiae C. bertholletiae
USA USA
CBS 190.84
JN205878
C. bertholletiae
USA
CBS 191.84 CNM-CM 3628 CNM-CM 3650 CNM-CM 6017
JN205879 JN205881 JN205880 JN989285
C. bertholletiae C. bertholletiae C. bertholletiae C. blakesleeana
USA Spain Spain Spain
178 UAMH 11661
JX661054 JX966103
C. echinulata C. echinulata
Russia USA
CNM-CM 5114
JN205884
C. elegans
Spain
human, lung, patient with leukemia human human, lung, autopsy with underlying chronic myelogenous leukemia human, heart, autopsy with lymphosarcoma human, right tibia human human human, lung, T-cell acute lymphoblastic leukemia human, sputum human, right middle meatus and left middle turbinate of the nasal cavity, acute leukemia human
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Isolates were grown for 3–5 days on MEA. DNA was extracted following the Quick CTAB extraction method [16]. Two gene regions, the rDNA ITS region and partial tef1α gene, were chosen for the sequence analysis. Primers included ITS1 and ITS4 for the ITS region and P12F (5 TAGGTACCTTCTGCTTACTAGCGTGTA-3 ) and P12R (5 -AATCTAGAGTTGGGCTTGGGTGAAG-3 ) for tef1α. Polymerase chain reaction (PCR) amplification for the ITS region was performed in a 12.5 μl reaction mixture containing 8 μl ddH2 O, 0.5 μl of 10 pmol of each primer, 1.25 μl of PCR buffer, 1.25 μl of deoxynucleotide triphosphate (1 mM), 0.5 μl of 5 U bioTaq polymerase (GC Biotech, Leiden), and 1 μl of template DNA. PCR amplification for the tef-1α gene was performed in 12.5 μl of reaction mixture containing 7.815 μl ddH2 O, 0.25 μl of 10 pmol of each primer, 1.25 μl PCR buffer, 0.625 μl dimethylsulfoxide, 1 μl of deoxynucleotide triphosphate (2.5 mM), 0.06 μl of 5 U Takara Taq polymerase (Takara Bio, Japan) and 1.25 μl of template DNA. PCR conditions were: one cycle of 5 min at 95o C, followed by 40 cycles of 1 min at 95o C, 1 min at 53o C (ITS) or 48o C (tef-1α) and 2 min at 72o C, followed by one cycle of 7 min at 72o C. The PCR products were visualized by electrophoresis on a 1% (w/v) agarose gel. Both strands of the PCR fragments were sequenced using the above-mentioned primers. Sequencing
Yu et al.
likelihood algorithm of the server version of RAxML-VIHPC v. 7.0.0 [19] as implemented on the Cipres portal. The robustness of all phylogenetic trees was assessed by bootstrap analyses with 1000 replicates. Distance matrices based on uncorrected distances and alignments of the entire genus were calculated for both loci using PAUP v. 4.0b10 (Swofford 2002) in order to access intra- and interspecific diversities.
Results
distinctly higher in ITS than in tef-1α. The highest intraspecific ITS heterogeneities were found in C. clavata and C. echinulata with 14.4% and 13.47%, respectively, and exceed interspecific distances between closely related species. Approximate clinical significance of species was evaluated on the basis of proven cases of which species identities were confirmed by sequencing. Of our set of strains studied, eleven originated from human sources (Table 2). Six of these concerned proven invasive cases: three strains were isolated from lung tissue, one from heart tissue, one from a right tibia, and one from a nasal cavity tissue. Five of the patients with deep infections had underlying hematological disorders including leukemia and lymphosarcoma, and presented with conditions of lowered immunity; the underlying disease of the sixth case is unknown. The strains causing invasive infections were identified as C. bertholletiae (four cases), C. blakesleeana (one case), and C. echinulata (one case). The remaining five strains were derived from clinical samples such as sputum but without proven infection. The clinical strains, except C. echinulata (GenBank JX966103, LSU) are marked with red strain numbers in Figures 1 and 2.
Discussion Like many other genera of Mucorales, Cunninghamella has traditionally been distinguished by its pronounced morphology [10,11]. Diagnostic characteristics have been color and texture of colonies, branching patterns of sporophores, shape and dimensions of vesicles and sporangioles, and presence and length of spines on sporangioles. In addition, homothallic vs. heterothalllic zygospore formation and maximum growth temperatures are recorded [10]. Zheng and Chen [10] were able to segregate the 12 currently accepted Cunninghamella species based on morphological traits, maximum growth temperature, and mating systems (homo- vs. heterothallism). In our study, all strains grew fast and had highly similar colony colors and textures. While some species have characteristic features, distinction of all species by microscopy requires extensive experience. In addition, some distinguishing characters such as the giant conidia of C. echinulata are not formed by all strains [20]. ITS is highly variable within the genus. In agreement with Su et al. [21] and Liu et al. [15] we found that the ITS tree topology also for our extended set of sequences is consistent with current species boundaries in Cunninghamella based on phenotypic characters (Fig. 1) except for C. septata that shared the clade with C. echinulata. Also in the tef-1α tree species are supported, and, in contrast to that of the ITS tree, the backbone of the tree is well resolved (Fig. 2) allowing conclusions on the relations among Cunninghamella
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Ten strains belonging to eight statistically supported species clades (below) were observed by culture and microscopy. The characteristics of Cunninghamella were consistent with the descriptions and key features given in the monograph of the genus [10]. Sequence variability was distinctly higher for the ITS than for tef-1α. The ITS alignment including the outgroup sequence had a length of 1160 nucleotides of which only 440 were conserved, while 615 were parsimonyinformative. The tef-1α alignment including the outgroup sequence had a length of 568 nucleotides of which 410 were conserved and of the variable ones 91 were parsimonyinformative. The currently accepted species were well supported in the trees of both markers (Figs 1, 2), except C. septata which was positioned in the C. echinulata clade of the ITS tree. The ITS1 sequence of C. septata contained a 37 bp long insertion. Different from the ITS tree, the backbone of the tef-1α tree was resolved due to a distinctly higher proportion of characters that are conserved for subgroups within Cunninghamella. Figure 2 shows phylogenetic relationships within Cunninghamella; the maximum growth temperatures of the species are taken from Zheng and Chen [10]. The two highly thermotolerant species, C. bertholletiae and C. echinulata, were found to be distantly related, suggesting that the ability to grow at temperatures above 40◦ C has evolved twice or was lost in the remaining species. The most variable species for both markers was C. echinulata. Four morphological varieties are known in this species: var. antarctica, var. echinulata, var. nodosa, and var. verticillata (syn. C. bainieri). The ex-type strains of the varieties differed in their ITS sequences and partly in their tef-1α sequences, but they did not form separate clades (Figs. 1, 2). In the ITS tree two supported subclades were observed within C. echinulata (highlighted in green and blue in Fig. 1). The blue subclade was also supported in the tef-1α tree (Fig. 2), while the green ITS subclade was not resolved in the tef-1α tree. Genetic distances among the eight species were 3.0%– 14.3% in the tef-1α gene and 9.1–50.0% in the ITS region. Intraspecific variabilities were 0–1.9% in tef-1α and 2.9%– 14.4% in ITS (data not shown); in general variability was
5
6
Medical Mycology, 2014, Vol. 00, No. 0
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Figure 1. RAxML phylogram of Cunninghamella based on the ITS region. Ambiguously aligned positions of the ITS were excluded by using Gblocks [17]. Branches with bootstrap values of 75% or higher are printed in bold. Ex-type strains of currently accepted taxa are printed in bold and designated by: T = ex-type strain, IT = ex-isotype strain, NT = ex-neotype strain. Clinical strains are marked by red strain and GenBank accession numbers.
Yu et al.
7
species. Cunninghamella polymorpha has been considered as a synonym of C. bertholletiae on morphological grounds and mating tests; our findings support this viewpoint. In the ITS tree C. septata clusters in the same clade as C. echinulata [15]. Nevertheless, Zheng and Chen [10] retained its species status because of strong differences between both species in morphology and maximum growth temperature. Its slow growth, the absence of aerial mycelium and the low maximum growth temperature rather might indicate that C. septata could be a growth-reduced mutant of C. echinulata. The 37 bp insertion in the ITS1 region suggests that a separate taxonomic entity is concerned. Cunninghamella echinulata has a very high degree of variability in ITS. CBS 148.29, morphologically assigned to var. verticillata [10], and CBS 656.85 formed a supported clade in both phylogenetic trees, suggesting the existence of an additional phylogenetic species. However, Zheng and Chen [10] reported the frequent formation of zygospores
of normal size, color, and ornamentation between var. echinulata and var. verticillata suggesting that both varieties belong to the same biological species. These data suggest the occurrence of high levels of genetic variability in mucoralean fungi. Seven out of 11 clinical strains available during our study clustered in Cunninghamella bertholletiae, and four of them were derived from proven invasive cases (Table 2). This matches with literature data where C. bertholletiae is seen as the prevalent clinical species [7]. Cunninghamella blakesleeana and C. echinulata were recently reported as causes of mucormycosis in two leukemic patients [9,11], and these identifications were confirmed by sequence data. One C. elegans strain had been isolated from human samples [3] but was not confirmed to cause disease. Compared with 42 cases caused by C. bertholletiae reviewed in literature [7], we found both sets of patients showing similar characteristics, such as hematological malignancy being the
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Figure 2. RAxML phylogram of Cunninghamella based on partial tef-1α sequences. Branches with bootstrap values of 75% or higher are printed in bold. Ex-type strains of currently accepted taxa are printed in bold and designated by: T = ex-type strain, NT = ex-neotype strain. Clinical strains are marked by red strain and GenBank accession numbers. Species specific maximum growth temperatures reported by Zheng and Chen [10] are mapped on the tree.
8
Acknowledgments We thank Ana Alastruey-Izquierdo for providing strains and for sharing unpublished data. Jin Yu was supported by project CDP11009 of the China Desk of the Royal Netherlands Academy of Sciences (KNAW).
References 1. Hesseltine CW, Ellis JJ. Mucorales. In: Ainsworth GC, Sparrow FK, eds. The Fungi: An Advanced Treatise, vol 5B. A Taxonomic Review with Keys: Basidiomycetes and Lower Fungi. New York: Academic Press, 1973: 187–217. 2. Hoffmann K, Pawłowska J, Walther G et al. The family structure of the Mucorales: a synoptic revision based on comprehensive multigene-genealogies. Persoonia 2013; 30: 57–76. 3. Walther G, Pawlowska J, Alastruey-Izquierdo A et al. DNA barcoding in Mucorales: an inventory of biodiversity. Persoonia 2013; 30: 11–47.
4. Gomes MRZ, Lewis RE, Kontoyiannis DP. Mucormycosis caused by unusual mucormycetes, non-Rhizopus, -Mucor, and Lichtheimia species. Clin Microbiol Rev 2011; 24: 411–445. 5. Zeilender S, Drenning D, Glauser FL, Bechard D. Fatal Cunninghamella bertholletiae infection in an immunocompetent patient. Chest 1990; 97: 1482–1483. 6. Jayasuriya NSS, Tilakaratne WM, Amaratunga EAPD, Ekanayake MKB. An unusual presentation of rhinofacial zygomycosis due to Cunninghamella sp. in an immunocompetent patient: a case report and literature review. Oral Diseases 2006; 12: 67–69. 7. Hsieh TT, Tseng HK, Sun PL, Wu YH, Chen GS. Disseminated zygomycosis caused by Cunninghamella bertholletiae in patient with hematological malignancy and review of published case reports. Mycopathologia 2013; 175: 99–106. 8. Vitale RG, De Hoog GS, Schwarz P et al. Antifungal susceptibility and phylogeny of opportunistic members of the order Mucorales. J Clin Microbiol 2012; 50: 66–75. 9. Pastor FJ, Ru´ız-Cendoya M, Pujol I et al. In vitro and in vivo antifungal susceptibilities of the Mucoralean fungus Cunninghamella. Antimicrob Agents Chemother 2010; 54: 4550–4555. 10. Zheng RY, Chen GQ. A monograph of Cunninghamella. Mycotaxon 2001; 80: 1–75. 11. Baijal U, Mehrotra BS. The genus Cunninghamella – a reassessment. Sydowia 1980; 33: 1–13. 12. Garc´ıa-Rodr´ıguez J, Quiles-Melero I, Humala-Barbier K et al. Isolation of Cunninghamella blakesleeana in an immunodepressed patient. Mycoses 2012; 55: 463–465. 13. Leblanc RE, Meriden Z, Sutton DA et al. Cunninghamella echinulata causing fatally invasive fungal sinusitis. Diagn Microbiol Infect Dis 2013; 76: 506–509. 14. Lemmert K, Losert H, Rickerts V et al. Identification of Cunninghamella spec. by molecular methods. Mycoses 2002; 45 Suppl. 1: 31–36. 15. Liu XY, Huang H, Zheng RY. Relationships within Cunninghamella based on sequence analysis of ITS r DNA. Mycotaxon 2001; 80: 77–95. 16. Dolatabadi S, Walther G, Gerrits van den Ende AHG, de Hoog GS. Diversity and delimitation of Rhizopus microsporus. Fungal Div 2014; 64: 145–163. 17. Rambaut A. 2002. Se–Al. http://tree.bio.ed.ac.uk/software/seal/. 18. Talavera G, Castresana J. Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. System Biol 2007; 56: 564–577. 19. Stamatakis A, Hoover P, Rougemont J. A rapid bootstrap algorithm for the RAxML web-servers. System Biol 2008; 75: 758–771. 20. Samson RA. Revision of the genus Cunninghamella. Koninkl Nederl Akad Wetensch Proc Series C 1969; 72: 322–335. 21. Su YC, Huang H, Liu XY, Zheng RY. Systematic relationship of several controversial Cunninghamella taxa inferred from sequence comparisons of ITS2 of rDNA. Mycol Res 1999; 103: 805–810. 22. Skiada A, Pagano L, Groll A et al. Zygomycosis in Europe: analysis of 230 cases accrued by the registry of the European Confederation of Medical Mycology (ECMM) Working Group on Zygomycosis between 2005 and 2007. Clin Microbiol Infect 2011; 17: 1859–1867.
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
most common underlying disease in Cunninghamella infections, while the respiratory system, especially the lung, was the most commonly involved organ. Cunninghamella bertholletiae, C. echinulata, C. intermedia, and C. multiverticillata are the Cunninghamella species that grow well at 37◦ C [10], but only the first two species have been reported to cause infections. The maximum growth temperature of C. elegans as determined by Zheng and Chen [10] ranges from 34◦ C to 36◦ C depending on the strain. The clinical strain CNM-CM5114 of C. elegans has a maximum growth temperature of 35◦ C (A. Alastruey-Izquierdo, pers. comm.) suggesting that the conditions inside the human body may affect the maximum growth temperatures of pathogenic fungi. In all four species the clinical strains cluster with the environmental ones suggesting that clinical isolates are environmental strains that cause opportunistic infections and renders a mammal-associated lifestyle unlikely. In conclusion, morphologically defined species are well recognized in ITS and tef-1α based phylogenies without contradictions between the topologies. The presence of a well-supported subclade within C. echinulata in the phylogenies of both loci suggests the inclusion of at least one additional phylogenetic species. However, formation of zygospores between subgroups rather supports the current concept of a large species with several varieties. Sequence variability was distinctly higher for the ITS than for tef-1α. Intraspecific variabilities are high, and in case of the ITS they can exceed interspecific distances between closely related species. Nevertheless, reliable molecular species identification using ITS as well as tef-1α is possible for all Cunninghamella species including the clinically relevant C. bertholletiae, C. blackesleeana, C. echinulata, and C. elegans.
Medical Mycology, 2014, Vol. 00, No. 0
Original Paper
DNA barcoding of clinically relevant Cunninghamella species Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Jin Yu1,2 , G. Walther3,4 , A. D. Van Diepeningen2 , A. H. G. Gerrits Van Den Ende2 , Ruo-Yu Li1 , T. A. A. Moussa5,6 , O. A. Almaghrabi5 and G. S. De Hoog1,2,5,7,∗ 1
Research Center for Medical Mycology, Peking University Health Science Center, Department of Dermatology and Venereology, Peking University First Hospital, Beijing, China, 2 CBS-KNAW Fungal Biodiversity Centre, Utrecht, The Netherlands, 3 Institute of Microbiology, Department of Microbiology and Molecular Biology, University of Jena, Jena, Germany, 4 Leibniz-Institute for Natural Product Research and Infection ¨ Biology – Hans-Knoll-Institute, Jena Microbial Resource Collection, Jena, Germany, 5 Biological Sciences Department, Faculty of Science (North Jeddah), King Abdulaziz University, Jeddah, Saudi Arabia, 6 Botany and Microbiology Department, Faculty of Science, Giza, Egypt and 7 Institute for Biodiversity and Ecosystem Dynamics, University of Amsterdam, Amsterdam, The Netherlands, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, China, Shanghai Institute of Medical Mycology, Changzheng Hospital, Second Military Medical University, Shanghai, China, Basic Pathology Department, Federal University of Parana´ State, Curitiba, Brazil *To whom the correspondence should be addressed: G. S. De Hoog, Centraalbureau voor Schimmelcultures, KNAW Fungal Biodiversity Centre, P. O. Box 85167, 3508 AD Utrecht, The Netherlands; E-mail: [email protected] Received 11 May 2014; Revised 30 September 2014; Accepted 18 October 2014
Abstract Mucormycosis caused, in part, by representatives of the genus Cunninghamella is a severe infection with high mortality in patients with impaired immunity. Several species have been described in the literature as etiologic agents. A DNA barcoding study using ITS rDNA and tef-1α provided concordance of molecular data with conventional characters. The currently accepted Cunninghamella species were well supported in phylogenetic trees of both markers except for C. septata with ITS that clustered in the C. echinulata clade. Sequence variability was distinctly higher for the ITS than for tef-1α. Intraspecific ITS variability of some of the species exceeded that between some closely related species, but the marker remained applicable for species identification. The most variable species for both markers was C. echinulata. Cunninghamella bertholletiae is the main pathogenic species; infections by C. blakesleeana, C. echinulata, and C. elegans are highly exceptional. Key words: Cunninghamella, ITS, tef-1α, mucormycosis.
Introduction Cunninghamella is a genus in the order Mucorales covering species with distinctive morphological features. The
erect sporophores with balloon-shaped branches producing synchronous single-spored sporangiola are unique within the fungi, and as a result, Cunninghamella was formerly
C The Author 2014. Published by Oxford University Press on behalf of The International Society for Human and Animal Mycology.
All rights reserved. For permissions, please e-mail: [email protected]
1
2
development of molecular diagnostics. In Cunninghamella, only one molecular phylogenetic tree based on the rDNA Internal Transcribed Spacer (ITS) of 20 strains has been published so far [15]. The study nicely shows the suitability of the ITS for species identification in Cunninghamella, but the intraspecific variability of many species cannot be assessed because of the low number of isolates per species that have been investigated. A recent DNA barcoding study of the Mucorales with focus on the Mucoraceae [3] provided numerous ITS sequences of Cunninghamella species without a phylogenetic analysis. The sequence data of this study revealed a wide range of intraspecific ITS sequence variability within the genus Cunninghamella ranging from 0.9% in C. blakesleeana to 13.3% in C. echinulata, suggesting the involvement of more than one species in the latter. The present article therefore addresses the questions: Are the species limits especially of the clinically relevant taxa well defined in ITS phylogenies including the entire known diversity of the species? Is the topology of the ITS trees supported by a second phylogenetic marker, the partial Translation Elongation Factor-1α (tef-1α)? And does the tree topology of tef-1α contradict the single species concept of species with a high intraspecific variability such as C. echinulata? We focus on the position of clinical strains in phylogenetic trees and discuss their ecology and hypothetical infective potential.
Materials and methods Strains Thirty-three isolates of Cunninghamella were studied, comprising eight species: C. bertholletiae, C. binariae, C. blakesleeana, C. clavata, C. echinulata, C. elegans, C. homothallica, and C. vesiculosa originating from the reference collection of the Centraalbureau voor Schimmelcultures (Utrecht, The Netherlands) (Table 1); the set included nine ex-type strains. Five isolates were from clinical origins, 12 were environmental, and 16 were from unknown sources. Lyophilized strains were transferred to 5% malt extract agar (MEA; Oxoid, Basingstoke, UK) in 8 cm culture plates and incubated at 30◦ C for 3 days.
Morphology One or two abundantly sporulating strains from each species were selected to examine macro- and microscopic morphology. Slides were made using cotton blue as a mounting fluid. Sporulation was determined in cultures on Potato Dextrose Agar (PDA; Oxoid) derived from inocula
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
classified as the only member of a separate family, the Cunninghamellaceae [1]. However, DNA sequence clearly demonstrates that Cunninghamella is monophyletic and the genera Absidia, Halteromyces, and Chlamydoabsidia are the closest relatives in the extended Cunninghamellaceae [2,3]. Cunninghamella species are saprobes that are isolated from soil, decaying fruit, or rotten wood. In humans, severe disseminated infections have occasionally been observed. But during the last decades there has been a significant increase in the number of mucoralean that is correlated with growing patient populations with impaired immunity and uncontrolled diabetes. Judging from literature data, Cunninghamella predominantly infects hematological patients. Recently Gomes et al. [4] published the following data relative to the underlying clinical conditions of patients with C. bertholletiae infections, in other words, leukemia 51%, diabetes mellitus 19%, nonmalignant hematological diseases 16%, deferoxamine-based therapy 12%, solid organ transplantation 9%, and other diseases 13%. In two cases Cunninghamella infections were reported from otherwise apparently healthy patients [5,6]. Since 1990, more than 40 cases of infection by Cunninghamella species have been published [7]. In an epidemiology study of mucormycoses in Europe, Cunninghamella proved to be a relatively rare etiologic agent of the disease accounting only for about 5% of the cases [22]. However, among opportunistic Mucorales, infections caused by C. bertholletiae have the poorest prognosis [4], possibly due to limited treatment options. The antifungals of choice against mucoralean infections, in other words, amphotericin B and posaconazole, showed reduced activity against Cunninghamella strains [8]. In vitro, C. echinulata and C. blakesleeana showed higher MIC values than C. bertholletiae against AmB, posaconazole and itraconazole [9,12]. Given the severity and high mortality of these infections, and species specific differences in the antifungal susceptibility [9], the distinction of species of the genus Cunninghamella is of prime importance. Currently 12 species are accepted in Cunninghamella [10,11], which are all taken to be saprobic. Four of these have been reported to cause human infection, with C. bertholletiae being the most frequent recovered, accounting for >90% of cases [7]. Cunninghamella elegans was isolated from a lung biopsy specimen [3], and recently, C. blakesleeana [12] and C. echinulata [13] were described causing infections in two patients with neutropenia; infections by the latter species had been reported earlier by Lemmert et al. [14]. The knowledge of molecular variability within the species and species limits is of prime importance to the
Medical Mycology, 2014, Vol. 00, No. 0
Yu et al.
3
Table 1. Data and GenBank accession numbers of Cunninghamella strains studied in this paper (T – ex-type strain, NT – ex-neotype strain). Name and status
Geography
Source
ITS
tef-1α
CBS 151.80
C. bertholletiae
USA
JN205875
KJ156472
CBS 182.84 CBS 190.84
C. bertholletiae C. bertholletiae
USA USA
JN205877 JN205878
KJ156481 KJ156485
CBS 191.84 CBS 372.95 CBS 373.95 CBS 779.68 CBS 186.84
C. bertholletiae C. bertholletiae C. bertholletiae C. bertholletiae C. bertholletiae
USA China China
Human, lung, patient with leukemia Human Human, heart, autopsy with lymphosarcoma (case 3 from Hutter) Human, right tibia Forest soil Rotten log
JN205879 JN205872 JN205873 JN205874 JN205876
KJ156489 KJ156488 KJ156497 KJ156471 KJ156499
CBS 693.68
C. bertholletiae (NT of C. polymorpha) C. blakesleeana C. blakesleeana C. blakesleeana C. blakesleeana (T)
JN205871, AF254931
KJ156490
JN205868 JN205867 JN205869 NR103567, JN205865, AF254932 JN205891 JN205890 KJ183115 AF254936, JN205895, JN942997 JN205893, AF254938
KJ156483 KJ156482 KJ156478 KJ156479
CBS 177.36 CBS 433.84 CBS 782.68 CBS 133.27
USA
Yugoslavia
Human, lung, autopsy with underlying chronic myelogenous leukemia Soil
Kuwait Switzerland
CBS 362.95 CBS 100178 CBS 148.29 CBS 156.28
C. clavata C. clavata (T) C. echinulata C. echinulata var. echinulata
China
Soil
CBS 545.75
C. echinulata var. antarctica (T) C. echinulata var. verticillata C. echinulata C. echinulata
Chile
Soil
C. echinulata C. echinulata C. echinulata C. elegans C. elegans (NT) C. elegans C. elegans C. elegans C. elegans C. elegans C. homothallica (T) C. vesiculosa (T)
India
CBS 595.68 CBS 596.68 CBS 656.85 CBS 694.68 CBS 766.68 CBS 770.68 CBS 158.28 CBS 160.28 CBS 167.53 CBS 318.78 CBS 481.66 CBS 773.68 CBS 781.68 CBS 168.53 CBS 989.96
AF254937 USA Egypt
Flower, Hibiscus (Malvaceae) Soil
JN205896, JN942996
Canada Turkey Brazil Czechoslovakia
Linum usitatissimum, seed Soil under shrub Soil
Japan
Soil
JN205894 KJ183114 JN205888 AF254928 JN205882 JN205886 JN205889 JN205887 JN205885 JN205863
India
Soil of Shorea robusta forest
JN205897
KJ156473 KJ156477 KJ156476 KJ156500
KJ156492
KJ156475 KJ156480 KJ156491 KJ156493 KJ156496 KJ156484 KJ156486 KJ156470 KJ156494 KJ156502 KJ156495 KJ156487 KJ156501 KJ156498 KJ156474
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Number
4
Medical Mycology, 2014, Vol. 00, No. 0
as mycelial fragments of growing colonies as well as from fresh spores from 3-day-old cultures incubated at 24◦ C for 3 days.
reactions were performed with a BigDye Terminator Cycle Sequence Ready Reaction Kit (Applied Biosystems) and analyzed on a Prism 3730XL Sequencer (Applied Biosystems). Newly obtained sequences have been deposited in GenBank (nos KJ156470 to KJ156502) and are boldfaced in Table 1.
DNA extraction, amplification, and sequencing GenBank search, alignment, and phylogenetic reconstruction All ITS sequences of Cunninghamella species that were available in GenBank were checked for their length, quality, and uniqueness. In total, 63 ITS sequences were retained including 40 sequences of the DNA barcoding study on Mucorales [3] that have not been analyzed phylogenetically. For 11 clinical Cunninghamella strains species identification was confirmed by our sequence analyses (Table 2). For the clinical strain UAMH 11661 of C. echinulata only a LSU sequence was available. In this case species identification was confirmed by alignment using a set of LSU sequences (data not shown). ITS and tef-1α sequence data were edited in SeqMan (DNAStar-Lasergene, Madison, WI, USA) and aligned using the server version of the MAFFT program (www.ebi.ac.uk/Tools/mafft) with manual correction using the program Se-Al v. 2.0a11 [17]. Halteromyces radiatus was chosen as outgroup because it was shown to belong together with Absidia and Chlamydoabsidia to the sister clade of Cunninghamella [2, 3]. The server version of Gblocks [18] was used to exclude positions of the ITS alignment that were too heterogeneous for reliable phylogenetic analyses. Molecular phylogenies were estimated using the maximum
Table 2. Clinical strains of Cunninghamella included in the phylogenetic analyses. Strain number
Genbank number
Species
Country
Source
CBS 151.80
JN205875
C. bertholletiae
USA
CBS 182.84 CBS 186.84
JN205877 JN205876
C. bertholletiae C. bertholletiae
USA USA
CBS 190.84
JN205878
C. bertholletiae
USA
CBS 191.84 CNM-CM 3628 CNM-CM 3650 CNM-CM 6017
JN205879 JN205881 JN205880 JN989285
C. bertholletiae C. bertholletiae C. bertholletiae C. blakesleeana
USA Spain Spain Spain
178 UAMH 11661
JX661054 JX966103
C. echinulata C. echinulata
Russia USA
CNM-CM 5114
JN205884
C. elegans
Spain
human, lung, patient with leukemia human human, lung, autopsy with underlying chronic myelogenous leukemia human, heart, autopsy with lymphosarcoma human, right tibia human human human, lung, T-cell acute lymphoblastic leukemia human, sputum human, right middle meatus and left middle turbinate of the nasal cavity, acute leukemia human
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Isolates were grown for 3–5 days on MEA. DNA was extracted following the Quick CTAB extraction method [16]. Two gene regions, the rDNA ITS region and partial tef1α gene, were chosen for the sequence analysis. Primers included ITS1 and ITS4 for the ITS region and P12F (5 TAGGTACCTTCTGCTTACTAGCGTGTA-3 ) and P12R (5 -AATCTAGAGTTGGGCTTGGGTGAAG-3 ) for tef1α. Polymerase chain reaction (PCR) amplification for the ITS region was performed in a 12.5 μl reaction mixture containing 8 μl ddH2 O, 0.5 μl of 10 pmol of each primer, 1.25 μl of PCR buffer, 1.25 μl of deoxynucleotide triphosphate (1 mM), 0.5 μl of 5 U bioTaq polymerase (GC Biotech, Leiden), and 1 μl of template DNA. PCR amplification for the tef-1α gene was performed in 12.5 μl of reaction mixture containing 7.815 μl ddH2 O, 0.25 μl of 10 pmol of each primer, 1.25 μl PCR buffer, 0.625 μl dimethylsulfoxide, 1 μl of deoxynucleotide triphosphate (2.5 mM), 0.06 μl of 5 U Takara Taq polymerase (Takara Bio, Japan) and 1.25 μl of template DNA. PCR conditions were: one cycle of 5 min at 95o C, followed by 40 cycles of 1 min at 95o C, 1 min at 53o C (ITS) or 48o C (tef-1α) and 2 min at 72o C, followed by one cycle of 7 min at 72o C. The PCR products were visualized by electrophoresis on a 1% (w/v) agarose gel. Both strands of the PCR fragments were sequenced using the above-mentioned primers. Sequencing
Yu et al.
likelihood algorithm of the server version of RAxML-VIHPC v. 7.0.0 [19] as implemented on the Cipres portal. The robustness of all phylogenetic trees was assessed by bootstrap analyses with 1000 replicates. Distance matrices based on uncorrected distances and alignments of the entire genus were calculated for both loci using PAUP v. 4.0b10 (Swofford 2002) in order to access intra- and interspecific diversities.
Results
distinctly higher in ITS than in tef-1α. The highest intraspecific ITS heterogeneities were found in C. clavata and C. echinulata with 14.4% and 13.47%, respectively, and exceed interspecific distances between closely related species. Approximate clinical significance of species was evaluated on the basis of proven cases of which species identities were confirmed by sequencing. Of our set of strains studied, eleven originated from human sources (Table 2). Six of these concerned proven invasive cases: three strains were isolated from lung tissue, one from heart tissue, one from a right tibia, and one from a nasal cavity tissue. Five of the patients with deep infections had underlying hematological disorders including leukemia and lymphosarcoma, and presented with conditions of lowered immunity; the underlying disease of the sixth case is unknown. The strains causing invasive infections were identified as C. bertholletiae (four cases), C. blakesleeana (one case), and C. echinulata (one case). The remaining five strains were derived from clinical samples such as sputum but without proven infection. The clinical strains, except C. echinulata (GenBank JX966103, LSU) are marked with red strain numbers in Figures 1 and 2.
Discussion Like many other genera of Mucorales, Cunninghamella has traditionally been distinguished by its pronounced morphology [10,11]. Diagnostic characteristics have been color and texture of colonies, branching patterns of sporophores, shape and dimensions of vesicles and sporangioles, and presence and length of spines on sporangioles. In addition, homothallic vs. heterothalllic zygospore formation and maximum growth temperatures are recorded [10]. Zheng and Chen [10] were able to segregate the 12 currently accepted Cunninghamella species based on morphological traits, maximum growth temperature, and mating systems (homo- vs. heterothallism). In our study, all strains grew fast and had highly similar colony colors and textures. While some species have characteristic features, distinction of all species by microscopy requires extensive experience. In addition, some distinguishing characters such as the giant conidia of C. echinulata are not formed by all strains [20]. ITS is highly variable within the genus. In agreement with Su et al. [21] and Liu et al. [15] we found that the ITS tree topology also for our extended set of sequences is consistent with current species boundaries in Cunninghamella based on phenotypic characters (Fig. 1) except for C. septata that shared the clade with C. echinulata. Also in the tef-1α tree species are supported, and, in contrast to that of the ITS tree, the backbone of the tree is well resolved (Fig. 2) allowing conclusions on the relations among Cunninghamella
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Ten strains belonging to eight statistically supported species clades (below) were observed by culture and microscopy. The characteristics of Cunninghamella were consistent with the descriptions and key features given in the monograph of the genus [10]. Sequence variability was distinctly higher for the ITS than for tef-1α. The ITS alignment including the outgroup sequence had a length of 1160 nucleotides of which only 440 were conserved, while 615 were parsimonyinformative. The tef-1α alignment including the outgroup sequence had a length of 568 nucleotides of which 410 were conserved and of the variable ones 91 were parsimonyinformative. The currently accepted species were well supported in the trees of both markers (Figs 1, 2), except C. septata which was positioned in the C. echinulata clade of the ITS tree. The ITS1 sequence of C. septata contained a 37 bp long insertion. Different from the ITS tree, the backbone of the tef-1α tree was resolved due to a distinctly higher proportion of characters that are conserved for subgroups within Cunninghamella. Figure 2 shows phylogenetic relationships within Cunninghamella; the maximum growth temperatures of the species are taken from Zheng and Chen [10]. The two highly thermotolerant species, C. bertholletiae and C. echinulata, were found to be distantly related, suggesting that the ability to grow at temperatures above 40◦ C has evolved twice or was lost in the remaining species. The most variable species for both markers was C. echinulata. Four morphological varieties are known in this species: var. antarctica, var. echinulata, var. nodosa, and var. verticillata (syn. C. bainieri). The ex-type strains of the varieties differed in their ITS sequences and partly in their tef-1α sequences, but they did not form separate clades (Figs. 1, 2). In the ITS tree two supported subclades were observed within C. echinulata (highlighted in green and blue in Fig. 1). The blue subclade was also supported in the tef-1α tree (Fig. 2), while the green ITS subclade was not resolved in the tef-1α tree. Genetic distances among the eight species were 3.0%– 14.3% in the tef-1α gene and 9.1–50.0% in the ITS region. Intraspecific variabilities were 0–1.9% in tef-1α and 2.9%– 14.4% in ITS (data not shown); in general variability was
5
6
Medical Mycology, 2014, Vol. 00, No. 0
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Figure 1. RAxML phylogram of Cunninghamella based on the ITS region. Ambiguously aligned positions of the ITS were excluded by using Gblocks [17]. Branches with bootstrap values of 75% or higher are printed in bold. Ex-type strains of currently accepted taxa are printed in bold and designated by: T = ex-type strain, IT = ex-isotype strain, NT = ex-neotype strain. Clinical strains are marked by red strain and GenBank accession numbers.
Yu et al.
7
species. Cunninghamella polymorpha has been considered as a synonym of C. bertholletiae on morphological grounds and mating tests; our findings support this viewpoint. In the ITS tree C. septata clusters in the same clade as C. echinulata [15]. Nevertheless, Zheng and Chen [10] retained its species status because of strong differences between both species in morphology and maximum growth temperature. Its slow growth, the absence of aerial mycelium and the low maximum growth temperature rather might indicate that C. septata could be a growth-reduced mutant of C. echinulata. The 37 bp insertion in the ITS1 region suggests that a separate taxonomic entity is concerned. Cunninghamella echinulata has a very high degree of variability in ITS. CBS 148.29, morphologically assigned to var. verticillata [10], and CBS 656.85 formed a supported clade in both phylogenetic trees, suggesting the existence of an additional phylogenetic species. However, Zheng and Chen [10] reported the frequent formation of zygospores
of normal size, color, and ornamentation between var. echinulata and var. verticillata suggesting that both varieties belong to the same biological species. These data suggest the occurrence of high levels of genetic variability in mucoralean fungi. Seven out of 11 clinical strains available during our study clustered in Cunninghamella bertholletiae, and four of them were derived from proven invasive cases (Table 2). This matches with literature data where C. bertholletiae is seen as the prevalent clinical species [7]. Cunninghamella blakesleeana and C. echinulata were recently reported as causes of mucormycosis in two leukemic patients [9,11], and these identifications were confirmed by sequence data. One C. elegans strain had been isolated from human samples [3] but was not confirmed to cause disease. Compared with 42 cases caused by C. bertholletiae reviewed in literature [7], we found both sets of patients showing similar characteristics, such as hematological malignancy being the
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
Figure 2. RAxML phylogram of Cunninghamella based on partial tef-1α sequences. Branches with bootstrap values of 75% or higher are printed in bold. Ex-type strains of currently accepted taxa are printed in bold and designated by: T = ex-type strain, NT = ex-neotype strain. Clinical strains are marked by red strain and GenBank accession numbers. Species specific maximum growth temperatures reported by Zheng and Chen [10] are mapped on the tree.
8
Acknowledgments We thank Ana Alastruey-Izquierdo for providing strains and for sharing unpublished data. Jin Yu was supported by project CDP11009 of the China Desk of the Royal Netherlands Academy of Sciences (KNAW).
References 1. Hesseltine CW, Ellis JJ. Mucorales. In: Ainsworth GC, Sparrow FK, eds. The Fungi: An Advanced Treatise, vol 5B. A Taxonomic Review with Keys: Basidiomycetes and Lower Fungi. New York: Academic Press, 1973: 187–217. 2. Hoffmann K, Pawłowska J, Walther G et al. The family structure of the Mucorales: a synoptic revision based on comprehensive multigene-genealogies. Persoonia 2013; 30: 57–76. 3. Walther G, Pawlowska J, Alastruey-Izquierdo A et al. DNA barcoding in Mucorales: an inventory of biodiversity. Persoonia 2013; 30: 11–47.
4. Gomes MRZ, Lewis RE, Kontoyiannis DP. Mucormycosis caused by unusual mucormycetes, non-Rhizopus, -Mucor, and Lichtheimia species. Clin Microbiol Rev 2011; 24: 411–445. 5. Zeilender S, Drenning D, Glauser FL, Bechard D. Fatal Cunninghamella bertholletiae infection in an immunocompetent patient. Chest 1990; 97: 1482–1483. 6. Jayasuriya NSS, Tilakaratne WM, Amaratunga EAPD, Ekanayake MKB. An unusual presentation of rhinofacial zygomycosis due to Cunninghamella sp. in an immunocompetent patient: a case report and literature review. Oral Diseases 2006; 12: 67–69. 7. Hsieh TT, Tseng HK, Sun PL, Wu YH, Chen GS. Disseminated zygomycosis caused by Cunninghamella bertholletiae in patient with hematological malignancy and review of published case reports. Mycopathologia 2013; 175: 99–106. 8. Vitale RG, De Hoog GS, Schwarz P et al. Antifungal susceptibility and phylogeny of opportunistic members of the order Mucorales. J Clin Microbiol 2012; 50: 66–75. 9. Pastor FJ, Ru´ız-Cendoya M, Pujol I et al. In vitro and in vivo antifungal susceptibilities of the Mucoralean fungus Cunninghamella. Antimicrob Agents Chemother 2010; 54: 4550–4555. 10. Zheng RY, Chen GQ. A monograph of Cunninghamella. Mycotaxon 2001; 80: 1–75. 11. Baijal U, Mehrotra BS. The genus Cunninghamella – a reassessment. Sydowia 1980; 33: 1–13. 12. Garc´ıa-Rodr´ıguez J, Quiles-Melero I, Humala-Barbier K et al. Isolation of Cunninghamella blakesleeana in an immunodepressed patient. Mycoses 2012; 55: 463–465. 13. Leblanc RE, Meriden Z, Sutton DA et al. Cunninghamella echinulata causing fatally invasive fungal sinusitis. Diagn Microbiol Infect Dis 2013; 76: 506–509. 14. Lemmert K, Losert H, Rickerts V et al. Identification of Cunninghamella spec. by molecular methods. Mycoses 2002; 45 Suppl. 1: 31–36. 15. Liu XY, Huang H, Zheng RY. Relationships within Cunninghamella based on sequence analysis of ITS r DNA. Mycotaxon 2001; 80: 77–95. 16. Dolatabadi S, Walther G, Gerrits van den Ende AHG, de Hoog GS. Diversity and delimitation of Rhizopus microsporus. Fungal Div 2014; 64: 145–163. 17. Rambaut A. 2002. Se–Al. http://tree.bio.ed.ac.uk/software/seal/. 18. Talavera G, Castresana J. Improvement of phylogenies after removing divergent and ambiguously aligned blocks from protein sequence alignments. System Biol 2007; 56: 564–577. 19. Stamatakis A, Hoover P, Rougemont J. A rapid bootstrap algorithm for the RAxML web-servers. System Biol 2008; 75: 758–771. 20. Samson RA. Revision of the genus Cunninghamella. Koninkl Nederl Akad Wetensch Proc Series C 1969; 72: 322–335. 21. Su YC, Huang H, Liu XY, Zheng RY. Systematic relationship of several controversial Cunninghamella taxa inferred from sequence comparisons of ITS2 of rDNA. Mycol Res 1999; 103: 805–810. 22. Skiada A, Pagano L, Groll A et al. Zygomycosis in Europe: analysis of 230 cases accrued by the registry of the European Confederation of Medical Mycology (ECMM) Working Group on Zygomycosis between 2005 and 2007. Clin Microbiol Infect 2011; 17: 1859–1867.
Downloaded from http://mmy.oxfordjournals.org/ at University of California, Santa Cruz on December 2, 2014
most common underlying disease in Cunninghamella infections, while the respiratory system, especially the lung, was the most commonly involved organ. Cunninghamella bertholletiae, C. echinulata, C. intermedia, and C. multiverticillata are the Cunninghamella species that grow well at 37◦ C [10], but only the first two species have been reported to cause infections. The maximum growth temperature of C. elegans as determined by Zheng and Chen [10] ranges from 34◦ C to 36◦ C depending on the strain. The clinical strain CNM-CM5114 of C. elegans has a maximum growth temperature of 35◦ C (A. Alastruey-Izquierdo, pers. comm.) suggesting that the conditions inside the human body may affect the maximum growth temperatures of pathogenic fungi. In all four species the clinical strains cluster with the environmental ones suggesting that clinical isolates are environmental strains that cause opportunistic infections and renders a mammal-associated lifestyle unlikely. In conclusion, morphologically defined species are well recognized in ITS and tef-1α based phylogenies without contradictions between the topologies. The presence of a well-supported subclade within C. echinulata in the phylogenies of both loci suggests the inclusion of at least one additional phylogenetic species. However, formation of zygospores between subgroups rather supports the current concept of a large species with several varieties. Sequence variability was distinctly higher for the ITS than for tef-1α. Intraspecific variabilities are high, and in case of the ITS they can exceed interspecific distances between closely related species. Nevertheless, reliable molecular species identification using ITS as well as tef-1α is possible for all Cunninghamella species including the clinically relevant C. bertholletiae, C. blackesleeana, C. echinulata, and C. elegans.
Medical Mycology, 2014, Vol. 00, No. 0
