|Home | About | Journals | Submit | Contact Us | Français|
The bone marrow of old adult mice (~2 yrs. old) have reduced B lymphopoiesis; however, whether the B1 pathway in adult bone marrow is also compromised in senescence is not known. Herein, we show that phenotypic (IgM-Lin-CD93+[AA4.1+] CD19+B220low/-) B1 progenitors are retained in old bone marrow even as B2 B cell precursors are reduced. Moreover, B1 progenitors from both young adult and old mice generated new B cells in vitro enriched for CD43 expression, likely due to their activation, and exhibited increased λ light chain usage and diminished levels of κ light chain expression. B1 progenitors were shown to have lower surrogate light chain (λ5) protein levels than did B2 pro-B cells in young mice and these levels decreased in both B1 and B2 precursor pools in old age. These results indicate that the B1 B cell pathway persists during old age in contrast to the B2 pathway. Moreover, B1 B cell progenitors generated new B cells in the adult bone marrow that have distinct surface phenotype and light chain usage. This is associated with decreased surrogate light chain expression, a characteristic held in common by B1 progenitors as well as B2 precursors in old mice.
Characteristically, old mice of several inbred strains exhibit decreased numbers of pre-B cells within their bone marrow (Riley et al., 1991; Stephan et al., 1996; Sherwood et al., 1998; Miller et al., 2003; Van der Put et al., 2003; Labrie et al., 2005; Alter-Wolf et al. 2009). Furthermore, earlier B cell precursors may also be diminished in old mice, including both pro-B cells and common lymphoid progenitor cells (CLPs) (Miller et al., 2003; Van der Put et al., 2003). This suggests that B lymphopoiesis is both defective and limited in old mice; this may affect the “read-out” of the B cell specificity repertoire and the maintenance of new B cells in senescence.
B cell precursors in adult bone marrow are generally regarded as generating new B cells belonging to the B2, or conventional follicular, B cell lineage (Allman et al., 1992; Hardy et al., 2001). These B2 B cells participate in adaptive immune responses to foreign antigens and provide memory. This is in contrast to B cell precursors found early in ontogeny where, during the fetal/neonatal period, B cell precursors predominantly yield B cells belonging to the B1 B cell subsets (Herzenberg, 2000). B1 B cells provide a front-line defense against common bacterial pathogens, but also are associated with poly- and self-reactivity. In young adult mice, B2 B cells are predominant while B1 B cells comprise only ~5% of total B cells in peripheral lymphoid tissues, e.g., spleen (Herzenberg, 2000; Weksler, 2000). In contrast, B1 B cells are the major B cell subset in the peritoneal cavity (Herzenberg, 2000). In old mice, the incidence of B1 B cells is markedly increased by ~3-fold in the periphery and expanded clonal B1 populations are often observed in peritoneal cavity and spleen (Stall, et al., 1988; Weksler, 2000).
Several minor pathways of B lymphopoiesis have recently been characterized in adult murine bone marrow (Montecino-Rodriguez et al., 2006; Yang et al., 2007). It has been demonstrated that adult murine bone marrow does maintain a limited capacity for generating B1 B cells, mainly of the B1b subset (Hardy et al., 1991). Recently, Montecino-Rodriguez et al., (2006) have characterized a bone marrow fraction containing precursors of the B1 lineage. These B1 lineage precursors are characterized by the absence of myeloid lineage antigenic markers, as well as having negligible CD45R/B220 antigen which ordinarily is expressed by early B lineage precursor cells. However, these B1 precursors do have CD19, the surface antigen regarded as indicative of B lineage commitment (Montecino-Rodriguez et al., 2006). In addition, common myeloid progenitors have also recently been demonstrated to have limited B lineage developmental potential (Yang et al., 2007). Therefore, adult bone marrow has multiple pathways that support B lymphopoiesis, albeit with different efficiencies and with differences in types of B cells generated from these distinct precursor pools. While B2 pathways of development are known to be compromised in old age (Riley et al., 1991; Stephan et al., 1996; Sherwood et al., 1998), whether the B1 pathway retained in adult bone marrow is similarly affected is not known.
In this report, we demonstrate that a population of CD19+B220low/- B1 progenitor cells within young adult murine bone marrow shows a significant, but limited, potential for B cell development. However, in senescent mice, the major B2 pathway leading to B cell development is diminished, but the pool of B1 progenitors is maintained. Furthermore, B cells generated from B1 progenitors in vitro, from both young and old mice, have altered surface phenotypes and also differ in light chain isotype usage from that seen in immature B cells derived from conventional pre-B cells. These observations suggest that the reduced competency for B lymphopoiesis seen in old mice affects the B2 lymphopoietic pathway, but not the B1 pathway. Moreover, our findings link the altered development of new B cells in old bone marrow as well as the unusual phenotype of immature B cells from the B1 progenitor pool to a common reduction in the surrogate light chain and possibly decreased reliance on the pre-B cell receptor (preBCR) checkpoint.
Young (2-4 months) and old (21-26 months) BALB/c mice were purchased from the National Institutes of Aging colony at Harlan Sprague Dawley, Indianapolis, IN. Mice with obvious abdominal tumors and/or splenomegaly in the thoracic or abdominal cavities were eliminated from the studies. All studies adhered to principles of laboratory animal care guidelines and were IACUC approved.
Femur and tibia pairs were flushed to harvest cells from the bone marrow as previously described (Riley et al., 1991). Red blood cells were removed by treatment with ACK (0.15 M NH4CL, 1mM KHCO3, 0.1 mM EDTA) for 5 minutes at room temperature followed by centrifugation to remove red cell debris. Bone marrow cells were counted and used for cell sorting, flow cytometry, or cell culture. For cell culture, cells were resuspended at 1×106/ml in RPMI-1640 (Gibco Life Technologies, Grand Island, NY), supplemented with 10% FCS (Sigma Aldrich, St. Louis, MO) plus 1% penicillin-streptomycin, 1% L-glutamine and 2-mercaptoethanol at 2×10-5M. Purified recombinant mouse IL-7 (rmIL-7, BioSource International, Camarillo, CA) was added at 5ng/ml and recombinant mouse stem cell factor (SCF) (BioSource International, Camarillo, CA) was added at 50ng/ml for 4-6 days after which cells were harvested and used for further analysis.
Mouse bone marrow cells freshly harvested or cultured were stained with the following antibodies that were labeled with an appropriate fluorochrome: IgM (II/41), CD43 (S7), B220 (RA3-6B2), CD19 (1D3), κ (187.1), λ(1-3) (R26-46) (BD Bioscience, San Diego, CA), CD93 (AA4.1), Gr-1 (RB6-8C5), Ter-119, CD3e (145-2C11), CD11b (M1/70), CD49b (DX5), c-kit (2B8), and CD8a (53-6.7) (e-Bioscience, San Diego, CA). For the cytoplasmic staining of μ chain, goat anti-mouse μ (Jackson ImmunoResearch) and, for λ5, LM34 (BD Biosciences) antibodies were employed. Cells were initially stained for surface markers, permeablized using BD Cytofix/Cytoperm (BD Bioscience), washed with PermWash (BD Bioscience), and followed by cytoplasmic stain addition. Cells were analyzed within 30 minutes of staining. Analysis was performed on an LSR II fluorescence flow cytometer (BD Bioscience). Gates to discriminate B220low/- cells were set based on isotype controls as well as background staining seen on CD19- bone marrow cells, analogous to that used by Montecino-Rodriguez et al. (2006). Between 5 × 105 (n=2) and 1 × 106 (n=6) bone marrow cells from individual old and young mice were acquired in the experiments.
Bone marrow cells isolated from tibia and femur pairs of young and old mice were sorted for early pre-B and sIgM-B220low/-CD19+ precursors. Cells were stained for surface IgM (II/41), CD43 (S7), B220 (RA3-6B2), CD19 (1D3) (BD Bioscience), CD93 (AA4.1) and c-kit (2B8) (e-Bioscience, San Diego, CA). Cells were sorted, as previously reported (Alter-Wolf et al., 2009) with a FACS Aria cell sorter (BD ImmunoCytometry, San Jose, CA) with purity ranging between 94-99%.
Previous reports have shown that pre-B cells, as well as pro-B cells, are often reduced in the bone marrow of old mice of several inbred strains, including the BALB/c strain (Riley et al., 1991; Stephan et al., 1996; Sherwood et al., 1998; Miller et al., 2003; Van der Put et al., 2003; Labrie et al., 2005). A panel of surface antigens, including the isoform of CD43 recognized by the S7 mAb; the membrane tyrosine kinase c-kit; and the C1q receptor protein AA4.1 (CD93), together with IgM and/or μ Ig heavy chain and the B lineage antigens CD19 and B220 (CD45R) allowed discrimination of different subsets of B cell progenitors, precursors, and B cells in mouse bone marrow (Fig. 1) (Hardy et al., 1991; Allman et al., 1992; Rolink et al., 1994).
As shown in Figure 1, old BALB/c mice (22-26 mo. old) have reduced numbers of bone marrow late stage pre-B cells (surface IgM-CD43-B220+) as well as early pre-B cells (IgM-cμ+AA4.1+CD43+B220+CD19+), consistent with previous results (Alter-Wolf et al., 2008). Pro-B cells (IgM-cμ-AA4.1+CD43+B220+CD19+) cells also tended to be low in number in old BALB/c bone marrow, albeit this varied among individual mice tested, consistent with previously published results (Van der Put et al., 2003). These B220+CD19+ B cell precursor populations are generally presumed as generating new B cells of the B2 subset (Hardy, et al., 2001; Herzenberg, 2000). Consequently, this lineage of B cell development in the bone marrow is significantly diminished in old BALB/c mice.
Recently, it has been reported that an ordinarily minor, but distinctive, B cell precursor pool is present in adult bone marrow that lacks surface Ig, expresses CD19, has negligible B220 surface antigen, and preferentially generates B1 lineage B cells (Montecino-Rodriguez et al., 2006). As shown in Figure 2A, B cells negative for surface IgM, but bearing CD19 with little or no B220, are observed in both young and old BALB/c bone marrow. These putative B1 progenitors occur at similar incidences and cell numbers in both young and old mice (Fig. 2B). The parity in B1 progenitors in young vs. old mice is apparent even when conventional B2 B cell lineage pro-B cells are substantially reduced (Fig. 2C). These CD19+B220low/- cells are also uniformly positive for AA4.1, but lack myeloid and T lymphocyte lineage markers, e.g., Gr-1, CD11b, Ter119, CD3e, CD8a, and CD49b (DX5) (data not shown). Although expressing the B cell commitment antigen CD19, the CD19+B220low/- cells do not express detectable cytoplasmic μ heavy chain, suggesting that they are at an early developmental stage (data not shown). Lack of these markers as well as cμ, but expression of AA4.1, on the CD19+B220low/- cells is analogous to the phenotype of the B1 B cell progenitor population found in adult murine bone marrow as reported by Montecino-Rodriguez, et al. (2006). Our next experiments assessed the efficiency of B cell generation as well as the phenotypes of B cells derived from the CD19+B220low/- B1 progenitors in young and old mice.
When cultured in IL-7/SCF supplemented media, conventional early pre-B cells (IgM-c-kit-AA4.1+CD43+B220+CD19+ cells; ~70% cμ+) proliferated extensively and generated significant numbers of new B cells over a 4 day period (Fig. 3). IgM-CD19+B220low/- cells from young adult bone marrow also proliferated substantially in culture. In contrast to pre-B cells, CD19+B220low/- B1 progenitors generated new B cells in vitro with considerably reduced efficiency (Fig. 3A-C). This may reflect the relative stages of development of the B1 progenitor cells as compared to isolated B2 pre-B cells. As indicated above, the former did not express cμ heavy chain and may represent an earlier (pro-B?) developmental stage. Pre-B cells, defined and isolated as described above, were ~70% cμ+ pre B cells. However, it is notable that in vitro cultures of B1 progenitor cells, after 4 days, possessed ~ 25% cμ+ pre-B cells which was comparable to that observed in cultures derived from B2 pre-B cells in culture (Fig. 3D). This indicates that μ heavy chain rearrangement and protein expression occur, as does pre-B cell expansion, during culture of the IgM-CD19+B220low/- B1 progenitor cells. Therefore, it is unlikely that the lower output of B cells derived in culture of B1 progenitor cells results solely from reduced numbers of pre-B cells in culture, but instead also reflects reduced efficiency of the pre-B to B cell transition.
The B cells derived in vitro from both conventional B2 pre-B cells and from B1 progenitor cells were uniformly positive (>95%) for AA4.1, indicative of their immaturity (data not shown). These B cells did not express CD5, CD23, or CD11b antigens during the 4 day culture period (data not shown). Immature B cells derived from conventional B2 pre-B cells also expressed little CD43 antigen as detected with the S7 monoclonal antibody (Fig. 4A,B). In contrast, immature B cells derived from the B1 progenitor cells expressed greater surface densities of CD43 and increased incidences of B cells with relatively high CD43 expression (CD43hi) (Fig. 4A,B). The immature B cells derived from B1 progenitors were also larger in size, as determined by forward angle light scatter, than were B cells derived from conventional B2 pre-B cells (Fig. 4C). Therefore, the B cell progeny of B2 pre-B cells differed in surface phenotype (e.g., CD43 expression) and cell size when compared to B cells derived from B1 progenitors. The immature B cells derived from these precursor subsets from old BALB/c mice were similar in phenotype to those cultured from young mice (Fig. 4D,E).
Typically, immature B cells in adult murine bone marrow predominantly utilize the κ light chain. This was also seen for B cells newly derived from conventional B2 pre-B cells in vitro (Fig. 5A,B). In contrast, B1 progenitors from young mice generated new B cells in vitro where the incidences of B cells with significantly lower expression of κ light chain as well as λ light chain usage were increased (Fig. 5A). Analogous precursor pools isolated from old mice produced patterns of light chain usage equivalent to those seen in young B cell precursors (Fig. 5B). We have recently shown that diminished usage of κ light chain was seen among immature IgM+AA4.1+ B cells from old mice in vivo, in particular from those old mice with extensive (>90%) loss of bone marrow pre-B cells (Alter-Wolf et al., 2009). Therefore, the phenotypes of immature B cells seen in old mice in vivo (Alter-Wolf et al., 2009) resembled that of B cells derived from the B1 progenitors in vitro more so than from the majority of conventional pre-B cells.
We have also observed that bone marrow B2 B cell precursors that are compromised in their expression of surrogate light chains (e.g., c-kit+ precursors from either wild-type or λ5 knock-out mice) preferentially generate new B cells in vitro that express CD43, have increased cell size, and have reduced κ expression similar to that seen in B cells derived in analogous cultures from B1 B cell progenitors (Alter-Wolf et al., 2009). This pool of preBCR compromised pre-B cells is little affected by old age, in contrast to the reductions seen in preBCR competent c-kit- pre-B cells, and is proportionally increased in old bone marrow (Alter-Wolf et al., 2009). Given the similarities in phenotype of immature B cells generated from preBCR deficient B2 precursors and B1 progenitors, we next assessed surrogate light chain content in bone marrow B1 vs. B2 B cell precursors.
B1 progenitors expressed lower levels of the surrogate light chain λ5 protein as revealed by cytoplasmic staining than did B2 pro-B cells (IgM-CD43+AA4.1+cμ-CD19+B220+) (Fig. 5C,D). These results suggest that the distinct properties of newly derived B cells from either λ5 knockout or wild-type c-kit+ B2 pre-B cells (Alter-Wolf et al., 2009), as well as B1 progenitors (this report), may have a common origin in deficient expression of the preBCR. B1 progenitors are also maintained at wild-type numbers in the bone marrow of λ5 gene knockout mice, indicating that surrogate light chain expression has no effect on development of IgM-CD19+B220low/- B1 progenitors (data not shown).
Notably, we have recently shown that the immature B cell population within the bone marrow of both young adult λ5 gene knockout mice and those old wild-type mice that are severely depleted of B2 pre-B cells (<20% of young adult numbers) and have significantly reduced surrogate light chain exhibit alterations in phenotype in vivo (Wilson et al., 2005; Alter-Wolf et al., 2009). This includes increased representation of CD43+ immature B cells as well as increased proportions of immature B cells with reduced κ light chain expression as well as κ+λ+ dual isotypes. Therefore, in vivo, immature B cells from old mice resemble the immature B cell populations derived from preBCR deficient c-kit+ B cell precursors and B1 progenitors, the same precursor populations preferentially retained in old bone marrow (this report and Alter-Wolf et al., 2009).
B lineage development is compromised in the bone marrow of mice during old age (Riley et al., 1991; Stephan et al., 1996; Sherwood et al., 1998; Miller et al., 2003; Van der Put et al., 2003; Labrie et al., 2005). This is most readily observed as a decline in the number of pre-B cells within the bone marrow, but more recent studies also indicate that early B cell precursors, including common lymphoid progenitors and pro-B cells, are also adversely affected by the aging process (Riley et al., 1991; Stephan et al., 1996; Sherwood et al., 1998; Miller et al., 2003; Van der Put et al., 2003; Labrie et al., 2005). The consequences of diminished B cell development in old mice are not well characterized, but may contribute to alterations in B cell longevity (Johnson et al., 2002; Kline et al., 1999; Labrie et al., 2005) and composition of the B cell receptor repertoire (Klinman et al., 1997; Wilson et al., 2005) during senescence. Whether similar alterations in B cell development are observed in humans is speculative; however, naïve B cells are reduced in number in the periphery in old age (Nunez et al., 1996; Huppert et al., 1999; Ogawa et al., 2000; Rossi et al., 2003)
Previously, it has been shown that old murine bone marrow cells deficient in surface IgM, when used as a source of pre-B cells, generated new B cells with substantial changes in the antibody repertoire to the phosphorylcholine (PC) epitope (Klinman et al., 1997). Changes in the antibody repertoire to PC are associated with susceptibility to pathogens, e.g., S. pneumoniae (Nicoletti et. al., 1993). More recently, we have shown that an increased proportion of immature B cells present in the bone marrow of old mice have a skewed Vh repertoire (over-representation of the anti-PC-antibody associated VhS107 heavy chains) and have unusual surface phenotypes with heightened expression of the CD43 antigen as well as increased, but variable, expression of CD5 and CD11b, all antigens associated with activation as well as B1 B cells (Wilson et al., 2003; 2005). These B cells also were larger in size and their phenotype was dependent upon expression of functional Bruton's tyrosine kinase (Btk), again consistent with a requisite for activation via the B cell receptor (Wilson et al., 2003; 2005). However, the origin of these partially activated, immature CD43+ B cells in the bone marrow of old mice was not addressed. Of interest, anti-PC B cells are also regarded as autoreactive (Kenny et. al., 2000) and it is likely that self-antigen drives the activation of a subset of immature B cells within the bone marrow with these surface characteristics. Indeed, this activated B cell population may contribute to increased autoreactivity characteristic of old age (Klinman, 1992). Consequently, the origin of this subset of activated immature B cells within the bone marrow of young and old mice is important, but as yet not known.
While the vast majority of B cells developing within the adult bone marrow is of the B2 type and will populate the follicular peripheral compartments, it is now appreciated that there are several minor pathways within the bone marrow that can also lead to B cell development (Montecino-Rodriguez et al., 2006; Yang et al., 2007). Common myeloid progenitors have been shown to have limited B cell developmental potential (Yang et al., 2007); furthermore, a subset of B cell precursors has been demonstrated that mainly yields B1 B cells (Montecino-Rodriguez et al., 2006).
The B cell progeny of B1 progenitor cells derived in vitro, in addition to increased CD43 expression, also were larger in size and we presume that they are partially activated. This contrasts with the phenotype of B cells derived from conventional B2 pre-B cells; therefore, we speculate that B1 progenitors preferentially develop into new B cells that more readily undergo activation to self or environmental antigens. Ordinarily, B cells derived from the B1 progenitor cell pool would only be a minor fraction of the immature B cell compartment. This would result from the low numbers of B1 progenitors in the bone marrow; moreover, their capacity to yield new B cells is considerably limited when compared to that of conventional B cell precursors as shown herein. However, in old mice where conventional B cell development is often highly reduced, by >80% or more, the maintenance of the B1 progenitor pool may result in their increased contribution to new B cell generation.
It is noteworthy that “antigen experienced” B cells are increased in the spleens of old mice; these include B1 B cells as well as marginal zone B cells, memory B cells, and B cells with characteristics of chronic activation (Johnson et. al., 2002). This raises the question as to whether the maintenance of the B1 pathway in old mice contributes to the increases seen in B1 cells in the periphery. Under normal circumstances, in young adult mice, it is unlikely that the bone marrow B1 pathway contributes substantially to the B1a (CD5+) compartment, but may contribute to the B1b (CD5-CD11b+) population which is much more effectively derived from adult bone marrow (Hardy and Hayakawa, 2001). However, as numbers of B2 B cell precursors become reduced in old mice, while B1 progenitors are maintained, the ordinarily minor contribution of this B cell pathway would likely be more pronounced. If applicable to the aging human, this may impact, both quantitatively and qualitatively, the types of B cells reconstituted during autologous transplants with bone marrow or stem cells derived from young adult vs. older donors.
The immature B cells produced from B1 progenitors have features typical of B1 B cells, e.g., high CD43 expression, greater size, and increased dual isotype (κ+λ+) BCR; in vivo, these same cells also express CD11b and often CD5 (Hardy and Hayakawa 1991; Hardy and Hayakawa 2001; Rezanka et al., 2005). However, it is of interest that precursors from the B2 lineage also can produce immature B cells with similar characteristics when preBCR expression and signaling is minimized as is likely for c-kit+ pre-B cells (Alter-Wolf et al., 2009). This suggests the possibility that an important molecular distinction in the development of B cell precursors directed to B1 vs. B2 cell fates may be the extent of preBCR assembly and function. That B1 progenitors have reduced surrogate light chain and undergo new B cell formation comparable to that seen previously from either preBCR compromised wild-type or surrogate light chain deficient λ5 knockout pre-B cells (Alter-Wolf et al., 2009) is consistent with this hypothesis. It has been suggested that fetal pre-B cells, rich in B1 precursors, undergo expansion optimally when preBCR signaling is more limited, in contrast to the pronounced dependence of adult pre-B cell expansion on strong preBCR signaling (Hardy et al., 2000; Wasserman et al., 1998). However, for B1 associated Vh genes, (e.g., Vh11), relative preBCR assembly correlates with capacity to promote proliferation (Yoshikawa, et al., 2008). We speculate that adult B1 precursors may express less surrogate light chain and this may impact expression of their antibody repertoires. More importantly, as shown herein, B1 progenitors, in contrast to B2 precursors, persist in old age.
While it has been clearly shown that B1 B cells derive from committed progenitors (Montecino-Rodriguez et al., 2006), it has also been demonstrated that B cells with some B1 characteristics can be generated via BCR stimulation (e.g., T cell independent antigens) of B2 cells or their precursors (Wortis and Berland, 2001). While the extent to which such B1-like B cells overlap in function with B1 progenitor-derived B1 B cells is not well characterized, our studies suggest that limited surrogate light chain expression among B2 precursors results in new B cells with several properties also shown by B cells derived from B1 progenitors. This suggests that the pool of B cells with some B1-like properties will be augmented in old bone marrow. We speculate that this results from the altered selection of the μ heavy chain repertoire at the pro-B to pre-B cell transition due to limited preBCR expression and signaling in old B2 precursors and in B1 progenitors. Alterations in the relative abundance of B1 and B2 precursor pools, together with changes in surrogate light chain expression and preBCR function, in old age may contribute to the changes seen in B cell output, as well as function, in senescence.
We gratefully acknowledge the assistance of Jim Phillips and the Flow Cytometry Core Facility at the Sylvester Comprehensive Cancer Center. We thank all members of the Riley and Blomberg laboratories for their support in the performance of these studies.
1Supported by NIH grants AG025256 and AI064591 to RLR and AG017618 to BBB
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.