sign in sign up
<<
Home , Nina Papavasiliou, Plasma cell, T cell

Long-Lived Plasma Cells Are Induced Early in Response to -Independent or T Cell-Dependent This information is current as of October 2, 2021. Alexandra Bortnick, Irene Chernova, William J. Quinn III, Monica Mugnier, Michael P. Cancro and David Allman J Immunol 2012; 188:5389-5396; Prepublished online 23 April 2012;

doi: 10.4049/jimmunol.1102808 Downloaded from http://www.jimmunol.org/content/188/11/5389

Supplementary http://www.jimmunol.org/content/suppl/2012/04/24/jimmunol.110280 Material 8.DC1 http://www.jimmunol.org/ References This article cites 48 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/188/11/5389.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on October 2, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Long-Lived Bone Marrow Plasma Cells Are Induced Early in Response to T Cell-Independent or T Cell-Dependent Antigens

Alexandra Bortnick,* Irene Chernova,* William J. Quinn, III,* Monica Mugnier,† Michael P. Cancro,* and David Allman*

The signals required to generate long-lived plasma cells remain unresolved. One widely cited model posits that long-lived plasma cells derive from germinal centers (GCs) in response to T cell-dependent (TD) Ags. Thus, T cell-independent (TI) Ags, which fail to sustain GCs, are considered ineffective at generating long-lived plasma cells. However, we show that long-lived -specific plasma cells are readily induced without formation of GCs. Long-lived plasma cells developed in T cell-deficient mice after a single with haptenated LPS, a widely used TI Ag. Long-lived plasma cells also formed in response to TD Ag when the GC response was experimentally prevented. These observations establish that long-lived plasma cells are induced in both TI and TD Downloaded from responses, and can arise independently of maturation in GCs. The Journal of Immunology, 2012, 188: 5389–5396.

ntibodies play pivotal roles in host defense and can cause and therefore more protective, Abs (12). This viewpoint is sup- disease when specific for self-antigens. Plasma cells, the ported by data showing that the bulk of BM plasma cells induced main source of Abs, are thought to consist of two pools: by TD Ag secrete high-affinity class-switched Abs (13). However,

A http://www.jimmunol.org/ short-lived cells that secrete low-affinity IgM Abs and form extra- the notion that long-lived plasma cells must derive from GCs follicular foci in the or ; and long-lived cells appears incompatible with recent data showing that low-affinity that secrete high-affinity, -switched Abs found mainly in IgM Abs can mediate long-term protection against certain mi- the bone marrow (BM) (1). Whereas short-lived plasma cells die crobial pathogens (14). within 3–5 d (2), long-lived plasma cells are thought to persist for Classically, Ags that do not engage Th cells are thought to months or years in mice and perhaps decades in people (3–6). generate transient IgM responses that reflect the activity of short- Sustained serum Ab titers due to long-lived plasma cells are es- lived plasma cells (2, 15–17). These T cell-independent (TI) Ags sential for protective against many pathogens (3, 7–9). may also generate short-lived GCs that fail to engender SHM Therefore, defining the events underlying the generation of long- and affinity maturation (18, 19). However, several recent studies by guest on October 2, 2021 lived plasma cells is essential for understanding how Ab-mediated challenge the long-standing paradigm that TI Ab responses are immunity is established and maintained. always short-lived. For instance, immunization of T cell-deficient Why some plasma cells achieve longevity whereas others are mice with the spirochete Borrelia hermsii induces long-term IgM- short-lived is not known. One current and widely cited model posits dependent protection (20, 21). IgM Abs induced by the intracel- that long-lived plasma cells emanate chiefly from germinal centers lular bacteria Franciscella tularensis and Ehrlichia muris also (GCs) in response to T cell-dependent (TD) Ags (4, 10). GCs are mediate long-term protection; however, these responses may re- microanatomical structures enriched with Ag-stimulated B cells flect the continuous generation of short-lived Ab-secreting cells undergoing class switch recombination, (22, 23). More recently, E. muris and the encapsulated bacterium (SHM), and affinity-based selection (11). Indeed, it was proposed Streptococcus pneumoniae were shown to elicit a long-standing recently that long-lived plasma cells are generated preferentially pool of Ag-specific BM plasma cells in T cell-sufficient mice (24– within the GC to favor the continuous secretion of high-affinity, 26). Although the latter work provides evidence that typical TD Ags are not unique in their ability to induce long-term Ab re- sponses, these studies did not address whether long-lived plasma *Department of Pathology and Laboratory Medicine, University of Pennsylvania cells can be generated in the absence of T cells. Therefore, School of Medicine, Philadelphia, PA 19104; and †Laboratory of Biol- ogy, The Rockefeller University, New York, NY 10065 whether T cell-derived signals are strictly required to generate Received for publication September 28, 2011. Accepted for publication March 22, long-lived plasma cells remains unclear. 2012. Our work addresses the capacity of TI and TD Ags to induce This work was supported by National Institutes of Health Grants T32 AI070099 (to long-lived plasma cells during the earliest phases of A.B.) and R21 AI090700 (to D.A.) and Department of Defense Grant W81WWXWH- differentiation. We show that haptenated LPS, a classic type 1 TI 10-1-0185 (to M.P.C.). Ag, readily induces a long-standing pool of BM plasma cells in Address correspondence and reprint requests to Dr. David Allman, University of mice that lack T cells. These Ab-secreting cells are detected for Pennsylvania, 36th & Hamilton Walk, 230 John Morgan Building, Philadelphia, PA 19104-6082. E-mail address: [email protected] .100 d after a single immunization, exhibit a t1/2 of 45–55 d, and The online version of this article contains supplemental material. arise despite an inability to detect Ag-induced GC B cells. These Abbreviations used in this article: BCL6, B cell gene-6; BM, bone mar- data challenge the long-standing notion that type 1 TI Ags fail to row; CGG, chicken g-globulin; GC, ; IR, ionizing radiation; NP, (4- induce the formation of long-lived plasma cells, and suggest that hydroxy-3-nitrophenyl)acetyl; SHM, somatic hypermutation; TD, T cell-dependent; long-lived plasma cells need not arise from GCs. Similarly, we TI, T cell-independent. show that long-lived plasma cells also form in response to a Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 standard TD Ag without undergoing maturation and selection in www.jimmunol.org/cgi/doi/10.4049/jimmunol.1102808 5390 LONG-LIVED PLASMA CELLS

GCs, as plasma cells secreting low-affinity IgM Abs persisted for Statistical analysis $100 d in mice in which we prevented GC formation early in Significant differences in plasma cell frequencies between two experimen- these responses. These findings indicate that maturation in GCs is tal groups were evaluated with the unpaired two-tailed t test, using Excel not requisite for achieving longevity in the plasma cell lineage and software. also suggest that competence to enter long-lived plasma cell pools is achieved early in TI and TD Ab responses. Results Persistence of LPS-induced TI plasma cells Materials and Methods 2 2 2 2 We immunized C57BL/6 (B6) or T cell-deficient B6.TcRb / d / Mice adults with the hapten (4-hydroxy-3-nitrophenyl)acetyl (NP) con- 2 2 2 2 C57BL/6 (B6), B6.TcRb / d / , and MD4 Ig transgenic (anti-HEL) jugated to LPS or CGG. We studied the Ab response to NP in B6 females (age 8–10 wk) were obtained from Jackson Laboratories. 2 2 background mice for two reasons. First, the kinetics of both AID / mice were provided by Dr. Nina Papavasiliou (The Rockefeller University). All animal procedures were approved by the University of plasma cell differentiation and GC-dependent somatic hypermu- Pennsylvania Office of Regulatory Affairs. tation and selection in response to NP are well established (29, 30). Second, NP-responsive B-lineage cells are readily identified Chimeras by flow cytometry with NP-conjugated fluorescent , and Hosts were exposed to whole-body radiation (900 R) 5 d post immunization, such cells can be resolved in conventional inbred mice without 6 and then given 1 3 10 BM cells i.v. from MD4 IgH+L transgenic mice in using Ig transgenes to increase frequencies of Ag-responsive which all B cells are specific for hen egg lysozyme (27). BM cells were depleted of CD3ε+ cells using a MACs LD column before i.v. transfer. All B cells (31). chimeras were maintained on antibiotic water for $6 wk post radiation. We quantified NP-specific plasma cell frequencies in the spleen Downloaded from Immunization of separate MD4-into-B6 chimeras with NP-chicken and BM at several time points after a single inoculation of B6 or g-globulin (CGG) or NP-LPS failed to elicit detectable NP-specific B6.TcRb2/2d2/2 mice with NP-LPS. Because the Ab response plasma cells as determined by ELISPOT (not shown). to NP-conjugates in B6-background mice is dominated by l+ B cells (30, 32), we assayed frequencies of plasma cells secreting l+ NP-binding Abs. Although, as expected, splenic NP-specific Mice 8–10 wk old were immunized i.p. with 50 mgNP16-CGG in alum or

plasma cell frequencies increased and then decreased exponen- http://www.jimmunol.org/ 50 mgNP0.6-LPS in PBS. tially between days 1 and 5 (15), NP-specific plasma cells were ELISPOT readily observed in the spleen and BM for $200 d in B6 mice and 2/2 2/2 MultiScreenHTS plates (Millipore) were coated with 10 mg per well of $91 d in B6.TcRb d mice (Fig. 1A, 1B). Furthermore, either goat anti-mouse Ig (H+L) (Southern Biotech), or NP33-BSA or NP4- although in the spleen background levels increased substantially BSA (BioSearch), in sodium bicarbonate buffer, and then blocked with 2% for mice immunized with NP-LPS .300 d previously, obscuring BSA/PBS. Cells were serially diluted across the plate and then incubated hapten-specific splenic plasma cells, in the BM we easily detected for 4–6 h at 37˚C. Biotin–goat-anti–Igl, goat–anti-IgM, or goat–anti-IgG1 + (Southern Biotech) diluted in block buffer was added, followed by three l NP-specific plasma cells well above background in mice that washes with 0.1% Tween 20 detergent, and a secondary incubation with were immunized .400 d previously (Supplemental Fig. 1). Of

ExtrAvidin-alkaline phosphatase (Sigma-Aldrich). Spots were detected note, as shown in Fig. 1B, on day 91 postimmunization frequen- by guest on October 2, 2021 using BCIP/NBT (Sigma-Aldrich) and scanned and counted with an cies of NP-specific plasma cells appeared to be greater in ImmunoSpot Analyzer (Cellular Technology). B6.TcRb2/2d2/2 mice than in B6 mice, although whether this and cell sorting Spleen and BM cells were harvested and stained with optimal dilutions of the indicated Abs, as described (28). Unless noted otherwise, all of the following reagents were purchased from eBiosciences: FITC–anti-Igl1–3 (R26-46; Pharmingen) and peanut lectin (Sigma-Aldrich); PE–anti-CD138 (281-2; Pharmingen); PE–Texas Red–anti-B220 (RA3-6B2); PE–Cy7– anti-CD4 (RM4-5), anti-CD8a (53-6.7), anti–Gr-1 (RB6-8C5), anti-F4/80 (BM8), and anti-TER119; allophycocyanin–Cy5.5–anti-CD19 (1D3); Alexa405 anti-IgD (11–26) and Biotin–anti-CD138 (281-2; Pharmingen). Biotinylated Abs were revealed with Streptavidin-APC-Cy7 (Pharmingen). NP-allophycocyanin was conjugated in our laboratory, using standard methods. Nonviable cells were eliminated from all analyses with the ultraviolet-excited DNA dye DAPI (Molecular Probes), and doublets were excluded from all analyses, using the combined width parameter of the forward and side scatter parameters. Flow cytometry was performed on a BD LSRII, and cell sorting was performed on a four-laser Aria (Becton Dickinson). Analysis was accomplished using FlowJo 8.8 (Tree Star). BrdU labeling B6 mice were immunized with NP-LPS and fed BrdU in the drinking water (0.5 mg/ml with 1% dextrose) for 3 d beginning immediately after im- FIGURE 1. Kinetics of the plasma cell response after NP-LPS immu- munization. Spleen and BM cells were stained with appropriate Abs, fixed nization. (A) NP-specific l+ plasma cells (PCs) in the spleen and BM were and permeabilized (Fix & Perm; Invitrogen), treated with 225 mg/ml quantified by ELISPOT after B6 adults were immunized once with NP- DNase (DN25; Sigma-Aldrich), stained intracellularly with FITC–anti- B 2/2 2/2 BrdU Abs and NP-APC, and washed twice before analysis on an LSR2 LPS in PBS. ( ) B6 (circles, solid line) or B6.TcRb d (squares, A A flow cytometer. Multiple files per sample were concatenated before data dashed line) adults were immunized and evaluated as in ( ). For ( ) and analysis using FlowJo 8.8. (B) all symbols represent data from an individual mouse. Note: On day 91, BM PC frequencies were significantly different between B6 and B6. CD40-CD154 blockade TcRb2/2d2/2 mice, as indicated by * (p = 0.002). At all other time points, On days 5, 7, and 9 post immunization, NP-CGG/alum–immunized mice were no significant differences in PC frequencies were found between these given i.v. inoculations (300 mg per injection) of anti-CD154 (MR-1) or control groups. (C) Representative ELISPOT wells from the BM (d 0, d 90) or 2 2 2 2 Hamster IgG (both from BioXcell), as described by Takahashi et al. (29). spleen (d 3) of B6.TcRb / d / mice. The Journal of Immunology 5391 difference is meaningful is, at present, unclear. We conclude that 94 d (range, 85–105 d) (6). Fig. 3 shows that NP-specific plasma a single immunization with NP-LPS is sufficient to induce the cells in NP-LPS–immunized and irradiated/reconstituted mice generation of hapten-specific BM plasma cells that are readily declined with an estimated t1/2 of 55 d (range 49-63 d). Therefore, detected .400 d later. by these measurements the half-life of NP-LPS–induced plasma cells is only modestly less than that observed during acute LCMV Decay rate of radioresistant TI plasma cells . Because residual Ag may induce naive B cells to contribute to Given that whole-body IR may change microenvironments in plasma cell pools after immunization, we sought to estimate the the BM and elsewhere, potentially altering plasma cell lifespan, decay rate of hapten-specific BM plasma cells following radiation- we also adopted an in vivo BrdU pulse-chase protocol to assess induced ablation of the functional pool. In this regard, the decay rate of BrdU+ NP-specific BM plasma cells without although past experiments show that long-lived plasma cells in exposing mice to IR. This general approach was used previously the BM are resistant to induced by ionizing radiation (IR) to gauge the half-life of Ag-specific plasma cells induced by a (6), it was unclear whether IR resistance is also a property of early TD Ag (5). We fed NP-LPS–immunized mice BrdU for 3 d, begin- plasma cells in the spleen. Significantly, whereas naive B cells ning at the time of immunization, and then assessed the fraction of readily died upon direct exposure to 200 R, the vast majority of BrdU+ NP-specific BM plasma cells at multiple time points later very early splenic plasma cells were resistant to IR-induced ap- by flow cytometry. As shown in Fig. 4, NP-specific BrdU+ plasma optosis, regardless of whether they were induced by immunizing cells were readily apparent in the BM between 34 and 94 d of the with NP-LPS or NP-CGG and exposed to doses up to 800 R (Fig. chase period. Therefore, many plasma cells formed within the first 2). It should be noted, consistent with the past observation that 3 d of the response to NP-LPS were able to enter and persist in the cultured GC B cells die rapidly without stimulation of CD40 (33), BM for at least 94 d. However, consistent with the data in Fig. 3, Downloaded from in our hands GC B cells died immediately in culture, with or the fraction of NP-specific BM plasma cells that were BrdU+ without IR (not shown). To assess the capacity of NP-LPS–in- declined progressively between days 34 and 94 of the chase pe- duced plasma cells to persist without input from naive B cells, we riod, resulting in an estimated t1/2 of 47 d (range, 41–59 d) (Fig. quantified NP-specific plasma cells out to 190 d after NP-LPS– 4B). Of interest, although surface expression of B220 is often used immunized B6 mice were exposed to 900 R whole-body IR. to identify early plasma cells (34), many hapten-specific BrdU+ Notably, past experiments using this approach indicate that the plasma cells continued to exhibit low but detectable surface levels http://www.jimmunol.org/ half-life of BM plasma cells induced after acute infection with of B220, even on day 94 of the chase period (Supplemental Fig. lymphocytic choriomeningitis , a complex TD viral Ag, is 2), suggesting that surface levels of B220 are not consistently

FIGURE 2. Early NP-specific plasma cells (PCs) by guest on October 2, 2021 are radioresistant. (A) Flow cytometric approach for sorting NP-specific CD138+ PCs from of mice immunized 7 d previously with NP-CGG/ alum. Cells from the CD138+ gate were sorted for subsequent irradiation experiments. In addition to NP-binding PCs, radiosensitivity was also deter- mined for naive IgD+ NP2 B220+ CD19+ B cells sorted from alum- and NP-CGG–immunized mice (gate not shown). (B) Alum- or NP-CGG/alum–im- munized mice were used as donors for the indicated cell populations, including NP-binding CD138+ PCs. Sorted cells were either rested (top) or subjected to 200 R (bottom) before resting. Viability was deter- mined 9 h later by flow cytometry, using DAPI. (C) Frequencies of cells secreting NP-specific Abs were determined by ELISPOT, using freshly sorted (open), sorted and cultured for 9 h (black), or sorted, exposed to 200R IR, and then rested for 9 h (gray) CD138+ NP-binding cells. (D) Separate experiment in which B6 adults immunized with NP-LPS in PBS or NP-CGG in alum were sacrificed on day 3 post immunization. Single splenocyte suspensions were exposed to 0, 200, 600, or 800 rads, rested 9 h, and then transferred to NP-coated ELISPOT plates to assay frequencies of NP-specific l+ PCs. Nonirra- diated samples are shown with closed symbols; ir- radiated samples with open symbols. Differences in PC frequencies in (C)and(D) were not statistically significant. 5392 LONG-LIVED PLASMA CELLS

Longevity of pre-GC plasma cells in TD responses Given that many NP-LPS–induced plasma cells become long lived without maturing in GCs, we also examined whether plasma cells induced with a TD Ag achieve longevity independently of the GC response. In this regard, several past studies suggest that B cell maturation in GCs is associated tightly with the formation of long-lived plasma cells in TD Ab responses. For instance, the BM is highly enriched for plasma cells synthesizing high-affinity isotype-switched Abs (13), and high-affinity GC B cells prefer- entially generate plasma cells (37). However, these studies do not exclude the possibility that low-affinity pre-GC plasma cells possess the potential to seed long-lived plasma cell pools in the BM. Furthermore, although B cells lacking the transcriptional repressor BCL6 fail to generate both GCs and long-lived BM plasma cells (38, 39), BCL6 mutation may lead to secondary effects that preempt the generation of long-lived plasma cells during the early phases of B cell differentiation. For these experiments, we sought to prevent the GC response on

FIGURE 3. Decay rate for NP-LPS–induced plasma cells. Cohorts of day 5 post immunization with NP-CGG, before the exponential Downloaded from NP-LPS–immunized B6 adults were left untouched or irradiated/recon- increase observed in responding GC B cells (Fig. 5B), and 3 d stituted, as described in Materials and Methods.(A) Shown are ELISPOT + before Ig somatic mutation events are detected in these cells (36). results for NP-specific l plasma cells (PCs) in the BM comparing non- First, we abrogated the GC response via whole-body IR. As ob- irradiated controls (circles with solid line) with irradiated (squares with served for naive B cells (Fig. 2), Ag-activated GC B cells in vivo dashed line) B6 mice. Lines are drawn through the mean for each group. On day 190, BM PC frequencies were significantly different between ir- were also sensitive to IR-induced cell death, as the NP-specific GC B cell pool was readily depleted upon exposure of NP-CGG– radiated and control mice, as indicated by * (p = 0.05). At all other time http://www.jimmunol.org/ points, no significant differences in PC frequencies were found between immunized mice to 900 R (Supplemental Fig. 3). Fig. 6A illus- these groups. (B) The decay rate of NP-LPS–induced PCs was illustrated trates the frequency of NP-specific plasma cells in the BM of mice by dividing the number of NP-specific PCs in each irradiated mouse by the irradiated with 900 R and reconstituted 5 d after immunization average number of NP-specific PCs in three to five controls for each in- with NP-CGG. As shown, Igl+ NP-specific plasma cells were dicated time point, as previously described by Slifka et al. (6). The half-life readily detected in the BM for $105 d post IR. Consistent with 3 of BM plasma cells post IR was calculated as follows: t1/2 =(T log 2)/ the absence of GC B cell maturation and selection, NP-specific log (starting quantity/ending quantity) where T = elapsed time. plasma cells in irradiated/reconstituted mice secreted low- affinity IgM+ Abs (Fig. 6B). Notably, however, a readily detect- + able number of IgM plasma cells were observed in the BM of by guest on October 2, 2021 downregulated on all plasma cells during the earliest phases of control (nonirradiated) mice. Second, we inhibited the generation plasma cell differentiation. Altogether, these data, along with the of Ag-induced GCs by blocking B cell–T cell interactions with analyses in Fig. 3, indicate that NP-LPS–induced plasma cells anti-CD154 blocking Abs, beginning on day 5 post immunization. ∼ possess a t1/2 of 50 d. As shown, despite prevention of GC formation via this strategy Lack of evidence for induction of GCs in T cell-deficient mice (Supplemental Fig. 4A, 4B), NP-specific BM plasma cells were detected for $91 d post immunization (Fig. 6C), and these cells Past work with carbohydrate-enriched (type 2) TI Ags indicate that did not secrete high-affinity IgG1 Abs (Fig. 6D). Significantly, we these Ags are sufficient to induce the transient development of GC- did not detect NP-specific GC B cells in any mouse given anti- like structures. These B cell clusters, however, resolve quickly, fail CD154 Abs throughout this study (Supplemental Fig. 4B). Of to support SHM, and require high-dose Ag for their formation (18, note, frequencies of BM l+ NP-specific plasma cells were de- 19). The capacity of LPS-based Ags to induce GC formation, creased significantly at the earliest time point (day 14) analyzed, though, was less clear. To address whether TI GCs are induced by perhaps suggesting suboptimal clonal expansion due to CD40 NP-LPS immunization, we quantified Ag-specific GCs in B6 or 2/2 2/2 blockade. Regardless, these results suggest that maturation and B6.TcRb d mice immunized in parallel with this type 1 TI selection of high-affinity B cells in GCs are not requisite for Ag and compared these results with those in B6 mice immunized propagating long-lived plasma cells in TD Ab responses. In fur- with NP-CGG. Hapten-specific GC B cells were identified as NP- + high ther support of this notion, we also detected hapten-specific binding CD19 PNA cells that lacked surface IgD expression plasma cells in the BM of NP-CGG–immunized mice lacking (31, 35). Consistent with past data (36), NP-specific GC B cells activation-induced deaminase, which is strictly required for the were not detected in B6 mice immunized with NP-CGG until after generation of high-affinity B cells by SHM (40) (Supplemental day 5 post immunization (Fig. 5B). By contrast, low frequencies Fig. 4C). Collectively, these data suggest that competence to enter of GC B cells were identified in B6 mice inoculated with NP-LPS the long-lived BM plasma cell pool occurs in early plasma cells 3 d post immunization, but not at later time points. However, it + high and is not associated with maturation and affinity-based selection should be noted that frequencies of NP-specific CD19 PNA in the GC. B cells in B6 mice immunized with NP-LPS were highly variable, perhaps owing to the inefficiency of such responses in the absence of cognate T cell help. In contrast, we were unable to detect GC Discussion B cells in NP-LPS–immunized B6.TcRb2/2d2/2 mice at all time TI Ags have long been deemed incapable of inducing the gener- points examined (Fig. 5A, 5B). These data, together with the data ation of long-lived plasma cells (1). However, our results show that in Figs. 1, 3, and 4, establish that long-lived TI plasma cells can many early plasma cells, including those induced with an LPS- form independently of B cell maturation in GCs. based conjugate, persist for extended periods in the BM at fre- The Journal of Immunology 5393 Downloaded from

FIGURE 4. Pulse-chase kinetics for hapten-specific NP-LPS–induced plasma cells. NP-LPS–immunized B6 adults were fed BrdU in the drinking water http://www.jimmunol.org/ for 3 d post immunization. (A) The frequency of NP-binding splenic plasma cells that were BrdU+ on day 3 post immunization is shown. Data are the mean of four mice. (B) Frequencies of NP-binding BM plasma cells that were BrdU+ on days 34, 64, and 94 d of the chase are shown. Each circle represents an individual mouse, and the mean for each time point is indicated by a solid line. The half-life of BM plasma cells post IR was calculated as described for Fig. 3. (C) Representative flow cytometric data illustrating the identification of NP-binding BrdU-positive plasma cells in the BM at the indicated time points of the chase period. Note that BrdU+ cells were identified by comparing immunized controls that were not exposed to BrdU. The “Dump” channel includes Abs to CD4, CD8, F4/80, TER-119, and Gr-1. Each plot was derived from files containing 7–9 3 106 events. quencies that mirror TD responses, despite the significant con- surface B220 expression is typically considered a trait restricted to by guest on October 2, 2021 traction in Ag-specific plasma cell frequencies observed early in recently formed plasma cells (34), we found that many long-lived such responses (Fig. 1A) (15). Thus, many early plasma cells BM plasma cells exhibited a bimodal pattern for surface B220 achieve competence to enter long-lived BM plasma cell pools expression (Supplemental Fig. 2). Together, these findings un- early in TI responses and in the absence of T cell-derived signals derscore the need for increased understanding of the cellular dy- associated with the generation of long-lived immunity. These namics of BM plasma cells and the potential influence of Ag and findings contrast with the recently proposed “imprinting model” other factors on maintenance of the plasma cell pool. positing that TD Ags are unique in their capacity to induce B cells Although TI pathogens were previously thought to be ineffectual to yield long-lived plasma cells (12). Indeed, our findings suggest inducers of long-term protective immunity, recent results, including that TLR4 and T cell-derived signals such as those provided by those presented in this article, challenge this long-standing para- CD40 activation on B cells may provide comparable signals re- digm. TI Ab responses mediated by B1 B cells, for example, were quired for inducing plasma cell longevity. sufficient to provide long-term IgM-dependent protection against We used two different approaches to estimate the decay rate of B. hermsii (20, 21). However, whereas these investigators con- hapten-specific NP-LPS–induced plasma cells in the BM. We were cluded that long-term protection against this spirochete was ach- surprised to learn that, although hapten-specific plasma cells were ieved through the continual induction of short-lived plasma cells, readily detected in the BM for .400 d following a single im- an alternative possibility supported by our data is that a significant munization with NP-LPS (Fig. 1), hapten-specific BM plasma fraction of plasma cells induced by B. hermsii infection are long cells exhibited a t1/2 of ∼50 d (Figs. 3, 4). Whereas these calcu- lived. This perspective may apply to other TI pathogens, such as S. lations contrast substantially with the past notion that early plasma pneumoniae and E. muris. Indeed, recent work shows that plasma cells in TI responses fail to survive beyond 5 d of their generation cells induced via immunization of conventional mice with S. (1, 2, 41), they also suggest that the long-term maintenance of pneumoniae or E. muris colonize and persist in the BM for ex- serum Abs specific for such Ags cannot be explained solely by the tended periods (24–26). Whereas the cellular basis underlying generation of exceptionally long-lived plasma cells generated long-term protection in many of these systems is not fully un- early in such responses. Accordingly, we favor the notion that derstood, our data reveal the possibility that long-lived plasma the BM plasma cell pool consists of both short- and long-lived cells formed in the complete absence of T cells may play critical cells. This viewpoint is supported by recent work from Racine roles in responses to TI pathogens. et al. (24), who characterized a surface IgM+ BM plasmablast These recent studies, together with the conclusions drawn in this population with a potential role in maintaining Ab titers to the article, contrast substantially with the currently dominant model, intracellular bacterium E. muris. Further evidence for heteroge- which posits that T cell-regulated GC maturation and selection neity within the BM plasma cell pool stems from our analysis of events are associated with the propagation of long-lived plasma BrdU-retaining hapten-specific BM plasma cells. Thus, although cells (4, 10, 12). Why, then, were TI Ags thought to induce only 5394 LONG-LIVED PLASMA CELLS

2/2 2/2

FIGURE 5. Failure to detect GCs in immunized T cell-deficient mice. B6 or B6.TcRb d adults were immunized with either NP-LPS/PBS or NP- Downloaded from CGG/alum, and frequencies of NP-binding CD19+ PNAhigh GC B cells in the spleen were determined by flow cytometry. (A) Flow cytometric data for the indicated mice, representative of three to six mice per group. Left-most plots are pregated on viable (DAPI2) IgD2 cells. A total of 106 events were collected for each sample. Abs used for the “Dump” channel are as described for Fig. 4. (B) Numbers of NP-specific GC B cells at 3, 5, or 7 d were calculated by multiplying the frequency of hapten-binding GC B cells, using the gating strategy shown in (A) by the total number of splenocytes harvested in each mouse. Results are expressed as means +/2 SEM of three to six mice per group performed over two separate experiments. Absolute numbers for days 3, 5, and 7 are graphed separately to better reveal the small number of PNA-binding B cells in NP-LPS–immunized mice at early time points, which fell below detection levels in all mice after day 5. http://www.jimmunol.org/ short-lived plasma cells? As reported by Fidler (15), following number of TI plasma cells enter the BM and persist for extended a robust clonal burst the number of hapten-specific NP-LPS–in- periods. Indeed, other past studies have documented small num- duced plasma cells in the spleen declines by ∼90% by day 5 in this bers of hapten-specific BM plasma cells 3 wk after immunization response (see Fig. 1A), suggesting that the vast majority of early with haptenated LPS, but these workers did not address whether NP-LPS–induced plasma cells are indeed short-lived. However, the plasma cells were maintained beyond 21 d, nor did they seek this observation does not exclude the possibility that a meaningful to measure the half-life of these cells (42). Furthermore, although by guest on October 2, 2021

FIGURE 6. Long-lived TD plasma cells (PCs) without maturation in GCs. (A) Frequencies of NP-specific l+ PCs post IR in the BM of B6 adults immunized with NP-CGG/alum and then left untouched (closed) or irradiated and reconstituted (open) on day 5 post immunization (see Materials and

Methods). (B) BM cells from the mice in (A) on day 105 post IR were assayed by ELISPOT, using either NP33-BSA–coated plates evaluated with anti-IgM + + specific Abs (left)orNP4-BSA–coated plates evaluated with anti-IgG1–specific Abs. Differences in l and IgM PC frequencies between irradiated and control groups were not significant. The p value for differences in frequencies of NP-specific IgG1+ PCs is shown on the graph. (C and D) NP-CGG/alum– immunized B6 adults were given three injections of anti-CD154 (MR-1) Abs (open) or hamster IgG (closed), beginning 5 d post immunization (see + Materials and Methods). (C) Shown are BM NP-specific l PCs assayed with NP33-BSA. Differences between MR1-treated and control mice were sig- nificant on day 14 (p = 0.02) and day 190 (p = 0.04). At all other time points, no significant differences were found between these groups. (D) NP-specific + + IgM PCs were assayed with NP33-BSA (left) or NP-specific IgG1 PCs were evaluated with NP4-BSA (right). The p values for statistically significant differences are shown on graphs. All results are expressed as means 6 SEM of three to four mice per group. The Journal of Immunology 5395 the majority of early plasma cells induced by either TI or TD Ags 7. Hammarlund, E., M. W. Lewis, S. G. Hansen, L. I. Strelow, J. A. Nelson, G. J. Sexton, J. M. Hanifin, and M. K. Slifka. 2003. Duration of antiviral im- may die before exiting the spleen (2, 41, 43, 44), only 9% of early munity after smallpox vaccination. Nat. Med. 9: 1131–1137. splenic plasma cells in a TD response show clear signs of apo- 8. Kjeldsen, K., O. Simonsen, and I. Heron. 1985. Immunity against diphtheria ptosis (43). By quantifying frequencies of hapten-specific plasma 25–30 years after primary vaccination in childhood. Lancet 1: 900–902. 9. Simonsen, O., K. Kjeldsen, and I. Heron. 1984. Immunity against tetanus and cells in the BM following immunization with LPS or - effect of revaccination 25–30 years after primary vaccination. Lancet 2: 1240– based hapten conjugates, we have found that plasma cells per- 1242. sist in the BM for extended periods in response to both TI and TD 10. Crotty, S., R. J. Johnston, and S. P. Schoenberger. 2010. Effectors and memories: Bcl-6 and Blimp-1 in T and B lymphocyte differentiation. Nat. Immunol. 11: Ags and without undergoing GC maturation and selection. 114–120. Our data also suggest that to TI self-antigens, 11. Tarlinton, D. M. 2008. Evolution in miniature: selection, survival and distribu- tion of reactive cells in the germinal centre. Immunol. Cell Biol. 86: 133– such as rheumatoid factor Abs induced by DNA-chromatin com- 138. plexes (45), might arise from long-lived plasma cells despite a 12.Amanna,I.J.,andM.K.Slifka.2010.Mechanisms that determine plasma cell fully tolerant T cell compartment. This notion emphasizes the lifespan and the duration of . Immunol. Rev. 236: 125– 138. importance of central and peripheral B cell tolerance mechanisms, 13. Smith, K. G., A. Light, G. J. Nossal, and D. M. Tarlinton. 1997. The extent of especially in scenarios in which such self-antigens are sufficient affinity maturation differs between the memory and -forming cell to induce somatic hypermutation without engaging Th cells (46). compartments in the primary . EMBO J. 16: 2996–3006. 14. Racine, R., and G. M. Winslow. 2009. IgM in microbial : taken for The notion that most Ags are capable of inducing long-lived granted? Immunol. Lett. 125: 79–85. plasma cells may also apply to TI self-antigens underlying the 15. Fidler, J. M. 1975. In vivo immune response to TNP hapten coupled to - independent carrier lipopolysaccharide. Cell. Immunol. 16: 223–236. formation of polyreactive natural Abs, which have been shown to 16. Burlen, O., A. Coutinho, and A. A. Freitas. 1988. Long-lasting thymus- be protective against several pathogens (47). independent immune responses to anti- lipopolysaccharide conjugates We used resistance to radiation-induced cell death as a tool require continuous B cell renewal. Eur. J. Immunol. 18: 1433–1439. Downloaded from 17. Garcı´a de Vinuesa, C., P. O’Leary, D. M. Sze, K. M. Toellner, and to investigate plasma cell longevity in the context of TI and TD I. C. MacLennan. 1999. T-independent type 2 antigens induce B cell prolifera- immune responses. In this context, we were surprised to find that tion in multiple splenic sites, but exponential growth is confined to extra- radioresistance in plasma cells is established exceptionally early follicular foci. Eur. J. Immunol. 29: 1314–1323. 18. Lentz, V. M., and T. Manser. 2001. Cutting edge: germinal centers can be in- during these responses. It should be noted that, at present, the duced in the absence of T cells. J. Immunol. 167: 15–20. underlying mechanisms allowing plasma cells to avoid death de- 19. de Vinuesa, C. G., M. C. Cook, J. Ball, M. Drew, Y. Sunners, M. Cascalho,

M. Wabl, G. G. Klaus, and I. C. MacLennan. 2000. Germinal centers without http://www.jimmunol.org/ spite widespread DNA damage are not understood. However, other T cells. J. Exp. Med. 191: 485–494. studies have shown that antiapoptotic DNA repair pathways are 20. Alugupalli, K. R., R. M. Gerstein, J. Chen, E. Szomolanyi-Tsuda, induced by TLR signaling in memory CD4+ T cells (48). Hence, R. T. Woodland, and J. M. Leong. 2003. The resolution of relapsing bor- reliosis requires IgM and is concurrent with expansion of B1b . TLR signaling may also play a similar role in activated B cells. J. Immunol. 170: 3819–3827. In sum, we propose that LPS-based Ags readily elicit the for- 21. Alugupalli, K. R., J. M. Leong, R. T. Woodland, M. Muramatsu, T. Honjo, and mation of a long-standing pool of plasma cells secreting low- R. M. Gerstein. 2004. B1b lymphocytes confer T cell-independent long-lasting immunity. Immunity 21: 379–390. affinity but high-avidity IgM Abs that may play fundamental roles 22. Bitsaktsis, C., B. Nandi, R. Racine, K. C. MacNamara, and G. Winslow. 2007. in combating certain pathogens. Accordingly, we suggest that the T-cell-independent humoral immunity is sufficient for protection against fatal role of such plasma cells in host protection be re-evaluated. In intracellular ehrlichia infection. Infect. Immun. 75: 4933–4941. 23. Cole, L. E., Y. Yang, K. L. Elkins, E. T. Fernandez, N. Qureshi, M. J. Shlomchik, by guest on October 2, 2021 addition, the control of gene expression patterns in short- and long- L. A. Herzenberg, L. A. Herzenberg, and S. N. Vogel. 2009. Antigen-specific lived plasma cells, the signals required to induce the formation B-1a induced by Francisella tularensis LPS provide long-term pro- tection against F. tularensis LVS challenge. Proc. Natl. Acad. Sci. USA 106: of long-lived cells, and, conversely, the mechanisms underlying 4343–4348. the apparent loss of many plasma cells during early phases of Ab 24. Racine, R., M. McLaughlin, D. D. Jones, S. T. Wittmer, K. C. MacNamara, responses all remain important areas of investigation. D. L. Woodland, and G. M. Winslow. 2011. IgM production by bone marrow plasmablasts contributes to long-term protection against intracellular bacterial infection. J. Immunol. 186: 1011–1021. 25. Taillardet, M., G. Haffar, P. Mondie`re, M. J. Asensio, H. Gheit, N. Burdin, Acknowledgments T. Defrance, and L. Genestier. 2009. The thymus-independent immunity con- We thank Dr. Avinash Bhandoola for helpful discussions and critically ferred by a pneumococcal polysaccharide is mediated by long-lived plasma cells. reviewing the manuscript, Dr. Nina Papavasiliou for generously providing 114: 4432–4440. mice, and Charles Pletcher, Drew Bantley, and other members of the 26. Foote, J. B., T. I. Mahmoud, A. M. Vale, and J. F. Kearney. 2012. Long-term maintenance of polysaccharide-specific antibodies by IgM-secreting cells. J. Abramson Center Flow Cytometry and Cell Sorting Shared Re- Immunol. 188: 57–67. source Laboratory for expert technical support. 27. Goodnow, C. C., J. Crosbie, S. Adelstein, T. B. Lavoie, S. J. Smith-Gill, R. A. Brink, H. Pritchard-Briscoe, J. S. Wotherspoon, R. H. Loblay, K. Raphael, et al. 1988. Altered immunoglobulin expression and functional silencing of self- Disclosures reactive B lymphocytes in transgenic mice. Nature 334: 676–682. The authors have no financial conflicts of interest. 28. Lindsley, R. C., M. Thomas, B. Srivastava, and D. Allman. 2007. Generation of peripheral B cells occurs via two spatially and temporally distinct pathways. Blood 109: 2521–2528. 29. Takahashi, Y., P. R. Dutta, D. M. Cerasoli, and G. Kelsoe. 1998. In situ studies of References the primary immune response to (4-hydroxy-3-nitrophenyl)acetyl. V. Affinity 1. Radbruch, A., G. Muehlinghaus, E. O. Luger, A. Inamine, K. G. Smith, maturation develops in two stages of . J. Exp. Med. 187: 885– T. Do¨rner, and F. Hiepe. 2006. Competence and competition: the challenge of 895. becoming a long-lived plasma cell. Nat. Rev. Immunol. 6: 741–750. 30. Jacob, J., R. Kassir, and G. Kelsoe. 1991. In situ studies of the primary immune 2. Ho, F., J. E. Lortan, I. C. MacLennan, and M. Khan. 1986. Distinct short-lived response to (4-hydroxy-3-nitrophenyl)acetyl. I. The architecture and dynamics of and long-lived antibody-producing cell populations. Eur. J. Immunol. 16: 1297– responding cell populations. J. Exp. Med. 173: 1165–1175. 1301. 31. McHeyzer-Williams, M. G., M. J. McLean, P. A. Lalor, and G. J. Nossal. 1993. 3. Amanna, I. J., N. E. Carlson, and M. K. Slifka. 2007. Duration of humoral Antigen-driven B cell differentiation in vivo. J. Exp. Med. 178: 295–307. immunity to common viral and vaccine antigens. N. Engl. J. Med. 357: 1903– 32. Jack, R. S., T. Imanishi-Kari, and K. Rajewsky. 1977. Idiotypic analysis of the 1915. response of C57BL/6 mice to the (4-hydroxy-3-nitrophenyl)acetyl group. Eur. J. 4. Goodnow, C. C., C. G. Vinuesa, K. L. Randall, F. Mackay, and R. Brink. 2010. Immunol. 7: 559–565. Control systems and decision making for antibody production. Nat. Immunol. 11: 33. Liu, Y. J., D. E. Joshua, G. T. Williams, C. A. Smith, J. Gordon, and 681–688. I. C. MacLennan. 1989. Mechanism of antigen-driven selection in germinal 5. Manz, R. A., A. Thiel, and A. Radbruch. 1997. Lifetime of plasma cells in the centres. Nature 342: 929–931. bone marrow. Nature 388: 133–134. 34. Kallies, A., J. Hasbold, D. M. Tarlinton, W. Dietrich, L. M. Corcoran, 6. Slifka, M. K., R. Antia, J. K. Whitmire, and R. Ahmed. 1998. Humoral immunity P. D. Hodgkin, and S. L. Nutt. 2004. Plasma cell ontogeny defined by quanti- due to long-lived plasma cells. Immunity 8: 363–372. tative changes in blimp-1 expression. J. Exp. Med. 200: 967–977. 5396 LONG-LIVED PLASMA CELLS

35. McHeyzer-Williams, L. J., M. Cool, and M. G. McHeyzer-Williams. 2000. 42. Koch, G., B. D. Lok, A. van Oudenaren, and R. Benner. 1982. The capacity and Antigen-specific B cell memory: expression and replenishment of a novel mechanism of bone marrow antibody formation by thymus-independent anti- b220(-) compartment. J. Exp. Med. 191: 1149–1166. gens. J. Immunol. 128: 1497–1501. 36. Jacob, J., J. Przylepa, C. Miller, and G. Kelsoe. 1993. In situ studies of the 43. Smith, K. G., T. D. Hewitson, G. J. Nossal, and D. M. Tarlinton. 1996. The primary immune response to (4-hydroxy-3-nitrophenyl)acetyl. III. The kinetics phenotype and fate of the antibody-forming cells of the splenic foci. Eur. J. of V region mutation and selection in germinal center B cells. J. Exp. Med. 178: Immunol. 26: 444–448. 1293–1307. 44. Garcı´a De Vinuesa, C., A. Gulbranson-Judge, M. Khan, P. O’Leary, 37. Phan, T. G., D. Paus, T. D. Chan, M. L. Turner, S. L. Nutt, A. Basten, and M. Cascalho, M. Wabl, G. G. Klaus, M. J. Owen, and I. C. MacLennan. 1999. R. Brink. 2006. High affinity germinal center B cells are actively selected into Dendritic cells associated with plasmablast survival. Eur. J. Immunol. 29: 3712– the plasma cell compartment. J. Exp. Med. 203: 2419–2424. 3721. 38. Dent, A. L., A. L. Shaffer, X. Yu, D. Allman, and L. M. Staudt. 1997. Control of 45. Herlands, R. A., S. R. Christensen, R. A. Sweet, U. Hershberg, and inflammation, expression, and germinal center formation by BCL-6. M. J. Shlomchik. 2008. T cell-independent and toll-like -dependent Science 276: 589–592. antigen-driven activation of autoreactive B cells. Immunity 29: 249–260. 39. Toyama, H., S. Okada, M. Hatano, Y. Takahashi, N. Takeda, H. Ichii, 46. William, J., C. Euler, S. Christensen, and M. J. Shlomchik. 2002. Evolution of T. Takemori, Y. Kuroda, and T. Tokuhisa. 2002. Memory B cells without somatic responses via somatic hypermutation outside of germinal centers. hypermutation are generated from Bcl6-deficient B cells. Immunity 17: 329–339. Science 297: 2066–2070. 40. Muramatsu, M., K. Kinoshita, S. Fagarasan, S. Yamada, Y. Shinkai, and 47. Ochsenbein, A. F., T. Fehr, C. Lutz, M. Suter, F. Brombacher, H. Hengartner, and T. Honjo. 2000. Class switch recombination and hypermutation require R. M. Zinkernagel. 1999. Control of early viral and bacterial distribution and activation-induced deaminase (AID), a potential RNA editing enzyme. disease by natural antibodies. Science 286: 2156–2159. Cell 102: 553–563. 48. Zheng, L., N. Asprodites, A. H. Keene, P. Rodriguez, K. D. Brown, and 41. Sze, D. M., K. M. Toellner, C. Garcı´a de Vinuesa, D. R. Taylor, and E. Davila. 2008. TLR9 engagement on CD4 T lymphocytes represses gamma- I. C. MacLennan. 2000. Intrinsic constraint on plasmablast growth and extrinsic radiation-induced apoptosis through activation of checkpoint kinase response limits of plasma cell survival. J. Exp. Med. 192: 813–821. elements. Blood 111: 2704–2713. Downloaded from http://www.jimmunol.org/ by guest on October 2, 2021