The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, U.
Right arrow Articles by Roeder, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, U.
Right arrow Articles by Roeder, R. G.
The Journal of Immunology, 2000, 165: 6825-6832.
Copyright © 2000 by The American Association of Immunologists

Genetic Analyses of NFKB1 and OCA-B Function: Defects in B Cells, Serum IgM Level, and Antibody Responses in Nfkb1-/-Oca-b-/- Mice1

Unkyu Kim, Cary S. Gunther and Robert G. Roeder2

Laboratory of Biochemistry and Molecular Biology, Rockefeller University, 1230 York Avenue, New York, NY 10021


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Defined patterns of gene expression during cell differentiation are likely to be ensured by multiple factors playing redundant roles. By generating mice deficient in both NFKB1 and OCA-B, we show here that the two transcription factors are required for B-1 cell differentiation and serum IgM production. In addition, relative to Nfkb1-/- or Oca-b-/- mice, the Nfkb1-/-Oca-b-/- mice show a decrease in conventional B cell frequencies in the spleen and augmented reductions in T-independent and T-dependent Ab responses. These results suggest that NFKB1 and OCA-B play compensatory roles in multiple aspects of B cell differentiation.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
In the past few years, targeted gene disruption studies in mice have implicated many genes in B cell differentiation. In addition to the signaling and receptor proteins, a host of transcription factors that ultimately regulate expression of such proteins have been found to be essential for normal B cell differentiation (1). Thus, disruptions of genes encoding transcription factors PU.1, Ikaros, early B cell factor, E2A, and B cell lineage-specific activator protein (BSAP)3 (Pax5) all block early B cell differentiation in the bone marrow. In contrast, genes encoding transcription factors such as Spi-B, Oct-2, NF-{kappa}B, Pip-1, and OCA-B are required for Ag-dependent B cell differentiation in the periphery.

OCA-B (also called Bob1 and OBF-1) is a transcription coactivator that functions in conjunction with Oct family transcription factors Oct-1 and Oct-2 to generate full activation of Ig promoters in vitro and in transient transfection assays (2, 3, 4, 5, 6). Surprisingly, mice lacking OCA-B can undergo normal Ag-independent B cell differentiation and produce normal levels of serum IgM, indicating that OCA-B function is not required for early B cell development and serum IgM production (7, 8, 9). However, OCA-B is essential for the Ag-dependent phase of B cell differentiation, as Oca-b-/- mice show severe deficiencies in serum IgG, IgA, and IgE levels, reduced Ab responses to T-independent (TI) and T-dependent (TD) Ags, and an inability to form germinal centers (7, 8, 9). Because Oca-b-/- B cells can undergo normal isotype switching when cultured in the presence of LPS and IL-4, but show reduced levels of transcripts from corresponding loci (7), the reduced levels of secondary isotypes appear to result from defects in transcription from switched loci.

Consistent with indications that transcription of heavy chain genes is driven by an intronic enhancer in early B cell differentiation and by the 3' enhancer in late B cell differentiation, it was recently shown that the IgH 3' enhancers are less active in Oca-b-/- B cells than in normal cells under certain conditions (10). This may reflect direct function of OCA-B through octamer sequence elements present in the 3' IgH enhancer (11) or an enhancer-specific synergy with OCA-B-activated promoters. Because B cell activation by CD40 and IL-4 induces OCA-B expression (12, 13), OCA-B also may regulate factors required for B cell activation that in turn induce 3' IgH enhancer function. Oca-b-/- splenic B cells are deficient in B cell receptor (BCR) signaling, because mutant B cells show decreased proliferation following surface IgM cross-linking (7). Furthermore, Oca-b-/- B cells, unlike normal B cells, do not exhibit repression of the 3' IgH enhancer following surface Ig cross-linking (10). Because a number of studies suggest that BCR signaling is required for normal differentiation and maintenance of B cells in the bone marrow and periphery (14, 15), inefficient BCR signaling by Oca-b-/- B cells is likely to affect the differentiation and survival capabilities of these cells.

Studies of mice lacking members of the NF-{kappa}B/Rel transcription factor family have shown that, like OCA-B, these proteins function in late B cell differentiation (16, 17). Thus, B cells in Nfkb1 (p50/p105) null mutant mice undergo normal Ag-independent differentiation, but show a severely reduced ability to proliferate in response to LPS stimulation and a moderately reduced ability to proliferate upon surface IgM cross-linking (18). Like OCA-B-deficient mice, mice lacking NFKB1 have reduced levels of serum IgG, IgA, and IgE but normal levels of IgM. These mutant mice do not respond well to TD Ags and are more susceptible to certain bacterial infections. However, in contrast to Oca-b-/- B cells, Nfkb1-/- B cells are defective in both transcription at IgG3 and IgE germline loci and isotype switching to IgG3, IgE, and IgA (19), but not in germinal center formation. Double-null mutations in Nfkb1 (p50/p105) and Nfkb2 (p52/p100) result in a B cell differentiation block earlier than that seen with single-null mutations, as the double-mutant mice contain severely reduced levels of B cells in the spleen (20). Adoptive transfer studies showed that this defect is B cell autonomous and that double-null B cells are functionally deficient and, like Oca-b-/- B cells, unable to form germinal centers. Therefore, NFKB1 and NFKB2 play compensatory roles in generating normal numbers of splenic B cells and in germinal center formation.

It has been proposed that cell lineage originates from stochastic low-level expression of lineage-affiliated genes in the progenitor cells (21, 22, 23). Defined patterns of gene expression within a lineage may be ensured by redundancy and feedback mechanisms involving multiple proteins. The criteria for functional redundancy are that mutations in single genes individually produce no phenotype, whereas mutation of all paralogs produces a strong phenotype (21). Therefore, a protein may be involved in pathways that are not disrupted by a mutation in that protein if its normal function is fully compensated by other factors playing redundant roles. Consequently, analyses of single mutations produce only a partial list of protein functions, and more extensive insight into a protein function can be gained by creating multiple mutations in genes that are suspected of playing redundant roles. Such functional redundancy has been observed not only among proteins within the same family (i.e., NFKB1 and NFKB2) but also between the two unrelated transcription factors early B cell factor and E2A (24). Considering the fact that both NFKB1 and OCA-B are required for late B cell differentiation, we wished to determine whether these two transcription factors participate in redundant pathways during late B cell differentiation. To this end, we generated mice lacking both NFKB1 and OCA-B by crossing the Nfkb1-/- mice with Oca-b-/- mice. Interestingly, the double-mutant mice showed severe deficiencies in B-1 cell development and serum IgM. Moreover, conventional B (B-2) cell frequencies in the spleen also were reduced in Nfkb1-/-Oca-b-/- mice. These results suggest that NFKB1 and OCA-B are involved in functionally redundant pathways that ensure proper B cell differentiation.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Generation of the Nfkb1-/-Oca-b-/- mice

Oca-b-/- mice (7) were mated with the Nfkb1-/- mice (18) mice (stock number 002849; B6, 129-Nfkb1tm1Bal; The Jackson Laboratory, Bar Harbor, ME) to generate mice heterozygous for Oca-b and Nfkb1 mutations. Interbreeding the heterozygotes generated homozygous double-knockout mice. All mice including the wild type were maintained in a C57BL and 129SV hybrid background in specific pathogen-free conditions. The following primers were used to identify the genotypes by PCR: primer set A, sense, TTGCAAACCTGGGAATACTTCATGT, antisense, GTGAGTTGCCGGTCTCCTCCGCCTT; set B, sense, CGTGACATTAGTGATGAGTGGTGGA, antisense, CCATACCAGGGCGTTCGAGTCA; set C, sense, CTCCCCTACCCGGTAGAATTGACC, antisense, CTGCTGGAATGCTATAGAGTGGTTC. Typically, 1 µg of tail DNA was amplified by 30 cycles of 45 s at 94°C, 45 s at 55°C, and 20–45 s at 72°C. For phenotype analyses, either littermates or age- and gender-matched 5- to 9-wk-old mice were used.

B cell purification and RT-PCR

B cells were purified from spleen by immunomagnetic cell sorting as follows. A single-cell suspension from spleen was washed with PBS/0.3% BSA and incubated with biotinylated anti-B220 Ab (PharMingen, San Diego, CA; 0.1 µg per 106 cells) at 4°C for 45 min. Cells were washed twice with PBS/2 mM EDTA to remove excess Ab. Cells were then incubated with streptavidin-conjugated M280 Dynabeads (Dynal, Great Neck, NY) at 4°C for 45 min. Labeled cells were collected on a magnet and washed twice with PBS/2 mM EDTA/0.5% BSA. Trizol (Life Technologies, Rockville, MD) was added to purified splenocytes. Total RNA was prepared following the manufacturer’s protocol. Total RNA was treated with DNase I to remove any contaminating DNA. Complementary DNA was prepared from 2 µg total RNA using random hexamer and avian myeloblastosis virus reverse transcriptase (Promega, Madison, WI). The cDNA amount was normalized by amplifying the ubiquitous GAPDH mRNA. Two 3-fold serial dilutions of cDNA per genotype were used to ensure that amplification was in a linear range. A typical PCR involved 28 cycles of 45 s at 94°C, 45 s at 55°C, and 30–60 s at 72°C. The following gene-specific primers were used: BLR-1 sense, ATGAACTACCCACTAACCCTGGAC; BLR-1 antisense, CTAGAAGGTGGTGAGGGAAGTAGC; Oct-2 sense, ACCTCCAGCCACCTGCTCAGTTC; Oct-2 antisense, TTGTGCTCAGACTGCTAGAAGCT; CD36 sense, GTGCTGATCCTTTCAGAGTCTC; CD36 antisense, CTGAGGAATGGATCTTTGTAAC; BSAP sense, CTACAGGCTCCGTGACGCAG; BSAP antisense, GTCTCGGCCTGTGAAATAGG; GAPDH sense, AAGCTTGTCATCAACGGGAAGCCCATC; GAPDH antisense, GATGTTCTGGGCAGCCCCACGGCCATC.

Flow cytometry

For bone marrow analyses, marrow cells were flushed out from the thighbone using PBS. Splenocytes were prepared by squashing the spleen using the top of a plunger from a 1-ml syringe. Peritoneal lavage was obtained by flooding the peritoneal cavity with 5 ml PBS. RBC were lysed using 160 mM NH4Cl/10 mM KHCO3/0.1 mM EDTA, and debris was filtered out with a 40-µm nylon membrane. After blocking the Fc receptors with FcBlock (PharMingen), cells were incubated with fluorescently labeled antisera on ice. Cells were washed with PBS/0.3% BSA, fixed with 1% formaldehyde (methanol-free EM grade; Polysciences, Warrington, PA), and analyzed on a FACScan using CellQuest software (Becton Dickinson, Mountain View, CA). Lymphocytes were gated according to the forward- and side-scatter patterns to exclude large granular cells and debris. For each experiment, isotype controls were used as the negative staining controls. Quadrants were drawn according to the negative controls and confirmed by single-stain controls. Numbers represent the percentage of gated cells in the region or quadrant. Data are illustrated as dot plots or 20% probability contour plots. In each case, a representative data set out of five separate experiments is shown. The following antisera recognizing mouse cell surface markers were used: from PharMingen, anti-B220, clone RA3-6b2; anti-IgM, clone II/41; anti-CD4, clone H129.19; anti-CD8, clone 53-6.7; anti-CD11b, clone M1/70; anti-IgD, clone 11-26c.2a; anti-CD5, clone 53-7.3; anti-CD43, clone S7; anti-CD23, clone B3B4; from Southern Biotechnology Associates (Birmingham, AL), goat anti-IgM R-PE conjugate; and from Caltag (South San Francisco, CA), anti-CD11b Tricolor conjugate, clone M1/70.15.

Immunization and determination of serum Ig levels

Total Ig levels (IgM and IgG) were determined by sandwich ELISA using a mouse isotyping kit (Southern Biotechnology Associates). For IgA measurements, anti-mouse IgA (Southern Biotechnology Associates) was used as a capture Ab, and biotinylated anti-mouse IgA (Sigma) followed by avidin-HRP conjugate was used as detecting Ab. For all experiments, eight 1:5 serially diluted sera and standard Igs were added to the plate. Ig concentrations were determined using a standard curve generated from standard Igs. For TI Ab responses, mice were i.p. injected with 50 µg DNP-conjugated Ficoll (Biosearch Technologies, Novato, CA). Blood was collected before and 10 days after the immunization. TD responses were analyzed by injecting mice with 100 µg DNP-keyhole limpet hemocyanin (KLH; Calbiochem, La Jolla, CA) with CFA. After 3 wk, mice were injected again with 10 µg DNP-KLH with IFA. Blood was collected before and 10 days after the first and second injections. Ab to DNP was measured by ELISA using plates coated with either DNP-OVA (OVA; Biosearch Technologies) or DNP-gammaglobulin (Calbiochem). Eight 1:5 serially diluted sera were used to determine the titer. The DNP-specific titer was defined as fold serum dilution that resulted in background absorbance.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Generation of mice deficient in both OCA-B and NFKB1

Mice lacking both NFKB1 (p50/p105) and OCA-B were made by crossing Oca-b-/- mice with Nfkb1-/- mice (Fig. 1Go). As expected, the Nfkb1-/-Oca-b-/- mice expressed neither NFKB1 nor OCA-B mRNA, as determined by RT-PCR assays (Fig. 1GoB). Recently, it was reported that the promoter of the B cell-specific chemokine receptor gene Blr-1 (Cxcr-5) contains NF-{kappa}B and octamer sites and that this promoter is synergistically activated by the NFKB1/RelA heterodimer and the Oct/OCA-B complex (25). The BLR-1 mRNA level was decreased 2- to 3-fold in the Oca-b-/- splenic B cells, indicating that OCA-B is indeed necessary for full activation of this promoter (Fig. 1GoB, data not shown). In contrast, the Nfkb1-/- splenic B cells expressed normal levels of BLR-1 mRNA, suggesting that NFKB1 is not required for activation of the Blr-1 promoter. In the Nfkb1-/-Oca-b-/- splenic B cells, the level of BLR-1 mRNA was decreased to a level comparable to that in Oca-b-/- B cells. This suggests that OCA-B is necessary for BLR-1 expression whereas NFKB1 is not. In contrast, mRNAs for OCT-2 and its presumptive target CD36 (26) were present at normal levels in both the single- and double-mutant mice.



View larger version (38K):
[in this window]
[in a new window]
  		FIGURE 1. Genotyping and RT-PCR assay. A, A representative set of PCR amplifications from tail DNA of progenies of a mating between Nfkb1+/-Oca-b+/- and Nfkb1-/-Oca-b+/- mice. PCR primer set A amplifies a 150-bp region within the Nfkb1 exon 6 that is replaced with the phosphoglycerate kinase NEO cassette and missing in the mutated locus. Likewise, primer set B amplifies a 900-bp region that is missing in the mutated Oca-b locus. Primer set C amplifies a 1500-bp region that is present only in the mutated Oca-b locus. B, RT-PCR analyses of B220+ splenic B cells. RT-PCR using 1 µl and 3 µl of normalized cDNA are shown for each genotype with no DNA (-) controls. Reverse transcription (RT) reaction was duplicated in the absence of RT, and 3 µl of RT reaction was amplified by PCR (RT-). Northern blot analyses using the 1.2-kb BLR-1 cDNA probe amplified by RT-PCR show that Oca-b-/- splenocytes contain 2- to 3-fold less BLR-1 mRNA (data not shown).

 
Mice lacking OCA-B and NFKB1 contain reduced numbers of splenic B cells

To study B cell development in the absence of OCA-B and NFKB1, we analyzed cell surface marker expression in lymphoid organs by flow cytometry. Similar numbers of nucleated bone marrow cells were obtained from wild-type and mutant mice (24 ± 9 x 106 for wild type, 19 ± 8 x 106 for Nfkb1-/-, 25 ± 18 x 106 for Oca-b-/-, and 26 ± 8 x 106 for Nfkb1-/-Oca-b-/-). Previous studies indicated that B cell differentiation in the bone marrow proceeds normally in the absence of either NFKB1 or OCA-B (7, 8, 9, 18). To find out whether NFKB1 and OCA-B play redundant roles in early B cell differentiation, we analyzed B cell differentiation in the bone marrow of Nfkb1-/-Oca-b-/- mice. The double-mutant mice contained slightly reduced proportions of B220+IgM+ immature B cells (Fig. 2GoA). However, analyses of B220 and CD43 expression did not show any obvious differences (data not shown), indicating that early B cell differentiation is largely normal in the absence of both NFKB1 and OCA-B. Previous studies of Oca-b-/- mice indicated that there is a reduction of the recirculating B cell population in bone marrow (8, 9, 27). This deficiency in recirculating B cells was most pronounced in young (3- to 4-wk-old) mice, while in older (20-wk) mice the recirculating B cell population was largely restored. In the present study, when mice were analyzed between 5 and 9 wk of age, the recirculating B cell population (defined as B220highIgDhigh cells) was significantly reduced in both Nfkb1-/- and Oca-b-/- mice, as well as in Nfkb1-/-Oca-b-/- mice (Fig. 2GoA).



View larger version (71K):
[in this window]
[in a new window]
  		FIGURE 2. A, Recirculating B cell number is severely diminished in Nfkb1-/-Oca-b-/- bone marrow. Flow cytometry of bone marrow cells of 5- to 9-wk-old mice doubly stained with anti-B220-PE/anti-IgM FITC or anti-B220-PE/anti-IgD-FITC. The percentage of B220+IgD+ cells in the upper right quadrant is 14 for wild type, 10 for Nfkb1-/-, 9 for Oca-b-/-, and 6 for Nfkb1-/-Oca-b-/-. B, Splenic B cell frequencies are reduced in Nfkb1-/-Oca-b-/- mice. Flow cytometry of splenocytes labeled with anti-B220-PE/anti-IgM-FITC, anti-IgM-PE/anti-IgD-FITC, or anti-CD4-PE/anti-CD8-FITC.

 
The immature B cells that have completed differentiation in the bone marrow exit to the blood stream and home to the spleen. In the spleen, only a fraction of immature B cells from bone marrow are selectively included in the long-lived recirculating B cell pool. The total number of nucleated splenocytes was reduced 2- to 3-fold in the mutant mice (24 ± 11 x 106 for wild-type, 13 ± 6 x 106 for Nfkb1-/-, 9 ± 5 x 106 for Oca-b-/-, and 12 ± 4 x 106 for Nfkb1-/-Oca-b-/-). However, frequencies of the B220+IgM+ and IgM+IgD+ B cell populations were quite normal in the Nfkb1-/- and Oca-b-/- spleens (Fig. 2GoB). Interestingly, the Nfkb1-/-Oca-b-/- mice showed 2- to 3-fold reductions in frequencies of both the B220+IgM+ and IgM+IgD+ B cell populations (Fig. 2GoB). Therefore, NFKB1 and OCA-B play redundant roles in the generation and/or maintenance of normal splenic B cell frequencies.

B splenocytes can be further subdivided according to IgM and IgD expression patterns (28). Transitional B cells recently arrived from the bone marrow are IgMhighIgDlow (T1) and subsequently IgMhighIgDhigh (T2), whereas mature recirculating B cells are IgMlowIgDhigh. An increase in the transitional B cell population with a concomitant decrease in the mature B cell population would indicate that there is a developmental block preventing B cell progression from the transitional to the mature B cell stage. For example, B cell differentiation is blocked at the T2 to mature B cell transition in B220 null and X-linked immunodeficiency (XID) mice (28). However, in the present case the Nfkb1-/-Oca-b-/- mice exhibited defects in both transitional and mature B cell populations (Fig. 2GoB). This suggests that the developmental block is at or before the arrival of the transitional B cells in the spleen of Nfkb1-/-Oca-b-/- mice. Alternatively, there might be a defect in survival rate of the mutant B cells. T cell differentiation to CD4 or CD8 single-positive stages occurs normally in the double-knockout thymus (data not shown). However, there is up to a 2-fold increase in the frequency of CD8+ cells in the Nfkb1-/-Oca-b-/- spleen (Fig. 2GoB). This altered CD4+/CD8+ frequency in the Nfkb1-/-Oca-b-/- spleen might indicate a dependency of CD4+ T cell maintenance on B cell function. However, we cannot exclude the possibility that NFKB1 and OCA-B might directly function in redundant T cell pathways that maintain normal CD4+/CD8+ frequencies.

B-1 cell development is defective in the absence of NFKB1 and OCA-B

In addition to the well-studied conventional B cells (B-2 cells), there is a separate population of B cells (B-1 cells) that is found mainly in the adult peritoneal and pleural cavities (29, 30, 31, 32). B-1 cells make a major contribution to serum IgM, IgG3, and IgA levels and respond well to TI Ags such as bacterial coat Ags and autoantigens. Recent studies showed that natural IgM produced by B-1 cells is crucial for complement-mediated inflammatory diseases (33). B-1 cells are subdivided into B-1a and B-1b populations on the basis of cell surface markers and progenitor sources (34). All B-1 cells in the peritoneum are CD11b (MAC1)+, IgM+, and CD23-. B-1a cells express CD5 on the cell surface, whereas B-1b cells do not. Adoptive transfer studies show that the precursors for B-1a cells are restricted to fetal liver and omentum, while progenitors for B-1b cells are found in adult bone marrow as well as in fetal liver and omentum (34). An alternative view proposes that B-1 cells are generated from activated B-2 cells rather than from a separate lineage (35, 36). In addition, recent studies using the hen egg lysozyme transgenic mouse model system have shown that anergic B cells can assume a B-1a cell phenotype by up-regulating the expression of CD5, which inhibits the BCR signal induced by the autoantigen (37). Regardless of the origin of the B-1 cells, several recent studies showed that B-1 cell differentiation and survival are critically dependent on positive selection through BCR signaling (38, 39, 40).

We studied B-1 cell development in the various mutant mice by analyzing cell surface expression of CD5 and CD11b in the peritoneum (Fig. 3Go). Staining for the total B-1 cells (CD11b+IgM+) indicated that Oca-b-/- mice contained an increased frequency of B-1 cells in the peritoneum (Fig. 3GoA). In contrast, there was a 2- to 3-fold reduction in the frequency of CD11b+IgM+ B-1 cells in mice lacking NFKB1 (Fig. 3GoA). In fact, the absolute numbers of CD11b+IgM+ cells were 2- to 3-fold elevated in the Oca-b-/- mice and 2- to 3-fold reduced in the Nfkb1-/- mice (1.2 ± 0.6 x 106 for wild-type, 0.5 ± 0.3 x 106 for Nfkb1-/-, 3 ± 2 x 106 for Oca-b-/-, and 0.05 ± 0.04 x 106 for Nfkb1-/-Oca-b-/-). Three-color flow cytometry showed that this reduction in the total B-1 cell population in Nfkb1-/- mice is specifically due to a severe reduction in the CD5+IgM+ B-1a cell population, whereas the CD5-IgM+ cell population appears to be unaffected by the Nfkb1 mutation (Fig. 3GoB, upper panel). The CD11b expression patterns of the IgM+CD5- gated fraction showed that Nfkb1-/- mice contained a normal proportion of B-1b (CD11b+CD5-IgM+) cells (Fig. 3GoB, lower panel). Thus, our analyses indicate that NFKB1 is selectively required for B-1a cell differentiation. In Oca-b-/- mice, there was an increase in the B-1a (CD5+IgM+) population (Fig. 3GoB, upper panel) and a slight decrease in the CD11b+CD5-IgM+ population (Fig. 3GoB, lower panel). Interestingly, there was a drastic loss of the IgMhighCD11b+CD5- population in the Oca-b-/- mice (boxed, Fig. 3GoB, lower panel), indicating that a subset of B-1b cells are reduced in the absence of OCA-B. Importantly, Nfkb1-/-Oca-b-/- mice had severe deficiencies in both the B-1a and the B-1b cell populations in the peritoneum (Fig. 3Go, A and B). In fact, the reduction in B-1 cells in Nfkb1-/-Oca-b-/- mice was comparable to that seen in the XID mice (Fig. 3Go, A and B). Taken together, these results suggest that NFKB1 and OCA-B are required for B-1 cell differentiation and that each factor selectively affects either B-1a (NFKB1) or B-1b (OCA-B) cell differentiation.



View larger version (56K):
[in this window]
[in a new window]
  		FIGURE 3. B-1 cell differentiation is defective in Nfkb1-/-Oca-b-/- mice. A, Flow cytometry of CD11b-PE and IgM-FITC double-stained peritoneal cells. As a comparison, peritoneal cells from XID mice (CBA/CaHN-Btkxid/J; The Jackson Laboratory) exhibiting a severe reduction in B-1 cells are shown. B, Peritoneal cell staining with anti-CD5-PE, anti-IgM-FITC, and anti-CD11b-Tricolor. Lower panels show profiles of cells gated on the CD5-IgM+ population (boxed). Frequencies of cells within the IgMhighCD11b+ fraction indicated by the rectangular regions in the lower panel are: 32% (wild type), 41% (Nfkb1-/-), 11% (Oca-b-/-), 2% (Nfkb1-/-Oca-b-/-), and 7% (XID).

 
Mice lacking OCA-B and NFKB1 produce severely reduced levels of Igs, including IgM, and deficient TI and TD Ab responses

We measured the total serum Ig levels to see whether the reduced number of Nfkb1-/-Oca-b-/- B cells in the periphery function normally (Fig. 4Go). As expected from the single-knockout phenotypes, the secondary heavy chain isotypes were all reduced, albeit to variable degrees, in the Nfkb1-/-Oca-b-/- mice. Thus, whereas IgG1 and IgG2b levels were slightly increased in Nfkb1-/-Oca-b-/- mice compared with Oca-b-/- mice, IgG2b, IgG3, and IgA levels were decreased even further. It is unclear whether these slight differences are functionally significant. Surprisingly, however, the IgM levels in the Nfkb1-/-Oca-b-/- mice were greatly reduced, to <10% of the wild-type level. Because single-knockout mice produce normal levels of IgM, this is another phenotype that reflects redundant functions of OCA-B and NFKB1. Considering that B-1 cells produce a major portion of serum IgM, the reductions in IgM levels might be due to the virtual absence of B-1 cells in the Nfkb1-/-Oca-b-/- mice.



View larger version (11K):
[in this window]
[in a new window]
  		FIGURE 4. Nfkb1-/-Oca-b-/- mice contain severely reduced levels of serum IgM, IgG, and IgA. Isotype-specific ELISA showing the serum Ig levels. Wild type, {circ}; Nfkb1-/-, {triangleup}; Oca-b-/-, {square}; Nfkb1-/-Oca-b-/-, •. Bars denote the average values. *, Two mice had IgG2a levels below the detection limit.

 
To further test whether the Nfkb1-/-Oca-b-/- B cells are functional in vivo, we measured Ab responses to TI (DNP-Ficoll) and TD (DNP-KLH) Ags. As shown in Fig. 5GoA, the Oca-b-/- mice showed diminished IgM (18-fold) and IgG3 (90-fold) responses to TI Ag. This is in agreement with previous reports (8). In the present study, mice lacking NFKB1 also showed decreased TI responses (4-fold for IgM and 8-fold for IgG3), although the degree of reduction was less than in Oca-b-/- mice (Fig. 5GoA). Moreover, the Nfkb1-/-Oca-b-/- mice showed synergistic reductions in TI responses, as no DNP-specific antisera were detected above preimmune levels. Similarly, and consistent with previous reports, TD responses were greatly reduced in both the Nfkb1-/- (6-fold for IgM, 60-fold for IgG1, and 400-fold for IgA responses) and the Oca-b-/- (4-fold for IgM, 500-fold for IgG1, and 600-fold for IgA responses) mice (Fig. 5GoB). Again, Nfkb1-/-Oca-b-/- mice showed even greater defects and produced very little DNP-specific antisera. These results suggest that NFKB1 and OCA-B play functionally redundant roles in TI and TD Ab responses.



View larger version (43K):
[in this window]
[in a new window]
  		FIGURE 5. TI and TD Ab responses are defective in the Nfkb1-/-Oca-b-/- mice. Anti-DNP Ab levels were measured by ELISA in mice injected with the TI Ag DNP-Ficoll (A) or the TD Ag DNP-KLH (B). Average values of groups of three mice per genotype are shown with SDs.

 
Previous studies showed that Oca-b-/- mice lacked germinal centers in both spleen and lymph nodes, whereas Nfkb1-/- mice could form germinal centers in response to TD Ags. The Nfkb1-/-Oca-b-/- mice did not form any germinal centers when immunized with the TD Ag DNP-KLH (data not shown). Therefore, OCA-B is still required for germinal center formation in an NFKB1-deficient background.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
This study describes analyses of B cell differentiation and immune functions in mice lacking both OCA-B and NFKB1. Both of these transcription factors have been shown to be crucial for late B cell maturation and function and both are induced upon B cell activation. Therefore, these two transcription factors are likely to function in redundant or interacting feedback loops that ensure proper B cell responses during the Ag-driven maturation process. As detailed below and summarized in Table IGo, the analyses of the double-null mutation have revealed functional redundancies between NFKB1 and OCA-B and novel roles for these two key transcription factors in B cell differentiation.


View this table:
[in this window]
[in a new window]
  		Table I. Genetic Analysis of B Cell Pathways Mediated by NFKB1 and OCA-Ba

 
B-1 cell differentiation and serum IgM levels

Our present study shows that Nfkb1-/- mice have a severely reduced B-1a subpopulation (CD5+CD11b+IgM+) but normal levels of B-1b cells (CD5-CD11b+IgM+) (Fig. 3Go, A and B). In contrast, Oca-b-/- mice contain elevated numbers of B-1a cells and reduced numbers of IgMhigh B-1b cells (Fig. 3Go, A and B). Interestingly, the double-knockout mice contain drastically reduced levels of both B-1a and B-1b cells (Fig. 3Go). This suggests that NFKB1 and OCA-B are required for B-1 cell differentiation. This also suggests that although B-1a and B-1b cells are considered to be closely related, each are under a distinct transcriptional control. Currently, only limited information is available on the transcriptional regulation of B-1 cell differentiation. Analyses of mice lacking either Oct-2 or Pax-5 (BSAP) have shown that these transcription factors are required for B-1 cell differentiation (41, 42). However, in these studies on Oct-2-/- and Pax-5-/- mice, only CD5+ B-1a cells were analyzed, and no information is available on the requirement of these factors in B-1b cell differentiation. Therefore, ours is the first report to show that two distinct transcription factors regulate B-1a or B-1b cell differentiation. There is growing evidence that positive selection through the BCR plays a crucial role in B-1 cell differentiation (38, 39, 40). Therefore, the B-1 cell deficiencies observed in the mutant mice might be due in part to defective BCR signaling in the mutant B cells (7, 10, 18).

It has been proposed that B-1 cells make a major contribution to serum IgM levels. For example, depletion of B-1 cells by anti-IL-10 treatment results in a 90% reduction in serum IgM levels, whereas IgM responses to TD Ag by conventional B cells are normal (43). Therefore, it is likely that reductions in serum IgM levels in the Nfkb1-/-Oca-b-/- mice (Fig. 4Go) are, in large part, a consequence of the severe reduction in the B-1 populations. The inability of conventional Nfkb1-/-Oca-b-/- B cells to generate specific IgM Abs against TI and TD Ags (Fig. 5Go) may contribute further to the deficiencies in the total serum IgM levels.

Conventional (B-2) B cell differentiation

In the spleen, B220+IgM+ B cell frequencies are relatively normal in the single-knockout mice (Fig. 2GoB). In contrast, the double mutation causes a 2- to 3-fold reduction in the B220+IgM+ B cell frequency. These results suggest that OCA-B and NFKB1 play redundant roles in the generation and/or maintenance of normal frequency of conventional B cells in the periphery. While analyzing the cell surface expression of IgM and IgD in the spleen, we noticed that the mean intensity of both IgM and IgD was consistently reduced in Oca-b-/- and Nfkb1-/-Oca-b-/- mice. In fact, this reduction is quite obvious in the contour plot of the double staining (Fig. 2GoB). We do not know the precise reason for this reduction, but it is possible that the peripheral B cells in the Oca-b-/- and the Nfkb1-/-Oca-b-/- mice are abnormal. Thus, in addition to being reduced in number, the peripheral B cells in Nfkb1-/-Oca-b-/- mice may be functionally distinct from the B cells in wild-type mice. Detailed cellular analyses are under way to study the activation processes in mutant B cells.

Defects in switched isotype levels, Ab responses, and germinal center formation

Single-null mutations in either Nfkb1 or Oca-b lead to decreased serum IgG and IgA levels. In the Oca-b-/- mice these defects may reflect essential OCA-B function both in germinal center formation (7, 8, 9) and in optimal 3' enhancer-dependent expression of switched heavy chain loci (10). In contrast, the decreased serum isotype levels in Nfkb1-/- mice may reflect the fact that unlike OCA-B, NFKB1 is required for germline immunoglobulin transcription and isotype switching, but not for germinal center formation (18, 19). It is currently unknown whether NFKB1 plays a role in transcription of the switched isotype loci through the 3' IgH enhancer. However, the presence of functional NF-{kappa}B sites along with octamer sites in the 3' IgH enhancer (11) suggests that transcription of switched isotype loci might be coordinately regulated by NFKB1 and OCA-B.

Ab responses to either TI (DNP-Ficoll) or TD (DNP-KLH) Ags are significantly reduced in the single-knockout mice. In the Nfkb1-/-Oca-b-/- mice, the responses are further reduced, with very little anti-DNP antisera detectable (Fig. 5Go). Functional redundancy between NFKB1 and OCA-B is obvious in both TI and TD responses, as the single-knockout mice produce significantly reduced (4- to 600-fold) yet measurable amounts of DNP-specific antisera, whereas the Nfkb1-/-Oca-b-/- mice produce very little anti-DNP antisera above the preimmune level. These results suggest that OCA-B and NFKB1 play functionally redundant roles in the production of TI and TD Ab responses. The roles played by these factors in isotype switching, germinal center formation, and 3' IgH enhancer-dependent transcription may be part of this process. However, the degree of the defect produced by the double-null mutation is more than the simple sum of defects caused by the single mutations. Therefore, it is likely that NFKB1 and OCA-B play additional roles in generating specific Ab responses.

Collaboration between NFKB1 and OCA-B

Currently, the only known target gene that is coordinately regulated by NFKB1 and OCA-B is the B cell chemokine receptor Blr-1 (25). Our analyses of BLR-1 mRNA levels in the single- and double-mutant B cells suggest that NFKB1 function on this promoter is compensated by other NF-{kappa}B family proteins (Fig. 1GoB). Considering the defects in the Nfkb1-/-Oca-b-/- mice, it is possible that there are other target promoters/enhancers that are directly and synergistically regulated by NFKB1 and OCA-B. It is also possible that NFKB1 and OCA-B activate separate target genes that regulate the same downstream genes within a B cell differentiation pathway. A large-scale gene expression profiling of the mutant mice should yield interesting information on potential target genes and pathways coordinately regulated by NFKB1 and OCA-B.


   Acknowledgments
 
We thank J. Shou for histological examinations and instructions on peritoneal lavage preparation. We also thank the Hospital for Special Surgery flow cytometry core facility for assistance with FACS analyses.


   Footnotes
 
1 This work was supported by National Institutes of Health Grant CA42567 (to R.G.R.). C.S.G. was supported by National Institutes of Health Medical Scientist Training Program Grant GM07739, National Institutes of Health Grant 5T32CA09673, and institutional funds. U.K. was supported by a Leukemia and Lymphoma Society Special Fellowship. Back

2 Address correspondence and reprint requests to Dr. Robert G. Roeder, Laboratory of Biochemistry and Molecular Biology, Rockefeller University, 1230 York Avenue, New York, NY 10021. Back

3 Abbreviations used in this paper: BSAP, B cell lineage-specific activator protein; TI, T independent; TD, T dependent; BCR, B cell receptor; KLH, keyhole limpet hemocyanin; XID, X-linked immunodeficiency; RT, reverse transcription. Back

Received for publication June 12, 2000. Accepted for publication September 21, 2000.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Reya, T., R. Grosschedl. 1998. Transcriptional regulation of B-cell differentiation. Curr. Opin. Immunol. 10:158.[Medline]
  2. Luo, Y., H. Fujii, T. Gerster, R. G. Roeder. 1992. A novel B cell-derived coactivator potentiates the activation of immunoglobulin promoters by octamer-binding transcription factors. Cell 71:231.[Medline]
  3. Luo, Y., R. G. Roeder. 1995. Cloning, functional characterization, and mechanism of action of the B-cell-specific transcriptional coactivator OCA-B. Mol. Cell Biol. 15:4115.[Abstract]
  4. Strubin, M., J. W. Newell, P. Matthias. 1995. OBF-1, a novel B cell-specific coactivator that stimulates immunoglobulin promoter activity through association with octamer-binding proteins. Cell. 80:497.[Medline]
  5. Gstaiger, M., L. Knoepfel, O. Georgiev, W. Schaffner, C. M. Hovens. 1995. A B-cell coactivator of octamer-binding transcription factors. Nature 373:360.[Medline]
  6. Luo, Y., H. Ge, S. Stevens, H. Xiao, R. G. Roeder. 1998. Coactivation by OCA-B: definition of critical regions and synergism with general cofactors. Mol. Cell Biol. 18:3803.[Abstract/Free Full Text]
  7. Kim, U., X. F. Qin, S. Gong, S. Stevens, Y. Luo, M. Nussenzweig, R. G. Roeder. 1996. The B-cell-specific transcription coactivator OCA-B/OBF-1/Bob-1 is essential for normal production of immunoglobulin isotypes. Nature 383:542.[Medline]
  8. Schubart, D. B., A. Rolink, M. H. Kosco-Vilbois, F. Botteri, P. Matthias. 1996. B-cell-specific coactivator OBF-1/OCA-B/Bob1 required for immune response and germinal centre formation. Nature 383:538.[Medline]
  9. Nielsen, P. J., O. Georgiev, B. Lorenz, W. Schaffner. 1996. B lymphocytes are impaired in mice lacking the transcriptional co-activator Bob1/OCA-B/OBF1. Eur. J. Immunol. 26:3214.[Medline]
  10. Stevens, S., J. Ong, U. Kim, L. A. Eckhardt, R. G. Roeder. 2000. The role of OCA-B in 3' IgH enhancer function. J. Immunol. 164:5306.[Abstract/Free Full Text]
  11. Michaelson, J. S., M. Singh, C. M. Snapper, W. C. Sha, D. Baltimore, B. K. Birshtein. 1996. Regulation of 3' IgH enhancers by a common set of factors, including {kappa}B-binding proteins. J. Immunol. 156:2828.[Abstract]
  12. Qin, X. F., A. Reichlin, Y. Luo, R. G. Roeder, M. C. Nussenzweig. 1998. OCA-B integrates B cell antigen receptor-, CD40L- and IL 4-mediated signals for the germinal center pathway of B cell development. EMBO J. 17:5066.[Medline]
  13. Stevens, S., L. Wang, R. G. Roeder. 2000. Functional analysis of the OCA-B promoter. J. Immunol. 164:6372.[Abstract/Free Full Text]
  14. Benschop, R. J., J. C. Cambier. 1999. B cell development: signal transduction by antigen receptors and their surrogates. Curr. Opin. Immunol. 11:143.[Medline]
  15. Lam, K. P., R. Kuhn, K. Rajewsky. 1997. In vivo ablation of surface immunoglobulin on mature B cells by inducible gene targeting results in rapid cell death. Cell 90:1073.[Medline]
  16. Ghosh, S., M. J. May, E. B. Kopp. 1998. NF-{kappa}B and Rel proteins: evolutionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16:225.[Medline]
  17. Gerondakis, S., R. Grumont, I. Rourke, M. Grossmann. 1998. The regulation and roles of Rel/NF-{kappa}B transcription factors during lymphocyte activation. Curr. Opin. Immunol. 10:353.[Medline]
  18. Sha, W. C., H.-C. Liou, E. Tuomanen, D. Baltimore. 1995. Targeted disruption of the p50 subunit of NF-{kappa}B leads to multifocal defects in immune response. Cell 80:321.[Medline]
  19. Snapper, C. M., P. Zelazowski, F. R. Rosas, M. R. Kehry, M. Tian, D. Baltimore. 1996. B cells from p50/NF-{kappa}B knockout mice have selective defects in proliferation, differentiation, germline CH transcription, and Ig class switching. J. Immunol. 156:183.[Abstract]
  20. Franzoso, G., L. Carlson, L. Xing, L. Poljak, E. W. Shores, K. D. Brown, A. Leonardi, T. Tran, B. Boyce, U. Siebenlist. 1997. Requirement of NF-{kappa}B in osteoclast and B cell development. Genes Dev. 11:3482.[Abstract/Free Full Text]
  21. McAdams, H. H., A. Arkin. 1999. It’s a noisy business! Genetic regulation at the nanomolar scale. Trends Genet. 15:65.[Medline]
  22. Reiner, S. L., R. A. Seder. 1999. Dealing from the evolutionary pawnshop: how lymphocytes make decisions. Immunity 11:1.[Medline]
  23. Enver, T., M. Greaves. 1998. Loops, lineage, and leukemia. Cell. 94:9.[Medline]
  24. O’Riordan, M., R. Grosschedl. 1999. Coordinate regulation of B cell differentiation by the transcription factors EBF and E2A. Immunity 11:21.[Medline]
  25. Wolf, I., V. Pevzner, E. Kaiser, G. Bernhardt, E. Claudio, U. Siebenlist, R. Forster, M. Lipp. 1998. Downstream activation of a TATA-less promoter by Oct-2., Bob1, and NF-{kappa}B directs expression of the homing receptor BLR1 to mature B cells. J. Biol. Chem. 273:28831.[Abstract/Free Full Text]
  26. Pfisterer, P., J. Hess, T. Wirth. 1997. Identification of target genes of the lymphoid-specific transcription factor Oct2. Immunobiology 198:217.[Medline]
  27. Schubart, D. B., A. Rolink, K. Schubart, P. Matthias. 2000. Cutting edge: lack of peripheral B cells and severe agammaglobulinemia in mice simultaneously lacking Bruton’s tyrosine kinase and the B cell-specific transcriptional coactivator OBF-1. J. Immunol. 164:18.[Abstract/Free Full Text]
  28. Loder, F., B. Mutschler, R. J. Ray, C. J. Paige, P. Sideras, R. Torres, M. C. Lamers, R. Carsetti. 1999. B cell development in the spleen takes place in discrete steps and is determined by the quality of B cell receptor-derived signals. J. Exp. Med. 190:75.[Abstract/Free Full Text]
  29. Kantor, A. B., L. A. Herzenberg. 1993. Origin of murine B cell lineages. Annu. Rev. Immunol. 11:501.[Medline]
  30. Hayakawa, K., R. R. Hardy. 2000. Development and function of B-1 cells. Curr. Opin. Immunol. 12:346.[Medline]
  31. Tarakhovsky, A.. 1997. Bar Mitzvah for B-1 cells: how will they grow up?. J. Exp. Med. 185:981.[Free Full Text]
  32. Youinou, P., C. Jamin, P. M. Lydyard. 1999. CD5 expression in human B-cell populations. Immunol. Today 20:312.[Medline]
  33. Carroll, M. C., A. P. Prodeus. 1999. Linkages of innate and adaptive immunity. Curr. Opin. Immunol. 10:36.
  34. Kantor, A. B., A. M. Stall, S. Adams, L. A. Herzenberg, L. A. Herzenberg. 1992. Differential development of progenitor activity for three B-cell lineages. Proc. Natl. Acad. Sci. USA 89:3320.[Abstract/Free Full Text]
  35. Clarke, S. H., L. W. Arnold. 1998. B-1 cell development: evidence for an uncommitted immunoglobulin(Ig)M+ B cell precursor in B-1 cell differentiation. J. Exp. Med. 187:1325.[Abstract/Free Full Text]
  36. Huang, C. A., C. Henry, J. Iacomini, T. Imanishi-Kari, H. H. Wortis. 1996. Adult bone marrow contains precursors for CD5+ B cells. Eur. J. Immunol. 26:2537.[Medline]
  37. Hippen, K. L., L. E. Tze, T. W. Behrens. 2000. CD5 maintains tolerance in anergic B cells. J. Exp. Med. 191:883.[Abstract/Free Full Text]
  38. Lam, K. P., K. Rajewsky. 1999. B cell antigen receptor specificity and surface density together determine B-1 versus B-2 cell development. J. Exp. Med. 190:471.[Abstract/Free Full Text]
  39. Hayakawa, K., M. Asano, S. A. Shinton, M. Gui, D. Allman, C. L. Stewart, J. Silver, R. R. Hardy. 1999. Positive selection of natural autoreactive B cells. Science 285:113.[Abstract/Free Full Text]
  40. Chumley, M. J., J. M. Dal Porto, S. Kawaguchi, J. C. Cambier, D. Nemazee, R. R. Hardy. 2000. A VH11V{kappa}9 B cell antigen receptor drives generation of CD5+ B cells both in vivo and in vitro. J. Immunol. 164:4586.[Abstract/Free Full Text]
  41. Humbert, P. O., L. M. Corcoran. 1997. Oct-2 gene disruption eliminates the peritoneal B-1 lymphocyte lineage and attenuates B-2 cell maturation and function. J. Immunol. 159:5273.[Abstract]
  42. Urbanek, P., Z. Q. Wang, I. Fetka, E. F. Wagner, M. Busslinger. 1994. Complete block of early B cell differentiation and altered patterning of the posterior midbrain in mice lacking Pax5/BSAP. Cell 79:901.[Medline]
  43. Ishida, H., T. Muchamuel, S. Sakaguchi, S. Andrade, S. Menon, M. Howard. 1994. Continuous administration of anti-interleukin 10 antibodies delays onset of autoimmunity in NZB/W F1 mice. J. Exp. Med. 179:305.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
X. Yu, R. Siegel, and R. G. Roeder
Interaction of the B Cell-specific Transcriptional Coactivator OCA-B and Galectin-1 and a Possible Role in Regulating BCR-mediated B Cell Proliferation
J. Biol. Chem., June 2, 2006; 281(22): 15505 - 15516.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
J. R. Podojil, N. W. Kin, and V. M. Sanders
CD86 and {beta}2-Adrenergic Receptor Signaling Pathways, Respectively, Increase Oct-2 and OCA-B Expression and Binding to the 3'-IgH Enhancer in B Cells
J. Biol. Chem., May 28, 2004; 279(22): 23394 - 23404.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
U. Kim, R. Siegel, X. Ren, C. S. Gunther, T. Gaasterland, and R. G. Roeder
Identification of transcription coactivator OCA-B-dependent genes involved in antigen-dependent B cell differentiation by cDNA array analyses
PNAS, July 22, 2003; 100(15): 8868 - 8873.
[Abstract] [Full Text] [PDF]

Home page
 J. Appl. Physiol.Home page
G. F. Elphick, J. Wieseler-Frank, B. N. Greenwood, J. Campisi, and M. Fleshner
B-1 cell (CD5+/CD11b+) numbers and nIgM levels are elevated in physically active vs. sedentary rats
J Appl Physiol, July 1, 2003; 95(1): 199 - 206.
[Abstract] [Full Text] [PDF]

Home page
 J. Immunol.Home page
T. L. Rothstein
Two B-1 or Not To Be One
J. Immunol., May 1, 2002; 168(9): 4257 - 4261.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kim, U.
Right arrow Articles by Roeder, R. G.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kim, U.
Right arrow Articles by Roeder, R. G.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS