Multilocus Phylogeny and Antifungal Susceptibility of Aspergillus Section Circumdati from Clinical Samples and Description of A. pseudosclerotiorum sp. nov.

ABSTRACT

A multilocus phylogenetic study was carried out to assess species identity of a set of 34 clinical isolates from Aspergillus section Circumdati from the United States and to determine their in vitro antifungal susceptibility against eight antifungal drugs. The genetic markers used were the internal transcribed spacer (ITS) region, and fragments of the beta-tubulin (BenA), calmodulin (CaM), and RNA polymerase II second largest subunit (RPB2) genes. The drugs tested were amphotericin B, itraconazole, posaconazole, voriconazole, anidulafungin, caspofungin, micafungin, and terbinafine. The most common species sampled was A. westerdijkiae (29.4%), followed by a novel species, which was described here as A. pseudosclerotiorum (23.5%). Other species identified were A. sclerotiorum (17.6%), A. ochraceus (8.8%), A. subramanianii (8.8%), and A. insulicola and A. ochraceopetaliformis, with two isolates (5.9%) of each. The drugs that showed the most potent activity were caspofungin, micafungin, and terbinafine, while amphotericin B showed the least activity.

INTRODUCTION

Section Circumdati includes aspergilli with biseriate conidial heads in shades of yellow to ochre, with mostly globose vesicles, and sclerotia variable in shape and color (1–3). It contains 26 species (3), with A. ochraceus being the best known and described as an important producer of many extrolites, including the mycotoxin ochratoxin A (3–5). This metabolite has nephrotoxic, immunosuppressive, teratogenic, and carcinogenic properties (6, 7) and is commonly found in coffee, rice, beverages, and other contaminated foodstuffs (3, 8). Several species in this section have been involved in different types of infections, such as onychomycosis, caused by A. insulicola, A. melleus, A. ochraceopetaliformis, A. persii, A. sclerotiorum, and A. westerdijkiae (9–14); otomycosis, caused by A. sclerotiorum (15); skin infection, caused by A. westerdijkiae (12); and pulmonary aspergillosis and osteomyelitis, caused by A. ochraceus (16, 17). Moreover, A. ochraceus, A. sclerotiorum, and A. westerdijkiae have been repetitively isolated from clinical specimens of immunocompromised patients, although in such cases their pathogenic role is uncertain (18–22).

There are few data on the in vitro antifungal susceptibility of species within section Circumdati. The azoles, especially itraconazole, appear to have good activity against A. ochraceus and A. sclerotiorum (18, 23). In contrast, amphotericin B shows limited activity against species in this section (18, 23, 24), particularly against A. westerdijkiae (25).

Identification of Aspergillus species, traditionally based on morphological and physiological aspects (2), has changed recently with the use of DNA sequencing and multilocus analyses (26). Therefore, to assess the diversity of clinically relevant species within this section, a set of isolates with features characteristic of Circumdati section were identified molecularly. These clinical isolates were recovered between 2003 and 2015 in a U.S. reference laboratory. Moreover, the antifungal susceptibility of the most frequent species was determined against eight antifungal drugs.

RESULTS

Single-gene analyses of sequences revealed similar topologies for all of them, especially for the terminal branches. The internal transcribed spacer (ITS) marker was the least informative, being unable to discriminate between closely related species. However, the most basal clades could still be discerned in the analysis of this region, providing useful data in the concatenated tree. A limitation of the concatenated analysis that included all of the species in the Circumdati section was the lack of RNA polymerase II second largest subunit (RPB2) sequences for the ex-type strains of A. affinis, A. occultus, A. pulvericola, A. salwaensis, A. sesamicola, and A. westlandensis. However, analyses of the other three markers, i.e., ITS, beta-tubulin (BenA), and calmodulin (CaM), unequivocally demonstrated that none of the strains studied here corresponded to any of the above-mentioned species.

The final concatenated sequence alignment, with 58 strains and the 4 sequenced markers, consisted of 2,451 bp (ITS, 482 bp; BenA, 470 bp; CaM, 481 bp; RPB2, 1,018 bp), of which 941 sites were variable (ITS, 85; BenA, 250; CaM, 231; RPB2, 375) and 686 parsimony informative (ITS, 57; BenA, 182; CaM, 159; RPB2, 288). Topology trees inferred by the two phylogenetic methods were basically the same, with only minor differences in the support values of the internal nodes. The ML phylogenetic tree and the bootstrap and posterior probability values (Fig. 1) show that 26 of the strains included in this study clustered with the ex-type strains of six species from section Circumdati, i.e., A. westerdijkiae (n = 10; 29.4%), A. sclerotiorum (n = 6; 17.6%), A. ochraceus (n = 3; 8.8%), A. subramanianii (n = 3; 8.6%), A. insulicola (n = 2; 5.7%), and A. ochraceopetaliformis (n = 2; 5.9%). Interestingly, a group of eight isolates (25.7%) formed a well-supported clade together with sequences of two unidentified Aspergillus strains (NRRL 35028 and NRRL 35056). This clade represents an undescribed species, proposed here as Aspergillus pseudosclerotiorum.

FIG 1.

Maximum likelihood tree obtained from analysis of combined ITS, BenA, CaM, and RPB2 data set. Branch lengths are proportional to phylogenetic distance. Bootstrap support values/Bayesian posterior probability scores over 70/0.95 are indicated on the nodes. Fully supported branches (100/1) and ex-type strains are shown in boldface. UTHSCSA, University of Texas Health Science Center (San Antonio, Texas, USA).

The isolates examined here showed typical morphology of section Circumdati and matched those of the respective species. We found, however, that identification to the species level based only on phenotypic characteristics is difficult, but combining some of the phenotypic characteristics can make this feasible (Table 1). Among the species identified here, A. westerdijkiae and A. ochraceus were the only ones with finely roughened conidia; these two species could be distinguished from each other by the lack of or only slight growth at 37°C (0 to 9 mm) for A. westerdijkiae, while A. ochraceus reached 23 to 26 mm in diameter in 7 days at the same temperature. The other species identified here had smooth-walled conidia. In addition, A. insulicola was the only species that did not produce sclerotia but did produce a reddish-brown soluble pigment on Czapek yeast autolysate agar (CYA); A. subramanianii showed good growth at 37°C (39 to 46 mm in 7 days); the colonies of A. ochraceopetaliformis had dense white mycelial areas and poor sporulation after 7 days; and A. sclerotiorum produced yellow (3A7) to brownish-orange (6C3) colonies, which reached 56 to 58 mm in diameter in 7 days on CYA, with white sclerotia, abundant sporulation, and profuse growth at 37°C (32 to 36 mm). Aspergillus pseudosclerotiorum shares similar morphological features with A. sclerotiorum but with a slightly lower growth rate at 25°C (45 to 55 diameter in 7 days) and at 37°C (22 to 38 mm), smaller metulae (3 to 9 by 2.5 to 6 μm, compared with 7 to 15 by 4 to 7 μm in A. sclerotiorum), and its sclerotia become yellow to orange yellow with age.

TABLE 1.

Key morphological features of Aspergillus section Circumdati species identified in this study

Species	Sclerotium	Metula dimensions (μm)	Conidial ornamentation	CYA colony diam (mm) in 7 days at:
				25°C	37°C
A. insulicola	Absent	6.5–12 by 3–5	Smooth	46–49	14–15
A. ochraceopetaliformis	Present	9–18 by 3.5–6	Smooth	38–46	27–29
A. ochraceus	Present	7–14 by 3–6	Finely roughened	44–49	23–26
A. pseudosclerotiorum	Present	3–9 by 2.5–6	Smooth	45–55	22–38
A. sclerotiorum	Present	7–15 by 4–7	Smooth	56–58	32–36
A. subramanianii	Present	8.5–14 by 3.5–6.5	Smooth	52–53	39–46
A. westerdijkiae	Present	8–18 by 4–7	Finely roughened	41–51	0–9

In vitro susceptibility testing showed that the drugs with the most potent activity against all of the isolates tested were caspofungin (CFG), micafungin (MFG), and terbinafine (TBF), while amphotericin B (AMB) showed the lowest activity. The azoles (itraconazole [ITC], posaconazole [PSC], and voriconazole [VRC]), showed good activity in general, with the exception of ITC against A. sclerotiorum. Interestingly, according to statistical analyses based on the Mann-Whitney test, the ITC MIC values showed significant differences between A. sclerotiorum, A. ochraceus, and A. westerdijkiae (GM of 11.31 μg/ml, 1.0 μg/ml, and 0.46 μg/ml, respectively; P < 0.05); however, differences were not significant between A. sclerotiorum and A. pseudosclerotiorum (0.89 μg/ml; P = 0.06) and A. subramanianii (4.0 μg/ml; P = 0.43). Regarding the new species, in general the drugs tested showed good activity against A. pseudosclerotiorum. Higher MIC values were observed only for AMB and VRC. Results of the in vitro susceptibility test are summarized in Table 2.

TABLE 2.

Results of in vitro antifungal susceptibility test for 30 isolates of Aspergillus section Circumdati

Species (no. of isolates) and parametera	MIC or MEC (μg/ml) forb:
	AMB	AFG	CFG	MFG	ITC	PSC	VRC	TBF
A. ochraceus (3)
GM	16.0	0.25	0.04	0.03	1.0	0.31	2.0	0.03
MIC range	16.0	0.12–0.5	0.03–0.06	0.03	1.0	0.25–0.5	2.0	0.03
Mode	16.0	0.5	0.03	0.03	1.0	0.25	2.0	0.03
A. subramanianii (3)
GM	>16.0	0.10	0.03	0.03	4.0	0.79	4.0	0.03
MIC range	16.0–>16.0	0.03–0.25	0.03	0.03	4.0	0.5–1.0	4.0	0.03
Mode	>16.0	0.25	0.03	0.03	4.0	1.0	4.0	0.03
A. sclerotiorum (6)
GM	4.76	0.03	0.04	0.03	11.31	1.0	3.36	0.03
MIC range	4.0–8.0	0.03	0.03–0.06	0.03	4.0–>16.0	1.0	2.0–4.0	0.03
Mode	4.0	0.03	0.03	0.03	>16.0	1.0	4.0	0.03
A. pseudosclerotiorum (8)
GM	5.04	0.04	0.03	0.03	0.89	0.25	1.41	0.03
MIC range	2.0–>16	0.03–0.12	0.03–0.06	0.03	0.25–>16.0	0.12–0.5	1.0–2.0	0.03
Mode	4.0	0.03	0.03	0.03	0.5	0.25	2.0	0.03
A. westerdijkiae (10)
GM	>16.0	0.14	0.03	0.03	0.46	0.29	1.08	0.03
MIC range	>16.0	0.03–1.0	0.03–0.06	0.03–0.06	0.12–1.0	0.12–0.5	1.0–2.0	0.03
Mode	>16.0	0.25	0.03	0.03	0.5	0.25	1.0	0.03
MIC90	>16.0	0.5	0.06	0.06	0.5	0.5	1.0	0.03
Total (30)
GM	12.82	0.08	0.03	0.03	1.28	0.39	1.74	0.03
MIC range	2.0–>16.0	0.03–1.0	0.03–0.06	0.03–0.06	0.12–>16.0	0.12–1.0	1.0–4.0	0.03
Mode	>16.0	0.03	0.03	0.03	0.5	0.25	1.0	0.03
MIC90	>16.0	0.5	0.06	0.03	4.0	1.0	4.0	0.03

^aGM, geometric mean.

^bAMB, amphotericin B; AFG, anidulafungin; CFG, caspofungin; MFG, micafungin; ITC, itraconazole; PSC, posaconazole; VRC, voriconazole; TBF, terbinafine; MEC, minimum effective concentration for AFG, CFG, and MFG.

Taxonomy.

Aspergillus pseudosclerotiorum J. P. Z. Siqueira, Deanna A. Sutton & Gené sp. nov. (MycoBank accession no. MB818572) (Fig. 2). Etymology: the name refers to the morphological similarity to A. sclerotiorum. Holotype: USA, Pennsylvania, isolated from lung biopsy specimen (human), D. A. Sutton, 2014 (CBS H-22808; culture ex-types: UTHSCSA DI15-13, FMR 14449, CBS 141845).

FIG 2.

Morphological features of Aspergillus pseudosclerotiorum sp. nov. (UTHSCA DI 15-13 [a to n] and UTHSCSA DI16-383 [o]). (a, b, e, and f) Front and reverse of colonies on CYA and MEA, respectively, after 7 days at 25°C. (c, d, g, and h) Front of colonies on DG18, OA, YES, and CREA, respectively, after 7 days at 25°C. (i) Enlarged view of conidial heads on CYA after 7 days at 25°C. (j) Sclerotia on CYA after 14 days at 25°C. (k) Conidia. (l) Conidiophores and a sclerotium. (m) Detail of conidiophore stipe. (n and o) Details of conidial heads. Scale bars: 10 μm (k, m, n, and o) and 100 μm (l).

Colonies on CYA at 7 days reached 45- to 55-mm diameter at 25°C; at 30°C exhibited optimum growth, reaching 55- to 64-mm diameter; at 37°C reached 22- to 38-mm diameter; and at 40°C showed restricted growth. Colonies on CYA were pale yellow (3A3) to reddish white (7A3) at the center, white toward the periphery, cottony to floccose, and usually granulose due to the presence of abundant sclerotia, margin fimbriate; reverse yellow (3A7) to greyish yellow (3B5); colorless exudates present in most isolates; little soluble pigment produced, yellow (3A6), or absent. On malt extract agar (MEA), colonies similar to CYA but with slower growth, reaching 34 to 42 mm at 7 days. On yeast extract sucrose agar (YES), colonies showed fastest growth, reaching 56 to 66 mm at 7 days, white, cottony to floccose, with abundant sclerotia; reverse yellow (3A6) to greyish yellow (4B5), sulcate; exudates abundant, colorless to yellowish white (3A2). On dichloran 18% glycerol agar (DG18), colonies reaching 28 to 34 mm at 7 days, with white to light orange (5A4) compact center, and white fluffy mycelium toward periphery; reverse yellowish white (3A2) to pale yellow (3A3); sporulation sparsely produced only with age; sclerotia absent. On oatmeal agar (OA), colonies reaching 24 to 27 mm at 7 days, yellowish white (3A2) to greyish yellow (4B4), sandy to dusty, with a more compact center, margin regular; reverse yellowish white (4A2) to greyish yellow (4B6). On creatine-sucrose agar (CREA), colonies reaching 22 to 28 mm at 7 days, white, dense at the center, sparse aerial mycelium toward the periphery; acid production absent. Micromorphology consisting of conidiophores with biseriate and radiating conidial heads; stipes septate with rough walls, subhyaline to pale brown, 120 to 980 μm long by 2.5 to 8 μm wide; vesicles mainly globose, occasionally subglobose, 7- to 31-μm diameter; metulae cylindrical, 3 to 9 by 2.5 to 6 μm, usually covering 100% of vesicle, with exception of the strain UTHSCSA DI16-383, which covered 75% of vesicle; phialides ampulliform, 4.5 to 8 by 1.25 to 3 μm; conidia globose, smooth-walled, 1.5- to 3-μm diameter; sclerotia present (except in UTHSCSA DI16-380), 150- to 507-μm diameter, white to light orange (5A4), becoming yellow (3A6) to orange yellow (4A6) with age.

DISCUSSION

In this study, we identified a total of six species in the section Circumdati from clinical samples, some of which contained a relatively large number of isolates. Although their role as etiologic agents in these cases is unknown, detection of 34 isolates of this section over a period of 12 years in a single reference center, together with some reports on infections produced by members of this section in the same period (15, 17, 18, 22, 27), highlights the importance of these fungi in the clinical setting. The degree of morphological similarity among the species of the Circumdati section, as with other groups of Aspergillus, requires DNA sequencing analysis for a definitive identification.

As was mentioned above, the most common Aspergillus species in the set of isolates studied here was A. westerdijkiae, a species described in 2004 and known to produce ochratoxin (28). It is noteworthy that the A. ochraceus strain from which ochratoxin A was discovered was later reidentified as A. westerdijkiae. This means that some isolates reported as A. ochraceus, especially the ones identified before 2004, in fact may be A. westerdijkiae (29). Growth rates at 37°C can be a useful feature to differentiate between these species without sequencing (3). Aspergillus westerdijkiae is commonly found in environmental samples (30) and as a food (31) and indoor contaminant (31–35). In the clinical setting, A. westerdijkiae has been linked to superficial infections (12) and isolated from sputum of immunocompromised patients in Tunisia (19). In our case, this species was mainly identified from respiratory specimens but also from a nail and in a sample from a marine animal (Table 3).

TABLE 3.

List of Aspergillus section Circumdati species, their isolate information, sequences generated in this study, and those retrieved from GenBank^d

Species	Isolate no.a	Originb	Yr	GenBank/EMBL accession no.c
				ITS	BenA	CaM	RPB2
A. affinis	ATCC MYA-4773T			GU721090	GU721092	GU721091
A. auricomus	NRRL 391 T			EF661411	EF661320	EF661379	EF661301
A. bridgeri	NRRL 13000 T			EF661404	EF661335	EF661358	EF661290
A. cretensis	NRRL 35672 T			FJ491572	AY819977	FJ491534	EF661311
A. elegans	NRRL 4850 T			EF661414	EF661349	EF661390	EF661316
A. fresenii	NRRL 407 T			EF661409	EF661341	EF661382	EF661296
A. insulicola	NRRL 6138 T			EF661430	EF661353	EF661396	EF661286
	UTHSCSA DI16–374	Marine	2003	LT574681	LT574716	LT574751	LT574786
	UTHSCSA DI16-402	Marine	2009	LT574682	LT574717	LT574752	LT574787
A. melleus	NRRL 5103 T			EF661425	EF661326	EF661391	EF661309
A. muricatus	NRRL 35674 T			EF661434	EF661356	EF661377	EF661314
A. neobridgeri	NRRL 13078 T			EF661410	EF661345	EF661359	EF661298
A. occultus	CBS 137330 T			KJ775443	KJ775061	KJ775239
A. ochraceopetaliformis	NRRL 4752 T			EF661429	EF661350	EF661388	EF661283
	UTHSCSA DI16-387	BAL	2006	LT574683	LT574718	LT574753	LT574788
	UTHSCSA DI16-392	Marine	2007	LT574684	LT574719	LT574754	LT574789
A. ochraceus	NRRL 398 T			EF661419	EF661322	EF661381	EF661302
	UTHSCSA DI15-10	BAL	2012	LT574686	LT574721	LT574756	LT574791
	UTHSCSA DI15-11	Heart valve	2013	LT574687	LT574722	LT574757	LT574792
	UTHSCSA DI16-384	Ear	2006	LT574685	LT574720	LT574755	LT574790
A. ostianus	NRRL 420 T			EF661421	EF661324	EF661385	EF661304
A. pallidofulvus	NRRL 4789 T			EF661423	EF661328	EF661389	EF661306
A. persii	NRRL 35669 T			FJ491580	AY819988	FJ491559	EF661295
A. pseudoelegans	CBS 112796 T			FJ491590	AY819962	FJ491552	EF661282
A. pseudosclerotiorum	NRRL 35028			EF661407	EF661343	EF661362	EF661293
	NRRL 35056			EF661405	EF661344	EF661364	EF661294
	UTHSCSA DI15-13 T	Lung biopsy	2014	LT574713	LT574748	LT574783	LT574818
	UTHSCSA DI15-14	BAL	2014	LT574714	LT574749	LT574784	LT574819
	UTHSCSA DI15-15	Lung tissue	2015	LT574715	LT574750	LT574785	LT574820
	UTHSCSA DI16-373	Sputum	2003	LT574707	LT574742	LT574777	LT574812
	UTHSCSA DI16-380	BAL	2006	LT574708	LT574743	LT574778	LT574813
	UTHSCSA DI16-383	BAL	2006	LT574709	LT574744	LT574779	LT574814
	UTHSCSA DI16-385	Sputum	2006	LT574710	LT574745	LT574780	LT574815
	UTHSCSA DI16-386	Lung mass	2006	LT574711	LT574746	LT574781	LT574816
A. pulvericola	CBS 137327 T			KJ775440	KJ775055	KJ775236
A. robustus	NRRL 6362 T			EF661176	EU014101	EF661357	EF661033
A. roseoglobulosus	NRRL 4565 T			FJ491583	AY819984	FJ491555	EF661299
A. salwaensis	DTO 297B3 T			KJ775447	KJ775056	KJ775244
A. sclerotiorum	NRRL 415 T			EF661400	EF661337	EF661384	EF661287
	UTHSCSA DI15-12	Sputum	2014	LT574693	LT574728	LT574763	LT574798
	UTHSCSA DI16-395	Sputum	2007	LT574688	LT574723	LT574758	LT574793
	UTHSCSA DI16-398	BAL	2008	LT574689	LT574724	LT574759	LT574794
	UTHSCSA DI16-404	Sputum	2009	LT574690	LT574725	LT574760	LT574795
	UTHSCSA DI16-399	BAL	2009	LT574691	LT574726	LT574761	LT574796
	UTHSCSA DI16-409	Eye	2014	LT574692	LT574727	LT574762	LT574797
A. sesamicola	CBS 137324 T			KJ775437	KJ775063	KJ775233
A. steynii	NRRL 35675 T			EF661416	EF661347	EF661378	JN121428
A. subramanianii	NRRL 6161 T			EF661403	EF661339	EF661397	EF661289
	UTHSCSA DI16-378	Lung tissue	2005	LT574694	LT574729	LT574764	LT574799
	UTHSCSA DI16-389	Wound	2006	LT574695	LT574730	LT574765	LT574800
	UTHSCSA DI16-390	Foot	2006	LT574696	LT574731	LT574766	LT574801
A. tanneri	NRRL 62425 T			JN853798	JN896582	JN896583	JN896585
A. westerdijkiae	NRRL 3174 T			EF661427	EF661329	EF661360	EF661307
	UTHSCSA DI15-5	BAL	2014	LT574703	LT574738	LT574773	LT574808
	UTHSCSA DI15-6	Sputum	2014	LT574704	LT574739	LT574774	LT574809
	UTHSCSA DI15-7	Nail	2015	LT574705	LT574740	LT574775	LT574810
	UTHSCSA DI15-8	Marine	2011	LT574706	LT574741	LT574776	LT574811
	UTHSCSA DI16-376	Unknown	2004	LT574697	LT574732	LT574767	LT574802
	UTHSCSA DI16-377	Unknown	2004	LT574698	LT574733	LT574768	LT574803
	UTHSCSA DI16-379	BAL	2005	LT574699	LT574734	LT574769	LT574804
	UTHSCSA DI16-388	Lung mass	2006	LT574700	LT574735	LT574770	LT574805
	UTHSCSA DI16-391	Lung nodule	2007	LT574701	LT574736	LT574771	LT574806
	UTHSCSA DI16-393	Sputum	2007	LT574702	LT574737	LT574772	LT574807
A. westlandensis	CBS 137321 T			KJ775434	KJ775066	KJ775230

^aATCC, American Type Culture Collection; CBS, CBS-KNAW Fungal Biodiversity Centre (Utrecht, the Netherlands); DTO, Applied and Industrial Mycology Department Collection (Utrecht, Netherlands); NRRL, Agriculture Research Service Culture Collection (Peoria, NY); UTHSCSA, University of Texas Health Science Center (San Antonio, TX). A superscript T indicates an ex-type strain.

^bBAL, bronchoalveolar lavage fluid specimens.

^cITS, internal transcribed spacer regions of the rDNA and 5.8S region; BenA, β-tubulin; CaM:, calmodulin; RPB2, partial RNA polymerase II, second largest subunit.

^dSequences generated in this study are in boldface.

It is worth noting that the second most frequently identified species in the present study was a novel one, A. pseudosclerotiorum. This species is closely related to A. bridgeri, A. persii, A. salwaensis, A. sclerotiorum, and A. subramanianii. While these species could not be discriminated from each other using the ITS-based fungal barcode, A. pseudosclerotiorum was noted to have unique sequences for the other three markers (BenA, CaM, and RPB2). Phenotypically, A. pseudosclerotiorum generally can be distinguished from the above-mentioned aspergilli by its growth rate on different media and temperatures, colony pigmentation, and degree of sporulation, as well as sclerotia and conidiophore features. Aspergillus bridgeri produces brown colonies (3, 36). A. persii grows faster on OA (35- to 38-mm diameter in 7 days) and DG18 (45- to 50-mm diameter in 7 days) (3). Aspergillus salwaensis produces a characteristic yellowish-orange soluble pigment and usually has conidiophores with vesicles flattened at the apex (3). Aspergillus subramanianii grows faster on CYA at 37°C (39- to 46-mm diameter in 7 days). Aspergillus sclerotiorum grows faster on CYA at 25°C (54- to 57-mm diameter in 7 days), and at 37°C (32- to 36-mm diameter at 37°C) it shows a higher level of sporulation and its sclerotia are white to cream colored. However, one of the eight isolates of A. pseudosclerotiorum (UTHSCSA DI16-380), which showed 99.6% similarity with the other isolates, produced atypical colonies (i.e., brownish and profusely sporulated). The size of metulae is also a diagnostic feature for A. pseudosclerotiorum, because they are smaller (3 to 9 by 2.5 to 6 μm) than those of the related species (6.5 to 10 by 3.5 to 5.5 μm in A. bridgeri, 9 to 17.5 by 4 to 7.5 μm in A. persii, 8 to 21 by 3.5 to 6 μm in A. salwaensis, 8 to 16 by 4.5 to 7 μm in A. sclerotiorum, and 9 to 14 by 4 to 6.5 μm in A. subramanianii) (3). Although all isolates of A. pseudosclerotiorum were from the human respiratory tract (i.e., BAL fluid samples, sputum, and lung tissue), further studies are needed to determine the pathogenic role of this new fungus.

The third most common species sampled was A. sclerotiorum, which has been reported to cause superficial infections, such as onychomycosis and otomycosis (10, 14, 15). Here, most of the isolates were also from the human respiratory tract. Aspergillus sclerotiorum is found worldwide, commonly isolated from soil, and reported as a species of biotechnological importance due to its ability to produce a wide range of compounds (37–39).

The best-known species in the section, A. ochraceus, was poorly represented in this study (8.8%). In contrast, it is commonly found on coffee, rice, dried fruits, and nuts (8, 40, 41) and is capable of producing different metabolites (42–44). It was reported previously in pulmonary infections based on morphological identifications (16, 20). More recently, it has been identified in a case of osteomyelitis (17) and has also been isolated from immunocompromised patients (18, 19). Carpagnano et al. often found A. ochraceus in exhaled breath condensate of lung cancer patients (27). In other mammals, it was associated with a case of otomycosis in a dog (45). Here, the three isolates were from different clinical origins (i.e., BAL fluid, ear, and heart valve).

Of the three other species identified, A. insulicola and A. ochraceopetaliformis have been reported from cases of onychomycoses (9, 12), while A. subramanianii was recovered for the first time from clinical specimens. Concerning the latter species, it is noteworthy that two isolates (UTHSCSA DI16-378 and UTHSCSA DI16-389) formed a clade slightly separate from the other A. subramanianii isolates (Fig. 1); however, the genetic identity (99.3%) with the ex-type strain and phenotypic similarity confirm their identification as A. subramanianii. This species could be considered a potential agent of human infections because of its ability to grow at 37°C and the deep-tissue origin of the isolates (lung tissue and wound).

Data available on the in vitro susceptibility of section Circumdati aspergilli against antifungal drugs are limited to a few reports with a small number of isolates tested. Here, the three echinocandins and TBF exhibited potent activity against the fungi tested. Similar results were obtained in our previous study on Aspergillus section Versicolores (46). TBF also has been reported to be highly effective in vitro against clinically relevant Aspergillus species, such as A. flavus, A. niger, A. nidulans, or A. terreus, and even against numerous isolates of A. fumigatus sensu stricto (47–49). To our knowledge, however, there is no previous information available on the activity of TBF against section Circumdati species. Results observed for echinocandins, especially MFG and anidulafungin (AFG), could be expected since, in general, they have been reported to be effective in vitro on Aspergillus species (50, 51). With respect to Circumdati aspergilli, Arabatzis et al. (18) tested three echinocandins against two isolates of A. ochraceus and one of A. sclerotiorum and reported high MICs only for CFG. In contrast, Gheith et al. (21) tested CFG against one isolate of A. ochraceus and one of A. westerdijkiae and reported low MICs, which is similar to our findings. AMB showed the least activity against the isolates tested, especially for A. ochraceus, A. subramanianii, and A. westerdijkiae. High AMB MICs were also observed for species in section Circumdati (i.e., A. melleus, A. ochraceous, and A. pallidofulvus), recently identified from human clinical specimens in India, in contrast to the results obtained in the same study for most isolates of A. fumigatus, A. flavus, and A. terreus, which were susceptible to antifungals tested there (51). PSC was the azole with the most potent activity against the strains tested, which agrees with Alastruey-Izquierdo et al. (25), Gheith et al. (21), and Masih et al. (51); however, the study of Arabatzis et al. (18) showed higher MICs for PSC. Recently, Babamahmoodi et al. (17) reported a case of osteomyelitis by A. ochraceus, for which the strain showed azole MICs (PSC, 0.032 μg/ml; VRC and ITC, 1.0 μg/ml) similar to ours (Table 2), and the patient improved after 4 months of treatment with VRC.

In conclusion, taxonomic studies are very important to assess the distribution of fungal species and their identity in clinical settings. In our study of clinical isolates within section Circumdati from a reference collection in the United States, we not only identified A. subramanianii as being associated with human specimens for the first time but also described a new taxon, A. pseudosclerotiorum, as one of the most frequent species of the section in this set of isolates. However, data from more isolates are needed to determine more reliable MICs of the different antifungal drugs against the species of this section and to determine the pathogenic role of these fungi in human and animal infections.

MATERIALS AND METHODS

Fungal isolates.

A total of 34 Aspergillus isolates received from the Fungus Testing Laboratory at the University of Texas Health Science Center (San Antonio, TX, USA) were investigated. Based on morphological features, the isolates were identified as belonging to section Circumdati. Most isolates studied were from human clinical specimens, mainly from the respiratory tract (n = 22; 64.7%), although other human clinical sources were noted as well (n = 8; 23.5%). In addition, four isolates were from marine animals (Table 3).

Morphological characterization.

The isolates were characterized morphologically by following the criteria recommended by Samson et al. (1). Briefly, colony morphology and growth rates were determined after 7 days of incubation on CYA (Becton, Dickinson and Company, Sparks, MD, USA) at 25°C and 37°C and on MEA (Pronadisa, Madrid, Spain) at 25°C. After 10 to 14 days of incubation, microscopic structures were examined and measured from MEA cultures in wet mounts with 60% lactic acid and a drop of 70% ethanol to wash out the excess conidia. A minimum of 20 of each structure was measured in order to cover all of the size ranges. Photographs were made using a Zeiss Axio Imager M1 light microscope (Zeiss, Oberkochen, Germany) with a mounted DeltaPix Infinity X digital camera using Nomarski differential interference contrast and phase-contrast optics.

DNA extraction, amplification, and sequencing.

Total genomic DNA was extracted from MEA cultures after 7 days of incubation at 25°C using the FastDNA kit and the FastPrep instrument (MP Biomedicals, Irvine, CA, USA) according to the manufacturer's specifications. Four genetic markers were amplified, i.e., the ITS region of the rRNA, which comprises ITS1, the 5.8S gene, and ITS2 regions, and fragments of the BenA, CaM, and RPB2 genes (1, 26). The primers used were ITS5 and ITS4 for the ITS region (52), Bt2a and Bt2b for BenA (53), Cmd5 and Cmd6 for CaM (54), and 5F and 7CR for RPB2 (55). PCR products were sequenced in both directions, using the same primers, at Macrogen Europe (Macrogen Inc., Amsterdam, Netherlands). Sequences were assembled and edited using SeqMan v.7.0.0 (DNASTAR, Madison, WI, USA).

Molecular identification and phylogenetic analysis.

Phylogenetic analyses were first performed individually for each gene. Since the topologies proved to be congruent with the incongruence length difference test (56), a concatenated analysis was performed. Sequences of the ex-type strains of all the species in section Circumdati were obtained from GenBank and added to the analyses. Aspergillus tanneri (section Tanneri) and A. robustus (section Robusti) were used as outgroups. In addition, GenBank sequences of two strains identified only as Aspergillus spp. (NRRL 35028 and NRRL 35026) were also added to the analyses because they formed a distinct lineage in section Circumdati (26). For multiple-sequence alignment, ClustalW was used together with MUSCLE in MEGA v.6 (57), followed by manual adjustments. The maximum likelihood (ML) analysis was conducted with MEGA v.6, as well as to estimate the best nucleotide substitution model. Support of the internal branches was assessed by the bootstrap method with 1,000 replications, where values of ≥70 were considered significant. Bayesian inference (BI) was performed using MrBayes v.3.1.2 (58). The evolutionary model that best fit each gene was assessed by MrModelTest (59). Markov chain Monte Carlo (MCMC) sampling was performed with two simultaneous runs for 1 million generations, with samples taken every 100 generations. The 50% majority rule consensus trees and posterior probability (pp) values were calculated after removing the first 25% of the resulting trees for burn-in. A pp value of ≥0.95 was considered significant.

Antifungal susceptibility testing.

Isolates of the most frequent Aspergillus species identified here were tested against eight antifungal drugs using the methods in the CLSI M38-A2 reference standard (60). The antifungal agents, obtained as pure powders, were AMB (Sigma-Aldrich Quimica S.A., Madrid, Spain), ITC (Jansen Pharmaceuticals, Beerse, Belgium), PSC (Schering-Plough Research Institut, NJ, USA), VRC (Pfizer S.A., Madrid, Spain), AFG (Pfizer S.A., Madrid, Spain), CFG (Merk & Co., Inc., Rahway, USA), MFG (Astellas Pharma, Madrid, Spain), and TBF. The MIC was defined as the lowest drug concentration that produced 100% inhibition of visible fungal growth for AMB and the azoles (ITC, PSC, and VRC) and 80% for TBF. The minimum effective concentration (MEC) was determined for the echinocandins (AFG, CFG, and MFG) and was defined microscopically as the lowest concentration of drug that permitted growth of small, rounded, compact hyphal forms, as opposed to the long, unbranched hyphal clusters that were seen in the growth control. The quality control strain Candida krusei ATCC 6258 was used in each test, and the MIC values were according to CLSI guideline ranges. All tests were carried out in duplicate, on different days, to assess reproducibility. Statistical analyses were performed using Prism software for Windows v.6.0 (GraphPad Software, San Diego, CA).

Accession number(s).

Newly generated sequences from this study were deposited in GenBank/EMBL databases under the accession numbers listed in Table 3 and in MycoBank under accession number MB818572.