Background

The pathogenesis of IgA nephropathy is still obscure. The aim of this study was to investigate whether the fundamental pathogenesis of IgA nephropathy lies in bone marrow stem cells (BMCs).

Methods

We used donors of two different strains for bone marrow transplantation (BMT) into mice with a high content of serum IgA (ddY strain, HIGA mice), a murine model of IgA nephropathy. One group (B6HIGA, N = 5) received BMCs of C57BL/6j (B6) mice, and the other (HIGAHIGA, N = 8) were reconstituted with BMCs of HIGA mice.

Results

Twenty-six weeks after BMT, in B6HIGA mice, mesangial deposits of IgA and C3 were statistically milder than those in HIGAHIGA mice. Light microscopic observations disclosed that glomerular sclerosis and mesangial matrix expansion in B6HIGA mice were decreased compared with those in HIGAHIGA mice. These B6HIGA mice also excreted less urinary albumin than HIGAHIGA mice. Furthermore, serum levels of IgA in B6HIGA mice were markedly lower than those in HIGAHIGA mice. Size analysis of serum IgA revealed that macromolecular IgA were notably lower in B6HIGA mice than in HIGAHIGA mice.

Conclusions

Our results suggest that qualitative and quantitative changes of serum IgA are determined at the level of stem cells, and that BMT from normal donors can attenuate glomerular lesions in HIGA mice. This approach may offer a new avenue to study the pathogenesis of IgA nephropathy.

Methods

Mice

Female HIGA mice were obtained from Nippon Shinyaku Co. Ltd. (Kyoto, Japan). Female B6 mice were purchased from Japan Sankyo Labo Service Corporation, Inc. (Tokyo, Japan). All of the animals were maintained on autoclaved standard laboratory chow under specific pathogen-free conditions in the animal facility at Tokyo Metropolitan Institute of Gerontology. All surgical procedures conformed to the guidelines established by the U.S. Department of Health and Human Services, as published by the National Institutes of Health (NIH publication no. 85-23, revised 1985).

Experimental protocol

One week before BMT (pre-BMT), serum and urine were obtained from recipient HIGA mice (22 to 24 weeks of age), and then, after anesthesia with pentobarbital, approximately 2 mm cubes of renal sections were removed surgically. On the day of cell transfusion, recipients were lethally irradiated (8.5 Gy/mouse) and then divided into two groups: One group (designated as B6HIGA, N = 5) received BMCs of 25-week-old B6 mice, and the other (designated as HIGAHIGA, N = 8) were reconstituted with BMCs of 27-week-old HIGA mice. Five to six hours after irradiation, recipient HIGA mice (23 to 25 weeks of age) were injected intravenously with respective 1 10⁷ T-cell–depleted BMCs. From six weeks after transplantation, we obtained serum samples every four weeks. Twenty-six weeks after the reconstitution (approximately 50-week-old, post-BMT), all mice were sacrificed to obtain serum, urine, kidney specimens, and spleen cells. Serum and urine samples for immunological assay were kept at -80°C until use. Kidney specimens for immunohistopathology were embedded in OCT compound (Miles Scientific, Napperville, IL, USA) and quickly frozen in dry ice and acetone. Spleen cells were subjected to immunocytological analysis on the day of sacrifice.

Irradiation and T-cell–depleted bone marrow cells

Preliminary experiments showed that a single 8.5 Gy dose of total body irradiation from a ⁶⁰Co source killed all HIGA mice by three weeks after the irradiation if subsequent BMT was not carried out. In spleens of these irradiated mice, no hematopoietic colonies were visible macroscopically 14 days after irradiation, indicating that no pluripotential stem cells were present¹⁴. Conversely, autologous BMT rescued all HIGA mice exposed to 8.5 Gy without side effects such as gastrointestinal bleeding or infection. Based on these results, all recipient HIGA mice were irradiated with 8.5 Gy.

Bone marrow cells from pelvis, femoral, and peroneal bones were incubated with anti–Thy-1.2 Ab (F7D5; Serotec Ltd., Oxford, UK) at a 1:1000 dilution on ice for 30 minutes and then reacted with rabbit complement (Cedar Lane, Ontario, Canada) at 37°C for 30 minutes. Preliminary studies confirmed that this procedure eliminated more than 98% of the T cells, as revealed by immunofluorescent staining.

FACS analysis

Spleen cells were gently homogenized and depleted of red blood cells by Tris-buffered ammonium chloride (pH 7.2). To prevent nonspecific binding of Abs to Fc receptors, spleen cells were incubated with purified mouse IgG (Zymed Laboratories, Inc., San Francisco, CA, USA) on ice for 30 minutes. After washing, phycoerythrin-conjugated monoclonal mouse anti-H-2D^b Ab (Pharmingen, San Diego, CA, USA) was reacted with the pretreated cells on ice for 30 minutes. Then cells were resuspended in 1% paraformaldehyde containing 0.1% azide. Stained cells were analyzed by FACScan (EPICS CS; Coulter Electronics, Hialeah, FL, USA).

Histopathological analysis

For light microscopy, tissues embedded in paraffin were stained with periodic acid-Schiff (PAS) reagent. For assessment of histological changes, the frequency of globally or segmentally sclerosed glomeruli was calculated in each section by dividing the number of affected glomeruli by the total number of glomeruli. The extent in mesangial matrix increase was assessed for each glomerulus by scoring from 0 to 3: 0 = normal, 1 = mild (<one third of glomerular tuft area), 2 = moderate (<two thirds of glomerular tuft area), and 3 = severe (>two thirds of glomerular tuft area). The glomeruli were scored in the average of 56 glomeruli (range, 22 to 102 glomeruli) cross sections per specimen, and a value for each individual animal represents an averaged score in each section.

Cryostat sections were cut 4 m thick and fixed in cold acetone for 10 minutes. For direct immunofluorescence, the sections were stained with fluorescein isothiocyanate (FITC)-labeled goat anti-mouse IgA, IgG, IgM, and complement C3 Abs (Cappel Research Products, Durham, NC, USA) appropriately diluted in phosphate-buffered saline (PBS) at 37°C for one hour. Immunohistopathological analysis was carried out under a Zeiss Axiophot fluorescence photomicroscope. The relative amount of staining was scored semiquantitatively as follows¹²: 0 = negative staining, 1 = slight (focal and weak) staining, 2 = moderate staining, and 3 = marked (diffuse and intense) staining. More than 20 glomeruli per specimen were scored in this manner, and the mean was calculated by the total score of affected glomeruli by the total number of glomeruli as fluorescent intensity (FI) in each individual animal.

Urinary albumin concentration

Concentrations of excreted urinary albumin were determined by the single radial immunodiffusion method (SRID)¹⁵. In brief, urine samples and a mouse albumin reference (Cappel Research Products) as a standard were applied to wells on agarose gel-containing polyclonal rabbit anti-mouse albumin Ab (Cappel Research Products). By this method, we could minimally detect 60 g/ml of mouse albumin. Because many samples were in an undetectable range, excreted urinary albumin concentrations measured by SRID were grouped for the statistical analysis as 0, 0 to 99 g/ml; 1, 100 to 199 g/ml; 2, 200 to 299 g/ml; and 3, 300 to 399 g/ml.

Enzyme-linked immunosorbent assay

Concentrations of mouse immunoglobulins were measured by enzyme-linked immunosorbent assay (ELISA) as follows. Ninety-six-well plates (IMMULON 2; Dynex Technologies, Inc., Chantilly, VA, USA) had been previously coated with goat affinity-purified antimouse immunoglobulin Abs (Cappel Research Products). After reacting with PBS containing 1% skim milk powder to block nonspecific reactivity of sera, the plates were incubated with appropriately diluted samples and immunoglobulin reference sera (ICN Pharmaceuticals Inc., Costa Mesa, CA, USA) and then washed with PBS containing 0.05% Tween20 (PBS/Tween). Adsorbed immunoglobulin was incubated with peroxidase-labeled goat antimouse immunoglobulin Ab (Zymed Laboratories). After washing with PBS/Tween, the residual peroxidase was reacted with TMB peroxidase substrate (Kirkegaard & Perry Laboratories Inc., Gaithersburg, MD, USA) according to manufacturer's instructions. Absorbance in each well was measured with a MTP-100 microplate reader (Corona Electric Co., Tokyo, Japan) at a wavelength of 450 nm.

Fast protein, peptide, and polynucleotide liquid chromatography

Ten-times diluted serum samples pooled from mice of each group were subjected to the fast protein, peptide, and polynucleotide liquid chromatography (FPLC) system (Pharmacia, Uppsala, Sweden) on a Superdex 200 HR 10/30 (Pharmacia). Molecular weight markers were thyroglobulin (669 kDa) and aldolase (158 kDa; Pharmacia). Samples were applied to the FPLC column at a flow rate of 0.4 ml/min, and fractionated samples were collected every one minute in 0.1 M phosphate buffer, pH 7.0, containing 0.1 M NaCl.

Statistical analysis

All values are expressed as means SEM. The two-group paired t-test was used to compare the changes of histology at pre- and post-BMT in each individual animal; the unpaired t-test was used to analyze ELISA and histology data between transplanted animals, and the Mann–Whitney test was used to compare urinary data. A P value of less than 0.05 was considered statistically significant.

RESULTS

Donor bone marrow stem cells repopulated in recipients

To confirm that BMCs from each donor successfully repopulated in recipients, H-2 typing was performed 26 weeks after BMT. Splenic lymphocytes were stained with anti–H-2D^b Ab and then analyzed by FACS. Lymphocytes of nontransplanted B6 mice, whose MHC class I was H-2D^b, showed positive staining Figure 1a, whereas lymphocytes in HIGA mice, whose MHC type has not yet been determined, were negative for H-2D^b Figure 1b. This method enabled us to recognize whether lymphocytes in recipients were B6 or HIGA in origin. Accordingly, lymphocytes in HIGA mice received BMCs of B6 mice (B6HIGA) were positive like those of B6 mice Figure 1c. HIGA recipients of BMCs of HIGA mice (HIGAHIGA) had H-2D^b-negative lymphocytes Figure 1d. Thus, BMCs transferred to B6HIGA mice successfully replaced those of the HIGA type.

Figure 1.

FACS analysis of lymphocytes in spleens 26 weeks after bone marrow transplantation (BMT). Splenic lymphocytes depleted of red blood cells were stained with anti–H-2D^b antibodies (Ab) and were then subjected to FACS. Results represent (A) 50-week-old B6 mouse, (B) 50-week-old HIGA mouse, (C) B6HIGA mouse 26 weeks after BMT, and (D) HIGAHIGA mouse 26 weeks after BMT.

Full figure and legend (19K)

Immunodeposition decreased in glomeruli of B6HIGA mice

Because BMCs of B6HIGA mice originated from B6 mice, we next examined mesangial depositions of immunoglobulins and C3 by immunofluorescent staining. At the post-BMT period (26 weeks after BMT; when mice were approximately 50 weeks old), mesangial deposition of IgA was weak in B6HIGA mice Figure 2a. At the same time, IgA deposits in HIGAHIGA mice were observed diffusely and globally in the mesangial area Figure 2b.

Figure 2.

Representative immunofluorescence photographs of IgA staining in renal tissues (original magnification 200) of (A) B6HIGA mouse at post-BMT and (B) HIGAHIGA mouse at post-BMT. Frozen sections were stained directly with FITC-labeled antimouse IgA Ab. At post-BMT, glomeruli in B6HIGA mice showed weak IgA deposition. In contrast, HIGAHIGA mice at post-BMT had global, diffuse mesangial IgA deposits.

Full figure and legend (26K)

When we then scored the intensity of fluorescence (FI), mesangial IgA depositions in both B6HIGA (N = 5) and HIGAHIGA (N = 8) mice were similarly slight at pre-BMT (one week before BMT; Figure 3 A, B). In B6HIGA mice, the FI values of IgA depositions at post-BMT were not statistically different from those at pre-BMT (1.17 0.14 vs. 0.79 0.03, P > 0.05; Figure 3a). The degree of mesangial IgA depositions in B6HIGA mice at post-BMT was as same as those in nontransplanted normal B6 mice at the age of 50 weeks (1.34 0.32). In sharp contrast, all HIGAHIGA mice had much higher FI scores for mesangial IgA deposition at post-BMT than those at pre-BMT (2.49 0.16 vs. 0.84 0.05, P < 0.0001; Figure 3b). It should be noted that FI scores of mesangial IgA deposition in nontransplanted HIGA mice at 50 weeks old were comparable to those of post-BMT HIGAHIGA mice (2.39 0.24). At post-BMT, mesangial IgA depositions in B6HIGA mice were statistically milder than those in HIGAHIGA mice (P < 0.01) or age-matched HIGA mice. Clearly, mesangial IgA depositions of B6HIGA mice took a turn for the better compared with those of HIGAHIGA mice or with nontransplanted HIGA mice.

Figure 3.

Comparison of mesangial immunofluorescent intensity and mesangial matrix expansion scores before and after BMT. (A) IgA staining in B6HIGA mice (N = 5). (B) IgA staining in HIGAHIGA mice (N = 8). (C) C3 staining in B6HIGA mice. (D) C3 staining in HIGAHIGA mice. (E) Mesangial matrix expansion scores of B6HIGA mice. (F) Mesangial matrix expansion scores of HIGAHIGA mice. Bars are means SEM. Renal sections biopsied before and 26 weeks after BMT were stained for IgA and C3, followed by scoring more than 20 glomeruli per section and calculating the values of FI as described in the Methods section. The degree of mesangial matrix expansion was semiquantitatively calculated as detailed there. P values are indicated between pre- and post-BMT in each group (paired t-test) or between B6HIGA mice at post-BMT and HIGAHIGA mice at post-BMT (unpaired t-test).

Full figure and legend (43K)

Next, mesangial complement C3 deposition was analyzed because such deposits universally accompany human and murine IgA nephropathy^5,12. At pre-BMT, mesangial depositions of C3 were seldom apparent in either group of mice. In B6HIGA mice, FI scores of C3 at post-BMT were comparable to those at pre-BMT (0.17 0.08 vs. 0.0 0.0, P > 0.05; Figure 3c). However, in HIGAHIGA mice, scores of C3 deposits were considerably higher after than before BMT (1.08 0.25 vs. 0.02 0.02, P < 0.01; Figure 3d). FI scores of mesangial C3 depositions in B6HIGA mice were statistically lower than those in HIGAHIGA mice (P < 0.05 at post-BMT).

At post-BMT, mesangial depositions of IgG in B6HIGA mice were also milder than those in HIGAHIGA mice without a statistically significant difference (1.28 0.32 vs. 1.80 0.21, P > 0.05), whereas mesangial IgM depositions in B6HIGA mice were the same as those in HIGAHIGA mice (1.38 0.32 vs. 1.49 0.27).

Glomerular sclerotic lesion and mesangial matrix expansion diminished in B6HIGA mice

Because immunofluorescent staining revealed that BMT from normal B6 mice attenuated mesangial IgA and C3 deposition in HIGA mice, this effect was analyzed with light microscopy. In both groups, no glomeruli had sclerotic lesions at pre-BMT. At post-BMT period, few glomeruli in B6HIGA mice had sclerotic lesions. In fact, the frequency of glomerulosclerotic lesions in B6HIGA mice Figure 4a was significantly lower than that in HIGAHIGA mice (0.40 0.40 vs. 9.8 1.5%, P < 0.001; Figure 4b). Next, we scored the degree of mesangial matrix expansion and calculated the values of matrix expansion scores as described in the Methods section. In B6HIGA recipients of BMT, scores at post-BMT were not statistically higher than those at pre-BMT (0.90 0.11 vs. 0.69 0.10, P > 0.05; Figure 3e). Conversely, the mesangial matrix expansion scores of HIGAHIGA mice were considerably higher after than before BMT (1.49 0.10 vs. 0.76 0.06, P < 0.01; Figure 3f). At post-BMT, the mesangial matrix expansion scores of B6HIGA mice were statistically lower than those of HIGAHIGA mice (P < 0.01). BMT from normal mice also improved the glomerular changes of HIGA mice, as judged by light microscopy.

Figure 4.

Representative microscopic photographs in renal tissues at post-BMT (PAS staining, original magnification 400). (A) Glomeruli in B6HIGA mouse showed a normal appearance. (B) In contrast, glomeruli in the HIGAHIGA mouse had severe mesangial matrix expansion and segmental glomerulosclerosis.

Full figure and legend (198K)

Decreased concentration of excreted albuminuria from B6HIGA mice

Next, we examined how these histological improvements effected urinary albumin excretion. Albumin excretion from HIGA mice was virtually undetectable at pre-BMT (<60 g/ml; Figure 5). However, at post-BMT, the scores of albuminuria in B6HIGA mice were statistically lower than those from HIGAHIGA mice (P < 0.05; Figure 5). Therefore, in addition to ameliorating the renal lesions, BMT also reduced albuminuria, an important clinical feature of IgA nephropathy.

Figure 5.

Comparison of the concentration of urinary albumin. The concentration of albumin was measured by the single radial immunodiffusion method and was then scored for statistical analysis. This figure shows albuminuria in (A) B6HIGA mice and in (B) HIGAHIGA mice at pre- and post-BMT.

Full figure and legend (11K)

Serum IgA levels, which increase with age in HIGA mice, were decreased by bone marrow transplantation from B6 mice

To investigate the immunological components participating in these effects of BMT, we next calculated serum immunoglobulin levels by ELISA. Serum IgA levels Figure 6a in B6HIGA mice were much lower than those in HIGAHIGA mice throughout the experimental course after BMT (402 65 vs. 1519 93 mg/dl, P < 0.01 at post-BMT). However, serum IgM levels Figure 6b in B6HIGA mice were significantly higher than in HIGAHIGA mice from 14 weeks after BMT (85.0 9.8 vs. 48.1 2.2 mg/day, P < 0.01 at post-BMT). Serum IgG levels Figure 6c in both groups changed similarly throughout the experimental course (1128 79 vs. 1242 74 mg/dl, P > 0.05 at post-BMT).

Figure 6.

Serum immunoglobulin levels determined by ELISA for (A) IgA, (B) IgM, and (C) IgG in B6HIGA mice () and in HIGAHIGA mice (). Values are means SEM; *P < 0.05 vs. HIGAHIGA mice. Serum samples were obtained before BMT and every four weeks after BMT.

Full figure and legend (24K)

Bone marrow transplantation from B6 mice decreased serum macromolecular IgA of transplanted HIGA mice

To assess the size of serum IgA as a determinant of mesangial deposition at post-BMT, IgA level in samples fractionated by FPLC were measured by ELISA. Preliminary examination revealed that most of these IgA molecules in nontransplanted HIGA mice at 25 and 50 weeks of age were macromolecules (molecular weight> 500 kDa). We also confirmed that macromolecular IgA was not predominant in serum of normal B6 mice at 25 and 50 weeks old. As in the HIGA mice, macromolecular IgA with a molecular weight of approximately 670 kDa predominated in HIGAHIGA mice Figure 7. However, in B6HIGA mice, levels of macromolecular IgA were markedly lower than in HIGAHIGA mice and were equivalent to those of monomeric IgA with a molecular weight of 160 kDa Figure 7.

Figure 7.

Analysis of IgA size in sera from B6HIGA mice () and HIGAHIGA mice (). Serum samples collected from mice of each group were fractionated by FPLC (0.4 ml/min, 1 ml/tube), and the concentrations of IgA in fractions were determined by ELISA. Monomeric IgA has been identified as about 160 kDa²⁴.

Full figure and legend (13K)

DISCUSSION

When we compared IgA nephropathy-prone HIGA mice transplanted with BMCs from normal B6 mice (B6HIGA) to their counterparts given HIGA BMCs to probe the origin of IgA nephropathy, our results showed clear-cut reductions in mesangial IgA and complement C3 deposition and a decrease in glomerular sclerotic lesions and mesangial matrix expansion only in the recipients of B6 cells. These B6HIGA mice also excreted less urinary albumin than the HIGAHIGA controls. Subsequent analysis of immunological changes revealed that the serum IgA level, especially macromolecular IgA, in B6HIGA mice was much lower than in HIGAHIGA mice.

For successful BMT and efficient observation of its effects, we needed a murine model in which mesangial IgA deposition occurred by the 25th week of life. The essential factors were that the thymus is considered crucial to immunological reconstitution after BMT¹⁶, but thymic atrophy starts in mice at 25 weeks of age and worsens drastically thereafter¹⁷. Because HIGA mice, which originated from IgA nephropathy-prone ddY mice and produce especially high levels of serum IgA, exhibited mesangial IgA deposits at 25 weeks of age and these deposits become more marked with age¹², we chose them for this study.

In general, T-cell–depleted BMCs are the cells used in BMT^18,19. For the generation of naive T cells capable of differentiating into antigen-specific T lymphocytes through education in thymus, two to three months are required after BMT^20,21. In this study, 26 weeks after the transplantation of T-cell–depleted BMCs, donor BMCs were successfully regenerated in recipients, and the cell surface phenotype of lymphocytes in B6HIGA mice completely changed from the HIGA type to the B6 type Figure 1. Additionally, the IgG level in B6HIGA mice was comparable, and the IgM content was superior to that in HIGAHIGA mice Figure 6 or age-matched, unmanipulated HIGA mice (data not shown). These findings contradict some reports that immunodeficiency occasionally occurred in mice after lethal irradiation followed by transfer of multiple histocompatibility complex-mismatched BMT²². In our transplant recipients, no immunological defect was evident. Therefore, we considered it feasible to examine the role of BMCs in the pathogenesis of IgA nephropathy by analyzing mice 26 weeks after BMT.

As a result of BMT, mesangial IgA depositions in B6HIGA mice were statistically weaker than those in HIGAHIGA mice or age-matched HIGA mice Figure 2 and 3. Until now, mesangial IgA deposition has been attributed to a high production of IgA²³, the presence of serum macromolecular IgA^6,13, the defective clearance of IgA⁹, and abnormal cytokine production⁸. To pursue the study of factors that attenuate mesangial IgA deposition in B6HIGA mice, we first analyzed the nature of serum IgA in this model.

As shown in Figure 6, serum IgA levels in B6HIGA mice were much lower than in HIGAHIGA mice. Others believe that not only in murine but also in human IgA nephropathy, mesangial IgA deposition correlates better with the amount of macromolecular IgA^13,24,25, because of the macromolecular form's (a) reduced clearance from the circulation⁹, (b) high affinity to mesangial cell⁶, and (c) enhancement of interleukin-6 expression, which induces immunoglobulin class switching to IgA^26,27. For these reasons, we next analyzed the molecular size of serum IgA in our model and found that serum macromolecular IgA levels in B6HIGA mice were notably lower than in HIGAHIGA mice Figure 6. Indeed, as reported in patients with AIDS, the increment of macromolecular IgA was not necessarily accompanied by mesangial IgA deposits^28,29. However, it is quite possible that the marked decrement of serum macromolecular IgA could attenuate mesangial IgA deposition in B6HIGA mice because macromolecular IgA, which is a predominant constitute of the deposition¹², is decreased in the circulation. Our study may indicate that macromolecular IgA produced in mice with IgA nephropathy has a tendency to deposit into the mesangial area, unlike that of AIDS patients.

In addition to its role in IgA deposition, macromolecular IgA exerts its action variously in IgA nephropathy. Macromolecular IgA is a potent activator of complement³⁰, an inducer of matrix expansion^26,31, and one of the molecules responsible for inducing proteinuria³². In this study, weaker mesangial C3 deposition, decreased glomerular sclerosis, milder matrix expansion, and less excretion of proteinuria were observed in B6HIGA mice than the other experimental group, presumably because the B6HIGA subjects had less macromolecular IgA.

Because BMCs transplanted into both B6HIGA mice and HIGAHIGA mice were equally exposed to the same environmental factors, the immunological differences in the two groups must come from variations in the BMCs. Accordingly, qualitative and quantitative changes of serum IgA may be determined at the level of stem cells. In support of this notion, we have obtained a result that BMT from IgA nephropathy-prone ddY mice to normal B6 mice increased the recipients' serum macromolecular IgA and glomerular IgA deposition³³. We do not deny the involvement of antigens in the pathogenesis of IgA nephropathy. However, we consider that the antigens may merely elicit the basic factor that lies in BMCs. Still obscure is what factor(s) of BMCs, not of the environment, reduces macromolecular IgA in B6HIGA mice. We speculate that bone marrow-derived cells, such as T cells^7,34,35, B cells^36,37, Kupffer cells^9,38,39,40 and mucosal intraepithelial cells^5,41,42, may contribute to determine qualitative and quantitative changes of serum IgA in HIGA mice and that the defects of these cells might be determined at the stem cell level. It was also thought that BMT from normal donors improved the abnormalities of bone marrow-derived cells and attenuated the glomerular lesions in these mice. We are now attempting to clarify which of these cells and what factors are defective at the stem cell level and are most important for increasing serum levels of macromolecular IgA in IgA nephropathy.

This study revealed that qualitative and quantitative changes in serum IgA are determined at the level of stem cells, and that disordered BMCs in mice with IgA nephropathy are responsible for high serum levels of IgA, especially macromolecular IgA, enough to cause glomerular lesions. Furthermore, our data implicate that BMT is therapeutically beneficial for murine IgA nephropathy and warrants a pioneering approach to study the pathogenesis of IgA nephropathy.

References

1. Berger, J, Hinglais, N: Intercapillary deposits of IgA-IgG. J Urol Nephrol 1968 74: 694–695,  | ChemPort |
2. D'Amico, G: Clinical features and natural history in adults with IgA nephropathy. Am J Kidney Dis 1988 12: 353–357,  | PubMed | ChemPort |
3. Chauveau, D, Droz, D: Follow-up evaluation of the first patients with IgA nephropathy described at Necker Hospital. Contrib Nephrol 1993 104: 1–5,  | PubMed | ChemPort |
4. Ibels, LS, Gyory, AZ: IgA nephropathy: Analysis of the natural history, important factors in the progression of renal disease, and a review of the literature. Medicine (Baltimore) 1994 73: 79–102,  | PubMed | ChemPort |
5. Galla, JH: IgA nephropathy. Kidney Int 1995 47: 377–387,  | PubMed | ISI | ChemPort |
6. Lai, KN, To, WY, Li, PK, Leung, JC: Increased binding of polymeric lambda-IgA to cultured human mesangial cells in IgA nephropathy. Kidney Int 1996 49: 839–845,  | PubMed | ISI | ChemPort |
7. Lai, KN, Ho, RT, Lai, CK, Chan, CH, Li, PK: Increase of both circulating Th1 and Th2 T lymphocyte subsets in IgA nephropathy. Clin Exp Immunol 1994 96: 116–121,  | PubMed | ISI | ChemPort |
8. de Caestecker, MP, Bottomley, M, Telfer, BA, Hutchinson, IV, Vose, BM, Ballardie, FW: Detection of abnormal peripheral blood mononuclear cell cytokine networks in human IgA nephropathy. Kidney Int 1993 44: 1298–1308,  | PubMed | ChemPort |
9. Rifai, A, Schena, FP, Montinaro, V, Mele, M, D'Addabbo, A, Nitti, L, Pezzullo, JC: Clearance kinetics and fate of macromolecular IgA in patients with IgA nephropathy. Lab Invest 1989 61: 381–388,  | PubMed | ISI | ChemPort |
10. Berger, J: Recurrence of IgA nephropathy in renal allografts. Am J Kidney Dis 1988 12: 371–372,  | PubMed | ISI | ChemPort |
11. Silva, FG, Chander, P, Pirani, CL, Hardy, MA: Disappearance of glomerular mesangial IgA deposits after renal allograft transplantation (letter).Transplantation 1982 33: 241–246,  | PubMed | ChemPort |
12. Muso, E, Yoshida, H, Takeuchi, E, Yashiro, M, Matsushima, H, Oyama, A, Suyama, K, Kawamura, T, Kamata, T, Miyawaki, S, Izui, S, Sasayama, S: Enhanced production of glomerular extracellular matrix in a new mouse strain of high serum IgA ddY mice. Kidney Int 1996 50: 1946–1957,  | PubMed | ISI | ChemPort |
13. Monteiro, RC, Halbwachs, ML, Roque, BM, Noel, LH, Berger, J, Lesavre, P: Charge and size of mesangial IgA in IgA nephropathy. Kidney Int 1985 28: 666–671,  | PubMed | ISI | ChemPort |
14. Magli, MC, Iscove, NN, Odartchenko, N: Transient nature of early haematopoietic spleen colonies. Nature 1982 295: 527–529,  | Article | PubMed | ISI | ChemPort |
15. Mancini, G, Carbonara, AO, Heremans, JF: Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry 1965 2: 235–254,  | Article | PubMed | ChemPort |
16. Miller, JFAP, Doak, SMA, Cross, AM: Role of the thymus in recovery of the immune mechanism in the irradiated adult mouse. Proc Soc Exp Biol Med 1963 112: 785–792,  | ISI |
17. Aspinall, R: Age-associated thymic atrophy in the mouse is due to a deficiency affecting rearrangement of the TCR during intrathymic T cell development. J Immunol 1997 158: 3037–3045,  | PubMed | ISI | ChemPort |
18. Ikehara, S, Good, RA, Nakamura, T, Sekita, K, Inoue, S, Oo, MM, Muso, E, Ogawa, K, Hamashima, Y: Rationale for bone marrow transplantation in the treatment of autoimmune diseases. Proc Natl Acad Sci USA 1985 82: 2483–2487,  | PubMed | ChemPort |
19. Ikehara, S, Yasumizu, R, Inaba, M, Izui, S, Hayakawa, K, Sekita, K, Toki, J, Sugiura, K, Iwai, H, Nakamura, T, Muso, E, Hamashima, Y, Good, RA: Long-term observations of autoimmune-prone mice treated for autoimmune disease by allogeneic bone marrow transplantation. Proc Natl Acad Sci USA 1989 86: 3306–3310,  | PubMed | ChemPort |
20. Parkman, R, Weinberg, KI: Immunological reconstitution following bone marrow transplantation. Immunol Rev 1997 157: 73–78,  | PubMed | ISI | ChemPort |
21. Ault, KA, Antin, JH, Ginsburg, D, Orkin, SH, Rappeport, JM, Keohan, ML, Martin, P, Smith, BR: Phenotype of recovering lymphoid cell populations after marrow transplantation. J Exp Med 1985 161: 1483–1502,  | Article | PubMed | ISI | ChemPort |
22. Wang, BY, Cherry, El-Badri, NS, Good, RA: Prevention of development of autoimmune disease in BXSB mice by mixed bone marrow transplantation. Proc Natl Acad Sci USA 1997 94: 12065–12069,  | Article | PubMed | ChemPort |
23. Emancipator, SN, Lamm, ME: IgA nephropathy: Overproduction or decreased clearance of immune complexes? Lab Invest 1989 61: 365–367,  | PubMed | ISI | ChemPort |
24. Muso, E, Yoshida, H, Takeuchi, E, Shimada, T, Yashiro, M, Sugiyama, T, Kawai, C: Pathogenic role of polyclonal and polymeric IgA in a murine model of mesangial proliferative glomerulonephritis with IgA deposition. Clin Exp Immunol 1991 84: 459–465,  | PubMed | ISI | ChemPort |
25. Rifai, A, Small, PJ, Teague, PO, Ayoub, EM: Experimental IgA nephropathy. J Exp Med 1979 150: 1161–1173,  | Article | PubMed | ISI | ChemPort |
26. van den Dobbelsteen, ME, van der Woude, FJ, Schroeijers, WE, van den Wall Bake, AWL, van Es, LA, Daha, MR: Binding of dimeric and polymeric IgA to rat renal mesangial cells enhances the release of interleukin 6. Kidney Int 1994 46: 512–519,  | PubMed | ChemPort |
27. Beagley, KW, Eldridge, JH, Lee, F, Kiyono, H, Everson, MP, Koopman, WJ, Hirano, T, Kishimoto, T, McGhee, JR: Interleukins and IgA synthesis: Human and murine interleukin 6 induce high rate IgA secretion in IgA-committed B cells. J Exp Med 1989 169: 2133–2148,  | Article | PubMed | ISI | ChemPort |
28. Jackson, S, Tarkowski, A, Collins, JE, Dawson, LM, Schrohenloher, RE, Kotler, DP, Koopman, WJ: Occurrence of polymeric IgA1 rheumatoid factor in the acquired immune deficiency syndrome. J Clin Immunol 1988 8: 390–396,  | Article | PubMed | ISI | ChemPort |
29. Beaufils, H, Jouanneau, C, Katlama, C, Sazdovitch, V, Hauw, JJ: HIV-associated IgA nephropathy: A post-mortem study. Nephrol Dial Transplant 1995 10: 35–38,  | PubMed | ISI | ChemPort |
30. Bogers, WM, Stad, RK, van Es, LA, Daha, MR: Immunoglobulin A: Interaction with complement, phagocytic cells and endothelial cells. Complement Inflamm 1991 8: 347–358,  | PubMed | ISI | ChemPort |
31. Ruef, C, Kashgarian, M, Coleman, DL: Mesangial cell-matrix interactions: Effects on mesangial cell growth and cytokine secretion. Am J Pathol 1992 141: 429–439,  | PubMed | ISI | ChemPort |
32. Stad, RK, Bruijn, JA, van Gijlswijk-Janssen, DJ, van Es, LA, Daha, MR: An acute model for IgA-mediated glomerular inflammation in rats induced by monoclonal polymeric rat IgA antibodies. Clin Exp Immunol 1993 92: 514–521,  | PubMed | ISI | ChemPort |
33. Imasawa, T, Utsunomiya, Y, Kawamura, T, Nagasawa, R, Maruyama, N, Sakai, O: Evidence suggesting the involvement of hematopoietic stem cells in the pathogenesis of IgA nephropathy. Biochem Biophys Res Commun 1998 249: 605–611,  | Article | PubMed | ISI | ChemPort |
34. Sakai, H, Miyazaki, M, Endoh, M, Nomoto, Y: Increase of IgA-specific switch T cells in patients with IgA nephropathy. Clin Exp Immunol 1989 78: 378–382,  | PubMed | ISI | ChemPort |
35. Chintalacharuvu, SR, Emancipator, SN: The glycosylation of IgA produced by murine B cells is altered by Th2 cytokines. J Immunol 1997 159: 2327–2333,  | PubMed | ISI | ChemPort |
36. Allen, AC, Topham, PS, Harper, SJ, Feehally, J: Leucocyte beta 1,3 galactosyltransferase activity in IgA nephropathy. Nephrol Dial Transplant 1997 12: 701–706,  | Article | PubMed | ISI | ChemPort |
37. Kokubo, T, Hiki, Y, Iwase, H, Horii, A, Tanaka, A, Nishikido, J, Hotta, K, Kobayashi, Y: Evidence for involvement of IgA1 hinge glycopeptide in the IgA1-IgA1 interaction in IgA nephropathy. J Am Soc Nephrol 1997 8: 915–919,  | PubMed | ISI | ChemPort |
38. Gale, RP, Sparkes, RS, Golde, DW: Bone marrow origin of hepatic macrophages (Kupffer cells) in humans. Science 1978 201: 937–938,  | PubMed | ISI | ChemPort |
39. Bogers, WM, Stad, RK, Janssen, DJ, Prins, FA, van Rooijen, N, van Es, LA, Daha, MR: Kupffer cell depletion in vivo results in clearance of large-sized IgA aggregates in rats by liver endothelial cells. Clin Exp Immunol 1991 85: 128–136,  | PubMed | ISI | ChemPort |
40. Grossetête, B, Launay, P, Lehuen, A, Jungers, P, Bach, JF, Monteiro, RC: Down-regulation of Fc receptors on blood cells of IgA nephropathy patients: Evidence for a negative regulatory role of serum IgA. Kidney Int 1998 53: 1321–1335,  | Article | PubMed | ISI | ChemPort |
41. Mosley, RL, Styre, D, Klein, JR: Differentiation and functional maturation of bone marrow-derived intestinal epithelial T cells expressing membrane T cell receptor in athymic radiation chimeras. J Immunol 1990 145: 1369–1375,  | PubMed | ISI | ChemPort |
42. Denning, GM: IL-4 and IFN-gamma synergistically increase total polymeric IgA receptor levels in human intestinal epithelial cells: Role of protein tyrosine kinases. J Immunol 1996 156: 4807–4814,  | PubMed | ISI | ChemPort |

Acknowledgments

We thank Mrs. Minori Kamada for help with FACS analysis, Dr. Chen Guang Ping for analysis of renal pathology, Dr. Tatsuo Shimosawa and Dr. Taro Kogure for suggestions about manuscript preparation, Ms. Phyllis Minick and Dr. Masao Okazaki for critical editing of the manuscript, and the staff of radiotherapeutics at the Tokyo Metropolitan Institute of Gerontology.