FEMS Immunology and Medical Microbiology 45 (2005) 423–428

www.fems-microbiology.org

Electrophoresis karyotype and chromosome-length polymorphism

of Histoplasma capsulatum clinical isolates from Latin America
Cristina Elena Canteros a,*, Marı́a Fernanda Zuiani a, Viviana Ritacco a,b,
Diego E. Perrotta a, Marı́a Rocı́o Reyes-Montes c, Julio Granados d,
Gerardo Zúñiga e, Maria Lucia Taylor c, Graciela Davel a
a
Departamento Micologı´a, INEI ANLIS ‘‘Dr. Carlos G. Malbrán’’, Av. Velez Sarsfield 563, 1281 Buenos Aires, Argentina
b
Consejo Nacional de Investigaciones Cientı́ficas y Técnicas (CONICET), Buenos Aires, Argentina
c
Departamento de Microbiologı´a-Parasitologı´a, Facultad de Medicina, Universidad Nacional Autónoma México, México DF, 04510, Mexico
d
Instituto Nacional de Ciencias Médicas y Nutrición ‘‘Salvador Zubirán’’, México DF, 14000, Mexico
e
Departamento de Zoologı´a, Escuela Nacional de Ciencias Biológicas, Instituto Politécnico Nacional, México DF, 11340, Mexico

Received 27 April 2005; accepted 27 May 2005

First published online 19 July 2005

Abstract

Intact chromosomes of 19 clinical isolates of Histoplasma capsulatum recently obtained in Argentina, Mexico and Guatemala and
the laboratory reference strain G186B from Panama were analyzed using pulsed-field gel electrophoresis. Chromosomal banding
patterns of the human isolates revealed 5–7 bands, ranging from 1.3 to 10 Mbp in size. Strain G186B showed five bands of approx-
imately 1.1, 2.8, 3.3, 5.4 and 9.7 Mbp. Thirteen different electrokaryotypes were identified, indicating that the genome of H. capsu-
latum varies widely in nature, as observed previously in laboratory strains. No definite association was found between
electrokaryotype and geographical or clinical source.

2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Histoplasma capsulatum; Karyotyping; Pulsed-field gel electrophoresis; Chromosomes

1. Introduction malignancy, or immunosuppressive treatment. Although

Histoplasma capsulatum is a dimorphic fungus that
has been recognized as an important worldwide patho-
gen, agent of histoplasmosis. The disease, which in some
Latin American regions should be considered a public
health threat, presents a wide diversity of clinical mani-
festations. However, in most individuals the infection is
asymptomatic or manifests itself as a self-limiting flu-
like syndrome [1]. Lately, the risk of progressive forms
of histoplasmosis has increased due to HIV infection,
*
Corresponding author. Tel./fax: +54 11 4302 5066.
E-mail address: ccanteros@anlis.gov.ar (C.E. Canteros).
the immunological status of the susceptible host is criti-
cal to define the clinical outcome of the infection, the ge-
netic traits of different fungal strains might also
determine differences in pathogenesis [2]. The knowledge
on the chromosomal band profile of clinical isolates
might therefore shed light on the role of fungal genetic
diversity in the evolution of different clinical forms of
the disease [3,4].
In the past and even now, chromosomes have been
studied by traditional methods, such as cytogenetics,
but when these methods were applied to characterize
fungal chromosomes, difficulties arose due to their small
size. In addition, these methodologies are laborious for
analyzing a great number of strains [3,4]. The separation

0928-8244/$22.00
2005 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.femsim.2005.05.015
424 C.E. Canteros et al. / FEMS Immunology and Medical Microbiology 45 (2005) 423–428

of intact chromosomes by pulsed-field gel electrophore- 2. Materials and methods

sis (PFGE) has become a well-established technique to
study the molecular karyotype of several fungal species.
Besides, the resolution of chromosomal DNA sizes by
PFGE has many other applications, including intra-spe-
cies strain identification, phylogenetic analysis, and
preparation of DNA for further genomic studies. The
technique permits to evidence chromosome-length poly-
morphism and to resolve chromosomal bands of small
size [3–5].
To our knowledge, the karyotype of H. capsulatum
has not been sufficiently explored. The only study using
PFGE techniques was reported by Steele et al. [6], and
showed variability in band size and migration among
three laboratory reference strains of H. capsulatum by
contour-clamped homogeneous electric field (CHEF)
and field inversion gel electrophoresis (FIGE). Later,
Carr and Shearer [7] used renaturation kinetics, genomic
reconstruction and flow cytometry to determine the
whole genome size, complexity and ploidy of two H.
capsulatum laboratory reference strains with a long his-
tory of in vitro manipulation.
As only a few laboratory strains have been karyo-
typed to date, the genome variability of H. capsulatum
has not been sufficiently explored. In order to enhance
the knowledge on chromosome-length polymorphism
of this pathogen, we compared the electrophoretic kary-
otypes of recent H. capsulatum human isolates from dif-
ferent bodily sites and geographical origins in Latin
America.
2.1. Isolates
Nineteen clinical isolates of H. capsulatum obtained
from 16 Latin America patients and reference strain
G186B (ATCC 26030) were converted to yeast-phase
and maintained by culturing in GYE-agar (2% glucose,
1% yeast-extract and 1.5% Bacto-agar) at 37 C in 5%
CO2 atmosphere. After primary isolation, subcultures
were kept to a minimum. The year, country and speci-
men source of isolates are described in Table 1. Species
identification was based on morphologic characteristics,
thermal dimorphism and exoantigen production [1,8].
2.2. Chromosomal DNA plugs
The method described by Perrotta et al. [9] was ap-
plied with slight modifications. After 72 h growth at
37 C and 5% CO2 in slant-cultures, yeast-cells were sus-
pended in 100 ml GYE broth and incubated at 37 C for
72 h in a reciprocal shaker at 120 rpm. Cells were har-
vested after centrifuging at 800g–10 min, washed in ster-
ile distilled water and resuspended in 230 ll of 50 mM
EDTA, pH 8.0. Per each 100 mg of yeast-cells, 6.5 ll
(84 U) of lyticase (Promega Biosciences, Inc., San Luis
Obispo, CA) and 83 ll of Trichoderma harzianum lytic
enzyme (6 lg ll 1) (Promega) were added. The yeast
cells were immobilized in 2% (wt/vol) low melting-
temperature agarose (Gibco, Grand Island Biological

Table 1
Characteristics of studied H. capsulatum clinical isolates and a reference strain
P
Patient Isolate Country of Year of Specimen Chromosomal band Number of of PFGE Karyotype
origin isolation size range (Mbp) chromosomal bands bands (Mbp) profile
1 01738 Argentina 2001 Mucose 2.7–8.5 5 24 I
01739 Argentina 2001 Blood 2.7–8.5 5 24 I
2 92590 Argentina 1992 Bone marrow 2.7–8.5 5 25 I
3 993267 Argentina 1999 ND 2.7–8.5 5 26 I
4 00205 Argentina 2000 Blood 3.0–8.5 5 26 I
00335 Argentina 2000 Skin 3.0–8.5 5 26 I
5 01870 Argentina 2001 Bone marrow 1.8–9.8 7 36 II
01869 Argentina 2001 Sputum 1.8–9.8 7 36 II
6 H.1.12.96 Guatemala 1996 Skin 2.0–8.5 6 28 III
7 993446 Argentina 1999 Skin 1.5–9.5 7 36 IV
8 EH-316 Mexico 1993 Blood 2.7–10.0 5 32 V
9 EH-317 Mexico 1992 Blood 2.7–10.0 5 32 V
10 01558 Argentina 2001 Skin 1.8–9.5 6 32 VI
11 EH-326 Mexico 1996 ND 2.7–9.7 5 28 VII
12 EH-325 Mexico 1996 Peritoneal fluid 1.3–9.7 6 29 VIII
13 EH-327 Mexico 1991 ND 3.0–10.0 5 29 IX
14 00334 Argentina 2000 CSF 2.5–9.0 5 27 X
15 EH-324 Mexico 1994 ND 3.5–9.5 5 29 XI
16 90455 Argentina 1990 Mucose 2.5–9.7 6 37 XII
Ref. strain G186B Panama 1967 ND 1.1–9.7 5 22 XIII
ND, not determined and CSF, cerebrospinal fluid.
 C.E. Canteros et al. / FEMS Immunology and Medical Microbiology 45 (2005) 423–428
Co., NY) in 125 mM EDTA, pH 8.0, and dispensed into
molds to form the plugs. Plugs were incubated 8 h at
28 C in 50 mM EDTA, pH 8.0, under gentle shaking.
Spheroplasts immobilized within plugs were disrupted
with 10 ml lysis solution (500 mM EDTA, 1% [wt/vol]
N-laurylsarcosine (Sigma Chemical Co., St. Louis,
MO) and 0.5 mg ml 1 proteinase K (Gibco), for 18 h
at 50–55 C. Plugs were then washed three times at
50 C and twice at 28 C with TE buffer (10 mM Tris–
HCl and 1 mM EDTA, pH 8.0) and stored at 4 C in
50 mM EDTA pH 8.0 until required.
2.3. Pulsed-field gel electrophoresis
Separation of chromosomal bands was performed
under PFGE conditions in a CHEF DR III apparatus
(Bio-Rad Laboratories, Richmond, CA). In preliminary
experiments, different agarose concentration, voltage,
pulse rate, and run time were assayed. The best chromo-
somal band resolution was achieved by using 0.7% aga-
rose (Pulsed-Field Agarose, Bio-Rad) in 1· TAE buffer
(40 mM Tris–HCl, 40 mM acetate and 2 mM EDTA,
pH 8.0) and 72 h electrophoresis in three block-phases
(phase 1 pulse 4500 s, 1.5 V cm 1, 24 h; phase 2 pulses
ramped from 3600 to 2700 s, 1.5 V cm 1 30 h; phase 3
pulses ramped from 2700 to 750 s, 1.8 V cm 1 18 h). In
some cases, the presence of smaller chromosomal bands
was resolved by raising agarose concentration to 1%.
Size standards were Schizosacharomyces pombe (5.7–
3.5 Mbp) and Hansenula wingeii (3.13–1.05 Mbp) (Bio-
Rad). After electrophoresis, the gel was stained with
0.5 lg ml 1 ethidium bromide (Sigma) for 30 min and
washed in distilled water for 1 h [10]. All the experiments
were performed by triplicate in different runs. Digital
images of gels were obtained with a Bio-Rad Gel Doc
1000.
2.4. Analysis
The software Kodak Digital Science ID Image Anal-
ysis (Kodak, NY) was employed to estimate chromo-
somal band sizes. The size (Mbp) of each H.
capsulatum chromosomal band was calculated based
on migration compared to that of the size standard.
The size of the whole genome of each tested H. capsula-
tum isolate was calculated by the sum of each individual
chromosomal band size, generated in PFGE conditions
[11,12].
The relatedness of strains was determined by analyz-
ing the PFGE banding patterns with the software Bio-
Numerics (Applied Maths, Kortrijk, Belgium). S.
pombe was run in the first and last lanes of each gel as
an external standard for intra and inter-experiment nor-
malization. A reference system was defined based on
band positions of this strain in the first gel entered to
the software and all subsequent gel images were normal-
425
ized according to the reference system. The Dice coeffi-
cient was used to analyze banding pattern similarities
and the unweighted pair group method with arithmetic
averages to analyze clustering. As chromosomal images
were broad and diffuse, a fairly high band position tol-
erance (2.5%) was chosen to match visual inspection.
The cophenetic correlation was calculated to evaluate
the robustness of clusters. Strains were defined to belong
to the same karyotype when showing 100% similarity
and identical number of chromosomal bands in such
analysis conditions.
3. Results and discussion
The present study is the first attempt of using PFGE
to explore the genome size, the chromosomal band num-
ber and the genetic linkage of human H. capsulatum iso-
lates obtained from different bodily sites and
geographical sources. This method had not been applied
to H. capsulatum karyotyping since the early study on
laboratory strains by Steele et al. [6]. In general, we con-
firmed the results of that report regarding intra-species
genomic variability.
Table 1 describes the characteristics of the isolates
and their karyotypes. The dendrogram, gel images and
schematic PFGE patterns of H. capsulatum clinical iso-
lates from Argentina, Mexico and Guatemala are pre-
sented in Fig. 1 together with that of the reference
strain G186B from Panama.
As reported for other fungal species, some H. capsu-
latum chromosomal bands in our study exceeded the
range size of the external reference strain S. pombe
and therefore their sizes had to be estimated tentatively
by extrapolation [11,13]. Besides, certain bands
displayed high fluorescence intensity after ethidium bro-
mide staining, which could be attributed to co-migration
of chromosomes [11,14]. By changing PFGE switching
intervals and gel concentrations as suggested by Chavez
et al. [15], we succeeded in resolving the doublets in
some cases, like strains 01869 and 01870 (see the distinct
low-size pair of bands). In other cases (the smallest band
in strains 90455, 993267) no evidence of duplication was
found under the different electrophoretic conditions
tried (data not shown). We considered these bands as
single in the analysis but the presence of doublets cannot
be totally ruled out.
According to our results, the karyotype of strain
G186B showed five chromosomal bands, being the
smallest 1.1 Mbp in size. Using similar, but not identi-
cal, PFGE techniques, Steele et al. [6] ascribed four
chromosomal bands to this strain, including an excep-
tionally small one of 0.5 Mbp. These discrepancies in re-
sults might be due to differences in experimental
conditions and analysis accuracy. They also show that
it is necessary to adhere to a rigorous standardization
426 C.E. Canteros et al. / FEMS Immunology and Medical Microbiology 45 (2005) 423–428

Fig. 1. Dendrogram, normalized gel images and schematic representation of PFGE patterns of 19 recent of H. capsulatum and the reference strain
G186B. Dendrogram constructed using Dice coefficient and the unweighted pair group method of arithmetic averages. Cophenetic values are shown
on each node. MX: Mexico, PA: Panama, GT: Guatemala, AR: Argentina. Estimated chromosome band sizes are expressed in megabase pairs
(Mbp). h Isolates from different clinical sources of the same patient.

in order to build a database with reproducible and inter-
changeable information on H. capsulatum PFGE karyo-
types. And yet, it should be stressed that the strength of
PFGE lies not in the precise assessment of chromosomal
band number but in the discrimination of whole band-
ing patterns among genetically different isolates pro-
vided electrophoretic conditions are fixed.
To our knowledge, Carr and Shearer [7] were the only
authors that assessed the whole genome size of H. cap-
sulatum and they did it in a few laboratory strains. We
estimated in 22 Mbp the whole genome size of the refer-
ence strain G186B, matching closely the size attributed
by these authors to the same strain using more accurate
techniques. This agreement supports the use of PFGE as
a tool for calculating tentative genome size.
In our study, the observed number/size of bands and
the whole genome size were not homogeneous among re-
cent human isolates, but ranged around values reported
for common laboratory strains. The origin of such
divergences among wild strains of the same species is
still unclear and might be related with chromosomal
rearrangement involving gain or loss of genomic mate-
rial in the natural environment and/or the infected host.
This hypothesis has been postulated by Carr and
Shearer for H. capsulatum [7] and already demonstrated
for other fungal species by other authors [3,4].
According to our results, at least three isolates from
Argentina showed as many as seven chromosomal
bands, the number previously attributed to the Downs
strain [6]. If total DNA mass were to be considered, se-
ven isolates had estimated genome sizes equal or larger
than the 32 Mbp previously estimated for Downs strain
[7] (see Table 1). Possessing the same genome size or
number of chromosomal bands, however, is not suffi-
cient evidence to infer genomic similarity. The ability
of PFGE of identifying Downs-like genotypes is yet to
be defined, as well as the biological relevance of such
finding.
The cophenetic correlation is a measure of the agree-
ment between the similarity represented in the dendro-
gram and the actual degree of similarity expressed by
the Dice coefficient. A minimum 70% correlation indi-
cates that the dendrogram does not greatly distort the ori-
ginal structure in the input data [16]. As indicated on each
node in Fig. 1, all the cophenetic correlations obtained in
our study were above this value, thus reassuring the
robustness of the clustering obtained (72–100%).
Thirteen distinct electrokaryotypes were observed,
supporting inter-strain chromosomal polymorphism, as
it has been described for several pathogenic fungi,
including other dimorphic species [3,4,12,17–19]. Profile
differences were more marked in chromosomal DNA
 C.E. Canteros et al. / FEMS Immunology and Medical Microbiology 45 (2005) 423–428
band sizes than in their numbers and no variation was
found between isolates obtained from different bodily
sites in a single patient (Fig. 1). We observed that certain
isolates from Argentina with the same number of chro-
mosomal bands clustered distantly, as shown in Fig. 1,
karyotypes I and X. This polymorphism was only due
to differences in band sizes. Provided that these subtle
differences are consistent and meaningful, PFGE could
become a useful tool to discriminate between strains of
H. capsulatum sharing a geographical niche.
In spite of the apparently high chromosome-length
polymorphism, three clusters of identical patterns were
identified. Karyotype I included isolates from four
Argentinean patients. Karyotype II was composed of
two isolates obtained from different bodily sites in a sin-
gle patient. In karyotype V segregated two isolates from
two different Mexican patients with no record of epide-
miological link with each other.
In this study, the electrophoretic profile of the refer-
ence strain G186B was almost 50% different from the
profiles of the clinical isolates. The divergence of its
chromosomal banding pattern from those of wild strains
suggests that this laboratory strain may have undergone
genomic rearrangements throughout its long-term his-
tory of in vitro passages. In addition, its small genome
size points towards a loss rather than a gain in genomic
material, compared with natural isolates.
It is interesting to note that all H. capsulatum isolates
from Mexico showed substantial polymorphism among
each other whereas a number of isolates from Argentina
formed two closer groups. Our results regarding the ge-
netic homogeneity of strains isolated in Argentina sup-
port previous findings by RAPD-PCR [20]. More
recently, in a phylogeographical study involving DNA se-
quence variation of protein-coding genes, Kasuga et al.
[21] found clonality among strains from Argentina and
a wider genetic diversity among strains from Mexico.
In sum, the present PFGE study indicates that the
genome of H. capsulatum varies widely. This research
also shows that the chromosomal band number and
the estimated genome sizes of clinical isolates are in
range with those previously reported for highly domesti-
cated laboratory strains. No clear relationship was
found between electrokaryotypes and bodily site or geo-
graphical source, either due to the small number of iso-
lates analyzed or to unsuitability of the method for
answering epidemiological or phylogenetic questions.
Further studies are required to establish whether the
karyotypes here described can be assigned to genuine ge-
netic linkage groups.
Acknowledgment
The study was partially supported by CONACyT
Ref. No. 34443.
427
References
[1] Tewary, R., Wheat, L.J. and Ajello, L. (1998) Agents of
histoplasmosis. In: Medical Mycology. Topley and WilsonÕs,
Microbiology and Microbial (Ajello, L. and Hay, R.J., Eds.), 9th
Edn. pp. 373–407. Arnold and Oxford University Press, New
York.
[2] Retallack, D.M. and Woods, J.P. (1999) Molecular epidemiology,
pathogenesis, and genetics of the dimorphic fungus Histoplasma
capsulatum. Microbes Infect. 1, 817–825.
[3] Zolan, M.E. (1995) Chromosome-length polymorphism in fungi.
Microbiol. Rev. 59, 686–698.
[4] Fierro, F. and Martı́n, J.F. (1999) Molecular mechanisms of
chromosomal rearrangement in fungi. Crit. Rev. Microbiol. 25,
1–17.
[5] Beadle, J., Wright, M., McNeely, L. and Bennett, J.W. (2003)
Electrophoretic karyotype analysis in fungi. Adv. Appl. Micro-
bial. 53, 243–270.
[6] Steele, P.E., Carle, G.F., Kobayashi, G.S. and Medoff, G. (1989)
Electrophoretic analyses of Histoplasma capsulatum chromosomal
DNA. Mol. Cell. Biol. 9, 983–987.
[7] Carr, J. and Shearer, G. (1998) Genome size, complexity and
ploidy of pathogenic fungus Histoplasma capsulatum. J. Bacteriol.
180, 6697–6703.
[8] Standard, P.G. and Kaufman, L. (1976) Specific immunological
test for the rapid identification of members of the genus
Histoplasma. J. Clin. Microbiol. 3, 191–199.
[9] Perrotta, D., Rodero, L., Demkura, H., Canteros, C. and Davel,
G. (2002) Cariotipos electroforéticos y análisis de fragmentos de
restricción de ADN genómico: su utilidad como herramienta en
estudios epidemiológicos de Candida parapsilosis. Rev. Argent.
Microbiol. 34, 29–39.
[10] Sambrook, J. and Russell, D.W. (2001) Gel electrophoresis of
DNA and pulse-field agarose electrophoresis. In Molecular
Cloning: A Laboratory Manual, pp. 5.55–5.67. Cold Spring
Harbor Laboratory Press, Cold Spring Harbor.
[11] Nagy, A., Palágyi, Z., Vastag, M., Ferenczy, L. and Vágvölgyi, C.
(2000) Electrophoretic karyotypes of some related Mucor species.
Antonie Van Leeuwenhoek 78, 33–37.
[12] Montoya, A.E., Alvarez, A.L., Moreno, M.N., Restrepo, A.
and McEwen, J.G. (1999) Electrophoretic karyotype of envi-
ronmental isolates of Paracoccidioides brasiliensis. Med. Mycol.
37, 219–222.
[13] Higashiyama, T. and Yamada, T. (1991) Electrophoretic karyo-
typing and chromosomal gene mapping of Chlorella. Nucleic
Acids Res. 19, 6191–6195.
[14] Rincones, J., Meinhardt, L.W., Vidal, B.C. and Pereira, G.A.
(2003) Electrophoresis karyotypes analysis of Crinipellis pernici-
osa, the causal agent of witches broon disease of Theobroma
cacao. Mycol. Res. 107, 452–458.
[15] Chavez, R., Fierro, F., Gordillo, F., Martı́n, J.F. and
Eyzaguirre, J. (2001) Electrophoretic karyotypes of the
filamentous fungus Penicillum purpurogenum and chromosomal
location of several xylanolytic genes. FEMS Microbiol. Lett.
205, 379–383.
[16] MacEllistrem, M.C., Pass, M., Elliott, J.A., Whitney, C.G. and
Harrison, L.H. (2004) Clonal grups of Penicillin-nonsusceptible
Streptococcus pneumoniae in Baltimore, Maryland: a population-
based, molecular epidemiologic study. J. Clin. Microbiol. 38,
4367–4372.
[17] Pan, S. and Cole, G.T. (1992) Electrophoretic karyotypes of
clinical isolates of Coccidioides immitis. Infect. Immun. 60, 4872–
4880.
[18] Montoya, A.E., Moreno, M.N., Restrepo, A. and McEwen, J.G.
(1997) Electrophoretic karyotype of clinical isolates of Paracoc-
cidioides brasiliensis. Fungal Genet. Biol. 21, 223–227.
428 C.E. Canteros et al. / FEMS Immunology and Medical Microbiology 45 (2005) 423–428

[19] Tateishi, T., Murayama, S.Y., Otsuka, F. and Yamaguchi, H.
(1996) Karyotyping by PFGE of clinical isolates of Sporothrix
schenckii. FEMS Immunol. Med. Microbiol. 13, 147–154.
[20] Perrotta, D., Abrantes, R., Canteros, C., Rodero, L. and Davel,
G. (2001) Caracterización molecular de aislamientos clı́nicos
autóctonos Histoplasma capsulatum var. capsulatum mediante
RAPD-PCR. Rev. Argent. Microbiol. 33, 160–166.
[21] Kasuga, T., White, T.J., Koenig, G., McEwen, J., Restrepo, A.,
Castañeda, E., Da Silva-Lacaz, C., Heins-Vaccari, E.M., De
Freitas, R.S., Zancopé-Oliveira, R.M., Qin, Z., Negroni, R.,
Carter, D.A., Mikami, Y., Tamura, M., Taylor, M.L., Miller,
G.F., Poonwan, N. and Taylor, J.W. (2003) Phylogeography of
the fungal pathogen Histoplasma capsulatum. Mol. Ecol. 12,
3383–3401.