Correspondence Claudio Cermelli, Dipartimento di Scienze di Sanità Pubblica, Via Campi 287, 41100 Modena, Italy. Tel: +39 059 205 5457; fax: +39 059 205 5483; email: firstname.lastname@example.org
We investigated the interplay occurring between pathogens in the course of dual infections, using an in vitro model in which the THP-1 monocytic cell line is first infected with HSV-1 and then exposed to Ca or Cn. These three pathogens share some pathogenic features: they cause opportunistic infections, target macrophages and are neurotropic. Here, we show that HSV-1-infected THP-1 cells exhibited augmented phagocytosis against the two opportunistic fungi but reduced capability to counteract fungal infection: the better ingestion by monocytes was followed by facilitated fungal survival and replication. Reduced IL-12 production was also observed. Cytofluorimetric analysis showed that HSV-1-infected monocytes exhibit: (i) downregulated TLR-2 and TLR-4, critical structures in fungal recognition; (ii) reduced expression of CD38 and CD69, known to be important markers of monocyte activation; and (iii) enhanced expression of apoptosis and necrosis markers, in the absence of altered cell proliferation. Overall, these findings imply that HSV-1 infection prevents monocyte activation, thus leading to a significant dysfunction of the monocyte-mediated anti-Candida response; HSV-1 induced apoptosis and necrosis of monocytes further contribute to this impairment.
Macrophages play a key role as first-line defense against microorganisms and represent a crucial link between innate and adaptive immunity. Macrophage response strictly depends on an adequate recognition of the pathogen, consisting of non-successful clonal interaction(s) between some PRR on host cells and well-conserved structures on the microbial surface, the so-called PAMP (1–3). Among numerous PRR, the TLR have acquired increasing importance (4, 5). TLR-2 and TLR-4 are important PRR in Ca infections (6–10), whereas TLR-2 seems primarily involved in Cn recognition (11, 12). Upon binding, phagocytosis is accomplished and followed by release of pro-inflammatory cytokines. Increased phagocytosis and release of reactive oxygen species likely cooperate in eliminating the pathogen (13, 14).
TLR also play an important role in the innate response against viruses, including HSV-1. TLR-2 recognizes HSV-1 envelope glycoproteins whereas TLR-9 recognizes CpG motifs present in the viral genome (15). Triggering of these PRR by HSV-1 induces the release of numerous cyto-chemokines that exert a dual effect as precious first-line response, and as deleterious signals heavily contributing to HSV-1 pathogenesis (16).
Despite the complex network of soluble and cellular antimicrobial systems, pathogens may elaborate successful strategies through which macrophage-mediated defenses are eluded. In this respect, two important opportunistic fungal pathogens, such as Ca and Cn, have been described as facultative intracellular parasites targeting macrophages within which they are capable of survival and replication. It is reasonable to assume that, in vivo, any event resulting in imbalanced macrophage homeostasis may directly affect the outcome of infections by these fungi, because of the well-established role of professional phagocytes as first-line defense (17) and their potential significance as reservoir for survival and dissemination of fungal cells. In this respect, polymicrobic infections, occurring especially in immunocompromised hosts, may be of relevance (18–23). It is possible that pathogen–pathogen interactions occur, directly or indirectly, disturbing host cell functions and, in turn, exacerbating disease outcome. Recent reports describe clinical cases of dual infections by HSV-1 and different fungi, including Ca and Cn (20, 21).
By an in vitro model of dual infection, we recently reported that human HHV-6 dysregulates monocyte-mediated anticryptococcal defenses in vitro with an overall pro-cryptococcal effect (24). Here, we show that monocytes exposed to HSV-1 and then infected with either Ca or Cn exhibit augmented phagocytosis, but their capability to counteract fungal survival and replication is reduced. Some of the molecular mechanisms possibly involved in such dysfunction have been investigated and will be discussed.
MATERIALS AND METHODS
The human acute monocytic leukemia cell line THP-1 was maintained in RPMI-1640 medium with 10% FBS, 2 mm l-glutamine, 100 U/mL penicillin and 100 μg/mL streptomycin, at 37 °C and 5% CO2. Cells were split twice a week in fresh medium.
We used a clinical isolate obtained from blisters of a case of herpes labialis, initially identified by indirect IFA using a MAb specific for HSV-1. This strain was laboratory adapted through serial passages (>50) on VERO cells. Viral stocks on THP-1 cells were prepared and harvested when most cells (>80%) showed a cytopathic effect: the supernatants from the infected cultures were centrifuged and aliquoted for −80 °C storage. The viral stocks were titrated by plaque assay on VERO cells. Stocks for mock infection were prepared in the same manner from the supernatants of uninfected THP-1 cells.
The non-germinative low-virulence strain PCa2 of Ca was used (25). This strain was a kind gift from Dr Kerridge, Department of Biochemistry, University of Cambridge, UK. For the experiments with Cryptococcus we used a clinical isolate of Cn var. neoformans (serotype D) (laboratory name: 1526), obtained from the cerebral spinal fluid of a patient with meningitis (26). Both fungi were maintained by weekly passages on Sabouraud Dextrose agar plates.
Evaluation of HSV-1-infected cells and measurement of viral load
THP-1 cells were infected with HSV-1 at a 10 PFU/cell MOI and incubated overnight. The rate of HSV-1-infected cells was then evaluated by IFA and by RealTime PCR TaqMan (Applera-Applied Biosystems, Monza, Italy). IFA was carried out according to a protocol reported elsewhere (27), using a MAb against HSV-1 gC glycoprotein (Argene, Varilhes, France). For quantitative PCR, DNA was extracted from 7 × 105 cells by a QIAmp DNA Mini Kit (QIAGEN, Milan, Italy) and then HSV-1 DNA quantified by TaqMan with an ABI Prism 7000 (Applied Biosystems, Milan, Italy). The primers used to amplify HSV-1 DNA target the gD gene according to a protocol by Weidmann (28). In order to obtain a standard curve for target quantitation, a 2055-bp insert of the gD gene was cloned into a ptZ18U vector. Beta-actin was used as a housekeeping gene for DNA content normalization.
The phagocytosis assay was carried out as previously detailed (29). THP-1 cells were seeded at a concentration of 106 cells/mL and then infected with HSV-1 as described above. Uninfected cells (negative controls) were exposed to mock inocula. The following day, HSV-1-infected and -uninfected cultures were exposed to Ca or Cn (107 cell/mL, effector: target = 1:10) in 24-well plates and then incubated for 2 and 4 hr. Other effector: target ratios (E:T ratio) were also tested (1:5, 1:20). An excess of microorganisms was removed by centrifugation of the cell suspension on a Ficoll cushion at 300 g for 10 min. The cells at the interface were recovered and washed. Fungal uptake was directly evaluated in May–Grunwald–Giemsa-stained cytospin preparations. A minimum of 200 monocytes were scored, and any cells containing one or more yeasts were counted as phagocytic. Phagocytosis index was also calculated as the mean value of internalized fungal cells/phagocytic cell (i.e. total phagocytosed fungi/total number of phagocytic cells).
Anticandidal activity was evaluated by means of the CFU inhibition assay. HSV-1-infected and -uninfected THP-1 cultures (5 × 105 cells/mL) were exposed to Ca or Cn (5 × 104 cells/mL, E:T = 10:1) in 96-well microplates. After 6 hr of incubation at 37 °C, Triton X-100 was added at 0.2% final concentration. Serial dilutions from each well were plated on Sabouraud agar plates in triplicate. After 48 hr of incubation at room temperature (RT) the CFU number was assayed. Control cultures consisted of Ca or Cn without effector cells. The results reported as arithmetical mean of CFU number were expressed as percentage of anticryptococcal activity according to the formula:
IL-12 and nitric oxide quantification
Supernatants of THP-1 cells infected with HSV-1 or mock inoculum for 16 hr and then exposed to Ca were assayed for IL-12 and nitric oxide (NO). For IL-12 quantification, a commercial sandwich ELISA kit (Euroclone IL-12 p40, Euroclone, Pero, Milan, Italy) was used according to the manufacturer's instructions. NO was assayed by the Griess reaction, measuring the stable oxidation product NO2− 24 and 48 hr after virus infection. Optical density was read at 570 nm in a UV/VIS spectrophotometer (Beckman DU530; Beckman Coulter, Cassina de’ Pecchi, Milan, Italy). A standard curve, prepared using sodium nitrite in a range between 3 and 200 μm, allowed us to determine the amount of NO product in each sample. As positive controls, cells treated with LPS from Escherichia coli serotype 0128:B12 (1 μg/mL) and with human recombinant γ-interferon (100 U/mL) were assayed.
For the simultaneous analysis of changes in mitochondrial membrane potential and other markers related to cell death, we applied a recently developed multicolor flow cytometric technique (30). Briefly, THP-1 (5 × 105) cells were suspended in RPMI-1640 and stained with 2.5 μg/mL JC-1 (Invitrogen, Carlsbad, CA, USA) for 10 min at RT. Cells were then washed with PBS and resuspended in 195 μL Annexin binding buffer (10 mm HEPES, 140 mm NaCl and 2.5 mm CaCl2, pH 7.4); 5 μL Alexa 647-conjugated Annexin-V (Invitrogen) was added for phosphatidyl serine detection. Cells were then incubated for 15 min at room temperature, washed with 800 μL Annexin binding buffer (as detailed above) and resuspended in 1.5 mL of the same buffer. Propidium iodide (PI) (Sigma Aldrich, St Louis, MO, USA) was added at a final concentration of 1 μg/mL 2 min before acquisition on a cytometer to exclude debris and dead cells from the analysis (31).
For cell surface antigen detection, the following MAbs were used: anti-CD38 PE (BD Pharmingen, Franklin Lakes, NJ, USA); anti-CD95 PE (eBiosciences, San Diego, CA, USA); anti-CD69 PE, anti-CD11b APC, anti-TLR-2 PE, anti-TLR-4 PE, anti-CD54 APC and anti-CD106 APC (Serotec, Oxford, UK). THP-1 (5 × 105) cells were stained by using standard methodologies described elsewhere (32). The absolute expression of each antigen was calculated by subtracting the median fluorescence intensity (MFI) of the unstained sample (i.e. that with an irrelevant mAb conjugated with the same fluorochrome in order to eliminate the interference of non-specific binding to Fc receptors) from the MFI of the corresponding test sample. Both values were expressed in linear scale.
To assess cell proliferation, the probe 5-(6)-carboxyfluorescein diacetate, succinimidyl ester (CFSE) (Invitrogen) was used. THP-1 cells were resuspended at a density of 1 × 106/mL in PBS and stained with 2 μm CFSE for 10 min at 37 °C; afterwards, cells were washed with complete culture medium in the presence of 10% FBS to remove unbound CFSE. A sample was immediately collected and CFSE intensity was determined (before culture condition, Fig. 1). Cells were then infected with HSV-1 as reported above, incubated overnight and exposed to Ca. Finally, cultures were harvested and CFSE dilution was determined. PI (final concentration 1 μg/mL) was added to the cell suspension to remove dead cells. Relative cell proliferation was calculated by subtracting the MFI of the CFSE-stained sample prior to incubation from the MFI of the test samples. Both values were expressed in linear scale.
Flow cytometric analysis was carried out using a CyFlow ML (Partec GmbH, Münster, Germany). Fluorescence compensation was carried out by using samples stained with single fluorochromes. Data were analyzed by FlowJo 6.3 software (Treestar, Ashland, OR, USA).
As detailed in the figure legends, data reported in figures and tables are either the results of a representative experiment or the mean values (± standard deviation) from at least three different experiments performed. Phagocytosis and killing results were analyzed by the Mann–Whitney U-test. When analyzing cell surface antigen expression and apoptosis markers the two-tailed Student's t-test was used. Results were considered significant when P < 0.05.
HSV-1 can productively infect THP-1 cells
IFA and real-time PCR experiments were carried out to evaluate the rate of HSV-1-infected cells and the viral load in THP-1 cells exposed overnight to cell-free viral stocks. As shown in Table 1, using a virus inoculum of 10 PFU/cell, more than 80% of THP-1 cells were IFA positive 16 hr after infection. DNA quantitation disclosed the presence of 3 × 106 HSV-1 DNA copies/105 cells. Uninfected control cells exposed to mock inocula were consistently negative (data not shown).
Table 1. HSV-1 infection of THP-1 cells
Hours after infection
†% of positive cells; ‡genome copy/cell.
THP-1 cells (106 cells/mL) were infected with HSV-1 at a 10 PFU/cell MOI. IFA was carried out 6 and 16 hr post-infection using a monoclonal antibody against gC protein. real-time PCR TaqMan was carried out after 16 hr incubation using primers and probe recognizing the gD gene.
n.t., not tested.
HSV-1 infection enhances phagocytic activity
In order to investigate whether HSV-1 infection could influence macrophage effector functions, THP-1 cells were exposed to virus or mock inocula and then assessed for phagocytosis against Ca. As shown in Figure 2, THP-1 cells infected overnight with HSV-1 disclosed a remarkable increase in the percentage of phagocytic cells in comparison with mock-treated THP-1 cells both in the 2 hr (49.8%vs 36%) and the 4 hr (72%vs 51.5%) phagocytosis assay. The phagocytosis index was also increased: 2.1 vs 4.3 at 2 hr, 2.4 vs 5.1 at 4 hr. A similar trend was observed when the phagocytic activity was assessed against Cn. Both the percentage of phagocytic cells and the phagocytosis index were always higher in HSV-1-infected cells than in the uninfected controls, both in the 2 and the 4 hr assay (Fig. 3). Similar results were obtained with different ET ratios (data not shown).
HSV-1 infection impairs antifungal activity
THP-1 cells exposed overnight to HSV-1 or mock inocula were tested for antifungal activity against Ca and Cn by a 6 hr CFU inhibition assay. The anticandidal activity of HSV-1-infected cells disclosed a drastic decrement down to negative values in comparison with mock-treated THP-1 cells (−42.2%vs 9.3%). Similar results were obtained in terms of anticryptococcal activity: again, a decrease to negative levels was observed in HSV-1 macrophages with respect to uninfected controls (−60% vs 2%) (Fig. 4).
HSV-1 infection downloads IL-12 production
The levels of IL-12 and NO were measured in the culture media of THP-1 cells exposed overnight to HSV-1 or mock inocula and then to Ca. NO was not produced up to 48 hr after virus infection, nor by cells stimulated with LPS or IFN (data not shown). Table 2 shows the amounts of IL-12 detected 24 hr after HSV-1 infection: in both cell groups infected by HSV-1, IL-12 was decreased in comparison with uninfected cells.
Table 2. IL-12 production by THP-1 cells
†Expressed as pg/mL; ‡1 μg/mL; §100 U/mL.
THP-1 cells (106 cells/mL) were infected with HSV-1 at a 10 PFU/cell MOI or exposed to mock inocula. After overnight incubation, cells were infected with Ca (effector: target ratios [E:T = 1:10]) in 24-well plates. After 24 hr from virus infection, supernatants were tested for IL-12 by a commercial ELISA kit.
195.5 ± 20.4
THP-1 + HSV-1
THP-1 + HSV-1 +Ca
156.2 ± 25.5
THP-1 + LPS‡
THP-1 + IFN-γ§
349.5 ± 38.0
HSV-1 induces apoptosis in THP-1 cells
In order to asses whether the HSV-1-induced dysfunction in killing by THP-1 cells could be due to changes in cell viability, multiple apoptotic markers were simultaneously investigated by polychromatic flow cytometry. THP-1 cells, exposed overnight to HSV-1 or mock inocula, were challenged with Ca for an additional 4 hr. As shown in Figure 5(a) and in Figure 6, unlike mock treatment, overnight infection with HSV-1 significantly increased the number of both apoptotic (AnxV+/PI−) and necrotic (AnxV+/PI+) cells. Moreover, induction of apoptosis was associated with altered mitochondrial membrane potential as shown by the loss of ΔΨm in these cells (Fig. 5b). Infection with Ca per se did not trigger cell death or alterations in ΔΨm nor potentiated HSV-1-induced apoptosis or loss of ΔΨm in THP-1 cells.
HSV-1 infection downregulates activation markers and PRR
We finally investigated HSV-1 and mock-treated THP-1 cells for the expression of surface markers known to be involved in cell activation and adhesion. Dead cells were excluded from the analysis. Figure 7 shows that THP-1 cells constitutively expressed CD69 and CD38 activation markers but not CD95. In addition, THP-1 cells expressed low levels of the adhesion molecules, CD11b and CD54, but not CD106. TLR-2 and TLR-4 were also abundantly expressed. Overnight infection with HSV-1 downregulated CD69, CD38 and TLR-2; indeed, MFI values for these antigens (mean ± SEM of five experiments) were: 1033 ± 140 vs 430 ± 83 (P < 0.02) for CD69, 1487 ± 112 vs 327 ± 43 (P < 0.02) for CD38, and 818 ± 61 vs 323 ± 71 (P < 0.01) for TLR-2. The same trend was also observed for TLR-4, even though the difference did not reach a statistical significance (ctrl vs HSV-1 MFI: 354.0 ± 101.7 vs 211.7 ± 27.7; P > 0.05). CD11b and CD54 were not influenced by the infection with HSV-1. The presence of Ca alone or as co-infection did not alter the expression of any of the antigens analyzed, nor modulated the pattern of cell surface markers induced by HSV-1.
CD38 and CD69 are directly involved in cell activation and are also expressed by proliferating cells (33–38). In order to investigate whether the HSV-1-induced CD69 and CD38 downregulation was due to the virus infection per se or rather was a consequence of a possible inhibitory effect on cellular proliferation, THP-1 cells were first stained with CFSE, successively infected overnight with HSV-1 and then exposed or not to Ca. CFSE dilution was then determined by flow cytometry. At the moment of analysis, necrotic cells were discriminated by gating on PI negative cells in order to exclude any possible artifact on CFSE distribution due to dye loss from this population. As shown in Figure 1, the initial level of CFSE (indicated as ‘Before culture’ in figure) was diluted in HSV-1-infected cells at the same extent as in control cells even though a consistent amount of AnxV+/PI− cells could be detected after infection with HSV-1 (Figs 5,6). Similarly, infection with Ca did not influence CFSE dilution. These data exclude proliferation arrest as a cause for decreased expression of activation antigens.
The present study demonstrates that macrophages can be productively infected by HSV-1 and that such an event is accompanied by altered antifungal defenses; namely, that monocytes exert augmented phagocytosis and decreased killing against Ca and Cn. The efficiency of HSV-1 infection in THP-1 cells is demonstrated by the high rate of Ag-positive cells in IFA and by the real-time PCR results. Moreover, titration by plaque assay on VERO cells provided direct evidence that HSV-1 infection of THP-1 cells is productive (data not shown). We have recently observed an analogous phenomenon in macrophages exposed to HHV-6 or to a transmissible cytotoxic activity (39, 40). Also, in those cases, macrophages display augmented phagocytosis but decreased killing against fungi; such effects have consistently been observed using either human or murine cell lines or, rather, primary macrophages (i.e. peritoneal macrophages). Taken together, our previous and the present data suggest that, irrespective of the different noxae used, a similar trend of phagocyte-mediated dysfunctions may occur. It is worth noting that both the fungal pathogens and the viral agents used in our studies commonly target macrophages, within which facultative or obligate intracellular replication is pursued. As recently established by Alvarez et al. (41), Cn is able to survive and replicate inside macrophages and eventually escape via phagolysosomal extrusion. Accordingly, the finding that herpesvirus-infected cells exhibit increased phagocytosis but lack killing ability suggests that, in the course of dual infections, the virulence of a facultative intracellular pathogen such as Cn may be greatly enhanced by the co-presence of a viral agent.
The flow cytometry analysis provides information on the mechanisms involved in the monocyte dysfunction described above. Here, we demonstrate that HSV-1-infected monocytes showed: (i) downregulation of TLR-2 and TLR-4; (ii) reduced expression of CD38 and CD69; and (iii) enhanced levels of apoptosis and necrosis markers. TLR-2 and TLR-4 are widely expressed in immune cells and act as key receptors for different fungi, including candida and cryptococci (7, 9, 10). In particular, TLR-2 interacts with candida β-glucans and phospholipomannans, whereas TLR-4 recognizes fungal mannans (8). Following PRR-PAMP recognition, cytokine production is triggered, thus leading to activation of innate immunity as well as polarization of a lymphocytic response towards adaptive immunity. Furthermore, TLR-2 and TLR-4 likely play a role in non-opsonic phagocytosis. Preliminary results of a microarray study we are carrying out suggest that both opsonin- and non-opsonin-mediated phagocytosis are greatly affected by HSV-1 in a rather complex manner: on the whole, opsonic phagocytosis seems to be prevalently inhibited, whereas many genes involved in non-opsonin-mediated phagocytosis are upregulated (C. Cermelli et al., unpubl. data). In a previous work, we demonstrated that genetically engineered TLR-2−/− macrophages show increased Ca phagocytosis with respect to the wild-type TLR-2+/+ counterpart (6). Those data imply that TLR-2 is either detrimental or, rather, it may interfere with the activity of other receptor(s) directly involved in candida recognition/ingestion. The present results, coupling the HSV-1-induced downregulation of TLR-2 with the increased fungal ingestion, further support the idea that TLR-2 may play a detrimental role in macrophages handling fungal pathogens. Nevertheless, unlike our previous report on TLR-2−/− macrophages (6), in the present model, TLR-2 downregulation is coupled with a consistent impairment of antifungal activity. This unexpected result can be interpreted in the light of the significant downregulation of CD38 and CD69 observed in HSV-1-infected macrophages. Both such surface molecules are important markers of monocyte activation (33–38). CD38 is a 45 kDa multifunctional ectoenzyme endowed with ADP-ribosyl cyclase and hydrolase activities, contributing to the regulation of calcium exchange (42). The CD38 molecule plays a central role in monocyte activation, proliferation and selectin-type adhesion to endothelial cells (43). Moreover, CD69 is a member of the natural killer cell gene complex family of signal-transducing receptors (37). As CD38, it is involved in calcium intake as well as in different other pathways, including monocyte-mediated production of nitric oxide, a key molecule in killing of the ingested pathogens (44). In our model, the CD38 and CD69 downregulation induced by HSV-1 may indicate a hyporesponsiveness of THP-1 cells, that, in turn, may account for the failure in the clearance of the phagocytosed fungi. The link between TLR-2 downregulation and this impairment of antifungal defenses is also strengthened by the lack of IL-12 production in HSV-1-infected THP-1 cells: IL-12 is a fundamental pro-inflammatory and immunomodulating cytokine playing a significant antifungal role (45). As also seen by others (46), TLR-2-deficient monocytes disclose a decrease in IL-12 production as well impaired antifungal defenses. Moreover, the increment in apoptotic and necrotic cells occurring in the HSV-1-infected population may further contribute to the observed dysfunction. The analysis of CFSE dilution by cell cultures after exclusion of necrotic cells indicates that cell death is not accompanied by altered proliferation in the cell population spared by apoptosis. This result, excluding proliferation arrest as a cause of the decreased expression of activation antigens, supports the hypothesis of failure in cell activation. Similarly, HSV-1 has been recently demonstrated to inhibit dendritic cell maturation and activation before apoptosis is induced (47).
In conclusion, herpesviruses are known to greatly impact host immune response (16, 47–51). Here, we show that HSV-1 causes dysfunction of macrophage-mediated antifungal defenses leading to a disturbance of an important branch of the innate immunity. Although the in vivo correlates of these interactions cannot be automatically assumed, it is intriguing to speculate an increased risk of disseminated mycoses during active viral infections. It is worth noting that the pathogens used in our study commonly affect the central nervous system: HSV-1 and Cn are responsible for severe cases of encephalitis and meningitis, mainly observed in immunocompromised hosts, whereas invasive candidiasis may evolve as meningoencephalitis in neonates and transplant patients. If our in vitro data mirror in vivo situations in which phagocytes, due to whatever factor, exhibit increased microbial ingestion followed by inefficient killing, we may speculate that such a phenomenon will favor pathogen survival and dissemination and, ultimately, disease progression.
Different from what we have shown here, Barton (52) showed that murine γ-herpesvirus 68 upregulates γ-interferon production and activates macrophages during latent infection, thus allowing protection against Listeria monocytogenes and Yersinia pestis. Thus, the interplay between phylogenetically distant microorganisms causing dual infection has to be rather complex, not always unidirectional and brings clear-cut advantages to one or both pathogens. Further investigations are needed for a deeper understanding of the mutual relationships between microorganisms and their role in the outcome of polimicrobic diseases.
This work was supported by a grant by MIUR, PRIN 2005, N. 2005068754–003.