Age-associated changes in renal glomeruli of mice

Experimental Gerontology, Vol. 24, pp. 237-249, 1989

0531-5565/89 $3.00 + .00 Copyright© 1989PergamonPressplc

Printed in the USA. All rightsreserved.

AGE-ASSOCIATED

CHANGES

IN RENAL

GLOMERULI

OF MICE

WAKO YUMURA,I'2 NOBUHIRO SUGINO,2 RYUJI NAGASAWA,1,3 SACHIHOKUBO,t KATSUIKU HIROKAWAl and NAOKI MARUYAMA1 1Department of Pathology, Tokyo Metropolitan Institute of Gerontology, 2Department of Internal Medicine, Tokyo Women's Medical College and 3Department of Internal Medicine, Saitama Medical College, Tokyo, Japan

Abstract -- To investigate age-associated changes in renal glomeruli of C57BL/6 female mice, we used a single radial immunodiffusion method to measure albumin excretion. Up to 100 mg/dl in urine samples was regarded as microalbuminuria. The mean amount of urinary albumin increased from 14.0 mg/dl at 6 months to 151.1 mg/dl at 24 months of age. Microalbuminuria occurred in 64.6% of tested mice by the time they were 24 months old, and 10% of the mice had marked albuminuria (more than 100 mg/dl) at that time. Parallel morphological study showed that renal mesangial changes were also age-dependent. Mesangial cell proliferation and spike lesions in glomerular capillary walls appeared in aged mice with microalbuminuria, and were then followed by diffuse glomerular sclerosis accompanied by marked albuminuria. Histological scores on damage in the renal mesangium with changes of glomerular basement membrane increased significantly with age from a mean score of 0 at 6 months to 3.24 at 24 months of age. Immunofluorescent study showed a marked deposition of IgG and IgM, but no complement component C3 in enlarged mesangium. Electron microscopic examination of diffuse sclerotic glomeruli in aged mice revealed amyloid substances. These results suggest that assays of albuminuria could be a useful method for early detection of age-associated renal deterioration. Key Words: proteinuria, aibuminuria, aging, renal glomerular sclerosis, renal amyloidosis INTRODUCTION OF THE many functional and/or morphological changes o f human organs that appear with aging, some of the more important are thymic involution, senile dementia, atherosclerosis, and deterioration of renal function, the latter being critical to physiological homeostasis. The cause of this renal degeneration might be age-associated glomerular sclerosis (Kirschbaum et al., 1949; Gude and Lupton, 1960; Elema and Arends, 1975; Hirokawa, 1975; Bolton et al., 1976), and microalbuminuria has been regarded as a convenient early signal of such renal dysfunction (Berg, 1965). Several studies have focused on the relationship between urinary protein excretion and renal morphological changes. Couser and Stilmant (1975), who studied

Correspondence to: N. Maruyama, Department of Pathology, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakaecho, Itabashi-ku, Tokyo 173, Japan. (Received 1 October 1987; Accepted 22 November 1988) 237

238

w . YUMURA et al.

age-associated morphological changes in the rat kidney, reported that 90% of test rats were proteinuric at 24 months of age. Most of those rats developed the nephrotic syndrome resulting from focal and segmental glomerular sclerosis. Moreover, Brandis and colleagues (1987) observed age-dependent glomerulosclerosis and proteinuria in rats of the Milan Normohypertensive strain, but not of the Milan Hypertensive strain. They concluded that the rats without prominent proteinuria may be protected against the development of an age-dependent glomerulosclerosis. These studies suggested that the major site of age-associated renal changes is the glomeruli. In humans, microalbuminuria has been recorded as an increased glomerular filtration rate in patients at early stages of diabetes mellitus and in aged individuals (Mogensen, 1984). As yet, there is no definitive report on the relationship between albuminuria and morphological changes or renal glomeruli in aged mice without a particular disease. Since the life span of mice is shorter than that of rats, we chose this species for study of age-related changes of renal glomeruli. As the present report describes, we observed increased albuminuria and progression of glomerulosclerosis with aging. This age-associated glomerulopathy resulted primarily from an increase in mesangial cells. MATERIALS AND METHODS

Mice Fifty-eight female C57BL/6 mice ranging from 3 to 24 months of age were sacrificed for the histological study. Urinary samples for measurement of albumin were collected from 259 mice. They were purchased from the Shizuoka Animal Breeding Co. Ltd. (Hamamatsu, Japan) at 4 weeks of age and raised in SPF mouse colony of Tokyo Metropolitan Institute of Gerontology. The mice had free access to a standard chow diet and tap water until tested.

Measurement of albuminuria Albumin concentrations in randomly sampled spot urine were measured by the single radial immunodiffusion method using rabbit anti-mouse albumin serum. Purified mouse albumin (Cappel Laboratories, Cochranville, PA, USA) was used to make a standard curve for albumin concentrations. Urine samples containing less than 100 mg/dl of albumin were arbitrarily defined as microalbuminuric. The lowest level of detection was 13.9 mg/dl urinary albumin in this assay system.

SDS-PAGE analysis Mouse urine were analyzed with 10% SDS-PAGE described by Laemmli (1970). Electrophoresed bands were visualized by Coomassie Brilliant Blue staining.

Histological examination The kidney sections were fixed in 10% buffered formalin (pH 7.4) and embedded in paraffin. The stains used for examining these 4 I~m sections were hematoxylin and eosin (H & E), periodic acid Sciff (PAS), and Congo red.

Histological scoring of renal glomeruli The specimens were examined by two observers who knew neither the animals' age nor their

AGE-ASSOCIATED

CHANGES

IN RENAL

239

GLOMERULI

2000'

1000'

| al

500"

eoooo

:3 ooo

ooo

oo

oooooo

). i-

100ooo

50-

•

o0 ewl ooo

20-

• --~ZSZ--ZZ.Z.Z

--~--:7.Z

Age (months) Occurrence of albuminuria with aging in C57BL/6 female mice. Horizontal bar represents the mean amounts in each age.

FIG. 1.

state of albuminuria. Sections containing at least 25 glomeruli each were graded 0 to 5 according mainly to mesangial changes. Mesangial changes are defined as the proliferation of mesangial cells and enlargement of mesangial area. Definitive spike lesions were also evaluated as changes in the glomerular basement membrane (GBM). The following criteria were used for scoring the degree of glomerular changes: grade 0, no apparent glomerular changes; grade 1, focal, segmental, and mild proliferation of mesangial cells (more than 2 nuclei in cell) or mild enlargement of mesangial area without cell proliferation; grade 2, diffuse mesangial cell proliferation and enlargement of mesangial area; grade 3, marked increase of mesangial cells (more than 3 nuclei in a cell) and/or enlarged mesangial areas (more than one-third of glomeruli); grade 4, severe mesangial cell proliferation (more than 5 nuclei in a cell), and appearance of focal and segmental mesangial sclerosis; grade 5, diffuse mesangial sclerosis with over 75% of glomeruli showing almost complete hyalinization.

Immunofluorescence study A piece of each kidney was rapidly frozen in liquid nitrogen and stored at - 60 °C. Sections were cut 4 to 6 Ixm thick with a cryostat and air dried. They were stained with FITC-labeled rabbit antisera (Cappel Laboratories) against mouse immunoglobulins (IgG, IgA, and IgM) and the complement component C3 at room temperature for 1 h, and then washed in PBS for 15 min, and mounted with buffered glycerol by adding p-phenylenediamine (Platt and Michael, 1983). The stained sections were examined in a dark field fluorescence microscope (Nikon Co., Tokyo, Japan).

240

w. YUMURA et al.

Age 6

12

(months) 18

24

94K

6,K

,3K

FIG, 2. SDS-PAGEanalysisof urinary protein of aged mice.

Electron microscopy A piece of the kidney tissue from each sample was fixed in 1% glutaraldehyde, and postfixed in 1% osmic acid. The tissue blocks were embedded in Epon 812, then cut into ultrathin sections before staining with uranyl acetate and lead citrate. These sections were viewed with a HITACHI 800 electron microscope. RESULTS

Age-associated occurrence of albuminuria To investigate the renal damage associated with aging of mice, we evaluated albumin excretion into urine by using a semiquantitative immunodiffusion method. Figure 1 shows the amount of albumin reactive with rabbit anti-mouse albumin antiserum in urine obtained from

AGE-ASSOCIATED CHANGES IN RENAL GLOMERULI

241

F~G. 3. Age-associated morphological changes of murine glomeruli at various ages. A) No remarkable changefrom normal at 3 months;B) Diffuseproliferation of mesangial cells in a mouse with microalbuminuriaat 24 months;C) Scleroticlesion and cell proliferation in mesangiumof the mouse with marked albuminuria at 24 months; D) Advancedsclerotic lesion in the same mouseas shown in C. * q C57BL/6 female mice. The mean amount of urinary albumin increased with aging: 14.0 mg/dl at 6 months of age, 17.6 mg/dl at 12 months, 37.0 mg/dl at 18 months, and 151.7 mg/dl at 24 months. This so-called microalbuminuria (less than 100 mg/dl) occurred in 3.3% of tested mice at 6 months of age, 20.7% at 12 months, 57.3% at 18 months, and 64.6% at 24 months. Marked albuminuria (more than 100 mg/dl) developed in 6 of 78 mice (7.6%) and 16 of 96 mice (16.6%) at 18 and 24 months of age, respectively. To confirm that the major component of this urinary protein was albumin, SDS-PAGE analysis was performed. The resulting staining profile demonstrated that the major urinary component found in mice 18 and 24 months had a molecular weight of 68 000 (Fig. 2). This finding indicates that the microalbuminuria defined here can be regarded as proteinuria.

Age-associated morphological changes of renal glomeruli By microscopy, kidneys of mice examined at 3 months of age had no remarkable changes in glomeruli, tubuli, vessels, or interstitium (Fig. 3A). However, by 6 months of age, mild mesangial cell proliferation first appeared segmentally, but mesangial sclerotic lesions were not

242

w . YUMURA et al.

P
I -P<005-']1

[ 5-

o o or)

4-

t@@

(J

o

•

3-

0000

.m

"1" E

o0

~) 2nr

1-

=1= u 1~2

•@@@

lit

l@

It 18

24

A g e (months)

FIG. 4. Quantitative light microscopic evaluation of renal changes in 12-, 18-, and 24-month-old C57BL/6 mice. The constituent of each renal histological score is described in Materials and Methods. Histological score 0 to 5 correspondto grade 0 to 5 respectively in the text. The mean score at 24 months was significantlyhigher than those at 12 and 18 months (p < 0.05 when compared by using Student's t test). yet apparent. Such proliferation was diffuse after 12 months of age. At 24 months, mice with microalbuminuria showed an increase in numbers of mesangial cells but no sclerosis (Fig. 3B). Capillary walls had thickened and undergone spike formation due to the accumulations of eosinophilic and PAS-positive materials. Moreover, mice with marked albuminuria showed a significant proliferation of mesangial cells. Some glomeruli developed focal and segmental sclerosis with a decrease in proliferative changes (Fig. 3C). Such mesangial sclerotic lesions were diffuse; however, there was no adhesion of glomeruli to Bowman's capsule (Fig. 3D). In some mice with marked albuminuria, a few cells infiltrated the medulla. Tubular cast formation with interstitial fibrosis was also found. Eosinophilic and slightly PAS positive substances accumulated in mesangial areas (Fig. 3D). Congo red staining was positive in the mesangial screlotic lesion's, and also focally positive in the interstitium. To evaluate the age-associated changes objectively, we graded renal specimens according to the criteria described in Materials and Methods. As Fig. 4 shows, scores of degenerative

AGE-ASSOCIATEDCHANGESIN RENALGLOMERULI

FIG. 5A. Deposition of immunoglobulins in glomeruli. Mesangial deposition of IgM in some lobuli at 3 months. Ftc. 5B. Diffuse granular deposition of IgM in mesangium and partially along capillary walls in the 24-month-old mouse with microalbuminuria described in Fig. 3.

243

244

w. YUMURA et al.

FI6. 5C. Marked deposition of IgG in sclerotic mesangiallesions in the 24-month-oldmouse with marked albuminuria. changes increased with aging. The mean score was 0 at 6 months of age, 0.95 at 12 months, 1.66 at 18 months, and 3.24 at 24 months. The mean score at 24 months was significantly higher than those at 12 and 18 months (Student's t test; p < 0.05). To examine the deposition of immunoglobulins or C3 in glomeruli, an immunofluorescence study was performed. A few granular deposits of IgG and IgM were observed in focal and segmental patterns in the glomerular mesangium of young mice (Fig. 5A). However, there was no significant deposition of IgA or C3. In a 24-month-old mice with microalbuminuria, deposition of IgG and IgM was marked in the mesangium and GBM of most glomeruli (Fig. 5B). In the mice with marked albuminuria, lumpy deposition of IgG and IgM was prominent in enlarged mesangial sclerotic lesion (Fig. 5C). Only a trace amount of IgA deposition was visible in the glomeruli, and no C3 deposition was obvious. Ultrastructural study demonstrated no thickening of GBM at 3 months of age. The average of width was 1770 +_ 64 (1 SD)/~ (Fig. 6A). With aging, the GBM increased in thickness until it reached an average value of 4050 +_ 474 (1 SD)/~ at 24 months (Fig. 6B). We found a small amount of electron-dense deposits in the mesangia of mice with microalbuminuria. Foot-process fusion of epithelial cell was present in the glomeruli. No marked deposition was observed along the subepithelial or subendothelial sides of the GBM. On the other hand, newly synthesized basement membrane appeared in the sclerotic lesions of mice with marked albuminuria (Fig, 6C). These mesangial sclerotic lesions contained translucent materials and fibrillar structures, but no increase of mesangial matrix (Fig. 6D).

AGE-ASSOCIATEDCHANGESIN RENALGLOMERULI

FIG. 6A. GBM and mesangia of aging mice. No thickening of GBM and no electron-dense deposits were visible at 3 months of age. FiG. 6B. Marked diffuse increase in thickening of basement membrane with segmental nodules and foot process of epithelial cells in 24-month-old mouse with microalburninuria.

245

246

w. YUMURAet al.

F16. 6C. Glomerula sclerosis with newly synthesized basement membrane. The foot process of epithelial podocytes is widely fused. Some segments in sclerotic mesangial lesions show low electron density. FIc. 6D. High power view of the segmental mesangial lesion from the same glomerula structure are apparent but little increase of the mesangial matrix.

AGE-ASSOCIATED CHANGES IN RENAL GLOMERUL1

247

DISCUSSION In the present study, we investigated the relationship between albuminuria and glomerular sclerosis in aged C57BL/6 mice. It seemed appropriate to use this strain, which is raised in our facility, because these animals do not develop any particular disease in the course of senescence. Our test confirmed that the C57BL/6 mice undergo an increase in urinary protein excretion with aging. Since SDS-PAGE analysis of the excreted urinary protein showed that the major component was 68K in molecular weight (Fig. 2), one can regard proteinuria as the albuminuria used here as a criterion of renal deterioration. Subsequent morphological study suggested that a numerical increase of mesangial cells could be the cause of this albuminuria, since the two events coincided temporally (Figs. 1 and 4). The increase of mesangial cells first appeared at younger age in mice than in rats, and the resultant glomerular sclerotic lesions differed in the two species. That is, the distribution of glomerular sclerosis was focal in rats, but diffuse in mice. However, the precise mechanism of this finding has not been clarified. Increased glomerular permeability is believed to play a critical role in the development of sclerosis. Several factors may enhance glomerular permeability, such as injection of aminonucleoside of puromycin or a state of serum sickness (Vernier et al., 1959; Ericsson and Andres, 1961). Thereby, mesangial uptake of soluble materials or particles from the circulation is enhanced (Mauer et al., 1972). Lannigan (1963) also suggested that the mesangial changes following the increase of GBM permeability is the lesion preceding focal glomerular sclerosis. Additionally, the alteration of GBM components may affect glomerular permeability. Cohen and Ku (1983) reported that a decrease in the sulfation of glycosaminoglycans in the aged human kidney can reduce the net negative charge of the GBM, and render it more permeable to macromolecules. In electron microscopic analysis, we observed GBM thickening in the association with mesangial cell proliferation (Fig. 6B), suggesting a change in GBM components. Although the mechanism of increased GBM permeability is wholly obscure, the outcome may be albuminuria. Additionally, other age-associated factors might be involved in this process. The mesangium presumably functions as part of the reticuloendothelial system, taking up and clearing aggregated serum proteins or immune complexes of large molecular weight derived from the subendothelial space (Michael et al., 1967, Stilmant et al., 1975). Accumulations of these substances are known to affect mesangial function. For example serum components can deposit in glomeruli of aged mice (Linder et al., 1972; Shimizu et al., 1977). Granular depositions of immunoglobulin and complements have been demonstrated in early lesions of glomeruler capillary walls and mesangia (Linder et al., 1972), The murine leukemia viral antigen with gamma globulin were found in C57BL x C3H F 1 hybrid mice over 20 months of age (Peter, 1973). Markhan et al. (1973) examined normal mice at 24 to 32 weeks of age and reported that 95.4% had IgM deposits, 85.6% had IgG, 62.2% had IgA, and 78.7% had C3. These reports suggest that immunoglobulins deposited in the kidneys derived from circulating immune complexes in many strains of mice. Although, in our study, marked deposition of IgG and IgM was clear in the mesangia of most glomeruli, staining for C3 was faint. However, the deposition of C3 is prominent in the glomeruli of autoimmune disease-prone NZB/WF 1 mice, which develop immune complex glomerulonephritis (Helyer and Howie, 1968). Therefore, the results cited here suggest that the immunoglobulins deposited in the kidneys of aged C57BL/6 mice did not originate from circulating immune complexes. However, since these mice become hyperglobulinemic with age (data not shown), one can postulate that exudation and/or entrapment of their accumulated serum components are caused by a decrease of mesangial

248

w. YUMURA et al.

function due to glomerular sclerosis. Systemic amyloidosis is a well-known occurrence in various organs of aged individuals. There are several reports of amyloid substances in kidneys of aged mice (Dunn 1944; Thung, 1957a). We observed diffuse glomerular sclerosis with positive Congo red staining, indicative of amyloid's effect. However, those previous studies reported that amyloid deposition was restricted mainly to tubuli and interstitium, but not found in glomeruli. In our ultrastructual study, the site of fibrillar deposits corresponded with those of Congo red staining. Our study on aged mice has demonstrated morphological changes of the renal glomerulus with age. Those changes include mesangial cell proliferation, thickening of capillary walls and glomerular sclerosis with deposition of immunoglobulins and amyloid fibrills. Acknowledgments -- We thank Ms. Setsuko Handa, Ms. Eiko Moriizumi, and Ms. Yoshie Urano for their excellent

technical assistance. The excellent editorial assistance of Ms. Phyllis Minick is gratefully acknowledged. This study was performed through Special Coordination Funds of the Science and Technology Agency of the Japanese Government. REFERENCES BERG, B.N. Spontaneousnephrosis with proteinuria,hyperglobulinemiaand hypercholesterolemiain the rat. Proc. Soc. Exp. Biol. Med. 119, 417-420, 1965. BOLTON, W.K., BENTON, F.R., MACLAY,J.G., and STURGILL,B.C. Spontaneousglomerularsclerosisin aging Sprague-Dawley rats. I. Lesions associated with mesangialIgM deposits. Am. J. Pathol. 85, 277-300, 1976. BRANDIS, A., BIANCH, G., REALE, E., HELMCHEN, U., and KUHN, K. Age-dependentglomerulosclerosisand proteinuriaoccurringin rats of the Milan normotensivestrain and not in rats of the Milan hypertensivestrain. Lab. Invest. 55, 234-240, 1987. COUSER, W.G. and STILMANT,M.M. Mesangiallesionsand focal glomemlarsclerosisin the agingrat. Lab. Invest. 33, 491-501, 1975. COHEN, M.P. and KU, L. Age-related changes in sulfation of basement membrane glycosaminoglycans.Exp. Gerontol. 18, 447-450, 1983. DUNN, T.B. Relationshipof amyloidinfiltrationand renal disease in mice. J. Natl. Cancer Inst. 5, 17-28, 1944. ELMA, J.D. and ARENDS, A. Focal and segmentalglomerularhyalinosisand sclerosis in the rat. Lab. Invest. 33, 554-561, 1975. ERICSSON, J.L.E. and ANDRES, G.A. Electronmicroscopicstudies of the developmentof the glomerularlesionsin aminonucleosidenephrosis. Am. J. Pathol. 39, 643-663, 1961. GUDE, W.D. and LUPTON, A.C. Spontaneousglomerulonephritisin aging RF mice. Gerontology 15, 373-376, 1960. HEYLER, B.J. and HOWIE, J.B. The immunologyand pathology of NZB mice. Adv. lmmunol. 9, 215-266, 1968. HIROKAWA, K. Characterizationof age-associated kidney disease in Wister rats. Mech. Ageing Dev. 4, 301-316, 1975. KIRSCHBAUM, A., BELL, E.T., GORDON, J., and MINN, M. Spontaneousand inducedglomerulonephritisin an inbred strain of mice. J. Lab. Clin. Med. 34, 209-220, 1949. LAEMMLI, U.K. Cleavage of structural proteins during the assembly of the head of bacteriophage T. Nature 227, 680-683, 1970. LANNIGAN, R. The productionof chronicrenal disease in rat by a single intravenousinjectionof aminonucleosideof puromycinand the effect of low dosage continuoushydrocortisone. Br, J. Exp. Pathol. 44, 326--333, 1963. LINDER, E., PASTERNACK,A., and EDGINGTON,T.S. Pathology and immunologyof age-associated disease of mice and autologous immunecomplex pathogenesisof the associated renal disease. Clin. Immunol. lmmunopathol. 1, 104-121, 1972, MARKHAN, R.V., SUTHERLAND,J.C., and MARDINEY, M.R. The ubiquitousoccurrence of immunecomplex localization in the renal glomeruliof normal mice. Lab. Invest. 29, 111-120, 1973. MAUER, S.M., FISH, A.J., BLAU, E.B., and MICHAEL, A.F. The glomerularmesangium.I. Kinetic studies of macromolecularuptake in normal and nephroticrats. J. Clin. Invest. 51, 1092-1101, 1972. MICHAEL, A.F., FISH, A.J., and GOOD, R.A. Glomerularlocalizationand transportof aggregated proteinsin mice. Lab. Invest. 17, 14--29, 1967.

AGE-ASSOCIATEDCHANGESIN RENALGLOMERULI

249

MOGENSEN, C.E. Microalbuminuria predicts clinical proteinuria and early mortality in maturity-onset diabetes. N. Engl. J. Med. 310, 356-360, 1984. NEUHAUS, O.W. and FLORY, W. Age-dependent changes in the excretion of urinary proteins by the rat. Nephron. 22, 570-576, 1978. PETER, C.P. Possible immune origin of age-related pathological changes in long-lived mice. J. Gerontol. 28, 265-275, 1973. PLATI', J.L. and MICHAEL, A.F. Retardation of fading and enhancement of intensity of immunofluorescence by p-phenylendiamine. J. Histochem. Cytochem. 31, 840-842, 1983. SHIMIZU, F., ABE, F., ITO, K., and KAWAMURA, S. On the age-associated presence of immunoglobulin and complement in the renal glomeruli of mice. Contro. Nephrol. 6, 79-93, 1977. STILMANT, M.M., CAUSER, W.G., and COTRAN, R.S. Experimental glomerulonephritis in the mouse associated with mesangial deposition of autologous ferritin immune complexes. Lab. Invest. 32, 746--756, 1975. THUNG, P.J. The relation between amyloid and ageing in comparative pathology. Gerontologia 1, 234-254, 1957a. THUNG, P.J. Senile amyloidsis in mice. Gerontologia 1, 259-279, 1957b. VERNIER, R.L., PAPERMASTER, B.W., and GOOD, R.A. Aminonucleoside nephrosis. 1. Electron microscopic study of the renal lesion in rats. J. Exp. Med. 109, 115-125, 1959.