Abstract
FcγRIIB1 molecules serve as negative feedback regulator for B cell Ag receptor-elicited activation of B cells; thus, any impaired FcγRIIB1 function may possibly be related to aberrant B cell activation. We earlier found deletion polymorphism in the Fcgr2b promoter region among mouse strains in which systemic autoimmune disease-prone NZB, BXSB, MRL, and autoimmune diabetes-prone nonobese diabetic, but not NZW, BALB/c, and C57BL/6 mice have two identical deletion sites, consisting of 13 and 3 nucleotides. In this study, we established congenic C57BL/6 mice for NZB-type Fcgr2b allele and found that NZB-type allele down-regulates FcγRIIB1 expression levels in germinal center B cells and up-regulates IgG Ab responses. We did luciferase reporter assays to determine whether NZB-type deletion polymorphism affects transcriptional regulation of Fcgr2b gene. Although NZW- and BALB/c-derived segments from position −302 to +585 of Fcgr2b upstream region produced significant levels of luciferase activities, only a limited activity was detected in the NZB-derived sequence. EMSA and Southwestern analysis revealed that defect in transcription activity in the NZB-derived segment is likely due to absence of transactivation by AP-4, which binds to the polymorphic 13 nucleotide deletion site. Our data imply that because of the deficient AP-4 binding, the NZB-type Fcgr2b allele polymorphism results in up-regulation of IgG Ab responses through down-regulation of FcγRIIB1 expression levels in germinal center B cells, and that such polymorphism may possibly form the basis of autoimmune susceptibility in combination with other background contributing genes.
Low-affinity receptors for the Fc portion of IgG, FcγRIIB, and FcγRIII transduce opposing activation/inhibition signals upon ligation by IgG immune complexes (ICs)3 (1, 2, 3, 4). Although FcγRIII exists as an oligomeric complex together with γ- and ζ-chain homo- or heterodimers involved in cell activation through immunoreceptor tyrosine-based activation motif, FcγRIIB is a monomeric receptor with immunoreceptor tyrosine-based inhibition motif in the cytoplasmic region. There are known two membrane-bound isoforms of FcγRIIB, FcγRIIB1, and FcγRIIB2, and these are encoded by the same gene, but show distinct cell-type distribution and function. FcγRIIB1 molecule, which is predominantly expressed on B cells, has complete domains from all exons, and upon cross-linking with B cell Ag receptor (BCR) by IgG ICs, it acts as a negative feedback regulator by inhibiting BCR-elicited activation signal through immunoreceptor tyrosine-based inhibition motif. In contrast, FcγRIIB2 is mainly expressed on macrophages and a domain encoded by the first intracytoplasmic exon is missing, due to alternative splicing of mRNA transcripts. Such a defect allows FcγRIIB2 to internalize IgG ICs for Ag presentation mainly in phagocytosis.
Susceptibility to autoimmune disease is determined by combined effects of alleles at multiple loci. Several lines of evidence suggest that some could be related to intrinsic hyperresponsiveness of B cells to signals for proliferation and maturation (5). B cell responses are controlled by signaling thresholds through the BCR complex (6). Thus, allelic variants associated with functional deficits in negative feedback regulation of BCR signaling would be worth examining for their association with susceptibility to autoimmune diseases. Genetic studies of our own and others provided evidence that one of the susceptibility alleles to systemic lupus erythematosus (SLE) was mapped in the vicinity of Fcgr2b gene on the telomeric region on chromosome 1, in both murine models (7, 8, 9, 10, 11, 12) and human SLE (13). Thus, we focused on allelic polymorphisms of Fcgr2b as a pivotal genetic element for aberrant B cell activation in SLE.
In previous studies, we found that mouse strains can be divided in two groups, based on the presence or absence of allelic polymorphisms in the Fcgr2b promoter region, consisting of 13 and 3 nucleotide deletion sites (14). Intriguingly, spontaneous animal models of autoimmune diseases, i.e., SLE-prone NZB, BXSB, MRL, and diabetes-prone nonobese diabetic (NOD), all shared identical deletion sites in the Fcgr2b promoter region. The polymorphic promoter region is located in the region containing the reported regulatory element for transcription of Fcgr2b gene in B cells (15), and a 13 nucleotide deletion site in the NZB-type Fcgr2b promoter region includes putative transcription factor-binding sequences, i.e., the AP-4-binding site (16) and S box (17). Of note was the finding that the FcγRIIB1 expression in mice having this polymorphism was abnormally down-regulated in B cells, particularly in follicular germinal centers (GCs), areas of class switching and affinity maturation of B cells, while expression in non-GC B cells was practically identical (10, 14). The importance of this finding was supported by genetic studies using (NZB × NZW)F1 × NZW backcross mice, indicating that the NZB-type Fcgr2b allele was significantly linked to hyper-IgG and IgG anti-DNA Abs in quantitative trait loci analysis (10). Therefore, the FcγRIIB promoter polymorphism may possibly predispose to SLE through GC B cells down-regulating FcγRIIB1 expression upon autoantigen stimulations, thus escaping negative signals for IgG production. Nevertheless, there is no direct evidence that the polymorphic regions are indeed contributing to transcriptional regulation of Fcgr2b gene. In the present study, we highlighted this issue and did luciferase reporter assays, together with EMSAs and Southwestern analysis to detect nuclear binding protein, and to confirm the pivotal role of promoter region polymorphisms in transcriptional regulation of the Fcgr2b gene.
Materials and Methods
Mice
As the 129Sv strain of mice was reported to have NZB-type fcgr2b allele (18), C57BL/6 (B6) congenic mice for this allele were generated by introducing Fcgr2b allele from 129Sv with selective back-crossing of (129Sv × B6)F1 to B6 for 10 generations. FcγRII-deficient B6 mice were obtained by back-crossing the originally constructed deficient mice on a hybrid (129Sv × B6) background to B6 for 12 generations (19). All mice used were housed under identical conditions. If necessary, mice were immunized twice at 10 and 13 wk of age by giving an i.p. injection of 100 μg per mouse of keyhole limpet hemocyanin (KLH) in CFA and IFA, respectively.
Genotyping
DNA was extracted from the mouse tail tissue. Genotyping for microsatellite markers and the FcγRIIB promoter region was done using PCR. Microsatellite primers were purchased from Research Genetics (Huntsville, AL). Genotyping for FcγRIIB promoter region was done using 5′ and 3′ primers: 5′-GTAAGTGGTTGTGGGTACCTTTATT-3′ and 5′-CGCAGCTCAGAAGTCATTTGCCTCA-3′. PCRs were run in 20 μl total volume containing 40 ng of genomic DNA. A three temperature PCR protocol (94°C, 65°C, and 72°C) was used for 35 cycles in a GeneAmp 9600 Thermal Cycler (PerkinElmer/Cetus, Norwalk, CT). PCR products were diluted 2-fold with loading buffer consisting of xylene cyanol and bromophenol blue dyes in 50% glycerin and were run on 18% polyacrylamide gels. After electrophoresis, gels were visualized with ethidium bromide staining.
Flow cytometry analysis
For analysis of FcγRIIB1 expression levels on B cells, aliquots of 1 × 106 spleen cells were first stained with PE-conjugated mAb 2.4G2 in PBS containing 1% BSA and 0.1% sodium azide for 30 min at 4°C. After incubation, the cells were washed in PBS with 0.2% BSA and then stained with FITC-labeled B cell-specific anti-B220 (6B2) mAb and biotinylated peanut agglutinin (PNA). After washing, the cells were further stained with streptavidin-conjugated allophycocyanin. For FcγRIIB2 expression levels on peritoneal macrophages, peritoneal exudate cells were double-stained with Alexa 488-conjugated mAb F4/80 and PE-conjugated 2.4G2 in PBS containing 1% BSA and 0.1% sodium azide for 30 min at 4°C. After the cells were washed and fixed in PBS containing 1% formaldehyde, fluorescence intensity was measured using a FACStar flow cytometer and CellQuest software (BD Biosciences, Mountain View, CA).
Measurement of serum Ab levels
Serum levels of IgG anti-KLH Abs were quantified using ELISA plates coated with 5 μg/ml of KLH (Sigma-Aldrich, St. Louis, MO). KLH-binding activities were expressed in units, referring to a standard curve obtained by serial dilutions of a standard serum pool from hyperimmunized NZB mice, containing 1000 U activities/ml.
Serum levels of IgG class autoantibodies to dsDNA were also determined using ELISA plates coated with 5 μg/ml of dsDNA derived from the calf thymus (Sigma-Aldrich). Binding activities were expressed in units, referring to a standard curve obtained by serial dilution of a standard serum pool from (NZB × NZW)F1 mice >8 mo of age, containing 1000 U activities/ml.
Plasmid construction and nucleotide sequence analysis
The 5′-flanking sequences of Fcgr2b gene (nucleotides from position −302 to +76 or position −302 to +585 relative to the transcription start site) were generated by PCR from genomic DNA, using appropriate primers derived from the sequences published by Hogarth et al. (16). The PCR products were cloned into the upstream of the luciferase coding region of the pGL3-Basic vector (Promega, Madison, WI) to yield luciferase constructs carrying each mouse Fcgr2b gene promoter.
For sequencing, we used these PCR products and the dideoxy chain termination method with Taq dye primer cycle sequencing kits (Applied Biosystems, Foster City, CA), according to the manufacturer’s instruction.
Luciferase assays
Bal17 (murine B cell line) and P388D1 (murine macrophage-like cell line) cells were kindly provided by Dr. T. Tsubata (Department of Immunology, Medical Research Institute, Tokyo Medical and Dental University, Tokyo, Japan), and Dr. K. Okumura (Department of Immunology, Juntendo University School of Medicine, Tokyo, Japan), respectively. For stable transfections, cells grown to subconfluence in RPMI 1640 with 10% FBS were washed twice with PBS and resuspended at the density of 1.5 × 107 cells/ml. A total of 700 μl of cells were transfected with 20 μg of the 2.9-kb SalI-SacI fragment derived from each luciferase construct and 1 μg of the 5.7-kb EcoRI fragment derived from pSV2-neo (20), including the neomycin-resistant gene by electroporation using Gene pulser (Bio-Rad, Hercules, CA) (350 V/950 μF for Bal17 cells, 400 V/950 μF for P388D1 cells). Two days after electroporation, medium was replaced with that containing 1 mg/ml (Bal17 cells) or 400 μg/ml (P388D1 cells) of G418 (Calbiochem, La Jolla, CA). After a 14-day selection period, luciferase assays were done. Luciferase activity was measured for a 20-s time course with the use of a luminometer (Lumat LB 9507; Berthold, Wildbad, Germany), and then normalized by gene dosage estimated by Southern blot hybridization using the 1.7-kb NcoI-XbaI fragment from the pGL3-Basic vector as a probe.
Preparation of nuclear protein extracts and EMSA
Nuclear protein extracts were prepared as described by Schreiber et al. (21). The oligonucleotides were chemically synthesized, annealed after being heated to 100°C, and gel purified. The resultant probes (Table I⇓) were labeled with [γ-32P]ATP using T4 polynucleotide kinase. Binding reactions and electrophoresis were done essentially according to Fujihara et al. (22). Nuclear protein extracts (5 μg) with 2 μg of poly(dI-dC) (Amersham Pharmacia Biotech, Piscataway, NJ) and labeled DNA (100,000 cpm) were mixed in 25 μl of binding buffer (10 mM Tris-HCl (pH 7.5), 40 mM NaCl, 1 mM EDTA, 1 mM 2-ME, 4% glycerol, 0.1% Nonidet P-40, 1 mM DTT, 1 mg/ml BSA) then incubated for 30 min on ice after mixing. The DNA-protein complexes were separated on a native 4.5% polyacrylamide gel in 0.25 × Tris-borate EDTA buffer, then the gels were fixed, vacuum dried, and exposed for autoradiography.
Sequences of synthesized oligonucleotide probes for EMSA and Southwestern analysis
Southwestern blotting
A total of 50 μg of nuclear extracts obtained from P388D1 cells were separated on 10% SDS-polyacrylamide gels at 4°C for 120 min (20 mA). The separated proteins were then electrically transferred overnight to a polyvinylidene difluoride (PVDF) membrane at 4°C (30 V) in buffer containing 25 mM Tris, 190 mM glycine, and 20% methanol. After transfer, the membrane was soaked in Z′ buffer (25 mM HEPES-KOH (pH 7.6), 12.5 mM MgCl2, 20% glycerol, 0.1% Nonidet P-40, 0.1 M KCl, 10 μM ZnSO4, 1 mM DTT) with 6 M guanidine-HCl, and was washed five times with Z′ buffer containing graded concentrations of guanidine-HCl to obtain a final concentration of 0.1 M guanidine-HCl. The PVDF membrane was finally washed twice with Z′ buffer without guanidine-HCl, blocked with 3% skim milk in Z′ buffer for 30 min, and preequilibrated in binding buffer (0.25% skim milk in Z′ buffer) for 5 min before addition of a probe. dsDNA oligonucleotide probe, end-labeled with 1 × 107 cpm 32P, was diluted in 2 ml binding buffer, and the PVDF membrane was incubated with probes for 30 min at 4°C with continuous agitation. The membrane was washed three times with Z′ buffer and then three times with 0.3 M KCl in Z′ buffer, each time for 5 min. After air drying, the membrane was autoradiographed.
RT-PCR analysis for AP-4 transcripts
Total RNA was isolated from cell lines Bal17 and P388D1, and T cell-depleted splenic B cells from 2- and 10-mo-old NZB mice using ISOGEN (Nippon Gene, Tokyo, Japan). To obtain a T cell-depleted B cell population, spleen cells were treated with a mixture of the rat mAbs to CD4, CD8, and Thy-1.2 plus rabbit complement at 37°C for 45 min. First-strand cDNA was synthesized using a random hexamer and Moloney murine leukemia virus reverse transcriptase (Invitrogen, Carlsbad, CA). The specific PCR products for AP-4 were amplified using primers (5′-GATGATGCTGGAAGAGCAGGTGCGC-3′ and 5′-CCCTCAATGTGCTGGATGGCCTGCAC-3′) in three-temperature system (94°C, 57°C, and 72°C) for 35 cycles. Amplification of GAPDH served as a reference to normalize data for the total amount of RNA input. Primers for GAPDH were 5′-GGACTGCCCAGAACATCATCCCTGC-3′ and 5′-TTAGTGGGCCCCTGGATGCCAGCTT-3′. The products were electrophoresed in 2% agarose gel and visualized after ethidium bromide staining.
Statistics
Statistical analysis was done using Student’s t test. A value of p < 0.05 was considered to have a statistical significance.
Results
Effects of Fcgr2b allele polymorphisms on FcγRIIB expression levels in immune cells and IgG Ab responses
To provide evidence that the marked down-regulation of FcγRIIB1 expression levels on GC B cells is linked to the NZB-type Fcgr2b promoter polymorphism, but not to other background genetic elements, we established B6 mice congenic for NZB-type Fcgr2b allele. As the 129Sv strain shares the NZB-type Fcgr2b allele (18), congenic mice were generated by selective back-crossing of (129Sv × B6)F1 to B6 for 10 generations. The 129Sv interval on chromosome 1 introduced into B6 mice extended from ∼88 (D1 Mit15)-92 (Fcgr2b) cM, which were verified to be negative for the 129Sv allele of Fcer1g, Cd48, Ly9, Fcer1a, Crp, Sap, Ifi202, Ifi203, Ifi204, Cr2, and Adprp (Fig. 1⇓). These congenic B6 mice, together with control B6 and B6.FcγRIIB−/− mice at 10 and 13 wk of age, were immunized twice with KLH. One week after the second immunization, expression levels of FcγRIIB1 on B220-positive splenic B cells were examined, using 2.4G2 mAb (FcγRIIB/III; Ref. 23). As shown in Fig. 2⇓A, both PNAlow non-GC B cells and PNAhigh GC B cells from B6.FcγRIIB−/− mice were negative for 2.4G2 staining; thus, fluorescence intensity of 2.4G2 on B cells represents the expression of FcγRIIB1, but not FcγRIII.
Gene segment introduced into the B6 strain on chromosome 1 in B6 congenic strain with NZB-type Fcgr2b allele. Filled bar was introduced from the 129Sv strain by selective backcrossing for 10 generations. Genotypes were determined by microsatellite markers and Fcgr2b allele polymorphism.
A, Comparisons of representative 2.4G2 staining levels between splenic PNAlow non-GC B cells and PNAhigh GC B cells from KLH-immunized B6, congenic B6 with NZB-type Fcgr2b allele, and B6.FcγRIIB−/− mice. Spleen cells were stained with PE-labeled 2.4G2 (anti-FcγRIIB/III) mAb, FITC-labeled B cell-specific anti-B220 mAb, and biotinylated PNA, followed by streptavidin-allophycocyanin. Because B cells from B6.FcγRIIB−/− mice were negative for 2.4G2 staining, 2.4G2 expression on B cells reflects the level of FcγRIIB1 isoform. The filled histogram indicates background staining. B, Comparisons of representative 2.4G2 staining levels on F4/80 positive peritoneal macrophages from nonimmunized and KLH-immunized B6, congenic B6 with NZB-type Fcgr2b allele, and B6.FcγRIIB−/− mice. The filled histogram indicates the background staining.
Comparisons of the results between congenic B6 and control B6 mice showed that although FcγRIIB1 expression levels of GC B cells in B6 mice were up-regulated, those in congenic B6 mice were markedly down-regulated, as compared with levels on non-GC B cells, which were virtually identical between the two strains (Fig. 2⇑A). Based on the mean channel of 2.4G2 fluorescence intensity, the FcγRIIB1 expression level on GC B cells was ∼10 times lower in congenic B6 than in control B6 mice. Fig. 3⇓ compares dual-color immunofluorescent staining patterns of splenic tissue sections with 2.4G2 mAb and PNA, representing marked differences in the staining intensity of 2.4G2 on GC B cells between control B6 and congenic B6 mice, high in the former and low in the latter. As for follicular dendritic cells (FDCs), intensities of 2.4G2 were almost identical between B6 and congenic B6 mice and negative in B6.FcγRIIB−/− mice. This indicates that FDCs in these mice are positive for either FcγRIIB1 or FcγRIIB2, or both, and negative for FcγRIII, at least at the level of tissue immunofluorescence.
Representative dual-color immunofluorescence staining of spleen sections from KLH-immunized B6, congenic B6 with NZB-type Fcgr2b allele, and B6.FcγRIIB−/− mice. Green and red colors depict staining for 2.4G2 mAb and PNA, respectively. FDCs were positive for 2.4G2 in both B6 and congenic B6, but not in B6.FcγRIIB−/− mice. Note that the level of FcγRIIB1 on GC B cells in a congenic B6 mouse was markedly down-regulated, compared with findings in a B6 mouse.
Fig. 2⇑B compares 2.4G2 staining levels on F4/80-positive peritoneal macrophages among three strains of mice. As there was no remarkable difference in 2.4G2 staining intensities on macrophages from nonimmunized B6, congenic B6, and B6.FcγRIIB−/−, the staining appeared to mainly reflect FcγRIII expression, although small amounts of FcγRIIB2 molecules would also be expressed on macrophages from B6 and congenic B6 mice. Levels of FcγRIIB2 in congenic B6 macrophages were slightly lower than seen in B6 mice. This difference was clearer in macrophages from immunized mice.
Fig. 4⇓A compares serum IgG anti-KLH Ab levels among KLH-immunized B6, congenic B6, and B6.FcγRIIB−/− mice. As compared with the levels in control B6 mice, those in congenic B6 mice showed significantly higher anti-KLH Ab responses. The levels in B6.FcγRIIB−/− mice were highest. Similar results were obtained with serum levels of spontaneously produced IgG anti-DNA Abs in these three strains of mice at 4 mo of age (Fig. 4⇓B).
A, Comparisons of serum levels of IgG anti-KLH Abs among KLH-immunized B6, B6 congenic with NZB-type Fcgr2b allele, and B6.FcγRIIB−/− mice at 14 wk of age. B, Comparisons of serum levels of spontaneously produced IgG anti-DNA Abs among B6, B6 congenic with NZB-type Fcgr2b allele, and B6.FcγRIIB−/− mice at 16 wk of age. Each horizontal bar represents the mean ± SE. ∗, p < 0.05; **, p < 0.001.
Transcriptional regulation by polymorphic Fcgr2b promoter region
To determine whether NZB-type deletion polymorphisms of the Fcgr2b promoter region contribute to transcriptional regulation of the Fcgr2b gene, luciferase reporter assay was done using an NZW-derived Fcgr2b upstream gene fragment connected in front of the luciferase reporter gene of pGL3-Basic vector (Fig. 5⇓A). For stable transfections, B cell line (Bal17) and macrophage-like cell line (P388D1) were transfected with the 2.9-kb SalI-SacI fragment containing the Fcgr2b upstream gene fragment and luciferase reporter gene. As it was reported that the Fcgr2b gene fragment from position −229 to +70 was fully active in the A20/2J B cell line in a short-term transfection assay (15), luciferase activity was first examined with a construct containing the Fcgr2b upstream fragment from position −302 to +76. However, this fragment was not long enough, and a longer one from position −302 to +585 drove transcription activities in stable transfectants of both Bal17 B cell and P388D1 macrophage-like cell lines.
Luciferase reporter assay for transcription activity by the Fcgr2b upstream region gene. Bal17 or P388D1 cell line was transfected with the NZW-derived Fcgr2b upstream fragment from −302 to +76 (+1: transcription starting site) or from −302 to +585 connected with luciferase reporter gene (A), or with the fragment from −302 to +585 derived from NZW, BALB/c, or NZB connected with luciferase reporter gene (B). After 14 days of selection of stable transfectants, cells were lysed and assayed for luciferase enzyme activity. Luciferase activity was normalized by gene dosage estimated by Southern blot hybridization, using the 1.7-kb NcoI-XbaI fragment from the pGL3-Basic vector as a probe. Each bar represents the mean ± SE of three separate experiments in duplicate.
As there are no data available on strain differences in nucleotide sequences of the Fcgr2b gene fragment from position −302 to +585, complete nucleotide sequences were examined in several mouse strains (Fig. 6⇓). In addition to the previously detected 13 and 3 nucleotide deletion sites in the Fcgr2b promoter region in NZB and 129Sv strains, these strains have an insertion of 16 nucleotides in the first intron in close vicinity to exon 2, as compared with BALB/c, B6, and NZW strains. Only a limited number of single nucleotide polymorphism was detected among mouse strains.
Comparison of Fcgr2b upstream nucleotide sequence among mouse strains. A dash indicates no difference from the BALB/c-derived sequence, and asterisks indicate deleted nucleotides. Nucleotides are numbered relative to the transcription starting site (+1). Transcribed sequences are given in capital letters and underlined. Putative sequences for transcription factor binding sites are boxed.
We then tested strain differences in luciferase activity, using constructs containing Fcgr2b gene fragments from position −302 to +585. The result showed that the luciferase reporter gene connected with NZW- or BALB/c- derived sequence produced significant levels of luciferase activities in both Bal17 and P388D1 cell lines; however, only a limited activity was detected in the NZB-derived sequence (Fig. 5⇑B). These findings indicate that polymorphic regions in the Fcgr2b upstream fragment appear to contribute to transcriptional regulation of Fcgr2b gene in both B cells and macrophages.
Absence of transcription factor binding to the NZB-type Fcgr2b promoter region
To demonstrate if a decreased transcription activity in the NZB-type Fcgr2b upstream gene is due to a defect in binding interactions with nuclear proteins, EMSA was done. Oligonucleotide probes with sequences of the polymorphic promoter region and first intron were synthesized to identify which regions are involved in nuclear factor binding (Table I⇑). First, the promoter probe derived from NZW sequence (NZW(P)) or the promoter probe from NZB sequence with 13 and 3 nucleotide deletion sites (NZB(P)) was radiolabeled, mixed with extracted nuclear proteins from P388D1 cell lines, and electrophoresed on acrylamide gels. In addition to the free probe, the specific band reflecting probe-nuclear factor complex was detected using a radiolabeled NZW(P) probe, in which a cold NZW(P), but not a NZB(P), probe inhibited formation of this specific band. However, when radiolabeled NZB(P) probe was used, this specific band was not detected (Fig. 7⇓A). Thus, the nuclear factor-binding site appeared to be deleted in NZB(P) probe. The same result was obtained by using nuclear proteins extracted from B cell line Bal17 (data not shown).
Detection of specific nuclear protein binding by EMSA. A, Radiolabeled promoter probe with NZW sequence (NZW(P)) or NZB sequence (NZB(P)) was mixed with a nuclear extract and electrophoresed. For competitive inhibition assay, we added 200-fold molar excess of unlabeled NZB(P) or NZW(P) probe competitors. The arrow and the arrowhead indicate the specific band composed of probe and nuclear proteins and free probes, respectively. B, The same experiment was done using intron probes with the NZW sequence (NZW(I)) and the NZB sequence (NZB(I)). There was no specific band for the NZB-type insertion site. Each probe used is shown in Table I⇑.
The same experiment was done, using the first intron probe derived from the NZW sequence (NZW(I)) or from the NZB sequence, including the 16 nucleotide insertion site (NZB(I)). As shown in Fig. 7⇑B, the same band pattern was detected in NZW(I) and NZB(I) probes, and was inhibited by both cold NZB(I) and NZW(I) competitors. Thus, this band formation seems to be derived from common sequence (probably corresponding to putative GAL4 binding sequence as shown in Table I⇑) involved in NZB(I) and NZW(I), and is not specific for NZB-type insertion site.
As the 13-nucleotide deletion site in the NZB promoter region includes the putative AP-4 binding site (16), we then asked if the nuclear factor detected in the above EMSA is AP-4. As shown in Fig. 8⇓A, specific bands with the same molecular mass were observed using a synthesized radiolabeled probe corresponding to AP-4 binding consensus sequence (AP-4) (Table I⇑; Ref. 24) and the NZW(P) probe in EMSA, and these bands were inhibited by both AP-4 and NZW(P) cold competitors. To further ask if the involved nuclear factor is indeed AP-4, Southwestern assay was done (Fig. 8⇓B). When the nuclear extract was electrophoresed and hybridized with a radiolabeled AP-4 probe (AP-4), a specific band of ∼48 kDa was detected. A band of the same size was detected using a radiolabeled NZW(P), but not a NZB(P), probe. Taken collectively, it seems likely that AP-4 binds to the polymorphic Fcgr2 promoter region corresponding to the 13-nucleotide deletion site in NZB-type Fcgr2b allele and promotes transcription activity. Fig. 9⇓ shows RT-PCR analysis for detecting AP-4 transcripts, demonstrating that the transcripts were detected in both Bal17 and P388D1 cell lines and splenic B cells from NZB mice. The levels were much higher in P388D1 than in Bal17. Splenic B cells from aged NZB mice showed higher levels of AP-4 transcripts than those from young NZB mice, in parallel with numerous spontaneous GC formation in aged NZB mice (10).
AP-4 is a candidate transcription factor binding to the upstream region of Fcgr2b gene from NZW, but not from NZB, mice. A, EMSA was done using mixtures of nuclear extract with radiolabeled AP-4 consensus oligonucleotide (AP-4) or the NZW(P) probe. For competitive inhibition assay, we added 200-fold molar excess of unlabeled probe competitors. The arrow indicates the specific band composed of probe and putative AP-4 transcription factor, which was inhibited by both cold AP-4 and NZW(P) probes. B, Southwestern analysis was done using the AP-4, NZW(P), and NZB(P) probes described in Table I⇑. Nuclear extract was electrophoresed and hybridized with a radiolabeled probe. Note that a band with the same molecular mass was detected using radiolabeled AP-4 and NZW(P), but not NZB(P), probes.
A representative result of RT-PCR amplification of AP-4 messages from 1) P388D1; 2) Bal17; 3) splenic B cells from 2-mo-old NZB; and 4) splenic B cells from 10-mo-old NZB mice. Amplification of GAPDH served to check RNA amounts. PCR products were electrophoresed on 2% agarose gels and visualized after ethidium bromide staining. Numbers of amplified bases are 335 and 210 bp for AP-4 and GAPDH, respectively.
Discussion
There are Fcgr2b promoter region polymorphisms among mouse strains, and two deletion sites of 13 and 3 nucleotides are shared with SLE-prone NZB, BXSB, MRL, nonobese diabetes-prone NOD, and nonautoimmune 129Sv strains of mice (14, 18, 25). Previous analysis documented that there are several putative transcription factor-binding sites, such as Sp1 site, glucocorticoid response element, AP-4 site, and S box in Fcgr2b upstream region (15, 16, 17), and that the 13-nucleotide deletion site is interrupted by the AP-4 binding site and the S box (14, 18). In the present studies, we characterized transcription activity of the Fcgr2b upstream region gene and found that the polymorphic promoter region is likely to be transactivated by AP-4, a member of the basic helix-loop-helix family of transcription factors. The identity of AP-4 as a transactivator of Fcgr2b gene needs further studies; however, the molecular mass of ∼48 kDa determined by Southwestern analysis strongly suggests that AP-4 is the possible transcription factor (24). In addition to deletion polymorphisms in the promoter region, our sequence analysis identified an insertion polymorphism of 16 nucleotides in the first intron in mice with the NZB-type Fcgr2b allele. This insertion site did not have consensus sequences for known transcription factors, and indeed did not show any binding activity to nuclear proteins in the present studies.
Bonnerot et al. (15) used A20/2J B cell line and MMC-1 mast cell line in chloramphenicolacetyl transferase assays, and found that different promoter fragments determine transcription of the Fcgr2b gene in these two cell lines. The fragment of positions from −229 to +70 relative to the transcription start site was shown to be able to drive the transcription of Fcgr2b gene in B cell line, but not in mast cell line. In our studies, the fragment of positions from −302 to +76 was not long enough and a longer one from −302 to +585 did gain the transcription activity in both Bal17 B cell and P388D1 macrophage-like cell lines. The reason for this discrepancy is unclear, but may relate to differences in assay systems in which the short-term transcription assay used by Bonnerot et al. (15) and a stable transfection assay by our own, and also in cell lines used. In any instance, the polymorphic promoter region among mouse strains is located in the fragment containing the regulatory element, and appears to function for transcription of the Fcgr2b gene in both B cells and macrophages.
Establishment of the B6 strain congenic for the NZB-type Fcgr2b allele made it feasible to examine the in vivo effect of the Fcgr2b allele polymorphism in the same B6 genetic background. For this purpose, we took advantage of the B6 congenic strain carrying the 129Sv-derived Fcgr2b allele, since the upstream Fcgr2b region gene sequence was almost identical between NZB and 129Sv, except for one single nucleotide polymorphism in the second intron (Fig. 6⇑). When immunized with KLH, FcγRIIB1 expression levels on non-GC B cells were high and identical between both strains of mice, yet those on activated GC B cells were markedly down-regulated in congenic B6 mice and up-regulated in control B6 mice, as compared with findings in non-GC B cells. This down-regulation of FcγRIIB1 expression levels was associated with the up-regulation of IgG anti-KLH Ab responses. FcγRIIB2 expression levels on macrophages from congenic B6 mice were also low, as compared with levels in control B6 mice. However, the effect of FcγRIIB2 expression levels on macrophage function is controversial. There is a report of the possibility that impairment of FcγRIIB2 expression on macrophages from mice with the NZB-type Fcgr2b allele contributes to elevated serum levels of IgG through the diminished catabolism of ICs (25). In contrast, another report showed an increase in the phagocytosis of opsonized SRBC by macrophages from mice with the NZB-type Fcgr2b allele (18). Thus, the exact role of FcγRIIB2 molecules in macrophage function remains to be determined. As far as the role in B cells is concerned, we propose that the NZB-type polymorphic Fcgr2b promoter region is related to enhanced IgG responses through down-regulation of FcγRIIB1 expression levels on activated B cells in GCs, where affinity maturation and Ig class switch occur, thus escaping negative signals for BCR-elicited activation. In vivo studies also showed that the polymorphic region of Fcgr2b allele does not affect FcγRIIB1 expression levels on resting non-GC B cells, which suggests that an unidentified different transcriptional regulation mechanism is operative in resting stage B cells.
We previously mapped an allele susceptible for hyper-IgG, including IgG anti-DNA Abs, Hig-1, to close vicinity of the Fcgr2b gene on NZB chromosome 1, and proposed that the NZB-type Fcgr2b promoter polymorphism forms the basis of one aspect of SLE susceptibility in NZB and (NZB × NZW) F1 mice (5, 10). This is consistent with the present finding that serum levels of IgG anti-DNA Abs were significantly higher in Fcgr2b-congenic B6 than in control B6 mice, albeit the levels being significantly lower than in B6.FcγRIIB−/− mice. At present, it is not clear if congenic B6 mice with the NZB-type Fcgr2b allele develop SLE, and the study is under investigation in our laboratory. Significance of the NZB-type Fcgr2b promoter polymorphism in the pathogenesis of autoantibody-mediated autoimmune diseases may require further investigations. 129Sv mice neither produce IgG anti-DNA Abs nor develop the autoimmune disease, despite carrying the NZB-type Fcgr2b allele. Contribution of the NZB-type Fcgr2b allele to the pathogenesis of NOD diabetes is unknown, as the pathogenesis is thought to be mainly mediated by T cells, although there are reports showing a possible involvement of Abs in diabetes (26, 27). Bolland and Ravetch (28) reported that B6.FcγRIIB−/− mice with aging spontaneously develop SLE-like autoimmune disease; however, the disease does not occur in BALB.FcγRIIB−/− mice. All these findings are probably related to the premise that autoimmune disease is a complex multigenic disease in which susceptibility arises from the combined effects of multiple contributing genes, and different autoimmune manifestations are controlled separately by different sets of susceptibility alleles (29, 30, 31). Function of the Fcgr2b gene appears to be a kind of immune regulatory gene, and it is feasible to speculate that the NZB-type Fcgr2b promoter polymorphism could be selected evolutionally for natural defense against pathogens. Such polymorphism may in turn form the basis of autoantibody-mediated autoimmune diseases as one of multiple susceptibility alleles.
In earlier genetic studies, it was shown that the telomeric chromosome 1 interval from not only NZB and BXSB, but also NZW and SWR, strains contributes to IgG autoantibody production (7, 8, 9, 10, 11, 12), and that several potential candidate genes, such as Cr2, Adprp, and Ifi202 may be involved (32, 33, 34). In our congenic B6 mice with the NZB-type Fcgr2b allele, the interval introduced into B6 mice does not include these genes, thus indicating that, at least in our congenic B6 mice, the observed enhanced IgG responses are probably due to the NZB-type Fcgr2b allele and not other suggested candidate genes. Studies by Morel et al. (33) using congenic B6 mice carrying telomeric chromosome 1 segments derived from NZM2410 mice, a recombinant inbred strain generated from crosses of NZB and NZW strains by Rudofsky et al. (35), yielded evidence that three loci, Sle 1a, Sle 1b, and Sle 1c, but not the interval containing the Fcgr2b gene, are independently associated with loss of tolerance to chromatin. However, it is to be noted that the chromosomal region containing Fcgr2b gene in NZM mice is derived from the NZW, but not the NZB strain. It is highly feasible that there are several other important candidate genes related to the susceptibility to SLE in the vicinity of the Fcgr2b gene on chromosome 1, and intensive congenic dissection will be useful to search for other susceptibility allele located in this region of chromosome 1.
Acknowledgments
We thank Dr. Robert Tjian for kind help in Ap-4 experiment and M. Ohara for language assistance.
Footnotes
-
↵1 This work was supported in part by grants-in-aid for Scientific Research and High Technology Center Grant from the Ministry of Education, Science, Technology, Sports and Culture, Japan.
-
↵2 Address correspondence and reprint requests to Dr. Sachiko Hirose, Department of Pathology, Juntendo University School of Medicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113-8421, Japan. E-mail address: sacchi{at}med.juntendo.ac.jp
-
↵3 Abbreviations used in this paper: IC, immune complex; BCR, B cell Ag receptor; FDC, follicular dendritic cell; GC, germinal center; KLH, keyhole limpet hemocyanin; PNA, peanut agglutinin; SLE, systemic lupus erythematosus; NOD, nonobese diabetic; PVDF, polyvinylidene difluoride.
- Received May 9, 2002.
- Accepted August 12, 2002.
- Copyright © 2002 by The American Association of Immunologists
References
In this issue
