HIV-1 Single-Stranded RNA Induces CXCL13 Secretion in Monocytes via TLR7 Activation and Plasmacytoid −Derived Type I IFN This information is current as of October 2, 2021. Kristen W. Cohen, Anne-Sophie Dugast, Galit Alter, M. Juliana McElrath and Leonidas Stamatatos J Immunol 2015; 194:2769-2775; Prepublished online 9 February 2015;

doi: 10.4049/jimmunol.1400952 Downloaded from http://www.jimmunol.org/content/194/6/2769

References This article cites 49 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/194/6/2769.full#ref-list-1 http://www.jimmunol.org/

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

by guest on October 2, 2021 • Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2015 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

HIV-1 Single-Stranded RNA Induces CXCL13 Secretion in Human Monocytes via TLR7 Activation and Plasmacytoid Dendritic Cell–Derived Type I IFN

Kristen W. Cohen,*,†,1 Anne-Sophie Dugast,‡ Galit Alter,‡ M. Juliana McElrath,x,1 and Leonidas Stamatatos*,†,1

Elevated levels of the CXCL13 have been observed in the plasma of chronically HIV-1–infected subjects and have been correlated with plasma viremia, which in turn has been linked to progressive dysregulation of humoral responses. In this study we sought to identify mechanisms of CXCL13 induction in response to HIV-1 infection. Plasma levels of CXCL13 in HIV-1–infected antiretroviral therapy–naive subjects correlated with viral load and were higher compared with antiretroviral therapy–treated HIV-1–infected and HIV-1–uninfected subjects. To elucidate the relationship between HIV-1 viremia and CXCL13 plasma levels, PBMCs from uninfected donors were stimulated with HIV-1 infectious virions, HIV-1 ssRNA, TLR 7 and 8 agonists, or IFN-a. Downloaded from The cellular sources of CXCL13 were determined by intracellular staining of cell populations. CXCL13 was produced by monocytes after stimulation with TLR 7 and 8 ligands or HIV-1–derived ssRNA. CXCL13 production by monocytes required TLR7 activation of plasmacytoid dendritic cells and secretion of type I IFN. IFN-a alone was sufficient to induce CXCL13 expression in human monocytes. In sum, we identified a novel mechanism of HIV-1–induced CXCL13 secretion—one caused by TLR7 induction of type I IFN by plasmacytoid dendritic cells and subsequent IFN stimulation of monocytes. Our findings are relevant in understanding how HIV-1 infection leads to immune dysregulation and provide the opportunity to develop and test http://www.jimmunol.org/ potential therapeutic interventions. The Journal of Immunology, 2015, 194: 2769–2775.

characteristic immunological defect of HIV-1 infection is high-affinity Ab-secreting memory B and plasma cells are generated progressive humoral dysregulation (reviewed in Refs. 1, 2), (15, 16). Conversely, aberrant CXCL13 secretion has been impli- A which includes changes in the frequencies of specific cated in the pathogenesis of many chronic inflammatory conditions, subsets (3), and results in hypergammaglobulinemia (4) and in the including various infections and autoimmune disorders associated impaired induction of de novo Ab responses to vaccination (5). Al- with dysregulated lymphoid genesis and humoral responses (17–20). though evidence exists that some of the above-mentioned defects are Elevated levels of CXCL13 have been observed in the plasma of by guest on October 2, 2021 associated with viremia (6–8), the specific mechanisms of B cell chronically HIV-1–infected subjects and have been correlated with dysregulation have yet to be elucidated. Understanding such mech- viremia (21–24). In addition, it has been shown that CXCL13 anisms may assist in the development of therapeutic interventions to plasma levels decline after suppression of virus replication by anti- restore B cell functionality in chronic HIV-1 infection. retroviral therapy (ART) (21). Furthermore, increased CXCL13 CXCL13 is a chemokine crucial for the development of secondary plasma levels were found to be associated with reduced chemokine lymphoid structures, where it is secreted by follicular dendritic cells CXCR5 expression on B cells, with alterations in the che- (DCs) in response to receptor activation (9, 10). CXCL13 motactic potential of B cells, and correlated inversely with the fre- is chemotactic for cells expressing the receptor CXCR5 (9, 11), quency of circulating Tfh cells during chronic HIV-1 infection (22, including mature B cells and T follicular helper (Tfh) cells (12–14), 24). It appears, therefore, that CXCL13 is a factor linked with the and is expressed at high levels in the B cell follicles of lymphoid dysregulation of B cell function during HIV-1 infection. However, it organs (9). Importantly, CXCL13 facilitates the comigration of is unclear how HIV-1 infection leads to increased plasma CXCL13 B cells and Tfh cells into B cell follicles and germinal centers, where concentrations and how changes in CXCL13 plasma concentration are linked with B cell dysregulation during HIV-1 infection. Al- though increased transcriptional expression of CXCL13 is detected *Seattle Biomedical Research Institute, Seattle, WA 98109; †Department of Global Health, University of Washington, Seattle, WA 98195; ‡Ragon Institute of MGH, in B cells from HIV-1–infected subjects, secretion of CXCL13 by MIT and Harvard, Cambridge, MA 02139; and xDepartment of Medicine, University those B cells is detectable only at low levels and only after stimu- of Washington, Seattle, WA 98195 lation for 6 d with CD40 ligation, CpG, IL-2, and IL-10 (22). 1 Current address: Vaccine and Infectious Disease Division, Fred Hutchinson Therefore, the reported increases in CXCL13 concentrations during Research Center, Seattle, WA. HIV-1 infection are likely not due to increases in CXCL13 pro- Received for publication April 14, 2014. Accepted for publication January 8, 2015. duction by B cells. In contrast, in biopsies from these This work was supported by National Institutes of Health Grants AI081625 (to L.S.) same subjects, the majority of CXCL13 was found in macrophages and AI080289 (to M.J.M.). and immature DCs (22). Whether these two cell types secrete Address correspondence and reprint requests to Dr. Leonidas Stamatatos at the cur- rent address: Fred Hutchinson Cancer Research Center, 1100 Fairview Avenue North, CXCL13 in the periphery is unknown. It is also not understood Room E5-420, Seattle, WA 98109. E-mail address: [email protected] whether CXCL13 expression by macrophages and immature DCs is Abbreviations used in this article: ART, antiretroviral therapy; AZT, azidothymidine; the direct result of their infection by HIV-1 or is due to a bystander DC, dendritic cell; mDC, myeloid DC; pDC, plasmacytoid DC; Tfh, T follicular effect of immune activation that takes place during HIV-1 infection. helper; TLR7/8, TLRs 7 and 8. In this regard, it is known that HIV-1–derived RNA triggers Copyright Ó 2015 by The American Association of Immunologists, Inc. 0022-1767/15/$25.00 cytokine secretion directly through activation of TLRs 7 and 8 www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400952 2770 HIV-1 INDUCTION OF CXCL13 IN MONOCYTES

(TLR7/8) (25, 26), which are predominantly expressed by APCs mAbs against CD19, CD3, CD14, CD11c, and CD123 (BD Biosciences) such as monocytes, monocyte-derived DCs, and myeloid cells (27). and analyzed using an LSR II flow cytometer. In addition, monocytes, macrophages, and monocyte-derived DCs In vitro stimulation of PBMCs with TLR ligands have been identified as inducible producers of CXCL13 by TLR2 A total of 2 million freshly isolated PBMCs or 2 3 105 monocytes incubated and TLR4 activation (17, 20, 28–30). TLRs 2, 4, 7, and 8 share over- with or without 25,000 purified pDCs were cultured in 1 ml complete media lapping signaling pathways and downstream transcription factors. and treated for 1–4 d with the TLR7/8 agonist R848 (Invivogen) at 1 mg/ml. Therefore, increases in CXCL13 plasma concentrations during Increasing concentrations of HIV-1 long terminal repeat–derived GU-rich HIV-1 infection could be due to secretion of this chemokine by any ssRNA40/LyoVec (59-GCCCGUCUGUUGUGUGACUC-39) and negative (or all) of these cells through TLR-activated pathways. control RNA, in which U nucleotides were replaced with adenosine, ssRNA41/LyoVec (59-GCCCGACAGAAGAGAGACAC-39) (26) (Inviv- In this article, we report that CXCL13 is secreted by monocytes in ogen), and recombinant human IFN-a2A (PBL IFN) were added to cells. response to stimulation with HIV-1 virions, HIV-1–derived ssRNA, The soluble IFNR B18R (Affymetrix) was added (at 0.1 mg/ml) to cells and the TLR7/8 ligand R848. Of interest, maximal CXCL13 directly preceding stimulation with the above TLR7/8 agonists to neutralize m production required plasmacytoid DC (pDC) secretion of IFN-I. type I IFN in the supernatant. For time course experiments, 500 l super- natant was removed and replaced with fresh media at 24-h intervals. Therefore, our study reveals a novel mechanism of CXCL13 in- Concentrations of CXCL13 and IFN-a in culture supernatants were deter- duction in human monocytes via the TLR7 stimulation of pDCs mined by ELISA according to the manufacturers’ instructions (IFN-a and secretion of IFN-I and subsequent IFN-induced production of from PBL Biomedical Laboratories; CXCL13 kits from R&D systems). CXCL13 in monocytes. Quantification of mRNA by real-time PCR Materials and Methods Total RNA was isolated from monocytes and monocyte-depleted PBMCs, using the RNeasy Kit (QIAGEN), and was reverse transcribed into cDNA Downloaded from Study participants using the QuantiTect Reverse Transcription Kit. Briefly, genomic DNA was removed by DNase digestion by incubation with gDNAWipeout Buffer. First- A total of 50 HIV-1–infected subjects and 10 HIV-1–negative subjects m included in the plasma CXCL13 determination studies were enrolled in strand cDNA synthesis was carried out for 30 min at 42˚C in 20 lsolution protocols at the Massachusetts General Hospital (Boston, MA). Additional containing Quantiscript Reverse Transcriptase, Quantiscript RT Buffer, and RT Primer Mix (QIAGEN), followed by denaturation for 3 min at 95˚C. HIV-1–negative subjects were enrolled in a control cohort at Seattle Bio- m medical Research Institute, and HIV-1–infected adult subjects were enrolled Then, 4 l cDNA templates were used for real-time PCR reactions to quantify CXCL13 (assay number Hs00757930_m1) in a 96-well plate. Each in either the Seattle HIV Vaccine Clinic Long-Term Non-Progressor or http://www.jimmunol.org/ reaction was carried out in 20 ml solution containing 10 mL TaqMan reaction Natural Progression cohorts at the Fred Hutchinson Cancer Research Center m (Seattle, WA). The latter subjects were chronically HIV-1 infected with mix and 1 l TaqMan FAM dye–labeled MGB probe using a 7500 Fast Real- Time PCR machine (Applied Biosystems). Fold change of CXCL13 a median viral load of 1773 RNA copies per milliliter (range: 320–9999) and ΔΔ aCD4+ T cell count of 483 CD4+ T cells per cubic millimeter (range: 261– expression was calculated by the CT method using GAPDH (assay 1330) at the time of the blood draw, and were not receiving ART. The rel- number Hs02758991_g1) as an internal control and compared with unsti- evant institutional review boards approved all human subject protocols, and mulated controls. all subjects provided written informed consent before enrollment. PBMCs Statistical analysis were isolated within 4 h of venipuncture by density-gradient centrifugation. In vitro stimulation experiments were performed on fresh PBMCs from HIV- Data are presented as the medians of a minimum of two independent negative and HIV-1–infected subjects. Plasma samples were aliquoted and experiments carried out using PBMCs from different donors, unless stated frozen at 280˚C. Concentrations of CXCL13 in thawed plasma samples were otherwise. Two-tailed Mann–Whitney U tests, Wilcox-paired nonparametric by guest on October 2, 2021 determined by ELISA according to the manufacturer’s instructions (CXCL13 tests, and Spearman’s correlations were calculated using GraphPad Quantikine ELISA Kit; R&D systems). Prism 5.02 software (GraphPad Software). The p values , 0.05 were considered significant. Intracellular cytokine staining For stimulation with HIV-1 virions, PBMCs were isolated from HIV- Results negative donors and resuspended at 2 3 106/ml in RPMI 1640 supple- HIV-1 induces CXCL13 production mented with 10% heat-inactivated FCS, 2 mM L-glutamine, 100 U/ml penicillin, and 100 U/ml streptomycin (cRPMI). HIV-1 YU2 infectious To confirm the relationship between HIV-1 viremia and induction molecular clones were added to the PBMCs at 50 ng/ml P24 for 18 h at 37∘C of CXCL13, we compared levels of CXCL13 and viral load in the and 5% CO2. Brefeldin A (Sigma-Aldrich) was added at a final concentra- plasma of chronically HIV-1–infected subjects. We found a signif- tion of 5 mg/ml and incubated for an additional 18 h. Samples were stained icant correlation between viral load and CXCL13 plasma concen- for extracellular markers, fixed, permeabilized, and stained for FACS anal- ysis following the manufacturer’s recommendations (Becton Dickinson trations of untreated chronically HIV-1–infected subjects (Fig. 1A; Cytofix/Cytoperm Kit). Samples were surface stained with the following Abs r =0.55,p = 0.006), in agreement with previous reports (21–23). to distinguish cell subsets: CD11c (Pe-Cy5), CD14 (FITC), HLA-DR Similarly, we found that subjects with suppressed viral load by (Allophycocyanin-Cy7), CD3 (Qdot 800), CD19 (Qdot 605), and CD123 ART had significantly lower levels of plasma CXCL13 (Fig. 1B), (Brilliant Violet). Samples were stained intracellularly for CXCL13 (Allo- suggesting that viral replication drives the production of CXCL13 phycocyanin; R&D Systems). Cell subsets were defined as follows: monocytes (live/dead aqua2,CD32, and not the reverse. Overall, the CXCL13 levels following ART in CD192,CD562, HLA-DR+,CD11c+, CD1232,andCD14+), myeloid DCs HIV-1–infected subjects were higher than those observed in HIV- (mDCs) (live/dead aqua2,CD32,CD192,CD562, HLA-DR+, CD11c+, 2 2 2 2 2 2 1–negative control subjects. To determine whether HIV-1 induces CD123 ,andCD14 ), and pDCs (live/dead aqua ,CD3 ,CD19 ,CD56 , CXCL13 production directly and to define potential mechanism(s) HLA-DR+,CD11c+,CD123+,andCD142). They were acquired using a Becton Dickinson LSR II flow cytometer and analyzed using FlowJo by which viremia contributes to elevated CXCL13 expression, we (TreeStar). Abs were from Becton Dickinson unless otherwise stated. incubated PBMCs from uninfected donors with HIV-1 replication– competent virions and monitored for increases in CXCL13 by in- Isolation of cell subsets tracellular cytokine staining in different cell subsets. Indeed, we The pDCs were depleted from human PBMCs using a CD123 Microbead observed that CXCL13 expression was significantly increased in Kit (Miltenyi Biotec). Because CD123 is also expressed on basophils, pDCs monocytes (Fig. 1C, 1D), but not in other cell subsets, such as were positively selected using CD304 (BDCA-4/Neuropilin-1) magnetic mDCs, pDCs (Fig. 1C, 1D), or T and B cells (data not shown). bead isolation (Miltenyi Biotec). Monocytes were isolated from PBMCs using a monocyte negative selection enrichment kit and were depleted using Monocytes produce CXCL13 through TLR7/8 activation a CD14 positive selection kit (Stem Cell Technologies). All PBMCs and cell subsets were cultured in complete RPMI media as described above. Cell As mentioned above, HIV-1–derived ssRNA is known to activate purity was consistently .95%; it was confirmed by surface staining with TLR7/8 pathways (26, 31). Because monocytes produce CXCL13 The Journal of Immunology 2771 Downloaded from

FIGURE 1. CXCL13 plasma levels in chronically HIV-1–infected subjects. (A) A correlation of CXCL13 plasma levels in chronically HIV-1–infected untreated subjects with contemporaneous plasma viremia. Correlation coefficient and significance were calculated using a Spearman rank-order correlation

for nonparametric data. (B) CXCL13 concentrations were measured in the plasma of HIV-1–infected subjects: 20 chronic untreated (CU) and 30 chronic http://www.jimmunol.org/ ART-treated (CT), compared with 10 HIV-1–negative control subjects (HIV-NEG). The horizontal lines are at the median. The stated p values were calculated using the Mann–Whitney U test for nonparametric data and considered significant if *p , 0.05, **p , 0.01, or ***p , 0.0001. PBMCs from four HIV-negative donors were stimulated with HIV-1 YU2 virions for 36 h and evaluated by intracellular staining by flow cytometry for CXCL13. (C) Histograms depict the expression of CXCL13 in different cell populations after stimulation with HIV-1 virions, for a representative sample. (D) A summary of the expression of CXCL13 in monocytes, mDCs, and pDCs in PBMCs after stimulation with HIV-1 virions for HIV-1–uninfected subjects (n = 4). Significance was determined by the Kruskal–Wallis test, *p , 0.05. upon stimulation of the TLR 2 and 4 pathways (17, 29, 30), and be- results were obtained after 2 d of PBMC or monocyte stimulation by guest on October 2, 2021 cause TLRs 2, 4, 7, and 8 share overlapping signaling pathways and with ssRNA or R848. A time course experiment indicated that a downstream transcription factors (27), we hypothesized that HIV-1 significant induction of CXCL13 production by PBMCs was ob- infection of monocytes may induce CXCL13 through TLR7 and/or served as early as 1 d post stimulation (p = 0.03) that peaked at day 3 TLR8 activation. To investigate this possibility, we treated PBMCs (Fig. 2D). The isolated monocytes secreted significantly less from HIV-negative subjects with ssRNA derived from the long ter- CXCL13 than the stimulated PBMCs, and the levels of CXCL13 minal repeat of HIV-1 (ssRNA40) at different concentrations and detected in the stimulated monocytes were not significant compared measured CXCL13 production after 2 d of stimulation. We observed with the unstimulated monocytes. In addition, CXCL13 secretion that CXCL13 was detectable in response to stimulation with HIV-1 was almost completely abrogated in the monocyte-depleted PBMCs ssRNA, in a dose-dependent manner (Fig. 2A). In contrast, CXCL13 under these stimulatory conditions (Fig. 2D). The above results led secretion was not observed with a control ssRNA in which uridines us to conclude that monocytes seem to be necessary, but not suffi- have been replaced with adenosines (ssRNA41) (Fig. 2A). However, cient, to promote the secretion of high concentrations of CXCL13 in it has been shown that infection with HIV-1 modulates the immune response to TLR7/8 stimulation in human PBMCs. We envisioned response to TLR stimulation. Therefore, to determine whether TLR7/ the two most likely possibilities to explain these results: 1) mono- 8 induction of CXCL13 secretion is relevant in the context of HIV-1 cytes are the major producers of CXCL13, but additional cell types, infection, we stimulated PBMCs from untreated HIV-1–infected depleted when monocytes were enriched, are necessary for optimal subjects with HIV-1–derived ssRNA, control ssRNA, or the TLR7/8 CXCL13 production by monocytes and 2) monocytes are not the agonist R848 (Fig. 2B). We also stimulated PBMCs in the presence major producers of CXCL13 but are critical for CXCL13 production of the antiretroviral drug azidothymidine (AZT), to prevent endog- by another cell subset in PBMCs. To determine which of these two enous HIV-1 replication in vitro. Addition of either ssRNA or R848 possibilities was responsible for the above observations, we stimu- resulted in significant CXCL13 secretion (p = 0.008). The presence lated PBMCs from HIV-negative donors with R848 for 2 d and of AZT had no discernible effect on CXCL13 secretion, an indication subsequently isolated the monocytes. Then, we compared the levels that the observed increase in CXCL13 secretion was primarily due of CXCL13 mRNA expression in the monocytes and monocyte- to the addition of exogenous ssRNA. depleted PBMCs by real-time PCR. We observed significant up- To define the role of monocytes in the observed induction of regulation of CXCL13 mRNA expression in the monocytes, but not CXCL13 via TLR7/8 activation, we stimulated isolated monocytes in the monocyte-depleted PBMCs (Fig. 2E). Together, these results from HIV-1–negative subjects with HIV-1 ssRNA or with the TLR7/ confirmed that 1) monocytes were the primary cells secreting 8 agonist R848. Of interest, although TLR7/8 stimulation of purified CXCL13 under TLR7/8 stimulation of PBMCs, 2) TLR7/8 stim- monocytes resulted in CXCL13 secretion, the amounts of CXCL13 ulation of monocytes is sufficient for the production of CXCL13, produced were significantly lower than those observed when the and 3) an intermediary cell type in PBMCs exists that is required corresponding PBMCs were similarly stimulated (Fig. 2C). These for optimal CXCL13 secretion by monocytes. 2772 HIV-1 INDUCTION OF CXCL13 IN MONOCYTES Downloaded from

FIGURE 2. Monocytes secrete CXCL13 following TLR7/8 stimulation of PBMCs. (A) PBMCs isolated from HIV-negative subjects were stimulated with

increasing doses of HIV-1 ssRNA or with a negative control ssRNA. CXCL13 levels in the supernatants were determined after 2 d of stimulation. The lines http://www.jimmunol.org/ represent linear regression curves. (B) PBMCs isolated from HIV-1–infected subjects were stimulated with ssRNA derived from HIV-1 or control ssRNA (1 mg/ml) or the TLR7/8 agonist R848 (1 mg/ml) in the presence or absence of AZT (5 mM). CXCL13 levels in the supernatants were determined 2 d post stimulation. **p , 0.01. (C) PBMCs or isolated monocytes from HIV-negative subjects were stimulated with ssRNA derived from HIV-1 or control ssRNA at indicated concentrations. CXCL13 levels in the supernatants were determined after 2 d post stimulation. **p , 0.05. (D) PBMCs, monocytes, and monocyte- depleted PBMCs from HIV-1–negative subjects were stimulated with R848. CXCL13 levels in the supernatants were determined by ELISA at the indicated time points. The horizontal lines are at the median. The p values were calculated using the Mann–Whitney U test for nonparametric data. *p , 0.05. (E)PBMCs from HIV-negative subjects were stimulated with R848 for 2 d. Monocytes were isolated from the PBMCs, RNA was extracted from both the monocytes and the monocyte-depleted PBMCs, and CXCL13 transcripts were measured by real-time PCR. Fold change was calculated by ΔΔCT using GAPDH as internal control and compared with unstimulated controls. Values represent the mean of triplicates and SEM. The stated p values were calculated by paired t test, *p , 0.05. ns, not significant. by guest on October 2, 2021

Type I IFN is necessary and sufficient for CXCL13 secretion by these cells, we isolated pDCs and monocytes, recombined them, To identify the intermediate cell subtype necessary for the optimal and stimulated them with a TLR7/8 agonist (R848); after 2 d we production of CXCL13 by monocytes upon activation, we performed measured CXCL13 in the supernatant (Fig. 3E). The recombined targeted cell depletions from PBMCs. We focused on pDCs because pDCs and monocytes recapitulated the level of CXCL13 detected in pDCs express TLR7 and are known to secrete significant quantities the PBMC response to TLR7/8 stimulation (Fig. 2B, Fig. 3E). of —IFN-a in particular—in response to TLR7 stimula- tion (25, 32–34). A significant decrease in CXCL13 production in Discussion PBMC cultures was observed when pDCs were depleted from the The chemokine CXCL13 is the ligand for the receptor CXCR5, PBMCs (Fig. 3A). To determine whether pDC-produced IFN-I was expressed on B and Tfh cells, and is critical for follicle development, involved in the production of CXCL13 by monocytes, we stimu- affinity maturation of B cells, and organization of secondary lym- lated PBMCs with TLR7/8 agonists after depleting pDCs or in the phoid architecture. Although follicular DCs have historically been presence of the soluble IFN-I receptor B18R, which neutralizes considered the main producers of CXCL13, monocytes, macro- IFN-I. Under these conditions, a significant reduction of CXCL13 phages, and mDCs have also been shown to secrete CXCL13 in production was observed (Fig. 3A). response to bacterial-derived TLR2/4 ligands (17, 19, 20, 28–30). To determine whether IFN-a alone was sufficient to induce Previous reports demonstrated increases in plasma levels of CXCL13 production, we stimulated PBMCs, pDC-depleted PBMCs, CXCL13 during HIV-1 infection and a positive correlation between or purified monocytes from HIV-negative subjects with exogenous the plasma CXCL13 concentration and plasma viremia (21–24). In recombinant IFN-a. In all cases, we observed that CXCL13 ex- chronic HIV-1 infection, a correlation has been observed between pression was induced in an IFN-a dose–dependent manner (Fig. 3B). plasma CXCL13 concentrations and changes in the chemotactic IFN-a–mediatedCXCL13expressionwasabrogatedwhenthesol- potential of B cells (22), with a loss of circulating Tfh cells (24). uble IFNR B18R was added to the cell culture during stimulation However, the mechanistic connections between HIV-1 viremia and (Fig. 3B). Furthermore, similar levels of CXCL13 were secreted CXCL13 production are unknown. from PBMCs stimulated with the TLR7/8 agonist R848 or with In this study, we confirmed the correlation between plasma HIV-1 the TLR7 agonist R837 (Fig. 3B). This finding suggests that TLR8 viremia and CXCL13 plasma concentrations in untreated chronic activation is most likely dispensable for maximal secretion of HIV-1 infection and identified a novel mechanism of HIV-induced CXCL13. To further confirm the role of pDC-secreted IFN-a in the CXCL13 secretion. We report that HIV-1 induces potent mono- stimulation of monocytes and the subsequent secretion of CXCL13 cyte production of CXCL13 and that this occurs through TLR7- The Journal of Immunology 2773 Downloaded from http://www.jimmunol.org/

FIGURE 3. Type I IFN is required for maximal induction of CXCL13 in monocytes. (A) PBMCs and pDC-depleted PBMCs from HIV-1–negative subjects (the same subjects from Fig. 2D) were stimulated for 4 d with either media alone or R848 in the presence or absence of the IFN-I–neutralizing B18R. CXCL13 levels in the supernatants were determined by ELISA at the indicated time points. The horizontal lines are the medians. The p values were calculated using the Mann–Whitney U test for nonparametric data. *p , 0.05. (B) PBMCs, pDC-depleted PBMCs, monocyte-depleted PBMCs, and isolated monocytes from HIV-negative subjects were stimulated with the TLR agonists R848 and R837, or the indicated amounts of recombinant IFN- a2A, and CXCL13 concentrations in the cell supernatants were determined. (C) IFN-a concentrations were determined and (D) compared with the levels of CXCL13 detected in the supernatants of the unstimulated, TLR-, or IFN-stimulated cultures of PBMCs or pDC-depleted PBMCs from HIV-negative subjects. Spearman’s rank correlation was used to determine the significance of the correlation. The line is a linear regression curve. (E) PBMC-isolated monocytes combined with or without purified autologous pDCs were stimulated for 2 d with R848, and the CXCL13 levels were determined by ELISA. The by guest on October 2, 2021 horizontal lines are at the median. induced IFN-I secretion from pDCs and subsequent IFN-I stimula- with increased B cell (22). In addition, the fact that tion of monocytes. Specifically, we identified HIV-1–derived ssRNA, CXCL13 expression can be upregulated in peripheral monocytes which is a TLR7/8 agonist, as a potent inducer of CXCL13 secretion by IFN-I activation may indicate a similar mechanism of CXCL13 from PBMCs. We note that although we established that monocytes expression in lymphatic tissues where substantial HIV-1 replica- were the main producers of CXCL13 in PBMCs, under the stimu- tion occurs (35). Although we did not have access to lymph node latory conditions used in this study, and that maximum CXCL13 samples from HIV-infected subjects, in previous studies of lymph secretion was dependent on pDC-secreted IFN-I, it is possible that node biopsies from HIV-1–infected subjects, the majority of low CXCL13 amounts are secreted by other cell types. In addition, CXCL13+ cells were detected in the T-dependent zone and colo- the promoter region of CXCL13 contains the binding site of the calized with CD68+ monocytes or macrophages and CD1a+ im- transcription factor ISGF-3, which is downstream of IFN-I, further mature mDCs (22). pDCs have been shown to accumulate in the corroborating the potential induction of CXCL13 secretion by IFN lymph nodes of HIV-infected subjects and to secrete IFN-a (36, signaling. IFN-a was sufficient to induce high levels of CXCL13 37). Likewise, IFN-I–inducible have been reported to be secretion from isolated monocytes. This observation indicates that increased in the lymph nodes of SIV-infected macaques and HIV- direct contact between pDCs and monocytes upon TLR7/8 stimu- 1–infected (36, 38–40), and several studies have addressed lation of these two cell types is probably not required for secretion the potential contribution of TLR7-mediated activation of pDCs of CXCL13 by the monocytes. Future experiments will be necessary and induction of IFN-I to chronic immune activation, immune cell to better define to what extent direct contact between monocytes and dysfunction, disruption of lymphoid architecture, and reduced de pDCs contributes to the observed CXCL13 secretion. Therefore, the novo Ab responses (31, 34, 41–45). Furthermore, the accumulation IFN-I–dependent TLR7/8-mediated upregulation of CXCL13 in of Tfh cells and memory B cells, which express CXCR5 and mi- monocytes may explain the elevated levels of CXCL13 found in the grate in response to CXCL13, has been described in the lymph periphery during chronic HIV-1 infection. Finally, we were able to nodes of chronically HIV-1–infected humans and some SIV- confirm that the HIV-1–derived ssRNAs and the TLR7/8 agonist infected macaques (40, 46–48). The accumulation of Tfh cells in R848 were still able to induce CXCL13 secretion in PBMCs from lymph nodes during chronic HIV-1 infection has also been associ- untreated chronically HIV-1–infected subjects. ated with skewing of the B cell compartment and hyper- Previous studies have already demonstrated that increased gammaglobulinemia (46, 49). Taken together, our results suggest CXCL13 during chronic HIV-1 infection is associated with reduced that HIV-1–induced IFN-I–dependent CXCL13 secretion may expression of the cognate receptor CXCR5 on mature B cells and contribute to the observed disruption of the lymphoid architecture 2774 HIV-1 INDUCTION OF CXCL13 IN MONOCYTES and dysregulation of humoral responses during untreated chronic 23. Regidor, D. L., R. Detels, E. C. Breen, D. P. Widney, L. P. Jacobson, F. Palella, C. R. Rinaldo, J. H. Bream, and O. Martı´nez-Maza. 2011. Effect of highly active HIV-1infectionbyincreasingmigration and accumulation of antiretroviral therapy on biomarkers of B-lymphocyte activation and inflam- mature B and Tfh cells in lymphoid organs. mation. AIDS 25: 303–314. 24. Boswell, K. L., R. Paris, E. Boritz, D. Ambrozak, T. Yamamoto, S. Darko, K. Wloka, A. Wheatley, S. Narpala, A. McDermott, et al. 2014. Loss of circu- Disclosures lating CD4 T cells with B cell helper function during chronic HIV infection. PLoS Pathog. 10: e1003853. The authors have no financial conflicts of interest. 25. Diebold, S. S., T. Kaisho, H. Hemmi, S. Akira, and C. Reis e Sousa. 2004. Innate antiviral responses by means of TLR7-mediated recognition of single-stranded RNA. Science 303: 1529–1531. References 26. Heil, F., H. Hemmi, H. Hochrein, F. Ampenberger, C. Kirschning, S. Akira, G. Lipford, H. Wagner, and S. Bauer. 2004. Species-specific recognition of 1. Amu, S., N. Ruffin, B. Rethi, and F. Chiodi. 2013. Impairment of B-cell func- single-stranded RNA via toll-like receptor 7 and 8. Science 303: 1526–1529. tions during HIV-1 infection. AIDS 27: 2323–2334. 27. Kawai, T., and S. Akira. 2011. Toll-like receptors and their crosstalk with other 2. Moir, S., and A. S. Fauci. 2013. Insights into B cells and HIV-specific B-cell innate receptors in infection and immunity. Immunity 34: 637–650. responses in HIV-infected individuals. Immunol. Rev. 254: 207–224. 28. Perrier, P., F. O. Martinez, M. Locati, G. Bianchi, M. Nebuloni, G. Vago, 3. Ho, J., S. Moir, A. Malaspina, M. L. Howell, W. Wang, A. C. DiPoto, F. Bazzoni, S. Sozzani, P. Allavena, and A. Mantovani. 2004. Distinct tran- M. A. O’Shea, G. A. Roby, R. Kwan, J. M. Mican, et al. 2006. Two overrep- scriptional programs activated by -10 with or without lipopolysac- resented B cell populations in HIV-infected individuals undergo apoptosis by charide in dendritic cells: induction of the B cell-activating chemokine, CXC different mechanisms. Proc. Natl. Acad. Sci. USA 103: 19436–19441. chemokine ligand 13. J. Immunol. 172: 7031–7042. 4. De Milito, A., A. Nilsson, K. Titanji, R. Thorstensson, E. Reizenstein, M. Narita, 29. Rupprecht, T. A., C. J. Kirschning, B. Popp, S. Kastenbauer, V. Fingerle, S. Grutzmeier, A. So¨nnerborg, and F. Chiodi. 2004. Mechanisms of hyper- H. W. Pfister, and U. Koedel. 2007. Borrelia garinii induces CXCL13 production gammaglobulinemia and impaired antigen-specific humoral immunity in HIV-1 in human monocytes through Toll-like receptor 2. Infect. Immun. 75: 4351–4356. infection. Blood 103: 2180–2186. 30. Moreth, K., R. Brodbeck, A. Babelova, N. Gretz, T. Spieker, J. Zeng-Brouwers, 5. Malaspina, A., S. Moir, S. M. Orsega, J. Vasquez, N. J. Miller, E. T. Donoghue, J. Pfeilschifter, M. F. Young, R. M. Schaefer, and L. Schaefer. 2010. The pro- S. Kottilil, M. Gezmu, D. Follmann, G. M. Vodeiko, et al. 2005. Compromised Downloaded from teoglycan biglycan regulates expression of the B cell chemoattractant CXCL13 B cell responses to influenza vaccination in HIV-infected individuals. J. Infect. and aggravates murine lupus nephritis. J. Clin. Invest. 120: 4251–4272. Dis. 191: 1442–1450. 31. Alter, G., T. J. Suscovich, N. Teigen, A. Meier, H. Streeck, C. Brander, and 6. Moir, S., A. Malaspina, J. Ho, W. Wang, A. C. Dipoto, M. A. O’Shea, G. Roby, M. Altfeld. 2007. Single-stranded RNA derived from HIV-1 serves as a potent J. M. Mican, S. Kottilil, T. W. Chun, et al. 2008. Normalization of B cell counts activator of NK cells. J. Immunol. 178: 7658–7666. and subpopulations after antiretroviral therapy in chronic HIV disease. J. Infect. 32. Beignon, A. S., K. McKenna, M. Skoberne, O. Manches, I. DaSilva, Dis. 197: 572–579. D. G. Kavanagh, M. Larsson, R. J. Gorelick, J. D. Lifson, and N. Bhardwaj. 7. Sciaranghella, G., N. Tong, A. E. Mahan, T. J. Suscovich, and G. Alter. 2013. 2005. Endocytosis of HIV-1 activates plasmacytoid dendritic cells via Toll-like

Decoupling activation and exhaustion of B cells in spontaneous controllers of http://www.jimmunol.org/ receptor–viral RNA interactions. J. Clin. Invest. 115: 3265–3275. HIV infection. AIDS 27: 175–180. 33. Gibson, S. J., J. M. Lindh, T. R. Riter, R. M. Gleason, L. M. Rogers, A. E. Fuller, 8. Pensieroso, S., L. Galli, S. Nozza, N. Ruffin, A. Castagna, G. Tambussi, J. L. Oesterich, K. B. Gorden, X. Qiu, S. W. McKane, et al. 2002. Plasmacytoid B. Hejdeman, D. Misciagna, A. Riva, M. Malnati, et al. 2013. B-cell dendritic cells produce cytokines and mature in response to the TLR7 agonists, subset alterations and correlated factors in HIV-1 infection. AIDS 27: 1209–1217. imiquimod and resiquimod. Cell. Immunol. 218: 74–86. 9. Legler, D. F., M. Loetscher, R. S. Roos, I. Clark-Lewis, M. Baggiolini, and 34. Simmons, R. P., E. P. Scully, E. E. Groden, K. B. Arnold, J. J. Chang, K. Lane, B. Moser. 1998. B cell-attracting chemokine 1, a human CXC chemokine expressed in lymphoid tissues, selectively attracts B lymphocytes via BLR1/ J. Lifson, E. Rosenberg, D. A. Lauffenburger, and M. Altfeld. 2013. HIV-1 in- CXCR5. J. Exp. Med. 187: 655–660. fection induces strong production of IP-10 through TLR7/9-dependent pathways. 10. Ngo, V. N., H. Korner, M. D. Gunn, K. N. Schmidt, D. S. Riminton, AIDS 27: 2505–2517 M. D. Cooper, J. L. Browning, J. D. Sedgwick, and J. G. Cyster. 1999. Lym- 35. Pantaleo, G., C. Graziosi, J. F. Demarest, L. Butini, M. Montroni, C. H. Fox, photoxin alpha/beta and are required for stromal cell ex- J. M. Orenstein, D. P. Kotler, and A. S. Fauci. 1993. HIV infection is active and progressive in lymphoid tissue during the clinically latent stage of disease.

pression of homing in B and T cell areas of the . J. Exp. Med. by guest on October 2, 2021 189: 403–412. Nature 362: 355–358. 11. Gunn, M. D., V. N. Ngo, K. M. Ansel, E. H. Ekland, J. G. Cyster, and 36. Lehmann, C., M. Lafferty, A. Garzino-Demo, N. Jung, P. Hartmann, L. T. Williams. 1998. A B-cell-homing chemokine made in lymphoid follicles G. Fa¨tkenheuer, J. S. Wolf, J. van Lunzen, and F. Romerio. 2010. Plasmacytoid activates Burkitt’s receptor-1. Nature 391: 799–803. dendritic cells accumulate and secrete alpha in lymph nodes of HIV-1 12. Fo¨rster, R., T. Emrich, E. Kremmer, and M. Lipp. 1994. Expression of the patients. PLoS ONE 5: e11110. G-protein–coupled receptor BLR1 defines mature, recirculating B cells and 37. O’Brien, M., O. Manches, R. L. Sabado, S. J. Baranda, Y. Wang, I. Marie, a subset of T-helper memory cells. Blood 84: 830–840. L. Rolnitzky, M. Markowitz, D. M. Margolis, D. Levy, and N. Bhardwaj. 2011. 13. Schaerli, P., K. Willimann, A. B. Lang, M. Lipp, P. Loetscher, and B. Moser. Spatiotemporal trafficking of HIV in human plasmacytoid dendritic cells defines a 2000. CXC 5 expression defines follicular homing T cells a persistently IFN- -producing and partially matured phenotype. J. Clin. Invest. with B cell helper function. J. Exp. Med. 192: 1553–1562. 121: 1088–1101. 14. Breitfeld, D., L. Ohl, E. Kremmer, J. Ellwart, F. Sallusto, M. Lipp, and 38. Sanghavi, S. K., and T. A. Reinhart. 2005. Increased expression of TLR3 in R. Fo¨rster. 2000. Follicular B helper T cells express CXC chemokine receptor 5, lymph nodes during simian immunodeficiency virus infection: implications for localize to B cell follicles, and support immunoglobulin production. J. Exp. Med. inflammation and immunodeficiency. J. Immunol. 175: 5314–5323. 192: 1545–1552. 39. Malleret, B., B. Mane´glier, I. Karlsson, P. Lebon, M. Nascimbeni, L. Perie´, 15. Allen, C. D., K. M. Ansel, C. Low, R. Lesley, H. Tamamura, N. Fujii, and P. Brochard, B. Delache, J. Calvo, T. Andrieu, et al. 2008. Primary infection with J. G. Cyster. 2004. Germinal center dark and zone organization is mediated simian immunodeficiency virus: plasmacytoid dendritic cell homing to lymph by CXCR4 and CXCR5. Nat. Immunol. 5: 943–952. nodes, type I interferon, and immune suppression. Blood 112: 4598–4608. 16. Jacob, J., G. Kelsoe, K. Rajewsky, and U. Weiss. 1991. Intraclonal generation of 40. Petrovas, C., T. Yamamoto, M. Y. Gerner, K. L. Boswell, K. Wloka, E. C. Smith, antibody mutants in germinal centres. Nature 354: 389–392. D. R. Ambrozak, N. G. Sandler, K. J. Timmer, X. Sun, et al. 2012. CD4 T 17. Carlsen, H. S., E. S. Baekkevold, H. C. Morton, G. Haraldsen, and P. Brandtzaeg. follicular helper cell dynamics during SIV infection. J. Clin. Invest. 122: 3281– 2004. Monocyte-like and mature macrophages produce CXCL13 (B cell- 3294. attracting chemokine 1) in inflammatory lesions with lymphoid neogenesis. 41. Hardy, G. A., S. Sieg, B. Rodriguez, D. Anthony, R. Asaad, W. Jiang, J. Mudd, Blood 104: 3021–3027. T. Schacker, N. T. Funderburg, H. A. Pilch-Cooper, et al. 2013. Interferon-a is 18. Wong, C. K., P. T. Wong, L. S. Tam, E. K. Li, D. P. Chen, and C. W. Lam. 2010. the primary plasma type-I IFN in HIV-1 infection and correlates with immune Elevated production of B cell chemokine CXCL13 is correlated with systemic activation and disease markers. PLoS ONE 8: e56527. lupus erythematosus disease activity. J. Clin. Immunol. 30: 45–52. 42. Meier, A., G. Alter, N. Frahm, H. Sidhu, B. Li, A. Bagchi, N. Teigen, H. Streeck, 19. Shomer, N. H., J. G. Fox, A. E. Juedes, and N. H. Ruddle. 2003. Helicobacter- H. J. Stellbrink, J. Hellman, et al. 2007. MyD88-dependent immune activation induced chronic active lymphoid aggregates have characteristics of tertiary mediated by human immunodeficiency virus type 1-encoded Toll-like receptor lymphoid tissue. Infect. Immun. 71: 3572–3577. ligands. J. Virol. 81: 8180–8191. 20. Vermi, W., F. Facchetti, E. Riboldi, H. Heine, S. Scutera, S. Stornello, 43. Meier, A., A. Bagchi, H. K. Sidhu, G. Alter, T. J. Suscovich, D. G. Kavanagh, D. Ravarino, P. Cappello, M. Giovarelli, R. Badolato, et al. 2006. Role of den- H. Streeck, M. A. Brockman, S. LeGall, J. Hellman, and M. Altfeld. 2008. dritic cell-derived CXCL13 in the pathogenesis of Bartonella henselae B-rich Upregulation of PD- on monocytes and dendritic cells by HIV-1 derived TLR granuloma. Blood 107: 454–462. ligands. AIDS 22: 655–658. 21. Widney, D. P., E. C. Breen, W. J. Boscardin, S. G. Kitchen, J. M. Alcantar, 44. Chang, J. J., M. Woods, R. J. Lindsay, E. H. Doyle, M. Griesbeck, E. S. Chan, J. B. Smith, J. A. Zack, R. Detels, and O. Martinez-Maza. 2005. Serum levels of G. K. Robbins, R. J. Bosch, and M. Altfeld. 2013. Higher expression of several the homeostatic B cell chemokine, CXCL13, are elevated during HIV infection. interferon-stimulated genes in HIV-1-infected females after adjusting for the J. Interferon Cytokine Res. 25: 702–706. level of viral replication. J. Infect. Dis. 208: 830–838. 22. Cagigi, A., F. Mowafi, L. V. Phuong Dang, K. Tenner-Racz, A. Atlas, 45. Baenziger, S., M. Heikenwalder, P. Johansen, E. Schlaepfer, U. Hofer, S. Grutzmeier, P. Racz, F. Chiodi, and A. Nilsson. 2008. Altered expression of R. C. Miller, S. Diemand, K. Honda, T. M. Kundig, A. Aguzzi, and R. F. Speck. the receptor-ligand pair CXCR5/CXCL13 in B cells during chronic HIV-1 in- 2009. Triggering TLR7 in mice induces immune activation and lymphoid system fection. Blood 112: 4401–4410. disruption, resembling HIV-mediated pathology. Blood 113: 377–388. The Journal of Immunology 2775

46. Lindqvist, M., J. van Lunzen, D. Z. Soghoian, B. D. Kuhl, S. Ranasinghe, 48. Xu, Y., C. Weatherall, M. Bailey, S. Alcantara, R. De Rose, J. Estaquier, G. Kranias, M. D. Flanders, S. Cutler, N. Yudanin, M. I. Muller, et al. 2012. K. Wilson, K. Suzuki, J. Corbeil, D. A. Cooper, et al. 2013. Simian immuno- Expansion of HIV-specific T follicular helper cells in chronic HIV infection. J. deficiency virus infects follicular helper CD4 T cells in lymphoid tissues during Clin. Invest. 122: 3271–3280. pathogenic infection of pigtail macaques. J. Virol. 87: 3760–3773. 47. Perreau, M., A. L. Savoye, E. De Crignis, J. M. Corpataux, R. Cubas, 49. Hong, J. J., P. K. Amancha, K. Rogers, A. A. Ansari, and F. Villinger. 2012. E. K. Haddad, L. De Leval, C. Graziosi, and G. Pantaleo. 2013. Follicular helper Spatial alterations between CD4(+) T follicular helper, B, and CD8(+) T cells T cells serve as the major CD4 T cell compartment for HIV-1 infection, repli- during simian immunodeficiency virus infection: T/B cell homeostasis, activa- cation, and production. J. Exp. Med. 210: 143–156. tion, and potential mechanism for viral escape. J. Immunol. 188: 3247–3256. Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021