Lysophosphatidic Acid (LPA) Receptor 5 Inhibits B Cell Antigen Receptor Signaling and Antibody Response

Abstract

Lysophospholipids have emerged as biologically important chemoattractants capable of directing lymphocyte development, trafficking and localization. Lysophosphatidic acid (LPA) is a major lysophospholipid found systemically and whose levels are elevated in certain pathological settings such as cancer and infections. Here, we demonstrate that BCR signal transduction by mature murine B cells is inhibited upon LPA engagement of the LPA₅ (GPR92) receptor via a Gα12/13 – Arhgef1 pathway. The inhibition of BCR signaling by LPA₅ manifests by impaired intracellular calcium store release and most likely by interfering with inositol 1,4,5-trisphosphate receptor activity. We further show that LPA₅ also limits antigen-specific induction of CD69 and CD86 expression and that LPA₅-deficient B cells display enhanced antibody responses. Thus, these data show that LPA₅ negatively regulates BCR signaling, B cell activation and immune response. Our findings extend the influence of lysophospholipids on immune function and suggest that alterations in LPA levels likely influence adaptive humoral immunity.

INTRODUCTION

Signals transmitted by the B cell antigen receptor are not only required for the antibody response but also for the development and survival of B lymphocytes (1, 2). BCR signaling by mature B cells can be either positively or negatively regulated by additional surface co-receptors depending on the developmental stage of the B cell, the nature of antigen and the microenvironment where the antigen is encountered (3, 4). For example, BCR signaling is more effective when complement-decorated antigen simultaneously engages both the BCR and CD21/CD19 co-receptor complex as compared to BCR signaling alone (5). In contrast, a B cell encountering antigen bound by IgG-signals simultaneously via both the BCR and FcγRIIB; signals transmitted by FcγRIIB dampen BCR signaling, thus attenuating the antibody response (3, 6).

In addition to these established BCR co-receptors, lymphocytes also express G protein-coupled receptors (GPCR) for chemokines and lysophospholipids and both these ligands have been reported to modify lymphocyte antigen receptor signaling (7, 8). GPCRs signal primarily via associated αβγ heterotrimeric G-proteins and a pertussis toxin-insensitive Gα subunit has long been known to regulate BCR signaling (9, 10). However, neither the identity of this Gα protein nor the mechanism(s) by which any of these GPCR regulate antigen receptor signaling in B lymphocytes has been established.

Sphingosine-1-phosphate (S1P) and lysophosphatidic acid (LPA) are biologically active serum lysophospholipids that can signal extracellularly by engaging cognate GPCR expressed by diverse cell types (11). Notably, S1P has emerged as an important chemoattractant that guides leukocytes during development, homeostasis and inflammation (12-16). LPA is another major lysophospholipid that can signal via six established LPA GPCR, LPA_1-6, each capable of associating with members from distinct Gα families (17, 18). These LPA GPCR are encoded by two gene clusters that includes LPA_1-2-3 in the endothelial differentiation gene family, whereas LPA_4-5-6 are closely related to the purinergic GPCR subfamily (18).

The plasma and tissue concentration of LPA is in the hundred nanomolar range whereas, in serum, inflammatory exudates or tumor cell effusates it can reach as high as 10 μM (11, 17, 19, 20). The biological function of LPA has received most attention with respect to cancer where it has been shown to promote cell migration, proliferation and survival of a number of diverse cancer cell types (18, 21, 22). Moreover, LPA levels have been found to be significantly elevated with infection (23), inflammation (24) and particular cancers (19, 22). These data have led to the notion that LPA contributes to the promotion and metastasis of cancer (22) and has focused attention on modulating LPA in vivo as a possible therapeutic approach (18).

In this study, we show that the LPA₅ GPCR expressed by mature B lymphocytes negatively regulates BCR signaling by inhibiting calcium release from intracellular stores via a LPA₅ – Gα12/13 – Arhgef1 pathway. We further demonstrate that LPA also diminishes the activation and antibody response of antigen-specific B cells upon engaging cognate antigen. Cumulatively, these data show that LPA directly regulates B lymphocyte activation and function via the LPA₅ GPCR that serves as a negative co-receptor for the BCR.

MATERIALS AND METHODS

Mice

C57BL/6 (Jackson Labs), C57BL/6-Igh^B1-8/B1-8 mice(25) (gift of Dr. Klaus Rajewsky, Harvard University), Arhgef1^−/−(26), Lpar2^−/− mice (27) (gift of Dr. Jerold Chun, Scripps Research Institute), B6.C20 mice (C57BL/6 mice congenic for Igh^a) (gift of Dr. Leonore Herzenberg, Stanford University) and Lpar5^−/− mice were bred and maintained within the Biological Resource Center at NJH and used in accordance with the regulations of the Institutional Animal Care and Use Committee. Lpar5^−/− mice were backcrossed to C57BL/6 for at least 3-4 generations before these analyses.

Generation of Lpar5^−/− mice

The Lpar5 targeting vector was derived using the Lambda KOS system (28). The Lambda KOS phage library was screened by PCR using Lpar5-specific primers: forward 5′-GTTCTGCCTGGGCGTGTG-3′ and reverse 5′-GCCAGCAGGAGGCGCAC-3′. Five pKOS genomic clones, (pKOS-28, pKOS-58, pKOS-68, pKOS-81, and pKOS-92) were isolated and confirmed by sequence and restriction analysis. Gene-specific arms (5′-CTACGATGCCTCAGACTAATTTCTCTTCCCACCTGGACAT-3′) and (5′-CTGCACACTAGAGCTGGAGTTGTTTCAAAGTCCAAGTAC-3′) were appended by PCR to a yeast selection cassette containing the URA3 marker. The yeast selection cassette and pKOS-58 were co-transformed into yeast, and clones that had undergone homologous recombination to replace a 1200 bp region containing the single coding exon with the yeast selection cassette were isolated. The yeast cassette was subsequently replaced with a LacZ/Neo selection cassette to complete the Lpar5 targeting vector. The Not I linearized targeting vector was electroporated into 129/SvEv^Brd (Lex-1) ES cells. G418/FIAU resistant ES cell clones were isolated, and correctly targeted clones were identified and confirmed by Southern analysis. Five targeted ES cell clones were identified and microinjected into C57BL/6 (albino) blastocysts to generate chimeric animals which were bred to C57BL/6 (albino) females, and the resulting heterozygous offspring were interbred to produce homozygous Lpar5 deficient mice. Mating of Lpar5^+/− mice generated pups of the three possible genotypes with ratios that fit well with normal Mendelian frequencies. Lpar5^−/− mice exhibited no substantial difference in growth rate and size.

Antibodies and flow cytometry

Rabbit anti-mouse IgG F(ab’)₂ (Zymed Labortories) and goat anti-mouse IgM F(ab’)₂ (Southern Biotechnology Associates) were used to signal via the BCR on A20 B lymphoma cells and primary B cells, respectively. Antibodies used for western blot analyses were anti-β-actin (AC-40; Sigma); anti-Syk, anti-Btk, anti-PLCγ2, anti-phospho-Syk (mouse Tyr519/520 or Tyr323), anti-phospho-Btk (Ser180; 3D3), and anti-phospho-PLCγ2 (Tyr759) from Cell Signaling Technology; rabbit anti-Gα12 (S-20), anti-Gα13 (A-20), anti-Gαq (E-17), HRP-labeled goat anti-mouse IgG and donkey anti-rabbit IgG from Santa Cruz Biotechnology.

The following fluorescently-labeled monoclonal antibodies were used for flow cytometric analysis: B220 (RA3-6B2; BD Pharmingen and eBioscience), CD69 (H1.2F3; eBioscience), CD86 (GL1; eBioscience), Igλ (JC5-1; Southern Biotech), CD23 (B3B4; BD Pharmingen) and CD21 (7G6; BD Pharmingen). NP-specific B cells were gated on 7AAD-negative viable cells and identified by flow cytometry after staining with Alexa647-coupled NP-CGG (Biosearch Technologies). Flow cytometric analyses were performed on either a FACSCalibur (BD Bioscience) or a BD LSRII (BD Bioscience) and analyzed with FlowJo v8.8.6 software (Tree Star, Inc.).

Western blot analysis

Splenic B cells were isolated by negative enrichment using anti-CD43 magnetic beads (Miltenyi Biotec), resuspended in FBS-free DMEM and incubated for 15 minutes at 37°C. Cells were then stimulated with 20 μg/ml anti-IgM F(ab’)₂ in the presence of 20 μM LPA or, in some experiments, with 1 μM Syk inhibitor, BAY61-3606 (EMD Millipore) for the indicated times and quenched in ice water. Cells were lysed in RIPA buffer and used to prepare whole cell lysates as described previously (29). Briefly, 30μl whole cell lysates were mixed with 6 μl 2X SDS loading buffer, and run on SDS-PAGE after boiling at 90-95°C for 5 min. After SDS-PAGE, proteins were transferred to nitrocellulose membranes (Bio-Rad) and blocked with 5% nonfat milk-TBS solution at room temperature for 30 minutes. Membranes were then incubated with primary antibody in 5% nonfat milk-TBS solution for 1 hour at room temperature or overnight at 4° C, washed 3 times with TTBS, incubated with HRP-labeled secondary antibody in 5% nonfat milk-TBS solution for 45 minutes, and washed 4 times with TTBS. Target proteins were detected by standard chemiluminescence and quantified by densitometric analysis using ImageJ (National Institutes of Health) software.

Intracellular calcium measurements

Approximately 0.5-1 × 10⁷ cells per ml were resuspended in IMDM (Invitrogen) with 2.5% fatty acid-free (faf) BSA plus 5 μM Indo-1-AM (Molecular Probes) and incubated at 37° C for 30 min. Subsequently, cells were washed twice with IMDM with faf 2.5% BSA, and resuspended at 2 × 10⁶ cells per ml in IMDM with faf 2.5% BSA. Cell aliquots were stored at room temperature for analysis. The Indo-1-AM 400/475 fluorescence ratio was acquired as a function of time, using a LSR (Becton Dickinson) or Mo-Flo (Cytomation) cytometer. After a baseline reading of 30 seconds, A20 or splenic B cells were stimulated with anti-IgG F(ab’)₂ (2 μg/ml) or anti-IgM F(ab’)₂ (20 μg/ml), respectively, with or without 1-20 μM lysophosphatidic acid (16:0, 1-palmitoyl-2-hydroxy-sn-glycero-3-phosphate, Avanti Polar Lipids; critical micelle concentration = 0.54 mM), and readings were then continued for an additional 5-6 minutes. We note that the maximum concentration of LPA used in experiments, 20 μM, is >25-fold below the critical micelle concentration. In some experiments 18:0, 18:1 and 20:4 species of LPA, phosphatidic acid 16:0 (1,2-dihexadecanoyl-sn-glycero-3-phosphate) or phosphatidylglycerol 16:0 (1,2-dipalmitoyl-sn-glycero-3-phospho-(1′-rac-glycerol)) were used and purchased from Avanti Polar Lipids. To distinguish ER calcium release from extracellular calcium influx, 4 mM EGTA was added to cell solution before stimulation and 8 mM Ca²⁺ was then added 210 seconds after stimulation. When used to discharge intracellular calcium stores, 1 μM thapsigargin (Alomone Labs) was added to cells to empty ER calcium stores.

In some experiments A20 cells were loaded with caged-IP₃ (cat. no. cag-iso-2-145; SiChem GmbH, Germany) and Indo-1-AM and resuspended in IMDM supplemented with 2.5% faf BSA and 4 μM EGTA. Ca²⁺ concentration was recorded for 45 seconds to establish basal Ca²⁺ levels after which samples were exposed to ultraviolet light (330 ± 25nm) for 1 second to release IP₃ and cells were further monitored for an additional 3 minutes.

LPA receptor expression

Lpar mRNA expression was determined in mature follicular B cells identified by surface phenotype as CD23+ CD21^intermediate and purified using a MoFlo (Dako Cytomation) cell sorter. RNA was isolated using TRIzol (Invitrogen Life Technologies), and DNA removed using a DNA-free kit (Ambion). cDNA was prepared from equivalent amounts of RNA using a SuperScript III First-Strand Synthesis System for RT-PCR (Invitrogen Life Technologies). Quantitative PCR amplification was performed using Platinum SYBR Green qPCR SuperMix-UDG (Invitrogen Life Technologies) and detected on an MJ Research DNA Engine Opticon 2 real-time PCR machine. Primers for the LPA receptors were the following:

• LPA₁-forward CTGTGGTCATTGTGCTTGGTG

• LPA₁-reverse CATTAGGGTTCTCGTTGCGC

• LPA₂-forward GGCTGCACTGGGTCTGGG

• LPA₂-reverse GCTGACGTGCTCCGCCAT

• LPA₃-forward GCGCACAGGAATGGGAGAG

• LPA₃-reverse GAGCTGGAGGATGTTGGGAG

Primers from LPA₄ and LPA₅ were purchased from Super Array Biosciences (Frederick, MD). The cycle threshold (Ct) was determined for each experiment based on background from control samples (no cDNA). Expression of Lpar mRNA of interest was normalized to HPRT mRNA and calculated using the formula 2^−ΔCt × 100.

Cell culture and transfection

A20 B lymphoma cells (ATCC; TIB-208), HEK-293FT (Invitrogen) cells and Phoenix retroviral packaging cells were maintained in DMEM (Mediatech, Herndon, VA) supplemented with 10% fetal calf serum, 1000 units/mL penicillin, 100 μg/mL streptomycin, and Glutamax. HEK-293FT cells were transfected with retroviral shRNA vectors and packaging vectors using PolyFect transfection reagent (QIAGEN). Approximately 12 hours later, fresh medium was added and retroviral-containing supernantant was collected 24 hours later (36 hours after transfection) and used to transduce A20 cells.

To establish stable A20 cell lines constitutively expressing either Gαq shRNA (A20/shGaq) or doxycycline-inducible Gα12 and Gα13 shRNA (A20-shG12/13) cells were infected using HEK293FT supernatants previously transfected with the appropriate lentiviral vectors. Fluorescent-positive cells were sorted by flow cytometry on day 5 after infection and cultured in standard medium. After another 5 days, cells were again sorted and cultured until cells were 100% GFP (A20/shGaq) or Venus (A20-shG12/13) positive. These stable cell lines were then used for further experiments. To inhibit the expression of Gα12 and Gα13, these corresponding stable cell lines were cultured in medium containing 1 μg/ml doxycycline (Sigma) for 5 days. Plasmids (and source) used in these experiments were pLX_mU6-Gq-I-UGIH (ATCC; 10629198) for Gαq knockdown and pSLIK-Venus-TmiR-G12-G13 (30) (gift of Drs. Iain Fraser and Mel Simon, Cal Tech) for simultaneous doxycycline induced knockdown of Gα12 and Gα13. The control shRNA vector pLX-mU6-I-UGIH for pLX-mU6-Gq-I-UGIH was constructed by deleting the Gαq targeting sequence by site-directed mutagenesis (Stratagene).

To generate LPA receptor shRNA vectors, the following shRNA sequences were cloned into the pMSCV-LTRmiR30-PIG retroviral vector obtained from Open Biosystems (Huntsville, AL). Lpar5 sense oligo: TGCTGTTGACAGTGAGCGAGCGAGATACACATCGTTTGCATGTAGTGAAGCCACAG ATGTACATGCAAACGATGTGTATCTCGCGTGCCTACTGCCTCGGA or luciferase control oligo: TGCTGTTGACAGTGAGCGAGCTTACGCTGAGTACTTCGAGTAGTGAAGCCACAGATG TACTCGAAGTACTCAGCGTAAGCGTGCCTACTGCCTCGGA. These vectors contain an internal ribosomal entry site (IRES) with green fluorescent protein (GFP) as a reporter and a puromycin resistance gene. For virus production, Phoenix cells were transfected with 10 μg retroviral constructs and 2 μg of pCL-ECO retroviral packaging construct per plate using Lipofectamine (Invitrogen). Medium was replaced after 24 hours and supernatant was collected 48 and 72 hours after transfection.

Immature B cells expressing LPA₅ or control shRNA were produced by retroviral transduction of 3-83Igi, H-2d bone marrow cells (31) using spin-infection. 3-83Igi, H-2d bone marrow cells were isolated (day 0) and cultured in complete IMDM supplemented with 50-100 Units/ml IL-7 for 2 days as described (13). On days 2 and 3, 8 × 10⁶ cells per well of 24 well plate were retrovirally transduced with 1 mL of Phoenix cell supernatants, 0.5 mL of complete IMDM, 3.2 μg/mL of polybrene and IL-7 during centrifugation at 2300 RPM at ~22°C for 120 minutes. Two hours after spin-infection, medium was replaced with complete IMDM supplemented with IL-7. Cells were cultured with IMDM supplemented with IL-7 on days 4 and 5. On day 5, before cells were used in calcium assays, cells were washed of IL-7 and incubated for 2 hours in calcium medium for serum starvation.

In vitro B Cell Activation

Single cell suspensions were prepared from the spleens of 8 week old C57BL/6-Igh^B1-8/B1-8 mice after red blood cell lysis and B cells purified using negative selection with anti-CD43 magnetic microbeads and an autoMACS (Miltenyi Biotec). B cell purity as determined by flow cytometric analysis was at all times >95%. B cells were resuspended at 2 × 10⁶ cells per ml and stimulated with 10 ng/ml of NP₂₈-Ficoll (Biosearch Technologies) in the presence of 10 ng/ml recombinant BAFF (R&D Systems) to promote B cell survival and with or without 20 μM LPA or 50 μM sterile-filtered OTP. After 24 hours, cells were collected, stained, and analyzed for the expression of Igλ, CD86 and CD69 by flow cytometry.

Mixed Bone Marrow Chimeras

Mixed bone marrow adoptive transfers were performed as described(32). Briefly, bone marrow cells were isolated from 10-12 week-old Lpar5^−/− (Igh^b), Lpar5^+/+ littermates (Igh^b) and B6.C20 (Igh^a) donor mice, mixed at a ratio of either 1:1 or 1:4 (Igh^b:Igh^a) and a total of 1 × 10⁶ cells injected i.v. into 6-8 week old C57BL/6 recipients that had been lethally irradiated (2 × 500 rad). At 5-6 weeks post-reconstitution, PBLs from chimeric mice were analyzed by flow cytometric analysis to ensure engraftment of both IgM^b and IgM^a-expressing B cells and to measure reconstitution frequencies in the experimental chimeras as compared to the control chimeras. Recipient mice reconstituted with IgM^b-expressing Lpar5^−/− and Lpar5^+/+ B cells at frequencies between 25-35% (1:1 transferred ratio) or 5-25% (1:4 transferred ratio) were immunized at 6-7 weeks post-reconstitution and 7 days later serum NP-specific antibody responses were measured by ELISA and the total number and frequency of splenic IgM^b and IgM^a-expressing B cells enumerated.

Immunizations, ELISA and ELISpot

Mice were immunized i.p. as indicated with 5 μg NP (4-Hydroxy-3-nitrophenylacetic) hapten conjugated to Amino-Ethyl-Carboxy-Methyl-FICOLL (NP₂₈-Ficoll; Biosearch Technologies). Serum was obtained at day 0 before immunization and 7 days later and NP-specific (allotype-specific) IgM ELISAs and ELISpots performed as described previously (32).

Statistical and Data Analyses

Data were graphed and analyzed using GraphPad Prism v5.0a (GraphPad Software). Statistical significance was assessed with a Student’s t test with unequal variance, and the appropriateness of one-tailed or two-tailed significance was determined on an individual experiment basis. P values < 0.05 were considered significant.

RESULTS

LPA inhibits antigen receptor-induced calcium mobilization

The S1P and LPA lysophospholipids have previously been shown to modulate B and T cell antigen receptor proliferation and cytokine secretion (7, 33-35). However, the mechanism(s) by which these lysophospholipids regulate antigen receptor signaling is not established and we questioned whether LPA could influence early BCR signaling events. LPA alone at concentrations as high as 20 μM did not mobilize intracellular Ca²⁺ in either the A20 B lymphoma cell line or murine splenic B lymphocytes (Figure 1A), similar to a previous report (35). However, the ability of the BCR to signal an increase in intracellular Ca²⁺ was inhibited when B cells were simultaneously treated with LPA in a dose-dependent manner including at pathophysiological concentrations (Figure 1A). Multiple molecular species of LPA exist in biological fluids (19), and we found that 16:0, 18:0, 18:1 and 20:4 LPA all similarly inhibited BCR-mediated increase of intracellular Ca²⁺ (unpublished data). LPA 16:0 was subsequently used for all experiments and we note that the critical micelle concentration for 16:0 LPA is 0.54 mM and considerably higher than 20 μM, the highest LPA concentration used in our studies. LPA inhibition of anti-IgM stimulated intracellular calcium mobilization by splenic B cells was also specific inasmuch as neither phosphatidic acid or phosphatidylglycerol were able to alter BCR calcium signaling (Supplemental Figure 1A). In the presence of 20 μM LPA, peak intracellular calcium levels of wild type splenic B cells were routinely and significantly (n=5, p<.001) reduced to 75% the levels of that observed in the absence of LPA. Furthermore, if B cells were pre-treated with LPA for more than 30 minutes in advance, this inhibition was lost (Supplemental Figure 1B) and presumably as a result of GPCR desensitization. Thus, LPA negatively and specifically regulates B cell antigen receptor Ca²⁺ signaling, presumably by engaging its cognate GPCR(s) expressed on the B cell surface.

Figure 1.

LPA inhibits BCR-mediated intracellular Ca²⁺ mobilization. (A) A20 B cells and primary splenic B cells were stimulated (arrow) with F(ab’)₂ anti-IgG (2 μg/ml) or anti-IgM (20 μg/ml), respectively, in the absence or presence of increasing concentrations of LPA and intracellular Ca²⁺ levels measured by flow cytometry over time. LPA (20 μM) in the absence of BCR signaling is shown as a thin dotted line. Representative of 4 independent experiments. (B) LPA does not influence anti-Ig-induced phosphorylation of Syk, Btk or PLCγ2. Primary splenic B cells were stimulated with F(ab’)₂ anti-IgM (20 μg/ml) in the presence or absence of 20 μM LPA for the indicated times. After stimulation, total and phosphorylated Syk, Btk, and PLCγ2 were measured in whole cell lysates. Numbers below western blots indicate fold-induction over normalized time 0 protein levels. Representative of 3 experiments. (C) LPA inhibits IP₃-induced intracellular Ca²⁺ release. A20 cells were loaded with caged-IP₃ and in the presence of 4 mM EGTA, Ca²⁺ concentration was recorded for 45 seconds to establish basal Ca²⁺ levels. Samples were then exposed to ultraviolet light (330 ± 25nm) for 1 second to release caged IP₃ and further monitored for an additional 3 minutes (solid line). Dotted line represents samples that were treated with 5 μM LPA prior to UV exposure. Representative of two experiments with similar results.

The increase in intracellular Ca²⁺ levels after BCR stimulation ultimately results from the activation of phospholipase C-γ2 (PLCγ2)(36, 37). PLCγ2 is activated upon tyrosine phosphorylation and hydrolyzes phosphatidylinositol-4,5-bisphosphate (PIP₂) to produce inositol 1,4,5-trisphosphate (IP₃) that subsequently engages IP₃ receptors (IP₃Rs). IP₃Rs are ligand-gated Ca²⁺ release channels expressed on the surface of the endoplasmic reticulum (ER) that mediate the release of intracellular Ca²⁺ stores on binding IP₃ (reviewed in refs.(36-38). In B cells, Syk, Lyn and Btk tyrosine kinase activities are required for the activation of PLCγ2 (36). To assess whether LPA regulates BCR-mediated Ca²⁺ mobilization by altering the activation of PLCγ2 or activating tyrosine kinases, splenic B cells were isolated and stimulated with anti-IgM F(ab’)₂ in the absence or presence of LPA. Figure 1B (and Supplemental Figure 2) shows that tyrosine phosphorylation of Syk, Btk, and PLCγ2 was similar in the absence or presence of 20 μM LPA suggesting that Syk, Btk or PLCγ2 activation are not influenced by LPA signaling. These data further suggest that LPA inhibition of BCR Ca²⁺ signaling is downstream of PLCγ2 activation.

IP₃ concentration is tightly regulated by inositol lipid kinases and phosphatases (39) and decreased IP₃ production can lead to impaired intracellular Ca²⁺ mobilization. To address whether LPA inhibition of Ca²⁺ mobilization resulted from altered IP₃ production, we exploited a membrane-permeable ester of IP₃ (ci-IP₃/PM) to introduce exogenous photoactivatable IP₃ into the cytosol after UV exposure. A20 B cells were loaded with ci-IP₃/PM and the release of Ca²⁺ from intracellular stores was measured after liberating IP₃ in the absence of BCR stimulation. As expected, introduction of exogenous IP₃ into A20 cells resulted in a rapid and transient increase in intracellular Ca²⁺ (Figure 1C). However, in the presence of 5 μM LPA this Ca²⁺ increase was considerably reduced providing strong evidence that LPA inhibition of intracellular Ca²⁺ mobilization was independent of IP₃ production. These data indicate that LPA regulates BCR signaling downstream of PLCγ2 activation and IP₃ generation.

LPA inhibits calcium release from intracellular stores

Elevated intracellular Ca²⁺ levels mediated by BCR signaling is a consequence of both the release of Ca²⁺ from intracellular stores and the entry of extracellular Ca²⁺ following activation of store-operated Ca²⁺ channels (SOC) in the plasma membrane. We next evaluated whether LPA regulated either or both of these Ca²⁺ mobilization events.

To measure Ca²⁺ release from intracellular stores, A20 cells were stimulated with anti-IgG F(ab’)₂ in the presence of EGTA to prevent extracellular Ca²⁺ influx. Once Ca²⁺ concentrations returned to baseline levels after BCR stimulation, extracellular Ca²⁺ was restored and plasma membrane SOC channel activity measured as Ca²⁺ entry. These initial experiments revealed that both the release of Ca²⁺ from intracellular stores and extracellular Ca²⁺ entry were inhibited by LPA in a dose dependent manner (Figure 2A). Specifically, increasing LPA concentration progressively diminished both the magnitude and duration of intracellular Ca²⁺ release. Furthermore, the degree to which Ca²⁺ store-release was inhibited by LPA correlated with the amount of subsequent extracellular Ca²⁺ influx.

Figure 2.

LPA inhibits intracellular Ca²⁺ stores release. A) A20 cells were loaded with Indo-1-AM, resuspended in 4 mM EGTA and intracellular Ca²⁺ concentration measured. At the time point indicated by the first arrow, cells were stimulated with 2 μg/ml F(ab’)₂ anti-IgG alone (black line) or with the indicated increasing concentrations of LPA. Once Ca²⁺ concentration returned to baseline levels extracellular Ca²⁺ was restored (second arrow) and Ca²⁺ entry measured for approximately 3 additional minutes. B) Indo-1-loaded A20 cells were resuspended in EGTA and treated with thapsigargin (1 μM) alone (black line) or with 1 μM (blue line), 5 μM (red line) or 20 μM (green line) LPA. Once intracellular Ca²⁺ levels returned to baseline, extracellular Ca²⁺ was again restored and SOC channel activity measured as Ca²⁺ entry. C) Indo-1-loaded A20 cells were resuspended in EGTA and treated (first arrow) with thapsigargin alone (red line). As intracellular Ca²⁺ stores were discharging, some cells were additionally treated (second arrow) with anti-Ig alone (black line) or together with 5 μM LPA (green line).

Ca²⁺ concentration within the ER is monitored by STIM1and upon depletion of ER Ca²⁺ stores, STIM1 activates the plasma membrane SOC facilitating entry of extracellular Ca²⁺ (38, 40, 41). Although both Ca²⁺ store release and extracellular influx appeared attenuated by LPA, we asked whether the reduced extracellular Ca²⁺ entry might be an indirect consequence of reduced store release. ER Ca²⁺ stores can be artificially depleted with thapsigargin, a pharmacologic inhibitor of the ER resident Ca²⁺ ATPase (42) that functions to refill ER stores by transporting cytosolic Ca²⁺ back into the ER. To assess whether LPA can directly inhibit SOC channel activity, intracellular Ca²⁺ concentration was measured in A20 cells treated with thapsigargin in the absence or presence of 1, 5, or 20 μM LPA. Thapsigargin-treated A20 cells slowly discharge Ca²⁺ from intracellular stores resulting in elevated cytosolic Ca²⁺ and subsequent SOC activation and extracellular Ca²⁺ entry (Figure 2B). However, under these same conditions, LPA was unable to inhibit extracellular Ca²⁺ influx at any concentration. These data suggest that LPA does not directly inhibit store-operated Ca²⁺ channels following BCR signaling.

Our data thus far indicate that LPA inhibits BCR Ca²⁺ signaling by reducing the total amount of Ca²⁺ stores release and the length of time in which it is released (Figure 2A). Upon BCR-mediated IP₃ production, ER Ca²⁺ concentration is determined by the opposing activities of the IP₃R releasing Ca²⁺ into the cytosol and the Ca²⁺ ATPase transporting cytosolic Ca²⁺ back into the ER. Thus, we reasoned that the inhibition of Ca²⁺ store release by LPA could manifest either by inhibiting IP₃R-induced Ca²⁺ release or by enhancing Ca²⁺ ATPase activity. To discriminate between these possibilities, we tested whether LPA inhibition of BCR Ca²⁺ signaling required Ca²⁺ ATPase activity. Thus, A20 cells were treated with thapsigargin in the presence of EGTA to inhibit the Ca²⁺ ATPase and initiate ER stores depletion. During thapsigargin treatment, A20 cells were stimulated via the BCR in the presence or absence of LPA. These results show that in the absence of LPA, thapsigargin-mediated release of Ca²⁺ from intracellular stores was further increased by BCR stimulation, presumably as a consequence of the additional ER Ca²⁺ release mediated by the IP₃Rs engaging BCR-generated IP₃ (Figure 2C). However, in the presence of LPA this additional increase in cytosolic Ca²⁺ was inhibited. These findings demonstrate that LPA can still reduce BCR Ca²⁺ signaling when the Ca²⁺ ATPase is inhibited, excluding a direct effect of LPA on the Ca²⁺ ATPase.

Considered together, the results from the above experiments provide compelling evidence that LPA regulates BCR Ca²⁺ signaling by inhibition of IP₃R-induced Ca²⁺ store release.

LPA signals via LPA₅ to inhibit BCR-mediated calcium signaling

Of the five initially established LPA receptors (17, 18), mature follicular B lymphocytes and A20 B cells predominantly express LPA₂ and LPA₅ as determined by quantitative PCR (Figure 3A and B). To determine which of these LPA receptors regulates BCR signaling we first examined LPA inhibition of BCR signaling in B cells from LPA₂ knockout mice (27) and found that LPA maintained the ability to inhibit BCR Ca²⁺ signaling in Lpar2^−/− B cells (Supplemental Figure 3). This indicates LPA₂ does not contribute to LPA regulation of BCR signaling. To determine whether LPA₅ participated in this regulation we initially used RNA interference to inhibit LPA₅ expression. As shown, stable expression of an LPA₅ short hairpin RNA (shRNA) in the A20 B cell line specifically reduced LPA₅ expression by 67% relative to a control short hairpin and parental cells (Figure 3B). Furthermore, BCR Ca²⁺ signaling was no longer attenuated by LPA in either A20 cells or primary B cells expressing shRNAs for LPA₅ (Figure 3C). These findings show that LPA₅ expressed by mature B cells is responsible for repressing BCR signaling.

Figure 3.

The LPA₅ receptor inhibits BCR signaling. A) Follicular B cells were evaluated for the expression of LPA_1-5 as measured by quantitative PCR. B) LPA₅ and LPA₂ expression was measure by quantitative PCR in parental A20 cells (open bar) or cells transduced with a retrovirally encoded control shRNA (grey bar) or LPA₅ shRNA (black bar). C) BCR-mediated Ca²⁺ mobilization was measured in A20 cells (top) and in primary B cells (bottom) that had been transduced with retroviruses encoding either a control (black lines) or LPA₅ (grey lines) shRNA and in the absence (solid lines) or presence (dotted lines) of 20 μM LPA.

A Gα12/13-associated LPA receptor inhibits BCR calcium signaling

The LPA₅ receptor signals by associating with either Gαq or Gα12/13 heterotrimeric subunits (17, 18, 43). To investigate which of these Gα heterotrimeric G-protein families was responsible for LPA₅ inhibition of BCR signaling we again used RNA interference. We initially stably expressed a shRNA specific for Gαq in A20 cells (A20/shGαq) that led to an approximate 98% reduction in Gαq expression (Figure 4A). Despite this significant reduction in Gαq expression, 5 μM LPA still suppressed BCR-mediated ER Ca²⁺ release in A20/shGαq cells to a level comparable to the control shRNA cell line or parental A20 cells (Figure 4B). These data indicate that LPA₅ does not associate with Gαq to regulate B cell antigen receptor Ca²⁺ signaling.

Figure 4.

LPA₅ signals via a Gα12/13 – Arhgef1 signaling pathway to repress BCR calcium signaling. A) Western blot analysis of Gαq from stable A20 transfectants harboring a control vector (A20shCTL) or expressing a Gαq shRNA (A20shGαq). Beta-actin is shown as loading control. B) Intracellular Ca²⁺ stores release after anti-Ig stimulation (arrow) in A20 parental cells (black solid line) or A20 shRNA transfectants (grey solid line) alone or in the presence of 5 μM LPA (dotted lines). Ca²⁺ mobilization for A20shCTL is shown on the left and for A20shGαq is shown on the right. Representative of 3 independent experiments. C) Western blot analysis of Gα12 and Gα13 in A20-shG12/13 cells before (-) and after (+) doxycycline (DOX) treatment. Beta-actin is shown as loading control. D) Intracellular Ca²⁺ stores release was measured in parental A20 (dotted line) and A20-shG12/13 (solid line) in the presence of EGTA after anti-Ig stimulation alone (top panels) or with 5 μM (middle panels) or 20 μM (bottom panels) LPA. Intracellular Ca²⁺ stores release was also measured in the absence (left column) or presence (right column) of doxycycline. Data shown is representative of similar results from 3 independent experiments. E) BCR-mediated intracellular Ca²⁺ mobilization was measured in splenic B cells from C57BL/6 (top panel) and Arhgef1^−/− (bottom panel) mice in the absence (solid line) or presence (dotted line) of 20 μM LPA.

The Gα12 family is composed of Gα12 and Gα13 (44) and the participation of these Gα subunits in LPA regulation of BCR signaling was initially evaluated individually. However, constitutive repression of Gα12 or Gα13 expression in A20 cells could not be maintained for more than two passages (unpublished data). Thus, we exploited an inducible knockdown platform to conditionally repress Gα12 and Gα13 simultaneously upon treatment with doxycycline (30). Doxycycline-regulated expression of microRNA-like shRNAs that target Gα12 and Gα13 were introduced into A20 cells with a single lentiviral infection and transduced cells (A20-shG12/13) were monitored by the expression of the Venus fluorescent protein. Figure 4C shows that upon doxycycline treatment, Gα12 and Gα13 expression was reduced by 92% and 88%, respectively, compared to untreated cells. To assess the contribution of Gα12 and Gα13 to LPA₅ receptor inhibition of BCR signaling, a mix of Venus-positive transduced cells and Venus-negative parental A20 cells were treated or not with 1 μg/ml doxycycline for an additional 5 days. Subsequently, parental A20 (Venus-negative) and A20-shG12/13 (Venus-positive) cells were simultaneously evaluated for Ca²⁺ stores release following BCR stimulation (Figure 4D). In the absence of doxycycline, both parental A20 and A20-shG12/13 cells released Ca²⁺ from intracellular stores to the same level when stimulated through the BCR with anti-IgG; this level was similarly inhibited by LPA, indicating that lentiviral transduction and/or expression of Venus did not alter BCR or LPA₅ signaling. In contrast, similar BCR stimulation of doxycycline-treated cells abolished the ability of LPA to inhibit BCR Ca²⁺ signaling in A20-shG12/13 but not parental A20 cells (Figure 4D; right panels). Considered together, these data provide compelling support that LPA₅ associates with Gα12/13 heterotrimeric proteins to negatively regulate BCR-mediated Ca²⁺ intracellular stores release.

Arhgef1 is required for LPA Inhibition of BCR calcium signaling

Gα12/13-associated GPCR stimulate a family of signaling molecules referred to as RGS-containing RhoGEFs, whose family members are Arhgef1 (also known as murine Lsc or human p115RhoGEF), Arhgef12 (LARG), and Arhgef11 (PDZ-RhoGEF)(45). In addition to functioning as RhoA guanine nucleotide exchange factors (GEFs), these family members also harbor regulator of G-protein signaling (RGS) domains that function as GTPase activating proteins for GTP-bound Gα12/13 subunits, effectively terminating signaling by the these Gα subunits.

To examine the contribution of Arhgef1 in the signaling pathway from LPA receptor to regulation of intracellular stores Ca²⁺ concentration we evaluated B lymphocytes from Arhgef1-deficient mice (26). Figure 4E shows that LPA is unable to inhibit BCR Ca²⁺ signaling in Arhgef1-deficient primary splenic B cells. This demonstrates that Arhgef1 is also required for LPA inhibition of BCR signaling. Therefore, we conclude that LPA₅ inhibits BCR signaling through a Gα12/13-associated heterotrimeric G-protein and Arhgef1.

Antigen-mediated B cell activation and antibody response is inhibited by LPA

We next asked if LPA inhibition of BCR-induced Ca²⁺ signaling had more distal biological consequences on B cell activation and antibody response after antigen-specific BCR signaling. To investigate this we used Igh^B1-8/B1-8 mice, whose B cells express the same VDJ (B1-8) rearrangement inserted by gene-targeting where rearrangements naturally occur at the Igh loci (25, 32). Thus, all Igh^B1-8/B1-8 B cells express a B1-8 Ig heavy chain that when paired with an Igλ light chain generates specificity for the 4-hydroxy-3-nitrophenylacetyl (NP) hapten (46) (Figure 5A). Stimulation of splenic Igh^B1-8/B1-8 B cells in vitro with 10 ng/ml of NP-Ficoll resulted in increased intracellular Ca²⁺ levels in Igh^B1-8/B1-8 Igλ⁺ B cells that was again blunted in the presence of 20 μM LPA (Figure 5B). Thus, LPA inhibits BCR-mediated intracellular Ca²⁺ mobilization after antigen-specific stimulation.

Figure 5.

LPA inhibits antigen-specific B cell activation. A) Histogram of Igλ+ (filled grey) and Igλ-negative (open) B220⁺ Igh^B1-8/B1-8 B cells labeled with FITC-NP hapten. B) Igλ+ NP-specific B cells were treated with 10 ng/ml of NP-Ficoll and intracellular Ca²⁺ measured over time in the presence (dotted line) or absence (solid line) of 20 μM LPA. C) Primary splenic Igh^B1-8/B1-8 B cells were isolated and stimulated with 10 ng/ml NP-Ficoll for 24 hours and CD69 expression measured by flow cytometry on antigen-specific Igλ+ (filled grey) and antigen-nonspecific Igλ-negative (open) B cells. D) CD69 (left panel) and CD86 (right panel) expression on viable antigen-specific Igλ+ Igh^B1-8/B1-8 B cells 24 hours after stimulation with NP-Ficoll alone (grey filled histogram) or with 20 μM LPA (dotted line) or 50 μM OTP (solid line). Data in each panel are representative of at least 4 independent experiments.

Antigen-specific B cell activation was tested by evaluating the expression of the CD69 and CD86 activation antigens on splenic Igh^B1-8/B1-8 B cells treated with 10 ng/ml NP-Ficoll in the presence or absence of LPA or a synthetic metabolically stable LPA analog, octadecenyl thiophosphate (OTP). OTP is a potent agonist of LPA₅ (47) with a 10-fold selectivity over LPA (48) and that also efficiently inhibits BCR-mediated intracellular Ca²⁺ mobilization (unpublished data). As expected, 24 hours after NP-Ficoll stimulation of splenic B cells in vitro, the expression of CD69 was increased on antigen-specific Igλ⁺ Igh^B1-8/B1-8 but not antigen non-specific Igλ⁻ B cells (Figure 5C). Furthermore, when B cells were stimulated with specific antigen, the expression of both CD69 and CD86 activation antigens was inhibited in the presence of either LPA or OTP (Figure 5D).

These data thus far demonstrate that LPA signaling via LPA₅ inhibits both BCR-induced intracellular calcium mobilization and expression of activation antigens. To test if LPA₅ acts normally in vivo to inhibit the antibody response, we obtained LPA₅-deficient mice and initially evaluated the follicular (FO) and marginal zone (MZ) B cell populations in the spleen. Lpar5^−/− MZ and FO B cells are present at normal frequencies (Figure 6A) and appear to localize normally within the spleen as evidenced by the IgM^highIgD⁻ MZ B cells in the marginal zone surrounding the IgM⁺IgD⁺ FO B cells in the follicles (Figure 6B). Both FO and MZ B cells are inhibited in intracellular Ca²⁺ mobilization by LPA after BCR stimulation (Figure 6C), but in the absence of LPA₅, LPA is unable to suppress BCR-induced intracellular Ca²⁺ response (Figure 6D). These data indicate FO and MZ B cell populations develop normally in the absence of LPA₅ and confirm that LPA₅ is the LPA receptor responsible for negatively regulating BCR signaling.

Figure 6.

LPA₅ suppresses the TI-2 antibody response to NP-Ficoll. A) Percent transitional T1, follicular and marginal zone B cell populations in the spleen of Lpar5^+/+ control littermate (open bars) and Lpar5^−/− (filled bars) mice. B) Representative immunofluorescence histology of Lpar5^+/+ (left) and Lpar5^−/− (right) spleen using antibodies against IgM (red), IgD (blue) and CD3 (green). C) Wild type splenic B cells were isolated and treated with F(ab’)₂ anti-IgM (10 μg/ml) in the presence (dotted line) or absence (solid line) of 20 μM LPA and intracellular Ca²⁺ levels measured over time in cells gated based on CD21 and CD1d expression to identify FO (top) and MZ (bottom) B cells. D) B220⁺ B cells from Lpar5^−/− (bottom) or Lpar5^+/+ (top) littermates were stimulated with 10 μg/ml F(ab’)₂ anti-IgM without (solid line) or with (dotted line) 20 μM LPA and intracellular Ca²⁺ levels measured over time. E) Lpar5^−/−, wild type littermate or C57BL/6 mice were immunized with 5 μg NP-Ficoll and NP-specific IgM levels determined by ELISA. Data are shown as day 4 NP-specific IgM over day 0 pre-immune NP-specific IgM levels for 6-10 mice immunized in 3 different experiments. Day 0 pre-immune NP-specific IgM concentrations (μg/ml) were 11.9±8.6 for controls and 8.0±4.7 for mutants and not statistically different. *p value < 0.02 using a one-tailed Student’s t-test with unequal variance. F) NP-specific IgM^b concentration was measured in the serum of mixed bone marrow chimeras 7 days after immunization with 5 μg/ml NP-Ficoll and normalized to the number of reconstituted IgM^b-positive Lpar5^−/− (experimental chimeras; filled circles) or Lpar5^+/+ (littermate control chimeras; open circles) B cells. Data are combined from two independent experiments. G) NP-specific IgM^b ASCs enumerated in the spleen of adoptive mixed bone marrow chimeras 7 days after immunization from two independent experiments. Reconstitution frequencies and numbers of splenic IgM^b+ B cells and day 0 and 7 NP-specific IgM^b concentrations are shown for two independent mixed bone marrow chimeras in Supplemental Figure 4. Symbols represent individual chimeric mice, and horizontal bars indicate mean. Data are representative of three independent experiments. *p < .05 using a two-tailed Student’s t-test with unequal variance.

We next measured the antibody response by Lpar5^−/−, Lpar5^+/+ littermate controls and wild type mice to NP-Ficoll, a T-independent type 2 (TI-2) antigen that elicits antigen-specific IgM essentially exclusively from MZ B cells (32, 49-52). Four days after immunization, NP-specific antibody was measured in serum by ELISA and revealed that Lpar5^−/− mice produce twice the level of NP-specific IgM compared to littermate control and wild type C57BL/6 mice (Figure 6E). Although antigen-specific IgM production in response to NP-Ficoll is independent of T cells, the presence of (non-cognate) T cells, and possibly other hematopoietic cells, can influence the magnitude of this response. Thus, to more directly evaluate the role of LPA₅ expressed by B cells in the antibody response, we immunized mice with NP-Ficoll that had been lethally-irradiated and reconstituted with a mixture of bone marrow cells from Lpar5^−/− and wild type animals. Specifically, bone marrow from Lpar5^−/− (IgM^b) and B6.C20 (IgM^a) mice, or as control Lpar5^+/+ (IgM^b) and B6.C20 (IgM^a) mice, was used to reconstitute lethally irradiated wild type hosts, and the antibody response from each B cell population was measured using an ELISA that discriminates between IgM allotypes. Approximately 6 weeks after adoptive transfer, reconstituted mice were immunized with 5 μg NP-Ficoll and 7 days later NP-specific IgM^b antibody measured in the serum of individual recipient mice. To ensure that the antibody response reflected a difference in B cell function and not a difference in number of reconstituted B cells, we normalized the antibody response to the number of reconstituted mutant and wild type B cells. Importantly, in these adoptive mixed bone marrow chimeras splenic lymphoid architecture develops normally and IgM^b+ Lpar5^−/− B cells localize to both the B cell follicle and marginal zone similar to wild type B cells. The results from these experiments revealed that NP-specific IgM^b produced by Lpar5^−/− B cells was significantly increased relative to that produced by wild type B cells and in the presence of a wild type hematopoietic cells and host environment (Figure 6F). Furthermore, this elevated antibody response was also reflected by the increased number of NP-specific antibody secreting cells recovered in the spleen of immunized recipients 7 days after immunization (Figure 6G). From these data we conclude that in response to endogenous LPA concentrations LPA₅ suppresses the primary B cell antibody response to a TI-2 model antigen.

DISCUSSION

Extracellular bioactive lysophospholipids signal through GPCR expressed by diverse cell types to regulate a number of cell activities (11). In this report we show that LPA signals via the LPA₅ receptor to inhibit antigen-specific BCR signaling, subsequent B cell activation and in vivo antibody response. These findings extend the influence of lysophospholipids on immune cell function to include regulating the activation and function of B lymphocytes. Our data further provide the initial evidence that a GPCR can serve as a BCR co-receptor that negatively regulates antigen receptor signaling by B cells.

Both LPA and S1P have previously been shown to modulate antigen receptor signaling by T lymphocytes (7, 33, 34) although the precise lysophospholipid receptors or T cell signaling pathways mediating this regulation were not determined. Recently, we have demonstrated that LPA signaling via the LPA₅ receptor expressed by tumor-specific CD8 T cells suppresses tumor immunity (53). Our findings here show that physiological and pathophysiological concentrations of LPA act on the LPA₅ receptor to suppress BCR-mediated calcium release from intracellular stores. As a signaling second messenger in lymphocytes, calcium is required for a number of cellular activities that include the regulation of the NF-κB and NF-AT transcription factors (54, 55) as well as lymphocyte differentiation and effector function (37, 56). We show that LPA inhibition manifests early by preventing complete intracellular calcium mobilization and later suppresses antigen-specific induction of CD69 and CD86 expression and ultimately limits the early primary antibody response.

In B cells, the signaling pathway from the BCR leading to elevated intracellular calcium has been relatively well studied (36, 37). Our RNA interference experiments initially identified LPA₅ and Gα12/13 as the LPA receptor and associated Gα subunits that inhibit BCR calcium signaling by impairing IP₃R-mediated Ca²⁺ release from intracellular stores. This conclusion is supported by experiments evaluating signaling events receptor-proximal and distal to IP₃R activity in the presence of LPA. These results show BCR-induced tyrosine phosphorylation of several membrane-proximal signaling intermediates to, and including, PLCγ2 is comparable in the presence or absence of LPA and indicating LPA interferes with BCR signaling downstream of PLCγ2. Consistent with this we find in the presence of LPA intracellular stores release to be inhibited while SOC channel activity appears normal. The inhibition of Ca²⁺ stores release by LPA could in principle be accomplished by either suppressing IP₃R activity or by enhancing the calcium ATPase transport of calcium back into the ER. However, when the ATPase was inhibited by thapsagargin and ER calcium was being discharged, BCR signaling could further increase calcium stores release and LPA maintained the ability to inhibit this increase. This suggests that LPA does not act by enhancing ATPase activity. Further evidence that LPA suppresses IP₃R activity was provided by the introduction of exogenous IP₃ into B cells that leads to a transient increase in cytosolic Ca²⁺ that can be inhibited in the presence of LPA (Figure 1C). These findings considered together provide strong evidence that LPA₅ signaling impairs IP₃R-induced calcium release. These results do not, however, exclude that LPA may also inhibit IP₃ production further inhibiting intracellular store release. Precisely how IP₃R activity is inhibited by a LPA₅-Gα12/13-Arhgef1 signaling axis remains to be defined although we note that Gβγ dimers liberated from Gα_iβγ heterotrimers can associate and gate the IP₃R (57) providing precedence for regulation of IP₃R activity by GPCR signaling.

Immunization of LPA₅-deficient and sufficient mice revealed that LPA₅ is able to repress the early primary antibody response to a TI-2 model antigen and presumably in response to physiological levels of LPA. However, LPA₅ is also expressed by other hematopoietic and non-hematopoietic cell types, including T cells (43, 47), which are able to influence TI antibody responses (58). Thus, we further restricted our in vivo evaluation of LPA₅ function to (MZ) B cells with mixed bone marrow chimeras and found that Lpar5^−/− B cells mount a heightened antigen-specific primary antibody response compared to Lpar5^+/+ B cells and in competition with wild type B cells and in the presence of a wild type immune system. As MZ B cells dominate this TI-2 antibody response (32, 49-52) and we show that Lpar5^−/− MZ B cells develop at normal frequencies and localize appropriately, these data provide compelling evidence that LPA₅ signaling by B cells normally inhibits the primary antibody response to TI-2 antigens.

The physiological concentrations of LPA have been variably reportedly in the high nanomolar – low micromolar range (11, 17, 19, 20) and our in vivo findings suggest that B cell antigen receptor signaling and function are normally suppressed by these endogenous levels. On the other hand, our in vitro experiments required LPA at 5-20 μM concentrations to reveal a role for this lipid in regulating BCR signaling. However, we note that with regards to the concentration of LPA that promotes receptor signaling, it has been proposed that systemic LPA levels may be less relevant for LPA receptor signaling than locally produced microenvironmental levels (59, 60). Specifically, based on the ability of autotaxin, the secreted enzyme that generates extracellular LPA from abundantly available LPC, to associate with integrins and the recent autotaxin crystal structure (59, 60), it has been suggested that newly produced LPA is immediately delivered to LPA receptors on cells in the immediate microenvironment (61). Thus, local increased LPA production may have immunosuppressive activities that are not necessarily reflected by systemic concentrations. Furthermore, the elevated systemic LPA levels found in certain pathological settings would almost certainly promote increased LPA₅ signaling by antigen-specific B cells and further impair antibody responses. Clearly, a better understanding is needed of how LPA production is regulated in homeostatic and inflammatory settings.

With regards to lysophospholipid regulation of lymphocytes, the S1P lysophospholipid has received considerable attention for its role in regulating lymphocyte migration during development and homeostatic trafficking (12-16). In contrast, relatively little is understood about how LPA, an additional major lysophospholipid, regulates lymphocyte function although both T and B cells have been reported to express LPA receptors (33, 62, 63). We show B cells express LPA₂ and LPA₅ and consistent with the previous reported expression of LPA₂ on human B cells (7, 33-35) and prior to the identification of LPA₅. Our findings demonstrate that the LPA₅ GPCR negatively regulates BCR signaling and function and identify the release of calcium from intracellular stores as a likely point of inhibition in the BCR signaling pathway. Although it is not yet clear why the LPA lysophospholipid functions to inhibit B cell activation and antibody response, a plausible consequence of this suppression may be to prevent the activation of weak-affinity autoreactive B cells that are known to be selected into the mature peripheral B cell pool. Current experiments are exploring this possibility.

While attention on LPA has focused on its aberrant production by diverse cancer cell types and its ability to promote tumorigenesis (18, 21, 22), our data suggest that certain malignancies might exploit LPA production not only to promote tumorigenesis but also as a mechanism to inhibit adaptive immune responses (53). In the future it will be important to consider that the pathological or therapeutic settings that alter LPA levels or LPA receptor signaling will also likely modulate adaptive humoral immunity.