Received 7 March 2006/Accepted 3 August 2006

Transcription

1 JOURNAL OF VIROLOGY, Nov. 2006, p Vol. 80, No X/06/$ doi: /jvi Copyright 2006, American Society for Microbiology. All Rights Reserved. Administration of Fludarabine-Loaded Autologous Red Blood Cells in Simian Immunodeficiency Virus-Infected Sooty Mangabeys Depletes pstat-1-expressing Macrophages and Delays the Rebound of Viremia after Suspension of Antiretroviral Therapy B. Cervasi, 1 M. Paiardini, 1 S. Serafini, 1 A. Fraternale, 1 M. Menotta, 1 J. Engram, 2 B. Lawson, 3 S. I. Staprans, 3 G. Piedimonte, 4 C. F. Perno, 5 G. Silvestri, 2 * and M. Magnani 1 Institute of Biochemistry, University of Urbino, Urbino, Italy 1 ; Department of Pathology, University of Pennsylvania, Philadelphia, Pennsylvania 2 ; Emory Vaccine Center, Emory University, Atlanta, Georgia 3 ; LIPIN Laboratory, Biotechnology Section, University of Messina, Messina, Italy 4 ; and Department of Experimental Medicine, University of Rome Tor Vergata, Rome, Italy 5 Received 7 March 2006/Accepted 3 August 2006 A major limitation of highly active antiretroviral therapy is that it fails to eradicate human immunodeficiency virus (HIV) infection due to its limited effects on viral reservoirs carrying replication-competent HIV, including monocytes/macrophages (M/M). Therefore, therapeutic approaches aimed at targeting HIV-infected M/M may prove useful in the clinical management of HIV-infected patients. In previous studies, we have shown that administration of fludarabine-loaded red blood cells (RBC) in vitro selectively induces cell death in HIV-infected M/M via a pstat1-dependent pathway. To determine the in vivo efficacy of this novel therapeutic strategy, we treated six naturally simian immunodeficiency virus (SIV)-infected sooty mangabeys (SMs) with either 9-[2-(R)-(phosphonomethoxy)propyl]adenine (PMPA) only, fludarabine-loaded RBC only, or PMPA in association with fludarabine-loaded RBC. The rationale of this treatment was to target infected M/M with fludarabine-loaded RBC at a time when PMPA is suppressing viral replication taking place in activated CD4 T cells. In vivo administration of fludarabine-loaded RBC was well tolerated and did not induce any discernible side effect. Importantly, addition of fludarabine-loaded RBC to PMPA delayed the rebound of viral replication after suspension of therapy, thus suggesting a reduction in the size of SIV reservoirs. While administrations of fludarabine-loaded RBC did not induce any change in the CD4 or CD8 T-cell compartments, we observed, in chronically SIV-infected SMs, a selective depletion of M/M expressing pstat1. This study suggests that therapeutic strategies based on the administration of fludarabine-loaded RBC may be further explored as interventions aimed at reducing the size of the M/M reservoirs during chronic HIV infection. The hallmark of human immunodeficiency virus (HIV) infection in humans is a progressive depletion of CD4 T lymphocytes that is due to either direct killing of infected cells (viral cytopathicity) or activation-induced apoptosis of uninfected bystander cells (9, 30, 46, 47). In the majority of HIVinfected individuals, the introduction of highly active antiretroviral therapy (HAART) induces suppression of viral replication, which is associated with at least partial reconstitution of the CD4 T-cell pool and ultimately results in a significant decline in mortality/morbidity. Unfortunately, HAART does not eradicate the infection, and in fact, after its suspension, HIV RNA rapidly rebounds to pretherapy levels, indicating the existence of refractory reservoirs (17, 31, 40, 42, 49). Cellular sources of reemerging HIV include latently infected resting CD4 T cells as well as monocytes/macrophages (M/M) and dendritic cells. Infected M/M persist throughout the course of disease as long-term stable reservoirs able to produce large amounts of virions and to disseminate them in other cells and tissues (6, 22, 24, 33). As such, persistence of active viral replication in M/M despite prolonged antiretroviral * Corresponding author. Mailing address: Department of Pathology, University of Pennsylvania, 705 Stellar-Chance Lab, 422 Curie Blvd., Philadelphia, PA Phone: (215) Fax: (215) gsilvest@mail.med.upenn.edu. treatment represents a major obstacle to HIV eradication (8, 25, 34) It should also be noted that currently available drugs have poor antiviral activity against the M/M compartment (1). Collectively, these considerations define the rationale for designing therapeutic approaches aimed at eliminating, or at least reducing, the levels of active HIV replication in M/M in HIV-infected patients. In previous studies, we found that activation of the JAK/ STAT (signal transducers and activators of transcription) pathways is associated with survival of HIV-infected M/M (28). STAT proteins comprise a family of transcription factors that are normally located in the cytoplasm as inactive forms and that upon binding of extracellular signaling proteins (i.e., cytokine and growth factors) to specific cell surface receptors enter the nucleus and regulate several cellular events, including activation, differentiation, proliferation, cell survival, and apoptosis. Cytoplasmic STAT proteins are activated by tyrosine phosphorylation, a transient and tightly regulated process that results in dimerization, nuclear translocation, and transcriptional activation of genes that control the cellular response (2, 27, 41). We found that levels of phosphorylated STAT1 (pstat1) were three- to fivefold greater in in vitro HIV-infected M/M compared to uninfected M/M (28). Therefore, we proposed that, at least in this in vitro system, the 10335

2 10336 CERVASI ET AL. J. VIROL. enhanced expression and phosphorylation of STAT1 may play a role in the development of a persistent state of active HIV replication in M/M. This hypothesis also implies that pstat1 may be a useful target in designing treatment strategies to selectively eliminate persistently infected M/M and thereby provides a rationale for the use of 9-( -D-arabinofuranosyl)-2- fluoroadenine 5 -monophosphate (fludarabine), a potent antileukemic nucleoside analog that also acts as a potent inhibitor of cells with a low-growth fraction that overexpresses pstat1 (12). To selectively deliver fludarabine to M/M we used red blood cells (RBC) as carriers by taking advantage of the peculiar phagocytic capacity of M/M. Our group has validated the use of RBC as specific carriers for M/M with various compounds and in several different in vitro and in vivo models (13, 15, 28, 29). The fludarabine-loaded RBC technology allows us to overcome the nonspecific toxic effect of fludarabine on actively replicating cells (such as lymphocytes), which are more sensitive than resting cells (such as M/M) to the DNA polymerase inhibition induced by the drug (20, 21). Following its loading in human RBC, fludarabine is converted by the RBC kinases to the corresponding triphosphate derivative, which is the active form of the drug (14). Fludarabine loaded within RBC strongly and selectively contributes, in vitro, to the induction of cell death in HIV-infected M/M, without any cytopathic effect upon nonphagocytosing cells (28). In the present study we analyzed for the first time, in an in vivo model of simian immunodeficiency virus (SIV) infection, the safety and efficacy of a new therapeutic protocol based on the weekly administration of autologous RBC loaded with fludarabine, to clear the infected M/M compartment, in combination with daily administration of 9-[2-(R)-(phosphonomethoxy)propyl]adenine (PMPA), to suppress viral replication taking place in activating/proliferating CD4 T cells. As in vivo model, we used sooty mangabeys (Cercocebus atys) (SMs), a West African monkey species that is a natural host of SIV. Among natural SIV hosts, SMs are particularly relevant for two main reasons. First, cross-species transmission of SIVsmm, the virus infecting SMs, is responsible for the epidemic of HIV type 2 in humans. Second, SIVsmm is the origin of the SIVmac viruses, which cause AIDS in rhesus macaques, whose infection represents the most widely used animal model of HIV infection (3, 19, 37 39). Importantly, SIV infection of SMs is not associated with CD4 T-cell depletion and progression to AIDS, despite chronic levels of viremia that are as high as or higher than those observed in HIV-infected humans and SIVinfected rhesus macaques (4, 36, 44, 45). We chose naturally SIV-infected SMs as a nonhuman primate model to test the effects of fludarabine-loaded RBC on SIV reservoir compartments, based on preliminary data from our laboratory as well as from other groups (I. Pandrea and R. S. Veazey, personal communication) indicating the presence of SIV-infected M/M in the lymph nodes (LN) and intestines of these animals. We found that administration of fludarabine-loaded autologous RBC is a well-tolerated procedure that appears to delay the rebound of SIV replication which follows PMPA suspension, thus suggesting a reduction in the number of SIV reservoirs. In addition, we found that administration of fludarabine-loaded autologous RBC selectively depletes macrophages in chronically SIV-infected SMs, but not in uninfected ones, via a pstat1-dependent pathway. MATERIALS AND METHODS Animals. The 13 animals included in this study were selected from the colonies of SIV-infected and uninfected SMs housed at the Yerkes National Primate Research Center of Emory University, Atlanta, GA, and were maintained in accordance with NIH guidelines. All infected animals were studied during chronic infection. Based on longitudinal serologic surveys, the majority of infected SMs studied are known to have acquired SIV by 3 to 5 years of age (16). All animal studies were approved by the Emory University Institutional Animal Care and Usage Committee. The characteristics of the studied animals 3 weeks before the beginning of the study are shown in Table 1. Drugs. Fludarabine and PMPA were obtained from Schering S.p.a. (Segrate, Milan, Italy) and Gilead Sciences (Foster City, California), respectively. PMPA was administered intramuscularly at a dose of 30 mg/kg of body weight per day for 28 consecutive days. Fludarabine was encapsulated in SM RBC at a final concentration of mol/ml RBC and was administered weekly for four consecutive weeks. Lymph node biopsies. For lymph node biopsies, the skin over the inguinal or axillary lymph node was prepared for aseptic surgery. A small skin incision was made over the node, and blunt dissection was used to isolate and remove the node. The subcutis and skin were then closed with absorbable sutures. Either ketamine (10 mg/kg) or Telazol (4 mg/kg) was used for anesthesia. The frequency of administration was determined by the veterinarian performing the procedure so as to maximize animal safety and comfort. SIV viral load. Quantitative real-time reverse transcription-pcr assay to determine SIV viral load was performed as described previously (44). Encapsulation of fludarabine in SM RBC. SM RBC were loaded with fludarabine by a procedure involving hypotonic dialysis, isotonic resealing, and reannealing, as previously described (28). Some modifications were performed to adapt the procedure for SM RBC. In particular, compared to the loading procedure followed for human RBC, the osmolarity of dialysis buffer was increased and the dialysis time was prolonged. Targeting of fludarabine-loaded RBC to macrophages was achieved by inducing band 3 clustering as described previously (29). To study fludarabine metabolite formation in SM RBC, fludarabine-loaded RBC were incubated at 0.5% hematocrit (Ht) in RPMI 1640 medium containing fetal calf serum. At different times (0, 2 h, 16 h, 2 days, and 5 days) during incubation at 37 C in a 5% CO 2 atmosphere and under sterile conditions, 4-ml aliquots were processed to determine the concentrations of fludarabine and its metabolites. Briefly, fludarabine-loaded RBC were extracted with perchloric acid and analyzed by high-performance liquid chromatography as described previously (14), while media were submitted to solid-phase extraction using Isolute C 18 columns (International Sorben Technology, Mid-Glamorgan, United Kingdom) according to the manufacturer s instructions before being analyzed by high-performance liquid chromatography. Flow cytometry for surface and intracellular markers. Cells derived from peripheral blood (PB) and LN were isolated by gradient centrifugation. Fourcolor flow cytometric analysis was performed according to standard procedures, using a panel of monoclonal antibodies (MAbs) that were originally designed to detect human molecules but that we and others have shown to be cross-reactive with SMs (43, 44). The MAbs used included CD8-phycoerythrin (clone SK1), CD8-allophycocyanin (clone SK1), CD3-phycoerythrin (clone SP34-2), CD4- peridinin chlorophyll protein (clone L200), and Ki67-fluorescein isothiocyanate (FITC) (clone B56) (BD PharMingen, San Diego, CA) and CD14-FITC (clones RMO52 and MY4) (Beckman Coulter, Miami, FL). Samples used for Ki67 were surface stained, fixed and permeabilized using the PharMingen CytoFix/Perm kit, and stained intracellularly with the proper MAb and control. Flow cytometric acquisition and analysis of samples were performed on at least 100,000 events on a FACScalibur flow cytometer driven by the CellQuest software package (Becton Dickinson). Analysis of the acquired data was performed using FlowJo software (Tree Star, Inc., Ashland, OR). Western blot analysis. Expression levels of STAT1, STAT3, and STAT5 and of their corresponding phosphorylated forms in peripheral blood- and lymph node-derived cells were measured by Western blotting. Briefly, cells were lysed for 20 min on ice with 20 mm HEPES (ph 7.9), 25% glycerol, 0.42 M NaCl, 0.2 mm EDTA, 1.5 mm MgCl 2 containing 0.5% NP-40, 10 g/ml leupeptin, 10 g/ml pepstatin, 1 mm sodium fluoride, and 1 mm sodium orthovanadate. From the total protein extracted, 30 g was fractionated by 7% sodium dodecyl sulfatepolyacrylamide gel electrophoresis and then was electrically transferred to a nitrocellulose membrane. Blots were incubated with anti-stat1, anti-stat3, and anti-stat5 (1:1,000; Santa Cruz Biotechnology Inc., Santa Cruz, CA) and with anti-pstat1, anti-pstat3, and anti-pstat5 (1:1,000; Cell Signaling Technology, Danvers, MA) MAbs overnight at 4 C, followed by incubation with horseradish peroxidase-conjugated secondary antibody. Blots were treated with

3 VOL. 80, 2006 SELECTIVE MACROPHAGE DEPLETION IN SIV-INFECTED SMs TABLE 1. Immunologic and virologic characteristics of the animals included in the study Animal code Sex Age (yr) Viral load CD4 cell no. Treatment Amt ( g) of fludarabine a FBu Male 10 Negative 925 None FFs Male 7 Negative 710 None FYf Male 17 Negative 859 None FVs Female 8 174, None FZp Male 9 398,000 1,067 None FJs Female 8 1,420,000 1,572 None FGs Female 9 Negative 1,454 Fludarabine FLj Male 15 20, Fludarabine FQg Female ,000 1,276 Fludarabine FDy Male 6 31, PMPA FQu Male 8 38,800 1,003 PMPA FMr Male 10 36,450 1,077 Fludarabine PMPA FLv Male 7 49, Fludarabine PMPA a Quantity of fludarabine that each treated animal received in each infusion, expressed as mean standard deviation of the four administrations. enhanced chemiluminescence reagents, and proteins were detected and quantitated with the ChemiDoc system (Bio-Rad, Hercules, CA). Equal protein loading was confirmed by the level of actin protein present in the membrane tested with antiactin antibody (1:500; Sigma). Histology, immunohistochemistry (IHC), and indirect immunofluorescence analysis. Sections were cut from paraffin-embedded paraformaldehyde-fixed lymph node tissues, deparaffinized, and stained with hematoxylin and eosin. M/M and pstat1-positive cells were detected with NCL-Macro (1:100; Novacastra Laboratories, Newcastle, United Kingdom) and pstat1 monoclonal antibodies (1:100; Zymed Laboratories, San Francisco, CA), respectively. For indirect immunofluorescence staining, sections prepared as described above were deparaffinized, hydrated, treated for antigen unmasking, and incubated with the first monoclonal primary antibody, anti-pstat1, according to the manufacturer s instructions. Subsequently, the samples were incubated overnight 4 C with the second polyclonal primary antibody, anti p24 (1:100; Diatheva, Fano, Italy). Primary antibodies were detected with FITC- and tetramethyl rhodamine isocyanate-conjugated goat anti-mouse or -rabbit antibodies (1:200; Sigma, St. Louis, MO). DNA was stained with 4,6 -diamino-2-phenylindole (DAPI) (0.1 g/ml). Images were collected on a Leica DML equipped with a charge-coupled device camera. Statistical analysis. The Mann-Whitney U test was used for comparisons between groups, while correlations involving different sets of data within the same group were analyzed using either the standard Pearson correlation coefficient or Spearman s rank correlation test. Significance was assessed at P values of 0.01 and 0.05 levels. All analyses were performed using SAS software. FIG. 1. Stat1 activation in SIV-infected sooty mangabeys. (a) Expression of total and phosphorylated Stat1 in peripheral blood- and lymph node-derived cells from three SIV-infected (SIV ) and three uninfected (SIV ) SMs was measured by Western blot analysis. Histograms show the means standard deviations of band intensity, expressed as number of pixels. (b) The expression of Stat1 in lymph node tissues isolated from three SIV-infected and three uninfected SMs was determined by IHC. The results are reported as number of pstat1-positive cells per mm 2 of tissue. In panels a and b statistical analysis were performed between SIV-infected (black bars) and uninfected (gray bars) animals, and significant values are marked by asterisks.

4 10338 CERVASI ET AL. J. VIROL. FIG. 2. Treatment protocol and longitudinal analysis of viral load in SIV-infected sooty mangabey. (a) SMs were assigned to three groups treated with different protocols. Animals were studied for 64 days. At various time points through the study, plasma, peripheral blood, and lymph node were collected. Flu, fludarabine. (b) Temporal trend of viral load at various time points during the study in SIV-infected SMs treated with fludarabine (gray dotted line), PMPA (dark gray dashed line), and fludarabine plus PMPA (black). Viral load is presented as mean standard deviation of SIV RNA copies/milliliter of plasma of the two animals included in each group. Numbers in square bracket at day 36 and day 43 represent the fold increase in viral load compared to day 28 (end of treatment) found in group 2 and group 3. RESULTS Chronic SIV infection in SMs is associated with increased STAT1 activation in peripheral blood- and lymph node-derived cells. Our previous in vitro experiments indicated that HIV infection of M/M results in a marked increase of pstat1 expression in these cells, a finding that suggests that these cells may be specifically targeted by fludarabine in vivo. To determine whether increased STAT-1 expression is also present during in vivo SIV infection, we analyzed by Western blotting the intracellular concentrations of STAT1 and pstat1 in PB- and LN-derived cells isolated from three SIV-infected and three uninfected SMs. All infected animals, whose characteristics are listed in Table 1, were studied during the chronic stage of infection. SIV-infected SMs express significant higher level of pstat1 than uninfected ani-

5 VOL. 80, 2006 SELECTIVE MACROPHAGE DEPLETION IN SIV-INFECTED SMs TABLE 2. Hematological parameters in six SIV-infected SMs and one uninfected SM (FGs) treated with fludarabine-loaded RBC (FGs, FLj, and FQg), PMPA (FDy and FQu), or fludarabine-loaded RBC plus PMPA (FMr and FLv) a White blood cells (10 3 / l) No. of lymphocytes RBC (10 6 / l) Hgb (g/dl) Ht (%) MCV (fl) MCH (pg/cell) Animal Day 0 Day 30 Day 0 Day 30 Day 0 Day 30 Day 0 Day 30 Day 0 Day 30 Day 0 Day 30 Day 0 Day 30 FGs 8,800 8, ,600 6,468 FLj 3,400 3, ,142 2,052 FQg 6,900 6, ,200 4,941 FDy 6,400 5, ,213 3,120 FQu 5,300 4, ,127 3,528 FMr 5,600 4, ,858 2,975 FLv 5,700 5, ,359 3,599 a Values before (day 0) and 30 days after the initiation of treatment are reported. Hgb, hemoglobin; MCV, mean cell volume; MCH, mean cell hemoglobin. mals, in both PB (P 0.01)- and LN (P 0.05)-derived cells (Fig. 1a). In the PB of SIV-infected SMs we also found a significant increase (P 0.05) in the level of STAT1. As shown in Fig. 1b, the number of pstat1-positive cells per mm 2 of LN tissue (measured by IHC) was increased in SIV-infected SMs compared to uninfected animals. Importantly, the majority of these cells show morphological features that are typical of tissue macrophages. It is of note that no significant differences in the levels of either total or phosphorylated STAT3 and STAT5 were found between SIVinfected and uninfected SMs in either PB- or LN-derived cells (data not shown). In all, these results indicate that, similarly to what was observed in in vitro HIV-infected M/M, PB- and LNderived mononuclear cells of chronically SIV-infected SMs manifested increased pstat1 expression. Study design. Six naturally SIV-infected SMs were assigned to three groups and treated with the following protocols (Fig. 2a): two with four weekly administrations of fludarabineloaded autologous RBC (group 1), two with PMPA at a dose of 30 mg/kg per day for 28 consecutive days (group 2), and two with 30 mg/kg PMPA per day for 28 consecutive days and four weekly administrations of fludarabine-loaded autologous RBC (group 3). One uninfected SM was treated with fludarabineloaded autologous RBC as a control. Animals were studied for 64 days, during which time we collected plasma, PB, and LN at different time points (Fig. 2a). Fludarabine was encapsulated in SM RBC at a final concentration of mol/ml RBC. The study of fludarabine metabolites shows that in SM RBC, fludarabine is phosphorylated to the same active forms (fludarabine diphosphate and fludarabine triphosphate) that were found in human RBC (14). At the time of loading, both the fludarabine diphosphate and fludarabine triphosphate forms are present, and at 18 h, fludarabine triphosphate represents 23% of all metabolites (data not shown). Three SIVinfected and three uninfected SMs were left untreated and used for ex vivo analysis (Table 1). Administration of fludarabine-loaded RBC in SMs does not induce any significant toxic side effect. We carefully examined the possibility of toxic side effects associated with the abovedescribed protocol of administration of fludarabine-loaded autologous RBC alone or in association with PMPA. In particular, we evaluated several basic hematological parameters (white blood cells, RBC, hemoglobin, Ht, mean cell volume, mean cell hemoglobin, and lymphocyte number) at different time points (days 7, 14, 21, 30, 36, 43, and 64) in all treated SMs, and we found no significant difference from the pretreatment levels. Table 2 summarizes the hematological values before (day 0) and 30 days after the initiation of treatment, i.e., after four fludarabine administrations and 28 days of PMPA treatment. Similarly, no changes were observed in terms of either body weight variation or animal behavior, and all treated SMs are alive and in good health 2 years after treatment suspension. These results indicate that in vivo administration of fludarabine-loaded autologous RBC alone or in combination with PMPA is a safe intervention in this nonhuman primate model of SIV infection. Changes in viral replication induced by administration of fludarabine-loaded autologous RBC in SIV-infected SMs. We next sought to evaluate the effects of the three therapeutic approaches used (i.e., PMPA only, fludarabine-loaded RBC only, or the combination of the two) on the level of virus replication. To this end, we longitudinally measured plasma viral load in the six SIV-infected SMs at different time points before and after treatment. As shown in Fig. 2b, we found that daily PMPA administration was followed, in all four treated SMs, by a significant decline in viral load to levels below the limit of detection of our assay between day 14 and day 28 of treatment. As expected, the two SMs treated with fludarabineloaded RBC alone maintained viral replication at levels similar to the baseline, thus confirming that the administration of fludarabine within RBC does not act on actively replicating CD4 T cells, which are known to be a major source of virus production in SIV-infected SMs (G. Silvestri, unpublished observations). Importantly, we observed that after interruption of PMPA treatment, the two SIV-infected SMs treated only with PMPA showed a faster rebound of viral replication than the animals that received PMPA-plus-fludarabine-loaded RBC. In particular, the plasma viral load was significantly higher in the SMs treated with PMPA only at days 36 (P 0.05) and 43 (P 0.06) of follow-up. This faster rebound can be better perceived by calculating the ratio of viral loads between days 36 and 43 and the end of treatment (day 28). At days 36 and 43 animals treated with only PMPA increased their viral load, compared to that at day 28, by and 1,482-fold, respectively. In contrast, in SMs treated with PMPA-plus-fludarabine-loaded RBC, these variations were and 215-fold. It is of note that one animal (FMr) treated with PMPA-plus-fludarabine-loaded RBC maintained a lower level of viremia until day 64, i.e., the end of the study (data not shown). This delayed rebound in viral replication found after PMPA suspension in fludarabinetreated SMs suggests that this combined therapeutic approach

6 10340 CERVASI ET AL. J. VIROL.

7 VOL. 80, 2006 SELECTIVE MACROPHAGE DEPLETION IN SIV-INFECTED SMs FIG. 4. Administration of fludarabine-loaded RBC induces a significant reduction in the absolute number of M/M. Variations (expressed as percentages of initial values) in the absolute number of CD14 cells in the peripheral blood collected from SMs treated with fludarabine (animals FLj and FQg), PMPA (FDy and FQu), and fludarabine plus PMPA (FLv and FMr) were determined at various time points through the study. may have effectively reduced the number of viral reservoirs in naturally SIV-infected SMs. Treatment with fludarabine-loaded RBC does not induce any change in the levels of T cells and their main subsets. The delayed viral replication rebound found in SIV-infected SMs treated with fludarabine-loaded RBC and PMPA after PMPA suspension is compatible with two possible mechanisms: a reduction in the levels of total and actively proliferating CD4 T cells and/or a depletion in the number of long-lasting SIV reservoirs (i.e., M/M). To discriminate between these mechanisms, we performed a detailed four-color flow cytometric analysis of PB- and LN-derived cells isolated from the six SIV-infected SMs treated with the three different protocols. First, we measured the levels of CD4 and CD8 T cells in PB and LN and found that fludarabine treatment did not significantly change either the absolute number or the percentage of these T-cell subsets compared to those in PMPA-treated animals (Fig. 3). We next measured the levels of proliferating, i.e., Ki-67-positive, T cells and found that treatments with PMPA alone or PMPA-plus-fludarabine-loaded RBC had comparable effects on the levels of proliferating CD4 T cells in PB and LN (Fig. 3). In particular, the percentage of circulating CD4 Ki67 T cells decreased slightly during the 28 days of PMPA administration, as expected given the suppression of viral replication. After suspension of therapy, SIV-infected SMs treated with PMPA alone experienced an increase in the levels of proliferating CD4 T cells that temporally correlated with the rebound in viral replication. Interestingly, SMs treated with fludarabine-loaded RBC and PMPA also exhibited an increase in the levels of CD4 Ki67 T cells after PMPA suspension (Fig. 3), despite the delay in viral rebound (Fig. 2b). These data suggest that the different kinetics of viral replication rebound in SMs treated with PMPA alone and those treated with PMPA-plus-fludarabine-loaded RBC are unlikely to be due to direct effects of fludarabine on the actively proliferating CD4 T-cell compartment. In addition, no significant differences were found in the proliferation levels of PB- or LN-derived CD8 T lymphocytes between SIV-infected SMs that were treated with PMPA alone or with PMPA-plus-fludarabineloaded RBC (Fig. 3). Surprisingly, SMs treated with fludarabine experienced a significant increase only in the levels of proliferating CD4 and CD8 T cells, which was maintained until the end of the study (day 64). The increment in the pool of Ki-67 T cells in this group may be related to immune system activation induces by the administration of RBC in the context of active viral replication (no PMPA treatment). In the other two groups, this phenomenon may be counterbalanced by the decreased levels of T-cell proliferation that follow the PMPA-dependent control of viral replication. Fludarabine-loaded RBC induce a significant reduction in both the percentage and absolute number of M/M in SIV-infected SMs. To test the hypothesis that the delayed rebound in viral replication observed in SIV-infected SMs treated with PMPA-plus-fludarabine-loaded RBC is related FIG. 3. Longitudinal analysis of T-cell compartments in the peripheral blood and lymph nodes of treated animals. Total CD4 and CD8 T cells and their levels of proliferation in the peripheral blood (left) and lymph nodes (right) of SIV-infected SMs treated with fludarabine (gray dotted lines), PMPA (dark gray dashed lines), and fludarabine plus PMPA (black lines) were evaluated at different time points through the study. Values are depicted as mean standard deviations.

8 10342 CERVASI ET AL. J. VIROL. FIG. 5. Administration of fludarabine-loaded RBC in SMs significantly reduces SIV-infected M/M. (A) Fractions of cells expressing pstat1 at day 0, day 14, and day 31 after initiation of therapy. The upper panels show pstat1 staining of a lymph node biopsy obtained from a representative SM. The results for the three groups of SIV-infected SMs are expressed in the lower panel as percent variation from the initial value (set at 100%). Statistical analyses were performed between animals treated with fludarabine-loaded RBC and treated with only PMPA, and significant values are marked by asterisks. (B to F) Indirect fluorescence staining of p24 and pstat1 in LN tissue from a representative SIV-infected SM. (B) Visible light; (C) DAPI; (D) anti-p24 (FITC), (E) anti-pstat1 (tetramethyl rhodamine isocyanate); (F) merged illustration of collected images. to a decreased M/M reservoir, we measured the percentage and absolute number of M/M in PB of treated SMs by multiparametric flow cytometric analysis. In particular, we calculated the absolute number and percentage of cells that, based on forward scatter/side scatter, were gated as monocytes and also expressed CD14. We found that SIV-infected SMs treated with fludarabine-loaded RBC manifested a significant reduction in both the percentage and number of monocytes independent of PMPA administration (Fig. 4). In contrast, SMs treated with PMPA alone did not experience any significant decline in the levels of monocytes (Fig. 4). Importantly, SMs treated with fludarabine-loaded RBC plus PMPA showed significantly lower numbers of monocytes during days 28 to 42 after initiation of therapy, i.e., when the different rebound in viral replication was found (Fig. 2b), than did animals treated with PMPA alone. It is of note that treatment with fludarabine-loaded RBC did not modify levels of monocytes in the uninfected SM (data not shown), suggesting that only M/M from SIV-infected animals became a target for fludarabine. Administration of fludarabine-loaded RBC in SMs depletes SIV-infected M/M via a pstat1-dependent pathway. To further test the hypothesis that fludarabine-loaded RBC specifically deplete M/M that overexpress pstat1 as a result of their infection with SIV, we measured the fraction of macrophages expressing pstat1 in LN tissue by IHC. We analyzed LN samples collected from the six SIV-infected SMs at three different time points, i.e., before (day 0) and at day 14 and day 31 after initiation of therapy. pstat1-positive cells were identified by staining with a specific monoclonal antibody, whereas

9 VOL. 80, 2006 SELECTIVE MACROPHAGE DEPLETION IN SIV-INFECTED SMs FIG. 5 Continued. M/M were identified by morphological features. When treated with fludarabine-loaded RBC, and independently of PMPA administration, SIV-infected SMs showed a significant reduction (P 0.01) in the amount of M/M expressing pstat1 (Fig. 5A). Interestingly, no effects were found in SIV-infected animals treated with PMPA alone (Fig. 5A) or in the uninfected SM treated with fludarabine-loaded RBC (data not shown). Finally, to confirm that macrophages that overexpressed pstat1 and that are selectively depleted by fludarabine treatment are SIV infected, we performed, by fluorescence microscopy, p24 and pstat1 double staining in LN tissue isolated from infected animals. The results confirm that the vast majority of pstat1- positive cells are SIV-infected M/M, as indicated by the morphology of these cells and the expression of p24 viral protein (Fig. 5B to F). In all, these findings are compatible with the hypothesis that treatment with fludarabine-loaded autologous RBC is able to selectively deplete SIV-infected macrophages via a pstat1-dependent pathway in SMs. DISCUSSION The introduction of HAART has resulted in a major reduction of virus load and in a significant decline in mortality and morbidity among HIV-infected individuals (17, 18, 32, 49). Unfortunately, HAART is unable to eradicate HIV infection due to its limited effects on viral reservoirs carrying replicationcompetent HIV (5, 7, 10, 35). The existence of HIV reservoirs represents a major obstacle to eradication of HIV with current antiretroviral regimens and underlines the importance of exploring novel therapeutic approaches that specifically target

10 10344 CERVASI ET AL. J. VIROL. these reservoirs. In this perspective, an ideal anti HIV therapy should include drugs that protect new cells from infection (i.e., HAART) as well as agents able to reduce the reservoir of infected cells. The first reservoir that has been identified consists of resting CD4 memory T lymphocytes, and most efforts were aimed at designing treatment strategies able to reduce them (11, 26). Other reservoirs, however, and in particular M/M, are also involved in maintaining HIV infection in chronically treated patients and must be considered when protocols aimed at eradicating HIV infection are designed. Several lines of evidence indicate the importance of M/M in virus persistence: (i) macrophage and microglia in the central nervous systems of patients with AIDS dementia complex were identified as one of the first nonlymphocyte cell lineages that support viral replication (23, 48); (ii) in rhesus monkeys infected with a highly pathogenic virus such as SHIV DH12R, tissue M/M sustain high virus loads for several months after the depletion of CD4 T cells (22); (iii) in HIV-infected persons the number of infected M/M, compared to infected CD4 T lymphocytes, increases substantially over time (33); and (iv) M/M become exposed to virus very early in the infection, are relatively resistant to cytopathic effects of HIV, and are involved in antigen presentation to CD4 T cells (24, 50). The results of our previous in vitro study indicate that survival of human primary M/M during HIV infection is associated with activation of STAT1 pathway and that the HIVinfected M/M can be depleted with fludarabine, a nucleoside analog able to act on low-growth-fraction cells overexpressing pstat1 (12, 28). In the present study, we describe the results of an experiment aimed at testing the feasibility, safety, and efficacy of an antiviral therapeutic protocol based on the administration of PMPA in association with fludarabine-loaded RBC. We hypothesized that this approach may be of interest since it conceivably targets viral replication in both recently infected proliferating CD4 T cells (by PMPA) and persistently HIV-1-infected M/M (by fludarabine). To the best of our knowledge, this is the first time that this type of study has been conducted in vivo in a primate model of HIV infection. Although the current study was designed as an exploratory pilot trial involving a limited number of animals, several interesting indications emerged from the analysis of the results obtained. First, we found that, similarly to what was observed in pathological HIV infection in humans, natural, nonpathogenic SIV infection of SMs resulted in a significant activation of STAT1 signaling in M/M, thus providing a rationale for the use of fludarabine as a reservoir-reducing agent in this nonhuman primate model. Second, we found that administration of fludarabine-loaded autologous RBC is a feasible therapeutic approach that is not associated with any relevant toxicity in nonhuman primates. Third, we determined that SIV-infected SMs treated with four fludarabine-loaded RBC administrations plus PMPA showed a delayed rebound in SIV replication after suspension of treatment compared to animals treated with PMPA alone. It is worth noting that in both animals treated with fludarabine-loaded RBC plus PMPA, viremia remained at lower levels until day 43, i.e., 22 days after the last fludarabine administration, with one animal maintaining the lower level of SIV replication until the end of the study (day 64). Fourth, we found that the two SIV-infected SMs treated with fludarabine-loaded RBC only (i.e., without PMPA) did not show any change in viral replication, confirming that this intervention selectively targets M/M and is unlikely to have any effect on nonphagocyosting cells, such as the actively replicating CD4 T cells that are the major source of SIV production. Consistently, the absence of any measurable effects of fludarabine-loaded RBC plus PMPA on the level of activated/proliferating CD4 T cells suggests that the delayed rebound in viral replication is not due to direct effects of the drug on CD4 T cells. Importantly, the potential efficacy of the presently described therapeutic protocol was confirmed by the observed changes in the M/M compartment. In SMs, treatment with fludarabine-loaded RBC induced a significant reduction in the absolute number of SIV-infected M/M, which remained significantly lower than that in animals treated only with PMPA at the time when the different kinetics of viral rebound were observed. It is of note that treatment with fludarabine-loaded RBC selectively depleted macrophages in chronically SIV-infected SMs but not in the uninfected control. This specificity may be due to the fact that M/M overexpressed pstat1 following SIV infection, as demonstrated by IHC and indirect fluorescence analysis of LN tissues. In conclusion, this study suggests that anti-hiv therapeutic strategies based on the combination of antiretroviral (PMPA) and cytotoxic (fludarabine) drugs targeted to M/M may be well tolerated and effective in both suppressing the level of ongoing viral replication and reducing the reservoirs of infected cells. ACKNOWLEDGMENTS This work was supported by grants 30F.31 from the V Programma Nazionale di Ricerca sull AIDS, Istituto Superiore di Sanita, Rome, Italy, and FIRB RBNE01TBTR-001 to M.M. and by NIH grants A52775 and AI to G.S. We thank Stephanie Ehnert, Elizabeth Strobert, and Chris Souder for their assistance with animals and Andrew Fedanov for his assistance with IHC staining. REFERENCES 1. Belmonte, L., P. Bare, M. M. de Bracco, and B. H. Ruibal-Ares Reservoirs of HIV replication after successful combined antiretroviral treatment. Curr. Med. Chem. 10: Calo, V., M. Migliavacca, V. Bazan, M. Macaluso, M. Buscemi, N. Gebbia, and A. Russo Stat proteins: from normal of cellular events to tumorogenesis. J. Cell. Physiol. 197: Chen, Z., A. Luckay, D. L. Sodora, P. Telfer, P. Reed, A. Gettie, J. M. Kanu, R. F. Sadek, J. Yee, D. D. Ho, L. Zhang, and P. A. Marx Human immunodeficiency virus type 2 (HIV-2) seroprevalence and characterization of a distinct HIV-2 genetic subtype from the natural range of simian immunodeficiency virus-infected sooty mangabeys. J. Virol. 71: Chkrabarti, L. A., R. S. Lewin, L. Zhang, A. Gettie, A. Luckay, L. N. Martin, E. Skulsky, D. D. Ho, C. Cheng-Mayer, and P. A. Marx Normal T-cell turnover in sooty mangabeys harboring active simian immunodeficiency virus infection. J. Virol. 74: Chun, T. W., R. T. Davey, Jr., D. Engel, H. C. Lane, and A. S. Fauci Re-emergence of HIV after stopping therapy. Nature 401: Chun, T. W., R. T. Davey, Jr., M. Ostrowski, J. Shawn Justement, D. Engel, J. I. Mullins, and A. S. Fauci Relationship between pre-existing viral reservoirs and the re-emergence of plasma viremia after discontinuation of highly active anti-retroviral therapy. Nat. Med. 6: Chun, T. W., and A. S. Fauci Latent reservoirs of HIV: obstacles to eradication of virus. Proc. Natl. Acad. Sci. USA 96: Collman, R. G., C. F. Perno, S. M. Crowe, M. Stevenson, and L. J. Montaner HIV and cells of macrophage/dendritic lineage and other non-t cell reservoirs: new answers yield new questions. J. Leukoc. Biol. 74: Douek, D. C., L. J. Picker, and R. A. Koup T cell dynamics in HIV-1 infection. Annu. Rev. Immunol. 21: Finzi, D., J. Blankson, J. D. Siliciano, J. B. Margolick, K. Chadwick, T. Pierson, K. Smith, J. Lisziewicz, F. Lori, C. Flexner, T. C. Quinn, R. E. Chaisson, E. Rosenberg, B. Walker, S. Gange, J. Gallant, and R. F. Siliciano Latent infection of CD4 T cells provides a mechanism for life-long persistence of HIV-1, even in patients on effective combination therapy. Nat. Med. 5:

11 VOL. 80, 2006 SELECTIVE MACROPHAGE DEPLETION IN SIV-INFECTED SMs Finzi, D., M. Hermankova, T. Pierson, L. M. Carruth, C. Buck, R. E. Chaisson, T. C. Quinn, K. Chadwick, J. Margolick, R. Brookmeyer, J. Gallant, M. Markowitz, D. D. Ho, D. D. Richman, and R. F. Siliciano Identification of a reservoir for HIV-1 in patients on highly active antiretroviral therapy. Science 278: Frank, D. A., S. Mahajan, and J. Ritz Fludarabine-induced immunosuppression is associated with inhibition of Stat1 signaling. Nat. Med. 5: Fraternale, A., A. Casabinca, C. Orlandi, A. Cerasi, L. Chiarantini, G. Brandi, and M. Magnani Macrophage protection by addition of glutathione (GSH)-loaded erythrocytes to AZT and DDI in a murine AIDS model. Antiviral Res. 56: Fraternale, A., L. Rossi, and M. Magnani Encapsulation, metabolism and release of 2-fluoro-ara-AMP from human erythrocytes. Biochim. Biophys. Acta 1291: Fraternale, A., A. Tonelli, A. Casabianca, A. Casabianca, G. Vallanti, L. Chiarantini, G. F. Schivano, U. Benatti, A. De Flora, and M. Magnani Role of macrophage protection in the development of murine AIDS. J. Acquir. Immune Defic. Syndr. 21: Fultz, P. N., T. P. Gordon, D. C. Anderson, and H. M. McClure Prevalence of natural infection with simian immunodeficiency virus and simian T-cell leukemia virus type 1 in a breeding colony of sooty mangabey monkeys. AIDS 188: Furtado, M. R., D. S. Callaway, J. P. Phair, K. J. Kunsstman, J. L. Stanton, C. A., Macken, A. S. Perelson, and S. M. Wolinsky Persistence of HIV-1 transcription in peripheral-blood mononuclear cells in patients receiving antiretroviral therapy. N. Engl. J. Med. 340: Gulick, R. M., J. W. Mellors, D. Havlir, J. J. Eron, C. Gonzalez, D. McMahon, D. D. Richman, F. T. Valentine, L. Jonas, A. Meibohm, E. A. Emini, and J. A. Chodakewitz Treatment with indinavir, zidovudine, and lamivudine in adults with human immunodeficiency virus infection and prior antiretroviral therapy. N. Engl. J. Med. 337: Hahn, B. H., G. M. Shaw, K. M. De Cock, and P. M. Sharp AIDS as a zoonosis: scientific and public health implications. Science 287: Huang, P., D. Farquhar, and W. Plunkett Selective action of 3 -azido- 3 -deoxythymidine 5 -triphosphate on viral reserve transcriptase and human DNA polymerases. J. Biol. Chem. 265: Huang. P., and W. Plunkett Action of 9-beta-D-arabinofuranosyl-2- fluoroadenine on RNA metabolism. Mol. Pharmacol. 39: Igarashi, T., C. R. Brown, Y. Endo, A. Buckler-White, R. Plishka, N. Bioschofberger, V. Hirsch, and M. A. Martin Macrophage are the principal reservoir and sustain high virus loads in rhesus macaques after the depletion of CD4 T cells by highly pathogenic simian immunodeficiency virus/hiv type 1 chimera (SHIV): implications for HIV-1 infections of humans. Proc. Natl. Acad. Sci. USA 98: Koenig, S. H., H. E. Gendelman, J. M. Orenstein, M. C. Dal Canto, G. H. Pezeshkpour, M. Yungbluth, F. Janotta, A. Aksamit, M. A. Martin, and A. S. Fauci Detection of AIDS virus in macrophages in brain tissue from AIDS patients with encephalopathy. Science 233: Lambrotte, O., Y. Taoufik, M. G. de Goer, C. Wallon, C. Goujard, and J. F. Delfraissy Detection of infectious HIV in circulating monocytes from patients on prolonged highly active antiretroviral therapy. J. Acquir. Immune Defic. Syndr. 23: Lee, C., Q. H. Liu, B. Tomkowiez, Y. Yi, B. D. Freedman, and R. G. Collman Macrophage activation through CCR5- and CXCR4-mediated gp120- elicited signaling pathways. J. Leukoc. Biol. 74: Lehrman, G., I. B. Hogue, S. Palmer, C. Jennings, C. A. Spina, A. Wiegand, A. L. Landay, R. W. Coombs, D. D. Richman, J. W. Mellors, J. M. Coffin, R. J. Bosch, and D. M. Margolis Depletion of latent HIV-1 infection in vivo: a proof-of-concept study. Lancet 366: Levy, D. E., and J. E. Darnell, Jr Stats: transcriptional control and biological impact. Nat. Rev. Mol. Cell. Biol. 3: Magnani, M., M. Balestra, A. Fraternale, S. Aquaro, M. Paiardini, B. Cervasi, A. Casabianca, E. Garaci, and C. F. Perno Drug-loaded red blood cell-mediated clearance of HIV-1 macrophage reservoir by selective inhibition of STAT1 expression. J. Leukoc. Biol. 74: Magnani, M., L. Rossi, G. Brandi, G. F. Schiavano, M. Montroni, and G. Piedimonte Targeting antiretroviral nucleoside analogues in phosphorylated form to macrophage: in vitro and in vivo studies. Proc. Nalt. Acad. Sci. USA 89: McCune, J. M The dynamics of CD4 T-cell depletion in HIV disease. Nature 410: Natarajan, V., M Bosche, J. A. Metcalf, D. J. Ward, H. C. Lane, and J. A. Kovacs HIV-1 replication in patients with undetectable plasma virus receiving HAART. Highly active antiretroviral therapy. Lancet 353: Palella, F. J., Jr., K. M. Delaney, A. C. Moorman, M. O. Loveless, J. Fuhrer, G. A. Satten, D. J. Aschman, S. D. Holmberg, et al Declining, morbidity and mortality among patient with advanced human immunodeficiency virus infection. N. Engl. J. Med. 338: Perelson, A. S., P. Essunger, Y. Cao, M. Vesanen, A. Hurley, K. Saksela, M. Markowitz, and D. D. Ho Decay characteristics of HIV-1-infected compartments during combination therapy. Nature 387: Persidsky, Y., and H. E. Gendelman Mononuclear phagocyte immunity and the neuropathogenesis of HIV-1 infection. J. Leukoc. Biol. 74: Pierson, T., J. McArthur, and R. F. Siliciano Reservoirs for HIV-1: mechanisms for viral persistence in the presence of antiviral immune responses and antiretroviral therapy. Annu. Rev. Immunol. 18: Rey-Cuille, M. A., J. L. Berthir, M. C. Bomsel-Demontoy, Y. Chaduc, L. Montagnier, A. G. Hovanessian, and L. A. Chakrabarti Simian immunodeficiency virus replicates to high levels in sooty mangabeys without inducing disease. J. Virol. 72: Santiago, M. L., F. Range, B. F. Keele, Y. Li, E. Bailes, F. Bibollet-Ruche, C. Fruteau, R. Noe, M. Peeters, J. F. Brookfield, G. M. Shaw, P. M. Sharp, and B. H. Hahn Simian immunodeficiency virus infection in free-ranging sooty mangabeys (Cercocebus atys) from the Tai Forest, Cote d Ivoire: implications for the origin of epidemic human immunodeficiency virus type 2. J. Virol. 79: Sharp, P. M., E. Bailes, R. R. Chaudhuri, C. M. Rodenburg, M. O. Santiago, and B. H. Hahn The origins of acquired immune deficiency syndrome viruses: where and when? Philos. Trans. R. Soc. London B 356: Sharp, P. M., G. M. Shaw, and B. H. Hahn Simian immunodeficiency virus infection of chimpanzees. J. Virol. 79: Shehu-Xhilaga, M., G. Tachedjian, S. M. Crowe, and K. Kedzierska Antiretroviral compounds: mechanisms underlying failure of HAART to eradicate HIV-1. Curr. Med. Chem. 12: Shuai, K., and B. Liu Regulation of JAK-STAT signaling in the immune e system. Nat. Rev. Immunol. 3: Siliciano, J. D., and R. F. Siliciano A long-term latent reservoir for HIV-1: discovery and clinical implication. J. Antimicrob. Chemother. 54: Silvestri, G., A. Fedanov, S. Germon, N. Kozyr, W. J. Kaiser, D. A. Garber, H. McClure, M. B. Feinberg, and S. I. Staprans Divergent host responses during primary simian immunodeficiency virus SIVsm infecton of natural sooty mangabey and nonnatural rhesus macaque hosts. J. Virol. 79: Silvestri, G., D. L. Sodora, R. A. Koup, M. Paiardini, S. P. O Neil, H. M. McClure, S. I. Staprans, and M. B. Feinberg Nonpathogenic SIV infection of sooty mangabeys is characterized by limited bystander immunopathology despite chronic high-level viremia. Immunity 18: Silvestri, G Naturally SIV-infected sooty mangabeys: are we closer to understanding why they do not develop AIDS? J. Med. Primatol. 34: Sousa, A. E., J. Carneiro, M. Meier-Schellersheim, Z. Grossman, and R. M. Victorino CD4 T cell depletion is linked directly to immune activation in the pathogenesis of HIV-1 and HIV-2 but only indirectly to the viral load. J. Immunol. 169: Steveson, M HIV-1 pathogenesis. Nat. Med. 9: Wiley, C. A., R. D. Schrier, J. A. Nelson, P. W. Lampert, and, M. B. Oldstone Cellular localization of human immunodeficiency virus infection within the brains of acquired immunodeficiency syndrome patients. Proc. Natl. Acad. Sci. USA 83: Zhang, L., B. Ramratnam, K. Tenner-Racz, Y. He, M. Vesanen, S. Lewin, A. Talal, P. Racx, A. S. Perelson, B. T. Korber, M. Markowitz, and D. D. Ho Quantifying residual HIV-1 replication in patients receiving combination antiretroviral therapy. N. Engl. J. Med. 340: Zhang, Z., T. Schuler, M. Zupancic, S. Wietgrefe, K. A. Staskus, K. A. Reimann, T. A. Reinhart, M. Rogan, W. Cavert, C. J. Miller, R. S. Veazey, D. Notermans, S. Little, S. A. Danner, D. D. Richman, D. Havlir, J. Wong, H. L. Jordan, T. W. Schacker, P. Racz, K. Tenner-Racz, N. L. Letvin, S. Wolinsky, and A. T. Haase Sexual transmission and propagation of SIV and HIV in resting and activated CD4 T cells. Science 286: