Longevity Cells in the Maintenance of Plasma Cell the Role of Bone

The Role of Bone Marrow-Derived Stromal Cells in the Maintenance of Plasma Cell Longevity

This information is current as Heather A. Minges Wols, Gregory H. Underhill, Geoffrey S. of September 29, 2021. Kansas and Pamela L. Witte J Immunol 2002; 169:4213-4221; ; doi: 10.4049/jimmunol.169.8.4213 http://www.jimmunol.org/content/169/8/4213 Downloaded from

References This article cites 60 articles, 26 of which you can access for free at: http://www.jimmunol.org/content/169/8/4213.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

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

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

by guest on September 29, 2021 *average

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

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The Role of Bone Marrow-Derived Stromal Cells in the Maintenance of Plasma Cell Longevity1

Heather A. Minges Wols,*† Gregory H. Underhill,‡ Geoffrey S. Kansas,‡§ and Pamela L. Witte2*†

Protective circulating Abs originate primarily from long-lived plasma cells in the bone marrow. However, the molecular and cellular basis of plasma cell longevity is unknown. We investigated the capacity of primary bone marrow-derived stromal cells to maintain plasma cell viability in vitro. Plasma cells puriﬁed from the bone marrow or lymph nodes died rapidly when plated in media, but a subpopulation of plasma cells survived and secreted high levels of Ab for up to 4 wk when cocultured with stromal cells. Ab secretion was inhibited by the addition of anti-very late Ag-4 to plasma cell/stromal cell cocultures indicating that direct interactions occur and are necessary between stromal cells and plasma cells. The addition of rIL-6 to plasma cells cultured in

media alone partially relieved the sharp decline in Ab secretion observed in the absence of stromal cells. Moreover, when stromal Downloaded from cells from IL-6؊/؊ mice were used in plasma cell/stromal cell cocultures, Ab levels decreased 80% after 7 days as compared with wild-type stromal cells. Further, IL-6 mRNA message was induced in stromal cells by coculture with plasma cells. These data indicate that bone marrow plasma cells are not intrinsically long-lived, but rather that plasma cells contact and modify bone marrow stromal cells to provide survival factors. The Journal of Immunology, 2002, 169: 4213–4221.

he humoral branch of the immune system is critical for for Ͼ300 days post viral infection (12). In the same studies, http://www.jimmunol.org/ providing protective circulating Abs. In humans, Ag-spe- plasma cells transferred to naive mice maintained serum Ab levels T cific Ab has been detected in the serum for decades after for Ͼ120 days posttransfer, confirming the idea that plasma cells Ag encounter (1). However, because the half-life of circulating Ab elicited by a single Ag are long-lived and not a constantly replen- is estimated to be Ͻ3 wk (2Ð4), the Ab pool must be constantly ished population. Manz et al. (19) found that immunization of replenished to maintain serum Ab levels. The Ab pool is main- naive recipients with the original immunogen did not affect Ab tained by plasma cells, which continually secrete Abs. The fact secretion rates from the transferred plasma cells, nor was necessary that protective immunity is long-lasting, coupled with the short to maintain plasma cell longevity. half-life of Abs, implies either that plasma cells are long-lived cells or Despite the important role of plasma cells, they make up a small that the plasma cell pool is continually replenished to maintain certain percentage of the total cells in secondary lymphoid tissue and bone by guest on September 29, 2021 levels of secreted Ab. Evidence supporting the latter mechanism ex- marrow (12). The small number of Ab-secreting cells in these tis- ists (5Ð10), but recent studies strongly support the view that long-lived sues makes it difficult to isolate plasma cells for in vitro studies. To plasma cells provide much of the circulating Ab (11, 12). circumvent this problem, most studies have used myeloma cell It is well-established that the bone marrow becomes a major site lines or patient material. There are relatively few studies that have of Ab production (13Ð18). In the past few years, it has become characterized terminally differentiated plasma cells or the marrow apparent that the population of plasma cells that exists in the bone microenvironment that may support their longevity. One critical marrow is long-lived (11, 12). This hypothesis is supported by a element of the bone marrow microenvironment is the bone marrow study in mice showing that Ag-specific bone marrow plasma cells reticular stromal cell. Stromal cells provide growth factors as well survived Ͼ90 days postimmunization without turnover as assessed as cell contact-dependent signals needed for hemopoiesis, includ- using bromodeoxyuridine labeling (11). This finding suggested ing progression of B-lymphoid lineage cells in vivo and in vitro that bone marrow plasma cells are not a dynamic dividing popu- (20Ð25). It is possible that stromal cells also support plasma cell lation, but instead are long-lived cells that constitutively produce survival. Although the role of stromal cells in the maturation of Ab. Ag-specific bone marrow plasma cells have also been detected human plasmablasts into terminally differentiated plasma cells is supported by several studies (26Ð29), the role of stromal cells in maintaining plasma cell longevity remains largely unexplored. *Program for Immunology and Aging, Department of Cell Biology, Neurobiology, The goal of the present study was to determine whether plasma and Anatomy, and †Department of Microbiology and Immunology, Loyola University Medical Center, Maywood, IL 60153; ‡Department of Biomedical Engineering, cells need extrinsic components to survive and whether reticular Northwestern University, Evanston, IL 60201; and ¤Department of Microbiology- stromal cells are able to support plasma cell longevity. As we Immunology, Northwestern Medical School, Chicago, IL 60611 recently described, E-selectin/P-selectin double-deficient mice Received for publication January 23, 2002. Accepted for publication August 9, 2002. (E/PϪ/Ϫ) (30, 31) represent a source of plasma cells that can be The costs of publication of this article were defrayed in part by the payment of page purified in sufficient quantities for in vitro studies (32). Plasma charges. This article must therefore be hereby marked advertisement in accordance Ϫ/Ϫ with 18 U.S.C. Section 1734 solely to indicate this fact. cells from E/P mice exhibit classic morphology and phenotype 1 This work was supported by National Institutes of Health Grants RO1AG13874 and of terminally differentiated plasma cells and do not proliferate K07AG00997 (to P.L.W.) and HL58710 (to G.S.K.). H.A.M.W. was supported in part (32). They appear similar to plasma cells isolated from wild-type by National Institutes of Health Training Grant T32AI07508 and the Schmitt Disser- mice (19). In this study, we demonstrate that primary bone marrow tation Fellowship. reticular stromal cells are able to sustain viability and Ab produc- 2 Address correspondence and reprint requests to Dr. Pamela L. Witte, Department of tion by plasma cells isolated from either the lymph nodes of Cell Biology, Neurobiology, and Anatomy, Loyola University Medical Center, 2160 Ϫ Ϫ South First Avenue, Maywood, IL 60153. E-mail address: [email protected] E/P / mice or from wild-type bone marrow for at least 4 wk.

Further, both IL-6 and very late Ag (VLA)3-4 appear to have key adherent layers. Additionally, our anti-CD44 at 5 ␮g/ml was sufficient to roles in supporting plasma cell survival. inhibit hyaluronic acid binding to BW5147 cells by 100% (39). Blocking Abs were added at this concentration (5 ␮g/ml) to the plasma cell/stromal cell cocultures at the initiation of coculture and during each medium Materials and Methods change. Isotype-matched control Abs were purchased from BD PharMin- Animals gen and added to control cocultures (5 ␮g/ml). Female BALB/c mice and female C57BL/6 and C57BL/6tmIL-6Ϫ/Ϫkopf (IL- Ϫ Ϫ Ј 6 / ) (33) mice were purchased from Harlan Laboratories (Indianapolis, Cell cycle analysis and 5 -bromo-2-deoxyuridine (BrdU) Ϫ Ϫ IN) and from The Jackson Laboratory (Bar Harbor, ME), respectively, at incorporation of E/P / plasma cells 4Ð5 wk of age. Female BALB/c mice were purchased from the National Institute of Aging (Bethesda, MD) at 22 mo of age. Upon receipt, the Freshly isolated plasma cells cultured in the presence or absence of rmIL-6 were harvested and fixed in 100 ␮l1ϫ PBS, 100 ␮l heat-inactivated FBS, animals were housed at the Animal Research Facility (Loyola University ␮ Medical Center, Maywood, IL) under specific pathogen-free conditions. and 600 l 70% ice-cold ethanol for 60 min at 4¡C. The cells were washed E/PϪ/Ϫ on the C57BL/6 background were originally provided by Dr. D. twice and stained with 0.5 mg/ml propidium iodide (PI; Sigma-Aldrich, St. Bullard (University of Alabama, Birmingham, AL) and were bred and Louis, MO) with 0.1 mM EDTA and 0.05 mg/ml RNase A (Calbiochem, maintained at Northwestern University Medical School (Chicago, IL) (32). La Jolla, CA) for at least1hatroom temperature before FACS analysis. E/PϪ/Ϫ mice were used at 3Ð8 mo of age. Gates were set based on control lymphocyte samples stained in parallel and on the natural cellular population breaks. Samples were analyzed using Isolation of primary cultured stromal cells by flow cytometry CellQuest software. Additionally, freshly isolated plasma cells were cultured for 48 h in medium supplemented with 30 ␮g/ml BrdU (Sigma- Whitlock-type long-term bone marrow cultures that support B-lymphopoi- Aldrich) and in the presence or absence of rmIL-6. After 48 h, the cells esis (LTBMC-B) were initiated from pooled femoral and tibial bone mar- were harvested and stained according to the manufacturer’s suggested pro-

ϩ ϩ Ϫ Ϫ Downloaded from row of BALB/c, C57BL/6 / , or IL-6 / mice 4Ð5 wk of age and main- tocol (BD PharMingen). BrdU incorporation was determined by flow cy- tained by established methods (34, 35). After 4Ð6 wk, stromal cells were tometry analysis using Cell Quest software. FACS sorted (FACStarPlus; BD Biosciences, San Jose, CA) directly into ϫ 4 flat-bottom 96-well plates at a density of 1 10 cells/well based on ELISA differential forward vs side scatter and the lack of uptake of 1,1Ј-diocta- decyl-3,3,3Ј,3Ј-tetramethylindo-carbocyanine percholate-labeled acety- Isotype-specific ELISAs were modified from previous descriptions (40). lated low-density lipoprotein (Biomedical Technologies, Stoughton, MA) Supernatants collected at the indicated time points were frozen at Ϫ20¡C

as previously described (36). until analyzed by ELISA and were added undiluted or prediluted 1/2 or 1/4. http://www.jimmunol.org/ Biotin-conjugated rat anti-mouse isotype-specific Abs were used at 1 Plasma cell isolation ␮g/ml for detection and were developed with streptavidin-conjugated al- Plasma cells were isolated from E/PϪ/Ϫ cervical lymph node cells by de- kaline phosphatase (Southern Biotechnology Associates, Birmingham, AL) pletion of cells expressing CD5, CD11b, IgM, or B220 using magnetic and p-nitrophenyl phosphate (Sigma-Aldrich). Two-fold serial dilutions of beads (Miltenyi Biotec, Auburn, CA) (32). A purity of Ͼ94% was achieved isotype-specific standards were included on every test plate at a range of using this method. To isolate bone marrow plasma cells, a two-step isola- 1000Ð1.92 ng/ml. Samples were analyzed by quantitating values within the tion protocol was used. First, cells were purified based on CD138/Synde- linear range of the standard curve. The upper and lower limit of detection can-1 expression using biotinylated anti-CD138/Syndecan-1 (281Ð2; BD ranged from 1000Ð500 ng/ml to 1.92Ð7.8 ng/ml, respectively, in each as- PharMingen, San Diego, CA) (37) followed by an incubation with anti- say. No Ab was detected in the supernatants from stromal cells cultured in the absence of plasma cells. The following ELISA Abs were purchased

biotin microbeads (Miltenyi Biotec). Because bone marrow plasma cells by guest on September 29, 2021 are present in low numbers, the column-isolated cells were stained with from BD PharMingen: unlabeled rat anti-mouse IgG1 (A85-3), unlabeled streptavidin-allophycocyanin (BD PharMingen) and further purified by rat anti-mouse IgM (II/41), purified mouse IgG1 (A112-2), purified mouse FACS sorting for CD138ϩ cells. Aged BALB/c mice (22 mo old) were IgG2a (G155-178), purified mouse IgG2b (MPC-11), purified mouse IgG3 used because old mice have 10Ð20 times more plasma cells than young (A112-3), purified mouse IgM (G155-228), biotin-conjugated rat anti- mice (ϳ1Ð5% of total bone marrow cells). Isolated plasma cells were mouse IgG1 (A85-1), biotin-conjugated rat anti-mouse IgG2a (R19-15), placed onto slides by cytocentrifugation and stained with H&E to check for biotin-conjugated rat anti-mouse IgG2b (R12-3), biotin-conjugated rat anti- plasma cell morphology. More than 95% of cells with the morphology of mouse IgG3 (R40-82), and biotin-conjugated rat anti-mouse IgM differentiated plasma cells (small, dense, eccentric nucleus; prominent (R6-60.2). Additionally, unlabeled donkey anti-mouse IgG was purchased golgi; volumous cytoplasm; see also Fig. 1) were achieved using this from Jackson ImmunoResearch Laboratories (West Grove, PA). method. Enumeration of plasma cells Plasma cell cultures Plasma cells cultured in medium alone were harvested from individual Freshly isolated plasma cells were added to FACS-sorted reticular bone wells, pooled, and the numbers of viable cells were determined by trypan marrow stromal cells at a density of 5 ϫ 103 plasma cells/well with or blue exclusion. Total plasma cells present in cocultures were enumerated without 104 stromal cells/well, as indicated, in a total volume of 150 ␮l. by immunofluorescence microscopy for the presence of intracellular Ig L Supernatants (120 ␮l) were collected twice weekly and fresh media was chain. Plasma cells and stromal cells were harvested from wells at days 1, added after supernatant collection. In experiments using IL-6, recombinant 3, 7, and 14. Four wells per time point were pooled and placed onto slides mouse IL-6 (rmIL-6) (BD PharMingen) was added as indicated at the ini- by cytocentrifugation. Cells were fixed and permeabilized, then stained tiation of culture and during each medium change at the specified with FITC-R26-46 (anti-Ig␭) and FITC-R5-240 (anti-Ig␬) purchased from concentrations. BD PharMingen. At least 1000 cells were counted by phase microscopy for each slide and the number of FITCϩ cells were determined for each enu- Blocking Abs to adhesion molecules merated field using a Leitz Diaplan fluorescent scope (W. Nuhsbaum, Inc., McHenry, IL). The frequency of plasma cells per time point was deter- The following monoclonal blocking Abs were used: M/K2 (anti-CD106/ ϩ VCAM-1, protein G purified; American Type Culture Collection, Rock- mined by dividing the total number of FITC cells by the total number of ville, MD), PS2 (anti-CD49d/VLA-4, protein G-purified; American Type cells counted by phase microscopy. Culture Collection), and KM81 (anti-CD44/phagocytic gp-1, provided by Dr. M. Seigelman, University of Texas Health Science Center, Dallas, TX). ELISPOT to enumerate IgG1-secreting plasma cells The ability of these Abs to block adhesion was tested in our laboratories. IgG1-specific ELISPOTs were modified from previous descriptions (41). Briefly, the ability of anti-VCAM-1 and anti-VLA-4 to displace B cell Plates were blocked with 1ϫ PBS/1% BSA (Sigma-Aldrich) for a mini- precursors from the adherent layer in LTBMC-B was tested as described by mumof4hatroom temperature. Plasma cells were harvested from stromal Miyake et al. (38). Adding our Abs to the cultures at 5 ␮g/ml was sufficient cell cocultures and incubated on the plates overnight at 37¡C. Samples to release 81.6 and 73.68% of the lymphocytes, respectively, from the were serially diluted 1/3 and run in duplicate or triplicate. Bound Ab was detected and plates were developed and read as described previously (39). 3 Abbreviations used in this paper: VLA, very late Ag; rmIL-6, recombinant mouse Abs used for the ELISPOT assay were the same as those described above IL-6; LTBMC-B, long-term bone marrow culture for B-lineage; PI, propidium iodide; for Ig-specific ELISAs. Peroxidase-conjugated streptavidin was purchased BrdU, 5Ј-bromo-2-deoxyuridine. from BD PharMingen. The Journal of Immunology 4215

Real-time PCR for quantitation of IL-6 mRNA Statistical analysis RNA was isolated individually from plasma cells and stromal cells in co- Statistical signiﬁcance between the experimental groups was determined cultures. To release the plasma cells from the stromal cell adherent layer, using a paired Student t test; p Յ 0.05 was considered signiﬁcant. cultures were treated with 0.02% EDTA. Stromal cells were lysed directly in the wells and the lysate was collected for RNA isolation. RNA was isolated from both cell types using the Qiagen RNeasy kit (Valencia, CA) Results according to the manufacturer’s protocol. cDNA was synthesized using the Bone marrow stromal cells support plasma cell survival Applied Biosystems PCR Core kit (Branchburg, NJ). Real-time PCR analysis of IL-6 message was performed on a GeneAmp Reticular stromal cells form the cellular scaffold of the bone mar- 5700 sequence detection system (Applied Biosytems). Platinum QPCR Su- row and are important for supporting hemopoiesis. Because stro- permix-UDG (Life Technologies, Grand Island, NY), 12.5 ␮M Rox ref- mal cells are especially important for the maintenance of early erence dye (Life Technologies), SYBRGreen DNA binding dye diluted B-lineage cells, the ability of bone marrow stromal cells to en- ␮ 1/2000, forward and reverse primers (2.5 M in 10 mM Tris), UV-treated hance plasma cell survival was evaluated. Continued Ab secretion double-distilled H2O, and appropriate standard cDNA or sample cDNA Ϫ/Ϫ were added to each sample. RNA isolated from the stromal cell line BMS2 was used as a means of assessing plasma cell survival. The E/P (gift of P. Kincade, Oklahoma Medical Research Foundation, Oklahoma plasma cells secreted predominantly IgG1 Ab (Table I) (32). City, OK) stimulated with 25 ␮g/ml cyclohexamide and 10 ␮g/ml LPS was Therefore, IgG1 was used to assess plasma cell viability in all used as a positive control. Negative controls included samples without subsequent experiments with E/PϪ/Ϫ plasma cells. When plasma reverse transcriptase or RNA or cDNA. Controls, standards, and samples Ϫ/Ϫ were run in triplicate (20 ␮l volume) in a 96-well optical reaction plate and cells from E/P lymph nodes were cultured in the absence of a capped with optical caps (Applied Biosystems). Real-time oligonucleotides stromal cell layer, the level of secreted Ab was sharply reduced at

were designed using Primer Express software (Applied Biosystems) and time points as early as 24 h, compared with plasma cells cultured Downloaded from purchased from Integrated DNA Technologies (Coralville, IA) and Life with stromal cells, and by day 11 little or no Ab could be detected Ј Technologies: IL-6 5 primer, TGTATGAACAACGATGATGCACTT; (Fig. 1A). Specifically, no Ab was detected at day 7 in two of five IL-6 3Ј primer, ACTCTGGCTTTGTCTTTCTTGTTATCT; GAPDH 5Јprimer, GTGAGGCCGGTGCTGAGTAT; GAPDH 3Ј primer, TCATG experiments, and by day 11 average Ab levels were 5 ng/ml for all AGCCCTTCCACAATG. The real-time PCR thermal cycler profile was five experiments. In contrast, much higher levels of Ab were de- run as follows: one cycle at 50¡C for 2 min, one cycle at 95¡C for 10 min, tected in the plasma cell/stromal cell coculture supernatants for the 40 cycles of denaturing at 95¡C for 15 sec, and annealing and elongation ϳ first 3 days, followed by a significant drop by day 7. Ab levels http://www.jimmunol.org/ at 60¡C for 1 min, followed by a dissociation protocol run to test the melting temperature of the product. The data was analyzed using GeneAmp then fell gradually over the next 3 wk. Levels of IgG2a or IgG2b 5700 SDS software (Applied Biosystems). In each experiment, plasmids Ab followed a similar course indicating that the loss of IgG1 Ab containing the GAPDH or IL-6 PCR product were used to generate a stan- was not due to selective loss of this isotype or due to isotype dard curve to calculate the number of copies of transcript per microliter. switch (data not shown). These results indicate that bone marrow ϫ 5 ϫ The standards were 10-fold serial dilutions ranging from 1 10 to 1 stromal cells can maintain plasma cell viability and Ab secretion. 102 copies/␮l. IL-6 mRNA expression levels for each sample were calculated by determining the ratio of the number of IL-6 copies per microliter Recent studies by both Slifka et al. (12) and Manz et al. (11) to the number of GAPDH copies per microliter. Data are presented as the suggest that the majority of long-lived plasma cells reside in the fold-difference of IL-6 expression for each sample as compared with the bone marrow. Plasma cells isolated from either bone marrow of Ϫ Ϫ control samples with only stromal cells. wild-type mice or the lymph nodes of E/P / mice were morpho- by guest on September 29, 2021 Southern blot analysis of IL-6 RT-PCR products logically similar (Fig. 1), but the bone marrow plasma cells secreted primarily IgM. Despite the difference in the secreted Ab In some experiments, the relative amount of IL-6 mRNA message was isotype, a similar pattern of survival was observed between lymph assessed by semiquantitative RT-PCR followed by Southern blot of the RT-PCR products. RNA was isolated and cDNA synthesized as described node and bone marrow plasma cells. When bone marrow plasma above. Subsequent PCR analysis was performed using the manufacturer’s cells were cultured without stromal cells, Ab secretion declined suggested protocol and a GeneAmp 2400 thermacycler (PerkinElmer, quickly, falling to undetectable levels by day 11 (Fig. 1B). Con- Wellesley, MA). Primer oligonucleotides were synthesized by Integrated versely, when bone marrow plasma cells were cocultured with DNA Technologies: IL-6 5Ј primer, ATGAAGTTCCTCTCTGCAA GAGAC and IL-6 3Ј primer, CACTAGGTT TGCCGAGTAGATCTC as stromal cells, higher levels of Ab were observed, and Ab was still reported by Van Snick et al. (42); ␤-actin 5Ј primer, GTGGGGCGC detectable 3 wk after culture initiation. The pattern of decline of Ϫ Ϫ CCCAGGCACCA and ␤-actin 3Ј primer CTCCTTAATGTCACGCAC bone marrow plasma cells was also similar to E/P / plasma cells, GATTTC. DNA was transferred to a Duralose-UV membrane (Stratagene, with most of the plasma cell loss occurring early during culture. La Jolla, CA) and cross-linked with a UV stratalinker. The BMS2 IL-6 These data suggest that the plasma cells from E/PϪ/Ϫ or wild-type PCR product was used as the probe template and primed with random hexamers to produce a [32P]dCTP-labeled probe (Stratagene). The blot was mice are comparable and that stromal cells are able to prolong washed for 10 min at room temperature in 2ϫ SSC and 0.1% SDS, fol- plasma cell longevity of plasma cells from the two tissues lowed by a 60-min wash at 60¡C. Signals were visualized and quantified similarly. using a Typhoon 8600 PhosphorImager (Molecular Dynamics, Sunnyvale, To evaluate the frequency of plasma cells present in cocultures CA). ␤-Actin PCR products were stained for 30 min with Vistra Green diluted 1/105 in TE (Amersham Biosciences, Piscataway, NJ) and visual- with stromal cells, plasma cells were enumerated at various time ized using the Typhoon 8600 Fluorimager (Molecular Dynamics). Quan- points using trypan blue exclusion, immunofluorescence micros- titation was performed using ImageQuant software (Molecular Dynamics). copy for the presence of intracellular L chains, or by ELISPOT

Table I. Ab isotypes represented in plasma cell/stromal cell coculturesa

Isotype (ng/ml)

Experimental Group IgG1 IgG2a IgG2b IgG3 IgM

E/PϪ/Ϫ plasma cellsb 622.4 Ϯ 92.2 50.6 Ϯ 7.9 225.9 Ϯ 55.8 0.6 Ϯ 0.6 0.0 Bone marrow plasma cellsc 32.7 Ϯ 14.3 38.6 Ϯ 26.2 26.6 Ϯ 11.8 9.12 Ϯ 4.0 415.0 Ϯ 185.0

a Supernatants were collected at 24 h. b Data shown are the average Ϯ SEM of ﬁve independent experiments. c Data shown are the average Ϯ SEM of three independent experiments. 4216 BONE MARROW STROMAL CELLS MAINTAIN PLASMA CELL LONGEVITY Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021

FIGURE 1. Maintenance of Ab secretion from plasma cells when cocultured with stromal cells. Plasma cells isolated from the cervical lymph nodes of E/PϪ/Ϫ mice (A) or the bone marrow of BALB/c mice (B) were cocultured with FACS-sorted stromal cells. Supernatants were collected from the cultures twice weekly and then assayed for the presence of IgG1 (A) or IgM (B) Ab by ELISA. Data in A are shown as the average Ϯ SEM of six individual experiments. The results in B are from one representative experiment of two performed. Each experiment was performed from independent culture initiations for stromal cells and from independent plasma cell isolations. PC on SC, plasma cells cocultured with stromal cells; PC Only, plasma cells p Ͻ 0.05. Photomicrographs: plasma cells from E/PϪ/Ϫ lymph nodes or the bone marrow of wild-type mice were isolated ,ء .cultured in medium alone as described in Materials and Methods. Isolated plasma cells were placed onto slides by cytocentrifugation and stained with H&E. assay for IgG1. When plasma cells were cultured in the absence of process is stochastic or whether these stromal cells select for a stromal cells, their numbers decreased rapidly, falling to zero by subpopulation of plasma cells is not clear. day 7 as determined by trypan blue exclusion. Viable plasma cells were detected after day 7 only when plasma cells were cocultured with stromal cells. Using immunofluorescence microscopy, the Inhibition of plasma cell/stromal cell contact via VLA-4 frequency of intracellular IgLϩ cells in plasma cell/stromal cell decreases plasma cell viability cocultures decreased by 64% after 3 days, an additional 23% after To determine whether plasma cell survival is dependent on contact 7 days, followed by an 11% decrease after 2 wk (Table II). The with stromal cells, the requirement for stromal cell/plasma cell total frequency of plasma cells after 2 wk in culture with stromal binding through specific adhesion molecules was tested. The ad- cells was ϳ2% of the original input of 5000 plasma cells at day 0, hesion molecules VLA-4 and CD44 are expressed and active on yet the remaining cells continued to secrete relatively high levels the surface of E/PϪ/Ϫ plasma cells (32), and their ligands VCAM-1 of Ab. In separate experiments, the frequency of IgG1-specific and hyaluronate are found on the surface of stromal cells from plasma cells as determined by ELISPOT assay followed a similar LTBMC-B (38, 43). Addition of anti-CD44 or anti-VCAM-1 had trend: an 89% decrease observed at day 3, an 88% decrease at day little or no effect on Ab secretion (Fig. 2). In contrast, addition of 7, and a 99.7% decrease after 2 wk (Table II). Taken together, anti-VLA-4 resulted in a 40% decrease in Ab secretion after only these data suggested that the decreased amount of Ab in the su- 24 h, a 50% decrease after 3 days, and 70% by 7 days as compared pernatants corresponded to a decrease in the number of Ab-secret- with cultures that received no blocking Abs. These data indicate ing cells, rather than decreased Ab production from a large pop- that VLA-4 on the plasma cell surface is important for adhesion to ulation of surviving cells. Thus, a small population of plasma cells stromal cells and consequent plasma cell survival. Because the exhibits extended longevity in the presence of stromal cells, and addition of anti-VCAM-1 had no effect on Ab secretion, VLA-4 further, maintains their full capacity to secrete Ab. Whether this may bind stromal cells via another ligand, such as fibronectin. The Journal of Immunology 4217

Table II. Frequency of recovered plasma cells as a percent of inputa

Media Alone Coculture with Stromal Cells

% of plasma cells % of total plasma cells determined % of IgG1-speciﬁc plasma determined by trypan by cytoplasmic Ig L chain cells determined by Day blue exclusionb immunoﬂuorescencec ELISPOTd

0 100 (5000)e 100 (5000) 5000 (2500)f 1 14.3 Ϯ 9.6 (700) ND 20.8 (1050) 3 2.1 Ϯ 3.5 (100) 35.8 (1800) 11.2 (560) 7 0 13.1 (650) 12.0 (600) 11 ND ND 0.38 (19) 14 ND 2.4 (100) 0.30 (15)

a Total number of input E/PϪ/Ϫ plasma cells/well at day 0 ϭ 5000 cells. b Data is shown as the average Ϯ SEM of four independent experiments. c Data shown are the values from one representative experiment. d Data shown are the values from one representative experiment. e Valves in parentheses represent the calculated nmber of plasma cells/well remaining in culture at the indicated time point. f Based on data that ϳ50% of total input plasma cells are of the IgG1 isotype. Downloaded from Stromal cell-derived IL-6 is required to maintain plasma cell observed were due to IL-6-induced proliferation. Cell cycle anal- survival ysis was performed on plasma cells cultured in the presence or Several studies performed with normal and malignant human absence of rmIL-6 for 48 h using PI. No significant changes were plasma cells suggest that IL-6 is necessary for survival of human detected in the proportion of plasma cells in the S or G2/M phases plasma cells and myeloma cells (44, 45). We predicted that if IL-6 of cell cycle at any concentrations of rmIL-6 (Table III). The lack of IL-6-induced proliferation was further confirmed by adding is important for maintaining murine plasma cell survival in the http://www.jimmunol.org/ bone marrow, then plasma cell survival would be enhanced when BrdU to the cultures at day zero. No BrdU was incorporated during exogenous IL-6 is added to plasma cells cultured without stromal the 48-h culture in the presence or absence of exogenous IL-6 (data cells. To test this premise, Ab secretion was monitored after the not shown). Notably, the relative number of apoptotic events were Ն addition of varying concentrations of rmIL-6 (Fig. 3A). The markedly decreased when plasma cells received 500 pg/ml plasma cells did not constitutively make endogenous IL-6, which rmIL-6 (Fig. 3B). Therefore, the data suggest that the elevated was determined by the addition of neutralizing Ab to IL-6 and by levels of Ab observed when rmIL-6 was added was due to in- real-time PCR (data not shown and Fig. 4). The addition of exog- creased plasma cell survival and possibly also increased Ab secre- enous rmIL-6 maintained Ab secretion in a concentration-depen- tion per cell, but not due to increased cell number. dent manner. Ab was not detected after 4 days in wells with me- The data in Fig. 3A suggest that the addition of exogenous by guest on September 29, 2021 dium only, while Ab was still detectable at day 7 in wells that rmIL-6 can enhance plasma cell survival. To determine whether received rmIL-6. Because IL-6 can stimulate the proliferation of stromal cells provide IL-6 necessary for plasma cell longevity, many cell types, we determined whether the elevated levels of Ab plasma cells were cocultured with primary bone marrow reticular stromal cells from wild-type or IL-6 Ϫ/Ϫ mice (Fig. 3B). As early as 24 h, the amount of detectable Ab was reduced by ϳ33% in the supernatants from plasma cells cocultured with IL-6Ϫ/Ϫ stromal cells as compared with plasma cells cultured with wild-type stromal cells. Secretion continued to decrease rapidly in cocultures with IL-6Ϫ/Ϫstromal cells, declining to levels 85% of that detected in cocultures with wild-type stromal cells at the end of the first week. Ab levels from cocultures with IL-6Ϫ/Ϫ stromal cells continued to remain at this low level through day 14. This data suggests that IL-6 is an essential factor for plasma cell survival and that IL-6 is provided to plasma cells by stromal cells. Although the decline in Ab levels observed in the IL-6Ϫ/Ϫ stromal cell cultures was significant as compared with wild-type stromal cell cultures, it was not as dramatic as that observed when plasma cells were cultured in the absence of stromal cells. In the latter cultures, Ab was virtually undetectable by the end of the first week. This suggests FIGURE 2. IgG1 secretion from plasma cells cultured with anti- that the presence of other stromal cell-specific factors may be more adhesion molecule Abs. Plasma cells were cocultured with FACS-sorted important than IL-6 in the first 72 h of culture. stromal cells in the presence or absence of blocking Abs to VCAM-1, CD44, or VLA-4 (5 ␮g/ml). Supernatants were collected and pooled from wells twice weekly and IgG1 Ab secretion was assayed by ELISA. Data are IL-6 mRNA in stromal cells is induced in the presence of expressed as the percentage of IgG1 secretion as compared with plasma plasma cells cells cocultured with stromal cells in the absence of blocking Abs. Data The data in Fig. 3 suggested that plasma cells respond to IL-6 shown are the average of two independent experiments with anti-CD44 and anti-VCAM-1, and three independent experiments with anti-VLA-4 Ϯ provided by the stromal cells. However, previous studies in our lab p Ͻ 0.05 as compared with plasma cells cocultured with stromal showed that the primary cultured stromal cells from LTBMC-B do ,ء .SEM cells in the absence of blocking Abs. Data for cocultures receiving anti- not constitutively secrete IL-6 (determined by ELISA, data not VLA-4 was not statistically significant from plasma cells cultured in me- shown) or express IL-6 mRNA (Fig. 4 and Ref. 35). To determine dium alone. PC Only, plasma cells cultured in medium alone. whether IL-6 mRNA was induced in stromal cells upon contact 4218 BONE MARROW STROMAL CELLS MAINTAIN PLASMA CELL LONGEVITY Downloaded from

FIGURE 4. Induction of IL-6 mRNA in stromal cells after coculture http://www.jimmunol.org/ with plasma cells. A, Stromal cells were FACS-sorted from LTBMC-B and cocultured with plasma cells. At 48 h of coculture, plasma cells were removed from the stromal cells and RNA was isolated separately from each cell population. IL-6 mRNA levels were measured using real-time PCR. A standard curve was used to determine the number of copies per microliter for each sample. IL-6 mRNA levels for each sample were calculated by normalizing the number of copies of IL-6 to the number of copies of GAPDH. The fold-difference in IL-6 mRNA was determined by dividing

FIGURE 3. IL-6 is necessary for plasma cell viability. A, The effects of the normalized IL-6 value for each sample by the normalized IL-6 value for by guest on September 29, 2021 exogenous IL-6 on plasma cell survival in the absence of stromal cells. stromal cells cultured alone. B, In a separate experiment, IL-6 RT-PCR Plasma cells were cultured in the absence of an adherent cell layer and products were analyzed by Southern blot using the BMS2 IL-6 PCR prod- received increasing concentrations of exogenous rmIL-6 at the initiation of uct as the probe template. Signals were visualized and quantiﬁed using a culture. Supernatants were collected daily, pooled, and analyzed for the Typhoon PhosphorImager. ␤-Actin PCR products were visualized using a presence of IgG1 Ab. Exogenous rmIL-6 was reintroduced into the cultures Typhoon 8600 Fluorimager after a 30-min stain with Vistra Green. Lane 1, when supernatants were collected. Data is shown as one representative stromal cells alone; lane 2, plasma cells from coculture; lane 3, stromal experiment of two or three independent experiments for each concentration cells from coculture; lane 4, no template; lane 5, no reverse transcription; of exogenous rmIL-6. PC Only, plasma cells cultured in medium alone. B, lane 6, blank; lane 7, BMS2 cell line (positive control). IL-6 mRNA mes- Maintenance of Ab secretion after coculture with stromal cells from IL- sage levels from stromal cells from coculture are 16 times greater than 6Ϫ/Ϫ bone marrow. Plasma cells were cocultured alone or with a FACS- plasma cells from coculture; ␤-actin mRNA message levels from stromal sorted population of stromal cells obtained from wild-type or IL-6Ϫ/Ϫ cells or plasma cells from coculture are three times greater than SC alone. mice. Supernatants were collected and pooled from wells twice weekly and IgG1 Ab secretion assayed by ELISA. Data shown is one of three independent experiments. Two of three experiments exhibited the sharp decline in Ab secretion after 24 h as noted in the text; in one experiment Ab signal to the stromal cells that mediates changes in gene expression secretion was decreased by ϳ10% at 24 h, but diminished quickly there- and function of the stromal cell. after. By days 7 and 14, Ab secretion declined to similar levels in all three experiments performed. Discussion It is well-established that the majority of long-lived plasma cells reside in the bone marrow, an environment that promotes their with plasma cells, IL-6 mRNA levels in stromal cell cultures with terminal differentiation and fosters their survival (11Ð18). In the and without plasma cells were compared using quantitative real- current study, we addressed whether primary cultured bone mar- time PCR. As shown in Fig. 4A, IL-6 mRNA was increased 22- row reticular stromal cells are sufﬁcient to promote plasma cell fold in stromal cells 48 h after plasma cells were added to the longevity and maintain Ab secretion, and the mechanisms by culture as compared with stromal cells alone. No change in IL-6 which this occurs. mRNA was observed in plasma cells removed from stromal cell/ Plasma cells isolated from the lymph node of E/P Ϫ/Ϫ mice plasma cell cocultures, indicating that IL-6 was induced only in the appear to be a relevant model for studying plasma cell survival in stromal cells. Similarly, when IL-6 mRNA levels were assessed by vitro and appear phenotypically similar to wild-type plasma cells. conventional RT-PCR and Southern blot, IL-6 mRNA was only The plasma cells are in the G0/G1 phase of cell cycle suggesting detected when stromal cells were cocultured with plasma cells that they are terminally differentiated. In addition, their cell surface (Fig. 4B). No IL-6 was detected in RNA isolated from stromal phenotype (CD138/Syndecan-1ϩ, CD44ϩ, CD19Ϫ, MHC IIϪ, sur- cells cultured alone. Thus, contact with plasma cells may deliver a face IgϪ, CD11b/Mac-1Ϫ) and morphology resemble that of The Journal of Immunology 4219

Table III. The effects of exogenous IL-6 on plasma cell cell cycle support plasma cell viability, although this clearly cannot account statusa for all the observed plasma cell death. Additionally, the plasma cells present at the end of the fourth week of coculture may simply IL-6 Added to % of Cellsb be the most recently differentiated plasma cells. Finally, and per- Plasma Cells haps most physiologically, the death of the majority of plasma (pg/ml IL-6) Sub G0 G0/G1 S and G2/M cells may reﬂect the normal process of homeostasis in the immune 0 35.4 61.1 3.6 system, in which the majority of recently generated effector cells 200 21.7 75.9 2.4 500 12.9 84.2 3.2 die, leaving only a small subpopulation to contribute to immuno- 1000 9.3 84.0 6.7 logic memory. These various possibilities are currently under in-

Ϫ Ϫ vestigation in our laboratories. a E/P / plasma cells were cultured in the absence of an adherent layer and received increasing concentrations of exogenous rmIL-6 at the initiation of culture. The present studies also strongly suggest that IL-6 is crucial for Cells were pooled and harvested 48-h postculture initiation, stained with PI, and plasma cell viability and Ab secretion. The addition of exogenous analyzed by FACS to assess DNA content. b The percent of cells in each gate for one of two independent experiments is rIL-6 to plasma cells cultured without stromal cells results in in- indicated. creased longevity that is concentration-dependent. However, this action of IL-6 alone is weaker than stromal cells, because Ab levels still decrease more quickly than those from plasma cell/stromal plasma cells isolated from bone marrow of wild-type mice (Refs. cell cocultures (compare Figs. 1A and 3A). Although the concen- 19, 32, and 46; Fig. 1). tration used in these experiments is greater than reported circulat- Downloaded from We found that primary bone marrow stromal cells could maining levels of IL-6 (51), it is difﬁcult to mimic the local concentra- tain the viability of a subset of plasma cells, whether obtained from tion of IL-6 produced by stromal cells when they are in close Ϫ/Ϫ lymph nodes of E/P mice or bone marrow of wild-type mice. contact with plasma cells in the bone marrow microenvironment. In contrast, plasma cells cultured in media without stromal cells or In addition, IL-6Ϫ/Ϫ stromal cells are substantially impaired in exogenous growth factors die rapidly, suggesting that plasma cell their ability to support plasma cell viability and Ab secretion. longevity is not an intrinsic quality (47). Further, this ﬁnding Taken together, these results strongly imply that IL-6 is critical for shows that the ability of stromal cells to support plasma cell via- plasma cell viability. However, because Ab secretion was not com- http://www.jimmunol.org/ bility is independent of both the tissue source and the isotype of the Ϫ Ϫ Ϫ Ϫ pletely abrogated in IL-6 / cultures containing IL-6 / stromal plasma cells. However, the patterns obtained with the two cocul- cells, other factor(s) may be involved in plasma cell survival or can ture systems are not completely identical. The amount of Ab se- partially compensate for the absence of IL-6. Consistent with this, creted by bone marrow plasma cells cultured with stromal cells Ϫ Ϫ IL-6 / mice exhibit reduced but not absent humoral responses to stabilizes after 1 wk, whereas the amount of Ab secreted by Ϫ/Ϫ viral infection (33, 52). Whether plasma cell development or lon- E/P plasma cells continues to decline gradually over the same Ϫ Ϫ gevity per se is impaired in IL-6 / mice is unknown. time period. Notably, studies performed with human plasma cells Previous studies in our laboratory show that primary cultured showed there are distinct differences in the cell surface phenotypes

stromal cells do not constitutively secrete IL-6 in the absence of a by guest on September 29, 2021 and in the requirements for growth factors of Ab-secreting cells stimulus (35). Rather, our findings suggest that IL-6 is produced as isolated from tonsils, blood, or bone marrow (48, 49). More re- a result of a direct physical interaction between stromal cells and cently, a study by Medina et al. (50) confirmed that the cell surface plasma cells. We found, by real-time PCR, that stromal cells are phenotypes of Ab-secreting cells isolated from tonsils, peripheral blood, and bone marrow are heterogeneous in nature suggesting induced to transcribe IL-6 mRNA within 48 h of coculture with that the stages of maturity increase from tonsil to blood to bone plasma cells but that little or no IL-6 mRNA is detected in cultures marrow. Therefore, it is possible that bone marrow plasma cells of isolated stromal cells or purified plasma cells. Whether IL-6 are intrinsically longer lived than E/PϪ/Ϫ plasma cells derived induction in stromal cells is a result of signals transduced as a from lymph nodes, perhaps as a result of terminal differentiation consequence of plasma cell/stromal cell contact or from signals and conditioning in the bone marrow microenvironment, or by received from plasma cell-secreted proteins remains to be deter- having a higher fraction of cells committed to long-term survival. mined. However, the decrease in Ab secretion observed in the Ϫ/Ϫ The patterns of plasma cell viability in stromal cell cocultures IL-6 stromal cell/plasma cell cocultures during the first week is are striking and consistent. Ab levels drop rapidly during the first similar to that observed in cocultures that received anti-VLA-4, week and then much more gradually or not at all during the re- implying that IL-6 induction in stromal cells is adhesion-depen- maining 2Ð3 wk of culture, suggesting that the majority of plasma dent. It will be interesting to test whether blocking VLA-4 inter- cells die rapidly even in the presence of stromal cells. Several action with the stromal cell can influence IL-6 expression. possible factors may underlie the survival of only a subset of B cell lymphopoiesis in Whitlock-type LTBMC-B requires lym- plasma cells. There is a paucity of information on the components phocyte/stromal cell interactions mediated both by VLA-4/ necessary for the survival of terminally differentiated plasma cells. VCAM-1 interactions and CD44/hyaluronate interactions (38, 43, A recent study by Sze et al. (47) showed that long-lived plasma 53, 54). We found that the addition of blocking Abs directed cells appear not to develop from one specific pathway, but rather against VLA-4, but not VCAM-1 or CD44, result in decreased Ab are derived from a heterogeneous population of cells. The data secretion from plasma cells. These findings distinguish plasma suggest that the lifespan of plasma cells is not controlled intrinsi- cell/stromal cell interactions from pro-B cell/stromal cell interac- cally, but rather by signals received from the external environment. tions and suggest that plasma cell/stromal cell interactions via To that note, plasma cell survival may require other cells or factors VLA-4 are important for plasma cell survival. The importance of present in the bone marrow microenvironment in addition to the VLA-4 interaction with its ligand VCAM-1 has been demonstrated stromal cells used in this study. Alternatively, the relevant stromal for myeloma cell survival, hemopoiesis, and hemopoietic cell re- cells may represent only a fraction of those used in our cocultures tention within the bone marrow (23, 55Ð59). In contrast, inhibiting and may be limiting in these experiments. Plasma cell survival contact with anti-VCAM-1, a ligand for VLA-4, has no effect on may also be limited in part by the life of the stromal cells them- Ab secretion in our coculture system, suggesting that VLA-4 may selves or the ability of stromal cells to maintain their ability to interact with fibronectin or another unidentified ligand on stromal 4220 BONE MARROW STROMAL CELLS MAINTAIN PLASMA CELL LONGEVITY cells. The interaction of VLA-4 with fibronectin has been demon- 21. Kincade, P. W., P. L. Witte, and K. S. Landreth. 1987. Stromal cell and factor- strated with human CD38ϩ cells, which include plasma cells and dependent B lymphopoiesis in culture. Curr. Top. Microbiol. Immunol. 135:1. 22. Jacobsen, K., and D. G. Osmond. 1990. Microenvironmental organization and their precursors (60). stromal cell associations of B lymphocyte precursor cells in mouse bone marrow. Understanding the mechanism by which protective immunity is Eur. J. Immunol. 20:2395. 23. Funk, P. E., P. W. Kincade, and P. L. Witte. 1994. Native associations of early maintained is important for developing more efficient therapies to hematopoietic stem cells and stromal cells isolated in bone marrow cell aggre- induce long-term protective immunity. The goal in this study was gates. Blood 83:361. to determine the basic mechanism for promoting plasma cell lon- 24. Funk, P. E., and P. L. Witte. 1992. Enrichment of primary lymphocyte-supporting stromal cells and characterization of associated B lymphocyte progenitors. Eur. gevity. We found that bone marrow stromal cells are sufficient to J. Immunol. 22:1305. maintain the survival of a small population of plasma cells for at 25. LeBien, T. W. 1998. B-cell lymphopoiesis in mouse and man. Curr. Opin. Im- least 4 wk. Elements necessary to maintain plasma cell survival munol. 10:188. 26. Tew, J. G., R. M. DiLosa, G. F. Burton, M. H. Kosco, L. I. Kupp, A. Masuda, and include the adhesion through VLA-4 and the induction of IL-6 in A. K. Szakal. 1992. Germinal centers and antibody production in bone marrow. the stromal cells. Future studies using this experimental system Immunol. Rev. 126:99. 27. Merville, P., J. Dechanet, A. Desmouliere, I. Durand, O. de Bouteiller, should allow further dissection of critical components and events ϩ P. Garrone, J. Banchereau, and Y. J. Liu. 1996. Bcl-2 tonsillar plasma cells are necessary for plasma cell survival as well as the further elucidation rescued from apoptosis by bone marrow fibroblasts. J. Exp. Med. 183:227. of the role of bone marrow stromal cells in maintaining plasma cell 28. Kawano, M. M., K. Mihara, N. Huang, T. Tsujimoto, and A. Kuramoto. 1995. Differentiation of early plasma cells on bone marrow stromal cells requires in- longevity. terleukin-6 for escaping from apoptosis. Blood 85:487. 29. Roldan, E., and J. A. Brieva. 1991. Terminal differentiation of human bone mar- Acknowledgments row cells capable of spontaneous and high-rate immunoglobulin secretion: role of

bone marrow stromal cells and interleukin 6. Eur. J. Immunol. 21:2671. Downloaded from We thank Kirstin Parkin and Kara Johnson for critical reading of the manu- 30. Frenette, P. S., T. N. Mayadas, H. Rayburn, R. O. Hynes, and D. D. Wagner. script and helpful discussions, Dr. Phong Le for designing IL-6 real-time 1996. Susceptibility to infection and altered hematopoiesis in mice deficient in PCR primers, and Dr. Joseph Brewer for help with the Southern blots. both P- and E-selectins. Cell 84:563. Zheng Yu, Jeannette Pifer, David You, and Thomas Keech provided ex- 31. Bullard, D. C., E. J. Kunkel, H. Kubo, M. J. Hicks, I. Lorenzo, N. A. Doyle, cellent technical assistance. We also thank Patricia Simms, Dr. Tom Ellis, C. M. Doerschuk, K. Ley, and A. L. Beaudet. 1996. Infectious susceptibility and severe deficiency of leukocyte rolling and recruitment in E-selectin and P-selectin and the Loyola FACS Core Facility for assistance with the cell sorting. double mutant mice. J. Exp. Med. 183:2329. 32. Underhill, G. H., H. A. Minges-Wols, J. L. Fornek, P. L. Witte, and G. S. Kansas. References 2002. IgG plasma cells display a unique spectrum of leukocyte adhesion and http://www.jimmunol.org/ homing molecules. Blood 99:2905. 1. Slifka, M. K., and R. Ahmed. 1996. Long-term humoral immunity against vi- ruses: revisiting the issue of plasma cell longevity. Trends Microbiol. 4:394. 33. Kopf, M., H. Baumann, G. Freer, M. Freudenberg, M. Lamers, T. Kishimoto, 2. Fahey, J. L., and S. Sell. 1965. The immunoglobulins of mice: the metabolic R. Zinkernagel, H. Bluethmann, and G. Kohler. 1994. Impaired immune and (catabolic) properties of five immunoglobulin classes. J. Exp. Med. 122:41. acute-phase responses in interleukin-6-deficient mice. Nature 368:339. 3. Talbot, P. J., and M. J. Buchmeier. 1987. Catabolism of homologous murine 34. Whitlock, C. A., and O. N. Witte. 1982. Long-term culture of B lymphocytes and monoclonal hybridoma IgG antibodies in mice. Immunology 60:485. their precursors from murine bone marrow. Proc. Natl. Acad. Sci. USA 79:3608. 4. Vieira, P., and K. Rajewsky. 1988. The half-lives of serum immunoglobulins in 35. Witte, P. L., L. M. Frantsve, M. Hergott, and S. M. Rahbe. 1993. Cytokine adult mice. Eur. J. Immunol. 18:313. production and heterogeneity of primary stromal cells that support B lympho- 5. Ochsenbein, A. F., D. D. Pinschewer, S. Sierro, E. Horvath, H. Hengartner, and poiesis. Eur. J. Immunol. 23:1809. R. M. Zinkernagel. 2000. Protective long-term antibody memory by antigen- 36. Stephan, R. P., C. R. Reilly, and P. L. Witte. 1998. Impaired ability of bone marrow stromal cells to support B-lymphopoiesis with age. Blood 91:75. driven and T help-dependent differentiation of long-lived memory B cells to by guest on September 29, 2021 short-lived plasma cells independent of secondary lymphoid organs. Proc. Natl. 37. Sanderson, R. D., P. Lalor, and M. Bernfield. 1989. B lymphocytes express and Acad. Sci. USA 97:13263. lose syndecan at specific stages of differentiation. Cell Regulation 1:27. 6. Zinkernagel, R. M., M. F. Bachmann, T. M. Kundig, S. Oehen, H. Pirchet, and 38. Miyake, K., K. Medina, K. Ishihara, M. Kimoto, R. Auerbach, and P. W. Kincade. H. Hengartner. 1996. On immunological memory. Annu. Rev. Immunol. 14:333. 1991. A VCAM-like adhesion molecule on murine bone marrow stromal cells me- 7. Szakal, A. K., M. H. Kosco, and J. G. Tew. 1989. Microanatomy of lymphoid diates binding of lymphocyte precursors in culture. J. Cell Biol. 114:557. tissue during humoral immune responses: structure function relationships. Annu. 39. DeGrendele, H. C., P. Estess, L. J. Picker, and M. H. Siegelman. 1996. CD44 and Rev. Immunol. 7:91. its ligand hyaluronate mediate rolling under physiologic flow: a novel lympho- 8. MacLennan, I. C., Y. J. Liu, and G. D. Johnson. 1992. Maturation and dispersal cyte-endothelial cell primary adhesion pathway. J. Exp. Med. 183:1119. of B-cell clones during T cell-dependent antibody responses. Immunol. Rev. 126: 40. Sanders, V. M., R. A. Baker, D. S. Ramer-Quinn, D. J. Kasprowicz, B. A. Fuchs, 143. and N. E. Street. 1997. Differential expression of the ␤2-adrenergic receptor by 9. Gray, D., K. Siepmann, D. van Essen, J. Poudrier, M. Wykes, S. Jainandunsing, Th1 and Th2 clones: implications for cytokine production and B cell help. J. Im- S. Bergthorsdottir, and P. Dullforce. 1996. B-T lymphocyte interactions in the munol. 158:4200. generation and survival of memory cells. Immunol. Rev. 150:45. 41. Kamperschroer, C., and D. G. Quinn. 1999. Quantification of epitope-specific 10. Tew, J. G., M. H. Kosco, G. F. Burton, and A. K. Szakal. 1990. Follicular MHC class-II-restricted T cells following lymphocytic choriomeningitis virus dendritic cells as accessory cells. Immunol. Rev. 117:185. infection. Cell. Immunol. 193:134. 11. Manz, R. A., A. Thiel, and A. Radbruch. 1997. Lifetime of plasma cells in the 42. Van Snick, J., S. Cayphas, J. P. Szikora, J. C. Renauld, E. Van Roost, T. Boon, bone marrow. Nature 388:133. and R. J. Simpson. 1988. cDNA cloning of murine interleukin-HP1: homology 12. Slifka, M. K., R. Antia, J. K. Whitmire, and R. Ahmed. 1998. Humoral immunity with human interleukin 6. Eur. J. Immunol. 18:193. due to long-lived plasma cells. Immunity 8:363. 43. Miyake, K., C. B. Underhill, J. Lesley, and P. W. Kincade. 1990. Hyaluronate can 13. Benner, R., F. Meima, G. M. van der Meulen, and W. B. van Muiswinkel. 1974. function as a cell adhesion molecule and CD44 participates in hyaluronate rec- Antibody formation in mouse bone marrow. I. Evidence for the development of ognition. J. Exp. Med. 172:69. plaque-forming cells in situ. Immunology 26:247. 44. Kawano, M., T. Hirano, T. Matsuda, T. Taga, Y. Horii, K. Iwato, H. Asaoku, 14. Benner, R., W. Hijmans, and J. J. Haaijman. 1981. The bone marrow: the major B. Tang, O. Tanabe, and H. Tanaka. 1988. Autocrine generation and requirement source of serum immunoglobulins, but still a neglected site of antibody forma- of BSF-2/IL-6 for human multiple myelomas. Nature 332:83. tion. Clin. Exp. Immunol. 46:1. 45. Jego, G., N. Robillard, D. Puthier, M. Amiot, F. Accard, D. Pineau, 15. Dilosa, R. M., K. Maeda, A. Masuda, A. K. Szakal, and J. G. Tew. 1991. Ger- J. L. Harousseau, R. Bataille, and C. Pellat-Deceunynck. 1999. Reactive plas- minal center B cells and antibody production in the bone marrow. J. Immunol. macytoses are expansions of plasmablasts retaining the capacity to differentiate 146:4071. into plasma cells. Blood 94:701. 16. Ropke, C., H. P. Hougen, and N. B. Everett. 1975. Long-lived T and B lympho- 46. Calame, K. L. 2001. Plasma cells: finding new light at the end of B cell devel- cytes in the bone marrow and thoracic duct lymph of the mouse. Cell. Immunol. opment. Nat. Immunol. 2:1103. 15:82. 47. Sze, D. M., K. M. Toellner, C. Garcia de Vinuesa, D. R. Taylor, and 17. McMillan, R., R. L. Longmire, R. Yelenosky, J. E. Lang, V. Heath, and I. C. MacLennan. 2000. Intrinsic constraint on plasmablast growth and extrinsic C. G. Craddock. 1972. Immunoglobulin synthesis by human lymphoid tissues: limits of plasma cell survival. J. Exp. Med. 192:813. normal bone marrow as a major site of IgG production. J. Immunol. 109:1386. 48. Brieva, J. A., E. Roldan, M. L. De la Sen, and C. Rodriguez. 1991. Human in 18. Slifka, M. K., M. Matloubian, and R. Ahmed. 1995. Bone marrow is a major site vivo-induced spontaneous IgG-secreting cells from tonsil, blood and bone mar- of long-term antibody production after acute viral infection. J. Virol. 69:1895. row exhibit different phenotype and functional level of maturation. Immunology 19. Manz, R. A., M. Lohning, G. Cassese, A. Thiel, and A. Radbruch. 1998. Survival 72:580. of long-lived plasma cells is independent of antigen. Int. Immunol. 10:1703. 49. Brieva, J. A., E. Roldan, C. Rodriguez, and G. Navas. 1994. Human tonsil, blood 20. Witte, P. L., M. Robinson, A. Henley, M. G. Low, D. L. Stiers, S. Perkins, and bone marrow in vivo-induced B cells capable of spontaneous and high-rate R. A. Fleischman, and P. W. Kincade. 1987. Relationships between B-lineage immunoglobulin secretion in vitro: differences in the requirements for factors and lymphocytes and stromal cells in long-term bone marrow cultures. Eur. J. Im- for adherent and bone marrow stromal cells, as well as distinctive adhesion mol- munol. 17:1473. ecule expression. Eur. J. Immunol. 24:362. The Journal of Immunology 4221

50. Medina, F., C. Segundo, A. Campos-Caro, I. Gonzalez-Garcia, and J. A. Brieva. 56. Mazo, I. B., J. C. Gutierrez-Ramos, P. S. Frenette, R. O. Hynes, D. D. Wagner, 2002. The heterogeneity shown by human plasma cells from tonsil, blood, and and U. H. von Andrian. 1998. Hematopoietic progenitor cell rolling in bone bone marrow reveals graded stages of increasing maturity, but local profiles of marrow microvessels: parallel contributions by endothelial selectins and vascular adhesion molecule expression. Blood 99:2154. cell adhesion molecule 1. [Published erratum appears in 1998 J. Exp. Med. 188: 51. Faunce, D. E., M. S. Gregory, and E. J. Kovacs. 1998. Acute ethanol exposure 1001]. J. Exp. Med. 188:465. prior to thermal injury results in decreased T-cell responses mediated in part by 57. Frenette, P. S., S. Subbarao, I. B. Mazo, U. H. von Andrian, and D. D. Wagner. increased production of IL-6. Shock 10:135. 1998. Endothelial selectins and vascular cell adhesion molecule-1 promote he- 52. Kopf, M., A. Ramsay, F. Brombacher, H. Baumann, G. Freer, C. Galanos, matopoietic progenitor homing to bone marrow. Proc. Natl. Acad. Sci. USA 95: J. C. Gutierrez-Ramos, and G. Kohler. 1995. Pleiotropic defects of IL-6-deficient 14423. mice including early hematopoiesis, T and B cell function, and acute phase re- 58. Schweitzer, K. M., A. M. Drager, P. van der Valk, S. F. Thijsen, A. Zevenbergen, sponses. Ann. NY Acad. Sci. 762:308. A. P. Theijsmeijer, C. E. van der Schoot, and M. M. Langenhuijsen. 1996. Con- 53. Miyake, K., I. L. Weissman, J. S. Greenberger, and P. W. Kincade. 1991. Evi- stitutive expression of E-selectin and vascular cell adhesion molecule-1 on en- dence for a role of the integrin VLA-4 in lympho-hemopoiesis. J. Exp. Med. dothelial cells of hematopoietic tissues. Am. J. Pathol. 148:165. 173:599. 59. Jacobsen, K., J. Kravitz, P. W. Kincade, and D. G. Osmond. 1996. Adhesion 54. Miyake, K., K. L. Medina, S. Hayashi, S. Ono, T. Hamaoka, and P. W. Kincade. receptors on bone marrow stromal cells: in vivo expression of vascular cell ad- 1990. Monoclonal antibodies to Pgp-1/CD44 block lympho-hemopoiesis in long- hesion molecule-1 by reticular cells and sinusoidal endothelium in normal and term bone marrow cultures. J. Exp. Med. 171:477. gamma-irradiated mice. Blood 87:73. 55. Robledo, M. M., F. Sanz-Rodriguez, A. Hidalgo, and J. Teixido. 1998. Differ- 60. Roldan, E., A. Garcia-Pardo, and J. A. Brieva. 1992. VLA-4-fibronectin interac- ential use of very late antigen-4 and -5 integrins by hematopoietic precursors and tion is required for the terminal differentiation of human bone marrow cells ca- myeloma cells to adhere to transforming growth factor-␤1-treated bone marrow pable of spontaneous and high rate immunoglobulin secretion. J. Exp. Med. stroma. J. Biol. Chem. 273:12056. 175:1739. Downloaded from http://www.jimmunol.org/ by guest on September 29, 2021