Background

Galectins are characterized by specific affinity for -galactoside sugars, and they play a role in diverse biological processes, including cell adhesion, cell proliferation, and apoptosis. Galectin-1, -3, and -9 have been implicated in modulating the immune response.

Methods

Nephrotoxic serum nephritis, which is characterized by crescent formation and glomerular influx of CD8⁺ cells into glomerular capillaries, was induced in Wistar Kyoto (WKY) rats by injecting rabbit antiglomerular basement membrane serum. Following induction, the rats were treated either with phosphate-buffered saline or dexamethasone, galectin-1, galectin-3, or galectin-9 on alternate days and were sacrificed at day 14. At day 8, splenic lymphocytes were isolated and employed for terminal deoxytransferase-mediated uridine triphosphate nick end-labeling (TUNEL) assay to assess the degree of apoptosis, and the kidneys were utilized to determine the extent of influx of CD4⁺ and CD8⁺ cells and glomerular damage.

Results

Dexamethasone induced a marked apoptosis of splenic CD4⁺ and CD8⁺ cells, and it inhibited the production of anti-rabbit IgG and the influx of CD8⁺ cells and macrophages into the renal glomeruli. Crescent formation and excretion of urinary proteins were also reduced. Galectin-9 failed to induce apoptosis in the CD4⁺ cells; however, it induced apoptosis in the CD8⁺ cells and inhibited the infiltration of CD8⁺ cells. Although galectin-1 and -3 did not induce the apoptosis in the T cells, they inhibited the accumulation of macrophages in the renal glomeruli. Like dexamethasone, the galectins also reduced the crescentic formation, proliferation of glomerular cells, and excretion of urinary proteins.

Conclusions

Galectin-9 selectively induces apoptosis of the activated CD8⁺ cells, while the macrophage influx into the kidney is modulated by all three galectins. This finding raises an interesting possibility for the utility of galectins in the modulation of macrophages that are involved in immune-mediated glomerular diseases.

METHODS

Animals and preparation of rabbit nephrotoxic serum

Four-week-old female WKY rats, weighing approximately 100 g, were purchased from Charles River Japan (Atsugi, Kanagawa, Japan) and were used for the isolation of glomerular basement membranes (GBMs), which were used as an antigen for the induction of NTS nephritis. WKY kidneys were perfused with physiologic saline. They were then harvested, homogenized, and glomeruli isolated. An emulsion of equal volume of the homogenate and the Freund's complete adjuvant (Difco Laboratories, Detroit, MI, USA) was prepared and injected subcutaneously into rabbits twice a month for two months. Three weeks after the last injection, the rabbits were bled, and the antisera were isolated. The antiserum was decomplemented for 30 minutes at 56°C and adsorbed with freshly harvested rat red blood cells, and finally, it was used for the induction of nephritis of WKY rats.

Preparation of mouse recombinant protein of galectins-1, -3 and -9

Recombinant galectins (G1, G3, and G9) were prepared by using pTrcHis2 vector (Invitrogen, San Diego, CA, USA), as previously described^5,6. These recombinant proteins were fusion proteins with C-terminal c-myc epitope and (His)₆. In brief, the constructs of G1 (pTrcHis2/G1), G3 (pTrcHis2/G3), and G9 (pTrcHis2/G9)⁶ were transformed into the TOP 10 bacterial host (Invitrogen). A single transformed bacterial colony was grown in Luria-Bertani's broth, and protein synthesis was induced with the addition of 1 mmol/L isopropyl--D-thiogalactopyranoside (IPTG). The cell pellets were prepared, and they were lyzed in Tris-dithiothreitol (Tris-DTT) buffer (20 mmol/L Tris, pH 7.4, 5 mmol/L ethylenediaminetetraacetic acid (EDTA), 150 mmol/L sodium chloride, 1 mmol/L DTT), containing 1% Triton X-100, 10 mmol/L benzamidine, 10 mmol/L -amino-n-caproic acid, and 2 mmol/L phenylmethanesulfonyl fluoride. The lysates were centrifuged at 20,000 g at 4°C for 30 minutes. The supernatants were applied to 10 mL lactosyl-Sepharose column (Sigma, St. Louis, MO, USA). The unbound protein was washed, and fusion proteins were eluted with Tris-DTT buffer containing 200 mmol/L lactose. The eluted fractions were collected, dialyzed against phosphate-buffered saline (PBS) containing 1 mmol/L DTT, and kept in -70°C for further use. The samples were subjected to 12.5% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and all galectins gave single band with proper molecular weight⁵.

Induction of NTS nephritis

Nephrotoxic serum was injected intraperitoneally (2 mL/kg weight) into 40 WKY rats at day 1. WKY rats were then treated either with PBS containing 1 mmol/L DTT (N = 8) or dexamethasone (DEX; 0.17 mg/kg weight, N = 8), G1 (1 mg/kg weight, N = 8), G3 (1 mg/kg weight, N = 8) or G9 (1 mg/kg weight, N = 8) on alternate days for two weeks. Urine collection was carried out, and 24-hour urinary protein excretion was measured. The rats were sacrificed at day 14, and their kidney tissues were processed for light and immunofluorescent microscopy and for in situ terminal deoxytransferase (TdT)-mediated uridine triphosphate (dUTP) nick end labeling (TUNEL) reaction to assess the extent of apoptosis. In separate experiments, the additional 20 NTS nephritis and 20 normal WKY rats were prepared and treated with PBS containing 1 mmol/L DTT, DEX, G1, G3, and G9. The 10 groups (N = 4 each) were sacrificed at day 8, and their tissues were used for immunoperoxidase staining of CD4⁺ and CD8⁺ cells and fluorescence-activated cell sorter (FACS) analysis. Since the peak of the infiltration of CD8⁺ cells was observed at day 3 and gradually declined up to day 11, the animals were sacrificed at day 8.

Light microscopy

Tissues were fixed in 10% formaldehyde and embedded in paraffin, and 3 m thick sections were prepared. The sections were stained with periodic acid-Schiff (PAS). The diameter of 50 glomeruli of each animal were measured, and the number of total glomerular cells and the glomeruli with crescents were counted.

Immunofluorescence

Four-m thick cryostat sections were incubated with FITC-labeled goat anti-rabbit IgG or goat anti-rat IgG, goat anti-rat C3, or goat anti-rat fibrinogen (Cappel, Costa Mesa, CA, USA). The staining intensity was semiquantitatively graded as follows: (-) = no staining; (+) = weak staining; (++) = moderate staining; (+++) = strong staining; (++++) = very strong staining. In addition, the 2 m thick serial cryostat sections were prepared. For the staining for the proliferation marker Ki-67, the sections were fixed with acetone on ice and boiled for 10 minutes. They were incubated with mouse monoclonal antibody against Ki-67 (Dako, Glostrup, Denmark) for two hours. The adjacent sections were incubated with mouse monoclonal anti-rat monocytes/macrophages (ED1; Serotec, Oxford, UK) for one hour and mouse monoclonal anti–-smooth muscle actin (-SMA; Sigma) for one hour. They were finally incubated with FITC-conjugated anti-mouse IgG (Cappel) for 30 minutes. The numbers of Ki-67, ED1, and -SMA–positive cells per glomerular cross section were counted. Twenty glomeruli per rat were examined, and the average number of the positive cells in glomeruli was calculated.

Immunoperoxidase staining

The degree of influx of leukocytes into the glomeruli was assessed by using immunoperoxidase ABC kit (Vector Laboratories, Burlingame, CA, USA), as described previously²¹. Briefly, cryostat sections were first incubated for 60 minutes with mouse monoclonal antibodies against rat leukocytes (OX1), or rat monocytes/macrophages (ED1), rat CD4, or rat CD8 (Serotec Ltd.). The sections were then incubated with biotinylated horse anti-mouse IgG for 30 minutes at 22°C, followed by treatment with 3,3-diamino-benzidine and hydrogen peroxide. The cells with brownish reaction product were enumerated, and the number of OX1, ED1, CD4, and CD8 positive cells per glomerular cross section were counted. Twenty glomeruli per rat were examined, and the average number of positive cells in glomeruli was calculated.

In situ TUNEL assay in renal tissues

Apoptotic cells in renal tissues were visualized by in situ TUNEL method using FITC-dUPT (In Situ Cell Death Detection Kit; Boehringer Mannheim, Mannheim, Germany). Four micrometer thick cryostat sections of the kidney were prepared and fixed with 4% paraformaldehyde for 20 minutes at 22°C. Sections were then permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate for two minutes on ice. The sections were then immersed in the TUNEL reaction mixture and incubated in a humidified chamber for 60 minutes in dark. Fifty glomeruli per in each WKY rats were examined, and apoptotic cells in glomeruli were counted.

Enzyme-linked immunosorbent assays for circulating anti-rabbit IgG antibody

The titer of anti-rabbit IgG in rat serum was determined by enzyme-linked immunosorbent (ELISA). A 96-well plate was used, and each well was incubated with 50 L of normal rabbit IgG (10 mg/m; Chemicon, Temecula, CA, USA) overnight at 22°C. The next day, 50 L of 1:40-diluted serum (collected from rats at day 14) was added to the wells, and incubation was carried for two hours. This was followed by the addition of alkaline phosphatase-labeled rabbit anti-rat IgG (Southern Biotechnology Associates, Birmingham, AL, USA) at a dilution of 1:2000 and incubation for two hours. Finally, a p-nitrophenyl phosphate substrate solution (Sigma) was added and incubated for one hour, and absorbance readings at 405 nm, using the Microplate Reader (Japan Bio-Rad, Tokyo, Japan), were recorded.

FACS analysis

Mononuclear cells, isolated from spleen, were subjected to CD4 or CD8 staining and TUNEL reaction. First, spleens were homogenized and suspended in RPMI-1640 (GIBCO BRL, Grand Island, NY, USA). They were overlaid onto endotoxin-free Ficoll-Plaque (Pharmacia Biotech, Uppsala, Sweden) and centrifuged at 2000 r.p.m. for 20 minutes. Mononuclear cells were collected and washed two times in PBS at 4°C and adjusted to 2 10⁷ cells/mL. Cells were then stained with phycoerythrin-conjugated anti-rat CD4 antibody (Serotec Ltd.) or phycoerythrin-conjugated anti-rat CD8 antibody (Antigenix America Inc., Franklin Square, NY, USA) and fixed with 100 L of freshly prepared paraformaldehyde solution (4% in PBS, pH 7.4). Fixative was removed after centrifugation, and the cells were permeabilized with 0.1% Triton X-100 in 0.1% sodium citrate for two minutes on ice. Cells were rewashed twice with PBS and suspended in 50 L of TUNEL reaction mixture containing fluorescein-dUTP and TdT (In Situ Cell Death Detection Kit). After incubation in a humidified chamber in dark for 60 minutes at 37°C, cells were washed twice in PBS and adjusted to a final volume of 500 L. Finally, they were analyzed by two-color flow cytometry (FACS Caribur 3A; Becton Dickinson, Mountain View, CA, USA), and the percentage of apoptotic cells in the subpopulation of T cells was determined.

Distribution of exogenously injected galectin-9 in the kidney tissue

Galectin-9 (1 mg/kg weight) was injected into normal WKY rats, and they were sacrificed after four hours. The cryostat sections of kidneys were prepared, and they were incubated with anti-myc antibody (Invitrogen). They were then stained with rhodamine-labeled horse anti-mouse IgG (Vector Laboratories). They were further double stained with rabbit anti-G9 antibody^5,6 and FITC-labeled donkey anti-rabbit IgG (Chemicon). Secondary antibodies were already absorbed against serum protein, which minimized the cross reactivity to IgG of other species. Only exogenously injected fusion protein carrying the C-terminal myc was detected by anti-myc antibody, since the intrinsic G9 lacked the myc epitope.

Statistical analysis

The data were expressed as mean SD. Significance of difference (P values of less than 0.05) was determined by analysis of variance (ANOVA). Fischer's analysis was performed.

RESULTS

Urinary protein excretions

The control rats injected with normal rabbit serum showed minimal 24-hour urinary protein excretion (1.9 0.4 mg/day). The rats injected with NTS revealed minimal proteinuria during the first week, but in the second week, the urinary protein excretion progressively increased and was 35.8 7.1 mg/day at day 14. Daily urinary protein excretions were significantly suppressed in rats treated with DEX or galectins. The DEX treatment drastically reduced the urinary protein excretion (2.3 0.4 mg/day). Similarly, the treatment with G1, G3, and G9 resulted in a significant reduction of proteinuria, and respective values for 24-hour excretions were 5.6 1.8, 11.6 2.3, 9.5 4.5 mg/day Figure 2.

Figure 2.

Urinary protein excretion in Wistar Kyoto (WKY) rats with nephrotoxic serum (NTS) nephritis. Daily urinary protein excretions at day 14 were significantly suppressed by treatment of galectins (A) G1, (B) G3, or (C) G9, and (D) dexamethasone (DEX). DEX treatment reduced the urinary protein excretion (2.3 0.4 mg/day). The treatment of G1, G3, and G9 yielded a similar antiproteinuric effect (5.6 1.8, 11.6 2.3, and 9.5 4.5 mg/day, respectively). Symbols are: () PBS, WKY rats with NTS, and PBS containing 1 mmol/L DTT; () G1, G3 and G9, NTS nephritis WKY rats receiving G1, G3, and G9 in panels A-C, respectively; () DEX, NTS nephritis WKY rats treated with DEX (D); *P < 0.01 vs. PBS.

Full figure and legend (35K)

Light microscopy

Light microscopic examination of rat kidneys with NTS nephritis revealed hypertrophy and hypercellularity of most of the glomeruli Table 1. About 56% of the glomeruli had cellular crescents (Table 1 and Figure 3d), and severe necrotizing lesions were also observed in the glomeruli. By immunofluorescence microscopy, the glomeruli showed linear and granular capillary deposition of rat IgG and granular deposits of rat C3 and fibrin (Figure 3 a–c and Table 2).

Figure 3.

Immunofluorescence and light micrographs of kidney specimens from WKY rats with NTS nephritis. By immunofluorescence microscopy, the glomeruli of NTS nephritis rats reveal a capillary linear deposition of rat IgG and granular deposition of rat C3 and of fibrin (a–c). By light microscopy, glomeruli with cellular crescents are seen (d). In the DEX-treated group, the deposition of rat IgG, C3, and fibrin is completely inhibited (q–s), and the glomeruli show no morphologic abnormality (t). In galectin-treated groups (G1, G3, and G9), the deposition of rat IgG and C3 is not inhibited in the glomeruli (e, f, i, j, m, and n). However, the deposition of fibrin is notably inhibited (g, k, and o), and also the glomeruli show minimal proliferation of mesangial cells (h, l, and p). PBS is defined as WKY rats with NTS, and PBS containing 1 mmol/L DTT; G1, G3, and G9, NTS nephritis WKY rats receiving galectins-1, -3, and -9, respectively; DEX, NTS nephritis WKY rats treated with dexamethasone; bar, 100 m.

Full figure and legend (158K)

Table 1 - Glomerular hypertrophy and proliferation of various groups of Wistar Kyoto (WKY) rats.

Full table

Table 2 - Staining score of immunofluorescence in renal glomeruli of various groups of WKY rats.

Full table

In the DEX-treated group, less than 3% of the glomeruli had crescents, and no significant proliferation of cells or glomerular hypertrophy was observed Figure 3t. Also, the deposition of rat IgG, C3, and fibrin in the glomeruli was minimal (Fig. q, r, s and Table 2).

Galectin-treated rats (G1, G3, and G9) showed morphological changes similar to the DEX-treated group by immunofluorescence and light microscopy. Crescents were seen in 5 to 8% of the glomeruli, however, the hypercellularity of glomeruli was not observed Figures 3 e–t, and the deposition of rat IgG and C3 was not altered in glomeruli Table 2.

In all control and experimental groups, linear rabbit IgG in glomeruli was not altered (not shown).

Quantitative study of infiltrating cells and proliferation in the glomeruli

Immmunoperoxidase reaction was used to detect the presence of inflammatory cells in the glomerulus. In rats with NTS nephritis, at day 8, the number of CD8⁺ and CD4⁺ cells was 3.5 0.8/glomerulus and 0.9 0.2/glomerulus, respectively Figure 4a. At day 14, leukocyte infiltration into the glomeruli increased dramatically (28.3 3.4/glomerulus), and most of the cells were macrophages (25.0 2.6/glomerulus; Figure 4b). In the DEX-treated group, the infiltration of CD8⁺ cells decreased to 0.6 0.2/glomerulus and of macrophages to 2.3 0.1/glomerulus. G1 and G3 treatment failed to inhibit the infiltration of CD8⁺ cells; however, they significantly inhibited the infiltration of macrophages into glomeruli. In contrast, G9 treatment suppressed the infiltration of CD8⁺ cells to 1.2 0.2/glomerulus.

Figure 4.

Quantitative analyses of infiltrating cells in the glomeruli of WKY rats. (A) In NTS nephritis WKY rats, the influx of () CD8⁺ cells was noted in the glomeruli, while that of () CD4⁺ cells was relatively less at day 8. (B) At day 14, prominent leukocyte infiltration into the glomeruli was observed, and most of them were macrophages. Symbols are: () ED1, () OX1. DEX treatment inhibits the infiltration of both CD8⁺ cells and macrophages. G1 and G3 treatment failed to inhibit the infiltration of CD8⁺ cells; however, they significantly inhibited the infiltration of macrophages into glomeruli. In contrast, G9 treatment suppressed the infiltration of CD8⁺ cells and macrophages. Definitions of the groups are in the legend to Figure 3; *P < 0.01 vs. PBS.

Full figure and legend (17K)

The proliferation marker, that is, Ki-67, and cell type marker, ED1, were examined on the serial sections of kidney tissues. The immunoreactivity of Ki-67 in the glomeruli of PBS-treated NTS nephritis rats was seen on the nuclei, and most of them were also positive for ED1, that is, the marker for monocytes/macrophages Figure 5a, a and b. The ED1- and Ki-67–positive cells were abundant in PBS-treated NTS nephritis rats (24 8 and 22 7/glomerulus, respectively), and they were significantly suppressed in G1, G3, G9, and DEX-treated rats. In contrast, the distribution of -SMA was distinct from that of ED1 and Ki-67, and only a part of mesangial cells was positive for -SMA in PBS-treated NTS nephritis rats Figure 5a, c. The result indicated that mesangial cells were not actively proliferating, and the accumulation of the glomerular cells in NTS nephritis was due to the influx of macrophages into glomeruli and their proliferation.

Figure 5.

Immunofluorescence micrographs of kidney specimens and quantitative analysis of proliferating cells in glomeruli of WKY rats with NTS nephritis. (A) Immunostaining of serial sections indicated that Ki-67–positive cells were seen in the glomeruli of PBS-treated NTS nephritis rats (B), and most of them were also positive for ED1 (a), which is the marker for monocytes/macrophages. In contrast, the distribution of -SMA was distinct from that of ED1 and Ki-67, and only a part of mesangial cells was positive for -SMA in PBS-treated NTS nephritis rats (c). The smooth muscle cells of small artery was also positive for -SMA. (B) The () ED1- and () Ki-67–positive cells were abundant in PBS-treated NTS nephritis rats, and they were significantly suppressed in G1, G3, G9, and DEX-treated rats. (Bar, 100 m).

Full figure and legend (49K)

Circulating anti-rabbit IgG levels by ELISA

Dexamethasone treatment abolished the production of anti-rabbit IgG antibody. However, the administration of galectins did not affect the antibody production Figure 6.

Figure 6.

Circulating antirabbit IgG levels by ELISA. DEX treatment abolished the production of antirabbit IgG antibody. However, the administration of galectins did not affect the antibody production. Definitions of the groups are in the legend to Figure 3; *P < 0.01 vs. PBS.

Full figure and legend (6K)

In situ TUNEL assay of kidney tissue

In PBS-treated NTS nephritis rats, few apoptotic cells were seen in the glomeruli at day 14 Figure 7a. No increase of apoptotic cells was observed in kidney tissues of NTS nephritis rats treated either with DEX or galectins Figure 7b.

Figure 7.

In situ TUNEL assay of kidney tissues. (A) In NTS-treated control rat, TUNEL-positive apoptotic cells can be seen in the glomerulus (arrow; bar, 50 m). (B) Administration of DEX and galectins did not alter the number of apoptotic cells present in the glomeruli. Definitions of the groups are in the legend to Figure 3.

Full figure and legend (81K)

Apoptosis assay of T cells isolated from spleen

In PBS-treated normal rats, approximately 7% of both the CD4⁺ cells and CD8⁺ cells, isolated from spleen, underwent spontaneous apoptosis. In DEX-treated normal rats, the apoptosis was observed in approximately 25% of the CD4⁺ cells and approximately 28% of the CD8⁺ cells. No significant increase of apoptosis was observed in the galectin-treated normal rats (Table 3 and Figure 8).

Figure 8.

Apoptosis assay of T cells isolated from spleen. In DEX-treated normal rats, apoptosis was induced in approximately 25% of the CD4⁺ cells and in approximately 28% of the CD8⁺ cells. In DEX-treated NTS rats, a similar degree of apoptosis was seen in the CD4⁺ (20%) and CD8⁺ (26%) cells. Although G9 did not induce apoptosis in CD4⁺ and CD8⁺ cells in normal rats, it did induce apoptosis in CD8⁺ cells (17%) in NTS rats. Definitions of the groups are in the legend to Figure 3.

Full figure and legend (51K)

Table 3 - Apoptosis of T-cells isolated from spleen of WKY rats.

Full table

In NTS-treated rats, approximately 6% of the CD4⁺ cells and approximately 5% of the CD8⁺ cells isolated from spleen underwent spontaneous apoptosis, and the ratio of apoptotic cells was similar to that of normal PBS-treated rats. The DEX treatment increased apoptosis to approximately 20% for the CD4⁺ cells and approximately 26% for the CD8⁺ cells, which was similar to that observed in PBS-treated normal rats. G1 and G3 induced a basal level of apoptosis (5 to 9%) both in CD4⁺ cells and CD8⁺ cells; however, G9 induced an accentuated apoptotic response in CD8⁺ cells (approximately 17%), while the apoptosis in CD4⁺ cells remained close to the basal level (9%). These data indicate that DEX treatment induced apoptosis in both resting and activated T cells, while G9 induced a selective apoptotic response in the activated CD8⁺ splenic T cells in the NTS-treated rats.

Distribution of exogenously injected galectins in the kidney tissue

The exogenously injected G9 was detected with anti–c-myc antibody. Exogenous G9 was distributed on endothelial cells of glomeruli and peritubular capillaries (Figure 9, red). Somehow, partial population of endothelial cells of peritubular and glomerular capillaries revealed positive immunoreactivity. The immunoreactivity of intrinsic G9 was detected on glomerular cells, and it was distributed on both endothelial and mesangial cells (Figure 9, green) as reported previously⁶. The double-staining method indicated that exogenously injected G9 dominantly located glomerular capillary endothelial cells, and the distribution pattern was different from the localization of intrinsic G9.

Figure 9.

Distribution of intrinsic and exogenously injected G9. Normal WKY rats were injected with G9, which is the fusion protein with N-terminal c-myc and (His)₆ epitopes. In kidney tissues, exogenous G9 was detected with monoclonal anti-myc antibody and rhodamine-labeled horse anti-mouse IgG (red). Intrinsic G9 was detected with rabbit anti-G9 antibody^5,6 and FITC-labeled donkey anti-rabbit IgG (green). Exogenous G9 was detected on glomerular endothelial cells and endothelial cells of peritubular capillaries (red). The immunoreactivity of intrinsic G9 was also detected on glomerular cells, and it distributed on endothelial cells and mesangial cells.

Full figure and legend (169K)

DISCUSSION

The prevalence of specific interactions between lectins and oligosaccharides expressed on the plasmalemma of the thymus-derived T cells suggests a potential role of galectins in the biology of the mammalian immunoregulatory system. Among the various galectins, the role of G1, G3, and G9 has been certainly implicated in various processes related to the immune response. G1 induces apoptosis of phytohemagglutinin (PHA)-activated peripheral T cells¹⁶ and of the thymocytes¹⁷. Such an induction by the galectin seems to be specific and is conceivably mediated via its sugar-binding moiety on apoptosis-related receptors (death receptors)²² since apoptosis can be inhibited by lactose^16,17. It is likely that G9 as well induces apoptosis of thymocytes by a -galactoside-specific binding pathway⁵. The human homologue of the latter galectin was isolated by screening a Hodgkin's lymphoma cDNA library with a circulating autoantibody of patients with lymphoma²³, while the murine homologue was isolated from the embryonic mouse kidney⁵. Although G9 is widely distributed in various tissues, it is heavily expressed in lymphatic tissues, especially in thymic epithelial cells⁵, thus suggesting a functional relationship between these two galectins. G1 is a homodimer with carbohydrate-binding domains, and G9 is a single peptide chain with two carbohydrate-binding domains. By virtue of having two carbohydrate-binding domains, both the galectins, that is, G1 and G9, can cross-link two glycoconjugates on the cell surface. Whereas G3, which is a monomer that can self-associate at its N- or C-terminal domain, can mediate cross-linking of glycoprotein counter-receptor similar to G1 and G9^24,25. In addition, G3 is also expressed in the cytoplasm and exerts an anti-apoptotic effect in T cells. Such an effect may be related to the NWGR motif in G3 since a similar motif is present in bcl-2, a molecule that is known to have an anti-apoptotic property¹⁸.

Among the various human glomerular diseases, the crescentic form of glomerulonephritis has a rapid and relentless progression of the disease that results in renal failure unless aggressive immunosuppressive therapy is instituted^26,27. Similar to the human form crescentic glomerulonephritis, the disease can be induced in WKY rats by a single injection of anti-GBM antibody^19,20. Besides the crescentic change, the glomerular lesions include the infiltration of CD8⁺ cells and macrophages into glomeruli¹⁹. The latter may be due to the CD8⁺ cell-induced up-regulated expression of intercellular adhesion molecule-1 (ICAM-1) in the endothelial cells, which facilitate the recruitment of macrophages into glomeruli²⁰. The severity of the disease is further accentuated by production of host anti-rabbit IgG antibody and its subsequent implantation onto the GBM. The evolution of these lesions of crescentic glomerulonephritis can be suppressed with the DEX treatment. The latter induced apoptosis in both CD4⁺ and CD8⁺ cells and abolished the anti-rabbit IgG antibody production and the influx of CD8⁺ cells and macrophages into the glomeruli. The apoptotic effect of DEX was dramatic, but it was not selective since it affected both the CD4⁺ and CD8⁺ cells even in normal WKY rats. In contrast, G9 selectively induced apoptosis of CD8⁺ cells and reduced the infiltration of CD8⁺ cells and of macrophages and ultimately the formation of crescents. G1 was expected to induce apoptosis of activated T cells but did not do so. This differential apoptotic effect observed in our investigation may be attributable to the difference in their structural characteristics. The G9, endowed with two carbohydrate-binding domains in a single polypeptide chain, does not need to undergo dimerization in order to cross-link with the ligands and thus can induce apoptosis in vitro at a relatively low concentration, that is, 1 mol/L⁶. On other hand, G1 induces a comparable degree of apoptosis in vitro at a concentration of 7 mol/L^16,28. Also, G1 needs to undergo homodimerization in order to cross-link the ligands and thereby to induce comparable degree of apoptosis¹⁶.

Although G1 and G3 failed to inhibit apoptosis of CD8⁺ cells, influx of CD8⁺ cells and the antibody production, it suppressed the macrophage infiltration, crescent formation, and urinary excretion of proteins. Here, one may speculate that the therapeutic effect may be mediated by several different mechanisms, including modulation of cell–cell or cell–matrix adhesion and proliferation. G1 and G3 are known to mediate cell–cell or cell–matrix adhesive interactions; however, the findings have not been consistent³. G1 promotes the adhesion of ovarian carcinoma cells to ECM²⁹, whereas it inhibits the adhesion of myoblast to laminin³⁰. Similarly, G3 mediates adhesion of the neutrophils³¹, but not of the melanoma cells to laminin^32,33. The latter binary action of G3 may be related to its concentration and expression and to the glycosylation of the counter-receptors³. Albeit these divergent actions, in general galectins seem to cross-link glycoconjugates of different cells or of cell and ECM to facilitate adhesion. However, in the presence of the high concentration of galectins, the cell surface glycoproteins on the same cells get cross-linked, and the divalent carbohydrate-binding domains of galectins become occupied. As a result, they lose their adhesive potential³³. In any instance, the inhibitory effect of G1 and G3 on the macrophage influx in NTS-nephritis as observed in WKY rats can be explained on the basis of their strategic expression within the glomerulus and the circulating inflammatory cells. G1 is expressed in endothelial cells, and it is up-regulated by interleukin-1, transforming growth factor-, and interferon-³⁴, while G3 is expressed in the monocytes and macrophages³⁵. It can be speculated here that the administered exogenous galectins reduced the adhesiveness by cross-linking the glycoproteins expressed on the same cell, that is, endothelium or macrophage, thus reducing the residency of macrophages within the renal glomerulus. Immunohistochemical detection of exogenous galectins on glomerular endothelial cells supports this notion. In addition to their adhesion potential, the galectins also modulate cell proliferation as indicated earlier. G1 has been found to promote as well as inhibit cell proliferation. Its overexpression has been reported to induce transformation of 3T3 fibroblasts³⁶, while the addition of G1 in the culture media leads to the inhibition in the replication of mouse embryonic fibroblasts³⁷. Such biphasic response in cell growth seems to be concentration dependent³⁸. Thus, it is conceivable that the exogenous administration of galectins inhibited the macrophage infiltration in WKY rats, which ultimately ameliorate the glomerular injuries.

Although G9 is normally expressed in glomeruli both on endothelial and mesangial cells⁶, the administration of large amounts of galectins unexpectedly inhibit the macrophage infiltration to the glomeruli. Since administered galectins dominantly distributed on glomerular endothelial cells, the presence of higher concentration of galectins on the endothelial cells may alter the adhesiveness of the cells and may exert an inhibitory action. The unexpected suppression of macrophage infiltration into the glomeruli suggests that such a pharmacological effect may be common to all members of the galectin family, and another potential use of the galectins as therapeutic tools would be in their employment for the modulation of macrophage infiltration and proliferation in various diseases.

References

1. Barondes, SH, Castronovo, V, Cooper, DNW, Cummings, RD, Drickamer, K, Feizi, T, Gitt, MA, Hirabayashi, J, Hughes, C, Kasai, K, Leffler, H, Liu, F-T, Lotan, R, Mercurio, AM, Monsigny, M, Pillai, S, Poirer, F, Raz, A, Rigby, PWJ, Rini, JM & Wang, JL: Galectins: A family of animal -galactoside-binding lectins. Cell 1994 76:597–598,  | Article | PubMed | ISI | ChemPort |
2. Barondes, SJ, Cooper, DNW, Gitt, MA & Leffler, H: Galectins. J Biol Chem 1994 269:20807–20810,  | PubMed | ISI | ChemPort |
3. Perilo, NL, Marcus, ME & Baum, LG: Galectins: Versatile modulators of cell adhesion, cell proliferation, and cell death. J Mol Med 1998 76:402–412,  | Article | PubMed | ISI | ChemPort |
4. Varki, A: Selectin ligands. Proc Natl Acad Sci USA 1994 91:7390–7397,  | PubMed | ChemPort |
5. Wada, J & Kanwar, YS: Identification and characterization of galectin-9, a novel -galactoside-binding mammalian lectin. J Biol Chem 1997 272:6078–6086,  | PubMed | ISI | ChemPort |
6. Wada, J, Ota, K, Kumar, A & Wallner El Kanwar, YS: Developmental regulation, expression, and apoptotic potential of galectin-9, a -galactoside binding lectin. J Clin Invest 1997 99:2452–2461,  | PubMed | ISI | ChemPort |
7. Varki, A & Marth, J: Oligosaccharide in vertebrate development. Dev Biol 1995 6:127–138,  | ChemPort |
8. Orlandi, F, Saggiorate, E, Pivano, G, Puligheddu, B, Termine, A, Cappia, S, Degiuli, P & Angeli, A: Galectin-3 is a presurgical marker of human thyroid carcinoma. Cancer Res 1998 58:3015–3020,  | PubMed | ISI | ChemPort |
9. Bresalier, RS, Mazurek, N, Sternberg, LR, Byrd, JC, Yunker, CK, Nangia-Makker, P & Raz, A: Metastasis of human colon cancer is altered by modifying expression of the -galactoside-binding protein galectin-3. Gastroenterology 1998 115:287–296,  | PubMed | ISI | ChemPort |
10. Vlassara, H, Li, YM, Imani, F, Wojciechowicz, D, Yang, Z, Liu, F-T & Cerami, A: Identification of galectin-3 as a high affinity binding protein for advanced glycation end products (AGE): A new member of the AGE receptor complex. Mol Med 1995 1:634–646,  | PubMed | ISI | ChemPort |
11. Dagher, SF, Wang, JL & Patterson, RJ: Identification of galectin-3 as a factor in pre-mRNA splicing. Proc Natl Acad Sci USA 1995 92:1213–1217,  | PubMed | ChemPort |
12. Frigeri, LG, Robertson, MW & Liu, F-T: Expression of biologically active recombinant rat IgE-binding protein on E. coli. J Biol Chem 1990 265:20763–20769,  | PubMed | ISI | ChemPort |
13. Nachtigal, M, Al-Assaad, Z, Mayer, EP, Kim, K & Monsigny, M: Galectin-3 expression in human atherosclerosis. Am J Pathol 1998 152:1199–1208,  | PubMed | ISI | ChemPort |
14. Levi, G, Tarrab-Hazdai, R & Teichberg, VI: Prevention and therapy with electrolectin of experimental autoimmune myasthenia gravis in rabbits. Eur J Immunol 1983 13:500–507,  | PubMed | ISI | ChemPort |
15. Offner, H, Celnik, B, Bringman, TS, Casentini-Borocz, D, Nedwin, GE & Vandenbark, AA: Recombinant human -galactoside binding lectin suppresses clinical and histological signs of experimental autoimmune encephalomyelitis. Neuroimmunology 1990 28:177–184,  | ChemPort |
16. Perillo, NL, Pace, KE, Seilhamer, JJ & Baum, LG: Apoptosis of T cells mediated by galectin-1. Nature 1995 378:736–739,  | Article | PubMed | ISI | ChemPort |
17. Perillo, NL, Uittenbogaart, CH, Nguyen, JT & Baum, LG: Galectin-1, an endogenous lectin produced by thymic epithelial cells, induces apoptosis of human thymocytes. J Exp Med 1997 185:1851–1858,  | Article | PubMed | ISI | ChemPort |
18. Yang, R-Y, Hsu, DK & Liu, F-T: Expression of galectin-3 modulated T-cell growth and apoptosis. Proc Natl Acad Sci USA 1996 93:6737–6742,  | Article | PubMed | ChemPort |
19. Kawawsaki, K, Yaoita, E, Yamamoto, T & Kihara, I: Depletion of CD8 positive cells in nephrotoxic serum nephritis of WKY rats. Kidney Int 1992 41:1517–1526,  | PubMed | ISI | ChemPort |
20. Fujinaka, H, Yamamoto, T, Feng, L, Kawasaki, K, Yaoita, E, Hirose, S, Goto, S, Wilson, CB, Uchiyama, M & Kihara, I: Crucial role of CD8-positive lymphocytes in glomerular expression of ICAM-1 and cytokines in crescentic glomerulonephritis of WKY rats. J Immunol 1997 158:4978–4983,  | PubMed | ISI | ChemPort |
21. Kushiro, M, Shikata, K, Sugimoto, H, Shikata, Y, Miyatake, N, Wada, J, Miyasaka, M & Makino, H: Therapeutic effects of protacyclin analog on crescentic glomerulonephritis in rat. Kidney Int 1998 53:1314–1320,  | Article | PubMed | ISI | ChemPort |
22. Ashkenazi, A & Dixit, VM: Apoptosis control by death and decoy receptors. Curr Opin Cell Biol 1999 11:255–260,  | Article | PubMed | ISI | ChemPort |
23. Türeci, Ö, Schmitt, H, Fadle, N, Pfreundschuh, M & Sahin, U: Molecular definition of a novel galectin which is immunogenic in patients with Hodgkin's disease. J Biol Chem 1997 272:6416–6422,  | PubMed | ISI | ChemPort |
24. Inohara, H & Raz, A: Functional evidence that cell surface galectin-3 mediate homotypic cell adhesion. Cancer Res 1995 55:3267–3271,  | PubMed | ISI | ChemPort |
25. Yang, R-Y, Hill, PN, Hsu, DK & Liu, F-T: Role of carboxyl-terminal lectin domain in self-association of galectin-3. Biochemistry 1998 37:4086–4092,  | Article | PubMed | ISI | ChemPort |
26. Holzman, LB & Wiggins, RC: Consequence of glomerular injury. Glomerular crescent formation. Semin Nephrol 1991 11:346–353,  | PubMed | ISI | ChemPort |
27. Bolton, WK: Rapidly progressive glomerulonephritis. Semin Nephrol 1996 16:517–526,  | PubMed | ISI | ChemPort |
28. Cho, M & Cummings, RD: Galectin-1, a -galactoside-binding lectin in Chinese hamster ovary cells. I. Physical and chemical characterization. J Biol Chem 1995 270:5198–5206,  | Article | PubMed | ISI | ChemPort |
29. Allen, HJ, Sucato, D, Woynarowska, B, Gottstine, S, Sharma, A & Bernacki, RJ: Role of galaptin in ovarian carcinoma adhesion to extracellular matrix in vitro. J Cell Biol 1990 43:43–57,  | ChemPort |
30. Cooper, DNW, Massa, SM & Barondes, SH: Endogenous muscle lectin inhibits myoblast adhesion to laminin. J Cell Biol 1990 115:1437–1448,  | ISI |
31. Kuwabara, I & Liu, F-T: Galectin-3 promotes adhesion of human neutrophils to laminin. J Immunol 1996 156:3939–3944,  | PubMed | ISI | ChemPort |
32. van den Brule, FA, Buicu, C, Sobel, ME, Liu, F-T & Castronovo, V: Galectin-3, a laminin binding protein, fails to modulate adhesion of human melanoma cells to laminin. Neoplasma 1995 42:215–219,  | PubMed | ChemPort |
33. Ochieng, J, Leite-Browning, ML & Warfield, P: Regulation of cellular adhesion extracellular matrix proteins by galectin-3. Biochem Biophys Res Commun 1998 246:788–791,  | Article | PubMed | ISI | ChemPort |
34. Baum, LG, Seilhamer, JJ, Pang, M, Levine, WB, Beynon, D & Berliner, JA: Synthesis of an endogenous lectin, galectin-1 by human endothelial cells is up-regulated by endothelial cell activation. Glycoconj J 1995 12:63–68,  | Article | PubMed | ISI | ChemPort |
35. Liu, F-T, Hsu, DK, Zuberi, RI, Kuwabara, I, Chi, EY & Henderson, WRJ: Expression and function of galectin-3, a -galactoside-binding lectin in human monocytes and macrophages. Am J Pathol 1995 147:1016–1028,  | PubMed | ISI | ChemPort |
36. Yamaoka, K, Ohno, S, Kawasaki, H & Suzuki, K: Overexpression of a -galactoside-binding protein causes transformation of BALB 3T3 fibroblast cells. Biochem Biophys Res Commun 1991 179:272–279,  | Article | PubMed | ISI | ChemPort |
37. Wells, V & Mallucci, L: Identification of an autocrine negative growth factor: Mouse -galactoside-binding protein is a cytostatic factor and cell growth regulator. Cell 1991 64:91–97,  | Article | PubMed | ISI | ChemPort |
38. Adams, L, Scott, GK & Weinberg, CS: Biphasic modulation of cell growth by recombinant human galectin-1. Biochim Biophys Acta 1996 1312:137–144,  | Article | PubMed | ISI | ChemPort |

Acknowledgments

This work was supported by the Uehara Memorial Foundation, the Naito Foundation, the ONO Medical Foundation, and a Grant-in-Aid for Encouragement of Young Scientists, Ministry of Education, Science and Culture, Japan (10770199) to J. Wada; a National Institutes of Health grant DK 28492 to Y.S. Kanwar; a Grant-in-Aid for Scientific Research (B), Ministry of Education, Science and Culture, Japan (11470218); and the Uehara Memorial Foundation to H. Makino. H. Zhang was supported by International Society of Nephrology/Kirin Fellowship Award.