SLC4A7/NBCn1 maintains intracellular pH to support mTORC1 activity and antibody-secreting cell differentiation Authors: Ke Hu, Yin-Hu Wang, Anthony Y. Tao, Maxwell McDermott, Miki Jishage, Li Zhong, Thomas Boehringer, Marcus J. Hines, Sergei B. Koralov, Ebbe Boedtkjer, Stefan Feske Published: 2026 ABSTRACT Antibody-producing plasmablasts (PB) and plasma cells (PC) are critical for humoral immunity, autoimmunity and vaccine responses. Despite the importance of environmental stressors in regulating humoral immune responses, the influence of pH on PB and PC differentiation remains elusive. Here, we identify SLC4A7/NBCn1, a Na+/HCO3– cotransporter, as a selective regulator of PB differentiation in vitro. SLC4A7 deletion also impairs the formation of antibody secreting cells (ASCs) and antibody responses in mice in vivo following immunization and influenza A virus infection. Mechanistically, SLC4A7 deletion results in intracellular acidification and lysosomal alkalinization, and is associated with impaired function of the mechanistic target of rapamycin complex 1 (mTORC1). Enforcing mTORC1 activation in SLC4A7-deficient B cells or B cells in which intracellular pH is acidified by blocking Na+/H+ exchanger (NHE) function restores PB differentiation in vitro. Moreover, ASC differentiation and antibody responses are impaired under conditions of extracellular acidosis in vitro and in a mouse model of metabolic acidosis. Altogether, we identify a critical relationship between intracellular pH regulation through SLC4A7 and mTORC1-dependent ASC differentiation and humoral immunity. Introduction B cells play crucial roles in immunity against infection, vaccine responses and the pathogenesis of autoimmunity. A subset of activated B cells differentiates into antibody-secreting cells (ASCs), either short-lived plasmablasts (PBs) or more terminally differentiated plasma cells (PCs), which are typically long-lived1,2. Many of the transcriptional regulators and signaling pathways regulating ASC differentiation are well-described. Like other immune cells, B cells use ion channels and transporters (ICTs) to mediate the transport of ions such as Ca2+, Zn2+, Mg2+ and protons (H+) across cell membranes and to modulate cell functions including proliferation, differentiation and cell death3–7. Little is known about the function of ICTs in B cell mediated immunity and specifically their involvement in ASC development. To address this question, we conducted a forward genetic screen to identify ICTs that regulate PB differentiation in vitro. We identified the sodium (Na+)/bicarbonate (HCO3–) cotransporter NBCn1, or SLC4A7, and demonstrate that it is critical for intracellular pH (pHi) homeostasis during PB differentiation and humoral immune responses in vitro and in vivo. The regulation of pHi is important for proper cell function. Changes in pHi affect electrostatic protein-protein interactions and cellular processes such as cytoskeletal stability, organellar quality and cell metabolism8. Several mechanisms exist to tightly regulate pHi for instance in response to changes in extracellular pH (pHe)9,10. pHi is regulated by Na+/H+ exchangers (NHEs) encoded by the SLC9A subfamily, which under typical physiological conditions function as acid extruders by mediating net H+ export and Na⁺ import, and the voltage-gated proton channel Hv1 (encoded by HVCN1) that exports H+ from the cytosol11,12. Hv1 was shown to modulate the strength of B cell receptor (BCR) signaling by regulating the BCR-dependent generation of reactive oxygen species in endosomes and thus, the promotion of antibody production6. Whether Hv1 also influences B cell function by regulating pHi is not known. Another mechanism to control pHi is mediated by carbonic anhydrases (CA). They facilitate equilibration of the mobile CO2/HCO3– buffer13, and are important for delivering H+-equivalents to or removing H+-equivalents from sites of metabolic production, consumption, and transport. HCO3– concentrations in cells and thus, pHi are also regulated by members of the SLC4A subfamily which encode Na⁺/HCO₃– cotransporters (NBC) and Cl–/HCO₃– exchangers (AE). Whereas NBCs typically import HCO₃– together with Na⁺ and thus, alkalinize the cytosol, Cl–/HCO₃– exchangers export intracellular HCO₃⁻ in exchange for extracellular Cl– and thus, causes cytosolic acidification14. The Na+/HCO3– cotransporter NBCn1 (SLC4A7) is located in the plasma membrane and imports both HCO3– and Na+ to prevent cellular acidification15. It was shown to promote neuronal survival16, vasorelaxation17,18, and protection of the gastroduodenal mucosa19. In the human THP1 monocyte cell line, SLC4A7 promotes phagolysosomal acidification through the regulation of pHi20. Whether SLC4A7 is involved in pHi regulation in B cells and thus, B cell mediated immunity is unknown. More generally, the role of pHi homeostasis in B cells and ASCs is poorly understood. B cells, like other immune cells, operate in environments with different pHe which can differ vastly depending on tissue type and is altered in a number of disease states. Lymph nodes (LNs), an important site of B cell differentiation into ASCs, are significantly more acidic than the blood as demonstrated by intravital imaging-based measurements of pHe21. Because pHe influences pHi, it is likely that B cells possess mechanisms to regulate pHi to ensure proper cell function. B cells also encounter acidic pHe under various disease conditions including metabolic acidosis in chronic kidney disease (CKD) or diabetic ketoacidosis22,23, tissue inflammation and ischemia that are associated with hypoxia and lactic acid production24,25 and acidic tumor microenvironments26. The effects of metabolic acidosis on immune function have not been widely investigated in vivo but can be inferred from studies that have analyzed macrophages and T cells under low pH conditions in vitro, which demonstrated strong defects in immune cell activation, cytokine production and metabolism27–29. Whether extracellular acidity affects B cell functions including their differentiation into ASCs and antibody production is not known. Here we demonstrate that SLC4A7 is essential for PB differentiation in vitro and ASC development and antibody production in vivo in response to immunization and infection. Deletion of SLC4A7 in B cells results in intracellular acidification and alkalinization of lysosomes. Altered pHi and lysosomal pH (pHL) were associated with impaired mTORC1 activity, and promoting mTORC1 activity by expression or deletion of mTORC1 regulators restores PB and ASC differentiation in the absence of SLC4A7. The suppressive effects of cellular acidification on humoral immunity were not limited to SLC4A7 deletion, but were also observed following inhibition of NHEs in vitro or extracellular acidification in vitro and in an animal model of metabolic acidosis. Our study demonstrates that regulation of pHi by SLC4A7 is a critical determinant of mTORC1 function and, consequently, ASC development and humoral immune responses. Results A pooled CRISPR screen identifies SLC4A7 as a regulator of PB differentiation To identify ICTs that regulate PB differentiation in vitro, we designed a guide RNA (gRNA) CRISPR library targeting 1132 genes encoding known ICTs and regulators (Fig. 1A). The library also includes 57 negative and 31 positive controls, including known genes that control ASC differentiation such as Prdm1 (encoding BLIMP1) and Irf430,31. We isolated primary B cells from the spleen of Cas9-knock-in mice32 and stimulated them with anti-RP105 antibody33 to promote cell division and facilitate retroviral transduction with the gRNA library (Fig. 1B). RP105 stimulation alone did not induce PB differentiation as measured by CD138 (syndecan-1) expression (Figure S1A). Transduced B cells were then restimulated with 40LB feeder cells to induce B cell differentiation into CD138+ PBs (Figs. 1B and S1B). 40LB cells are NIH3T3 cells expressing CD40L and BAFF, which provide a microenvironment supportive of B cell activation, proliferation and PB differentiation in vitro34. Although in vitro-differentiated PBs are short-lived, the signaling programs driving ASC differentiation are largely conserved between in vitro and in vivo contexts. To identify ICTs that regulate PB differentiation, we sorted CD138+ and CD138‒ B cells for further analysis. In parallel, we conducted a CRISPR screen to identify ICTs that control B cell proliferation. B cells transduced with the CRISPR library were loaded with carboxyfluoroscein succinimidyl ester (CFSE), expanded on 40LB cells and sorted for lowest and highest 30% of CFSE intensity as a measure of cell division and thus, proliferation (Fig. S1B).Fig. 1SLC4A7 controls PB differentiation in vitro.A Composition of the CRISPR library targeting ion channels and transporters (ICTs). B Experimental design of CRISPR screens to identify ICTs regulating plasmablast (PB) differentiation or B cell proliferation in vitro. Cell proliferation was analyzed on day 2, PB differentiation was analyzed on day 4. C Scatterplot of MAGeCK Z-scores illustrating gRNA-mediated ICT depletion or enrichment in B cells, plotted against normalized log10 average (avg.) mRNA expression in PBs (top) or B cells (non-PB) (bottom) from RNA-seq data in this study (GSE269509). D Frequency of CD138+ PBs upon co-culture with 40LB feeder cells. Representative flow cytometry plots (left) and quantification (right). Purified B cells from WT or Slc4a7–/– mice were analyzed 4 days after co-culture with 40LB cells, IL-4, and IL-21 as shown in (B) and gated on DAPI– live cells. E Enzyme-linked immunosorbent assay (ELISA) of IgM secretion by WT or Slc4a7–/– B cells following anti-RP105 activation, 40LB co-culture (IL-4 and IL-21) and PB differentiation in vitro. Culture media was replaced with fresh media on day 2 of PB differentiation and supernatants were collected on day 4 for analysis. F Proliferation of WT and Slc4a7–/– B cells. Representative flow cytometry histograms (left) and frequencies of CFSElo cells (right). Naïve purified B cells were loaded with CFSE and analyzed after co-culture with 40LB cells and IL-4 for 2 days and gated on DAPI- live cells. The shaded histogram represents the CFSE loading control on day 0 without 40LB feeder cells. G Apoptosis of WT and Slc4a7–/– B cells. Naïve purified B cells were stained with DAPI and Annexin-V 4 days after co-culture with 40LB cells, IL-4, and IL-21 as shown in (B) and gated on DAPI− live cells. H Class switch recombination of WT and Slc4a7–/– B cells following 2 days of co-culture with 40LB cells and IL-4. DAPI− live cells were stained with anti-IgM and anti-IgG1 antibodies. Detailed gating strategies for (D–H) are shown in Fig. S9A. Data are from 9 (D–H) mice per cohort from three independent experiments. Each dot represents a biological replicate. Shown are the mean ± SEM. Statistical significance was tested by two-sided unpaired Student’s t-test. P-values are indicated above the bar graphs. Source data are provided as a Source data file. SLC4A7 controls PB differentiation in vitro. A Composition of the CRISPR library targeting ion channels and transporters (ICTs). B Experimental design of CRISPR screens to identify ICTs regulating plasmablast (PB) differentiation or B cell proliferation in vitro. Cell proliferation was analyzed on day 2, PB differentiation was analyzed on day 4. C Scatterplot of MAGeCK Z-scores illustrating gRNA-mediated ICT depletion or enrichment in B cells, plotted against normalized log10 average (avg.) mRNA expression in PBs (top) or B cells (non-PB) (bottom) from RNA-seq data in this study (GSE269509). D Frequency of CD138+ PBs upon co-culture with 40LB feeder cells. Representative flow cytometry plots (left) and quantification (right). Purified B cells from WT or Slc4a7–/– mice were analyzed 4 days after co-culture with 40LB cells, IL-4, and IL-21 as shown in (B) and gated on DAPI– live cells. E Enzyme-linked immunosorbent assay (ELISA) of IgM secretion by WT or Slc4a7–/– B cells following anti-RP105 activation, 40LB co-culture (IL-4 and IL-21) and PB differentiation in vitro. Culture media was replaced with fresh media on day 2 of PB differentiation and supernatants were collected on day 4 for analysis. F Proliferation of WT and Slc4a7–/– B cells. Representative flow cytometry histograms (left) and frequencies of CFSElo cells (right). Naïve purified B cells were loaded with CFSE and analyzed after co-culture with 40LB cells and IL-4 for 2 days and gated on DAPI- live cells. The shaded histogram represents the CFSE loading control on day 0 without 40LB feeder cells. G Apoptosis of WT and Slc4a7–/– B cells. Naïve purified B cells were stained with DAPI and Annexin-V 4 days after co-culture with 40LB cells, IL-4, and IL-21 as shown in (B) and gated on DAPI− live cells. H Class switch recombination of WT and Slc4a7–/– B cells following 2 days of co-culture with 40LB cells and IL-4. DAPI− live cells were stained with anti-IgM and anti-IgG1 antibodies. Detailed gating strategies for (D–H) are shown in Fig. S9A. Data are from 9 (D–H) mice per cohort from three independent experiments. Each dot represents a biological replicate. Shown are the mean ± SEM. Statistical significance was tested by two-sided unpaired Student’s t-test. P-values are indicated above the bar graphs. Source data are provided as a Source data file. We performed next generation sequencing of genomic DNA to analyze the enrichment or depletion of gRNAs targeting unique ICT genes in CD138+ vs CD138– and CFSElo vs CFSEhi cells. ICTs depleted in the CD138+ PB population relative to the CD138‒ non-PB population were considered positive regulators of PB differentiation. Similarly, ICTs depleted in the CFSElo vs the CFSEhi population were considered positive regulators of B cell proliferation and activation. As expected, gRNAs targeting ICTs not relevant for B cell function (negative controls) failed to be enriched or depleted in either screen (Fig. 1C). We identified 101 genes (ICTs and positive controls) whose deletion resulted in depletion of CD138+ cells (Fig. S1C). These included Prdm1, encoding BLIMP1, the master regulator of PB differentiation (Fig. 1C). Deletion of 112 genes resulted in depletion of proliferating CFSElo B cells (Figs. 1C and S1C). 28 genes were shared between the two screens, including the positive controls Eef2 and Irf4 (Figs. 1C and S1C). Among the 73 ICTs selectively depleted in the PB screen and unaffected in the proliferation screen were membrane proteins that regulate the transport of various organic and inorganic ions such as PO43− (Slc25a3), nucleotides (Slc25a25, Slc25a33), Ca2+ (Atp2a2), or Zn2+ (Slc39a7, Slc39a10) (Table S1). Slc4a7 was the top-ranked pH-regulating ICT in the screen. It encodes the Na+/HCO3− cotransporter NBCn1 that is ubiquitously expressed across most B cell subsets with moderately higher mRNA levels in GC centroblasts (GC-CBs), which are precursors of both short- and long-lived PCs and memory B cells, compared to other B cells (Fig. S1D). Expression levels of Slc4a7 were also among the highest in GC-CBs compared to the other ICT CRISPR screen hits (Fig. S1E). SLC4A7 was shown to regulate pHi in several cell types including HeLa cells, breast cancer cells and macrophages20,35,36, but the function of SLC4A7 in B cells, particularly ASCs, and humoral immunity is unknown. Deletion of SLC4A7 selectively impairs PB differentiation in vitro To validate the role of SLC4A7 in PB differentiation, we differentiated B cells isolated from the spleens of Slc4a7–/– mice and littermate controls into PBs by co-culturing them with 40LB cells in vitro. B cells from Slc4a7–/– mice lacked SLC4A7 protein expression (Fig. S2A) resulting in a significant reduction in the frequencies of CD138+ PBs (Fig. 1D) and IRF4+ Pax5– PBs (Fig. S2B) as well as decreased IgM antibody secretion by PBs (Fig. 1E). Of note, SLC4A7 deletion specifically affected terminal differentiation into PBs. Other B cell functions including proliferation (Fig. 1F), viability (Fig. 1G), and class-switch recombination of immunoglobulin heavy chains from IgM to IgG1 (Fig. 1H) were not altered in B cells from Slc4a7–/– mice compared to littermate controls. These findings are consistent with the fact that Slc4a7 was identified in the PB differentiation but not proliferation screen and further suggest that it selectively regulates ASC differentiation. To exclude the possibility that the defect in PB differentiation of B cells from Slc4a7–/– mice results from an earlier defect in B cell development, we analyzed immature and mature B cell populations in the bone marrow, peritoneal cavity and spleens of Slc4a7‒/‒ mice (ref. 37). The frequencies of pro-B, pre-B and immature B cells in the BM (Fig. S2C), B1, B2, follicular and marginal zone B cells in the spleen (Fig. S2D), and B1, B1a, B1b and B2 cells in the peritoneal cavity (Fig. S2E) were similar in Slc4a7‒/‒ mice and littermate controls. Collectively, these data indicate that SLC4A7 is critical for PB differentiation in vitro. To further investigate if SLC4A7 regulates the transition of B cells into PBs, we deleted it by CRISPR/Cas9 gene editing in anti-RP105-stimulated B cells using gRNAs targeting Slc4a7 (gSlc4a7). After two days to permit gene editing and protein knockdown, we co-cultured the transduced B cells with 40LB. SLC4A7 protein expression was almost completely abolished in gSlc4a7-transduced B cells as confirmed by immunoblotting using antibodies targeting N- and C-terminal domains of SLC4A7 (Fig. S3A, B). Deletion of SLC4A7 and co-culture with 40LB cells resulted in significantly reduced frequencies of CD138+ (Fig. S3C) or IRF4+Pax5– (Fig. S3D) PBs. A similar defect in the differentiation of CD138+ PBs was observed by deletion of Prdm1 as a positive control (Fig. S3E). Indeed, BLIMP1 protein expression was decreased in gSlc4a7-transduced PB compared to gCtrl (Fig. S3F). We also observed reduced IgM antibody secretion by SLC4A7-deficient PBs (Figure S3G). By contrast, the proliferation, viability and class switch recombination of gSlc4a7 transduced B cells were not impaired compared to controls (Fig. S3H–J). The failure of gSlc4a7-transduced B cells to differentiate into CD138+ PBs was not limited to 40LB co-culture conditions, but was also observed after CpG and LPS stimulation in vitro (Fig. S3K, L). By contrast, the proliferation of CpG or LPS + IL-4 stimulated SLC4A7 deficient cells was preserved (Fig. S3M, N). To distinguish between a bona fide defect in PB differentiation as opposed to reduced survival of PBs, which could potentially explain reduced PB frequencies in SLC4A7-deficient B cells, we sorted CD138+ PBs and CD138‒ non-PBs that had been differentiated in vitro and co-cultured them with 40LB cells for another 24 h (Fig. S3O). The numbers of CD138+ PBs and CD138‒ non-PBs (Fig. S3P) and the relative frequencies of live, pre-apoptotic and apoptotic B cells (Fig. S3Q) were comparable in SLC4A7 deficient and control cells. Lastly, we investigated the effects of pharmacologic inhibition of SLC4A7 activity on PB differentiation in vitro. S0859 is a generic inhibitor of Na+/HCO3‒ transporters, which has previously been leveraged to study SLC4A715. Incubation of B220+ B cells from WT mice with 50 μM S0859 for 4 days during PB differentiation with 40LB significantly reduced the frequencies of CD138+ PBs without causing increased cell death (Fig. S3R, S). S0859 did not further decrease the numbers of SLC4A7-deficient PBs compared to untreated controls (Fig. S3R). Although S0859 is not a selective blocker15, these data suggest that its effects in B cells are likely mediated by SLC4A7. Collectively, our data demonstrate that SLC4A7 selectively regulates PB differentiation in vitro without affecting other important B cell functions. SLC4A7 is required for ASC differentiation after influenza infection and immunization in vivo We next sought to determine the role of SLC4A7 in the differentiation of B cells into ASCs in vivo and thus, humoral immune response to immunization and infection. We first infected Slc4a7‒/‒ and WT mice with the laboratory-adapted X31 H3N2 strain of influenza A virus (IAV) (Fig. 2A). SLC4A7-deficient mice had normal frequencies of B220+ B cells (Fig. 2B) but significantly reduced numbers of CD138+ TACI+ ASCs (Fig. 2C)38, CD138+ B220lo ASCs (Figure S4A) and IRF4+ Pax5– ASCs (Fig. S4B) in their mediastinal lymph nodes at 14 days post-infection compared to littermate controls. Slc4a7‒/‒ mice also had significantly decreased serum levels of X31-specific IgM and IgG antibodies (Fig. 2D) and reduced numbers of ASCs secreting X31-specific IgG antibodies in their mediastinal LNs compared to controls (Fig. 2E). The frequencies of GL-7+ Fas+ GC B cells were only moderately reduced in SLC4A7-deficient mice (Fig. 2F). As a result, the ratio of ASCs to GC B cells was significantly reduced in Slc4a7‒/‒ mice compared to controls (Fig. S4C). Of note, the frequencies of CXCR5+ PD-1+ T follicular helper (Tfh) cells remained unchanged (Fig. 2G).Fig. 2SLC4A7 regulates ASC differentiation during influenza infection.A Experimental design for infection with the X31 (H3N2) strain of influenza A virus (IAV). B, C Analysis of B cells and ASCs isolated from the mediastinal lymph nodes (medLNs) of Slc4a7‒/‒ or WT mice 14 days after intranasal challenge with X31 influenza or treatment with PBS. B Frequencies of B220+CD138– B cells. C Frequencies of CD138+TACI+ ASCs. Gating strategies are shown in Fig. S10A. D Titers of X31-specific IgM (left) and IgG (right) antibodies in the serum of Slc4a7‒/‒ or WT mice measured by ELISA and absorbance (Abs.) at 405 nm. Dotted lines and open circles represent non-infected controls. E Detection of ASCs secreting X31-specific IgG antibodies. Cells isolated from the medLNs of Slc4a7‒/‒ or WT mice at day 14 after X31 infection were analyzed by enzyme-linked immunosorbent spot (ELISPOT) assay using plates coated with X31 viral particles. Representative images (left) and quantification (right). PBS control indicates mice that were challenged with PBS intranasally. Germinal center B cells (B220+Fas+GL-7+, in F) and T follicular helper cells (CD4+B220−CD44+Foxp3−CXCR5+PD-1+, in G) in medLNs of Slc4a7‒/‒ or WT mice 14 days after X31 infection. Representative flow cytometry plots (left) and quantification (right). Representative gating strategies are shown in Fig. S10A. H Experimental design to generate bone marrow chimeric (BMC) mice with B cell-specific deletion of Slc4a7. muMT–/–, immunoglobulin mu heavy chain (IgM) deficient mice. I Frequencies of CD138+B220lo ASCs in medLNs of WT and Slc4a7–/– BMCs, quantified relative to total live lymphocytes on day 8 after X31 infection. Representative flow cytometry plots (left) and quantification (right) of CD138+B220lo ASCs. Gating strategies are shown in Fig. S10B. Data are from 8 to 10 mice (B; WT X31, n = 10; Slc4a7‒/‒ X31, n = 8), 4–10 mice (C, F, G; WT X31, n = 10; Slc4a7‒/‒ X31, n = 8; WT PBS, n = 4; Slc4a7‒/‒ PBS, n = 4), 8–12 mice (D; WT, n = 12; Slc4a7‒/‒, n = 8), 8–10 mice (E; WT, n = 10; Slc4a7‒/‒, n = 8), and 10–12 mice (I; WT, n = 12; Slc4a7‒/‒, n = 10) from at least two independent experiments. Each dot represents one biological replicate. Data are shown as mean ± SEM. Statistical significance was determined by two-sided unpaired Student’s t-test (B–D, F, G, I), or two-way ANOVA (E). P values are indicated above the graphs. Source data are provided as a Source data file. SLC4A7 regulates ASC differentiation during influenza infection. A Experimental design for infection with the X31 (H3N2) strain of influenza A virus (IAV). B, C Analysis of B cells and ASCs isolated from the mediastinal lymph nodes (medLNs) of Slc4a7‒/‒ or WT mice 14 days after intranasal challenge with X31 influenza or treatment with PBS. B Frequencies of B220+CD138– B cells. C Frequencies of CD138+TACI+ ASCs. Gating strategies are shown in Fig. S10A. D Titers of X31-specific IgM (left) and IgG (right) antibodies in the serum of Slc4a7‒/‒ or WT mice measured by ELISA and absorbance (Abs.) at 405 nm. Dotted lines and open circles represent non-infected controls. E Detection of ASCs secreting X31-specific IgG antibodies. Cells isolated from the medLNs of Slc4a7‒/‒ or WT mice at day 14 after X31 infection were analyzed by enzyme-linked immunosorbent spot (ELISPOT) assay using plates coated with X31 viral particles. Representative images (left) and quantification (right). PBS control indicates mice that were challenged with PBS intranasally. Germinal center B cells (B220+Fas+GL-7+, in F) and T follicular helper cells (CD4+B220−CD44+Foxp3−CXCR5+PD-1+, in G) in medLNs of Slc4a7‒/‒ or WT mice 14 days after X31 infection. Representative flow cytometry plots (left) and quantification (right). Representative gating strategies are shown in Fig. S10A. H Experimental design to generate bone marrow chimeric (BMC) mice with B cell-specific deletion of Slc4a7. muMT–/–, immunoglobulin mu heavy chain (IgM) deficient mice. I Frequencies of CD138+B220lo ASCs in medLNs of WT and Slc4a7–/– BMCs, quantified relative to total live lymphocytes on day 8 after X31 infection. Representative flow cytometry plots (left) and quantification (right) of CD138+B220lo ASCs. Gating strategies are shown in Fig. S10B. Data are from 8 to 10 mice (B; WT X31, n = 10; Slc4a7‒/‒ X31, n = 8), 4–10 mice (C, F, G; WT X31, n = 10; Slc4a7‒/‒ X31, n = 8; WT PBS, n = 4; Slc4a7‒/‒ PBS, n = 4), 8–12 mice (D; WT, n = 12; Slc4a7‒/‒, n = 8), 8–10 mice (E; WT, n = 10; Slc4a7‒/‒, n = 8), and 10–12 mice (I; WT, n = 12; Slc4a7‒/‒, n = 10) from at least two independent experiments. Each dot represents one biological replicate. Data are shown as mean ± SEM. Statistical significance was determined by two-sided unpaired Student’s t-test (B–D, F, G, I), or two-way ANOVA (E). P values are indicated above the graphs. Source data are provided as a Source data file. To determine if the effects of SLC4A7 deletion on ASCs were intrinsic to B cells, we generated bone marrow (BM) chimeric mice in which Slc4a7 is deleted in the B cell compartment. BM cells from WT or Slc4a7‒/‒ mice were mixed with BM from B cell-deficient muMT‒/‒ mice at a 20:80 ratio to selectively delete SLC4A7 in B cells (Fig. 2H). We confirmed that Slc4a7 mRNA levels are significantly decreased in sorted B cells from Slc4a7‒/‒ BM chimeras compared to WT BM chimeras (Figure S4D). At day 8 after infection with X31 IAV, BMC mice had reduced numbers of CD138+ B220lo ASCs in their mediastinal lymph nodes (Fig. 2I). Decreased ASC frequencies were accompanied by lower serum levels of IAV-specific IgM and IgG antibodies (Fig. S4E) and exacerbated pulmonary inflammation in Slc4a7‒/‒ BM chimeric mice compared to controls (Fig. S4F, G). These findings are consistent with a B cell intrinsic role of SLC4A7 in ASC differentiation and humoral immunity. For an in-depth analysis of SLC4A7 effects on T-dependent antibody responses, we immunized WT and Slc4a7‒/‒ BM chimeric mice lacking SLC4A7 specifically in B cells with ovalbumin protein conjugated to 4-hydroxy-3-nitrophenylacetyl (NP-OVA) by intraperitoneal injection. Splenocytes were analyzed at days 4 and 14 post-immunization (p.i.) (Fig. 3A). Slc4a7‒/‒ BM chimeras had significantly reduced frequencies of CD138+ TACI+ (Fig. 3B) and IRF4+ PAX5‒ (Fig. S5A) ASCs at both days 4 and 14 p.i. compared to control chimeras. SLC4A7-deficient BM chimeric mice also had significantly reduced titers of NP-specific IgM and IgG antibodies in their serum compared to controls (Figs. 3C and S5B). The titers of both high-affinity NP4 and low-affinity NP25 specific antibodies were reduced in Slc4a7‒/‒ BM chimeras (Fig. 3C). Importantly, the numbers of ASCs secreting NP-specific IgG antibodies in the spleen on day 14 were also significantly reduced in Slc4a7‒/‒ BM chimeras compared to controls (Fig. 3D). By contrast, the frequencies of GL7+ Fas+ GC B cells were comparable in SLC4A7-deficient and control mice (Fig. 3E) consistent with a defect in ASC differentiation rather than in GC formation per se. The ratios of ASCs to GC B cells, however, were not significantly reduced on either day 4 or day 14 p.i. (Fig. S5C). As expected, the frequencies of CXCR5+ PD1+ T follicular helper (Tfh) cells were unchanged in the Slc4a7‒/‒ BM chimeras compared to controls (Fig. S5D).Fig. 3SLC4A7 regulates ASC differentiation after NP-OVA immunization.A Experimental design for NP-OVA immunization of Slc4a7‒/‒ and WT mixed bone marrow chimeric (BMC) mice. B Frequencies of CD138+TACI+ ASCs in spleens from Slc4a7‒/‒ or WT BMC mice on days 4 and 14 post-immunization with NP-OVA. Representative flow cytometry plots (left) and quantification of repeat experiments (right). Gating strategies are shown in Fig. S10C. C NP25-BSA and NP4-BSA-specific IgM (left) and IgG (right) antibodies in the serum of Slc4a7‒/‒ or WT BMC mice. Shown are absorbances (Abs.) at 1:180 (IgM) and 1:60 (IgG) dilution in the linear range of the ELISA titration curve. D Detection of ASCs secreting NP-specific IgG antibodies. Cells isolated from the spleens of Slc4a7‒/‒ or WT BMC mice at day 14 after NP-OVA immunization were analyzed by ELISPOT assay using NP25-BSA coated plates. Representative images (left) and quantification (right). E Frequencies of GL7+Fas+ germinal center (GC) B cells in the spleen of Slc4a7‒/‒ and WT BMC mice on days 4 and 14 post-immunization with NP-OVA. Representative flow cytometry plots (left) and quantification of repeat experiments (right). Gating strategies are shown in Fig. S10C. F Experimental design of NP-Ficoll immunization of Slc4a7‒/‒ and WT BMC mice. G Frequencies of CD45+CD138+B220loFSChi ASCs in the spleen of NP-Ficoll immunized BMC mice on day 14 post-immunization. Representative flow cytometry plots and quantification of repeat experiments (right). Gating strategies are shown in Fig. S10D. H Titers of NP25-specific IgM antibodies in the serum of mice on day 14 post-immunization measured by ELISA. Shown is the absorbance (Abs.) at 1:500 serum dilution based on a titration curve with 1:20, 1:100, 1:500, and 1:2500 dilutions. Data are from 8 to 9 mice (B, E; day 4 WT, n = 8; day 4 Slc4a7‒/‒, n = 8; day 14 WT, n = 9; day 14 Slc4a7‒/‒, n = 9), 9 mice (C, D; WT, n = 9; Slc4a7‒/‒, n = 9), 7 mice (G; WT, n = 7; Slc4a7‒/‒, n = 7), and 6–7 mice (H; WT, n = 6; Slc4a7‒/‒, n = 7) from at least two independent experiments. Each dot represents one biological replicate. Data are shown as mean ± SEM from at least two independent experiments. Statistical significance was assessed by two-sided unpaired Student’s t-test (B, C, E, G, H), or two-way ANOVA (D). P values are indicated above the graphs. Source data are provided as a Source data file. SLC4A7 regulates ASC differentiation after NP-OVA immunization. A Experimental design for NP-OVA immunization of Slc4a7‒/‒ and WT mixed bone marrow chimeric (BMC) mice. B Frequencies of CD138+TACI+ ASCs in spleens from Slc4a7‒/‒ or WT BMC mice on days 4 and 14 post-immunization with NP-OVA. Representative flow cytometry plots (left) and quantification of repeat experiments (right). Gating strategies are shown in Fig. S10C. C NP25-BSA and NP4-BSA-specific IgM (left) and IgG (right) antibodies in the serum of Slc4a7‒/‒ or WT BMC mice. Shown are absorbances (Abs.) at 1:180 (IgM) and 1:60 (IgG) dilution in the linear range of the ELISA titration curve. D Detection of ASCs secreting NP-specific IgG antibodies. Cells isolated from the spleens of Slc4a7‒/‒ or WT BMC mice at day 14 after NP-OVA immunization were analyzed by ELISPOT assay using NP25-BSA coated plates. Representative images (left) and quantification (right). E Frequencies of GL7+Fas+ germinal center (GC) B cells in the spleen of Slc4a7‒/‒ and WT BMC mice on days 4 and 14 post-immunization with NP-OVA. Representative flow cytometry plots (left) and quantification of repeat experiments (right). Gating strategies are shown in Fig. S10C. F Experimental design of NP-Ficoll immunization of Slc4a7‒/‒ and WT BMC mice. G Frequencies of CD45+CD138+B220loFSChi ASCs in the spleen of NP-Ficoll immunized BMC mice on day 14 post-immunization. Representative flow cytometry plots and quantification of repeat experiments (right). Gating strategies are shown in Fig. S10D. H Titers of NP25-specific IgM antibodies in the serum of mice on day 14 post-immunization measured by ELISA. Shown is the absorbance (Abs.) at 1:500 serum dilution based on a titration curve with 1:20, 1:100, 1:500, and 1:2500 dilutions. Data are from 8 to 9 mice (B, E; day 4 WT, n = 8; day 4 Slc4a7‒/‒, n = 8; day 14 WT, n = 9; day 14 Slc4a7‒/‒, n = 9), 9 mice (C, D; WT, n = 9; Slc4a7‒/‒, n = 9), 7 mice (G; WT, n = 7; Slc4a7‒/‒, n = 7), and 6–7 mice (H; WT, n = 6; Slc4a7‒/‒, n = 7) from at least two independent experiments. Each dot represents one biological replicate. Data are shown as mean ± SEM from at least two independent experiments. Statistical significance was assessed by two-sided unpaired Student’s t-test (B, C, E, G, H), or two-way ANOVA (D). P values are indicated above the graphs. Source data are provided as a Source data file. To exclude the possibility that deletion of SLC4A7 alters the proliferation or survival of B cells in vivo following immunization, we measured the frequencies of Ki67+ (Fig. S5E) and Annexin-V+ (Fig. S5F) B cells in the spleen on day 4 and 14 p.i. but did not observe significant differences between Slc4a7‒/‒ and control BM chimeras. Moreover, the deletion of SLC4A7 had no significant effects on the frequencies of non-class-switched IgM+ (Fig. S5G) and class-switched IgG1+ or IgA+ (Fig. S5H) B cells. Together, these findings indicate that SLC4A7 selectively promotes the formation of ASCs in a B cell-intrinsic manner, without affecting ASC survival, proliferation or antibody isotype switching. To determine whether SLC4A7 is required for ASC differentiation in the context of a T-independent immune response, we immunized Slc4a7‒/‒ and WT BM chimeric mice with NP-Ficoll (Fig. 3F). At 14 days p.i., SLC4A7-deficient BM chimeras had significantly reduced frequencies of ASCs in their spleens (Fig. 3G) and decreased serum levels of anti-NP25 IgM antibodies (Fig. 3H). Collectively, these data demonstrate that SLC4A7 promotes ASC differentiation and antibody production in response to both T-dependent and T-independent B cell responses. SLC4A7 mediates PB differentiation by regulating intracellular pH (pHi) SLC4A7 is located in the plasma membrane in other cell types and maintains an intracellular pH (pHi) homeostasis through the Na+-dependent import of bicarbonate (HCO3−)15,39. SLC4A7 tagged with hemagglutinin (HA) and ectopically expressed in B cells was predominantly found in or near the plasma membrane (Fig. 4A). To determine whether SLC4A7 regulates pHi in B cells, we expressed the genetically-encoded pH indicator pHluorin240 in B cells and differentiated them into PBs in vitro. Compared to wildtype controls, CD138– non-PBs and CD138+ PBs from Slc4a7–/– mice showed a significant decrease of pHi (Figs. 4B, C and S6A). To confirm that SLC4A7 regulates PB differentiation by maintaining a neutral pHi, we used another method to acidify B cells by inhibiting Na+/H+ exchangers (NHEs) in B cells. NHEs mediate the export of protons in exchange for Na+ import and NHE inhibition thus results in cellular acidification by proton retention11. B cells express transcripts of several NHE homologues, including Slc9a1 and Slc9a8, suggesting functional redundancy and potentially explaining why no NHE-encoding genes were identified in our CRISPR screens. We therefore inhibited NHE function pharmacologically using 5-(N-ethyl-N-isopropyl)-amiloride (EIPA), a non-selective blocker of multiple NHEs41. As expected, EIPA treatment reduced pHi in 40LB-stimulated B cells in vitro (Fig. 4D) without significantly affecting B cell apoptosis or cell cycle progression (Fig. S6B, C). Importantly, EIPA treatment significantly reduced the differentiation of 40LB-stimulated B cells into CD138+ PBs in vitro (Fig. 4E). EIPA also significantly reduced IgM antibody secretion by PBs in vitro (Fig. 4F). These findings indicate that intracellular acidification of B cells either by SLC4A7 deletion or NHE inhibition suppresses PB differentiation and antibody production in vitro.Fig. 4SLC4A7 promotes PB differentiation by regulating intracellular pH.A Representative immunofluorescence image of a murine B cell ectopically expressing HA-SLC4A7 after 24 h of CpG stimulation. HA-SLC4A7 was detected with PE-conjugated anti-HA antibody (red), and the nucleus was counterstained with DAPI (blue). Scale bar, 5 μm. Intracellular pH (pHi) of B cells from Slc4a7−/− and WT mice co-cultured with 40LB cells, IL-4 and IL-21 for 4 days. pHi was measured using pHluorin2 fluorescence at day 4. CD138– non-PBs (B) and CD138+ PBs (C). pHluorin2 fluorescence was calibrated to pHi using the calibration curve shown in Fig. S6A. D–F Effects of Na+/H+ exchanger (NHE) inhibition with EIPA on PB differentiation by co-culturing with 40LB cells, IL-4 and IL-21 for 4 days in vitro. D pHi in total B cells treated with the indicated concentrations of EIPA for 4 days and analyzed with pHluorin2. E Frequencies of CD138+ PBs. B cells were treated with DMSO or with indicated concentrations of EIPA for 4 days during 40LB co-culture. F IgM secretion by PBs in vitro following CpG stimulation and treatment with EIPA or DMSO for 2 days. Culture medium was collected for ELISA. G pHi in B cells from WT or Slc4a7−/− mice after overexpression of carbonic anhydrase 2 (CA2) or empty vector (EV) and 40LB co-culture for 4 days. pHi was measured using pHluorin2. Gated on DAPI−Annexin-V− live cells. H Frequencies of CD138+ PBs after expression of CA2 or empty vector (EV) in WT or Slc4a7−/− B cells on day 4 of 40LB co-culture. Gated on DAPI−Annexin-V- live cells. I Frequencies of CD138+ PBs after expression of CA2 or empty vector (EV) in WT B cells co-cultured for 4 days with 40LB cells and treated with or without EIPA. Gated on DAPI−Annexin-V− live cells. Data are from 9 mice (B, C, G–I), 6 mice (D, F), and 3–9 mice (E; 0 µM EIPA, n = 9; 2 µM EIPA, n = 3; 5 µM EIPA, n = 9; 10 µM EIPA, n = 6) per cohort. Each dot represents a biological replicate. Shown are the mean ± SEM. Statistical significance was tested by two-sided unpaired Student’s t-test (B, C, D, F), or one-way ANOVA with Bonferroni correction (E, H, I). P values are indicated above the graphs. Source data are provided as a Source data file. SLC4A7 promotes PB differentiation by regulating intracellular pH. A Representative immunofluorescence image of a murine B cell ectopically expressing HA-SLC4A7 after 24 h of CpG stimulation. HA-SLC4A7 was detected with PE-conjugated anti-HA antibody (red), and the nucleus was counterstained with DAPI (blue). Scale bar, 5 μm. Intracellular pH (pHi) of B cells from Slc4a7−/− and WT mice co-cultured with 40LB cells, IL-4 and IL-21 for 4 days. pHi was measured using pHluorin2 fluorescence at day 4. CD138– non-PBs (B) and CD138+ PBs (C). pHluorin2 fluorescence was calibrated to pHi using the calibration curve shown in Fig. S6A. D–F Effects of Na+/H+ exchanger (NHE) inhibition with EIPA on PB differentiation by co-culturing with 40LB cells, IL-4 and IL-21 for 4 days in vitro. D pHi in total B cells treated with the indicated concentrations of EIPA for 4 days and analyzed with pHluorin2. E Frequencies of CD138+ PBs. B cells were treated with DMSO or with indicated concentrations of EIPA for 4 days during 40LB co-culture. F IgM secretion by PBs in vitro following CpG stimulation and treatment with EIPA or DMSO for 2 days. Culture medium was collected for ELISA. G pHi in B cells from WT or Slc4a7−/− mice after overexpression of carbonic anhydrase 2 (CA2) or empty vector (EV) and 40LB co-culture for 4 days. pHi was measured using pHluorin2. Gated on DAPI−Annexin-V− live cells. H Frequencies of CD138+ PBs after expression of CA2 or empty vector (EV) in WT or Slc4a7−/− B cells on day 4 of 40LB co-culture. Gated on DAPI−Annexin-V- live cells. I Frequencies of CD138+ PBs after expression of CA2 or empty vector (EV) in WT B cells co-cultured for 4 days with 40LB cells and treated with or without EIPA. Gated on DAPI−Annexin-V− live cells. Data are from 9 mice (B, C, G–I), 6 mice (D, F), and 3–9 mice (E; 0 µM EIPA, n = 9; 2 µM EIPA, n = 3; 5 µM EIPA, n = 9; 10 µM EIPA, n = 6) per cohort. Each dot represents a biological replicate. Shown are the mean ± SEM. Statistical significance was tested by two-sided unpaired Student’s t-test (B, C, D, F), or one-way ANOVA with Bonferroni correction (E, H, I). P values are indicated above the graphs. Source data are provided as a Source data file. To demonstrate a causal relationship between pHi and PB differentiation, we sought to reverse the effects of SLC4A7 deletion on PB differentiation by ectopically expressing carbonic anhydrase 2 (CA2). CAs are a family of enzymes that catalyze the interconversion of HCO3‒ and CO2. CA inhibition typically promotes cytosolic acidification whereas CA overexpression results in cytosolic alkalinization42,43. These effects are most likely explained by the ability of CAs to accelerate the extrusion of metabolically generated acids from the cell. We found that ectopic expression of CA2 reversed the cellular acidification in B cells from Slc4a7–/– mice to levels comparable to those in WT B cells (Fig. 4G) and completely rescued the PB differentiation defect of Slc4a7–/– B cells (Fig. 4H). CA2 overexpression also partially protected WT B cells from the suppressive effects of EIPA-mediated NHE inhibition on PB differentiation (Fig. 4I). Collectively, these findings provide strong evidence for a causal link between cellular acidification and the suppression of PB differentiation in vitro. SLC4A7 facilitates PB differentiation by regulating mTORC1 function Many signaling pathways and cell functions can be affected by changes in pHi9,29,44, and we therefore sought to investigate the mechanisms underlying the effects of cellular acidification in the absence of SLC4A7 in B cells. To this end, we differentiated WT and Slc4a7‒/‒ B cells into PBs and analyzed gene expression patterns in sorted CD138+ PBs and CD138‒ non-PBs by RNA sequencing. Deletion of SLC4A7 resulted in hundreds of differentially expressed genes (DEGs) compared to WT controls in both PBs and non-PBs (Fig. 5A, B). Of note, we observed a significant upregulation of ICTs known to export H+ from the cytosol (Hvcn1 and Slc16a3) or import HCO3– (Slc4a8) in SLC4A7-deficient ASCs (Figure S6D), potentially as a compensatory mechanism to counter cellular acidification. Our pathway analyses of PBs showed a strong depletion of plasma cell-associated gene signatures in SLC4A7-deficient cells (Fig. 5C). This included the altered expression of positive regulators of ASC differentiation including Prdm1 (BLIMP1), Irf4, Xbp1, Stat3 or Eif2ak3 (PERK) and that of negative regulators such as Bcl6, Pax5, Bach2, or Id3 (Fig. 5D). We surmised that the dysregulation of these genes is reflective of a PB identity defect rather than a direct regulation of these genes by pHi downstream of SLC4A7. Subsequent analyses of DEGs identified a significant depletion of pathways in SLC4A7-deficient PBs that are related to N-linked glycosylation, DNA replication, protein translation and ribosome biogenesis (Fig. 5E). An upstream regulator analysis predicted the transcription factor (TF) MYC to potentially be responsible for differential gene expression in SLC4A7-deficient PBs (Fig. 5F). GSEA confirmed the negative enrichment of MYC target genes in PBs derived from Slc4a7−/− mice (Figs. 5G and S6E) suggesting decreased MYC function. Because MYC regulates protein translation and cell growth45, which are essential for ASC differentiation, we measured global protein translation using the SUnSET46 assay. PB were pulsed with puromycin and the incorporation of puromycin into nascent polypeptides was measured using an anti-puromycin antibody. Consistent with reduced MYC function or expression, we found that SLC4A7-deficient PBs have moderately, but significantly reduced puromycin incorporation (Fig. 5H), suggesting impaired protein translation.Fig. 5SLC4A7 facilitates PB differentiation by regulating mTORC1.A Volcano plots of differentially expressed genes (DEG) identified by RNA-sequencing in CD138– non-PBs (top) and CD138+ PBs (bottom) from Slc4a7−/− vs. WT B cells after 4 days of co-culture with 40LB feeder cells. DEGs were selected based on a false discovery rate (FDR) < 0.01 and fold change >1.5. B Venn diagram showing unique and overlapping DEGs in non-PBs and PBs. C GSEA plot showing depletion of the plasma cell signature from GSE12366 in Slc4a7‒/‒ and WT PBs. D Heatmap of selected genes associated with plasma cell differentiation. Z-scores were calculated using the normalized average expression per sample across all samples. E Top 15 pathways depleted in Slc4a7−/− vs. WT PBs, based on Gene Ontology (GO Biological Process analysis. F Top 10 Cancer and Genetic perturbation signatures enriched among downregulated genes in Slc4a7−/− vs. WT PBs. G GSEA plot of depleted MYC targets signature in Slc4a7−/− vs. WT PBs using the hallmark database. H Protein translation using the SUnSET assay in Slc4a7−/− and WT B cells following 40LB co-culture for 3 days. Representative flow cytometry histograms of puromycin incorporation (left) and quantification of mean fluorescence intensity (MFI) from repeat experiments (right). I GSEA plot of depleted mTORC1 signature in Slc4a7−/− vs. WT PBs from the LINCS L1000 project database. MFI of phospho-S6 (J) or phospho-mTORC1 (K) analyzed by flow cytometry in Slc4a7−/− and WT B cells co-cultured with 40LB feeder cells for 3 days. Representative flow cytometry histograms (left) and quantification of MFI from repeat experiments (right). Gated on DAPI- live cells. L Schematic illustration of the AKT-dependent pathway of mTORC1 activation. Frequencies of CD138+ PBs after ectopic myrAKT (M) or RHEBQ64L (N) expression in WT or Slc4a7−/− B cells followed by co-culture with 40LB cells for 4 days to induce PB differentiation. Cells gated on DAPI– live cells. O Frequencies of CD138+ PBs after ectopic expression of RHEBQ64L in WT B cells co-cultured with 40LB cells and treated with EIPA or DMSO for 4 days. Cells gated on DAPI– live cells. Pathway significance in (C, E–G, I) was assessed by permutation-based empirical testing of the enrichment score using a Kolmogorov–Smirnov-like statistic; tests in (E, F) were two-sided. Multiple-testing correction was performed using the Benjamini–Hochberg method (A–C, E–G, I). Unless otherwise indicated, all data are from B cells co-cultured with 40LB feeder cells for 4 days. Gating strategies for data in (H, J, K, M–O) are shown in Fig. S11A, B. Data are from 3 mice per cohort (A–G, I), 9 mice (H, J–N), and 6 mice (O) from 2 to 3 independent experiments. Each dot represents one biological replicate. Data are shown as mean ± SEM. Statistical significance was assessed by two-sided unpaired Student’s t-test (H, J, K) and one-way ANOVA with Bonferroni correction for (M–O). P values are indicated above the graphs. Source data are provided as a Source data file. SLC4A7 facilitates PB differentiation by regulating mTORC1. A Volcano plots of differentially expressed genes (DEG) identified by RNA-sequencing in CD138– non-PBs (top) and CD138+ PBs (bottom) from Slc4a7−/− vs. WT B cells after 4 days of co-culture with 40LB feeder cells. DEGs were selected based on a false discovery rate (FDR) < 0.01 and fold change >1.5. B Venn diagram showing unique and overlapping DEGs in non-PBs and PBs. C GSEA plot showing depletion of the plasma cell signature from GSE12366 in Slc4a7‒/‒ and WT PBs. D Heatmap of selected genes associated with plasma cell differentiation. Z-scores were calculated using the normalized average expression per sample across all samples. E Top 15 pathways depleted in Slc4a7−/− vs. WT PBs, based on Gene Ontology (GO Biological Process analysis. F Top 10 Cancer and Genetic perturbation signatures enriched among downregulated genes in Slc4a7−/− vs. WT PBs. G GSEA plot of depleted MYC targets signature in Slc4a7−/− vs. WT PBs using the hallmark database. H Protein translation using the SUnSET assay in Slc4a7−/− and WT B cells following 40LB co-culture for 3 days. Representative flow cytometry histograms of puromycin incorporation (left) and quantification of mean fluorescence intensity (MFI) from repeat experiments (right). I GSEA plot of depleted mTORC1 signature in Slc4a7−/− vs. WT PBs from the LINCS L1000 project database. MFI of phospho-S6 (J) or phospho-mTORC1 (K) analyzed by flow cytometry in Slc4a7−/− and WT B cells co-cultured with 40LB feeder cells for 3 days. Representative flow cytometry histograms (left) and quantification of MFI from repeat experiments (right). Gated on DAPI- live cells. L Schematic illustration of the AKT-dependent pathway of mTORC1 activation. Frequencies of CD138+ PBs after ectopic myrAKT (M) or RHEBQ64L (N) expression in WT or Slc4a7−/− B cells followed by co-culture with 40LB cells for 4 days to induce PB differentiation. Cells gated on DAPI– live cells. O Frequencies of CD138+ PBs after ectopic expression of RHEBQ64L in WT B cells co-cultured with 40LB cells and treated with EIPA or DMSO for 4 days. Cells gated on DAPI– live cells. Pathway significance in (C, E–G, I) was assessed by permutation-based empirical testing of the enrichment score using a Kolmogorov–Smirnov-like statistic; tests in (E, F) were two-sided. Multiple-testing correction was performed using the Benjamini–Hochberg method (A–C, E–G, I). Unless otherwise indicated, all data are from B cells co-cultured with 40LB feeder cells for 4 days. Gating strategies for data in (H, J, K, M–O) are shown in Fig. S11A, B. Data are from 3 mice per cohort (A–G, I), 9 mice (H, J–N), and 6 mice (O) from 2 to 3 independent experiments. Each dot represents one biological replicate. Data are shown as mean ± SEM. Statistical significance was assessed by two-sided unpaired Student’s t-test (H, J, K) and one-way ANOVA with Bonferroni correction for (M–O). P values are indicated above the graphs. Source data are provided as a Source data file. Protein translation critically depends on the function of mTORC147. Because mTORC1 and its upstream regulators are known to be pH-sensitive48,49, we hypothesized that SLC4A7 deletion may affect the activity of mTORC1 and thus, protein translation in differentiating PBs. GSEA showed a significant dysregulation of mTORC1 target genes in SLC4A7-deficient PBs compared to WT controls (Figs. 5I and S6F). To investigate directly if mTORC1 function is impaired in SLC4A7-deficient B cells, we measured the phosphorylation of ribosomal protein S6, a downstream effector of mTORC1 activity. S6 phosphorylation was significantly reduced in PB derived from Slc4a7‒/‒ mice compared to littermate controls (Fig. 5J). A similar reduction in S6 phosphorylation was found in B cells transduced with gSlc4a7 and differentiated into PB in vitro (Fig. S6G). We also observed significantly decreased phosphorylation of mTOR itself in Slc4a7‒/‒ PBs compared to controls (Fig. 5K). Collectively, these findings indicate that SLC4A7 regulates mTORC1 activity. To provide evidence for a causal link between SLC4A7, pHi, mTORC1 function, and PB differentiation, we sought to modulate mTORC1 function inSLC4A7-deficient B cells and thereby rescue their differentiation into PBs. In many cells including B cells mTORC1 activation is downstream of the serine/threonine kinase AKT, which inhibits tuberous sclerosis complex 1/2 (TSC1/2). TSC1/2 is an inhibitor of Ras homolog enriched in brain (RHEB), a direct activator of mTOR (Fig. 5L). Overexpression of a constitutively active form of AKT (myrAKT)50 rescued the impaired PB differentiation of SLC4A7-deficient B cells (Fig. 5M). Moreover, the overexpression of a constitutively active form of the direct mTORC1 activator RHEB (RHEBQ64L) also restored PB differentiation in the absence of SLC4A7 (Fig. 5N). Similar rescue effects of enforced mTORC1 activation by RHEBQ64L overexpression were observed in B cells treated with EIPA (to lower pHi) during PB differentiation upon either 40LB (Fig. 5O) or CpG (Fig. S6H) stimulation. Collectively, these data provide strong evidence that intracellular acidification in the absence of SLC4A7 or NHE function results in impaired mTORC1 activity and attenuated PB differentiation. Sustained mTORC1 activity is known to be important for B cell activation, growth and proliferation by acting through MYC-dependent signaling51,52. We were therefore surprised to not observe defects in cell proliferation or survival in SLC4A7-deficient B cells despite reduced mTORC1 activity. We tested the effects of mTORC1 inhibition with increasing concentrations of rapamycin on S6 phosphorylation and the viability, proliferation and PB differentiation of B cells53. Rapamycin concentrations between 0.1 and 1 nM resulted in a significant reduction in PB differentiation without noticeable effects on B cell proliferation or viability (Fig. S6I). The IC50 of rapamycin for PB differentiation was 0.47 nM, whereas those for proliferation and cell survival were 3.0 nM and >100 nM, respectively. The IC50 of rapamycin for S6 phosphorylation in B cells was 1.23 nM (Fig. S6J). These data suggest that PB differentiation is considerably more sensitive to reductions in mTORC1 activity than proliferation or cell survival. Taken together, we conclude that SLC4A7 deletion and the resulting intracellular acidification suppress the activity of mTORC1 and thus PB differentiation. Lysosomal alkalinization and reduced mTORC1 activity are associated with impaired PB differentiation We next sought to determine how pHi regulates mTORC1 activity. Intriguingly, the deletion of SLC4A7 and inhibition of NHEs not only resulted in lowered pHi but also caused lysosomal alkalinization (Fig. 6A–C). This finding is consistent with an earlier report in the human macrophage THP1 cell line, although the reasons for increased pHL were unclear20. The lysosomal lumen is maintained at a low pHL of ~4.5–5 by the proton-pumping activity of V-ATPases. Lysosomal proteases function optimally at low pHL to recycle amino acids, which mediates mTORC1 recruitment and its subsequent activation at the lysosomal surface54. The recruitment of mTORC1 to lysosomes is mediated by the Ragulator-Rag GTPase complex, which responds to amino acid availability. We confirmed that pharmacological alkalinization of lysosomes with chloroquine or bafilomycin A impairs S6 phosphorylation by mTORC1 in B cells (Fig. S7A). We hypothesized that lysosomal alkalinization in SLC4A7-deficient B cells impairs the nutrient-sensing arm of mTORC1 recruitment by dysregulating lysosome function. To measure the degradative function of lysosomes, we used the genetically encoded lysosomal-METRIQ probe55. Lysosomal function was reduced in SLC4A7-deficient B cells isolated from Slc4a7–/– mice (Fig. 6D) and B cells transduced with gSlc4a7 (Fig. S7B) consistent with their elevated pHL. Reduced lysosomal function was also observed in wildtype PBs upon NHE inhibition with EIPA (Fig. 6E). To demonstrate a causal link between elevated pHL in SLC4A7-deficient PBs and the lysosome-dependent function of mTORC1, we ectopically expressed WDR24, a component of the GATOR2 complex, or RAGBQ99L, an active mutant of RAGB in PBs from Slc4a7–/– mice. In response to amino acid availability, GATOR2 activates RAG GTPases, including RAGB, which mediate mTORC1 recruitment to the lysosome56 (Fig. 6A). Ectopic expression of RAGBQ99L (Figs. 6F and S7C) or WDR24 (Fig. 6G) partially rescued the defect in PB differentiation of SLC4A7-deficient B cells. RAGBQ99L expression also rescued PB differentiation of B cells treated with EIPA to inhibit NHE function (Fig. 6H). These findings suggest that lysosome alkalinization in SLC4A7-deficient B cells reduces lysosomal function and, thus PB differentiation through the regulation of mTORC1 activity.Fig. 6SLC4A7 deficient B cells exhibit lysosomal dysregulation.A Schematic illustration of mTORC1 regulation by lysosomal pathways and autophagy. B Measurements of lysosomal pH (pHL) using the genetically encoded pH sensor FIREpHLy in Slc4a7−/− or WT B cells. Representative histogram plots of Log10-transformed FIREpHLy signals (left) and quantification of non-transformed MFI of GFP/mCherry ratio (right). Histograms were normalized to total number of events. C Measurements of pHL using FIRE-pHLy in PBs treated with 5 μM EIPA or DMSO. D Lysosomal degradative function measured using the lysosomal-METRIQ probe in Slc4a7−/− or WT B cells co-cultured with 40LB cells for 4 days. The lysosomal-METRIQ signal was quantified as the mCherry to GFP fluorescence intensity ratio. Representative histogram plots of Log10-transformed Lysosomal-METRIQ ratios (left) and quantification of non-transformed mCherry/GFP MFI ratios (right). E Lysosomal-METRIQ signal in EIPA- or DMSO-treated B cells following 40LB co-culture for 4 days. Purified B cells were gated on DAPI– live cells. Gating strategies for (B–E) are shown in Fig. S11C. Frequencies of CD138+ PBs after ectopic expression of RAGBQ99L (F) or WDR24 (G) in WT or Slc4a7−/− B cells. Representative flow cytometry histograms (left) and quantification of CD138+ PB frequencies (right). H Frequencies of CD138+ PBs after ectopic expression of RAGBQ99L in WT B cells treated with 5 μM EIPA. Purified B cells were analyzed on day 4 of 40LB co-culture and gated on DAPI– live cells. I Frequencies of CD138+ PBs after ectopic expression of ATG5 in WT or Slc4a7−/− B cells. Representative flow cytometry histograms (left) and quantification of repeat experiments (right). J Frequencies of CD138+ PBs after ectopic expression of ATG5 in WT or Slc4a7−/− B cells co-cultured with 40LB cells and treated with DMSO or EIPA for 4 days. Purified B cells are gated on DAPI– live cells. Unless otherwise indicated, all B cells were co-cultured with 40LB feeder cells. Data are from 9 mice (B, D, F, G, I), 6 mice (C, H, J), and 6 and 3 mice (E; DMSO, n = 6; EIPA, n = 3) per cohort. Each dot represents one biological replicate. Data are shown as mean ± SEM. Statistical significance was assessed by two-sided unpaired Student’s t-test (B–E) and one-way ANOVA with Bonferroni correction for (F–J). P values are indicated above the graphs. Source data are provided as a Source data file. SLC4A7 deficient B cells exhibit lysosomal dysregulation. A Schematic illustration of mTORC1 regulation by lysosomal pathways and autophagy. B Measurements of lysosomal pH (pHL) using the genetically encoded pH sensor FIREpHLy in Slc4a7−/− or WT B cells. Representative histogram plots of Log10-transformed FIREpHLy signals (left) and quantification of non-transformed MFI of GFP/mCherry ratio (right). Histograms were normalized to total number of events. C Measurements of pHL using FIRE-pHLy in PBs treated with 5 μM EIPA or DMSO. D Lysosomal degradative function measured using the lysosomal-METRIQ probe in Slc4a7−/− or WT B cells co-cultured with 40LB cells for 4 days. The lysosomal-METRIQ signal was quantified as the mCherry to GFP fluorescence intensity ratio. Representative histogram plots of Log10-transformed Lysosomal-METRIQ ratios (left) and quantification of non-transformed mCherry/GFP MFI ratios (right). E Lysosomal-METRIQ signal in EIPA- or DMSO-treated B cells following 40LB co-culture for 4 days. Purified B cells were gated on DAPI– live cells. Gating strategies for (B–E) are shown in Fig. S11C. Frequencies of CD138+ PBs after ectopic expression of RAGBQ99L (F) or WDR24 (G) in WT or Slc4a7−/− B cells. Representative flow cytometry histograms (left) and quantification of CD138+ PB frequencies (right). H Frequencies of CD138+ PBs after ectopic expression of RAGBQ99L in WT B cells treated with 5 μM EIPA. Purified B cells were analyzed on day 4 of 40LB co-culture and gated on DAPI– live cells. I Frequencies of CD138+ PBs after ectopic expression of ATG5 in WT or Slc4a7−/− B cells. Representative flow cytometry histograms (left) and quantification of repeat experiments (right). J Frequencies of CD138+ PBs after ectopic expression of ATG5 in WT or Slc4a7−/− B cells co-cultured with 40LB cells and treated with DMSO or EIPA for 4 days. Purified B cells are gated on DAPI– live cells. Unless otherwise indicated, all B cells were co-cultured with 40LB feeder cells. Data are from 9 mice (B, D, F, G, I), 6 mice (C, H, J), and 6 and 3 mice (E; DMSO, n = 6; EIPA, n = 3) per cohort. Each dot represents one biological replicate. Data are shown as mean ± SEM. Statistical significance was assessed by two-sided unpaired Student’s t-test (B–E) and one-way ANOVA with Bonferroni correction for (F–J). P values are indicated above the graphs. Source data are provided as a Source data file. To provide further mechanistic support for the relationship between SLC4A7, lysosomal function, and PB differentiation, we investigated if increasing autophagy may restore PB differentiation of SLC4A7-deficient B cells. Upregulation of autophagy drives lysosomal catabolism to promote RAG-dependent mTORC1 recruitment and activity57. ATG5 is an indispensable component of the autophagic complex and is necessary for optimal PB differentiation and function58. We hypothesized that enhancing autophagy by overexpressing ATG5 may rescue the PB defect in SLC4A7-deficient cells. Ectopic expression of ATG5 indeed enhanced autophagy in primary murine CpG-stimulated B cells as measured by conversion of LC3B-I to LC3B-II (Fig. S7D). Importantly, it almost completely rescued PB differentiation of B cells from Slc4a7–/– mice (Fig. 6I) and wildtype B cells treated with EIPA (Fig. 6J). Together, these data provide evidence that lysosomal alkalinization in SLC4A7-deficient B cells may contribute to impaired PB differentiation by inhibiting mTORC1 on lysosomes. Extracellular acidosis impairs ASC differentiation and humoral immunity Our data suggests that SLC4A7 plays an important role in pHi homeostasis of ASC and humoral immunity. This is significant because pHi is influenced by extracellular pH (pHe) and B cells must possess mechanisms such as SLC4A7 and other acid-base transporters to maintain pHi within the physiological range10. Of note, pHe in lymph nodes, an important site of initial ASC differentiation, is significantly lower than in blood21. Extracellular acidification also occurs under disease conditions such as diabetic ketoacidosis or metabolic acidosis caused, for instance, by chronic kidney disease23. The effects of extracellular acidosis on humoral immunity are poorly understood and we therefore sought to analyze how extracellular acidification influences pHi in B cells, ASC differentiation and humoral immunity. We first used a model of metabolic acidosis to investigate the effects of pHe acidification on ASC differentiation in vivo. Wildtype mice were immunized with NP-OVA and treated with 0.5 M ammonium chloride (NH4Cl) in the drinking water to induce metabolic acidosis59 (Fig. 7A). On day 4 p.i., NH4Cl-treated mice showed a significant reduction in blood pHe by ~0.1 (Fig. 7B), which is comparable to the drop in pHe found in patients with severe metabolic acidosis23. This extracellular acidification in mice was accompanied by a strong reduction in the frequencies of CD138+ TACI+ (Fig. 7C) and IRF4+ Pax5– (Fig. S8A) ASCs in the spleen. By contrast, the frequencies of Fas+ GL-7+ GC B cells and CXCR5+ PD-1+ Tfh cells were not altered in NH4Cl-treated mice compared to controls (Fig. 7D, E). As a consequence, the ratio of ASC to GC B cells was significantly decreased upon NH4Cl treatment (Fig. S8B). Serum levels of IgM and IgG antibodies were significantly decreased after NH4Cl treatment compared to vehicle-treated controls (Fig. 7F, G). We also investigated the effects of metabolic acidosis on other immune cell populations in NH4Cl-treated mice. The frequencies of other immune cell populations including B220+ B cells, CD4+ and CD8+ T cells, NK cells, DCs, monocytes and neutrophils were not or only moderately altered in NH4Cl treated mice compared to controls (Fig. S8C). We did, however, observe significantly reduced total numbers of B cells, monocytes and cDC (Fig. S8D).Fig. 7ASC differentiation is sensitive to extracellular acidosis.A Experimental design for NP-OVA immunization and induction of metabolic acidosis by treatment of mice with 0.5 M NH4Cl in the drinking water. Splenocytes were analyzed on day 4 post-immunization. B Blood pH measurements in mice treated with NH4Cl or vehicle for 4 days. C Frequencies and total numbers of CD138+TACI+ ASCs in the spleen, quantified relative to total live lymphocytes. Representative contour plots (left) and quantification of ASCs from repeat experiments (right). Frequencies and total numbers of B220+Fas+GL-7+ GC B cells (D) and CD4+B220-CD44+Foxp3-CXCR5+PD-1+ Tfh cells (E) in the spleens of mice treated with NH4Cl or vehicle for 4 days. Representative flow cytometry plots (left) and quantification of repeat experiments (right). Gating strategies for (C–E) are shown in Fig. S12A, B. F Titers of low-affinity anti-NP25 IgG and high affinity anti-NP4 IgG antibodies from the serum of NP-OVA immunized mice treated with NH4Cl or vehicle. Absorbance (Abs.) measured by ELISA. G Quantification of NP-specific IgM and IgG antibody levels in the serum of NH4Cl- and vehicle-treated mice at 1:180 serum dilution. H–K Analysis of WT B cells cultured at extracellular pH (pHe) levels ranging from 6.9 to 7.7. Purified B cells were stimulated with anti-RP105 on day −3, transduced with pHluorin2 on day −2, seeded into the 40LB co-culture system on day 0, and analyzed on day 4 of PB differentiation. H Measurement of intracellular pH (pHi) using pHluorin2 in total B cells at day 4 of 40LB co-culture. I Frequencies of proliferating CFSElo cells on day 2 of 40LB co-culture. Gated on live DAPI– B cells. J Frequencies of live DAPI– B cells on day 2 of 40LB co-culture. K Frequencies of CD138+ PBs on day 4 of 40LB co-culture. L Frequencies of CD138+ PBs after ectopic expression of CA2 or empty vector (EV) in B cells and followed by 40LB co-culture at a medium pHe of 6.9 or 7.5 for 4 days. M Frequencies of p-S6+ B cells cultured in vitro at a media pHe of 6.9–7.7. Representative histograms (left) and quantification (right). Purified B cells were analyzed on day 3 of 40LB co-culture. Frequencies of CD138+ PBs after ectopic expression of RHEBQ64L (N), RAGBQ99L (O) or ATG5 (P), and co-culture with 40LB cells in media with pHe of 6.9 or 7.5. Purified B cells were analyzed on day 4. All data are from B cells co-cultured with 40LB feeder cells for 4 days unless otherwise stated. Data are from 10 (B–G), 6 (H–K), 9 (L), 5 (M) and 6 (N–P) mice per cohort from at least two independent experiments. Each dot represents a biological replicate. Shown are the mean ± SEM. Statistical significance was tested by two-sided unpaired Student’s t-test (B–G), and one-way ANOVA with Bonferroni correction (H–P). P values are indicated above the graphs. Source data are provided as a Source data file. ASC differentiation is sensitive to extracellular acidosis. A Experimental design for NP-OVA immunization and induction of metabolic acidosis by treatment of mice with 0.5 M NH4Cl in the drinking water. Splenocytes were analyzed on day 4 post-immunization. B Blood pH measurements in mice treated with NH4Cl or vehicle for 4 days. C Frequencies and total numbers of CD138+TACI+ ASCs in the spleen, quantified relative to total live lymphocytes. Representative contour plots (left) and quantification of ASCs from repeat experiments (right). Frequencies and total numbers of B220+Fas+GL-7+ GC B cells (D) and CD4+B220-CD44+Foxp3-CXCR5+PD-1+ Tfh cells (E) in the spleens of mice treated with NH4Cl or vehicle for 4 days. Representative flow cytometry plots (left) and quantification of repeat experiments (right). Gating strategies for (C–E) are shown in Fig. S12A, B. F Titers of low-affinity anti-NP25 IgG and high affinity anti-NP4 IgG antibodies from the serum of NP-OVA immunized mice treated with NH4Cl or vehicle. Absorbance (Abs.) measured by ELISA. G Quantification of NP-specific IgM and IgG antibody levels in the serum of NH4Cl- and vehicle-treated mice at 1:180 serum dilution. H–K Analysis of WT B cells cultured at extracellular pH (pHe) levels ranging from 6.9 to 7.7. Purified B cells were stimulated with anti-RP105 on day −3, transduced with pHluorin2 on day −2, seeded into the 40LB co-culture system on day 0, and analyzed on day 4 of PB differentiation. H Measurement of intracellular pH (pHi) using pHluorin2 in total B cells at day 4 of 40LB co-culture. I Frequencies of proliferating CFSElo cells on day 2 of 40LB co-culture. Gated on live DAPI– B cells. J Frequencies of live DAPI– B cells on day 2 of 40LB co-culture. K Frequencies of CD138+ PBs on day 4 of 40LB co-culture. L Frequencies of CD138+ PBs after ectopic expression of CA2 or empty vector (EV) in B cells and followed by 40LB co-culture at a medium pHe of 6.9 or 7.5 for 4 days. M Frequencies of p-S6+ B cells cultured in vitro at a media pHe of 6.9–7.7. Representative histograms (left) and quantification (right). Purified B cells were analyzed on day 3 of 40LB co-culture. Frequencies of CD138+ PBs after ectopic expression of RHEBQ64L (N), RAGBQ99L (O) or ATG5 (P), and co-culture with 40LB cells in media with pHe of 6.9 or 7.5. Purified B cells were analyzed on day 4. All data are from B cells co-cultured with 40LB feeder cells for 4 days unless otherwise stated. Data are from 10 (B–G), 6 (H–K), 9 (L), 5 (M) and 6 (N–P) mice per cohort from at least two independent experiments. Each dot represents a biological replicate. Shown are the mean ± SEM. Statistical significance was tested by two-sided unpaired Student’s t-test (B–G), and one-way ANOVA with Bonferroni correction (H–P). P values are indicated above the graphs. Source data are provided as a Source data file. To further investigate if low pHe directly acts on B cells and ASCs, we cultured wildtype B cells at varying pHe levels ranging from 6.9 to 7.7 during ASC differentiation by 40LB co-culture in vitro. Extracellular acidification resulted in a significant reduction in pHi (Fig. 7H) suggesting that neither SLC4A7 nor other acid-base transporters in B cells are fully sufficient to maintain pHi homeostasis in response to strong extracellular and subsequent intracellular acidification. Of note, low pHi did not affect B cell proliferation (Fig. 7I) or viability (Fig. 7J) consistent with the similarly negligible effect of SLC4A7 deletion (Fig. 1). By contrast, low pHe in vitro significantly decreased the frequencies of CD138+ ASCs (Fig. 7K) suggesting that ASC differentiation is directly sensitive to reduced pHe. The defect in ASC differentiation under acidic extracellular conditions could be partially rescued by ectopic expression of CA2 (Fig. 7L) similar to the effects of CA2 on SLC4A7-deficient and EIPA-treated B cells (Fig. 4). We next investigated if the effects of low pHe on ASC development are mediated by similar mechanisms as in Slc4a7-/- B cells, specifically reduced mTORC1 activity. Treatment of wildtype B cells in vitro under acidic extracellular conditions during ASC differentiation with 40LB resulted in a progressive decrease in S6 phosphorylation (Fig. 7M). To establish a causal link between low pHe, reduced S6 phosphorylation and impaired ASC differentiation, we ectopically expressed constitutively active mTORC1 regulators in wildtype B cells that were differentiated into ASC under acidic extracellular conditions. Both RHEBQ64L and RAGBQ99L completely rescued ASC differentiation even at pHe 6.9 (Fig. 7N, O). Moreover, promoting autophagy by ectopic expression of ATG5 also rescued ASC differentiation at pHe 6.9 (Fig. 7P). Collectively, our data demonstrate that extracellular acidosis impairs ASC differentiation in vitro and in vivo by altering pHi and suppressing mTORC1 activity. Discussion Here, we identified SLC4A7 as a critical regulator of ASC differentiation and humoral immune responses. We demonstrate that SLC4A7 regulates pHi and pHL in B cells and thus, controls the activity of the mTORC1 metabolic pathway which is required for ASC differentiation. Moreover, intracellular acidification under conditions of extracellular acidosis compromises ASC development and humoral immunity by suppressing mTORC1 function. These results identify pHi homeostasis as a fundamental driver of metabolic control in ASCs and establish acid-base regulation as a previously underappreciated mechanism shaping humoral immune responses (Fig. S14). SLC4A7 regulates pHi during ASC development A link between pHi and B cell function is not well established. Though a prior study had shown that the proton channel Hv1 regulates B cell function, its proposed function in B cells involved the BCR-induced production of reactive oxygen species, oxidation of the tyrosine phosphatase SHP-1 and modulation of BCR signaling strength6. The role of Hv1 in pHi homeostasis in B cells was not investigated. Here we show that SLC4A7 is essential for the regulation of pHi in B cells and their differentiation into ASCs in vitro and in vivo. Deletion of SLC4A7 in B cells resulted in cellular acidification, suppression of ASC differentiation and antibody production. Similar effects on ASC differentiation in vitro were observed by pharmacological inhibition in B cells with S0859, which is widely used as an inhibitor of Na+/HCO3− co-transport, although it is not isoform-selective for SLC4A7/NBCn115. Besides Na+/HCO3− co-transport by SLC4A7, it is likely that pHi homeostasis in ASCs is regulated by other channels and transporters. Indeed, pharmacological inhibition of Na+/H+ exchanger (NHE) function with EIPA resulted in reduced pHi and impaired ASC differentiation and antibody production. NHEs promote H+ export from cells and prevent cellular acidification, and several NHE genes are expressed in B cells, including Slc9a1 and Slc9a8. Neither gene was detected in our CRISPR screen for ICTs controlling ASC differentiation, potentially due to their redundant functions in controlling pHi. Of note, deletion of SLC4A7 resulted in significantly increased mRNA levels of several pHi regulating channels or transporters in ASCs, which included Slc16a3, a H+-dependent transporter of monocarboxylates such as L-lactate and pyruvate60, and Hvcn1 encoding Hv1. We speculate that both genes may be upregulated in SLC4A7-deficient cells in an attempt to maintain a neutral pHi. Their upregulation, however, does not seem to be sufficient to prevent cellular acidification in Slc4a7–/– cells, arguing for an essential role of SLC4A7 in ASCs and humoral immunity. SLC4A7 and pHi regulate mTORC1 activity Here we propose that the mechanism by which SLC4A7 and pHi exert their effects on ASC differentiation is through the regulation of mTORC1 activity. Studies in other cell types have shown a relationship between pHi and mTORC1, which was attributed to the pH dependence of upstream regulators such as TSC261, MAPK signaling62, and PI3K signaling49. This effect of low pHi on mTORC1 inhibition is not specifically related to SLC4A7 function because inhibition of NHEs with EIPA or acidification of the extracellular space to reduce pHi also resulted in decreased mTORC1 activity. We demonstrate that reduced pHi and mTORC1 activity are causally related to impaired ASC differentiation because the latter could be fully rescued by ectopic expression of constitutively active forms of the mTORC1 activators AKT and RHEB or by reversing low pHi to physiological levels by overexpression of CA2. Of note, mTORC1 was shown to promote SLC4A7 protein translation, thus enhancing HCO3– uptake35. It is possible that cellular acidification, for instance in conditions of extracellular acidosis, reduces mTORC1 activity, which in turn suppresses SLC4A7 expression, thus exacerbating cellular acidification and the ASC differentiation defect. Our data furthermore demonstrate that deletion of SLC4A7 in ASCs is associated with elevated pHL, which we hypothesize impairs mTORC1 function and ASC differentiation. The acidity of lysosomes (pHL ~4.5–5) is essential for optimal activity of lysosomal acid hydrolases, including proteases, which generate intralysosomal amino acids that contribute to mTORC1 recruitment and activation at the lysosomal surface. An involvement of alkalinized lysosomes in the mTORC1 defect of SLC4A7-deficient B cells is supported by the fact that ectopic expression of WDR24 or a constitutively active form of RAGB reverses the impaired ASC differentiation of SLC4A7-deficient B cells. WDR24 is a component of the GATOR2 complex, which positively regulates mTORC1 signaling by inhibiting GATOR1, a negative regulator of amino acid–dependent mTORC1 activation63. RAGB is a GTPase that functions downstream of amino acid sensing to promote mTORC1 recruitment and activation at the lysosome64. Moreover, we show that enhancing autophagy by overexpression of ATG5 also rescues ASC differentiation in vitro consistent with the fact that autophagic lysosomal degradation was shown to generate amino acids that promote RAG-dependent recruitment of mTORC1 to lysosomes and thus, its reactivation after starvation57. A limitation of this study is that we cannot at present provide an explanation for the opposing intracellular acidification yet lysosomal alkalinization in response to deletion of SLC4A7. Similar observations have been reported in other studies and cell-types20,65, and although seemingly paradoxical, several mechanisms may account for these findings. In THP1 cells, cytosolic acidification resulting from Slc4a7 deletion was shown to impair phagolysosomal acidification, which was attributed to impaired NADPH oxidase activity20. Moreover, V-ATPase function, which maintains lysosomal acidification, itself is pH dependent66. In yeast, cellular acidification was shown to perturb the assembly of the V1 and V0 subunits of the V-ATPase and thus, its function67. Regulation of mTORC1 activity by pHi promotes ASC differentiation mTORC1 function is essential for the generation of ASC as B cell-specific deletion of the gene encoding RAPTOR, an essential signaling adaptor for mTORC1, or treatment of wildtype mice with the mTORC1 inhibitor rapamycin abrogates the generation of ASCs in the spleen and bone marrow68. mTORC1 is also critical for B cell activation and proliferation after antigen receptor binding51. Although mTORC1 function depends on SLC4A7, we found that deletion of Slc4a7 does not affect the proliferation or viability of B cells in vitro but instead selectively impairs ASC differentiation. This apparent discrepancy is likely explained by the partial defect of mTORC1 activity caused by SLC4A7 deletion compared to the more complete mTORC1 inhibition reported in other studies. These findings suggest that ASC differentiation is particularly sensitive to small reductions in mTORC1 activity, whereas B cell proliferation can occur even under conditions of partial mTORC1 function. The molecular basis for this heightened sensitivity of ASC differentiation to mTORC1 remains to be elucidated. mTORC1 also plays a critical role in the differentiation and positive selection of GC B cells69. We did not, however, observe differences in GC B cell numbers in SLC4A7-deficient mice after influenza infection or immunization with a T-dependent antigen, or after induction of metabolic acidosis in immunized wildtype mice. A possible explanation again is that differentiation of GC B cells is not as sensitive to moderate reductions in mTORC1 activity compared to ASC differentiation. Because the focus of our study was on ASC differentiation, we did not comprehensively investigate the influence of SLC4A7 or extracellular acidification on GC B cell dynamics such as selection and affinity maturation. Since mTORC1 has a major influence on GC B cell selection, we suspect a more detailed analysis of this process may reveal interesting relationships between SLC4A7 and GC B cells. Moreover, future studies will need to evaluate the potential roles of SLC4A7 in other B cell subsets, such as memory B cells, and long-lived PCs in the bone marrow. Extracellular acidosis suppresses ASC differentiation and humoral immunity The role of extracellular pH in humoral immunity is largely unknown. Here, we provide new insights into the unique sensitivity of ASC differentiation to pHe. Our data demonstrate that low pHe reduces pHi, mTORC1 function and thus, suppresses ASC differentiation in vitro and in vivo. The channels and transporters responsible for cellular acidification in B cells encountering extracellular acidification are not known and a topic of future investigations. A potential explanation for reduced pHi in response to low pHe is that extracellular acidification alters the driving forces of pHi regulation in a manner that favors cellular acid loading while simultaneously inhibiting the net acid extrusion activity of SLC4A7 and several Na+/H+ exchangers, including NHE1 whose activity is inhibited by low pHe70. B cells encounter low pHe under many physiological and disease conditions. Intravital imaging studies of pHe demonstrated that lymph nodes at baseline are generally more acidic than blood, reaching pH levels as low as 5.821. How the acidic lymph node milieu affects ASC differentiation and which acid-base transporter may be involved in this process remains to be elucidated. Using a model of metabolic acidosis in mice, we here show that ASC differentiation is sensitive to low pHe in vivo. Metabolic acidosis is a hallmark of human diseases such as chronic kidney disease (CKD) or diabetic ketoacidosis. Direct evidence for a link between metabolic acidosis and altered humoral immunity in patients is lacking. Patients with CKD have reduced vaccination efficiency71 and immune dysfunction associated with an increased susceptibility to infections72. However, CKD is associated with many physiological derangements including uremia, thus complicating clinical investigations into the relationship between metabolic acidosis and humoral immunity. Cancer is another disease associated with extracellular acidosis, and acidification of the tumor microenvironment (TME) was shown to modulate antitumor immunity, in particular by affecting the function of effector T cells and myeloid cells26. Counteracting tumor acidosis by treatment with bicarbonate73–75 or inhibiting cancer cell acid extrusion with proton-pump blockers76 or through NBCn1 inhibition77 can inhibit tumor progression, invasion and metastasis. The effects of acidic TME on B cell function are not well understood, are intriguing to consider because tertiary lymphoid structures (TLS), including germinal center-like reactions and PC infiltration, have been observed in many tumors78,79. Translational significance The dependence of ASC differentiation and antibody production on pHi and pHe opens potential new therapeutic opportunities to modulate humoral immune responses in the context of vaccination, infection, antitumor immunity and autoimmune diseases such as multiple sclerosis (MS) and systemic lupus erythematosus (SLE). B cell-targeting therapies such as monoclonal antibodies against CD20 or CD19 and CAR T cells mediating B cell depletion have revolutionized the treatment of B cell malignancies and autoimmune diseases including SLE80. More ASC specific approaches are also being explored in antibody-mediated autoimmune diseases such as myasthenia gravis and SLE. The ASC-depleting anti-CD38 monoclonal antibody daratumumab has shown to be effective in patients with SLE and resulted in sustained clinical improvements81. It is tempting to speculate that the relative selective pH dependence of ASC differentiation can be exploited to suppress autoantibody production in SLE and other autoimmune diseases. SLC4A7 has been targeted in breast and lung cancer82,83, and for blood pressure control through regulation of arterial remodeling17. Many of these studies relied on an inhibitor of Na+/HCO3‒ transporters, S0859, which is not specific for SLC4A782,84. Recently, neutralizing antibodies against human SLC4A7 have been developed36 and represent a more selective tool to manipulate pHi and cell function. Future studies will investigate the effectiveness of genetic deletion and antibody-mediated modulation of SLC4A7 in the context of autoimmune diseases such as SLE. Methods Mice Slc4a7‒/‒ mice have been described previously17. Rosa26-Cas9 knock-in (strain 026179) and muMT-/- mice (strain 002288) were purchased from the Jackson Laboratory. All mice were maintained on a C57BL/6 genetic background and under specific pathogen-free conditions with a 12-h light/dark cycle at 25 oC and 55% humidity. Similar numbers of male and female mice (8-20 weeks old) were utilized for all experiments. Wild-type control and Slc4a7‒/‒ mice were bred and maintained separately, whereas control and experimental bone marrow chimeric mice were co-housed under identical conditions throughout the study. Mice had free access to food and water. They were healthy, with no notable defects in mobility, weight, or grooming. Mice were humanely euthanized by carbon dioxide (CO₂) inhalation followed by cervical dislocation to ensure death. All experiments were conducted in accordance with protocols approved by the Institutional Animal Care and Use Committee at New York University Grossman School of Medicine. PB differentiation in vitro Naïve B cells were isolated from mouse splenocytes by passing mechanically dissociated spleens through a 70-μm cell strainer, followed by red blood cell (RBC) removal using ammonium-chloride-potassium (ACK) lysis buffer (150 mM NH4Cl, 10 mM KHCO3, 0.1 mM Na2EDTA) and magnetic bead separation using CD43 Dynabeads (ThermoFisher Scientific, Cat no. 11422D). B cells were stimulated with 10 μg/mL anti-RP105 in complete RPMI containing 1X MEM non-essential amino acids, 1 mM sodium pyruvate, 10 mM HEPES, 100 μM β-mercaptoethanol. 24 h after stimulation, B cells were spin infected (1400 rcf, 90 min, 32 oC) with retroviral supernatant containing 10 μg/mL polybrene (Millipore Sigma, Cat no. 107689). Cells were returned to anti-RP105-containing media for another 48 h. After this timepoint, some cells were loaded with 5 μM CellTrace (ThermoFisher, Cat no. C34554) and re-stimulated with 20 μg/mL LPS (InvivoGen, Cat no. tlrl-eblps), 50 nM CpG ODN-1668 (InvivoGen, Cat no. tlrl-1668), 10 μg/mL anti-mouse IgM μ-chain (Jackson ImmunoResearch, Cat no. 115-006-075) plus 2 μg/mL anti-CD40 (Thermo Fisher Scientific, Cat no. 16-0401-82), or 40LB feeder cells in the presence of 2 ng/mL IL-4 (Peprotech). For PB differentiation using the 40LB co-culture system, B cells were cultured on a monolayer of 40LB feeder cells (obtained from Ari Melnik and as described in ref. 34) for 2 days and transferred to a fresh layer of 40LB cells in the presence of 10 ng/mL IL-21 (Peprotech, Cat no. 210-21) for another 2 days before analysis. For PB differentiation using CpG and LPS, cells were analyzed after two days of stimulation. Where indicated, EIPA or S0859 were added to B cells at day 0 concurrently with 40LB co-culture at concentrations and treatment durations specified in the figure legends. Chloroquine and bafilomycin A were added 1 h prior to sample collection. pHe was modulated using buffering agents: pH 7.5 (control medium), pH 7.7 with 5 mM Tris (tris(hydroxymethyl)aminomethane), pH 7.2 with 10 mM MES (2-(N-morpholino)ethanesulfonic acid), and pH 6.9 with 15 mM MES. Retroviral transduction Platinum-E (Plat-E) cells were transfected with retroviral expression plasmids encoding shRNAs (pLMPd), gRNAs (pMRI), or cDNA (pMSCV). The Plat-E retroviral packaging cell line (ATCC) was co-transfected with ecotropic packaging vector pCL-Eco for shRNAs or 10A1 for gRNAs. Retroviral supernatant was collected 48 and 72 h after transfection. The following guide sequences were used for retroviral transduction: gSlc4a7 (TCTGCAAAGGATCGAACCAG), gCtrl (TGTGTGACAGTATTAGCGAG), gPrdm1 (GAACTAATGCCGTACGG), gTsc1 (CACAGACAGTCAAACGACCG), and gTsc2 (CACAGGGTGATAATGAACAG). cDNAs for ectopic gene expression were bought from Addgene and cloned into the pMSCV vector: HA-WDR24-pRK5 (Addgene 46335), myrAkt-delta4-129 (Addgene 10841), pBad-CA TAG233 (Addgene 105838), pCW57.1 Lysosomal-METRIQ (Addgene 135401), pFUGW-FIRE-pHLy (Addgene 170774), pME pHluorin2 (Addgene 73794), pMXs-IP-EGFP-mAtg5 (Addgene 38196), pRK5 GFP RagB Q99L (Addgene 112749), RCAS(B)-Rheb-Q64L (Addgene 21050). Generation of custom gRNA library for ICTs We generated a custom library of 1132 ion channel and transporter (ICT) genes that includes channels, transporters, exchangers, pumps as well as some accessory proteins that facilitate their function. Selected genes encode transmembrane or membrane-associated proteins that facilitate the movement of inorganic and organic ions as well as organic metabolites across lipid bilayers. Three resources were used to select ICTs: (1) IUPHAR/BPS Guide to Pharmacology (https://www.guidetopharmacology.org), (2) HUGO Gene Nomenclature Committee (https://www.genenames.org/), and (3) UniProt (https://www.uniprot.org). The IUPHAR/BPS resource was used to identify solute carrier (SLC) genes. The HUGO resource was used to identify ICT genes known or proposed to mediate channel or transport function. To identify genes with known or predicted channel and transport function that do not belong to obvious ICT gene families, we bioinformatically scraped all remaining mouse and human genes from UniProt. For each gene, we collected functional information as well as gene ontologies. Genes whose functional information contained key terms related to channel or transport function or ion homeostasis were collected and manually reviewed for inclusion into the gRNA library. The resulting ICT gene list contains 1132 genes that are shared by mice and humans. The design and cloning of a small guide RNA (gRNA) library for CRISPR screens was done as follows. In addition to the 1132 ICT-related genes, we included 57 positive and 31 negative control genes. Positive control genes included genes with specific roles in B cells (e.g., Prdm1, Irf4 and Btk), as well as those with broad housekeeping roles (e.g., Eef2, Cdc123 and Polr2l). Negative control genes included those with negligible expression in B cells (e.g., Cd4, Cd8 and Bdnf) and those targeting non-mammalian genes (e.g., daf-16, slo-1 and mutS). Seven gRNAs targeting each gene were designed using CRISPick (https://portals.broadinstitute.org/gppx/crispick/public) and synthesized by Agilent. The gRNA pool was cloned into pMRI-Amt retroviral vector85 by Gibson assembly. Pooled CRISPR screen For the production of retroviral supernatants, Plat-E cells were co-transfected with the pMRI-based gRNA library and the 10A1 packaging vector. Supernatants were collected 36 and 48 h after transfection. 3–4 × 106 naive B cells were purified from spleen and enriched by negative selection with anti-CD43-magnetic beads. Retroviral transduction with the gRNA library, CellTrace loading, and co-culture in the 40LB system were performed as described above. After 2 days, 6 × 106 cells were sorted into 2 populations by flow cytometry: transduced, Ametrine-labeled (Amt+) cells with the highest (top 30%) and lowest (bottom 30%) CFSE mean fluorescence intensity (MFI) representing proliferating and non-proliferating cells, respectively. Another 6 × 106 cells were cultured on a fresh layer of 40LB feeder cells in the presence of 10 ng/mL IL-21 (Peprotech, Cat no. 210-21) to promote PB differentiation. After another 2 days, cells were sorted into 2 populations: transduced Amt+ CD138+ PBs and CD138– non-PBs. Throughout the culture process, the number of transduced cells was maintained above 1–2 × 106 to ensure at least 100x cell coverage of each gRNA. For the preparation of next generation sequencing libraries, sorted cells were lysed in buffer containing 10 mM Tris-HCl, 150 mM NaCl, 10 mM EDTA, 1% SDS at 55 oC for 16 h. Omnipur phenol:chloroform (EMD Millipore, Cat no. 6810400 ML) mix was emulsified into the lysate and samples were centrifuged (15,000 × g, 5 min) to isolate genomic DNA in the aqueous layer, which was then washed twice with a 24:1 chloroform:isoamyl alcohol mixture. Genomic DNA was precipitated with ethanol and reconstituted in ddH2O. Integrated proviruses containing gRNAs were amplified with Roche PCR master mix (Millipore Sigma, Cat no. 12140314001) using the following primers (for 26 cycles): forward (GAACCTCCTCGTTCGACCCC), reverse (GCACACCGGCCTTATTCCA). A second PCR reaction was performed to append primers containing the Illumina-compatible adapter sequence (forward primer adapter: AATGATACGGCGACCACCGAGATCT; reverse primer adapter: CAAGCAGAAGACGGCATACGAGAT), a unique 8 nucleotide barcode sequence for indexing, and the provirus complementary sequence (forward primer: TGGAGAAAAGCCTTGTTTG; reverse primer: CTAGGATCCTAGCTCTAAAAC). An example of a primer pair: forward primer, AATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACGCTCTTCCGATCTTAAGTAGAGTGGAGAAAAGCCTTGTTTG; reverse primer, CAAGCAGAAGACGGCATACGAGATAAGTAGAGGTGACTGGAGTTCAGACGTGTGCTCTTCCGATCTTCTAGGATCCTAGCTCTAAAAC. Next-generation sequencing of DNA libraries was performed following quality control of the amplified libraries using the Agilent 2200 TapeStation Bioanalyzer (Agilent Technologies). The libraries were sequenced as single-read 100 bp using a NextSeq 2000 Sequencer (Illumina). Sequencing results were trimmed with Cutadapt86, aligned using Bowtie 287, and gRNA enrichment analysis was performed using the Model-based Analysis of Genome-wide CRISPR-Cas9 Knockout (MAGeCK) package88. Hits in the PB differentiation screen were considered significant if they (i) fell into the top enrichment and depletion quartiles, and (ii) were expressed in wildtype PBs of our RNA-seq analysis (GSE269509) with a normalized count ≥30. Hits in the B cell proliferation screen were considered significant if they (i) fell into the top enrichment and depletion quartiles, and (ii) were expressed in wildtype non-PB of our RNA-seq analysis (GSE269509) with a normalized count ≥ 30. Gene expression analysis Publicly available mRNA expression data of mouse B cell subsets across different developmental stages were downloaded from the IMMGEN database (GSE109125) and analyzed for mRNA levels of Slc4a7 and the other 72 unique ICT hits from our PB CRISPR screen. For visualization, TPM (transcript per million) normalized counts were Log10 transformed. Measurements of intracellular pH, lysosomal pH, and lysosomal activity To measure intracellular pH (pHi), B cells were transduced with pHluorin2 (Addgene, cat. no. 73794), a genetically-encoded ratiometric pH sensor40. Fluorescent signals were collected by flow cytometry in the BV510 (405 nm excitation, 535 nm emission) and GFP (488 nm excitation, 535 nm emission) channels. The ratio of BV510 to GFP was calculated and pH values were calibrated to a standard curve. The standard curve was created by incubating pHluorin2-expressing B cells with 10 μM nigericin in a high-potassium buffer (140 mM KCl, 1 mM MgCl2, 2 mM CaCl2, 5 mM D-glucose) at varying pH levels from 4.1 to 10.1. Cells were incubated for 15 min to allow equilibration between cytosolic and extracellular pH, after which cells were analyzed by flow cytometry. To measure lysosomal pH (pHL), FIRE-pHLy (Addgene, Cat. no. 170774), a ratiometric genetically-encoded lysosome-localizing pH sensor89 was used. B cells were transduced as described above. Fluorescence signals were collected by flow cytometry. The ratio of GFP to mCherry (excitation 561 nm, emission 594 nm) was calculated, with low ratios corresponding to lower pHL values. To measure lysosomal activity, we used the genetically-encoded reporter Lysosomal-METRIQ (Addgene, cat. no. 135401)55. B cells were transduced with Lysosomal-METRIQ as described above and fluorescence signals were analyzed by flow cytometry. The ratio of mCherry to GFP was quantified, with low ratios corresponding to low lysosomal activity. Generation of bone marrow chimeras (BMC) To generate BMCs with B cell specific deletion of SLC4A7, C57BL/6 recipient mice were lethally irradiated (10 Gy split dose) and reconstituted with 2 × 106 bone marrow cells from Slc4a7‒/‒ mice and 8 × 106 bone marrow cells from B cell-deficient muMT‒/‒ mice by retro-orbital injection. Reconstituted mice were fed 0.2 mg/mL trimethoprim-sulfamethoxazole in drinking water for 2 weeks post-reconstitution. After 8–11 weeks, mice were analyzed to ensure successful reconstitution by measuring the frequency of CD19+ B cells, CD3+ T cells and CD11b+ myeloid cells among peripheral blood mononuclear cells and used for immunization or infection experiments. Immunization and influenza infection of mice Mice were immunized with the following reagents by intraperitoneal (i.p.) injection: 50 μg NP-Ficoll with 100 μg Poly-IC, or 50 μg NP-OVA in Alhydrogel 2% (InvivoGen, cat. no. 5808-47-01). For infection with the X31 (H3N2) strain of influenza A virus (IAV), virus stocks were diluted in PBS and administered intranasally at 1 × 105 pfu. For unimmunized control mice, PBS without virus was administered intranasally. X31 IAV was kindly provided by Dr. Paul G. Thomas (St. Jude Children’s Research Hospital)90. Lung histology Isolated lungs were washed in PBS, fixed in 4% paraformaldehyde (PFA) for 2–4 days, embedded in paraffin and cut into 5 μm tissue slices. Slides were stained with hematoxylin and eosin (H&E) in 95% ethanol using standard methods. Images were acquired with a Leica AT2 whole slide scanner. For image analysis, representative tissue scans were downscaled to a resolution of 1000 dots per inch (DPI). To identify pixels representing regions of immune infiltration compared to ventilated alveolar space, we used ilastik91, a random forest classifier for pixels that uses various non-specific image filters (e.g., Gaussian blur) as parameters. We trained the classifier on 3 lung sections using 100 trees and 37 features (representing 6 different image filters). With the resulting pixel classifications, images were then processed through the following steps: (1) background deletion to isolate lung tissue area, (2) quantification of pixels representing infiltrated lung regions and (3) quantification of pixels representing ventilated alveolar spaces. The severity of lung inflammation was quantified as the ratio of infiltrated areas to the total lung area. Metabolic acidosis Metabolic acidosis was induced in wildtype C57BL/6 mice by adding 0.5 M NH4Cl (plus 2% sucrose) to the drinking water. Vehicle control cohorts of mice were substituted with 2% sucrose alone. Animals were allowed to drink ad libitum and the ingested volumes were monitored to ensure equivalent ingestion between control and acidosis groups. Mice were administered 0.5 M NH4Cl on day 0 right after NP-OVA immunization, and analyzed on day 4 post-immunization. Blood was collected into heparin (Thermo Scientific, cat. no. A16198.MD) coated tubes via cheek blood puncture and pH was measured within 1 min of blood collection by PH 850 pH meter (APERA instruments, cat. no. AI5533). Flow cytometry B cells cultured in vitro or isolated from mice ex vivo were washed in PBS containing 2% fetal bovine serum and FcR were blocked by adding anti-CD16/32 (BioXcell, cat. no. BP0307) for 10 min. Surface proteins were stained with fluorescently-labeled antibodies at 4 °C for 15 min in the dark. For staining of intracellular proteins, cells were stained with LIVE/DEADTM Fixable Blue (ThermoFisher, Cat no. L23105) for dead cell discrimination according to manufacturer’s instructions. Cells were fixed with 4% PFA for 15 min and permeabilized with 0.2% NP-40 for 10 min. Cells were washed with PBS and incubated with antibodies overnight at 4 °C. For some transcription factor staining, cells were fixed and permeabilized using Foxp3 Transcription Factor Staining Buffer Set (eBioscience, cat. no. 00-5523-00) according to manufacturer’s instruction. Protein synthesis in B cell was measured by puromycin incorporation. Cells were stimulated under the indicated conditions and pulsed with 10 μg/mL puromycin (Invitrogen, A11138-03) for 30 min at 37 °C. Cells were then washed, surface stained as indicated, fixed with 4% paraformaldehyde for 15 min, permeabilized with 0.2% NP-40, and stained with anti-puromycin overnight at 4 °C. Puromycin incorporation was quantified as median fluorescence intensity (MFI) within gated populations. A detailed list of antibodies used is provided in Table S2. Samples were analyzed on a LSR Fortessa or a FACSymphony A5 flow cytometer (both BD Biosciences) and analyzed using FlowJo software (FlowJo v.10.10.1). Detailed gating strategies for all flow cytometry experiments are provided in Figs. S9–S12. Enzyme-linked immunosorbent assay (ELISA) For measurements of high and low-affinity anti-NP antibody levels, 10 μg/mL NP4-BSA or NP25-BSA were incubated on half-area ELISA plates (Corning, cat. no. 9018) overnight at 4 °C. For measurements of anti-X31 influenza antibodies, X31 virions were plated at a concentration of 1 × 106 p.f.u per well. For detection of IgM antibodies secreted by B cells in vitro, plates were coated with anti-mouse κ light chain (Southern Biotech, Cat no. 1170-01) and anti-mouse λ light chain (Southern Biotech, Cat no. 1175-01) at 2.5 μg/mL. Following incubation with the target antigen or capture antibody, plates were washed with PBS and blocked with 5% BSA for 1 h at room temperature. 50 μL of cell culture supernatants (diluted as indicated) or mouse sera (diluted as indicated) were added to the plates and incubated for 2 h at room temperature. Wells were then washed with PBS and incubated with alkaline phosphatase-conjugated anti-IgM (Southern Biotech, Cat no. 1021-04) or anti-IgG (Southern Biotech, Cat no. 1030-04) for 1 h at room temperature. Wells were washed again with PBS and incubated with p-Nitrophenyl Phosphate (PNPP, ThermoFisher, Cat no. 37620) substrate for 30 min. Colorimetric detection was performed at 405 nm using a FlexStation 3 plate reader (Molecular Devices). Enzyme-linked immunosorbent spot (ELISPOT) assay Multi-well PVDF plates (MilliporeSigma, Cat no. MSIP4W10) wells were activated with 70% ethanol and antigen coating was performed with either NP25-BSA (0.2 μg per well) or X31 influenza viral particles (1 × 105 VFU per well). For each well, 2 × 105 splenocytes or mediastinal lymph node cells were seeded and incubated in 5% CO2 at 37 °C for 18 h. Wells were then washed and incubated in primary anti-IgG-biotin antibody (1:5000; Southern Biotech, Cat no. 1030-08) for 2 h at room temperature. Secondary incubation with streptavidin-alkaline phosphatase (1:2000; Sigma Aldrich Cat no. 189732) was performed for 1 h at room temperature. Chromogenic reactions were induced by treating wells with NBT/BCIP substrate solution (ThermoFisher Scientific; Cat no. 34042) for 15 min. Spots were analyzed with an S6 Entry M2 ELISPOT reader (ImmunoSpot). Quantitative real-time PCR Total RNA was isolated from B cells using TRIzolTM Reagent (Invitrogen, cat. no. 15596026) and quantified using a Nanodrop 8000 spectrophotometer (Thermo Scientific). cDNA was synthesized using iScriptTM cDNA synthesis kit (Bio-Rad, cat. no. 1708890). Quantitative real-time PCR (qRT-PCR) was performed using the Maxima SYBR Green qPCR Master Mix (Thermo Scientific, cat. no. K0223). Transcript levels were normalized to the expression of housekeeping genes, either Gapdh, Rpl32, or Actb using the 2‒∆CT method92. QRT-PCR data were acquired and analyzed using a QuantStudio 5 Real-Time PCR system and data analysis software (Thermo Fisher Scientific). Primer sequences for Slc4a7 are as follows: forward primer (CGAGCAGATGAGACCGCT), reverse primer (ACGCCGACGACTCTCTTTAC). Immunoblotting Total cell lysates were prepared by incubating B cells on ice with lysis buffer containing 150 mM NaCl, 10 mM EDTA, 10 mM EGTA, 50 mM Tris, 1% Triton X-100, 1% protease inhibitor cocktail (Sigma, Cat no. P8340), and phosphatase inhibitor cocktail (Cell Signaling, Cat no. 5870). Cell debris was removed by centrifugation at 15,000 rcf for 5 min and protein extracts were solubilized with Laemmli sample buffer (5% β-mercaptoethanol, 0.01% bromophenol blue, 10% glycerol, 2% SDS, 63 mM Tris-HCl). Protein extracts were separated by SDS-PAGE using 4–20% NovexTM WedgeWellTM Tris-glycine mini protein gel (Invitrogen, cat. no. XP04205BOX) and transferred to polyvinylidene difluoride membranes (Bio-Rad, Cat no. 1704157). Membranes were blocked in 5% non-fat milk for at least 1 h. Membranes were incubated overnight with primary antibodies at 4 °C and immunoreactive bands were detected after incubation with horseradish peroxidase (HRP)-labeled secondary antibody (1:5000–1:10,000, Sigma, A9044) for 1 h at RT. Antibodies against epitopes in the N terminus (MEADGAGEQMRPLLTRGPD(C))93 and C terminus ((C)EDEPSKKYMDAETSL)94 of SLC4A7 were used at 1:1000 dilution. Other antibodies were used as follows: anti-β-actin (1:5000, Santa Cruz sc-1615), anti-GRP94 (1:1000, Enzo, ADI-SPA-850) and anti-LC3B (1:1000, Cell Signaling Technology, 43566T). Note that for detection of LC3B, cells were treated with either bafilomycin A (100 nM) or DMSO for 1 h prior to lysate preparation. Antibody binding was visualized by enhanced chemiluminescence using Clarity Max Western ECL Substrate (Bio-Rad, Cat no. 1705062) using an Amersham 680-GE system or ChemiDoc Imaging System (Bio-Rad). Alternatively, detection of some antibodies was performed after incubation with IRDye680RD secondary antibody (1:10,000, LI-COR, Cat no. 925-68072) for 1 h at RT and visualized with an Odyssey Fc Western Blot Detection System (LI-COR Bioscience) or a ChemiDoc Imaging System (Bio-Rad). Densitometric analysis of bands was performed using ImageJ. Uncropped immunoblots are provided in Fig. S13. RNA-sequencing Naïve B cells were isolated from the spleen of Slc4a7‒/‒ and control mice using CD43 magnetic selection beads. 5 × 104 purified B cells were seeded onto a monolayer of 40LB feeder cells in the presence of 2 ng/mL of IL-4 for 3 days followed by 10 ng/mL of IL-21 for 3 days, after which CD138+ PBs and CD138– non-PBs were enriched by flow cytometric cell sorting. RNA was purified using the RNeasy Micro RNA Isolation Kit (Qiagen, Cat no. 74104) and RNA-Seq libraries were prepared using the Clontech SMART-Seq HT Kit (Takara, Cat no. 634792). The amplified libraries were purified using AMPure beads (Beckman Coulter, Cat no. A63882), quantified using a Qubit 2.0 fluorometer (Life Technologies), and visualized on an Agilent Tapestation 2200. The libraries were pooled equimolarly, loaded onto the NovaSeq 6000 system (Illumina) and run as 50 nucleotide paired-end reads. For gene expression analysis, reads were aligned to the mouse reference genome (GRCm38.85 / mm10) using STAR aligner (v.2.5.0c)95, and read counts were generated with featureCounts as described96. Differential expression was performed with R/DESeq297, where differentially expressed genes (DEGs) were defined according to the following criteria: fold-change (FC) > 1.5 and false discovery rates (FDR) < 0.01. FDRs were adjusted using the Benjamini–Hochberg method, and FCs were shrunk using the ashr methods native to DESeq298. The Gene set enrichment analysis (GSEA) was performed with R/clusterprofiler99 using all genes (DEGs and non-DEGs) ranked according to their shrunken log2 FC comparing Slc4a7 ‒/‒ vs WT PBs. The specific gene sets used for the enrichment analyses were acquired from EnrichR100. Heatmaps were generated using ComplexHeatmap(v2.18.0). For heatmap visualization, DESeq2-normalized count values were row-normalized by z-score. Immunofluorescence B cells stimulated with CpG for 24 h were seeded on coverslips (No. 1.5) coated with 0.01% poly-L-lysine (Millipore Sigma, Cat no. P8920) for 20 min in RPMI1640 plus 10% FCS. Cells were fixed with 4% PFA for 15 min, permeabilized using 0.2% NP-40 in PBS with 5% BSA for 15 min and washed with PBS. Cells were incubated with Fc Block (BioXcell, Cat no. BP0307) for 1 h at room temperature and washed in PBS. Cells were incubated with 0.5 μg/mL anti-HA-PE antibody (BioLegend, Cat no 901517) overnight at 4 °C, after which cells were washed and incubated with 1 μg/mL DAPI (ThermoFisher, Cat no. D1306) in PBS for 10 min at room temperature. Cells were washed once more before being mounted with Glycergel mounting medium (Agilent, Cat no. C0563). Images were acquired using a Zeiss 880 confocal microscope and processed using ImageJ and Matlab/R2020 (Mathworks). Data analysis and statistics Data are presented as mean ± s.e.m. Sample sizes were not predetermined statistically, and normality was assessed using the Shapiro–Wilk test. Unless restricted by the genotype, animals and cell samples were assigned randomly to experimental conditions. Data collection and analysis were not blinded to the conditions of the experiments. Where specified, outlier analysis was performed to exclude datapoints more than three scaled median absolute deviations from the median. Apart from the transcriptomic data, all experimental hypotheses were pre-specified. Statistical significance was determined by using two-tailed, unpaired Student’s t-test, one-way ANOVA and two-way ANOVA as mentioned in the figure legends. Statistical analyses were performed using Prism, Matlab and R. P values are indicated above the graphs. P < 0.05 was considered statistically significant. Data visualization Flow cytometry data were analyzed and visualized using FlowJo, version 10.10.1 (BD Biosciences). Quantitative data were analyzed and visualized using GraphPad Prism, version 11.0.2 (GraphPad Software). Figures were assembled, edited, and formatted using Adobe Illustrator, version 30.6 (Adobe). Schematic illustrations were created using BioRender (BioRender.com), imported and further formatted in Adobe Illustrator. Reporting summary Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article. Supplementary information Supplementary Information Reporting Summary Transparent Peer Review file Supplementary Information Reporting Summary Transparent Peer Review file Source data Source Data Source Data