Maraviroc reduces the regulatory T-cell frequency in antiretroviral-naive HIV-infected subjects

The Journal of Infectious Diseases

Maraviroc reduces Regulatory T cells in ART-naïve HIV-infected subjects--Manuscript Draft--

Manuscript Number: 54235R2

Full Title: Maraviroc reduces Regulatory T cells in ART-naïve HIV-infected subjects

Short Title: Maraviroc reduces Treg

Article Type: Major Article

Section/Category: HIV/AIDS

Keywords: Regulatory T cells (Treg), Maraviroc (MVC), HIV, ART-naïves

Corresponding Author: Yolanda Maria Pacheco, PhDInstitute of Biomedicine of Seville (IBIS) University Hospital Virgen del Rocío, Seville,SpainSeville, Seville SPAIN

Corresponding Author SecondaryInformation:

Corresponding Author's Institution: Institute of Biomedicine of Seville (IBIS) University Hospital Virgen del Rocío, Seville,Spain

Corresponding Author's SecondaryInstitution:

First Author: María Mar Pozo-Balado

First Author Secondary Information:

Order of Authors: María Mar Pozo-Balado

Marta Martínez-Bonet

Isaac Rosado

Ezequiel Ruiz-Mateos

Gema Méndez-Lagares

María Mar Rodríguez-Méndez

Francisco Vidal

María Angeles Muñoz-Fernández

Yolanda Maria Pacheco, PhD

Manuel Leal

Order of Authors Secondary Information:

Manuscript Region of Origin: SPAIN

Abstract: Background / Aim:Maraviroc (MVC) is the first antiretroviral (ART) drug to target a human protein, theCCR5 coreceptor; however, the mechanisms of MVC-associated immunomodulation inHIV subjects remain to be elucidated. Regulatory T cells (Tregs) play a key role in HIV-associated immunopathology and are susceptible to MVC-mediated CCR5 blockade.Our aim was to evaluate the effect of MVC on Tregs.Methods:We compared the effect of an MVC-containing or sparing combined-ART (cART) onART-naïve HIV subjects on Tregs. Tregs were characterized as CD4+CD25hiFoxP3+on day 0, 8 and 30. Additional analysis on week 48 was performed in a subgroup ofpatients. The potential reduction of Tregs was also tested in vitro on MVC-treatedperipheral blood mononuclear cells. The suppressive function of Tregs was alsoanalyzed in MVC-treated Tregs.Results:

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation

1

TITLE PAGE 1

RUNNING TITLE: 2

Maraviroc reduces Treg 3

TITLE: 4

Maraviroc reduces Regulatory T cells in ART-naïve HIV-infected subjects 5

AUTHOR AND AFFILIATIONS INFORMATION: 6

María Mar Pozo-Balado1, Marta Martínez-Bonet2, Isaac Rosado1, Ezequiel Ruiz-Mateos1, Gema 7 Méndez-Lagares1-3, María Mar Rodríguez-Méndez1, Francisco Vidal4, María Angeles Muñoz-8 Fernández2, Yolanda María Pacheco1*# and Manuel Leal1* 9

* Both authors have contributed equally to this work 10

1 Laboratory of Immunovirology, Clinic Unit of Infectious Diseases, Microbiology and 11 Preventive Medicine, Institute of Biomedicine of Seville, IBiS, Virgen del Rocío University 12 Hospital/CSIC/University of Seville, Seville 41013,Spain 13

2 Laboratory of Molecular Immuno-Biology, Hospital General Universitario Gregorio Marañon, 14 Madrid 28007, Spain 15

3 Department of Medical Microbiology and Immunology, University of California Davis (UC 16 Davis), Davis, California 95616, USA 17

4 Infectious Diseases and HIV/AIDS Unit, Department of Internal Medicine, Hospital Universitari 18 de Tarragona Joan XXIII, Universitat Rovira i Virgili, IISPV, Tarragona, Spain 19

20

# CORRESPONDING AUTHOR 21

Yolanda M. Pacheco, PhD 22

Laboratory of Immunovirology 23

Institute of Biomedicine of Seville (IBiS) 24

Clinic Unit of Infectious Diseases, Microbiology and Preventive Medicine 25

Virgen del Rocío University Hospital 26

41013, Seville, Spain 27

E-mail address: [email protected] 28

Manuscript-cleanClick here to download Manuscript: Clean manuscritpt.doc

2

ALTERNATE CORRESPONDING AUTHOR 29

Manuel Leal 30

Laboratory of Immunovirology 31

Institute of Biomedicine of Seville (IBiS) 32

Clinic Unit of Infectious Diseases, Microbiology and Preventive Medicine 33

Virgen del Rocío University Hospital 34

41013, Seville, Spain 35

E-mail address: [email protected] 36

37

WORD COUNT: 2988 38

ABSTRACT: 18739

3

ABSTRACT 40

Background / Aim: 41

Maraviroc is the first antiretroviral (ART) drug to target a human protein, the CCR5 coreceptor; 42

however, the mechanisms of Maraviroc-associated immunomodulation in HIV subjects remain 43

to be elucidated. Regulatory T cells (Tregs) play a key role in HIV-associated immunopathology 44

and are susceptible to Maraviroc-mediated CCR5 blockade. Our aim was to evaluate the effect 45

of Maraviroc on Tregs. 46

Methods: 47

We compared the effect of an Maraviroc-containing or sparing combined-ART (cART) on ART-48

naïve HIV subjects on Tregs. Tregs were characterized as CD4+CD25hiFoxP3+ on day 0, 8 and 49

30. Additional analysis on week 48 was performed in a subgroup of patients. The potential 50

reduction of Tregs was also tested in vitro on Maraviroc-treated peripheral blood mononuclear 51

cells. The suppressive function of Tregs was also analyzed in Maraviroc-treated Tregs. 52

Results: 53

We found that Maraviroc significantly reduced Tregs both short-term and after one year of 54

treatment initiation. In vitro experiments showed a dose-dependent reduction of Tregs after 55

treatment of PBMCs with Maraviroc, although their suppressive function was not altered. 56

Conclusions: 57

These findings partially explain Maraviroc-associated immunomodulatory effects and open 58

new therapeutic expectations for the development of Tregs-depleting immunotherapies. 59

60

KEYWORDS 61

Regulatory T cells (Treg), Maraviroc , HIV, ART-naïves 62

4

INTRODUCTION 63

Maraviroc is the first antiretroviral drug to target a human protein, the CCR5 chemokine 64

receptor. This protein also serves as a coreceptor for the entry of R5-tropic human 65

immunodeficiency virus (HIV) strains into host cells. CCR5 plays a key role in lymphocyte 66

activation, homing and proliferation [1], priming adaptive immune responses, and promoting 67

the migration of CCR5-expressing cells to sites of infection and inflammation [2,3]. 68

This particular mechanism of action confers exceptional properties to Maraviroc. Thus, 69

independently of its antiviral effect [4], Maraviroc may also have immunomodulatory effects 70

[5]. Accordingly, Maraviroc has been related to increased CD4 T cell counts [6,7] and reduced 71

CD4 and CD8 T cell activation [8], sCD14 levels [9], HIV-associated chronic inflammation [3] and 72

apoptosis [10]. Moreover, it was recently found that Maraviroc increases antigen-specific T cell 73

responses against several vaccine-related antigens by altering the cytokine profile [11]. Despite 74

this evidence, the cellular and molecular basis of Maraviroc-associated immunomodulation in 75

HIV subjects has not yet been revealed. 76

Regulatory T cells (Tregs), which suppress the intensity and duration of immune responses 77

[12], play a key role in HIV-associated immunopathology and are involved in the suppression of 78

T cell activation, proliferation, inflammation and cytokine production [13]. An increased 79

proportion of Tregs has been described in HIV-infected subjects, although it is still not clear 80

whether Tregs are detrimental or beneficial [13]. We have shown an inverse relationship 81

between the frequency of Tregs and the qualitative and quantitative response to the hepatitis 82

B virus (HBV) vaccine in HIV-infected subjects [14], suggesting that therapeutic strategies 83

directed at reducing Tregs in HIV-infected subjects prior to HBV vaccination are advisable. 84

Combined antiretroviral therapy (cART) has been reported to have an impact on Tregs, but 85

there are contradictory results about its capacity to normalize Tregs levels [15-20], most likely 86

5

due to different characterization approaches for this subset [21]. Moreover, no current 87

information regarding the effect of new antiretroviral drugs, such as Maraviroc, exists. It 88

should be noted that like other immune cells, Tregs express CCR5 [8], which makes them 89

susceptible both to HIV infection and to Maraviroc-mediated CCR5 blockade. Interestingly, the 90

level of expression of this receptor is variable in different Treg subsets (e.g. higher in effector 91

Tregs than in naïve Tregs) [22]. Altogether, these data prompted us to hypothesize that Tregs 92

could be modulated by Maraviroc, which could partially explain the Maraviroc-associated 93

immunomodulatory effects on HIV-infected subjects. Our aim in the present study was to 94

analyze the effect of Maraviroc on the Treg subset in ART-naïve HIV-infected subjects starting 95

antiretroviral treatment. 96

6

MATERIALS AND METHODS 97

Patients, samples and study design 98

For the present study, we retrospectively selected consecutive ART-naïve HIV-infected subjects 99

that started combined antiretroviral treatment (cART) between January 2010 and December 100

2012 at the Infectious Diseases Service at the Virgen del Rocío University Hospital, Seville, 101

Spain. We included patients who were initiating both types of cART: i) an Maraviroc-containing 102

regimen including Maraviroc (150mg/24h) plus a ritonavir-boosted protease inhibitor (PI/rtv); 103

and ii) a conventional cART including a combination of two nucleoside analog reverse-104

transcriptase inhibitors (NRTIs) and one non-nucleoside reverse-transcriptase inhibitors 105

(NNRTI) (10/12 of subjects) or a PI (2/12 of subjects). Patients starting an Maraviroc-106

containing regimen had been first tested for their sensitivity to this drug by the Maraviroc 107

Clinical Test as previously described [23]. Briefly, the Maraviroc Clinical Test consisted of an 8-108

day monotherapy exposure to Maraviroc (300mg/12h) and was considered positive when a 109

viral load reduction of at least 1 log10 HIV RNA copies/mL was achieved at day 8 post-exposure. 110

On day 10, depending on the result of viral load test from day 8, a combination regimen 111

including Maraviroc was started in patients with positive test. Only asymptomatic ART-naïve 112

subjects without HCV and/or HBV coinfection, and with available frozen samples from baseline 113

(day 0), day 8 and day 30 post-antiretroviral exposure were included in the study (n=21). Nine 114

of them had started an Maraviroc-containing cART (Maraviroc group), and the remaining 12 115

had started a conventional Maraviroc-sparing cART (cART group). Long-term follow-up (week 116

48) was also analyzed in patients without treatment changes during the follow-up period, and 117

with available cryopreserved samples (11 patients out of 21). All patients had undetectable 118

viral load at this time point. 119

7

For in vitro assays, 8 additional ART-naïve patients were prospectively recruited, and their 120

blood was freshly processed. From these blood collections, suppression assays were 121

performed if enough Tregs could be isolated (3 patients out of 8). All study participants gave 122

informed consent, and the study was approved by the Committee of Ethics at the Virgen del 123

Rocío University Hospital. 124

Flow cytometry 125

Peripheral blood mononuclear cells (PBMC) were isolated from samples and 126

immunophenotyped using multicolor flow cytometry to characterize Tregs as 127

CD4+CD25hiFoxP3+ T cells, as previously described [21]. Anti-FoxP3-PE (e-Bioscience, San Diego, 128

CA), anti-IL2R1 (CD25)-FITC and anti-CD4-ECD (Beckman-Coulter, Florida, USA) were used. The 129

intracellular FoxP3 staining was performed using the FoxP3 Staining Set (eBioscience) 130

according to the manufacturer's instructions. Data were acquired on a Cytomics FC 500 flow 131

cytometer (Beckman-Coulter) and were analyzed using CXP software (Beckman-Coulter). One 132

million cells of each sample were stained, and a minimum of 200,000 events was acquired. 133

In vitro Maraviroc treatment of PBMC 134

To analyze the effect of Maraviroc on Tregs, PBMC from ART-naïve patients were isolated, 135

seeded into 96-well U-bottom plates (500,000 cells/well) and treated with different 136

concentrations of Maraviroc (UK-427857, kindly provided by Pfizer). Maraviroc was 137

reconstituted in DMSO and stored at 4°C until use. Maraviroc concentrations were chosen on 138

the basis of pharmacokinetic published data [24, 25], and the final concentration of DMSO in 139

all the wells was 0.1% for all concentrations. In preliminary time-course experiments (not 140

shown), we observed optimum effects after 3 days of incubation (1 to 5 days tested). The cells 141

were incubated for 3 days with 1 μM, 10 μM and 100 μM Maraviroc in complete medium 142

[RPMI with 10% FBS, 1% L-glutamine and penicillin (100 U/mL)/streptomycin (100 μg/mL)] at 143

8

37°C under 5% atmospheric CO2, and PBMC incubated with 0.1% DMSO in complete medium 144

were used as controls. Viability of the cells was assessed by Trypan blue exclusion dye (≥85% at 145

all Maraviroc concentrations). After treatment of PBMC, the Tregs subset was quantified by 146

flow cytometry as CD4+CD25hiFoxP3+ T cells as described above. 147

Assessment of suppressive function of Tregs in vitro 148

To evaluate the suppressive function of Tregs in vitro against total PBMC, a Tregs suppression 149

assay was performed following the manufacturer’s instructions (BD FastImmune Regulatory T 150

Cell Function Kit). Briefly, PBMC were isolated from cART-naïve patients and incubated in X-151

VIVO-15 medium with 10% human AB serum in 96-well U-bottom plates. To isolate Tregs, the 152

Human Regulatory T cell Isolation Kit II (Miltenyi Biotech) was used. Purified autologous 153

CD4+CD127-CD25hi T cells (>95%) were left untreated or were treated with 100 μM Maraviroc 154

for 24 h before the assay. They were then added at a 1:1 ratio with resting autologous PBMC. 155

PBMC alone were used as controls. The cells were then stimulated with anti-CD3 and anti-156

CD28 coated beads (Dynabeads Human T-Activator CD3/CD28, Invitrogen) at a ratio of 1:2 157

(beads/PBMC) in the presence of CD154-APC reagent (supplied by the kit). After 7 hours of 158

culturing at 37°C under 5% atmospheric CO2, the samples were stained with CD4-FITC/CD25-159

PE/CD3-PerCP-Cy5.5/CD69-PE-Cy7 (supplied by the kit), and the presence of early activation 160

markers (CD154 and CD69) in PBMC was assessed by flow cytometry. 161

Laboratory measurements 162

Absolute numbers of CD4 T cells were determined with an Epics XL-MCL flow cytometer 163

(Beckman-Coulter). Plasma HIV-1 RNA levels were measured by quantitative polymerase chain 164

reaction (qPCR) (COBAS Ampliprep/COBAS Taqman HIV-1 test, Roche Molecular Systems) 165

according to the manufacturer's instructions. The detection limit was <40 HIV RNA copies/mL. 166

The plasma samples were tested for HBV-related markers (HBsAg, anti-HBs, and anti-HBc) and 167

9

anti-hepatitis C virus (HCV) antibodies using an HBV enzyme-linked immunosorbent assay 168

(ELISA) and an HCV ELISA, respectively (Siemens Healthcare Diagnosis). Quantitative PCR 169

amplification was used for plasma HCV amplification (COBAS Amplicor, Roche Diagnosis), and 170

the test had a detection limit of 15 IU/mL. 171

Statistical analysis 172

Continuous variables were recorded as median values and interquartile ranges, and categorical 173

variables were recorded as numbers and percentages. For multiple longitudinal comparisons, 174

the Friedman test and Bonferroni adjustment to a series of post hoc Wilcoxon matched pairs 175

tests were applied. Comparisons between groups were made using the non-parametric Mann-176

Whitney U-test for continuous variables and the χ2 or Fisher exact test for categorical 177

variables, as appropriate. Statistical analyses were performed using SPSS (Statistical Package 178

for the Social Sciences) software (SPSS 18.0, Chicago, IL, USA), and the graphics were 179

generated with the Prism version 5.0 program (GraphPad Software, Inc.). Except for the 180

Bonferroni correction (p<0.017), p<0.05 were considered statistically significant for all tests. 181

10

RESULTS 182

Baseline characteristics of HIV-infected subjects 183

Baseline characteristics of patients included in the Maraviroc and cART groups are summarized 184

in Table 1. There were no differences between them with respect to sex, age, CD4 T cell 185

counts, CD4 T cell nadir, viral load (VL) or time from diagnosis. 186

Short-term reduction of Tregs in Maraviroc-treated patients 187

At baseline, the median frequency of Tregs was similar in the Maraviroc and cART groups (0.51 188

[0.41-0.65] and 0.57 [0.40-0.88], respectively; p=0.602). One month after initiating treatment, 189

a reduction in Tregs frequency was only observed in the Maraviroc group (Friedman test, 190

p=0.001 vs. p=0.127) (Fig. 1 A-B). This reduction was more pronounced during monotherapy 191

exposure to Maraviroc (day 0 vs. day 8, Wilcoxon test, p=0.007), followed by a gradual 192

decrease until day 30 (day 8 vs. day 30, Wilcoxon test, p=0.028). Similar results were found 193

when analyzing absolute Tregs numbers (Supplementary Table 1). Conversely, decreases in 194

viral load and increases in CD4 T cell counts during treatment were similar between groups 195

(Fig. 1 C-F). 196

Notably, graphs from both treatment groups showed very different profiles. Considering the 197

day 0-day 8 period, the Maraviroc group showed a consistent and pronounced reduction, 198

whereas the cART group showed a heterogeneous profile. We asked whether the absence of 199

Maraviroc could explain such a profile in the cART group. To address this issue, we analyzed 200

the spontaneous variation of Tregs frequency between baseline and day 8 in a group of 7 age- 201

and sex-matched, cART-naïve HIV patients not initiating treatment. We observed that the 202

untreated cART-naïve patients showed the same heterogeneous behavior for Tregs as the 203

cART-treated group (Supplementary figure 1A). These results support that a conventional cART 204

has no effect on the Treg subset, whereas Maraviroc induced a regular trend to reduce Tregs. 205

11

Finally, to check whether HIV infection could explain the heterogeneous profile in the cART 206

group, we analyzed the spontaneous variation of Tregs between baseline and day 8 from 5 207

age- and sex-matched subjects, and we observed no modifications in Tregs frequency 208

(Supplementary figure 1B) although, as expected, levels were much lower than in HIV subjects, 209

what could determine their absence of variability. 210

Long-term reduction of Tregs in Maraviroc-treated patients 211

We next assessed the long-term impact of either treatment, including Maraviroc or not, on the 212

Tregs subset. A subgroup of 5 patients in the Maraviroc group and 7 patients in the cART group 213

were analyzed because had available samples and no treatment changes during the 48 weeks 214

of treatment. The baseline characteristics of both subgroups were similar to those of the 215

overall study population (data not shown). All patients had undetectable viral load at this time 216

point. Thus, a significant reduction in HIV viremia (p<0.001 in both) and increase in CD4 T cell 217

counts was observed in the Maraviroc and cART groups (p=0.043 and p=0.013, respectively). 218

Both groups showed a reduction in Tregs frequency (Fig. 2A-B). However, Tregs returned to 219

healthy levels after 48 weeks of treatment only in the Maraviroc group (Fig. 2C). 220

Maraviroc reduces Tregs in a dose-dependent manner in vitro 221

We analyzed the effect of culturing freshly isolated PBMC from 8 cART-naïve patients with 222

Maraviroc on the frequency of Tregs. As observed in Figure 3, a significant dose-dependent 223

reduction in the frequency of Tregs (Fig. 3A) and a reduced FoxP3 expression of these cells (Fig. 224

3B) was observed in all tested samples, while the CD4 T cell frequency remained invariable 225

(Fig. 3C). 226

Maraviroc treatment does not affect the suppressive function of Tregs 227

12

We then investigated whether Tregs retain their suppressive activity after being treated with 228

Maraviroc. To do this, we measured expression of the early activation markers CD69 and 229

CD154 on CD4 T cells after PBMC stimulation in the presence of Tregs previously treated or not 230

with Maraviroc. As shown in Figure 4, Maraviroc-pretreated Tregs showed the same 231

suppressor ability as untreated Tregs. 232

13

DISCUSSION 233

In this study we show that in ART-naïve HIV-infected subjects, an Maraviroc-containing cART 234

mediated both short- and long-term reductions of Tregs and promoted a return to healthy 235

values after 48 weeks of treatment. Furthermore, since such reduction was observed even 236

during the initial monotherapy by Maraviroc Clinical Test, it is reasonably to think about a 237

Maraviroc-specific effect. Accordingly, in vitro experiments showed a dose-dependent 238

reduction of Tregs in Maraviroc-treated PBMC, while the suppressive function of Maraviroc-239

treated Tregs was not disturbed by the treatment. Our results contribute to the understanding 240

of the immunomodulatory properties of Maraviroc, and they reveal a novel effect on Tregs. 241

Despite some discordant results [18-20], which are likely due to different methods of 242

phenotyping Tregs [21], antiretroviral treatment has been shown to reduce Treg frequency 243

after a long-term period, although no short-term variations in this subset have been reported 244

in the literature [15-17]. Consistently, in our study, the control cART group (under an 245

Maraviroc-sparing cART) showed a long-term soft reduction of Treg frequency, but there was 246

no evidence of a reduction after either the first 8 or the first 30 days of treatment. It is 247

noteworthy that the control cART group regimen contained three active drugs instead of one. 248

In spite of this, Maraviroc-mediated Tregs reduction was evident as soon as day 8 after the 249

initiation of treatment, where it was administered as monotherapy, allowing us to conclude 250

that this effect was drug-exclusive. Such a specific effect could be explained by the differential 251

expression of CCR5 by Tregs compared to other cellular subsets. In fact, Tregs have been 252

shown to express the CCR5/CXCR4 ratio of coreceptors differently than T effector cells [26], 253

and activated Tregs more strongly upregulate CCR5 than activated Th1 and Th2 cells [27]. 254

Furthermore, Tregs are constituted by functionally different subsets with different pattern of 255

expression of CCR5, which could imply different susceptibility to Maraviroc. Thus, effector 256

14

Tregs express the highest levels of CCR5 whereas naïve Tregs are mainly CXCR4 expressing cells 257

[22]. 258

Importantly, this Maraviroc-mediated reduction of Tregs could partially explain the recent 259

observation of Maraviroc-mediated increases in cellular and humoral immune responses 260

against vaccination [11]. Accordingly, our previous data suggested a negative role of Tregs in 261

the response to HBV vaccination in HIV subjects [14]. Future studies will definitely explore if 262

such defective responses could be improved by using Maraviroc as a vaccine coadjuvant or by 263

including Maraviroc in the antiretroviral regimen of HIV subjects undergoing HBV 264

immunization. 265

Regarding the potential mechanisms involved in Maraviroc-mediated Tregs reduction, we can 266

only speculate at this moment. Considering our in vitro experiments, we cannot exclude a 267

direct cellular effect on Tregs. In fact, Maraviroc seems to induce downregulation of FoxP3 268

expression on Tregs, and we cannot rule out an increase of apoptosis of these cells, but it 269

could also act preferentially on other conventional T cells that could mediate this Tregs 270

reduction. However, due to the high viability of cultured PBMCs and the absence of fluctuation 271

in total CD4 T cells, a cytotoxic effect of Maraviroc can be ruled out. On the other hand, our in 272

vivo data could also be explained by the blockade of CCR5-dependent Treg migration, involving 273

primary lymphoid tissue retention of Tregs, and hence, their peripheral reduction [28]. 274

Notably, blockade of CCR5-dependent Treg chemotaxis has also been postulated as the main 275

mechanism of its immunomodulation in other scenarios. Thus, Maraviroc has already been 276

used as an anti-tumor drug in mouse models [32,34]. For example, in cancer, where Tregs 277

recruitment is associated with tumor progression [29-31], Maraviroc blocks metastasis of basal 278

breast cancer cells by blocking such recruitment [32, 33]. Furthermore, Maraviroc blocked 279

local Tregs suppression of immune responses against fungal and Leishmania infections [35, 280

15

36], and it appears to be an effective new strategy against graft-versus-host disease [37], 281

mediated by Tregs recruitment during transplants. 282

Conversely, it should be noted that the in vitro suppressive capacity of Tregs remained 283

invariable after Maraviroc pre-treatment. However, as a consequence of decreased Tregs 284

frequency and numbers, global immune suppression would be likely reduced, and the effector 285

response could be enhanced [11]. 286

This work has several limitations and opens new questions. The small sample size could have 287

prevented us to find a higher CD4 T cell increase in the Maraviroc group, which has been 288

described by other groups [6, 7]. On the other hand, our results cannot be extrapolated to 289

different scenarios of HIV infection other than ART-naïve subjects, where the effect of 290

Maraviroc should be investigated; and we do not know whether this effect is reversible, and a 291

Tregs expansion could be expected after the MCV-containing cART interruption [38]. 292

Additionally, these questions need to be addressed in a future: a) the specific effect of 293

Maraviroc on different Treg subsets (i.e., naïve and effectors), b) the Maraviroc-mediated 294

mechanisms of Treg reduction and c) the in vivo effect of Maraviroc treatment on Treg 295

function. 296

297

Finally, the observed dose-dependent in vitro effect of Maraviroc on Tregs, as well as the in 298

vivo effect during monotherapy (where the dose was four times higher than normal; 300 299

mg/each 12 h vs. 150 mg/each 24 h in Maraviroc-containing cART [39]), are important issues 300

considering HIV patients under Maraviroc could be exposed to lower drug doses than those 301

necessary for immunological benefit. We could speculate that a greater in vivo effect could be 302

observed when used at higher doses. 303

16

Our results shed light on the immunomodulatory properties of Maraviroc and have potential 304

clinical implications for the development of Tregs-depleting immunotherapies in different 305

scenarios, such as chronic infections, cancer and vaccination. 306

17

FUNDING 307

This work was supported by grants from the Fondo de Investigación Sanitaria [FIS; P11/02014], 308

the Spanish AIDS Research Network of Excellence [RIS; RD12/0017/0029, RD12/0017/0037], 309

the Ministerio de Sanidad, Política Social e Igualdad [EC11-520], the Fundación Progreso y 310

Salud [PI-0081-2011], and Pfizer/ViiV Healthcare [grant numbers WS843473, WS2425049]. 311

Y.M. Pacheco was supported by the Fondo de Investigación Sanitaria through the “Miguel 312

Servet” programs [CP07/00240, CPII13/00037], and by the Consejeria de Salud y Bienestar 313

Social of Junta de Andalucia through the “Nicolas Monardes” program [C-0010/13]. E Ruiz-314

Mateos was supported by the Fondo de Investigación Sanitaria through the “Miguel Servet” 315

program [CP08/0172]. 316

317

ACKNOWLEDGMENT 318

The authors thank all patients and volunteers for their collaboration, and Magdalena 319

Rodríguez Díez and Marien Gutiérrez Sancho for their clinical assistance to HIV-infected 320

subjects included in this study. 321

322

AUTHOR CONTRIBUTION 323

MM.P-B. contributed to the design, performed research, analyzed and interpreted data and 324

wrote the manuscript. M. M-B performed research and review the manuscript. I. R., G. M-L., 325

and MM. R-M. contributed to several experiments and review the manuscript. E. R-M., F. V. 326

and MA. M-F. contributed to interpret data and review the manuscript. YM.P and M.L. 327

designed the study, analyzed and interpreted data, drafted and revised the manuscript. 328

329

18

POTENTIAL CONFLICT OF INTEREST 330

E. R-M. and M. L. have received grants from Pfizer. 331

19

REFERENCES 332

1. Camargo JF, Quinones MP, Mummidi S, et al. CCR5 expression levels influence NFAT 333

translocation, IL-2 production, and subsequent signaling events during T lymphocyte 334

activation. J Immunol 2009; 182:171-82. 335

2. Lederman MM, Penn-Nicholson A, Cho M, Mosier D. Biology of CCR5 and its role in HIV 336

infection and treatment. JAMA 2006; 296:815-26. 337

3. Rossi R, Lichtner M, De Rosa A, et al. In vitro effect of anti-human immunodeficiency virus 338

CCR5 antagonist maraviroc on chemotactic activity of monocytes, macrophages and 339

dendritic cells. Clin Exp Immunol 2011; 166:184-90. 340

4. Fätkenheuer G, Pozniak AL, Johnson MA, et al. Efficacy of short-term monotherapy with 341

maraviroc, a new CCR5 antagonist, in patients infected with HIV-1. Nat Med 2005; 11:1170-342

2. 343

5. Wilkin TJ, Gulick RM. CCR5 antagonism in HIV infection: current concepts and future 344

opportunities. Annu Rev Med 2012; 63:81-93. 345

6. Asmuth DM, Goodrich J, Cooper DA, et al. CD4+ T-cell restoration after 48 weeks in the 346

maraviroc treatment-experienced trials MOTIVATE 1 and 2. J Acquir Immune Defic Syndr 347

2010; 54:394-7. 348

7. Saag M, Goodrich J, Fätkenheuer G, et al. A double-blind, placebo-controlled trial of 349

maraviroc in treatment-experienced patients infected with non-R5 HIV-1; A4001029 Study 350

Group. J Infect Dis 2009; 199:1638-47. 351

8. Funderburg N, Kalinowska M, Eason J, et al. Effects of maraviroc and efavirenz on markers 352

of immune activation and inflammation and associations with CD4+ cell rises in HIV-353

infected patients. PLoS One 2010; 5:e13188. 354

20

9. Romero-Sánchez MC, Machmach K, Gonzalez-Serna A, et al. Effect of maraviroc on HIV 355

disease progression-related biomarkers. Antimicrob Agents Chemother 2012; 56:5858-64. 356

10. Wilkin TJ, Lalama CM, McKinnon J, et al. A pilot trial of adding maraviroc to suppressive 357

antiretroviral therapy for suboptimal CD4⁺ T-cell recovery despite sustained virologic 358

suppression: ACTG A5256. J Infect Dis 2012; 206:534-42. 359

11. Westrop SJ, Moyle G, Jackson A, Nelson M, Mandalia S, Imami N. CCR5 antagonism impacts 360

vaccination response and immune profile in HIV-1 infection. Mol Med 2012; 18:1240-8. 361

12. Jiang H, Chess L. Regulation of immune responses by T-cells. N Engl J Med 2006; 354;1166-362

1176. 363

13. Chevalier MF, Weiss L. The split personality of regulatory T cells in HIV infection. Blood 364

2013; 121:29-37. 365

14. del Pozo-Balado M del M, Leal M, Méndez Lagares G, et al. Increased regulatory T-cell 366

counts in HIV-infected nonresponders to hepatitis B virus vaccine. J Infect Dis 2010; 367

202:362-9. 368

15. Bi X, Suzuki Y, Gatanaga H, Oka S. High frequency and proliferation of CD4+ FOXP3+ Treg in 369

HIV-1-infected patients with low CD4 counts. Eur J Immunol 2009; 39:301-9. 370

16. Montes M, Sanchez C, Lewis DE, et al. Normalization of FoxP3(+) regulatory T-cells in 371

response to effective antiretroviral therapy. J Infect Dis 2011; 203:496-9. 372

17. Zhang M, Zhang H, Zhang T, Ji Y, Jiao Y, Wu H. Longitudinal changes of peripheral blood DC 373

subsets and regulatory T-cells in Chinese chronic HIV-1-infected patients during 374

antiretroviral therapy. PLoS One 2012; 7:e37966. 375

18. Kolte L, Gaardbo JC, Skogstrand K, Ryder LP, Ersbøll AK, Nielsen SD. Increased levels of 376

regulatory T-cells (Treg) in human immunodeficiency virus-infected patients after 5 years of 377

21

highly active anti-retroviral therapy may be due to increased thymic production of naive 378

Treg. Clin Exp Immunol 2009; 155:44-52. 379

19. Gaardbo JC, Nielsen SD, Vedel SJ, et al. Regulatory T-cells in human immunodeficiency 380

virus-infected patients are elevated and independent of immunological and virological 381

status, as well as initiation of highly active anti-retroviral therapy. Clin Exp Immunol 2008; 382

154:80-6. 383

20. Lim A, Tan D, Price P, et al. Proportions of circulating T-cells with a regulatory cell 384

phenotype increase with HIV-associated immune activation and remain high on 385

antiretroviral therapy. AIDS 2007; 21:1525-34. 386

21. Del Pozo-Balado Mdel M, Leal M, Méndez-Lagares G, Pacheco YM. 387

CD4(+)CD25(+/hi)CD127(lo) phenotype does not accurately identify regulatory T-cells in all 388

populations of HIV-infected persons. J Infect Dis 2010; 201:331-5. 389

22. Arruvito L, Sabatté J, Pandolfi J et al. Analysis of suppressor and non-suppressor FOXP3+ T 390

cells in HIV-1-infected patients. PLoS One 2012; 7(12):e52580. 391

23. Genebat M, Ruiz-Mateos E, León JA, et al. Correlation between the Trofile test and 392

virological response to a short-term maraviroc exposure in HIV-infected patients. J 393

Antimicrob Chemother 2009; 64:845-9. 394

24. Emmelkamp JM, Rockstroh JK. CCR5 antagonists: comparison of efficacy, side effects, 395

pharmacokinetics and interactions--review of the literature. Eur J Med Res 2007; 12:409-17. 396

25. Lieberman-Blum SS, Fung HB, Bandres JC. Maraviroc: a CCR5-receptor antagonist for the 397

treatment of HIV-1 infection. Clin Ther 2008; 30:1228-50. 398

26. Moreno-Fernandez ME, Zapata W, Blackard JT, Franchini G, Chougnet CA. Human 399

regulatory T-cells are targets for human immunodeficiency Virus (HIV) infection, and their 400

susceptibility differs depending on the HIV type 1 strain. J Virol 2009; 83:12925-33. 401

22

27. Sebastiani S, Allavena P, Albanesi C, et al. Chemokine receptor expression and function in 402

CD4+ T lymphocytes with regulatory activity. J Immunol 2001; 166:996-1002. 403

28. Mailloux AW, Young MR. Regulatory T-cell trafficking: from thymic development to tumor-404

induced immune suppression. Crit Rev Immunol 2010; 30:435-47. 405

29. Chang LY, Lin YC, Mahalingam J, et al. Tumor-derived chemokine CCL5 enhances TGF-β-406

mediated killing of CD8(+) T-cells in colon cancer by T-regulatory cells. Cancer Res 2012; 407

72:1092-102. 408

30. Schlecker E, Stojanovic A, Eisen C, et al. Tumor-infiltrating monocytic myeloid-derived 409

suppressor cells mediate CCR5-dependent recruitment of regulatory T-cells favoring tumor 410

growth. J Immunol 2012; 189:5602-11. 411

31. Tan MC, Goedegebuure PS, Belt BA, et al. Disruption of CCR5-dependent homing of 412

regulatory T-cells inhibits tumor growth in a murine model of pancreatic cancer. J Immunol 413

2009; 182:1746-55. 414

32. Velasco-Velázquez M, Jiao X, De La Fuente M, et al. CCR5 antagonist blocks metastasis of 415

basal breast cancer cells. Cancer Res 2012; 72:3839-50. 416

33. Velasco-Velázquez M, Pestell RG. The CCL5/CCR5 axis promotes metastasis in basal breast 417

cancer. Oncoimmunology 2013; 2:e23660. 418

34. Byrne WL, Mills KH, Lederer JA, O'Sullivan GC. Targeting regulatory T-cells in cancer. Cancer 419

Res 2011; 71:6915-20. 420

35. Kroetz DN, Deepe GS. CCR5 Dictates the Equilibrium of Proinflammatory IL-17+ and 421

Regulatory Foxp3+ T-cells in Fungal Infection. J Immunol 2010; 184:5224–5231. 422

36. Yurchenko E, Tritt M, Hay V, Shevach EM, Belkaid Y, Piccirillo CA. CCR5-dependent homing 423

of naturally occurring CD4+ regulatory T-cells to sites of Leishmania major infection favors 424

pathogen persistence. J Exp Med 2006; 203:2451–2460. 425

23

37. Reshef R, Luger SM, Hexner EO, et al. Blockade of Lymphocyte Chemotaxis in Visceral 426

Graft-versus-Host Disease. N Engl J Med 2012; 367:135–145. 427

38. Weiss L, Piketty C, Assoumou L, et al. Relationship between regulatory T-cells and immune 428

activation in human immunodeficiency virus-infected patients interrupting antiretroviral 429

therapy.; ANRS 116 SALTO study group. PLoS One 2010; 5:e11659. 430

39. Mills A, Mildvan D, Podzamczer D, et al. Maraviroc once-daily nucleoside analog-sparing 431

regimen in treatment-naive patients: randomized, open-label pilot study. J Acquir Immune 432

Defic Syndr. 2013;62:164-70. 433

24

Table 1. Baseline Demographic and Clinical Characteristics of the Maraviroc and cART subjects 434

included in the study. 435

436

Characteristics Maraviroc group

(n=9) cART group (n=12) p value

Male sex 10 (83.3) 8 (88.9) 0.916

Age (years) 40.5 [27.3–45.5] 32.0 [28.5–35.0] 0.382

Time from diagnosis (weeks) 9.0 [7.0–16.0] 18.0 [4.50–52.5] 0.931

CD4+ count (cells/l) 409 [388–464] 360 [207-476] 0.456

Nadir (CD4+ T cells/l) 387 [363-468] 317 [139-441] 0.169

log VL (copies RNA/ml) 4.8 [4.2-5.1] 4.8 [4.2-5.3] 0.602

The data are number (%) of participants or median value [interquartile range] as appropriate, unless

otherwise noted. VL: Viral load. Mann-Whitney U-test was used for comparison between Maraviroc

and cART groups.

25

Supplementary Table 1. Comparison of absolute number of Tregs (cells/μl) between Maraviroc and cART groups. 437

438

439

440

441

442

The data are number (%) of participants or median value [interquartile range] as appropriate, unless otherwise indicated. Willcoxon test was used for paired 443

longitudinal comparisons in the same group. 444

Characteristics Tregs Day 0

(A)

Tregs Day 8

(B)

Tregs Day 30

(C)

Tregs Week 48

(D) p A vs B p A vs C p A vs D

Maraviroc (n=9) 20.9 [17.8 – 28.5] 15.6 [12.2 – 30.1] 16.7 [9.3 – 26.6] 17.3 [9.7 – 23.2] 0.008 0.008 0.048

cART (n=12) 18.1 [11.1 – 24.2] 19.3 [10.7 – 36.4] 16.7 [10.2 – 35.6] 21.3 [9.7 – 27.2] 0.386 0.093 0.043

26

Figure 1. Changes in Tregs frequency, viral load and CD4 T cell counts after the initiation of 445

treatment. The start of treatment resulted in a rapid reduction in Tregs cell frequency in the 446

Maraviroc group at day 8 and 30 (A), but no changes were observed in the cART group at any 447

time point (B). Both groups showed a similar sharp decrease in viral loads together with a 448

similar increase in CD4 T cell levels (C-F). Corresponding points to subjects who increased Treg 449

levels at day 8 have been highlighted with discontinuous lines both in panel 1B and 1D. Despite 450

some lines overlap, especially in panel 1B, all patients have been represented in all panels. 451

Wilcoxon test for longitudinal comparisons were assessed (p>0.05). 452

Figure 2. Tregs frequency after 48 weeks of successful treatment in Maraviroc (A) and cART (B) 453

groups. The frequency of Tregs was significantly higher at baseline in both study groups than in 454

age-matched healthy subjects, but no differences were found between the groups (p=0.89). 455

After 48 weeks of treatment, despite of all subjects showed undetectable viral load and 456

increases in CD4 T cell counts, Tregs frequency was returned to healthy levels only in the 457

Maraviroc group, where the levels were significantly lower than in the cART group (p=0.011) 458

(C). Continuous lines indicate Mann-Whitney U-tests, and dashed lines indicate Wilcoxon tests. 459

Figure 3. Effect of Maraviroc on Treg percentage (A), FoxP3 expression in Treg cells measured 460

as MFI (B) and CD4 T cell frequencies (C) in cultured PBMCs. Isolated peripheral blood 461

mononuclear cells (PBMCs) from 8 antiretroviral-naive HIV subjects were treated with 462

Maraviroc at 1, 10 and 100 M and cultured for 72 hours. Tregs and CD4 T cell frequencies 463

were assessed by flow cytometry. Values of percentages are relatives to the control, 464

considered as 100% (PBMCs in the presence of vehicle). A significant dose-dependent 465

reduction (Friedman test p<0.001) of Tregs and MFI of FoxP3 expression was observed, 466

whereas the CD4 T cell frequency remained invariable (Friedman test=0.120). Cell survival was 467

assessed by the Trypan blue exclusion dye test (>85% in all conditions). 468

27

Figure 4. Tregs suppression of the expression of early activation markers CD154 and CD69 on 469

CD4 T cells. The representative plots correspond to one experiment (A), although data were 470

collected from three different experiments for CD154 (B) and CD69 (C). Peripheral blood 471

mononuclear cells from 3 cART-naïve HIV subjects were stimulated with anti-CD3/anti-CD28 in 472

the absence or presence of autologous Tregs (both non-treated or Maraviroc-pretreated 473

during 24 h). Values are calculated following the manufacturer’s instructions (relative to the 474

control: PBMCs stimulated). Maraviroc-pretreated Tregs maintained the ability to suppress 475

CD4 T cell activation. US: unstimulated; ST: stimulated; Tregs NT: non-treated Tregs; Tregs 476

Maraviroc: Maraviroc pre-treated Tregs. 477

Supplementary Figure 1. Changes in the frequency of Tregs along 8 days of follow-up in 7 478

untreated cART-naïve HIV-infected subjects (A) and 5 healthy subjects (HS) (B). There were no 479

differences in the Tregs subset between day 0 and 8 in any control group (p>0.05; Wilcoxon 480

test). 481

Figure 1Click here to download high resolution image

Figure 2Click here to download high resolution image

Figure 3Click here to download high resolution image

Figure 4Click here to download high resolution image

Supplementary FigureClick here to download high resolution image