ISSN 0262-0898, Volume 27, Number 1
This article was published in the above mentioned Springer issue. The material, including all portions thereof, is protected by copyright; all rights are held exclusively by Springer Science + Business Media. The material is for personal use only; commercial use is not permitted. Unauthorized reproduction, transfer and/or use may be a violation of criminal as well as civil law.
Clin Exp Metastasis (2010) 27:35–42 DOI 10.1007/s10585-009-9300-9
Author's personal copy
RESEARCH PAPER
Candida albicans enhances experimental hepatic melanoma metastasis Juan Rodrı´guez-Cuesta • Fernando L. Hernando • Lorea Mendoza • Natalia Gallot • Ana Abad Dı´az de Cerio • Guillermo Martı´nez-de-Tejada • Fernando Vidal-Vanaclocha
Received: 30 July 2009 / Accepted: 1 December 2009 / Published online: 25 December 2009 Ó Springer Science+Business Media B.V. 2009
Abstract Candida albicans infections are very frequent in cancer patients, whose immune system is often compromised, but whether this fungal pathogen affects cancer progression is unknown. C. albicans infection involves endogenous production of inflammatory cytokines such as tumour necrosis factor alpha (TNF-a) and interleukin-18 (IL-18). Increased levels of these cytokines have already been correlated with metastasis of most common cancer types. In this study, a well-established model of IL-18dependent hepatic melanoma metastasis was used to study whether C. albicans can alter the ability of murine B16
Guillermo Martı´nez-de-Tejada and Fernando Vidal-Vanaclocha share senior authorship. J. Rodrı´guez-Cuesta G. Martı´nez-de-Tejada Department of Microbiology and Parasitology, University of Navarra, 31080 Pamplona, Spain F. L. Hernando A. A. D. de Cerio Department of Immunology, Microbiology and Parasitology, Basque Country University School of Sciences, 48940 Leioa, Bizkaia, Spain F. Vidal-Vanaclocha (&) Department of Cell Biology and Histology, Basque Country University School of Medicine & Dentistry, 48940 Leioa, Bizkaia, Spain e-mail:
[email protected] L. Mendoza N. Gallot Pharmakine Ltd., Bizkaia Technology Park, 48160 Derio, Bizkaia, Spain
melanoma (B16M) cells to colonize the liver. First, we determined the ability of intrasplenically (IS) injected B16M cells to metastasize into the liver of mice challenged with 5 9 104 C. albicans cells by three different routes (intravenous, IV; intrasplenic, IS; or intraperitoneal, IP) 12 h prior to injection of B16M cells. We demonstrated that C. albicans significantly increased metastasis of B16M cells with all three fungal injection routes. Pro-metastatic effects occurred when hepatic colonization with B16M cells place after the peak of TNF-a and IL-18 levels had been reached in the hepatic blood of fungal challenged mice. In a second set of experiments, mice were fungal challenged 4 days after injection of B16M cells. In these mice, C. albicans also potentiated the growth of established micro-metastases. Significantly, the fungal challenge had pro-metastatic effects without the C. albicans being able to reach the liver, suggesting that soluble factors can promote metastasis in remote sites. Mouse treatment with antifungal ketoconazol abrogated hepatic TNF-a stimulation by C. albicans and prevented the enhancement of hepatic metastasis in fungal challenged-mice. Therefore, the proinflammatory microenvironment generated by the host’s systemic response to C. albicans stimulates circulating cancer cells to metastasize in the liver. Keywords Melanoma Liver sinusoidal cells Metastasis Candida albicans TNF-a IL-18 Inflammation Tumour microenvironment
Introduction J. Rodrı´guez-Cuesta (&) Animal unit, Biosciences Co-operative Research Centre (CIC bioGUNE), Bizkaia Technology Park, 48160 Derio, Bizkaia, Spain e-mail:
[email protected]
Infections caused by certain viruses, bacteria and eukaryotic parasites are being increasingly recognized as triggering factors of malignant tumors [1]. It has been
123
36
Author's personal copy
estimated that such infections account for almost 18% of all cancer causes and their etiological agents include the bacterium Helicobacter pylori (5.5% of all cancers), human papilloma viruses (5.2%), Hepatitis B and C viruses (4.9%), Epstein Barr virus (1%), human immunodeficiency virus (\1%) and human herpes virus (\1%). Eukaryotic parasites appear to be less relevant as causative agents of cancer and fungal infections have not been reported to have carcinogenic potential [2]. Despite the high incidence of infections in cancer patients, little is known about the possible impact of microbial proliferation on metastatic progression. Previously, we and others have shown that injection of E. coli endotoxin (LPS) contributes to the metastatic progression of cancer cells in mice through the induction of proinflammatory cytokines, and that blockade of LPS-dependent IL-1 efficiently inhibits this pro-metastatic action [3, 4]. Consistent with these data, we recently reported a positive correlation between natural infections in laboratory mice, increased serum levels of tumour necrosis factor alpha (TNF-a) and interleukin-18 (IL-18), and incidence and growth of hepatic B16M metastasis [5], supporting previous reports on the pro-metastatic effects of inflammation. Although bacteria are the predominant source of hospital-acquired infections, the incidence of fungal infections is steadily increasing. The most important fungal pathogens belong to the Candida genus, which cause almost 80% of all hospital-acquired fungal infections and the fourth most common nosocomial agent. Among the Candida species, C. albicans is the most common life-threatening infective agent [6, 7]. These mycoses are particularly widespread and severe among cancer patients whose immune system is often compromised due to chemotherapy and other antitumour treatments. Despite the frequency of C. albicans infections in cancer patients [8, 9], it is unknown whether this pathogen might effect cancer progression and metastasis. C. albicans infection leads to production of endogenous inflammatory cytokines such as IL-18 and TNF-a, regulating host defense against candidiasis [10], and it has been reported that IL-18-deficient mice display an increased mortality due to C. albicans infection [11]. Serum levels of IL-18 also increase in cancer patients and this increase correlates with the occurrence and survival of metastases in some tumour types [12]. Therefore, while IL-18 and other pro-inflammatory cytokines play a central role in host defense against infectious and neoplastic diseases, they may also promote metastasis [13]. In this paper we show that C. albicans-infection may affect metastatic progression in cancer patients via a cytokine-dependent mechanism.
123
Clin Exp Metastasis (2010) 27:35–42
Materials and methods Animals Six to eight weeks old C57BL/6 male mice were purchased from Charles River Laboratories (Barcelona, Spain) and certified to be free from specific pathogens (SPF) including Candida spp. To further characterize the animal microflora, samples of mouth and hair from a representative group of animals were taken and incubated on Chromagar (CHROMagar Microbiology, Paris, France) and Candida ID medium (Biomerieux, Espan˜a S.A., Madrid, Spain) for 5 days at 37°C. These cultures gave no C. albicans-like colonies. During the course of the experiments mice were housed in polycarbonate cages containing wood chip bedding (29/12 Plus, Souralit, S.L., Logron˜o, Spain) and placed under conventional laboratory conditions. Animals were fed with rodent maintenance diet (AO4, Panlab, S.L., Barcelona, Spain) and provided with water ad libitum. Serum for cytokine analyses were obtained from the hepatic blood obtained through portal vein aspiration in animals previously anesthetized with 80 mg/kg of ketamine (Imalge`ne 500, Merial, Lyon, France) and 10 mg/kg of xylazine (Xilagesic 2%, Calier, Barcelona, Spain). Animal housing, care and the experimental procedures used conform with institutional guidelines and National and International policies. Fungal strain, culture conditions and experimental infections Candida albicans UPV 1413 was used for all experiments [14]. Yeast cells were grown for 24 h at 24°C in Sabouraud Dextrose Broth (SDB; Cultimed, Panreac Quı´mica SA, Barcelona, Spain) with constant shaking (120 rpm). Then, cells were harvested by centrifugation and a new suspension containing approximately 106 cells/ml was prepared in SDB and incubated for 12 h at 37°C. Cells from this suspension were washed twice with sterile water and suspended in 50 mM PBS at a concentration of 5 9 105 CFU/ ml. 0.1 ml of this suspension was used for each experimental infection. To evaluate the kinetics of colonization of mice by Candida, groups of mice (n = 5) were first injected (IP) with 5 9 104 cells of C. albicans. At four time points (10, 90 min, 2 h and 4 days), blood samples from anesthetized mice were obtained from the peripheral vessels and portal vein and plated onto Chromagar medium. Plates were incubated at 37°C for 5 days and then a viable count was performed. After blood extraction, animals were sacrificed by cervical dislocation, the peritoneal cavity was washed with 2 ml of sterile saline solution and 0.2 ml of the lavage
Clin Exp Metastasis (2010) 27:35–42
Author's personal copy
was cultured on Chromagar as mentioned above. Finally, the whole liver was dissected and homogenized with 1 ml of saline by using a stomacher blender and 0.2 ml of the homogenate was cultured on Chromagar medium and subjected to viable count. Antifungal treatment Animals received by oral route (gavage) 10 mg/kg of ketoconazol (Panfungol-Vet, Esteve, Barcelona, Spain) suspended in a 0.25% (w/v) solution of xanthan gum (Merck, Barcelona, Spain) in sterile saline.
37
Statistical analysis Experimental data were expressed as means ± SD. Statistical analyses were performed by using SPSS software (SPSS 10.0 for Windows, SPSS Inc. Chicago Ill.). For the comparison of parametric data Student ‘‘t’’ test was used whenever two sets of values had to be compared, whereas one-factor ANOVA was used to compare more than two groups of values. The comparison of non-parametric data was performed by using either the Mann–Whitney U test (for two groups) or the Kruskal–Wallis test (for more than 2 groups). The statistical significance was expressed as * (P \ 0.05) and ** (P \ 0.01).
Cancer cells Results The highly metastatic F10 sub-line of B16 melanoma (B16M) was cultured in endotoxin-free Dulbecco’s modified Eagle’s medium (Life Technologies, Inc., Grand Island, NY) supplemented with 10% fetal calf serum, and penicillin–streptomycin (100 U/ml penicillin, and 100 lg/ml streptomycin). Cultured cells were maintained and propagated as described previously [15]. Hepatic metastasis assay Unless indicated otherwise, hepatic metastases were produced by the IS injection into anesthetized mice (Nembutal, 50 mg/kg, IP) of 3 9 105 viable B16M cells suspended in 0.1 ml Hanks’ balanced salt solution. Mice were sacrificed by cervical dislocation on the 12th day after the injection of cancer cells. Liver tissue was fixed by immersion into a solution of 10% formaldehyde in phosphate buffer saline (pH 7.4) and paraffin-embedded. Fifteen 4-lm-thick tissue sections (separated 500 lm) of each liver were stained with hematoxylin-eosin. Densitometric analysis of digitized microscopic images was used to discriminate metastatic B16M from normal hepatic tissue and the number and average diameter of metastases were quantified using an integrated image analysis system (Olympus Microimage 4.0 capture kit; Olympus Optical Co., Hamburg, Germany) connected to an Olympus BX51TF microscope. The volume of liver occupied by metastases (as % of liver) was determined using stereological procedures described previously [15]. Cytokine analysis Concentration of cytokines in animal serum from hepatic blood was measured by using specific ELISA kits for mouse TNF-a and IL-18 (R&D Systems, Minneapolis, MN) according to the manufacturer’s instructions.
Influence of experimental infections by Candida albicans on the metastatic ability of B16M cells To study if infection by C. albicans alters the metastatic efficiency of B16M cells, we determined the ability of intrasplenically-injected melanoma cells to metastasize into the liver of SPF mice previously injected with 5 9 104 CFU of the fungal pathogen. Animals received firstly C. albicans by three different routes, namely intravenous (IV), intrasplenic (IS), or intraperitoneal (IP), and secondly cancer cells 12 h later. Assessment of hepatic colonization of B16M cells on day 12 post-injection revealed a marked increase (P \ 0.05) of metastatic volume in the liver of C. albicans-infected mice compared to non-infected animals (Fig. 1a). Specifically, this enhancement affected (P \ 0.05) the number of small diameter (\1 mm) metastases suggesting that C. albicans had generated a pro-metastatic microenvironment suitable for melanoma cell implantation and early metastatic growth (Fig. 1b). The pro-metastatic effect was of similar magnitude in all C. albicans-infected animals regardless of the injection route. Due to the lower experimental variability observed among animals receiving C. albicans intraperitoneally (see Fig. 1a), this route was chosen for subsequent infections. To evaluate whether infection by C. albicans lowers the inoculum size of B16M cells required for metastasis induction, a low dose (1.5 9 105 cells; IS) was used in mice pre-infected with C. albicans 12 h earlier. These results were compared to uninfected animals receiving an identical inoculum. As shown in Fig. 2a, C. albicans preinfection enabled a dose of B16M cells to metastasize at levels indistinguishable from unchallenged mice receiving the higher dose of cancer cells. As in the previous experiment, C. albicans infection specifically enhanced the number of small-diameter metastases (Fig. 2b), suggesting
123
Author's personal copy
38 80
Metastatic Volume (as % of liver volume)
Fig. 1 Influence of infection by C. albicans on the metastatic ability of B16M cells. Three groups of mice (n = 6) received C. albicans (Ca) by intravenous (IV), intrasplenic (IS) and intraperitoneal (IP) routes. Control groups (n = 6), received sterile saline. After 12 h, mice were injected (IS) with B16M cells. The hepatic metastasis volume was quantified on day 12 after cancer cell injection. a The histograms represent average values per experimental group. b Same data are arranged according to the size of the metastatic foci (*P \ 0.05)
B
*
60
*
* * *
40
40 20 20
0
0 Ca
Ca
IV
2,5
order to determine whether C. albicans pro-metastatic effect was dependent on the presence of fungal cells in the liver of infected animals. Mice were inoculated with 5 9 104 CFU of C. albicans by the IP route and, at different time intervals post-injection (10, 90 min, 12 h and 4 days), samples from peripheral and hepatic blood, and peritoneal fluid were taken and cultured on solid medium. In addition, liver samples that had been washed by repeated immersion in sterile saline, to prevent cross-contamination with peritoneal fluid during dissection, were homogenized and plated. The presence of C. albicans at all these body sites was quantified by viable count in vitro. As shown in Fig. 4, whereas viable C. albicans were detected in high numbers in the peritoneal fluid as late as 12 h post-inoculation, fungal cells were not recovered from peripheral blood, hepatic blood and liver at any time point.
80
Metastatic Volume (as % of liver volume)
1-2,5
Metastasis diameter (mm)
IP
IS
that new micro-metastases were promoted in the liver of infected animals. Because most patients have C. albicans infection at advanced stages of the malignant disease, the effect of C. albicans on already established micro-metastases was next investigated. As shown in Fig. 3, a statistically significant metastasis enhancement also occurred when C. albicans was injected 4 days after cancer cell injection, which suggests an additional stimulating effect of C. albicans on melanoma growth when microscopic metastases were already developed prior to fungal infection.
Fig. 2 Effect of infection by C. albicans on the minimum pro-metastatic dose of B16M cells. Mice received either 5 9 104 CFU of C. albicans (Ca) or sterile saline by IP injection. Twelve hours later, half of mice from each group received an IS injection of 1.5 9 105 B16M cells and the other half received twice that dose. The hepatic metastasis volume was quantified on day 12 after cancer cell injection. a The histograms represent the average values per experimental group (n = 8). b The same data arranged according to the size of the metastatic foci (*P \ 0.05)
60
*
A
Clin Exp Metastasis (2010) 27:35–42
Author's personal copy
Metastatic Volume (as % of liver volume)
80
** *
60
40
20
0
Ca
Ca
4 days
- 12 h
C. albicans inoculation relative to B16M cell injection
750
Peritoneal fluid
750
600
Peripheral blood
600
Hepatic blood
450
450
Liver
300
300
400
A UFC/g
UFC/mL
Fig. 3 Effect of the temporal spacing of C. albicans injection on its pro-metastatic effect. Two mouse groups (n = 8) were injected (IS) with 3 9 105 B16M cells and also received an IP injection containing 5 9 104 CFU of C. albicans (Ca) either 12 h before, or 4 days later, with respect to cancer cell injection. Control groups (n = 8) received saline solution instead of the fungus at the same time points that C. albicans inoculated groups. The hepatic metastasis volume was quantified on day 12 after cancer cell injection. The histogram represents average values per experimental group (P \ 0.01; *P \ 0.05)
150
150
39
increased in the hepatic blood of mice that received C. albicans compared with levels in those mice that were given saline (60 ± 12.2 vs. 33.6 ± 15.1 pg/ml, respectively, for TNF-a and 220.8 ± 23.2 vs. 130.3 ± 17.0 pg/ ml, respectively, for IL-18). Next, we studied the concentrations of TNF-a and IL-18 in serum from hepatic blood of animals used in experiments shown in Fig. 2. As presented in Fig. 5, regardless of the number of injected B16M cells both cytokines further increased on day 12 after cancer cell injection, and more significantly in animals inoculated with C. albicans in comparison with their respective noninfected controls. Cytokine levels in C. albicans-infected mice receiving the lower dose of cancer cells were similar to those measured in non-infected animals injected with the highly-metastatic cancer cell dose. Indeed, the concentration of TNF-a (Fig. 5a) and IL-18 (Fig. 5b) in each experimental group closely correlated with the magnitude of the hepatic volume occupied by metastases in that particular group (compare Figs. 2, 5), indicating that endogenous cytokine induction by tumor and C. albicansdependent mechanisms were synergistic.
*
300
TNF-α (pg/mL)
Clin Exp Metastasis (2010) 27:35–42
* 200
LLD
100 90 min
12 h
4 days
Fig. 4 Assessment of mouse colonization by C. albicans. Four groups of mice (n = 5) were injected (IP) with 5 9 104 CFU of C. albicans (Ca) and blood samples from anesthetized mice were obtained from the peripheral and hepatic vein at four time points (10, 90 min, 12 h and 4 days). After blood extraction, animals were sacrificed and samples from peritoneal fluid and liver were taken. Presence of Candida in the samples was quantified by colony counting. Results are expressed as CFU/ml for fluids, and as CFU/g for tissues. The dotted line indicates the Lower Limit of Detection (LLD)
Determination of proinflammatory cytokines in animals inoculated with C. albicans
0 1000
*
B
**
750
IL-18 (pg/mL)
10 min
500
250
0
1,5 x
IL-1, TNF-a and IL-18 are protective pro-inflammatory cytokines involved in host defense against C. albicans [10, 11]. Because endogenous release of these pro-inflammatory cytokines contributes to the hepatic colonization of B16M cells [16], the level of TNF-a and IL-18 was determined 12 h after C. albicans injection (Fig. 5). Interestingly, TNF-a and IL-18 significantly (P \ 0.05)
Ca
Ca
10 5
3 x 10 5
Injected B16M cells Fig. 5 Serum concentrations of TNF-a and IL-18 in C. albicansinfected hepatic metastasis-bearing mice. On day 12 after cancer cell injection, TNF-a (a) and IL-18 (b) concentrations were determined by ELISA in samples from hepatic blood collected from uninfected and C. albicans-infected mice (n = 8 per group). (**P \ 0.01; *P \ 0.05)
123
40
Author's personal copy
Effect of ketoconazol on C. albicans-induced hepatic B16M metastasis We hypothesized that if infection by C. albicans had a metastasis-enhancing effect on B16M cells, then treatment with an antifungal agent such as ketoconazol may interfere with that pro-metastatic activity. To test this hypothesis we took advantage of our previous observation that C. albicans exhibits a potent pro-metastatic activity on the melanoma cells when the fungus is inoculated 4 days after injection of the B16M cells (see right panel of Fig. 3). Thus, we injected all the animals first with cancer cells, 4 days later with C. albicans and then we started an 8 day-course therapy with ketoconazol by oral route (10 mg/kg). Metastatic colonization of the liver by the melanoma cells was assessed at the end of the antifungal therapy and compared with that of a duplicate group of animals that receive vehicle. In agreement with our previous observations, C. albicans-infected mice had a much larger volume of their liver occupied with metastases compared to noninfected animals (P \ 0.01; see Fig. 6a). These metastatic foci were mainly of small diameter (\1 mm; P \ 0.01) but some were also of medium size (1–2.5 mm; P \ 0.05). Interestingly, ketoconazol was able to lower the severity of the metastatic process down to levels indistinguishable from those of animals non-infected with C. albicans (see Fig. 6a). Moreover, such reduction involved both small and medium diameter metastatic foci (P \ 0.01 and P \ 0.05, respectively), as shown in Fig. 6b. On the other hand, control mice that received ketoconazol but that were not infected with C. albicans did not experience any amelioration of their metastatic process suggesting that ketoconazol does not have an anti-metastatic activity per se (see Fig. 6). Finally, as shown in Fig. 7, ketoconazoldependent reduction of C. albicans pro-metastatic activity correlated with a significant drop in the levels of TNF-a in hepatic blood. However, since concentration of this cytokine in animals treated versus those non-treated with the antifungal agent was similar, the ketoconazol antagonistic activity does not appear to depend on a potential ability of the drug to reduce levels of pro-inflammatory cytokines.
Discussion In this study, we used a well-established mouse model of IL-18-dependent hepatic melanoma metastasis [17, 18] to study whether C. albicans can alter the ability of B16M cells to colonize the liver. The liver was studied as a target because this organ is frequently colonized by C. albicans during chronic disseminated candidiasis, in cancer patients with prolonged, severe neutropenia following chemotherapy [6]. We demonstrated that C. albicans infection
123
Clin Exp Metastasis (2010) 27:35–42
increases the metastatic efficiency of this murine melanoma in the liver, irrespective of fungal injection route. We also showed that C. albicans infection increased IL-18 and TNF-a in the hepatic blood of mice inoculated with melanoma cells, and that treatment of C. albicans-injected mice with antifungal ketoconazol brought cytokine levels down to those detected in non-infected animals, and protected mice against infection-dependent metastases. Most of the cancer cells entering the metastatic process die or fail to colonize target organs for multiple reasons [19, 20] and therefore, the number of circulating cancer cells must reach certain threshold and need favourable micro-environmental factors to ensure metastatic colonization. Our results show that pro-metastatic effect of C. albicans operated by two clinically relevant mechanisms. First, by promoting micro-metastatic foci generation in the liver when hepatic colonization of cancer cells started 12 h after C. albicans injection. Second, by potentiating the growth of established micro-metastases, when C. albicans was inoculated 4 days after cancer cell injection. Under physiological conditions, pro-inflammatory cytokines such as TNF-a and IL-18 play essential roles as mediators of inflammation and immune responses aimed at counteracting infections [21]. In the present work, the injection of C. albicans involved a significant increase of IL-18 and TNF-a in the hepatic blood of healthy mice 12 h post-inoculation. More importantly, final levels of TNF-a and IL-18 accurately reflected the severity of the metastatic process and increments of TNF-a and IL-18 caused by B16M cell injection were much more prominent if animals were previously infected with C. albicans. Therefore, both C. albicans infection and cancer cell growth may synergistically increase the level of both cytokines, while ketoconazol concomitantly decrease both cytokine and metastasis level in C. albicans infected mice. Candida albicans has been shown to stimulate human endothelial cell synthesis of IL-8, TNF-a, IL-1a and IL-1b [22]. In the same report, these cytokines were found to induce the in vitro infected cultures to express E-selectin and vascular cell adhesion molecule-1 (VCAM-1), two molecules mediating circulating melanoma cell adhesion to capillary endothelium [16]. Previously, we reported that B16M cell-dependent factors induced hepatic sinusoidal cell production of IL-18 via TNF-a-induced IL-1. In turn, this pro-inflammatory hepatic microenvironment contributed to melanoma cell adhesion to micro-vascular hepatic endothelium via VCAM-1, and to rodent and human melanoma cell growth [15]. Therefore, at least for hepatic melanoma metastasis, implantation and developmentstimulating effects of C. albicans may primarily depend on the cytokine secretion of pro-inflammatory cytokines released by hepatic sinusoidal cells. In this context, we also have preliminary data showing that primary cultured
Author's personal copy
Clin Exp Metastasis (2010) 27:35–42
Metastatic Volume (as % of liver volume)
80
60
A
**
41
B
**
60 40 40 20 20
* 0
0
Ket
Ket
non -injected with C. albicans
injected with C. albicans
2,5
Metastasis diameter (mm) B16M + C. albicans + vehicle B16M + vehicle B16M + C. albicans + ketoconazol B16M + ketoconazol
Fig. 6 Effect of ketoconazol on pro-metastatic activity of C. albicans infection. Mice were injected (IS) with 3 9 105 B16M cells. Four days later they were injected (IP) with 5 9 104 CFU of C. albicans (Ca) and half of them were treated with ketoconazol (Ket) during 8 days (day 4 to day 12). Control mice received sterile saline 800
* **
600
TNF-α (pg/mL)
instead. The hepatic metastasis volume was quantified on day 12 after cancer cell injection. a The histograms represent average values per experimental group (n = 8). b Same data are arranged according to the size of the metastatic foci (**P \ 0.01; *P \ 0.05)
400
200
0
Ket
non -injected with C. albicans
Ket
injected with C. albicans
Fig. 7 Effect of ketoconazol on TNF-a concentration in normal and C. albicans-infected hepatic metastasis-bearing mice. On day 12 after cancer cell injection, TNF-a concentrations were determined by ELISA in samples from hepatic blood were collected from untreated and ketoconazol-treated, uninfected and C. albicans-infected mice (n = 8 per group). (**P \ 0.01; *P \ 0.05)
hepatic sinusoidal endothelium cells release IL-18 and other pro-inflammatory cytokines when co-cultured with C. albicans and in response to their mannosylated proteins [23]. Therefore, the new microenvironment generated by the hepatic sinusoidal cell response to C. albicans-derived factors may support the ability of a melanoma cell to metastasize at this site. Not surprisingly (see Fig. 2), C. albicans pre-infection enabled low dose of B16M cells to metastasize at levels indistinguishable from uninfected mice receiving a higher dose of cancer cells.
The identification of specific C. albicans factor(s) responsible for TNF-a and IL-18 induction was beyond the scope of this study. However, it is well known that cells of the innate immune system identify and respond to pathogens via specialized membrane-bound receptors of the Toll-like family which recognize specific molecular motifs in the microbial surface (the so called ‘‘pathogen-associated-molecular-patterns’’ (PAMPs) [24]. To date the C. albicans cell wall-associated PAMPs include b-glucans, mannans and phospholipomannan [25]. Although, we cannot rule out that some components released by C. albicans or bound to its cell surface might have an intrinsic pro-metastatic activity on cancer cells directly, we have preliminary evidence showing that C. albicans neither activates pro-inflammatory cytokine secretion nor induces cell growth in cultured B16M cells. In contrast, a cell-free extract containing mannoproteins purified from C. albicans cell surface activates cytokine production from mannose receptor-expressing hepatic sinusoidal endothelium cells which increases cancer cell adherence to this particular capillary bed [23]. Such soluble factors might diffuse through the bloodstream from infection sites to remote areas in the body, and induce a pro-metastatic micro-vascular inflammation at target organs for C. albicans factors. According to our data, this may occur from peritoneal cavity to liver, via mesenteric veins draining peritoneal fluids. In this case, mannose receptor-expressing hepatic sinusoidal cells may release pro-inflammatory cytokines in response to C. albicans factors. To this respect, we have
123
42
Author's personal copy
recently reported that pro-inflammatory cytokine activation of mannose receptor endocytosis mediates IL-10 up-regulation and interferon (IFN)-gamma down-regulation in the hepatic microenvironment, which in turn impairs antitumor cytotoxicity of liver sinusoidal lymphocytes during experimental colon carcinoma hepatic metastasis [26]. Therefore, mannose receptor-dependent anti-tumor inhibition may also be associated to inflammation induced by C. albicans-dependent factors and should not be discarded as an additional pro-metastatic mechanisms of C. albicans infection. In summary, we report that the fungal pathogen C. albicans may enhance hepatic melanoma metastasis both at early and late stages of the metastatic process. Other common malignancies such as colon, ovarian, pancreatic and breast cancer can also metastasize to the liver and their metastatic recurrence has been associated to increased level of IL-18 in peripheral blood [12]. Therefore, our results may have implications for other cancer types. Given that C. albicans is the most prevalent pathogen causing mycosis in cancer patients following immunosuppressive treatment [9] our results have particular significance. Further clinical and epidemiological studies are required to establish whether C. albicans infections are associated with a higher risk of metastasis in human patients. Acknowledgments This work was supported by grants from the Carlos III National Institute of Health (FIS program, no. PI031727), from the CICYT (no. SAF2006-09341) and from Basque Country University to consolidated research groups (no. UPV13641/2003 and GIU06/07) to Fernando Vidal-Vanaclocha and to Fernando L. Hernando (no. UE03/A04 and UE08/14). We acknowledge Dr. David Gubb for critical reading of the manuscript.
References 1. Herrera LA, Benı´tez-Bribiesca L, Mohar A et al (2005) Role of infectious diseases in human carcinogenesis. Environ Mol Mutagen 45(2–3):284–303 2. Parkin DM (2006) The global health burden infection-associated cancers in year 2002. Int J Cancer 118(12):3030–3044 3. Anasagasti MJ, Olaso E, Calvo F, Mendoza L et al (1997) Interleukin 1-dependent and -independent mouse melanoma metastases. J Natl Cancer Inst 89(9):645–651 4. Chirivi RG, Garofalo A, Padura IM et al (1993) Interleukin 1 receptor antagonist inhibits the augmentation of metastasis induced by interleukin 1 or lipopolysaccharide in a human melanoma/nude mouse system. Cancer Res 53(20):5051–5054 5. Rodriguez-Cuesta J, Vidal-Vanaclocha F, Mendoza L et al (2005) Effect of asymptomatic natural infections due to common mouse pathogens on the metastatic progression of B16 murine melanoma in C57BL/6 mice. Clin Exp Metastasis 22(7):549–558 6. DiNubile MJ, Hille D, Sable CA, Kartsonis NA (2005) Invasive candidiasis in cancer patients: observations from a randomized clinical trial. J Infect 50:443–449 7. Fradin C, Hube B (2006) Transcriptional profiling of Candida albicans in human blood. Microbe 1(2):76–80
123
Clin Exp Metastasis (2010) 27:35–42
8. Pagano L, Caira M, Candoni A et al (2006) The epidemiology of fungal infections in patients with hematologic malignancies: the SEIFEM-2004 study. Hematologica 91(8):1068–1075 9. Bodey GP, Mardani M, Hanna HA et al (2002) The epidemiology of Candida glabrata and Candida albicans fungemia in immunocompromised patients with cancer. Am J Med 112(5):380–385 10. Stuyt RJ, Netea MG, Verschueren I et al (2002) Role of interleukin-18 in host defense against disseminated Candida albicans infection. Infect Immun 70(6):3284–3286 11. Netea MG, Vonk AG, van den Hoven M et al (2003) Differential role of IL-18 and IL-12 in the host defense against disseminated Candida albicans infection. Eur J Immunol 33(12):3409–3417 12. Vidal-Vanaclocha F, Mendoza L, Telleria N et al (2006) Clinical and experimental approaches to the pathophysiology of interleukin-18 in cancer progression. Cancer Metastasis Rev 25(3):417–434 13. Dinarello CA (2006) The paradox of pro-inflammatory cytokines in cancer. Cancer Metastasis Rev 25(3):307–313 14. Barea PL, Calvo E, Rodriguez JA et al (1999) Characterization of Candida albicans antigenic determinants by two-dimensional polyacrilamide gel electrophoresis and enhanced chemiluminiscence. FEMS Immunol Med Microbiol 23(4):343–354 15. Vidal-Vanaclocha F, Amezaga C, Asumendi A et al (1994) Interleukin-1 receptor blockade reduces the number and size of murine B16 melanoma hepatic metastases. Cancer Res 54(10):2667–2672 16. Mendoza L, Carrascal T, De Luca M et al (2001) Hydrogen peroxide mediates vascular cell adhesion molecule-1 expression from interleukin-18-activated hepatic sinusoidal endothelium: implications for circulating cancer cell arrest in the murine liver. Hepatology 34(2):298–310 17. Vidal-Vanaclocha F, Fantuzzi G, Mendoza L et al (2000) IL-18 regulates IL-1beta-dependent hepatic melanoma metastasis via vascular cell adhesion molecule-1. Proc Natl Acad Sci USA 97(2):734–739 18. Carrascal MT, Mendoza L, Valcarcel M et al (2003) Interleukin18 binding protein reduces B16 melanoma hepatic metastasis by neutralizing adhesiveness and growth factors of sinusoidal endothelium. Cancer Res 63(2):491–497 19. Mehlen P, Puisieux A (2006) Metastasis: a question of life or death. Nat Rev 6:449–457 20. Weiss L (1990) Metastatic inefficiency. Adv Cancer Res 54:159– 211 21. Schreurs J (1993) Cytokines in infectious disease: the battle between host and pathogen. Curr Opin Biotechnol 4(6):727–733 22. Orozco AS, Zhou X, Filler SG (2000) Mechanisms of the proinflammatory response of endothelial cells to Candida albicans infection. Infect Inmmun (68):1134–1141 23. Ramirez A, Gallot N, Abad A et al. (2004) Variations in B16 melanoma cell adhesion to the mouse hepatic sinusoidal endothelium (HSE) and cytokines production were induced by different Candida albicans strains. In: Paper presented at the 12th international symposium on cells of the hepatic sinusoid, Bilbao, 5–9 Sep 2004 24. Werling D, Jungi TW (2003) TOLL-like receptors linking innate and adaptive immune response. Vet Immunol Immunopathol 91(1):1–12 25. Roeder A, Kirschning CJ, Rupec RA et al (2004) Toll-like receptors as key mediators in innate antifungal immunity. Med Mycol 42(6):485–498 26. Arteta B, Lasuen N, Vidal-Vanaclocha F (2007) Tumor-activated endothelial cells inhibit anti-tumor immunity during hepatic colon carcinoma metastasis via mannose receptor upregulation. In: Paper presented at the 4th international conference on tumor microenvironment: progression, Florence, 6–10 Mar 2007
