|Home | About | Journals | Submit | Contact Us | Français|
Mouse adenovirus type 1 (MAV-1) causes acute and persistent infections in mice, with high levels of virus found in the brain, spinal cord and spleen in acute infections. MAV-1 infects endothelial cells throughout the mouse, and monocytes/macrophages have also been implicated as targets of the virus. Here we determined the extent and functional importance of macrophage infection by MAV-1. Bone marrow-derived macrophages expressed MAV-1 mRNAs and proteins upon ex vivo infection. Adherent peritoneal macrophages from infected mice expressed viral mRNAs and produced infectious virus. Infected chemokine (C-C motif) receptor 2 (CCR2) knockout mice, which are defective for macrophage recruitment, did not show differences in survival or MAV-1 load compared to controls. In contrast, macrophage depletion using clodronate-loaded liposomes resulted in increased virus replication in spleens of a MAV-1-resistant mouse strain, BALB/cJ. Thus macrophages serve both as targets of infection and as effectors of the host response.
Mouse adenovirus type 1 (MAV-1) causes a fatal disease in immunocompetent newborn and adult mice at doses as low as 1-100 plaque forming units (PFU) (Guida et al., 1995; Ishibashi and Yasue, 1984; Kring et al., 1995). MAV-1 infection of immunodeficient mice, like human adenovirus infection of immunosuppressed patients, can result in pneumonia, hepatitis, encephalitis, gastroenteritis, and disseminated disease involving multiple organs (Carrigan, 1997; Charles et al., 1998; Moore et al., 2003; Moore et al., 2004; Pirofski et al., 1991).
To understand how MAV-1 disseminates and causes disease in immunocompetent and immunodeficient mice, it is important to know the cell types it infects and how infection affects their normal cell function. MAV-1 targets endothelial cells throughout the mouse, with the highest levels of virus being found in the brain, spinal cord, and spleen (Charles et al., 1998; Guida et al., 1995; Kajon et al., 1998; Kring et al., 1995; Moore et al., 2003). Cells of the monocyte lineage are potentially targets of infection by MAV-1, but their infection by MAV-1 has not been well studied previously. MAV-1 early region 3 (E3) nucleic acid was detected by in situ hybridization in cells histologically identified as macrophages (Kajon et al., 1998). MAV-1 E3 antigen was reported to be detected in Kupffer cells of MAV-1-infected mice (Lenaerts et al., 2005). Macrophage cell lines can be productively infected by MAV-1 in vitro (Spindler et al., unpublished).
Cells of the monocyte lineage are key effectors of the innate immune response. Circulating monocytes differentiate into macrophages, and dendritic cells (DCs) are also derived from monocytes (reviewed in Dale et al., 2008). There are three primary functions of monocytic phagocytes: antigen presentation, phagocytosis of debris and pathogens, and immunomodulation. Due to these functions, monocytic phagocytes also are effector and regulatory cells of the adaptive immune response. When monocytic cells become infected by viruses or bacteria, these innate and adaptive immune functions can be compromised.
To determine whether MAV-1 replicates in primary macrophages or whether it is simply phagocytosed by them, in this work we examined ex vivo infection of bone marrow-derived macrophages. We also isolated peritoneal macrophages and splenic macrophages from MAV-1-infected mice and demonstrated that MAV-1 early and late genes are expressed in both these cell types and infectious virus is produced in peritoneal macrophages. Since macrophages are effector cells of the innate response, and because they are a target of MAV-1 infection, we investigated whether biochemical depletion of macrophages compromised immune function and thus altered the outcome of MAV-1 infection. We quantitated viral loads in brains and spleens of mice of a MAV-1-resistant strain, BALB/cJ, whose macrophages had been depleted with clodronate-loaded liposomes. We also measured viral loads in brains and spleens of MAV-1-infected BALB/cJ mice knocked out for CCR2. Our data indicate that macrophages are both targets of MAV-1 and effectors of the immune response to infection.
We previously showed that MAV-1-specific nucleic acid is detectable by in situ hybridization in macrophages of outbred Swiss mice (Kajon et al., 1998). To determine whether the virus replicates in monocytes, we performed ex vivo infections of bone marrow-derived macrophages and DCs. We cultured primary mouse bone marrow cells from 3-week old mice with either granulocyte-macrophage colony-stimulating factor (GM-CSF/CSF2), which causes monocytes to differentiate into DCs (Zou and Tam, 2002), or with L cell media that provides macrophage colony-stimulating factor (M-CSF/CSF1) for macrophage differentiation (Tushinski et al., 1982). After six days of culture, an aliquot of the cells cultured to produce macrophages was stained for flow cytometry with antibodies to two macrophage surface markers, F4/80 and CD11b. The majority of cells present were positive for both markers (>99%, data not shown). Cells from the GM-CSF-treated cultures were incubated with magnetic beads coupled to CD11c antibody to positively select immature DCs. The CD11c selection routinely resulted in a population of cells that was >90% CD11c+ (data not shown). Some natural killer (NK) cells have also been shown to be CD11c+ (Laouar et al., 2005), so they may have been present in the CD11c+ population from spleens; it is not known whether NK cells are infected by MAV-1. The macrophages and CD11c+ cells from both bone marrow and spleen were then mock-infected or infected with MAV-1 at a multiplicity of infection (MOI) of 5 for 48 hr, and mRNA was isolated and analyzed by reverse transcriptase-polymerase chain reaction (RT)-PCR with MAV-1 E3 or mouse β-actin primers. We reasoned that if the virus were replicating and not simply present subsequent to phagocytosis, viral mRNA would be detected. Figure 1 shows the results for the bone marrow-derived CD11c+ cells (lanes 2 and 3) and macrophages (lanes 5 and 6). E3 mRNA was detected in all the infected samples, indicating that MAV-1 infected both macrophages and CD11c+ cells from bone marrow ex vivo.
We confirmed the result that MAV-1 is transcriptionally active in bone marrow-derived macrophages using intracellular virus staining (ICVS). We modified a protocol of Weaver and Kadan (2000), used for human adenoviruses. Bone marrow-derived macrophages were isolated and infected or mock infected as above, and then fixed, permeabilized, and stained with antibody to F4/80 and either CD11b or MAV-1 E3 gp11K protein. Controls showed that the majority of cells stained with both macrophage markers (Fig. 2C) and there was little background staining (Fig. 2A). F4/80+ cells from infected mice were specifically stained with the anti-E3 gp11K antibody (compare Figs. 2B and 2D). This confirms that macrophages were infected by MAV-1 and indicates that an early protein was expressed.
We next examined whether MAV-1 is replicating in macrophages in infected animals. We isolated peritoneal exudate cells (PECs) of SJL/JCr mice infected intraperitoneally (i.p.) with 104 PFU of MAV-1 for 4 days or from control mock-infected mice. The PECs were allowed to attach for ~3 hrs to tissue culture dishes; 1 hr after plating, 90% of adherent cells are macrophages (Davies and Gordon, 2005). RNA from the adherent cells was isolated and analyzed by RT-PCR with primers specific for genes for MAV-1 E3, MAV-1 hexon (a late viral capsid protein), or mouse β-actin. The results showed that the PECs from infected mice expressed E3 (Fig. 3, lanes 4, 5) and hexon mRNA (lane 9). Similar results were obtained at different times and virus doses in inbred and in outbred Swiss mice (data not shown). Adherent cells isolated from spleens of SJL/JCr infected mice were also positive for MAV-1 E3 (Fig. 3, lane 13). To demonstrate that infection of these cells was productive, we measured infectious virus yields from adherent peritoneal cells from infected SJL/JCr mice by 50% tissue culture infectious dose (TCID50) assays at 1, 48, and 96 hrs after cells had adhered (Fig. 4). Virus yields increased over time, indicating that adherent peritoneal cells were productively infected in SJL/JCr mice by MAV-1. SJL mice are highly susceptible to MAV-1, whereas BALB/c mice are less susceptible (Spindler et al., 2001; Welton et al., 2005). In a similar TCID50 assay of adherent peritoneal cells from BALB/c mice infected at the same dose and time as the SJL/JCr mice (104 PFU, assayed 4 days post infection), we detected no infectious virus (data not shown). However we found similar levels of virus yield from bone marrow macrophages of BALB/c and SJL/JCr mice infected ex vivo, 1.2 × 105 and 2.0 × 105 PFU/ml, respectively. These data suggest that the strain differences in susceptibility between SJL and BALB/c mice are manifested only in vivo in assays of macrophages. The higher susceptility of SJL mice to MAV-1 infection may be a cause or effect of the higher replication levels in vivo seen in peritoneal cells of those mice compared to BALB/c mice.
Based on the results indicating that MAV-1 infects macrophages in mice, we examined whether biochemical depletion of macrophages would alter the outcome of pathogenesis in mice after infection. We used clodronate-loaded liposomes, which deplete macrophages from liver, spleen, lymph nodes, and peritoneum, but not brain or blood monocytes (Van Rooijen, 1989). We treated BALB/cJ or SJL/J mice with i.p. injections of liposomes two days prior to i.p. infection with 102 PFU of MAV-1; these inbred mice are resistant and susceptible to MAV-1 infection, respectively (Spindler et al., 2001; Welton et al., 2005). Eight days post infection, mice were euthanized and organs were harvested for analysis. We confirmed the effectiveness of the clodronate treatment by flow cytometry with antibody to F4/80: clodronate liposomes routinely depleted splenic macrophages to 5-15% of untreated levels (data not shown). The treatment in both strains of mice did not alter clinical signs of infection (data not shown). A few SJL/J infected mice showed neurological signs (seizures) as they were transferred for euthanasia, but this was seen in both control and clodronate liposome-treated mice. Control (PBS-treated) MAV-1-infected BALB/cJ mice had low viral loads and SJL/J mice had high viral loads (Fig. 5A), consistent with these strains being resistant and susceptible to MAV-1, respectively (Spindler et al., 2001; Welton et al., 2005). Viral loads were measured by capture ELISA (Welton and Spindler, 2007). Treatment with clodronate liposomes did not alter the brain viral loads for either strain relative to controls: again levels were low in BALB/cJ mice and high in SJL/J mice (Fig. 5A). Using the Mann-Whitney U test, we determined there was no statistically significant difference between the mean brain viral loads of control and clodronate-treated mice for either mouse strain. This suggests that depletion of those populations of macrophages affected by clodronate liposomes (e.g., peritoneal and splenic macrophages) did not significantly alter the infection as measured in the brain.
In contrast, when we examined spleens, we found that clodronate liposome treatment did affect the infection in BALB/cJ mice (Fig. 5B). The median spleen viral load (ELISA value) for BALB/cJ mice treated with clodronate liposomes was 0.33 OD450 units, compared to 0.04 for controls. This difference was statistically significant by Mann-Whitney U test (P = 0.004). SJL/J mice did not have altered spleen viral loads upon clodronate liposome treatment; the high levels of virus in spleens were consistent with previous results for infection of this susceptible strain (Spindler et al., 2001). The results for BALB/cJ spleens suggest that peritoneal and splenic macrophages are protective against MAV-1 infection.
To address the importance of macrophage function in another way, we determined whether mice genetically defective in macrophage recruitment fared better or worse than control mice when infected with MAV-1. We used CCR2-/- mice, which lack the receptor for the chemokine CCL2 and other ligands (Kuziel et al., 1997). These mice are defective for recruitment of macrophages to the peritoneum (Boring et al., 1997; Kurihara et al., 1997; Kuziel et al., 1997), liver (Dambach et al., 2002; Hokeness et al., 2005), central nervous system (Held et al., 2004), and kidney (Kitagawa et al., 2004). Whether they are defective for recruitment to the lung is dependent on the model (Baran et al., 2007; Lu et al., 1998; Moore et al., 2001), and to our knowledge, effects on other organs have not been reported. CCR2-/- mice on a BALB/cJ strain background or wild-type BALB/cJ mice were infected i.p. with 102 or 106 PFU of MAV-1. Brains and spleens were harvested 8 days post infection and assayed for viral loads (Fig. 6). No significant differences in brain viral loads were seen between control BALB/cJ and CCR2-/- mice at either virus dose, although there were slightly higher virus levels in mice of both strains that received the 106 PFU dose compared to the 102 PFU dose. Similarly, there were no significant differences between spleen viral loads in BALB/cJ vs. CCR2-/- mice at either dose, though in both strains the mice that received 106 PFU had higher viral loads. In addition, CCR2-/- mice were given doses of 104, 105, or 106 PFU of MAV-1 and monitored for 21 days post infection. Morbidity and mortality due to MAV-1 infection usually occur within the first 3-14 days post infection (reviewed in Spindler et al., 2007). The mice survived to 21 days, exhibited no clinical signs, and had no virus detectable by ELISA in brains and spleens at the conclusion of the experiment (data not shown). Together, these data suggest that the lack of CCR2 did not render BALB/cJ mice more susceptible to MAV-1.
Prior to this work there was evidence that MAV-1 proteins and nucleic acid are found in cells of the monocyte lineage in infected mice (Kajon et al., 1998; Lenaerts et al., 2005). Because the methods by which the presence of the virus was detected are only moderately sensitive, we postulated that the virus is not simply phagocytosed, but that it replicates in the cells to levels such that nucleic acid and antigen are detectable. To test this, we used RT-PCR to detect expression of viral mRNAs, which would not be present in non-replicating (phagocytosed-only) virus particles. We first demonstrated that MAV-1 infects macrophages and bone marrow-derived CD11c+ cells (predominantly DCs) ex vivo; both early and late viral gene expression was detected when these monocyte-derived cells were isolated by standard methods and infected with MAV-1. The bone marrow-derived CD11c+ cell population was not 100% pure (it was > 90% CD11c+), and NK cells can also be CD11c+, so further experiments will be needed to determine whether DCs are infected ex vivo by MAV-1. The intracellular virus staining assay provided strong corroborating evidence of the finding that macrophages are expressing early MAV-1 genes: F4/80+, CD11b+ cells were positive for early protein expression by this assay.
We also examined whether MAV-1 infects macrophages in mice. We found that adherent peritoneal and splenic cells from infected mice expressed mRNA for MAV-1 E3, an early gene. It was possible that viral genes were expressed but infectious virus was not produced, so we also assayed for infectious virus. We were unable to detect virus production from adherent peritoneal cells by plaque assays, which have a limit of detection of 2 × 102 PFU/ml. Thus we turned to a more sensitive assay and demonstrated that these cells did produce a low level of infectious virus. This low level of virus production (1 TCID50 unit/300 peritoneal macrophages) suggests that replication is inefficient and/or that only a small subset of cells is capable of producing mature virus. For comparison, we have determined virus yield from permissive 3T6 fibroblasts to be 90 PFU/cell (data not shown). A block to efficient virus production could occur at many steps subsequent to early viral gene mRNA synthesis, including DNA replication, post-transcriptional events (protein translation and stability), virus maturation, and release. Since the adherent cell populations are enriched (>90%) for macrophages, it is possible that a non-macrophage adherent cell type is responsible for replication by the virus. However, given the results of the intracellular virus staining of F4/80+, CD11b+ cells, we believe macrophages are a target of MAV-1 infection and that they produce infectious virus.
It is perhaps not surprising that MAV-1 is found in mouse macrophages, because human adenoviruses (hAds) are rapidly taken up and cleared by alveolar macrophages when virus is administered to mice via the pulmonary route (Zsengellér et al., 2000), and by macrophages of the liver, Kupffer cells, when given intravenously (Alemany et al., 2000; Lieber et al., 1997; Tao et al., 2001; Wolff et al., 1997). This uptake by Kupffer cells has been associated with rapid cell death when very large doses of virus are given (Manickan et al., 2006). Uptake of hAds by mouse Kupffer cells involves scavenger receptors rather than integrins or the coxsackievirus and adenovirus receptor, and opsonization by natural IgM antibodies and complement contribute to clearance of the virus (Xu et al., 2008). To our knowledge there are no reports of hAd infection of macrophages in people, but in vitro studies indicate that the virus can bind and enter monocytic cells via integrins (Huang et al., 1996).
Because macrophages are effectors of innate and adaptive immunity, it seemed possible that depletion of macrophages might increase the severity of MAV-1 infection. Alternatively, because macrophages are targets of MAV-1 infection, their depletion might decrease disease severity. Depletion of macrophages generally makes animals more susceptible to pathogens, including fungi, parasites, bacteria, and viruses (Pinto et al., 1991; Qian et al., 1994; Vreden et al., 1993). We investigated the effect of depleting mice of liver, spleen, lymph node and peritoneal macrophages by treatment with clodronate-loaded liposomes (Van Rooijen, 1989). This treatment does not deplete brain or blood monocytes, though in the spleen it has been shown to deplete marginal DCs of the spleen (Leenen et al., 1998). We used a resistant mouse strain, BALB/cJ, to determine whether the treatment would increase pathogenesis (by removing immune effector cells), and a susceptible mouse strain, SJL/J, to determine if the treatment would decrease pathogenesis (by removing target cells) (Spindler et al., 2001; Welton et al., 2005). Viral loads in susceptible SJL/J mice were unchanged by treatment with the liposomes. In contrast, viral loads in the spleens of BALB/cJ mice were significantly higher in macrophage-depleted mice, approaching levels seen in spleens of susceptible SJL/J mice. Thus depletion of macrophages increased the severity of infection in resistant mice, as measured by spleen viral loads. The BALB/cJ brain viral loads did not differ between control- and liposome-treated mice. This could be due to the fact that brain macrophages are not depleted by the clodronate-liposome treatment. This result could also occur if macrophages do not play a role in controlling viral levels in the brain. The experiments described here do not distinguish between these possibilities, which are not mutually exclusive. However, we have observed that MAV-1 infection does result in increased numbers of macrophages in brains of C57BL/6NCr mice (Gralinski and Spindler, unpublished). Expression of F4/80 mRNA is increased in MAV-1-infected mouse brains relative to mock-infected mice (Weinberg et al., 2007). These results suggest there may be a protective role for macrophages in the brain. If macrophages contribute to dissemination of the virus from peritoneum to the brain, then there must also be a second mechanism for dissemination since macrophage depletion did not affect brain viral loads. Treatment of mice intraperitoneally by clodronate liposomes resulted in depletion of 85-95% of splenic macrophages; these numbers are comparable to results obtained by others (van Rooijen et al., 1989). Our data suggest that the residual macrophages were not able to protect BALB/cJ mice from viral replication in the spleen.
In other models that examine the role of macrophages in viral infection, depletion of macrophages by clodronate liposome treatment or other means also results in increased viral replication in the spleen or other organs. For example, in hAd infection of rats, when infection of Kupffer cells is blocked by treatment with polyinosinic acid, there is an increase in infection of surrounding liver cells (Haisma et al., 2008). Clodronate-liposome treatment of mice increases replication of mouse cytomegalovirus in the spleen (Hanson et al., 1999). In addressing whether macrophages function more as targets or effectors in mouse cytomegalovirus infection, the authors concluded that the latter role, particularly the production of cytokines, outweighs the role of macrophages as target cells. Their data suggest that the macrophages may function as “filters” or “sinks” for the virus, keeping other permissive cell types from being infected by the virus. A somewhat analogous result was found for measles virus, for which macrophages are a major target (Roscic-Mrkic et al., 2001). Mice are not a natural host for measles, but in transgenic mice expressing the measles receptor, human CD46, the most efficiently infected peripheral blood mononuclear cells are F4/80+ (macrophages). Depletion of splenic or lung macrophages in these mice resulted in enhanced viral replication in the spleen or lung, respectively. Furthermore, in each organ, there was a concomitant activation and infection of DCs. Thus macrophages contribute to protection of other immune cells from measles infection. We do not know whether macrophages similarly protect other immune cells, such as endothelial cells or DCs, from MAV-1 infection, or whether macrophages protect other surrounding cell types from MAV-1 infection, similar to the protective effect postulated for Kupffer cells in hAd infection (Haisma et al., 2008).
In our macrophage depletion experiments we assayed mice eight days after MAV-1 infection, and thus it is possible that the resulting increased viral load in BALB/cJ spleens was due to lack of both innate and adaptive functions of macrophages. For many cases where virus infection has been examined in the context of macrophage depletion, an early/innate role for macrophages in control of the infection has been deduced (Ben-Nathan et al., 1996; Pribul et al., 2008; Rivera et al., 2007; Seiler et al., 1997; Zisman et al., 1970; Zisman et al., 1971). However, macrophage depletion also appears to affect the adaptive immune response to some viruses, including ectromelia virus, mouse hepatitis virus A59, and hAd vectors examined in mice (Karupiah et al., 1996; Kuzmin et al., 1997; Kuzmin et al., 2001; Lieber et al., 1997; Shifrin et al., 2005; Wijburg et al., 1997).
In contrast to increased viral loads in the spleens of BALB/cJ mice following macrophage depletion, we saw no differences in spleen or brain viral loads of CCR2-/-mice infected with MAV-1 compared to BALB/cJ mice. Thus clodronate liposome-depleted mice are not functionally equivalent to mice that lack CCR2 with respect to MAV-1 infection. This may be due to differences in immune cell populations in the two models, e.g., DCs. In addition to depleting macrophages, clodronate liposome treatment depletes marginal DCs in the spleen but not CD8+ interdigitating DCs (Leenen et al., 1998). A subset of CCR2+ murine blood monocytes recruited to inflammatory sites can differentiate into DCs (Geissmann et al., 2003); this population would be absent in the CCR2-/- mice. CCR2-/- mice are defective for recruitment of macrophages to the peritoneum after a stimulus, but the resident numbers of macrophages in the peritoneum do not differ compared to wild-type mice (Boring et al., 1997; Kurihara et al., 1997; Kuziel et al., 1997). We suggest that the resident macrophages or other factors (e.g., cells or cytokines) in CCR2-/- mice were sufficient to protect mice from MAV-1 replication in the spleen upon i.p. infection. In most infection models using CCR2-/- mice there are increases in replication of the infectious agent compared to control mice, for example, infection with Toxoplasma gondii, Cryptococcus neoformans, Listeria monocytogenes, Aspergillus fumigatus, mouse cytomegalovirus, and neurotropic mouse hepatitis virus (Benevides et al., 2008; Blease et al., 2000; Chen et al., 2001; Dunay et al., 2008; Held et al., 2004; Hokeness et al., 2005; Kurihara et al., 1997; Traynor et al., 2000). However, in the case of influenza virus, infection of at least one CCR2-/- mouse strain showed no increases in influenza virus titers or delayed viral clearance, though there were clear changes in leukocyte recruitment compared to controls (Wareing et al., 2007). These authors suggested that the deficiency in CCR2 may have altered the development of immune responses without altering the outcome of influenza virus infection. Scott and Flynn (2002) found that for Mycobacterium tuberculosis infection, the dose affected whether differences were seen between CCR2-/- mice and controls. At low doses, the CCR2-deficient mice had bacterial loads similar to wild-type mice and successfully formed granulomas, whereas at high doses the mice were less able to control infection than wild-type mice. The authors suggest that the normal immune response may be greater than is necessary for control of the infection, such that deficiency in CCR2-/- does not have effects at low doses of bacteria. Our results for MAV-1 may have more similarity to the influenza case, because we found no difference in survival of CCR2-/-mice compared to wild-type mice even at very high doses that resulted in high viral loads in both mouse strains.
The highest levels of MAV-1 in infected mice are found in the brain and spleen (Charles et al., 1998; Guida et al., 1995; Kajon et al., 1998; Kring et al., 1995; Moore et al., 2003). MAV-1 causes encephalitis and breakdown of the blood-brain barrier (Gralinski and Spindler, unpublished; Guida et al., 1995; Kring et al., 1995). The finding that peritoneal macrophages produce infectious virus suggests that these cells may play a role in dissemination of the virus to the brain and spleen after i.p. inoculation, contributing to the pathogenesis of the virus. It will be important to determine how macrophages function in this regard and whether other monocytic cell types (e.g., DCs) are similarly productively infected and contribute to dissemination.
All animal work complied with relevant federal and institutional policies. All mice were maintained in microisolator housing with food and water provided ad libitum. Outbred Swiss mice were obtained from Harlan Laboratories. BALB/cJ and SJL/J mice were obtained from The Jackson Laboratory; SJL/JCr mice were obtained from Charles River Laboratories (NCI Animal Production Program). CCR2-/- mice were a kind gift of Beth Moore. These mice (Ccr2tm1Mae) were originally described by Kuziel et al. (1997) and were subsequently crossed for 8 generations to the BALB/c background. Mice were infected i.p. with the indicated dose of MAV-1 in 100 μl PBS or mock infected with conditioned media similarly diluted. Mice were euthanized by CO2 asphyxiation, and their organs were harvested. MAV-1 was grown and titrated in NIH 3T6 cells, a mouse fibroblast cell line (Cauthen et al., 2007).
Mice were euthanized and an excision was made to expose the peritoneum. Five ml of sterile PBS was injected into the peritoneal cavity using a 25 gauge needle, the abdomen was gently palpated and then the fluid containing peritoneal cells was removed. The cells were centrifuged 5 min at 300 × g, red blood cells lysed in lysis buffer (0.144 M NH4Cl, 0.017 M Tris, pH 7.2) for 2 min at room temperature, then centrifuged 5 min at 300 × g, washed twice in PBS, and resuspended in Dulbecco’s modified Eagle’s medium (DMEM). Cells were then counted, and 2 × 106 cells were plated in 10 ml DMEM containing 10% heat-inactivated calf serum in a 10-cm tissue culture plate, incubated for 3 hr at 37°C and non-adherent cells were removed.
Bone marrow was harvested by flushing femurs of 3-week old mice with 10 ml DMEM using a 26 gauge needle. Cells were centrifuged 3 min at 460 × g and resuspended in DMEM. For macrophage isolation, 5 × 106 cells were plated in 10 ml macrophage medium (DMEM containing 20% fetal bovine serum, 30% L929-conditioned media as a source of colony stimulating factor-1 (Tushinski et al., 1982), 2 mM glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin, and 55 μM 2-mercaptoethanol) in a 10-cm non-tissue culture treated Petri plate. L929-conditioned media was prepared by growing L929 cells for 5 days or until cells were confluent in DMEM with 10% fetal bovine serum, harvesting the media, filtering it through a 0.2 μm filter, and storing at -20°C until use. Cells were incubated at 37°C and fed on day 3 by adding10 ml macrophage media to the existing media and harvested on day 6.
For DC isolation, bone marrow cells were resuspended in DC media (DMEM containing 10% fetal bovine serum and 10 ng/ml GM-CSF [R&D, Minneapolis MN]) at 5 × 105 cells/ml. One ml was plated per well in 24-well plates, with incubation at 37°C. On days 2 and 4, 1 ml of DC media was added to the existing media. Cells were harvested on day 6. Cells were incubated for 15 min at 4°C with CD11c microbeads that were prepared by adding 100 μl of CD11c microbeads (Miltenyi Biotec, Auburn CA) per 108 cells. Cells were then centrifuged 10 min at 200 × g, resuspended in wash buffer (PBS plus 0.5% bovine serum albumin), centrifuged, resuspended in wash buffer, and passed through a MACS column (Miltenyi Biotec, Auburn CA).
Total RNA was extracted with TRI Reagent (Molecular Research Center, Inc., Cincinnati, OH) following the manufacturer’s instructions. Five μg of total RNA was reverse transcribed using random hexamer primers and Moloney murine leukemia virus reverse transcriptase (Invitrogen, San Diego, CA). Four μl of a 20 μl reverse transcriptase reaction were then amplified in a reaction volume of 20 μl, using GoTaq enzyme and buffer (Promega, Madison, WI) with denaturation 3 min at 94°C; 35 cycles of 30 sec at 94°C, 45 sec annealing at 55°C, 30 sec extension at 72°C; then a single 7 min extension at 72°C. Primers for MAV-1 E3 were MAVR24718 and MAVR25148 (Spindler et al., 2001); Hexon primers were HexFo2 (5’CACACCTTTCAGCGAGTTTCA3’) and HexRev2 (5’GCCGGGCCAGGGGTTCAA3’) (Weinberg et al., 2007); mouse actin primers were BetaActinUp (5’TGGAATCCTGTGGCATCCATGAAAC3’) and BetaActinDown (5’TAAAACGCAGCTCAGTAACAGTCCG3’). E3 and actin primers were designed to amplify across an intron sequence and thus enable analysis specifically of cDNA. Samples analyzed for hexon mRNA were treated with Superasin DNAse (Ambion, Austin, TX) during the reverse transciptase reaction. PCR products were analyzed on 7% acrylamide gels electrophoresed 2 hr at 200V.
Cl2MDP (clodronate) was a gift of Roche Diagnostics GmbH, Mannheim, Germany. Multilamellar liposomes containing Cl2MDP were prepared as described (Van Rooijen and Sanders, 1994). Each mouse was given either 400 μl of PBS or clodronate liposomes i.p. two days prior to infection. One hundred PFU of virus diluted in PBS or conditioned media (removed from confluent NIH3T6 cell monolayers 3-4 days after plating) similarly diluted was given i.p. in a 100 μl inoculum.
BALB/cJ and SJL/J splenocytes (7 × 106) were stained with antibodies to F4/80 (Caltag, Burlingame, CA). Stained samples were analyzed using a FACScan™ flow cytometer and CellQuest™ software (BD Biosciences, San Jose, CA) or FACSCanto™ flow cytometer (BD Biosciences, San Jose, CA) and FlowJo software (Ashland, OR). Efficient macrophage depletion was confirmed by quantitating F4/80+ cells and comparing their numbers in clodronate liposome-treated and -untreated (PBS) samples.
Viral loads from infected mice were quantitated by capture ELISA as previously described (Welton et al., 2005; Welton and Spindler, 2007). Previous experiments have shown that this method of quantitation has a high correlation with virus quantitation by plaque assay (Welton et al., 2005). An undiluted virus stock control (3.85 × 107 PFU/ml) was included to ascertain the upper limit of detection and to standardize assays performed at different times.
Yield of virus from adherent macrophages from peritoneal exudates of infected mice were quantitated by a 50% tissue culture infectious dose (TCID50) assay. Adherent cells (6 × 106) pooled from 15 SJL/JCr mice 4 days after infection with 104 PFU of MAV-1 i.p. were sampled at 1, 48, and 96 hr post plating. The cells were harvested in 5 ml of media, freeze-thawed three times, and centrifuged to remove debris. Serial 10-fold dilutions of the supernatants were plated (three or more wells per dilution) on 2 × 104 cells/well of mouse 3T6 cells in 96-well plates. Cytopathic effect was scored 4-6 days after infection. The TCID50 was calculated as the inverse of the dilution at which 50% of the wells showed cytopathic effect, calculated by the method of Reed & Muench (1938). A control virus stock of high titer was titrated by TCID50 and plaque assay, and 1 PFU corresponded to 68 TCID50 units.
We thank Linda Gooding for a helpful discussion early in the design of this project, Jenny Imperiale for technical assistance, and Beth Moore for the CCR2-/- mice, helpful discussions and comments on the manuscript. We also thank Linda Gooding, Mike Imperiale, Tien-Huei Hsu, and Jason Weinberg for review of the manuscript. We are grateful to the University of Michigan Center for Statistical Consultation and Research for advice on statistical analyses. This work was supported by NIH R01 AI023762 to K.R.S.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.