Abstract
Heparan sulfate (HS) modulates many cellular processes including adhesion, motility, ligand-receptor interaction, and proliferation. We have previously reported that murine B cells strongly upregulate cell surface HS upon exposure to type I interferon, TLR-ligands, or B cell receptor stimulation. To investigate the role of HS on B cells in vivo, we utilized EXT1lox/lox CD19-Cre conditional KO mice, which are incapable of synthesizing HS in B cells. We found that suppressing HS expression on B cells has no overt effect in B cell development, localization, or motility. However, we did observe that EXT1 conditional KO mice have decreased poly-reactive IgM in naïve aged mice relative to littermate control mice. Despite this decrease in poly-reactive IgM, EXT1 conditional KO mice mounted a normal B cell response to both model antigens and influenza infection. We also observed decreased plasma cells in EXT1 conditional KO mice after influenza infection. Although EXT1 conditional KO mice have decreased plasma cells, these mice still had comparable numbers of influenza-specific antibody secreting cells to littermate control mice. The findings presented here suggest that HS expression on B cells does not play a major role in B cell development or overall B cell function but instead might be involved in fine-tuning B-cell responses.
Introduction
Heparan sulfate (HS) is a post-translational modify cation comprised of a repeating disaccharide (uronic acid and N-acetyl glucosamine) that undergoes heterogeneous deacetylation, sulfation, and epimerization [1]. This glycosaminoglycan is attached to a core-protein, creating a heparan sulfate proteoglycan (HSPG) [2]. HSPGs are expressed on the surface of almost all mammalian cells and they modulate many cellular processes including cell adhesion, motility, migration, localization, and ligand-receptor interaction [2–5]. A number of lymphocyte-specific chemokines, cytokines, and growth factors contain HS binding domains (IL2-8, IL10, IL12, IFNγ, TNFα, MIP1β, APRIL, etc.)[5–13], suggesting that HS might be important for the coordination of adaptive immune responses.
B cells are an integral component of the immune system. Through the action of surface and secreted antibodies, B cells are capable of providing both innate and adaptive protection against invading pathogens [14, 15]. This is accomplished largely by the B-1 (CD19hi, CD5+) and B2 (CD19intermediate, CD5−) B cell subsets. B-1 B cells secrete poly-reactive IgM in the absence of antigen stimulation 14, 21. B-2 B cells are the classic B lymphocytes known to respond to thymus-dependent antigens by undergoing class switch recombination, somatic hypermutation, and clonal expansion. The combined effect of these two populations is that B-1 cell-derived IgM provides general innate-like protection against viral pathogens, while B-2 B cells contribute to the adaptive humoral response and immunological memory. Critical to initiating and maintaining a B-2 B cell response is B cell localization, cell-cell interaction, and responsiveness to numerous soluble ligands (chemokines and cytokines) [16–20]. This is illustrated by the concentration of B cells within the secondary lymphoid organs and their chemokine-dependent migratory patterns that result in the surveying of follicular dendritic cells for antigen [16, 21].
Recently, a number of groups have begun investigating the role of HS in immunity. Using distinct mouse models, Garner et al. and Reijmers et al. investigated the role of HS expression on B cells [22] [9, 23]. While Garner et al. found that HS only modestly affects B cell development, Reijmers and colleagues observed a dramatic phenotype in mice unable to express functional HS on cells of the hematopoietic lineage. In these mice, Reijmers et al. observed impairments in B cell development and in the antibody response. A recent report from our laboratory presented evidence that the expression of HS on B cells is tightly regulated [10]. Whereas naïve B cells express very low levels of HS, HS is strongly upregulated on the B cell surface following numerous stimuli that are present during different types of infections (such as IFN-I, toll-like receptor ligands, and B cell receptor-stimulators). Jarousse et al. also presented in vitro evidence that HS expression on B cells can increase B cell responsiveness to the B cell-specific cytokine APRIL, and as a result, HS-expressing B cells survived longer in culture and produced more IgA [10].
To assess the role of HS expression on B cells in vivo, we utilized a B cell-specific deletion of exostosin 1 (EXT1), an enzyme necessary for HS synthesis [24]. In this report, we show that expression of HS is not necessary for B cell development, motility, localization, or homing. We also determined that both the primary and secondary antibody response to infection or antigenic challenge proceeds normally in EXT1 cKO mice. Interestingly however, we show that older EXT1 cKO mice have lower titers of circulating poly-reactive IgM, which is suggestive of impaired B-1 B cell function. We also provide evidence that HS expression on B cells affects either the maintenance or generation of plasma cells induced at steady state, independent of immunological challenge. The data presented here suggest that, although EXT1 cKO mice do not show gross B cell defect,HS expression on B cells might be important for the fine-tuning regulation of antibody secreting cells.
Materials and Methods
Mice
All mice were housed and bred in pathogen-free conditions at the American Association of Laboratory Animal Care-approved animal facility at the Life Sciences Addition, or at the Northwest Animal Facility (NAF), both located at the University of California, Berkeley, CA, USA. All animal experiments were approved by the Animal Care and Use Committee at the University of California, Berkeley, CA, USA. C57Bl/6 were obtained from Jackson Laboratory. Actin-CFP and ubiquitin-GFP (C57BL/6-Tg(UBC-GFP)30Scha/J) mice were a kind gift from Professor Ellen Robey. Conditional KO mice were generated by crossing EXT1flox/flox (Jackson labs) mice to CD19-Cre+ (Jackson labs) mice (both on the C57Bl/6 background) to produce EXT1 conditional KO (EXT1flox/flox CD19-Cre+), and littermate controls (EXT1flox/+ CD19-Cre+). Breeding occurred using single mating pairs. Mouse pups were weekend 21 days after birth, and caged according to sex. Mice were fed standard pelleted food and water. Experiments were conducted with 6- to 25-week-old mice as indicated, in accordance with institutional guidelines for animal care and use. When Necessary, mice were euthanized by inhalation of carbon dioxide.
Isolation and culture of B cells
Lymphocytes were isolated from peripheral lymph nodes (inguinal, brachial, cervical, mediastinal, or axillary), the spleen, the bone marrow, or the peritoneal cavity of WT C57Bl/6, littermate control, EXT1 conditional KO, Ubiquitin-GFP, or EXT1 conditional KO-actin-CFP mice. Briefly, a single cell suspension was isolated and treated with red blood cell lysis buffer. For figures 1A, 3, and 4, B cells were purified using the EasySep negative selection mouse B cell enrichment kit (Stem Cell Technologies). B cell purity was assessed by flow cytometry using anti CD19 or B220 antibodies (we routinely observed over 98% purity). Cell viability was verified by 7AAD staining. For culturing of B cells, cells were resuspended at a density of 2×106 cells/ml in RPMI media containing 100mM MEM non-essential amino acids, 55 microM 2-Mercaptoethanol, 1 mM sodium pyruvate and 10 mM Hepes. When indicated, purified B cells were treated with Interferon beta (IFNβ◻◻ at 45U/ml unless indicated otherwise (PBL Interferon source). For evaluation of BM B cell subsets, BM was isolated from mice by flushing the marrow from the femurs. Bone marrow was separated from red blood cells by ficoll prior to staining.
Flow cytometry
Lymphocytes or purified splenic B cells were washed in PBS with 1% BSA and incubated with 1mg/ml purified rat anti-mouse CD16/CD32 (Fc block, BD Pharmingen) at 4°C for 15 min, prior to surface staining. B cells were detected with anti-CD19 or anti-B220 antibodies. HS was detected with the F58-10E4 clone from Seikagaku Corporation. The isotype control was a mouse IgM kappa (TEPC 183, Sigma). Bound antibody was revealed by staining with a FITC- or PE- conjugated anti-mouse IgMa (Igh-6a) monoclonal antibody (BD Pharmingen). Bone marrow progenitor B cells were defined as B220+, CD43 high, IgM low, and IgD low. Bone marrow pre B cells were defined as B220+, CD43 low, IgM/IgD low. Bone marrow immature B cells were defined as B220+, CD43−, IgM/IgD high. Splenic immature B cells were defined as CD19+, IgM−, IgD−. Splenic Transitional type I B cells were defined as CD19+, IgM+, IgD−. Splenic transitional type II B cells were defined as CD19+, IgM+, IgD+. Follicular mantle B cells were defined as CD19+, IgM−, IgD+. B-1 B cells were defined as CD19+, CD5+. Distinction of congenic CD45.1 and EXT1 cKO lymphocytes was done by staining for CD45.1 and CD45.2. For the homing experiments, fluorescently labeled B and T cells were identified with CD19 and CD3 antibodies, respectively. Germinal center B cells were identified as CD19+, Fas+, GL7+, and IgD−. Plasma cells were identified as CD19+, syndecan 1+, and IgD−. For all staining procedures, dead cells were excluded from the analysis by 7AAD staining. Data were acquired on a LSRII flow cytometer (BD Biosciences) and were analyzed with FlowJo software (Treestar).
Generation of BM chimeras
Wild-type female C57BL/6 congenic mice between 6-8 weeks were lethally irradiated with 900 rads. 24 hours later, the mice were injected intravenously with a 1:1 mixture of congenic WT CD45.1+ and EXT1 conditional KO CD45.2+ (2.5X106 of each genotype) bone marrow. 10 weeks post bone marrow transfer, reconstitution efficiency was checked by tail-vein bleeding and flow cytometry for peripheral B and T cells.
Homing of B cells to the spleen and peripheral lymph nodes
WT ubiquitin-GFP or EXT1cKO-actin-CFP mice were injected intravenously with either PBS or 0.2 mg Poly I:C. Twenty-four hours later, a single cell suspension of splenocytes were isolated, and red blood cells were lysed. Thirty million splenocytes of each genotype were then injected intravenously into a naïve WT mouse. Twenty minutes or 20 hours later, the indicated secondary lymphoid organs were harvested: at 20 minutes post transfer, spleens were isolated. At 20 hours post transfer, spleen, inguinal-, cervical-, brachial- and axillary-lymph nodes were isolated. A single cell suspension of each organ was isolated, and the ratio of B to T cells from the WT and EXT1 conditional KO donors were assessed by flow cytometry.
Localization of B cells in secondary lymphoid organs
Littermate control or EXT1cKO mice were injected intravenously with 0.2mg Poly I:C. 24 hours later, a single cell suspension of splenocytes were isolated, and red blood cells were lysed. B cells were purified using the Stem Cell Technologies kit, and labeled with either SNARF (Invitrogen) or CFSE (Invitrogen). 1X107 B cells from each genotype was intravenously injected into a naïve WT mouse. Twenty hours later, the mouse was sacrificed, the inguinal lymph nodes harvested, and cryosectioned. 10μM thick cryosections were transferred onto glass slides and imaged by microscopy.
Two-Photon imaging of B cells in the inguinal lymph nodes
B cells were purified from naïve littermate control or EXT1conditional KO mice, and labeled with either CFSE or SNARF before 1X107 of each was transferred into a naïve WT mouse. Three hours post transfer, recipient mice were intravenously injected with PBS or 0.2 mg poly I:C. 24 hours post PBS or poly I:C injection, mice were sacrificed, inguinal lymph nodes isolated, and explanted individual lymph nodes were maintained at 37 °C and perfused with oxygenated Dulbecco’s modified Eagle’s medium, and imaged using two-photon microscopy as previously described [25, 26]. Imaging volumes of 172 × 143 × 70–120 μm were scanned every 30 s for 30 min, with 2 μm z steps at tissue depths of 30–200 μm below the lymph node capsule using a custom-built two-photon microscope with a Spectra-Physics MaiTai Laser (Newport, Santa Clara, CA, USA) tuned to 920 nm. SNARF, CFSE, and autofluorescence were separated with 495 and 560 nm dichroic mirrors and a 645/75 nm band-pass filter to restrict the detection to the optimal wavelength for each fluorescent protein. For basic motility comparisons between samples, cell tracking was performed using Imaris software (Bitplane, Saint Paul, MN, USA), and speed (path length divided by time) was calculated using MATLAB software (Mathworks, Natick, MA, USA).
Virus and infections
Purified mouse-adapted influenza A/PR/8/34 (H1N1) was purchased from Advanced Biotechnologies Inc. (Cat. # 10-210-500), aliquoted, and stored at −80 degrees Celsius. For infection of mice, aliquots were diluted in sterile PBS. Mice were anesthetized with isoflurane and infected intranasally with 104 TCID50 in a 40ul volume. After intranasal inoculation, mice were allowed to recover on a heating pad. For re-infection/recall experiments, primary infection was performed as described above. Mice were then allowed to recover for a period of 10 weeks before they were re-infected with 5X104 TCID50 intranasaly. Mouse survival and weight were monitored daily. As our animal protocol mandates, mice were sacrificed if their weight dropped below 65% their pre-infection weight. For the influenza infection, no mice fell below this percentage of initial weight, though there were some deaths (see Figure 6).
Titering influenza virus from the lung
At the indicated times, mice were sacrificed and both lungs were harvested in 2 ml PBS. Lungs were then homogenized using a Polytron PT2100 homogenizer (Kinematica). Homogenate was then spun down, and the supernatant was used in an MDCK-agglutination assay as previously described (Cottey, Rowe, and Bender, 2001. (Current Protocols). Briefly, supernatant was serially diluted and added to MDCK cells for 24 hours, and then removed. 5 days later, hemagglutination of chicken red blood cells was used to determine the TCID50.
ELISA for influenza specific antibody isotypes
At the indicated times, mice were sacrificed and a terminal bleed was performed. Serum was separated using serum separator tubes (BD cat.# 02-675-188). ELISA plates were coated overnight at four degrees Celsius with heat-killed influenza (see above), diluted in PBS to 1μg/ml. Serially-diluted serum was added to the plates overnight, before detection of influenza specific antibodies with isotype specific alkaline-phosphotase conjugated antibodies. Substrate conversion (pNpP from Sigma; cat. # N2770) was monitored using a plate reader measuring at λ405nm.
Statistical Analysis
All statistical results are expressed as mean ± SEM. Statistical analysis was performed using a non-parametric Mann-Whitney test for comparison between two groups (Graphpad software). Differences were considered significant at p<0.05.
Conflict of Interest
The authors declare no financial or commercial conflict of interest.
Discussion
Numerous studies have shown that HS is important in determining how cells sense and respond to their extracellular environment [3],[37]. We have previously reported that naïve B cells do not express HS, but upon viral or bacterial infection, HS is rapidly upregulated [10]. However, little is known about how HS expression on B cells affects B cell biology. Various groups have presented data suggesting that HS biosynthesis and modification is regulated during B cell development and differentiation, and as a result, HS likely plays a role in B cell biology [38, 39]. To investigate the role of HS expression on B cells in vivo, we have characterized a transgenic mouse with a B cell-specific deletion of EXT1, an enzyme necessary for HS synthesis.
Development and Maturation of B cells in the absence of EXT1
Despite the fact that HS is not expressed from the pro-B cell stage onwards in EXT1 cKO mice, we did not observe any overt defects in B cell development in either the BM or in the spleen. However, a subtle role for HS expression in B cell development was revealed in mixed WT/EXT1 cKO BM chimeras. This developmental defect, observed only in a competitive context, is likely due to the expression of HS slightly above background in BM B cells (data not shown). The alterations in B cell subset populations observed in both the BM and spleen suggest partial blocks at the early stages in B cell development and maturation in the respective organ. In both the BM and the spleen, these partial blocks result in increased precursor B cell populations. These partial blocks suggest that the expression of HS during these stages, although not essential, aids in processes important for B cell maturation. It is tempting to speculate IL-7, a cytokine containing a heparan sulfate binding domain whose signaling is enhanced in the presence of exogenous HS, may account for the defect observed in BM B cell development [6] [40] [41].
B cell behavior in the absence of EXT1
The homing of B cells to secondary lymphoid organs is key for B cell surveillance of antigens and the development of the humoral immune response [31, 42, 43]. A number of groups have recently demonstrated the importance of HS and sulfated glycan expression on endothelial cells for lymphocyte homing. Both the Fukuda and the Esko groups have shown that the sulfated glycans expressed on endothelial cells serve as ligands for L-selectin expressed on lymphocytes, and interaction between L-selectin and sulfated glycans is important for lymphocyte sticking/rolling along high endothelial venules (HEV). Additionally, presentation of chemokines (such as CCL19 and CCL21) important for integrin-mediated adhesion and transcytosis were diminished in mice deficient for HS on HEVs [29, 44]. Given the importance of HS on HEVs for lymphocyte homing, it was surprising that we saw no defect in B cell homing, either immediate or accumulative, in EXT1-deficient B cells. It remains possible that by assessing only the B:T cell ratio within the secondary lymphoid organs, we may be missing HS-dependent differences in B cell sticking/rolling along the HEVs. Electrostatic repulsion between HS expressed on B cells and sulfated glycans expressed on HEVs may affect B cell rolling. Alternatively, the presence of HS on B cells may alter binding of HEV-presented chemokines such as SDF-1 and CCL19, and as a result, affect integrin activation [29] [31]. Further studies assessing the importance of HS expression on B cells as it relates to B cell rolling and integrin activation may reveal a role for HS in B cell homing.
To the best of our knowledge, we are the first to assess the role of HS expression on B cell localization within secondary lymphoid organs. When looking within peripheral lymph nodes (Fig 3F) and the spleen (data not shown), EXT1-deficient B cells appeared to be evenly distributed throughout the B cell follicle. This was surprising since T cell zone organizing chemokines CCL19 and CCL21 (also known as ELC and SLC, respectively), are known to bind HS [17, 33]. From our data, it does not appear that the presence of HS on B cells enhances responsiveness to these chemokines, as we did not see any difference in the density of B cells at the B cell/T cell border between littermate control and EXT1-deficient B cells. Perhaps only after upregulation of the CCL19/CCL21 receptor on B cells, as occurs upon B cell activation, does HS affect localization within the LN.
Because HS has been shown to affect both chemokine-mediated cell migration [29],[3, 44] and cell motility [45], we were surprised to find no defect in motility in our EXT1-deficient B cells. It is possible that HS expression on B cells can affect motility, however, heparan sulfate or other sulfated glycosaminoglycans present within the extracellular matrix of the lymph node [29] may compensate for the lack of HS on EXT1-deficient B cells. Alternatively, HS expression may affect B cell migration in a context other than that assessed here, or it may only affect B cells within certain parts of lymph nodes such as the border between the B cell follicle and the T cell zone border, where chemokine concentrations are different from those in the B cell follicle [17, 32].
Another facet of HS biology that must be considered is HS composition. That is, both the spacing and the degree to which HS chains are sulfated. A growing body of evidence suggests that sulfation along the HS chain is regulated in a cell- and tissue-specific manner [46],[47]. Importantly, it is now known that distinct sulfation patterns are associated with specific biological activities [37]. Characterization of sulfation patterns along HS chains on B cells may prove key in determining how HS affects B cell biology. Indeed, heparan sulfate modifying enzymes have been shown to be differentially regulated during B cell development and activation [22, 38].
Aged naive EXT1 cKO mice have lower titers of IgM in the serum
Although we observed no developmental defects in EXT1 cKO mice, we did observe that aged EXT1 cKO mice (over 16 weeks of age) exhibited lower circulating IgM. We were surprised to find that despite lower titers of poly-reactive IgM, EXT1 cKO mice had similar numbers of peritoneal (Figure 2E) and splenic B-1 B cells (data not shown). Although we cannot definitively attribute the lower IgM titers to impaired B-1 B cell antibody production, we hypothesize this is the case based on the poly-reactive nature of the decreased IgM. Furthermore, the B cells that respond to antigenic challenge, B2 B cells, are unimpaired in their ability to respond to both influenza and model antigens (discussed further below). If decreased B-1 B cell antibody production is indeed the cause of the observed lower IgM observed, this would indicate that HS expression on B cells affects function, but not maintenance of B-1 B cells.
How does HS affect antibody production in aged mice? Similar to what was seen in naïve B cells, HS expression on B cells (both CD5-postive and negative) from 6- and 20-week-old mice was slightly above background (Figure 5a and data not shown). Interestingly, Sindhava and colleagues have recently reported that mice lacking APRIL have decreased levels of B-1a and B-1b cells [48]. APRIL has been shown to regulate B cell survival and class switching in a HS-dependent manner [49]. Taken together, this data is in line with our observation that IgM levels are affected in EXT1 cKO mice. We hypothesize that HS and APRIL interact in a way that can regulate B-1 cell numbers and IgM levels. Perhaps the low level of HS expression observed here is key for positioning or responsiveness to factors important for antibody production.
The B cell response to influenza in the absence of heparan sulfate expression
As was seen with murine gammaherpesvirus 68 (MHV-68) and mouse cytomegalovirus, infection of mice with influenza induces high levels of HS on B cells early post infection [10] (Figure 5B). The strong induction of HS on B cells is likely due to the robust amount of IFN-I that influenza induces in the upper respiratory tract [10],[50]. Also similar to what was observed upon MHV-68 infection, HS expression on most B cells was transient following influenza infection. Notably however, HS expression on plasma cells was still moderately high 14 days pi. The level of HS expression on plasma cells was particularly interesting considering the decreased number of plasma cells present in EXT1 cKO mice post infection (discussed further below).
Both the early thymus-independent, as well as the Thymus-dependent B cell responses contribute to limiting influenza disease severity [36, 51]. Given that HS expression is induced on B cells early after influenza infection, we hypothesized that the early B cell response might be altered, and as a result, this might affect disease severity [15]. We were surprised to find that upon infection with influenza, both littermate control and EXT1 cKO mice tolerated the infection similarly. We also hypothesized that lack of HS expression might affect the generation or maintenance of effector B cells. Indeed, we found that after primary infection with influenza, lack of EXT1 in B cells negatively affects the percentage of plasma cells present. Similarly, we observed a decreased in both plasma cells and GC B cells during the recall response. This observation suggests that while expression of HS on B cells has only a mild affect on GC B cells (only during the recall response), HS expression affects either the generation and/or maintenance of plasma cells. The fact that the decreased percentage of plasma cells did not lead to a decrease in influenza-specific antibody titers is likely due to comparable numbers of influenza-specific antibody secreting cells (Supplemental Figure 3). This suggests that while the immediate generation of influenza-specific antibody secreting cells (ASCs) proceeds normally in EXT1 cKO mice, the total number of plasma cells is reduced.
A possible explanation for this discrepancy is that HS expression on B cell contributes to the maintenance of plasma cells (those plasma cells induced at steady state, independent of influenza infection). As a result, EXT1 cKO mice generate similar numbers of influenza-specific plasma cells, but the non-influenza-specific plasma cells are either not generated or not maintained. In the future, it will be important to delineate the contribution of EXT1 in the generation and or maintenance of antigen/pathogen specific plasma cells over time. The data presented here suggests that the expression of HS on B cells is not necessary to respond to, or clear, the viral pathogen influenza. However, HS expression on B cells does affect the total proportion of effector B cells.
Having observed decreased poly-reactive IgM in aged EXT1 cKO mice, we were surprised that these mice were able to survive and respond to influenza infection comparable to age-matched littermate control mice. given that B-1 B cells secrete both steady state and infection induced poly-reactive IgM. A percent of these IgM molecules function to neutralize influenza during early infection, and they have been shown to be important for surviving an influenza infection in mice15. Due to the fact that older EXT1 cKO mice had similar viral titers at 7 and 14 days post infection (data not shown), we believe that even with lower titers of IgM in EXT1 cKO mice, there is still sufficient quantities to control the infection. If we assume that the poly-reactive IgM continues to wane in aging EXT1 cKO mice, we hypothesize that these mice will eventually exhibit defects in controlling influenza. However, one thing to consider is that older mice exhibit impaired immune responses to influenza [52]. Thus, it would be crucial to examine EXT1 cKO mice before they exhibit this age-dependent defect, but when EXT1 cKO are sufficiently old to have minimal poly-reactive IgM.
Refining our understanding of the importance of heparan sulfate expression on B cells
It is important to note that two other groups have recently investigated the role of HS expression on B cells. Garner and colleagues generated an identical EXT1 cKO mouse, however, they observed a lower efficiency of EXT1 deletion than we report here (Figure 1a) [23]. Similar to the findings presented here, Garner and colleagues also observed no defect in B cell development, with the exception of a slight increase in the number of EXT1 cKO progenitor B cells in the BM. This discrepancy is likely explained by the different genotypes of littermate controls used (EXT1fl/fl CD19Cre− versus our EXT1fl/+ CD19Cre+). Upon immunization of EXT1 cKO with the model antigens DNP-KLH or DNP-Ficoll, Garner et al. observed no defect in the titers of antibodies generated, which is similar to what we observed.
More recently, Reijmers et al. investigated the role of GLCE, an epimerization enzyme important for HS maturation [9]. By transferring GLCE-deficient fetal liver hematopoietic stem cells into Rag-2−/−γc−/− mice (the authors could not use GLCE-null mice due to embryonic lethality of GLCE-deficiency), Reijmers and colleagues generated mice whose entire hematopoietic compartment is unable to express functional HS. Similar to what we observed in competitive BM chimeras, Reijmers et al. reported a defect in B cell reconstitution and B cell development. Because the expression of functional HS on other lymphoid cells was impaired, Reijmers et al. could not attribute this defect in B cell development solely to the lack of HS expression on B cells. We saw no developmental defect in EXT1 cKO mouse, and believe what we observed in BM chimeras may be an over estimation of the role of HS in B cell development.
Our observation that lack of HS expression negatively affects the proportion of plasma cells is similar to what Reijmers et al. observed after immunization with the model antigen TNP-KLH (Figure 7 and supplemental Figure 1B). Unique to our report is the use of a pathogen, which we show induces high levels of HS expression on B cells. Furthermore, our data allows us to definitively associate the decreased titers of poly-reactive IgM as well as the decreased plasma B cells to EXT1-deficiency in B cells. Given the level of HS expression on plasma cells, as well as the importance of HS expression for responsiveness to the B cell specific cytokine APRIL [10],[9], this result is consistent with the predicted role of HS expression on plasma cells.
Our observation that EXT1-deficiency in B cells does not affect titers of antibodies generated after influenza infection is distinct from Reijmers et al., who found that the amount of antibodies generated after immunization with the model antigen TNP-KLH was decreased. This discrepancy is likely explained by the use of a different mouse model.
Concluding Remarks
In this report, we provide evidence that the expression of HS plays a nuanced role in B cell biology. While most B cell processes we evaluated were unaffected by the absence of HS, we provide definitive evidence that HS on B cells can affect the quantity of poly-reactive IgM in older mice. Furthermore, we also show that the total proportion of plasma cells is reduced in EXT1 cKO mice. Although this notable reduction in plasma B cells did not impair the ability to cope with the viral pathogen influenza, we hypothesize that HS expression on B cells influences either the generation or maintenance of steady state plasma cells. As a result, we anticipate that it may be key for mounting B cell responses to certain classes of antigens and/or pathogens. Perhaps induction of HS is key in certain types of infections, or perhaps it is important at a certain point in the B cell response that was not assessed here. Although we did not see any defect in the thymus-independent antibody response when EXT1 cKO mice were immunized with NP-Ficoll or NP-LPS (Supplemental Figure 4), HS expression may be key early in the immune response, when HS is most highly expressed on B cells. Alternatively, HS expression on B cells may serve to increase interactions with antigens containing HS-binding domains. Indeed, many viruses are known to encode proteins containing HS-binding domains [53–56]. Further studies examining the response of EXT1 cKO mice to different pathogens, including those that bind HS to gain entry into target cells, may refine our understanding of HS expression on B cells.
Acknowledgements
We would like to thank Professor Ellen Robey, Jenny Ross, and Seong Ji Han, Heather Melichar and Paul Herzmark for help/tutoring with Two-photon imaging and data analysis.