ORIGINAL ARTICLE 10.1111/j.1469-0691.2009.02824.

Different virulence levels of the species of Sporothrix in a murine model

I. Arrillaga-Moncrieff1, J. Capilla2, E. Mayayo1, R. Marimon2, M. Mariné2, J. Gené2, J. Cano2 and J. Guarro2

1) Pathology Unit and 2) Microbiology Unit, Facultat de Medicina i Ciències de la Salut, Universitat Rovira i Virgili, Reus, Tarragona, Spain

Abstract

A comparative study on the experimental pathogenicity of five species of Sporothrix of clinical interest, Sporothrix albicans, Sporothrix
brasiliensis, Sporothrix globosa, Sporothrix mexicana, and Sporothrix schenckii sensu stricto, was performed using an immunocompetent
murine model. Two strains of each species and two levels of inoculum for each strain (2 · 107 and 2 · 104 conidia/animal) were tested
by intravenous inoculation of mice (ten per group). Mortality was caused by the low inoculum of one strain of S. brasiliensis only, and
the high inocula of S. brasiliensis and S. schenckii strains. Other inocula and other species tested did not kill any of the experimental
animals. Tissue burden studies showed fungal spread to kidneys, lungs, spleen, brain, and testicles. S. brasiliensis was recovered exten-
sively from all of the studied organs, and S. schenckii and S. globosa were recovered in lower amounts. Histopathological studies revealed
differences in the lesions, which ranged from local inflammation with a low number of fungal cells at the injection site in mice infected
with S. globosa, to massive infiltration of fungal cells in organs of those infected with S. brasiliensis. Our findings showed that S. brasiliensis
and S. schenckii were the most virulent species, and suggest that lesional mechanisms could be species-specific.

Keywords: Murine model, Sporothrix brasiliensis, Sporothrix schenckii, sporotrichosis, virulence

Original Submission: 10 September 2008; Revised Submission: 3 December 2008; Accepted: 4 December 2008
Editor: M. C. Arendrup
Article published online: 15 July 2009
Clin Microbiol Infect 2009; 15: 651–655

cally, have been proposed to be in the complex: Sporothrix

Corresponding author and reprint requests: J. Guarro, Unitat
brasiliensis and Sporothrix globosa are associated with human
de Microbiologia, Departament de Ciències Mèdiques Bàsiques,
Facultat de Medicina i Ciències de la Salut, Universitat Rovira i Virgili, infections, and Sporothrix mexicana has until now only been
Carrer Sant Llorenç 21, 43201 Reus, Tarragona, Spain identified among isolates of environmental origin [12].
E-mail: josep.guarro@urv.cat
Different models, including subcutaneous and intravenous
infection in mice [13,14] and subcutaneous infection in ham-
sters [15], have been used in virulence studies on sporothri-
Introduction
chosis. The use of animal models has demonstrated differences
in virulence between isolates from environmental and clinical
Sporotrichosis is a chronic subcutaneous disease caused by sources as well as differences between isolates of similar origin.
the dimorphic fungus Sporothrix schenckii, characterized by Following the proposal of three new species of Sporothrix, it is
the development of lymphatic nodules in humans and in not known whether the different degrees of severity attributed
some animals [1]. The fungus is found in habitats that are to S. schenckii correspond to different species of the complex
rich in organic matter and in regions with a warm and humid or whether they are caused by different isolates of the same
climate [2–4]. The infection is caused by traumatic inocula- species. We demonstrated previously that the species of the
tion of mycelia or conidia from the soil, wood, or plants [5]. complex showed different antifungal susceptibilities [16]. In the
Other studies have demonstrated varying mechanisms of present study, we compared the virulence of different species
infection, such as inhalation, insect bites, scratches by wild or of the complex in a murine model of disseminated infections.
domestic animals, and direct contact with exudates of feline
lesions [4,6–10]. S. schenckii had been considered to be a
Materials and Methods
well-defined species, with no significant intraspecific varia-
tions, but recent molecular studies have shown that this fun-
gus is a complex of phylogenetic species with different Strains and inocula preparation
geographical distributions [4,11]. In addition to S. schenckii Ten isolates of Sporothrix spp. were used in the study, two
sensu stricto, three new species, characterized phenotypi- isolates corresponding to each of the following species:

ª2009 The Authors

Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases
652 Clinical Microbiology and Infection, Volume 15 Number 7, July 2009 CMI

Sporothrix albicans, S. brasiliensis, S. globosa, S. mexicana, and Tissue burden study

S. schenckii (Table 1). The isolates were stored on corn- Three groups of animals (ten per group) were infected intra-
meal agar (30 g of corn, 15 g of agar, 1 L of tap water) venously with 2 · 107 conidia/animal of S. schenckii 1, S. brasili-
slants covered with paraffin oil, and prior to the study ensis 2, or S. globosa 1, respectively. Twenty days post-
they were cultured on potato dextrose agar (PDA) for infection, the mice were killed by CO2 anoxia, and spleen,
5–8 days at 30C. The inocula were prepared by flooding lungs, liver, testicles, kidneys and brain were removed asepti-
the surface of the agar plate with sterile saline, scraping cally. Approximately half of each organ was weighed, and
the sporulating mycelium with a culture loop, and drawing homogenized in 0.9% saline; ten-fold dilutions were then
up the resultant suspension with a sterile Pasteur pipette. placed into PDA for CFU determination.
The suspensions were then filtered once through sterile
gauze to remove hyphae. The conidia suspensions were Histopathology study
transferred to 200 mL of potato dextrose broth and incu- Half of each organ, plus skin from the tail, was fixed with
bated in an orbital shaker (150 rpm) at 30C for 4 days, 10% buffered formalin. Samples were dehydrated, paraffin-
filtered once through sterile gauze, and centrifuged at embedded, and sliced into 2-lm sections, which were stained
325 g for 10 min. The resulting pellet was washed three with haematoxylin and eosin, periodic acid–Schiff or Grocott
times in 0.9% saline by centrifugation (325 g for 10 min), methamine silver, and examined in a blinded fashion by light
and the conidia concentration was adjusted after counting microscopy.
with a haemocytometer. The viability of these inocula was
verified by plating dilutions of the suspension on PDA Statistical analysis
plates. Survival was compared using a two-tailed log-rank test.
Organ burden data were log10 transformed and compared by
Animals the two-tailed Mann–Whitney U-test, using GraphPad Prism 5
Six-week-old OF-1 male mice (ten per group) (Charles River, for Windows. p-Values £0.05 were considered to be signifi-
Criffa SA, Barcelona, Spain), weighing 28–30 g, were used. cant.
Animals were housed five per cage in standard boxes with
corncob bedding and free access to food and water. Condi-
Results
tions were approved by the Animal Welfare Committee of
the Faculty of Medicine of the University Rovira i Virgili.
Mortality
Mortality study The results show a clear correlation between the size of the
Twenty groups of animals were established, one for each iso- inoculum and the mortality rate. When the low inocula were
late and each of the two inocula. Infection was established tested, only S. brasiliensis 1 caused the death of all of the
intravenously with 200 lL of inoculum via the lateral tail mice between days 30 and 37; mice infected with the other
vein. For each isolate, two inocula were prepared, i.e. isolates were alive at the end of the experiment. All of the
2 · 107 conidia/animal (high) and 2 · 104 conidia/animal animals infected with the high inocula of S. albicans, S. globosa
(low), with the exception of S. brasiliensis 1, for which a high and S. mexicana survived to the end of the experiment, while
inoculum of 1.2 · 106 conidia/animal was prepared. Mortality all of the animals infected with the high inocula of S. brasilien-
was recorded daily for 40 days. sis and S. schenckii died between days 9 and 23, with no sig-
nificant differences (p >0.05) (Fig. 1).
TABLE 1. Isolates of Sporothrix included in the study, and
their origins Tissue burden
The tissue burdens of the different organs studied were con-
Species Isolate Reference no. Origin
sistent with the results of the survival study. Animals infected
Sporothrix schenckii 1 UTHSC 99-173 Biopsy tissue, hand, USA
with S. brasiliensis or S. schenckii showed the highest fungal
2 UTHSC 01-2137 Skin ulcer, USA loads, and those infected with S. globosa were significantly
Sporothrix brasiliensis 1 CBS 120339 Skin lesion, Brazil
2 IPEC 16919 Skin lesion, Brazil lower (p <0.001 in all of the organs) (Fig. 2). The liver and
Sporothrix globosa 1 CBS 120340 Face lesion, Spain
2 MCCL 220082 Unknown, India testicles were the organs most affected in animals infected
Sporothrix mexicana 1 CBS 120341 Soil, Mexixo
2 CBS 120342 Carnation leaves, Mexico
with S. schenckii. Fungal burden was significantly higher in ani-
Sporothrix albicans 1 CBS 302.73 Soil, UK mals infected with S. brasiliensis than in those infected with
2 KMU 4217 Wheat, China
S. schenckii in all studied organs (p 0.043 to <0.001).

ª2009 The Authors

Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases, CMI, 15, 651–655
CMI Arrillaga-Moncrieff et al. Virulence of Sporothrix spp. 653

Interestingly, S. brasiliensis showed marked brain tropism white nodules, approximately 1 mm in diameter, randomly
(7.31 ± 0.64 CFU/g), whereas this organ was minimally distributed on parenchyma of liver and spleen (Fig. 3b). An
affected by S. schenckii or S. globosa (3.84 ± 2.10 CFU/g and increase in testicular volume was observed, but changes
0.74 ± 1.33 CFU/g, respectively). Fifty per cent of the ani- were not noted macroscopically in other organs. Microscopi-
mals infected with S. globosa cleared the infection in all of cally, all of the organs of mice challenged with S. schenckii
the organs. showed granulomas with a necrotic centre (Fig. 4a), with
mature round and immature oval and cigar-shaped fungal
Histopathology cells (Fig. 4b,c). Inside the tissue vessels, some fungal thrombi
During the experimental period, mice infected with S. brasili- were present without observable infarction. Although granu-
ensis or S. schenckii showed cutaneous lesions of the tail lomatous lesions affected the paratesticular soft tissue, the
(Fig. 3a) and orchitis 10 days post-inoculation, whereas those testicles did not show any inflammatory response. No
infected with the other three species did not show any lesions were found in brain or meninges. The organs of mice
apparent lesion with either of the two inocula. Macroscopi- infected with S. brasiliensis showed necrotic areas, lacking
cally, only animals infected with S. schenckii showed lesions in inflammatory cells, with substitution of tissue by massive
internal organs, consisting of abundant but not confluent infiltration of fungal cells, mostly located in the Kupffer cells
(Fig. 4d–f). Paratesticular and meningeal tissues showed gran-
ulomatous lesions, with some fungal cells located in the cen-
tre of the granuloma.
No inflammation or fungal cells were observed in the
organs of mice infected with S. globosa. The tails of these
mice showed a subcutaneous inflammation with scattered
fungal cigar-shaped cells.

Discussion
FIG. 1. Survival of mice infected with Sporothrix spp. Mice were
infected with 2 · 107 conidia/animal, with the exception of Sporothrix Different studies using animal models have shown differences
brasiliensis 1 (2 · 104 conidia/animal). Ss, Sporothrix schenckii; Sb, S. bra- in virulence between Sporothrix isolates, but the causes
siliensis; Sg, Sporothrix globosa; Sm, Sporothrix mexicana; Sa, Sporothrix remain unclear. Studies evaluating the pigmentation, capacity
albicans. for growth at 37C and the clinical or environmental origin

FIG. 2. Scattergram of CFU/g of tissue

from mice infected with (a) Sporothrix
globosa 1, (b) Sporothrix schenckii 1, or
(c) Sporothrix brasiliensis 2. Each animal
was inoculated intravenously with
2 · 107 cells. The horizontal bars repre-
sent the means.

ª2009 The Authors

Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases, CMI, 15, 651–655
654 Clinical Microbiology and Infection, Volume 15 Number 7, July 2009 CMI

of the strains have shown a correlation with virulence. Mesa- S. globosa strains included in our study showed no pigmented
Arango et al. [4] found that environmental isolates generally conidia, did not grow at 37C, and were avirulent in our
showed higher virulence than isolates from clinical sources, murine model. More recently, it has been demonstrated that
independently of geographical origin. By contrast, Dixon isolates from cutaneous infections are more virulent than
et al. [13] studied clinical and environmental isolates from isolates from fixed infections in a systemic murine model
the largest US epidemic of sporotrichosis, and concluded [14,17].
that isolates with pigmented conidia and that grew at 37C Comparisons with previous studies are difficult to make,
were equivalently virulent in a murine model, independently as it not known which of the currently accepted species of
of their origin. Our results agree with these findings, as the Sporothrix are being referred to. Our results showed

(a) (b)

FIG. 3. Ulcerated lesion on tail (a) and

presence of nodules in liver (b) of OF-1
mice infected intravenously with
2 · 107 conidia/animal of Sporothrix
schenckii 1.

(a) (b)

(c) (d)

FIG. 4. Histopathological findings in liver

of OF-1 mice 20 days after intravenous
infection with 2 · 107 conidia of Sporo-
thrix spp. (a–c) Fungal cells and granulom-
atous lesions after infection with
(e) (f) Sporothrix schenckii 1 (haematoxylin and
eosin). (d, e) Massive infiltration of Sporo-
thrix brasiliensis 2 fungal cells replacing he-
patocytes. Arrows indicating conidia in
the kupffer cells (haematoxylin and
eosin). (f) Fungal cells of Sporothrix brasili-
ensis 2 on liver (periodic acid–Schiff).
Magnification ·25 (a, d) and ·400 (b, c,
e, f).

ª2009 The Authors

Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases, CMI, 15, 651–655
CMI Arrillaga-Moncrieff et al. Virulence of Sporothrix spp. 655

significant differences in the degree of virulence between 2. Mehta KI, Sharma NL, Kanga AK, Mahajan VK, Ranjan N. Isolation
of Sporothrix schenckii from the environmental sources of cutaneous
species: S. brasiliensis was clearly the most virulent species in
sporotrichosis patients in Himachal Pradesh, India: results of a pilot
terms of mortality, tissue burden, and tissue damage, fol- study. Mycoses 2007; 50: 496–501.
lowed by S. schenckii and then S. globosa. S. mexicana and 3. Mendoza M, Diaz E, Alvarado P, Romero E, Bastardo de Albornoz
S. albicans showed low or no virulence in our animal model. MC. Isolation of Sporothrix schenckii from environmental samples in
Venezuela. Rev Iberoam Micol 2007; 24: 317–319.
Skin, lungs, spleen and testicles have been reported as 4. Mesa-Arango AC, Del Rocio Reyes-Montes M, Perez-Mejia A et al.
major organs affected in experimental systemic or subcutane- Phenotyping and genotyping of Sporothrix schenckii isolates according
ous infection [14,18]. Our study demonstrated that S. brasili- to geographic origin and clinical form of sporotrichosis. J Clin Micro-
biol 2002; 40: 3004–3011.
ensis, S. schenckii and, to a lesser extent, S. globosa are able
5. Conias S, Wilson P. Epidemic cutaneous sporotrichosis: report of 16
to spread to other organs, such as liver, kidneys, and brain. cases in Queensland due to mouldy hay. Australas J Dermatol 1998;
Interestingly, fungal forms of S. schenckii were located at the 39: 34–37.
6. Bonasse J, Miller A, Civila E, Conti-Diaz IA. Cutaneous lymphangitic-
centre of liver granulomas, and S. brasiliensis conidia were
nodular sporotrichosis with multiple portals of entry. Med Cutan Ibero
found extensively in the Kupffer cells, showing similarity to Lat Am 1988; 16: 263–265.
Histoplasma capsulatum infections [19]. 7. Lopes-Bezerra LM, Schubach A, Costa RO. Sporothrix schenckii and
Now that differences are known to exist between the dif- sporotrichosis. An Acad Bras Cienc 2006; 78: 293–308.
8. Miller SD, Keeling JH. Ant sting sporotrichosis. Cutis 2002; 69: 439–
ferent species of the S. schenckii complex in their virulence 442.
and response to antifungals [16], it is important to differenti- 9. Schubach A, Schubach TM, Barros MB, Wanke B. Cat-transmitted
ate between these species in the clinical setting. sporotrichosis, Rio de Janeiro, Brazil. Emerg Infect Dis 2005; 11:
1952–1954.
This is crucial for the correct treatment of such infections
10. Schubach AO, Schubach TM, Barros MB. Epidemic cat-transmitted
and for a better understanding of their epidemiology. This sporotrichosis. N Engl J Med 2005; 353: 1185–1186.
task will be greatly facilitated by the use of recently 11. Marimon R, Gene J, Cano J, Trilles L, Dos Santos Lazera M, Guarro
J. Molecular phylogeny of Sporothrix schenckii. J Clin Microbiol 2006; 44:
described molecular and phenotypic markers [12].
3251–3256.
12. Marimon R, Cano J, Gene J, Sutton DA, Kawasaki M, Guarro J. Sporo-
thrix brasiliensis, S. globosa, and S. mexicana, three new Sporothrix spe-
Acknowledgements cies of clinical interest. J Clin Microbiol 2007; 45: 3198–3206.
13. Dixon DM, Duncan RA, Hurd NJ. Use of a mouse model to evaluate
clinical and environmental isolates of Sporothrix spp. from the largest
The authors are indebted to the curators of the culture US epidemic of sporotrichosis. J Clin Microbiol 1992; 30: 951–954.
collections indicated in Table 1 for supplying the strains used 14. Brito MM, Conceicao-Silva F, Morgado FN et al. Comparison of viru-
lence of different Sporothrix schenckii clinical isolates using experimen-
in the study.
tal murine model. Med Mycol 2007; 45: 721–729.
15. Charoenvit Y, Taylor RL. Experimental sporotrichosis in Syrian ham-
sters. Infect Immun 1979; 23: 366–372.
Transparency Declaration 16. Marimon R, Serena C, Gene J, Cano J, Guarro J. In vitro antifungal
susceptibilities of five species of Sporothrix. Antimicrob Agents Chemo-
ther 2008; 52: 732–734.
This study was supported by the Spanish Ministerio de 17. Kong X, Xiao T, Lin J, Wang Y, Chen HD. Relationships among geno-
Ciencia y Tecnologı́a, grants CGL 2004-00425/BOS and types, virulence and clinical forms of Sporothrix schenckii infection. Clin
Microbiol Infect 2006; 12: 1077–1081.
CGL 2005-07394. The authors declare that they have no 18. Tachibana T, Matsuyama T, Mitsuyama M. Characteristic infectivity of
conflicting interests in relation to this work. Sporothrix schenckii to mice depending on routes of infection and
inherent fungal pathogenicity. Med Mycol 1998; 36: 21–27.
19. Salfelder K. Atlas of fungal pathology. Dordrecht, the Netherlands:
References Kluwer Academic Publishers, 1990.

1. Ramos-e-Silva M, Vasconcelos C, Carneiro S, Cestari T. Sporotricho-

sis. Clin Dermatol 2007; 25: 181–187.

ª2009 The Authors

Journal Compilation ª2009 European Society of Clinical Microbiology and Infectious Diseases, CMI, 15, 651–655