Previously, we reported that Malassezia furfur, causing systemic fungal infection, was taken up into human monocytic cell line, THP-1, in a concentration-dependent manner. This fact suggested that M. furfur could activate phagocytes, such as monocyte and polymorphonuclear leukocyte. Thus we examined cytokine mRNA expression from human monocytic and granulocytic cell line, THP-1 and HL-60, stimulated with M. furfur by using reverse transcriptase-polymerase chain reaction (RT-PCR) and ELISA. We chose IL-1α, IL-6, IL-8, IL-12 and TNF-α as primers for THP-1, and IL-1α, IL-6 and IL-8 for HL-60. M. furfur induced the expression of IL-8 mRNA from THP-1 and HL-60 following incubation for 3 h, and also induced IL-1α mRNA from HL-60, although this induction was weaker than that of IL-8 mRNA. Furthermore, opsonized M. furfur induced stronger expression of IL-8 mRNA in comparison with intact M. furfur. IL-8 production from THP-1 and HL-60 was enhanced in a concentration- and incubation time-dependent manner. These facts strongly suggested that M. furfur could activate phagocytes, and could induce inflammatory responses in systemic infection.
Malassezia furfur is a lipophilic and dermatological pathogenic yeast, that normally exists on human skin, and causes tinea versicolor , seborrhoeic dermatitis , and atopic dermatitis . Moreover, systemic infection of M. furfur was initially reported in 1981 as a complication of total parenteral nutrition in neonates receiving long-term lipid emulsion through central venous catheters . Systemic infections by M. furfur were isolated from peritoneal fluid of patients with peritonitis undergoing continuous peritoneal dialysis, and broviac-catheter related M. furfur sepsis [5–7]. Fever, thrombocytopenia, sepsis were observed as clinical symptoms, and especially, premature neonates who developed pulmonary vasculitis died in systemic infection of M. furfur.
Previously, we reported that M. furfur activated complement system, constituted with classical and alternative pathways , and were taken up into human monocytic cell line, THP-1, via β-glucan receptor and mannose receptor, and furthermore, opsonized M. furfur might be also taken up via complement receptor . These facts suggested that M. furfur might induce the inflammatory response and activation of monocyte in host. Pathogenic fungi, such as Candida albicans, Saccharomyces cerevisiae and Cryptococcus neoformans, taken up into monocyte via above mentioned receptors, generally induce the cytokine production by the activation of phagocytic cells [10–13]. Cytokine production plays a key role for the induction of inflammatory responses and subsequent antigen-specific cellular immune responses against invasion of pathogenic fungal cells. Monocyte and polymorphonuclear leukocyte (PMN) release various bioactive peptides, reactive oxygen and nitrogen species, and are therefore primary defense systems against bacteria and fungi. However, the effect of M. furfur on immune responses during the systemic infection is not understood. We have therefore examined the expression of cytokine mRNAs from human monocytic and granulocytic cell lines, THP-1 and HL-60, stimulated with M. furfur.
2Materials and methods
M. furfur HIC 3321 and HIC 3343 were clinical isolates and C. albicans ATCC 10231 were used as pathogenic yeasts throughout these experiments.
Human monocytic cell line, THP-1, and human granulocytic cell line, HL-60, were used throughout, and purchased from Japan Cell Research Bank.
2.3Preparation of M. furfur
M. furfur was cultivated on Dixon's agar medium at 37°C for 1 week. After cultivation, the cells were harvested and suspended in RPMI 1640 (Nissui Seiyaku Co. Ltd.), and the obtained suspension was used as sample. These cell suspensions were heated in boiling water for 30 min, and used as heat-killed cells.
2.4Preparation of C. albicans
C. albicans was inoculated onto potato-dextrose agar medium and incubated at 30°C for 2 days. The formed colony was suspended in RPMI 1640, and the resulting suspension was used as sample. The cell suspension was heated in boiling water for 30 min, and used as heat-killed cells.
2.5Preparation of normal human serum
The blood was obtained from healthy volunteers, and was placed at room temperature for 2 h. The blood was centrifuged at 3000 rpm for 15 min, and the supernatant obtained was used as normal human serum.
2.6Preparation of opsonized yeast
5.0×107 cells ml−1 of yeast were mixed with 20% normal human serum for 30 min at 37°C. The mixture was centrifuged at 15 000 rpm for 5 min, and the obtained precipitate was used as opsonized yeast.
THP-1 was cultured in fetal calf serum coated 24-well plate at 1.0×106 cells per well in 1 ml of culture medium in the presence of 20 ng ml−1 of phorbol myristate acetate (PMA) for 7 days. After the culture, the adherent cells were washed three times with pre-warmed RPMI 1640.
HL-60 was cultured in 24-well plates at 1.0×106 cells per well in 1 ml of culture medium in the presence of 1 μM of all-trans retinoic acid for 7 days. After the culture, the cells were washed with pre-warmed RPMI 1640.
2.9Stimulation of human phagocytic cell line with M. furfur
Differentiated cells were cultured with 1.0×108 or 5.0×107 per well of M. furfur for 3 h for RT-PCR and 24 h for ELISA. After the incubation, the cells were used for assay of RT-PCR and ELISA.
Total cellular RNA was extracted from the cell suspension using ISOGEN (Wako Chemical Co.). The RNA fraction was dissolved in 20 μl of DEPC-treated water. cDNA was prepared using random hexamer and Moloney-Murine leukemia virus reverse transcriptase (MMLV RTase) (United States Biochemical Corporation). Samples were stored at 4°C until use. Aliquots of cDNA were amplified in 500 μl microcentrifuge tubes in the presence of 0.28 μM sense primer, 0.28 μM antisense primer, and 0.1075 units of Taq DNA polymerase in PCR buffer. The reaction mixture was overlaid with 3 drops of mineral oil, and PCR was performed in a Perkin-Elmer Cetus DNA Thermal Cycler (Connecticut, USA) for 30 to 40 cycles, each cycle consisted of 1 min of denaturation at 94°C, 2 min of annealing at 60°C, and 3 min of extension at 74°C. Reaction products were visualized by electrophoresis of 10 μl of the reaction mixture at 50 V for 40 min in a 1.2% agarose gel containing 0.5 μg of ethidium bromide per ml.
The primers used for this experiment were as follows: β-actin sense primer (s) 5′-TCCTGCGGCATCCACGAAACT-3′; antisense primer (a) 5′-GAAGCATTTGAGGTGGACGAT-3′; IL-1α (s) 5′-GTAAGCTATGGCCCACTCCAT-3′, 5′-TGACTTATAAGCACCCATGTC-3′; IL-6 (s) 5′-AACTCCTTCTCCACAAGCG-3′, 5′-TGGACTGCAGGAACTCCTT-3′; IL-8 (s) 5′-ATTTCAGCAGCTCTGTGTGAA-3′, 5′-TGAATTCTCAGCCCTCTTCAA-3′; IL-12p40 (s) 5′-GTCTATTCCGTTGTGTC-3′, 5′-CCACATTCCTACTTCTC-3′; TNF-α (s) 5′-ATGAGCACTGAAAGCATGATC-3′, (a) 5′-TCACAGGGCAATGATCCCAAAGTAGACCTGCCC-3′.
2.11ELISA for IL-8
A 96-well plate (Sumitomo Bakelite Co., Tokyo, Japan) was coated with anti-human IL-8 monoclonal antibody (Pharmingen, CA, USA) in 0.1 M bicarbonate buffer (pH 9.6) by incubation at 4°C overnight. The plate was washed with phosphate-buffered saline (PBS) containing 0.05% Tween 20 (Wako Pure Chemical Co.) (PBST) and blocked with 0.5% bovine serum albumin (Sigma Chemical Co.) (BPBST) at 37°C for 40 min. After washing, the plate was incubated with 50 μl of recombinant human IL-8 or samples at 37°C for 40 min. The plate was washed with PBST for three times and then treated with biotinylated anti-human IL-8 polyclonal antibody (Pharmingen, CA, USA) at 37°C for 40 min. After the incubation, the plate was treated with streptavidin peroxidase and TMB substrate system (KPL Inc., MD, USA). Color development was stopped with 1 N phosphoric acid and the optimal density at 450 nm was measured by using microplate reader (Corona Electric Co. Ltd.).
3.1Expression of cytokine mRNAs by human monocytic and granulocytic cell lines stimulated with M. furfur
As the first experiment, we examined cytokine mRNA expression by human monocytic cell line, THP-1, and granulocytic cell line, HL-60 stimulated with M. furfur and C. albicans. We used primers for human IL-1α, IL-6, IL-8, IL-12 and TNF-α in THP-1, and for IL-1α, IL-6 and IL-8 in HL-60. THP-1 stimulated with live or heat-killed cells of M. furfur at a concentration of 1.0×108 cells ml−1 for 3 h induced only the expression of IL-8 mRNA (Fig. 1). Furthermore, HL-60 stimulated with live or heat-killed cells of M. furfur at a concentration of 1.0×108 cells ml−1 induced the expression of IL-1α and IL-8 mRNAs, and the signal of IL-1α mRNA was weaker than that of IL-8 mRNA (Fig. 2). On the other hand, although heat-killed cells of C. albicans induced the expression of all of the used mRNAs, live cells of C. albicans induced only the expression of IL-1α and IL-8 mRNAs.
3.2Effect of opsonization on IL-8 mRNA expression from human monocytic cell lines stimulated with M. furfur
Previously, we reported that M. furfur could activate the complement system and be taken up into THP-1 via the β-glucan receptor and/or mannose receptor. Moreover, the uptake of M. furfur was enhanced by the opsonization with normal human serum. This fact strongly suggested the elevation of signal transduction in THP-1 following opsonization of M. furfur. Thus we examined the effect of opsonization of M. furfur on IL-8 mRNA expression from THP-1. As shown in Fig. 3, opsonized M. furfur induced IL-8 mRNA expression stronger than intact M. furfur.
3.3IL-8 production from human monocytic and granulocytic cell lines stimulated with M. furfur
As mentioned above, we found that M. furfur could induce the expression of IL-8 mRNAs from THP-1 and HL-60. We therefore examined the concentration and incubation time dependency of M. furfur on IL-8 production by using ELISA. As shown in Figs. 4 and 5, IL-8 production from THP-1 stimulated with M. furfur was enhanced in a concentration- and incubation time-dependent manner, and reached plateau level at the concentration of 5×107 ml−1. Furthermore, IL-8 production was also enhanced in a concentration- and incubation time-dependent manner in the case of HL-60, and live cells of M. furfur could induce the stronger activity than heat-killed ones (Figs. 6, 7).
Systemic infection of M. furfur was reported as a complication of total parenteral nutrition in neonates receiving long-term lipid emulsion through central venous catheters, and there have since been numerous reported cases of systemic infections by M. furfur[4–7]. However, the mechanism for exclusion of M. furfur from the host is not precisely examined until today. We previously reported that soluble mannan and β-glucan inhibit the uptake of M. furfur by human monocytic cell line, THP-1 . This suggested that M. furfur could activate phagocytes via both mannose and β-glucan receptors. We therefore examined the expression of cytokine mRNAs and subsequent cytokine production from human phagocytic cell lines stimulated with M. furfur. In this paper we demonstrated that M. furfur enhanced IL-8 production from both human monocytic and granulocytic cell lines at the transcriptional level. Live cells of M. furfur induced stronger activity of IL-8 production than heat-killed ones, and thus it is suggested that heat unstable substance, such as protein, might participate in IL-8 production.
Furthermore, the opsonization of M. furfur with normal human serum induced further IL-8 mRNA expression in comparison with intact M. furfur. These facts indicate the possibility that receptor-mediated signal transduction for IL-8 production in phagocytes would be induced by the stimulation with M. furfur. It is well known that bioactive IL-8 could induce the activation of polymorphonuclear leukocytes and the proliferation of T cells in addition to the chemotactic activity against polymorphonuclear leukocytes.
In addition, IL-8 production from HL-60 was stronger than that from THP-1. This phenomenon suggests that granulocyte would be important for the induction of inflammatory response and chemotaxis to infectious sites at the early stage of microbial infection in comparison with monocyte. Thus the activation of HL-60, human granulocytic cell line, would be related to the exclusion of invaded M. furfur from the host at the early stage of systemic infection. Furthermore, it is well known that fungal (1→3)-β-d-glucans possess various immunopharmacological activities, such as cytokine production and nitric oxide production [14–17]. It is thought that the cell wall of M. furfur is composed of mannan and β-glucan, although their structures were not examined in detail . We found, however, that M. furfur contains (1→3)- and (1→6)-β-d-glucans in cell wall by using 13C-NMR (unpublished data). Thus the cell wall β-glucan would participate in the expression of IL-8 mRNA from phagocytes and subsequent induction of inflammatory responses. Since opsonization with normal human serum induced the further enhancement of IL-8 mRNA expression from THP-1, complement receptor would be involved in the expression of IL-8 mRNA. We previously reported that the uptake of M. furfur by THP-1 was enhanced by the treatment with normal human serum, but not heat-inactivated serum . In addition, M. furfur opsonized with heat-inactivated serum also enhanced IL-8 production. These phenomena indicate that IL-8 production was not dependent on the uptake of M. furfur, but on the binding to THP-1. Thus natural antibody against M. furfur in human serum would bind to M. furfur and Fc receptor-mediated activation of phagocytes would participate in the transcriptional stage of IL-8 production in the case of the opsonization with heat-inactivated serum. Considering these facts, it is suggested that the expression of IL-8 mRNA from phagocytes stimulated with M. furfur would play a key role for the exclusion of M. furfur and induction of antigen-specific immune responses against systemic infection.
The authors wish to thank Mr. J. Nohjo and Mr. K. Sagawa for their technical assistance.