A Decline in CCL3-5 Chemokine Gene Expression during Primary Simian-Human Immunodeficiency Virus Infection

Background The CC-chemokines CCL3, CCL4 and CCL5 have been found to block the entry of CCR5-tropic HIV into host cells and to suppress the viral replication in vitro, but the in vivo role of endogenous CC-chemokines in HIV-1 infection is still incompletely understood. Methodology/Principle Findings In this study, the primate host CCL3, CCL4 and CCL5 gene expression was evaluated in response to simian-human immunodeficiency virus (SHIV) infection in rhesus macaque model. Five rhesus macaques were inoculated with CCR5-tropic SHIVSF162P4. The mRNA levels of CCL3, CCL4 and CCL5 were measured by real-time PCR at post inoculation day (PID) 0, 7, 14, 21, 35, 56 and 180 in peripheral blood. In addition, a selected subset of samples from CXCR4-tropic SHIVKu1-infected macaques was included with objective to compare the differences in CC-chemokine down-regulation caused by the two SHIVs. Gut-associated lymphoid tissues (GALT) collected from SHIVSF162P4-infected animals were also tested by flow cytometry and confocal microscopy to corroborate the gene expression results. Predictably, higher viral loads and CD4+ T cell losses were observed at PID 14 in macaques infected with SHIVKu1 than with SHIVSF162P4. A decline in CC-chemokine gene expression was also found during primary (PID 7-21), but not chronic (PID 180) stage of infection. Conclusions It was determined that A) SHIVSF162P4 down-regulated the CC-chemokine gene expression during acute stage of infection to a greater extent (p<0.05) than SHIVKu1, and B) such down-regulation was not paralleled with the CD4+ T cell depletion. Evaluation of CC-chemokine enhancing immunomodulators such as synthetic CpG-oligonucleotides could be explored in future HIV vaccine studies.


INTRODUCTION
Chemokines play a critical role in the host defense against viruses by mobilizing leukocytes to sites of infection, and they also play a homeostatic role in secondary lymphoid tissues [1]. Structurally, chemokines are divided into the C, CC, CXC, and CX 3 C subclasses on the basis of their N-terminal cysteine residues spacing [2]. The majority of known chemokines belong to CC or CXC category [2]. Since CC chemokine receptor 5 (CCR5) and CXC chemokine receptor 4 (CXCR4) were identified as major coreceptors for HIV-1 entry into target cells, it was suggested that these chemokine receptors and their ligands are involved in the transmission and replication of HIV-1 [3,4]. HIV-1 isolates that use CCR5 as co-receptor (R5 viruses) are predominantly transmitted and persist throughout the infection. By contrast, HIV-1 isolates that use CXCR4 (X4 viruses) commonly emerge in infected individuals at later stages of infection [3]. The importance of CCR5 for HIV-1 transmission was underscored by observation that certain individuals who had been repeatedly exposed to HIV-1 but remained uninfected had a defect in CCR5 expression [5]. It has been documented that mucosa-associated CD4+ T cells, macrophages, and dendritic cells express CCR5 which makes them prime targets for the virus [6][7][8].
Members of the CC-chemokine family, such as macrophage inflammatory protein (MIP)-1a (CCL3), MIP-1b (CCL4) and RANTES (CCL5) are the natural ligands for CCR5. They have been shown to inhibit HIV-1 entry into host cells in vitro by competing with viral Env protein for binding [9][10][11], and by down-regulating CCR5 surface expression [12]. However, the in vivo role of endogenous CC-chemokines in the pathogenesis of HIV-1 infection is still incompletely understood. A number of studies have analyzed levels of CCR5 ligands in the HIV-1/SIVinfected individuals or monkeys with conflicting results [13][14][15][16][17][18].
Some of these studies concluded that CCR5 ligands may exert protective role in HIV-1/SIV infection [13,[16][17][18], while others suggested an association between elevated levels of CCR5 ligands and HIV disease progression [14,15]. Most of these reports were based on data that correspond to chronic stages of HIV-1/SIV infection, while the relationship between host CC-chemokine expression and HIV-1 acute infection still needs to be studied.
In this study, we evaluated the host in vivo CCL3, CCL4 and CCL5 gene expression in response to CCR5-tropic SHIV infection. We utilized SHIV SF162P4 /rhesus macaque model and focused predominantly on acute stage of infection to measure the CC-chemokine gene expression changes in peripheral blood and GALT. In addition, we included a selected subset of samples from SHIV Ku1 -infected macaques in order to compare the differences between the extent of CC-chemokine down-regulation caused by CCR5-tropic SHIV SF162P4 and more pathogenic CXCR4-tropic SHIV Ku1 at the acute stage of infection.

Viral loads and CD4+ T cell counts in peripheral blood of SHIV SF162P4 -infected macaques
All five SHIV SF162P4 -inoculated macaques became infected as determined by individual viral load measurements. Highest viral loads were detected at PID 14 with an average of ,10 5 of viral copies (Fig. 1A). The lowest average peripheral CD4+ T cell counts were detected between PID 14 and 21 (Fig. 1B).
Down-regulation of CC-chemokine gene expression in peripheral blood of SHIV SF162P4 -infected macaques The changes of CCL3, CCL4 and CCL5 gene expression levels were studied in peripheral blood over the period of half year following the SHIV SF162P4 inoculation. Decreased expression of all three CC-chemokines was found upon infection. The lowest gene expression was detected at PID 14, coinciding with the peak of primary SHIV SF162P4 infection (Fig. 1A, C, D and E). By PID 14, CCL3, CCL4 and CCL5 mRNA levels decreased 5.9-13.6, 3.5-27.5, and 2.3-5.4-folds, respectively. These levels were significantly lower (p,0.005 for CCL3; p,0.05 for CCL4 and CCL5) than those prior to infection (Fig. 1C, D and E). However, such a decline of CC-chemokine gene expression occurred only temporarily, because starting from PID 21 and at later time points these gene expression values showed relative increase. By PID 180, expressions of all three chemokine genes were no longer significantly different from those prior to infection.

Down-regulation of CC-chemokine gene expression in GALT of SHIV SF162P4 -infected macaques
To determine whether the decline of CC-chemokine gene expression occurred also in GALT of SHIV SF162P4 -infected macaques, biopsy samples of colon, jejunum and MLN were obtained at PID 0, 14 and 180. At PID 14, CCL3 and CCL4 mRNA levels in colon, jejunum and MLN were lower (p#0.05) than those at PID 0 ( Fig. 2A and B). CCL5 mRNA decreased in jejunum (p,0.05) but not in colon and MLN. In agreement with results generated with PBMC, down-regulation of CC-chemokines occurred also in GALT at PID 14. By PID 180, expression of CCchemokines returned to pre-infection levels (CCL4 and CCL5), or even increased (CCL3).

In situ enumeration of CCL4+ cells in SHIV SF162P4infected macaques
To corroborate the mRNA results, and to determine the origin and number of CC-chemokine-producing cells in SHIV SF162P4infected macaques, confocal microscopy and MLN tissues were used. The CCL4 was selected as the CC-chemokine representative. The T lymphocytes (CD3+), macrophages (CD68+), and CCL4+ cells were identified directly in situ by tri-color staining (Figs. 3 and 4). Higher number of CD3+CCL4+ cells was seen at PID 0 than at PID 14 (Table 1) while CD3+p28+ cells were seen at PID 14 but not at PID 0 indicating the rapid spread of virus into GALTs following mucosal inoculation (Fig. 3). In addition to CD3+ cells, CD68+ macrophages were also identified as CCL4+ cells (Fig. 4). In agreement with real-time PCR data, the counts of CD3+CCL4+ and CD68+CCL4+ cells were lower (p,0.05) at PID 14 than they were at PID 0 (Table 1).

The extent of CC-chemokine gene down-regulation in blood of SHIV-infected macaques is independent from the level of CD4+ T cell depletion
We hypothesized that SHIV SF162P4 -induced CC-chemokine down-regulation at acute stage of infection was not proportionately related to peripheral CD4+ T cell depletion as a result of viral target cell destruction. Therefore, we included in this study a subset of samples obtained from SHIV Ku1 -infected macaques (PID 0 and 14). In accord with previous reports [19,20], it was confirmed that SHIV Ku1 infection was at PID 14 associated with greater plasma viral loads and more profound loss of peripheral CD4+ T cells than that caused by SHIV SF162P4 (Fig. 6A and B). Interestingly, SHIV SF162P4 caused by PID 14 5.9-13.6, 3.5-27.5, and 2.3-5.4fold CCL3, CCL4 and CCL5 gene down-regulation respectively, while only 2.8-6.2, 2.3-7.8, and 1.2-2.2-fold down-regulation was measured in SHIV Ku1 -infected macaques (Fig. 6C). These differences in CC-chemokine down-regulation between SHIV SF162P4and SHIV Ku1 -infected macaques were significant (p,0.05).

DISCUSSION
Main objective of our study was to evaluate the CC-chemokine gene expression changes in SHIV SF162P4 -infected macaques. We found that at primary/acute stage of infection, CC-chemokine genes were markedly down-regulated, which coincided with the peak of viremia. Gene expression results were consistent with direct measurements of chemokine production-corroborated by confocal microscopy and flow cytometry. Results generated with peripheral blood and GALT samples were also consistent. In agreement with our results, a decline in CCL4 gene expression by CD8+ intraepithelial lymphocytes (IEL) was observed at the acute stage of SIV infection [21]. Study by Wen and colleagues [22] also showed that expression of CCL3, CCL4 and CCL5 decreased in U937 promonocytes in response to HIV-1 infection. In addition, several studies reported up-regulation of CC-chemokines during chronic phase of HIV/SIV infection [14,15,23]. These reports are not in conflict with our results. Results shown here indicate however that down-regulation of CCR5 ligands occurred only at primary (PID 14) but not at chronic stage of SHIV SF164P4 infection (PID 180). At the chronic stage of SHIV SF164P4 infection, expression of the CCR5 ligands returned to pre-infection levels (CCL4 and CCL5), or even increased (CCL3) in all three GALTs.
There are several possible mechanisms involved in decline of CC-chemokines during primary SHIV infection. Since lympho-cytes are CC-chemokine producing cells, the depletion of CD4+ T lymphocytes could be one of the reasons. Based on measurable and significant CC-chemokine gene down-regulation caused by SHIV SF162P4 that is known to be less pathogenic in term of its ability to deplete the peripheral CD4+ T cells than other SHIVs, SIV or HIV, a subset of samples obtained from SHIV Ku1 -infected macaques (PID 0 and 14) was included in this study. It was hypothesized that SHIV SF162P4 -induced CC-chemokine downregulation is not directly related to peripheral CD4+ T cell depletion-as a result of viral target cell destruction. This hypothesis was conclusively corroborated by measurements of CC-chemokine gene expression at PID 14 when SHIV SF162P4 was found to cause significantly greater gene down-regulation than SHIV Ku1 . Such result suggests that decline of CC-chemokines cannot be solely attributed to CD4+ T cell depletion, but it is more complex and likely linked to other pathways triggered by acute SHIV infection. One of them might be related to SHIV gene products, which modulate transcriptional levels of host response genes [24]. For example, HIV accessory gene product Vpr has been reported to suppress the host CC-chemokine gene expression [25]. Another candidate pathway is the interaction between the HIV/SIV gp120 and the CCR5 chemokine co-receptors, which trigger not only viral entry but also other signal transduction cascades that are impacting host CC-chemokine gene expression [26,27]. Elucidating such pathways in context of CC-chemokine production can be subject of future studies.
CCL3, CCL4 and CCL5 function as natural ligands for CCR5, the major HIV/SIV co-receptor [3]. CC-chemokines are known to inhibit the CCR5-mediated HIV/SIV infection by blocking the viral entry into host cells [9,10]. In addition, CC-chemokines play a role in directing the cell movements necessary for the initiation of T cell immune responses [1], co-stimulate the T cell proliferation, and augment the cytolytic capacity of T and NK cells [28,29]. Thus, suppression of CC-chemokines during primary/acute stage of infection may be an event that facilitates the evasion of HIV/ SIV from the host immune response. It was demonstrated by several authors that production of CC-chemokines was associated with vaccine-mediated protection from SIV challenge [13,16,30]. Furthermore, systemic therapy with CC-chemokine homologues decreased SIV mac251 replication in rhesus monkeys [31]. These reports are in agreement with our findings by corroborating the notion that CC-chemokine down-regulation may facilitate virus spread during acute stage of infection.
It may be unexpected that CCL3, CCL4 and CCL5 are downregulated at the acute phase of SHIV infection, since most of the CC-chemokines belong to proinflammatory chemokines and are generally considered to be produced only in response to pathological conditions [32]. However, the constitutive expression of CCL3, CCL4 and CCL5 was detected in monocytes [33] and CD8+ intraepithelial lymphocytes [21]. In this study, highly sensitive real-time PCR was used to measure the expression of the above CC-chemokines in PBMC and GALTs before and after SHIV SF162P4 infection, and it was concluded that these chemokines are down-regulated during primary infection.
In summary, we hypothesize that an interventional upregulation of CC-chemokines might lead to better outcome of HIV/SIV primary infection. Evaluation of CC-chemokine enhancing immunomodulators such as synthetic CpG-oligonucleotides could be explored in future SIV/HIV vaccine studies.

Animals and virus
Five adult rhesus macaques (Macaca mulatta) of Indian origin were inoculated with 5,000 TCID 50 of CCR5-tropic SHIV SF162P4 via intrarectal route [19]. SHIV SF162P4 was obtained from the Simian Retrovirus Core Laboratory at the TNPRC (Tulane National Primate Research Center). Animals were housed at the TNPRC under biosafety level 2 conditions in accordance with the regulations of the American Association for Assessment and Accreditation of Laboratory Animal Care standards. In addition, five adult rhesus macaques of Indian origin were inoculated with CXCR4-tropic SHIV Ku1 by using the same route and dose as described above for SHIV SF162P4 . SHIV Ku1 was obtained from NIH (www.aidsreagent.org). Both SHIVs were expanded and titrated in vitro as described elsewhere [34,35].

RNA isolation and cDNA preparation
Total RNA was isolated from 20-30 mg of GALT tissue or 10610 6 of PBMC using the RNeasy mini kit (Qiagen, Valencia, CA). Prior to RNA isolation, tissue was homogenized by 3 pulses of sonication, each at 60 W for 3 sec. All samples were treated with RNase-free DNase (Qiagen). cDNA was prepared using the random hexamer primers (Integrated DNA Technologies, Skokie, IL) and M-MLV reverse transcriptase (Promega, Madison, WI) according to instructions of manufacturer.  followed. After the thermal cycles, melting curves of the amplicons were generated by heating the reaction mixture to 95uC for 15 sec, then to 60uC for 20 sec, and slowly increasing the temperature to 95uC over a period of 20 min. The fluorescence was measured as a function of temperature. All samples were run in duplicates and the PCRs for the housekeeping glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and the target genes (  [36] was used in calculating the fold-relationships in gene expression between the time points. Pre-infection (PID 0) was used as calibrator to generate the graphs shown in Fig 1 and 2.

Evaluation of viral loads by real-time PCR
Quantitative assessment of viral RNA load in plasma samples obtained from peripheral blood of SHIV-inoculated macaques was determined by real-time reverse transcriptase PCR as described in detail elsewhere [37]. The virus used in this study was accurately and equivalently quantified in this assay. The values were expressed as viral RNA copies per ml of plasma.

Enumeration of CCL4-producing cells in MLN by confocal microscopy
MLN biopsies obtained from 3 animals prior to infection (PID 0) and at PID 14 were cryosectioned to 10-16 mm and subjected to immunofluorescence staining as previously described [38]. Briefly, the tissues were blocked with 10% donkey serum (Sigma, St. Louis, MO) in PBS containing 0.2% fish skin gelatin (Sigma, St. Louis, MO). The sections were subsequently stained with primary antibodies to detect CCL4+ cells, SHIV-infected cells, T cells and macrophages (anti-CCL4, SIV gag p28, CD3, and CD68 antibodies were purchased from R&D Systems, Microbix, Dako and BD Pharmingen, respectively). Negative-control samples were stained with matched isotype control antibodies and no positive cells were identified (not shown). Primary staining was followed by incubation with secondary antibodies conjugated to fluoro-chromes. Images were captured by a Leica TCS SP2 laserscanning microscope (Leica Microsystems, Exton, PA) and viewed using the Leica imaging software [38].

Statistical analysis
CCL3, CCL4 and CCL5 mRNA levels, CCL4+ cell numbers in MLN and percentages of CCL4+ cells in peripheral blood CD4+ and CD8+ T cell populations were compared between selected time points of SHIV SF162P4 -infected macaques by Student's t test and a p,0.05 was considered statistically significant. In addition, the extent of CC-chemokine down-regulation in peripheral blood of SHIV SF162P4 -and SHIV Ku1 -infected macaques was compared at PID 14 by Student's t test (p,0.05).