Targeted Deletion of the IgA Constant Region in Mice Leads to IgA Deficiency with Alterations in Expression of Other Ig Isotypes

Targeting of the IgA locus in ES cells and generation of IgA-deficient mice

A plasmid (pm5′Iα) containing a 3.5-kb XbaI/AvrII DNA fragment immediately 5′ of the Iα exon was previously subcloned from the E-6 phage 7 into pKS⁻ (Stratagene, La Jolla, CA) 8 , and a plasmid (pm3′Cα(RI-XbaI)) containing a 2.9 kb EcoRI/XbaI DNA fragment from the 3′ end of the IgA constant region was subcloned from E-6 into pKS⁻. A targeting vector was constructed with the 3.5-kb XbaI/AvrII DNA fragment as the upstream homologous fragment and the 2.9-kb EcoRI/XbaI DNA fragment as the downstream homologous region (Fig. 1⇓). For positive selection a neomycin resistance cassette containing a PGK promoter and bovine growth hormone polyadenylation signal 9 was inserted between these two homologous regions in the same transcriptional orientation as the endogenous locus, while for negative selection a herpes simplex thymidine kinase (HSV-tk) gene 10 was inserted 5′ of the upstream homologous region.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

Generation of IgA-deficient mice by gene targeting. A, Strategy for deletion of the IgA switch and constant regions in ES cells. Maps of the germline Igh locus, targeting vector, Iα exon-deleted DNA, and IgA switch and constant region targeted DNA are shown. Restriction enzyme sites, predicted restriction fragment lengths, and probes used in Southern analysis are also shown. E, EcoRI; K, KpnI; S, SacI; Xh, XhoI. Homologous recombination between the targeting vector and genomic DNA from the Iα.2 ES cell clone (8) results in replacement of the HPRT minigene by the neo^r gene. As a consequence, the entire Iα exon, IgA switch region, and 5′ half of the IgA constant region are deleted. B, Southern analysis of KpnI/XhoI-digested DNA from targeted ES cell clones. A nylon filter containing KpnI/XhoI-digested DNA was hybridized with a probe from the 3′ region of Cε (Cε probe). Wild-type DNA gives an 8.8-kb fragment, while DNA from an Iα exon-deficient mouse (derived from Iα.2 ES cells) (8) gives a 10.7-kb fragment. Iα.2 ES cell clones in which the HPRT minigene-containing allele has been targeted contain a 9.6-kb fragment. C, Southern analysis of F₂ offspring derived from mating male chimeric mice with C57BL/6 female mice. Tail DNA was digested with KpnI/XhoI, electrophoresed, blotted to nylon, and analyzed with the Cε probe. One allele in the F₂ offspring is a wild-type allele derived from the C57BL/6 mother (Igh^b allotype; 6.0-kb fragment). Another allele, inherited from the chimeric male, is the wild-type 129/Sv allele (Igh^a allotype; 8.8-kb fragment). The last allele is the targeted allele in which the HPRT minigene has been replaced by the neo^r gene, with deletion of the Iα exon and the IgA switch and constant regions (9.6-kb fragment). Mice homozygous for deletion at the IgA locus demonstrate a single 9.6-kb targeted fragment recognized by the Cε probe. Lane 11 should be ++a/a (and not ++a/b).

Correctly targeted ES cells were injected into C57BL/6 blastocysts followed by implantation into foster mothers as previously described 11 . Chimeric males with the greatest degree of chimerism, assessed by the amount of agouti coat color, were bred to C57BL/6 or 129/Sv female mice. Black agouti offspring from these matings were analyzed for transmission of the targeted allele by Southern hybridization of tail DNA. Male and female heterozygous mice were bred to obtain mice homozygous for the targeted deletion.

Collection of serum and nasal, pulmonary, and intestinal secretions

Blood was collected from the axillary plexus of anesthetized mice and allowed to clot at room temperature, and serum was obtained following centrifugation and stored at −70°C until assayed. Nasal secretions were collected by instillation of 0.5 ml of PBS through a pipette tip inserted into the exposed trachea and directed toward the nasal cavity. Fluid was collected as it exited the nose. Pulmonary secretions were collected by inserting a pipette tip into the exposed trachea, ligating the trachea proximal to this site, and lavaging the lungs with 0.5 ml of PBS. Nasal and pulmonary secretions were tested for the presence of contaminating blood using Chemstrips (Boehringer Mannheim, Indianapolis, IN).

Intestinal secretions were collected from the small intestine using a modification of a previously described method 12 . Briefly, small intestines (from pylorus to cecum) were isolated and carefully filled for 10 min at room temperature with 2 ml of PBS containing a protease inhibitor solution consisting of 0.1 mg/ml soybean trypsin inhibitor (Sigma, St. Louis, MO), 50 mM EDTA, and 1 mM PMSF (Boehringer Mannheim). The intestinal content was then carefully transferred to a test tube, vigorously vortexed, and centrifuged for 10 min at 650 × g at 4°C. The supernatant was transferred to a microfuge tube, and PMSF was added to a final concentration of 1 mM. The solution was mixed by vortexing and was centrifuged at 13,000 × g for 20 min at 4°C. The resulting supernatant was mixed with PMSF and sodium azide to final concentrations of 1 mM and 0.01%, respectively, and was incubated for 15 min at 4°C. Thereafter, 50 ml of FCS was added per 1 ml of secretion, and the solution was centrifuged for an additional 20 min at 13,000 × g at 4°C. Supernatants were removed, and secretions were stored at −70°C until analysis.

Isolation of spleen and Peyer’s patch lymphocytes and in vitro culture of B cells

Spleen and Peyer’s patch non-T cells were obtained as previously described 13 . Briefly, a single cell suspension of spleen or Peyer’s patch cells was prepared, and RBCs were removed by hypotonic lysis. T cells were depleted by incubation with a mixture of anti-Thy 1.2 (New England Nuclear, Boston, MA; final concentration, 1/1000), anti-Lyt 2.2, and anti-L3T4 mAb (Accurate Chemical Co., Westbury, NY; final concentration for both, 1/1000) in cytotoxicity medium (RPMI 1640 with 0.2% BSA) followed by complement lysis (Low Tox-M rabbit complement, Accurate Chemical Co.).

Production of IgA in vitro was assessed as previously described 8 . Briefly, B cells (0.5–2 × 10⁶/ml) were cultured in RPMI 1640 (BioWhittaker, Walkersville, MD) supplemented with 10% FCS (BioWhittaker), 15 mM HEPES (Life Technologies, Grand Island, NY), 5% NCTC 109 medium (BioWhittaker), 5 × 10⁻⁵ M 2-ME (Sigma, St. Louis, MO), 2 mM glutamine (Life Technologies), and 100 U/ml penicillin/100 μg/ml streptomycin (Life Technologies) and incubated at 37°C and 6% CO₂ in a humidified atmosphere in 96-well flat-bottom plates or 25- or 75-cm² flasks (Corning, Corning, NY). LPS was added at the initiation of culture, while cytokines were added on day 1 of culture. Escherichia coli LPS (Sigma, 0127:B8) was used at a final concentration of 10 μg/ml, rIL-2 at 100 U/ml (PharMingen, San Diego, CA), rIL-5 at 100 U/ml 14 , and TGF-β (R&D Systems, Minneapolis, MN) at 1 ng/ml.

Measurement of Ig isotypes and Abs to hemagglutinin

Isotype-specific sandwich ELISAs were performed as previously described 8 using an mAb 7, 8, 9, 10, 11, 12, 13, 14 previously shown to be specific for mouse IgA 15 . Serum and secretory Ig levels are presented as the mean ± SD from analysis of three to six mice per group.

In vitro Ig production was measured in culture supernatants after 7 days of incubation. Cultures for in vitro Ig secretion by both spleen and Peyer’s patch B cells, for each condition analyzed were performed in triplicate. Two separate experiments were performed.

To determine the ability of IgA-deficient mice to develop serum and secretory Ab responses following immunization, mice were immunized with killed whole influenza virus (B-Yamagata) and cholera toxin (75 μg of hemagglutinin/10 μg of cholera toxin orally and 25 μg of hemagglutinin/1 μg of cholera toxin intranasally) on days 0, 11, and 20. Mice were sacrificed on day 33, and serum, nasal secretions and intestinal secretions were collected. Abs against B-Yamagata were measured by standard ELISA assays using plates coated with whole virus (2 μg/ml) in carbonate buffer. Abs of defined isotypes were detected using alkaline phosphatase-conjugated isotype-specific Abs (Southern Biotechnology Associates, Birmingham, AL). End-point titers were calculated based on the highest dilution giving an OD reading >2 SD above the same dilution from an unimmunized control animal.

Flow cytometric analysis of spleen and Peyer’s patch lymphocytes

Flow cytometric analysis was performed on Peyer’s patch and spleen lymphocytes as previously described 8 using a Coulter Profile II (Coulter, Hialeah, FL). Cells were stained with phycoerythrin-labeled goat anti-mouse IgM (Southern Biotechnology Associates) and fluorescein-labeled anti-mouse IgA 7, 8, 9, 10, 11, 12, 13, 14 . Forward and side scatter gates were set to include live lymphocytes, and 10,000 events were collected. Results shown are from one of two experiments that gave similar results.

Histopathology and immunohistochemistry

Portions of thymus, spleen, mesenteric lymph nodes, jejunum, and terminal ileum were placed in formalin. In addition, portions of jejunum and distal ileum were snap-frozen in Cryoform (International Equipment Co., Needham, MA) and liquid nitrogen. For histopathology, sections were routinely stained with hematoxylin and eosin, then analyzed by light microscopy. For immunohistochemistry, cryostat sections, cut 6 μm thick, were blocked with hydrogen peroxide, sodium azide, and BSA. They were then covered with FITC-labeled rat monoclonal anti-mouse IgA (Accurate Chemical Co.) at a 1/50 dilution in PBS at 0–4°C overnight. After rinsing in PBS, a horseradish peroxidase-conjugated secondary rabbit Ab against FITC (Dako, Carpinteria, CA) at a dilution of 1/50 was applied for 2 h at 21°C. After rinsing, color was developed with diaminobenzidine-nickel chloride as the substrate, and tissues were counterstained with Nuclear Fast Red.

RNA amplification by PCR for IFN-γ and IL-4

Total cellular RNA was extracted from fresh mouse lymphocytes (2–10 × 10⁶) or those stimulated or unstimulated with influenza virus for 3 or 7 days or with PHA (2 μg/ml) for 3 days using the RNeasy Mini Kit (Qiagen, Chatsworth, CA). First-strand cDNA preparation and PCR amplification were performed according to the established protocol (Perkin-Elmer/Cetus, Emeryville, CA) 16 . Briefly, cDNA synthesis by RT was performed using an RT kit from Perkin-Elmer/Cetus. The RT mixture was prepared by adding 4 μl of 25 mM MgCl₂, 2 μl of 10× PCR buffer II, 8 μl of 40 mM dNTP, 0.2 μl of RNase inhibitor, 1 μl of Moloney leukemia virus reverse transcriptase, 1 μl of random hexamers, and 1 μl of sample RNA (200 μg/ml) and was brought to 20 μl with RNase-free water. The mixture was overlayed with 50–100 μl of mineral oil (Sigma) and was incubated at room temperature for 10–20 min and in a Robocycler (PCR machine, Stratagene, La Jolla, CA) at 42°C for 15 min, at 99°C for 5 min, and at 6°C for 5 min. A PCR master mix for 40 PCR reactions (each 25 μl) was prepared by adding 100 μl of 10× PCR buffer, 80 μl of 40 mM dNTP, 2 μl of downstream and upstream primers (0.4 μM/ml of murine IFN-γ and IL-4 amplimer sets (Clontech, Palo Alto, CA)), 8 μl of AmpliTaq polymerase, and 684 μl of distilled water (PCR Kits, Perkin-Elmer/Cetus). A 22-μl aliquot of master mix was added to each of 0.5-ml microcentrifuge tube (10 tubes/4 groups for each primer set) followed by 2 μl of RT product. One microliter of 10-fold serially diluted IFN-γ and IL-4 mimics template (Clontech, Palo Alto, CA) was added to each tube in each group and was overlayed with 50 μl of mineral oil. The PCR reaction was performed in a thermal cycler (Robocycler, Stratagene) at 94°C for 2 min (one cycle) followed by 94°C for 1 min and 65°C for 1 min (30 cycles) and an extension at 65°C for 10 min (one cycle). The PCR products were analyzed using 1% agarose gel electrophoresis. The concentration of each cytokine mRNA was determined from overlapping bands of the mimics.

ELISA for IFN-γ and IL-4

Splenic lymphocytes were cultured at a concentration of 10⁶ cell/ml in RPMI 1640 supplemented with 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin (JRH Biosciences, Lenexa, KS), 5 × 10⁻⁵ M 2-ME, 10 mM HEPES buffer, and 10% heat-inactivated FBS with medium alone or PHA (2.5 μg/ml) in duplicate in a final volume of 2.0 ml in 12-well plates. Supernatants from day 3 mitogen cultures were harvested and tested for the presence of IFN-γ and IL-4 using a modification of the sandwich ELISA method 17 . Briefly, 96-well vinyl plastic plate (Nunc-Immuno Plate MaxiSorp, Nunc, Naperville, IL) wells were coated with 100 μl of solution (Pierce Coating Buffer, Pierce, Rockford, IL) containing 0.5–2 μg/ml of capture monoclonal rat anti-murine IFN-γ or IL-4 Ab (PharMingen, San Diego, CA). The plates were incubated overnight at 4°C and washed four times with Tris-NaCl buffer containing 0.05% Tween-20 (Tris-NaCl-Tween) using an automatic Titertek Microplate Washer (Titertek, ICN Biomedicals, Costa Mesa, CA). After blocking for 1 h with SuperBlock buffer in Tris-buffered saline (Pierce) and washing four times, serial dilutions of recombinant murine IFN-γ or IL-4 standard and undiluted test samples were added to the wells in quadruplicate, and the plates were incubated at 37°C for 1 h. After four washes, 100 μl of biotinylated rat anti-murine IFN-γ or IL-4 Ab (0.0–2 μg/ml in SuperBlock) was added and incubated at 37°C for 1 h. The plates were washed, and 100 μl of streptavidin alkaline phosphatase (Pierce) was added to each well, incubated at 37°C for 1 h, and then washed four times. Substrate (1–2 tablets of ρ-nitrophenol phosphate/10 ml of diethanolamine buffer, pH 9.8) was then added, and color was allowed to develop for 10–120 min. Absorbance in each well was read at a wavelength of 405 nm using a Molecular Devices automatic microplate reader (Menlo Park, CA). The data were collected using SOFTmax data reduction software (Molecular Devices). The murine recombinant IFN-γ and IL-4 were used to generate standard curves. The amount of cytokine in the test samples was extrapolated from the standard curves and was expressed as picograms per milliliter of cytokine.

Lymphoproliferative response assay

Splenic and mesenteric lymph node lymphocytes were cultured at a concentration of 10⁵ cell/microtiter well in RPMI 1640 supplemented with 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin (JRH Biosciences), 5 × 10⁻⁵ M 2-ME, 10 mM HEPES buffer, and 10% heat-inactivated FBS with medium alone, LPS (10 μg/ml), PHA (2.5 μg/ml), or Staphylococcus aureus lysate (SPL; Delmont Laboratories, Swarthmore, PA) in quadruplicate in a final volume of 0.2 ml. After 46 h, cells were pulsed with 0.5 μCi of [³H]TdR (6.7 mCi/ml; ICN Radiochemicals, Irvine, CA) for 18–24 h and harvested on glass-fiber filter mats using a semiautomatic harvester (Skatron, Sterling, VA). Radioactivity ([³H]TdR) incorporation was determined using a beta scintillation counter. Differences between means of counts per minute for mitogens and controls were determined by Student’s t test for each mitogen. Stimulation indexes for each mitogen and Ag (optimal concentration) obtained by dividing its mean counts per minute by the mean control counts per minute were compared among groups by ANOVA 18 .

Generation of mice with targeted deletion at the IgA locus

A replacement-type targeting vector was generated as shown in Fig. 1⇑A. This targeting vector contains a 5′ homologous DNA region located immediately upstream of the Iα exon transcription start site and a 3′ homologous DNA region containing the 3′ half of the IgA constant region downstream of the EcoRI site. Inserted between the two homologous regions of DNA is a neomycin resistance gene, while an HSV-tk gene is located upstream of the 5′ homologous region. Homologous recombination by this targeting vector in ES cells leads to deletion of the Iα exon, the entire IgA switch region, and the 5′ half of the Cα region. This approach was taken to ensure that class switch to IgA would be prevented and that any aberrant DNA rearrangement at that locus would nonetheless lead to a nonfunctional IgA molecule.

This targeting vector was transfected by electroporation into the Iα.2 ES cell clone 8 . As shown in Fig. 1⇑A, this ES cell clone had a previous targeting event at the IgA locus on one allele, leading to deletion of the Iα exon with replacement by a human HPRT minigene. The Iα.2 ES cell clone had previously been used to generate mice with targeted deletion of the Iα exon. This ES cell clone was used in the current studies because it allowed triple selection of correctly targeted ES cell clones, i.e., loss of the HPRT minigene by growth in 6-thioguanine (TG), gain of the neomycin resistance gene by growth in G418, and loss of the HSV-tk gene by growth in gancyclovir (Ganc). Of 1 × 10⁷ transfected Iα.2 ES cells, six triple-resistant (Tg^r, G418^r, Ganc^r) clones were obtained. DNA from these clones were analyzed by Southern blotting (Fig. 1⇑B), and five of the six clones are shown. Digestion with KpnI and XhoI results in a 8.8-kb DNA fragment recognized by a Cε probe located upstream of the targeted region at the wild-type locus and a 10.7-kb DNA fragment from the Iα.2 locus previously targeted by the HPRT minigene. Subsequent targeting of this latter locus by the targeting vector leads to a 9.6-kb DNA fragment recognized by the Cε probe. Of the six ES cell clones obtained, four had correct targeting at the IgA locus. Additional restriction digestion and Southern blotting were performed to confirm proper targeting (data not shown).

Two of the correctly targeted ES cell clones were injected into C57BL/6 blastocysts from which chimeric mice were obtained. Two chimeric male mice with >50% chimerism were bred to C57BL/6 females with successful germline transmission. Tail DNA from F₂ offspring was digested with KpnI and XhoI and analyzed by Southern blotting with the Cε probe (Fig. 1⇑C). Mice on the 129Sv background have the Igh^a allotype, whereas C57BL/6 mice have the Igh^b haplotype. As a consequence, a KpnI polymorphism is present at the Cε locus leading to an 8.8-kb DNA fragment in 129Sv mice and a 6.0-kb DNA fragment in C57BL/6 mice. Thus, three different sized DNA fragments recognized by the Cε probe were seen in the F₂ offspring-8.8 kb wild-type 129Sv allele, the 6.0-kb wild-type C57BL/6 allele, and the 9.6-kb 129Sv targeted allele. Shown in Fig. 1⇑C are two mice that are homozygous for the 9.6-kb 129Sv targeted allele.

Mice with targeted deletion of the IgA locus are completely deficient in IgA

To assess the effects of targeted deletion at the IgA locus on expression of IgA in vivo, the following analyses were performed. IgA levels were measured in the serum and nasal, pulmonary, and gastrointestinal secretions (Fig. 2⇓A). No detectable IgA (<80 ng/ml) was found in any of these locations in mice with targeted deletion at the IgA locus, in contrast to wild-type mice, which had readily detectable and substantial levels of IgA at all these sites.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Ig levels in serum and secretions of IgA-deficient mice. A, Total IgA levels were measured in serum and intestinal, nasal, and pulmonary secretions of wild-type and homozygous IgA^−/− mice. Serum and secretions were collected from 12- to 18-wk-old mice and analyzed by isotype-specific ELISA. B, Total IgM levels in serum and secretions of wild-type and homozygous IgA^−/− mice. C, Total IgG levels in serum and secretions of wild-type and homozygous IgA^−/− mice.

Further, when surface expression of IgA was evaluated on Peyer’s patch B cells, no detectable surface IgA-expressing B cells were found in the Peyer’s patches of mice with targeted deletion at the IgA locus (Fig. 3⇓). In contrast, 10–15% of lymphocytes in the Peyer’s patches of wild-type mice expressed surface IgA. Similarly, no surface IgA-expressing B cells were found in the spleen of targeted mice, in contrast to wild-type mice (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

Surface IgA expression on Peyer’s patch B cells from IgA-deficient mice. Peyer’s patch lymphocytes were obtained from wild-type and IgA^−/− mice. Lymphocytes were stained with phycoerythrin anti-IgM and fluorescein anti-IgA, then analyzed by flow cytometry. The percentages of cells expressing IgM and/or IgA are shown.

The inability of B cells from targeted mice to produce IgA was further assessed in in vitro assays of IgA class switch. Spleen and Peyer’s patch B cells from wild-type and targeted mice were cultured in vitro in the presence or the absence of LPS and TGF-β (Fig. 4⇓). B cells from wild-type mice secreted small amounts of IgA when cultured in medium or LPS. As expected, production of IgA by wild-type B cells was increased substantially in the presence of LPS and TGF-β. In contrast, B cells from both spleen and Peyer’s patches produced no detectable IgA (<80 ng/ml) even when cultured with LPS and TGF-β.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

In vitro production of IgA by B cells from IgA-deficient mice. B cells were obtained from the spleen (A) and Peyer’s patches (B) of wild-type and IgA^−/− mice. They were cultured in vitro for 7 days in medium alone or with LPS or LPS and TGF-β. The LPS and LPS plus TGF-β cultures were also supplemented with IL-2 and IL-5. Supernatants were then harvested and analyzed for IgA content by isotype-specific ELISA.

IgA deficiency leads to altered expression of other Ig isotypes

The levels of other Ig isotypes were measured in serum and nasal, pulmonary, and gastrointestinal secretions. In serum, IgM and IgG levels were modestly elevated in targeted animals, being approximately twice the level seen in wild-type mice (IgM, 382 ± 35 vs 179 ± 27 μg/ml (p < 0.001); IgG, 11,629 ± 1,908 vs 6,264 ± 842 μg/ml (p < 0.02); Fig. 2⇑, B and C). In the case of nasal and pulmonary secretions, no significant differences were seen in IgM or IgG levels in targeted animals in comparison to wild-type mice. In contrast, in intestinal secretions IgM levels were approximately 25 times higher in targeted animals (5.1 ± 4.7 μg/ml) than those in wild-type mice (0.2 ± 0.1 μg/ml). In addition, intestinal IgG levels were roughly 3 times higher in targeted animals (29 ± 4 μg/ml) than those in wild-type mice (8.2 ± 1.7 μg/ml; p < 0.005).

In addition, we measured IgG subclass levels in serum and secretions. Fig. 5⇓ shows the results for serum IgG subclasses. A twofold increase in serum IgG1 was seen in targeted mice (p < 0.01), while no difference was seen in serum IgG2a levels. In the case of IgG2b and IgG3, there was roughly a fivefold increase in IgG2b (p < 0.01) and a sevenfold decrease in IgG3 (p < 0.001) in the serum of targeted mice (p < 0.005). In gastrointestinal secretions, IgG1 was increased roughly fivefold (p < 0.0005), IgG2a was increased twofold (p < 0.01), IgG2b was unchanged, and IgG3 was decreased fivefold (data not shown). In nasal secretions, no change was seen in IgG1, IgG2a, or IgG2b; however, there was a 10-fold decrease in IgG3. Lastly, in pulmonary secretions no change in IgG1 or IgG2a was seen; however, there was a twofold increase in IgG2b and a three- to fourfold decrease in IgG3 (data not shown). It should be noted that when compared with wild-type littermate control mice, IgA^−/− knockout mice exhibited significantly decreased IgG3 levels; however, their IgG3 levels were normal compared with those in parental C57BL/6 and 129 mice (data not shown).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 5.

Serum IgG subclass levels in IgA-deficient mice. Serum was collected from 12- to 18-wk-old wild-type and IgA^−/− mouse. IgG subclass levels were analyzed by subclass-specific ELISA.

In the case of IgE, serum IgE levels were significantly decreased in targeted mice (≤3 ng/ml) compared with those in wild-type mice (64 ± 23 ng/ml). Likewise, pulmonary secretions from targeted mice had substantially lower levels of IgE (2.6 ng/ml) than wild-type mice (64 ± 39 ng/ml). In the case of nasal and intestinal secretion, no detectable IgE was found in either targeted or wild-type mice.

Targeted deletion of IgA does not affect development of lymphoid tissues or T and B lymphocytes

IgA-deficient mice developed normally with no apparent increased susceptibility to infection in conventional animal facilities. The size of litters from IgA-deficient mice was comparable to that in wild-type mice, and no increased perinatal mortality was evident. Routine histologic examination of thymus, spleen, mesenteric lymph nodes, jejunum, and terminal ileum showed no differences between IgA-deficient mice and age-matched wild-type controls. Phenotypic analysis of T and B cells in various lymphoid tissues of these mice revealed a normal ratio of T and B cell subsets, such as CD4 T cells, CD8 T cells, and IgM B cells, etc. (data not shown).

Histology and immunohistochemistry of Peyer’s patches and small intestine in IgA-deficient mice

Peyer’s patches were well developed in IgA-deficient mice, with prominent germinal centers (Fig. 6⇓). Nonetheless, no IgA-containing plasma cells were detected by immunohistochemistry in small intestine lamina propria of IgA-deficient mice, in contrast to frequent IgA plasma cells in the lamina propria of wild-type mice (Fig. 6⇓).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 6.

Histology of Peyer’s patches and IgA production in lamina propria of IgA-deficient mice. Formalin-fixed sections containing Peyer’s patches from wild-type (A) or IgA-deficient (B) mice were stained with hematoxylin-eosin as described in Materials and Methods (magnification, ×100). Frozen sections containing small intestinal villi from wild-type (C) or IgA-deficient (D) mice were stained for IgA as described in Materials and Methods (magnification, ×100).

Normal lymphoproliferative responses to T and B cell mitogens in IgA-deficient mice

Splenic and lymph node lymphocytes from IgA^+/+ and IgA^−/− mice were stimulated with PHA, a T cell mitogen; LPS, a B cell mitogen; PWM, a T cell-dependent B cell mitogen; or SPL, an ubiquitous Ag that stimulates both T and B cells. Both mice exhibited similar lymphoproliferative responses to all mitogens with identical stimulation index values, except for PHA, which led to significant decrease in proliferation in the IgA^−/− mice (Table I⇓). Lymph node lymphocytes exhibited a similar pattern of responses as splenic lymphocytes (data not shown).

Because of the decreased response to PHA, the lymphoproliferative responses to two other T cell mitogens, Con A and anti-CD3 mAb, were further evaluated. IgA^−/− mice exhibited a normal response to Con A but a significantly decreased response to anti-CD3 mAb (p < 0.05) compared with control littermates and parental C57BL/6 mice (data not shown). The basis for the differential responses to these T cell mitogens is currently being investigated.

IFN-γ and IL-4 production and mRNA expression

To assess the ability to produce and express two pivotal immunoregulatory cytokines for stimulation of Th1-type (IFN-γ) and Th2-type (IL-4) responses, spleen cells were stimulated with or without PHA for 4 days. Cell culture supernatants were harvested and tested for IFN-γ and IL-4 by ELISA. Unstimulated splenocytes from both IgA^−/− and IgA^+/+ mice produced undetectable levels of IFN-γ (Fig. 7⇓A). However, PHA-stimulated cells elaborated significant amounts of IFN-γ. IgA^−/− mice produced slightly lower levels of IFN-γ than IgA^+/+ mice (p = 0.13). In contrast, unstimulated cells from both mice produced marked levels of IL-4. While IgA^−/− mice produced statistically significantly higher amounts of IL-4 than IgA^+/+ mice (p < 0.05), the difference may not be biologically significant. PHA stimulation resulted in increased production of IL-4; however, both IgA^−/− and IgA^+/+ mice produced similar amounts (Fig. 7⇓B).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 7.

IFN-γ and IL-4 production by IgA^−/− mice. Splenic lymphocytes were cultured in medium alone or stimulated with 2 μg/ml of PHA for 3 days. The levels of IFN-γ (A) and IL-4 (B) in cell culture supernatants were determined by ELISA. Values (mean ± SEM) represent data for four individual mice per group. ∗, p < 0.01; ∗∗, p < 0.001.

To further assess IFN-γ and IL-4 production induced by PHA stimulation and its correlation with increased transcription of IFN-γ and IL-4 mRNA, quantitative RT-PCR analysis specific for IFN-γ and IL-4 was performed on RNA samples obtained from PHA-stimulated and unstimulated spleen cells. The intensity of the band for PCR product for IFN-γ- specific mRNA in both IgA^+/+ and IgA^−/− mice increased 10-fold following stimulation with PHA (5 × 10⁻¹⁹ mM/μg of RNA) compared with unstimulated spleen cells (5 × 10⁻²⁰ mM/μg of RNA; Fig. 8⇓, A and C). In contrast, IL-4 mRNA was detected at levels approximately 150 times higher than those of IFN-γ mRNA in unstimulated cells (5 × 10–18.5 mM/μg RNA) from both IgA^+/+ and IgA^−/− mice. Following stimulation with PHA, IL-4 mRNA expression increased approximately 5-fold (5 × 10⁻¹⁸ mM/μg RNA) in IgA^−/− mice but was unchanged in most IgA^+/+ mice (Fig. 8⇓, B and D).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 8.

Quantitative RT-PCR analysis of IFN-γ and IL-4 mRNA expression. Total RNA (0.2 μg) from IgA^+/+ (A andB) and IgA^−/− (C andD) splenic lymphocytes stimulated or unstimulated with PHA was used as templates for IFN-γ and IL-4 cDNA synthesis. Tenfold serial dilutions of sample cDNA and murine IFN-γ and IL-4 MIMIC templates (Clontech) were then amplified to quantitate the PCR products. The PCR products were loaded on a 1% agarose gel (with MIMICS and samples in opposite directions) and separated by electrophoresis. The relative quantities of the sample PCR products were estimated from coamplified MIMIC bands for IFN-γ (A and C) and IL-4 (B and D). Calculated values are: IFN-γ: unstimulated cells, 5 × 10⁻²⁰ mM/μg of RNA; PHA-stimulated cells, 5 × 10⁻¹⁹ mM/μg of RNA; IL-4: unstimulated cells, 5 × 10^−18.5 mM/μg of RNA; PHA-stimulated cells, 5 × 10⁻¹⁸ mM/μg of RNA. Shown are results for one mice representative of analysis performed on four individual mice per group.

The above data demonstrate that the pattern of IFN-γ and IL-4 secretion parallels mRNA expression. However, the data suggest that while IFN-γ is secreted only upon PHA stimulation, IL-4 is constitutively secreted in unstimulated cells, with increased secretion following PHA stimulation, especially among IgA^−/− mice. Furthermore, both IgA^+/+ and IgA^−/− mice exhibited generally similar patterns of IFN-γ and IL-4 secretion and mRNA expression.

Effect of IgA deficiency on systemic and mucosal Ab responses to influenza hemagglutinin

The ability of IgA-deficient mice to develop serum and secretory Ab responses following immunization was assessed by immunizing mice mucosally with killed whole influenza virus (B-Yamagata) and cholera toxin. Abs against B-Yamagata in serum, nasal secretions, and intestinal secretions were subsequently measured by ELISA. As shown in Fig. 9⇓, no detectable IgA Abs against influenza virus were detected in serum or nasal or intestinal secretions from IgA-deficient mice, whereas wild-type mice gave readily detectable levels. In contrast, IgM Abs in IgA-deficient mice were ≥4-fold higher in serum, nasal, and intestinal secretions (p < 0.01, p < 0.05, and p < 0.08, respectively) compared with those in wild-type mice. However, no significant differences were noted in IgG Abs between IgA deficient and wild-type mice. These studies indicate that IgA-deficient mice produce no detectable IgA Abs following mucosal vaccination; however, they produce elevated IgM Abs compared with wild-type mice.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 9.

Production of influenza Abs in serum and secretions following mucosal immunization. Wild-type and IgA^−/− mice were immunized both intranasally and orally with heat-killed influenza virus plus cholera toxin as described in Materials and Methods. Serum and secretions from these mice were analyzed by ELISA for Abs against influenza: IgA (A), IgM (B), and IgG (C).