Streptococcus mutans-induced nephritis in rabbits

0 downloads 4 Views 4MB Size Report
Oct 11, 1984 - (x 42,000) ..... 3rd edition. New York ... Dodge WF, Spargo BH, Travis LB, Srivastava RN, Car- ... immune complex nephritisand basic proteins.

Streptococcus mutans-Induced

BORIS ALBINI, MD, RUSSELL J. NISENGARD, DDS, PhD, I. GLURICH, BS, MIRDZA E. NEIDERS, DDS, and MURRAY W. STINSON, PhD

Neph

ritis in

Ra bbits

From the Departments of Microbiology, School of Medicine and Dentistry, and Departments of Periodontology and Oral Pathology, School of Dentistry, State University of New York at Buffalo, Buffalo, New York

Intravenous administration of disrupted Streptococcus mutans into rabbits over 23-76 weeks led to severe nephritis involving glomeruli, tubules, and interstitium. Light-microscopic observation of glomeruli documented diffuse endocapiliary proliferative glomerulonephritis accompanied often (65 %) by epithelial crescents. Electronmicroscopic observation revealed humps in glomeruli of 70% of kidney specimens. In the glomeruli of some rabbits, extensive fibrin deposits and sclerosis were evident. Immunofluorescence showed linear, granular, often ribbonlike or patchy immune deposits encompassing, in

order of decreasing frequency, C3, IgG, streptococcal antigen, IgA, and IgM. The histopathologic and immunohistologic features of the nephritis seen in rabbits given S mutans thus shows many features of Streptococcusassociated nephritides in man, in particular, the diffuse glomerular nephritis encountered in subacute bacterial endocarditis. Further, analysis of nephritis induced by administration of S mutans may have implications for the evaluation and purification of dental caries vaccines. (Am J Pathol 1985, 118:408-418)

IN MAN, nephritis may occur in the course or wake of infections with several species of streptococci.1-3 The main syndromes encompass acute poststreptococcal glomerulonephritis (APSGN) associated prominently with the "nephritogenic" strains of group A f3-hemolytic streptococci,4 6 but also with other streptococci, eg, of group GI; glomerulonephritis (GN) observed in subacute bacterial endocarditis (SBE)2; and interstitial nephritis,8 usually seen together with GN. Whereas APSGN has become a rather rare disease in Europe and the United States,9 it still is frequently encountered in tropical countries.10'11 The understanding of the pathogenesis of Streptococcus-associated nephritis (SAN) has gained much through studies of biopsy material, but reconstruction of the early development of the disease requires the study of a reliable animal model.12 This is underlined by the number of attempts made in the last three decades to establish such a model.13 Whereas nephritis has been observed in a recently developed rabbit model of SBE,14 no fully satisfactory results have been obtained with regard to an animal model reproducing the main pathognomonic features of APSGN in man.'2 '3 Models proposed up to now have reproduced

some of the features of APSGN,15 but usually showed very mild disease or were not easily reproducible.13 Streptococcus mutans is one of the four recognized species of the viridans group.16 These a-hemolytic microorganisms, lacking a well-defined group carbohydrate antigen, are found commonly in the mouth adherent to tooth surfaces. S mutans is consistently associated with supragingival dental caries.17 The use of S mutans antigens as components of dental caries vaccines is currently under trial. 18 This microorganism has also been demonstrated in patients with SBE.19-2' In the following we report on the histopathology and immunohistology of S mutans-induced nephritis observed in rabbits.

Supported by Public Health Service Grant DE05696 from the National Institute of Dental Research. Accepted for publication October 11, 1984. Address reprint requests to Boris Albini, MD, Department of Microbiology, Schools of Medicine and Dentistry, State University of New York at Buffalo, 240B Cary Hall, Buffalo, NY 14214.

408

STREPTOCOCCAL NEPHRITIS IN RABBITS

Vol. 118 * No. 3

Materials and Methods Animals Fourteen 4-month-old male New Zealand White rabbits were used for immunization with S mutans. The rabbits were purchased from a local dealer (Beckens Research Animal Farm, Sanborn, NY) and housed in the animal unit under veterinary surveillance and were given water and Purina pellets ad libitum. Three rabbits were not immunized and served as untreated controls.

Streptococcus mutans Preparations S mutans strains KIR (serotype d/g) and MT703 (serotype e) were grown in tryptic soy broth supplemented with 0.1% yeast extract (Difco Laboratories, Detroit, Mich) as previously described.22 Bacteria were disrupted by shaking with glass beads in a Braun cell homogenizer (Bronwill Scientific Inc., Rochester, NY) at 4 C.23 The suspension was heated at 70 C for 30 minutes, adjusted to 10 mg protein/ml in phosphatebuffered saline and stored at -20 C. Immunization The S mutans vaccine was evenly suspended by ultrasonication (Biosonic III, Bronwill Scientific) and injected intravenously into the rabbits by the immunization protocol proposed by van de Rijn and Bleiweis.23 The rabbits were given the injections thrice weekly with 0.5 mg protein the first week, 1.0 mg the second, and 1.5 mg the third week. The animals were bled at the end of the fourth week. This scheme was repeated at 4-week intervals. The rabbits injected with S mutans MT703 or S mutans KIR were designated Group A and Group B, respectively. Serum Creatinine and Urinalysis

Serum creatinine was determined in blood obtained from rabbits at sacrifice.24 Urine was collected once a week. The volume accumulating during 24 hours was documented, and the urine was checked for hematuria with the use of Hemastix (Ames, Miles Laboratories, Inc., Elkhart, Ind). The amount of protein in the urine was determined by the method of Sheffky and Stafford.25 Autopsy Rabbits were sacrificed by heart puncture after anesthesia with a barbiturate mixture (4 ml Diabutol, 1 mg Suritol in 5 ml 0.9% NaCl) administered intravenously.

409

An incision was made along the linea alba, and the kidneys were exposed. The macroscopic appearance of the kidneys was recorded. Tissue specimens taken for histologic studies were fixed in 10%o buffered formalin. Specimens for electron-microscopic study were minced into tissue blocks smaller than 1 mm in diameter and placed immediately into Karnovsky's fixative26; after 3 hours of incubation on ice, the fixative was removed and replaced by 0.1 M phosphate buffer. Specimens for immunofluorescence tests were placed on bolts, covered by OCT (Lab Tek Products, Naperville, Ill) and quickfrozen in liquid nitrogen. Light Microscopy Tissue was sectioned at 5 , thickness. Tissue sections were stained with hematoxylin and eosin, periodic acidSchiff (PAS), Masson's trichrome, PAS-silver-methenamine (PASM), methylgreen-pyronin, and picroMallory stains.2728 Preparations were screened for amyloid with Congo red and polarized light.29 The proliferation of glomerular cells was graded from 0 to V according to Baldwin et al,30 as described in the legend to Table 2. The increase of the mesangial matrix was graded A to D according to the same authors (see the legend to Table 2). The term "fibrosis" and "sclerosis" are used as defined by Zollinger and Mihatsch.28 Electron Microscopy Tissue blocks fixed in Karnovsky's fixative and stored in a 0.1 M phosphate buffer were embedded in Epon and stained with uranyl and lead acetate.28 Thin sections were placed on copper grids and viewed with a Siemens Elmiskop 101. Immunofluorescence Tests Tissue, snap-frozen in liquid nitrogen, was cut at - 20 C in an American Optical Cryostat (Cryocut, A&O, Buffalo, NY) at 4 M. The sections were stored at -20 C until used. Direct immunofluorescence tests were performed as described in detail elsewhere.3' The fluorescein isothiocyanate (FITC) antibody conjugates reacting with rabbit IgG, IgA, IgM, C3, and fibrinogen were obtained commercially (Cappel Laboratories, Westchester, Pa) and had molar F/P ratios of 2.1 to 2.8 and contained 2-4 U of antibody as determined following the procedure of Beutner, Sepulveda, and Barnett.32 The tissue was not fixed, and the test results were interpreted with the use of a Leitz-Wild Ortholux II immunofluorescence microscope equipped with epiillumination and a 100 W high-pressure mercury vapor lamp (Osram,

410

ALBINI ET AL

AJP

*

March 1985

Table 1-Rabbits Given Injections of S mutans MT703 (Group A) and S mutans Kl R (Group B): Clinical Parameters Group

n

A

7

Sacrifice (weeks) 38.3

+

23.8*

Oliguria

Hematuria

Proteinuria (mg/24 hours)

6/7

6/7

295.7 + 279.6*

(23-73)t B

7

Control

3

(43.2-748.8)t

56.3 + 22.1 (32-76)

6/7

60

0/3

6/7 0/3

216.0 ± 264.7 (28.8-403.2) 46.3 + 8.2

Creatininemia

(mg/dl) 4.2

+

2.4*

(1.6-6.4)t 2.6

+

0.8

(1.7-3.1) 0.9 ± 0.3

* Mean + SD. t Range of values.

ER.G.). Control preparations included blocking of staining by preincubation of tissue with unlabeled antiserum. Antiserum to S mutans KIR and MT703 was raised in chickens by intradermal immunization. The antibodies were precipitated with ammonium sulfate and conjugated to FITC according to routine procedures.32 Because both antisera reacted with KIR and MT703 streptococci, all tests were run with both reagents. The molar F/P ratio of the two preparations was 2.3 for the anti-KIR conjugate and 2.7 for the anti-MT703 conjugate. The conjugates did not react with normal rabbit kidney nor with kidney sections from rabbits with chronic serum sickness. Thioflavine-T fluorescence was used for assessment of the presence of amyloid.33 Results

Urinalysis, Creatininemia, and Appearance of Kidney at Autopsy A summary of the results on urinalyses and creatininemia is given in Table 1. In 47% of the rabbits, the kidneys were of normal size at autopsy. The kidneys of the other rabbits were enlarged (individual kidney weight up to 36 g). They were pale in the majority of rabbits. Forty-three percent of the rabbits had petechiae on the surface of the kidneys, and 290Wo had scars ranging from 1 to 8 mm in diameter and from 1 to 9 mm in depth.

Light Microscopy (Table 2) The presentation of kidney pathology in two representative animals given S mutans can be seen in Figures 1 and 2. In some rabbits, extensive areas of kidney tissue showed fibrosis and sclerosis (Figure 3). All but one rabbit showed increased numbers of polymorphonuclear leucocytes (PMNs) in glomeruli (up to 17 per glomerulus). The distribution of PMNs in glomeruli tended to be segmental, with large numbers accumulating in one or two lobes (Figure 4A). All rabbits showed hypercellularity in glomeruli (Figure 4A and C). Sixty-five percent of the rabbits had crescents (Figure 4C). In 43% of the rabbits, crescents were present in more than 30% of the glomeruli (Figure 2). The occlusion of glomerular capillaries varied from segmental to global. A majority of rabbits had axial or panmesangial matrix increase (Figure 4B). Bowman's capsule showed various degrees of thickening, splitting, or thinning. In some cases, wrinkling of the basement membrane and spikelike projections were seen (arrowheads, Figure 4B). The urinary space often was filled with proteinaceous fluid or fibrin (Table 2). Some glomeruli showed segmental capillary loop necrosis. No amyloid could be found with the Congo red stain on six selected kidney tissues. The periglomerular tissue was edematous, fibrotic, or sclerotic and infiltrated by PMNs or mononuclear cells, often including plasma cells. Tubular lumens often were widened, and the tubular epithelial cells were flattened (Figures 1 and 2). The

Figure 1-Rabbit 263. Glomeruli show moderate hypercellularity with lobulation and segmental occlusion of glomerular capillaries. The urinary space is free. There is focal interstitial edema and sparse and focal interstitial infiltration by inflammatory cells; focally, tubules are dilated. (PAS, x 80) Figure 2-Rabbit 240. Glomeruli show endotheliomesangial proliferation with lobulation and moderate to severe occlusion of capillaries. In addition, there is extensive epithelial proliferation with adhesions and numerous crescents (arrows). The interstitium is edematous and focally sclerotic and shows focal cellular infiltrates. Many tubules are atrophic and contain hyaline casts. (PAS, x 60) Figure 3-Rabbit 264. Large area of fibrosis and sclerosis. Only a few strands of cells persist in the place of tubules. Glomeruli show various degrees of disease. (Masson, x 65) Figure 4A-Rabbit 240. Endotheliomesangial proliferation, in a few areas glomerular capillary loop necrosis, and some PMNs. Periglomerular sclerosis and moderate cellular infiltration. B- Rabbit 257. Increase of mesangial matrix, thickening of the GBM, and Bowman's capsule (showing Some capillaries remain patent. (PAS, x 200) prominent spikes, arrowheads). Irregular thickness of the TBM. (PASM, x 320) C-Rabbit 240. Endotheliomesangial proliferation, few PMNs, and cellular crescent. (PAS, x 200) Figure 5-Rabbit 261. FITC-conjugated antiserum to rabbit IgG. Linear to ribbonlike staining of GBM. Segmentally, the granularity of the staining can be seen (arrows). (x 250) Figure 6-Rabbit 258. FITC-conjugated antiserum to rabbit C3. Coarsely granular staining of GBM and mesangium. (x 250) Figure 7-Rabbit 257. FITC-conjugated antiserum to S mutans raised in chicken. "Homogeneous" and linear staining of GBM and mesangium. (x250)

Vol. 118 * No. 3

STREPTOCOCCAL NEPHRITIS IN RABBITS

.

py

III

L, .1.

.#,4"

411

412

ALBINI ET AL

AJP * March 1985

Table 2-Light Microscopy

Pathologic lesion Glomeruli PMNs, > 4/glomerulus Fibrin Capillary loop Necrosis Thickened GBM Thickened Bowman's capsule Hypercellularity* 0

II IV V Mesangial matrix increaset A B

C D Tubular atrophy Interstitium Edema Sclerosis and fibrosis Cellular infiltrates

Table 3-Immunofluorescence Tests

Group A

Group B

Control

Group Group Site

6/7 2/7

7/7 0/7

0/3 0/3

2/7 4/7 4/7

1/7 6/7 5/7

0/3 0/3 0/3

0/7 2/7 0/7 1/7 0/7 4/7

0/7 0/7 0/7 0/7 5/7 2/7

3/3 0/3 0/3 0/3 0/3 0/3

1/7 3/7

3/7 6/7

0/7 3/7 2/7 2/7 7/7

3/3 0/3 0/3 0/3 0/3

7/7 6/7 7/7

7/7 7/7 7/7

0/3 0/3 0/3

0/7

"Grading slightly modified according to Baldwin et al.30 0, normal cellularity; I, increased number of nuclei in some mesangial regions; II, mesangial cell proliferation with occlusion of occasional peripheral capillary lumens; Ill, mesangial and endothelial cell proliferation with occlusion of up to 50% of capillary lumens; IV, mesangial and endothelial cell proliferation with occlusion of most capillary lumens, occasionally with crescent present (less than 30% of glomeruli); V, mesangial and endothelial cell proliferation with occlusion of most capillary lumens and many large crescents (more than 30% of glomeruli). t Grading according to Baldwin et al.30 A, normal or equivocal increase; B, increased coarse or condensed mesangial matrix; C, increased matrix in portions of lobules; D, obsolescent glomerulus.

lumen of tubules was sometimes filled with casts (Figure 2) or inflammatory and detached epithelial cells. The interstitium always was focally or diffusely expanded by edema, fibrosis, or sclerosis (Figures 1-3). Infiltration by PMNs and mononuclear cells was common. Vessels often showed intimal thickening with insudation of fibrin. Immunofluorescence (Table 3) The main patterns of glomerular staining are depicted in Figures 5-7 and Figure 8 (inset). Often, staining for IgG was ribbonlike, with some areas of granularity (arrows, Figure 5). Infrequently, the staining was finely granular. In the majority of rabbits, glomeruli showed diffuse and global staining, but in some there was focal and segmental granular staining. All but one rabbit had IgG in the GBM and mesangium. Staining for C3 was present in all rabbits studied and always was finely to coarsely granular (Figure 6). Fifty percent of the rabbits showed fibrinogen-associated antigen (FAA) seg-

Glomerulus GBM

Specificity

Pattern

A

B

Control

IgG

Granular Linear Granular Linear Granular Ribbonlike Granular Patchy "Homogeneous" Granular Linear

6/7 3/7 2/7 2/7 1/7 0/7 7/7 1/7 3/7 1/7 0/7 1/7 1/7 3/7 4/7 3/7

4/7 3/7 1/7 2/7 1/7 1/7 7/7 1/7 1/7 1/7 1/7 3/7 3/7 1/7 4/7 3/7

0/3

IgA

IgM C3 FAA* SAt Mesangium

Cells

IgG C3 SA IgG SA

0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3 0/3

* FAA, fibrinogen-associated antigens. t SA, S mutans antigens.

mentally and in small patches in the glomerular capillaries; in most animals, these deposits were small; extensive FAA deposits were detectable only in Rabbit 255. Sometimes there were deposits of immunoglobulin and C3 in cells of the glomeruli (inset, Figure 8). Glomeruli of 36Wo of the rabbits studied showed broad homogeneous and granular or patchy staining when incubated with antisera raised in chickens to S mutans (Figures 7 and 8). Staining often involved GBM, the mesangium, and cells in glomerular capillaries. When normal rabbit kidney was incubated with the antisera, no staining was observed in the glomeruli. S mutans, but not Actinobacillus actinomycetemcomitans, cells absorbed the serum antibodies that were reactive with deposits in GBM and mesangium, but not those reactive with glomerular cells, which indicates that the former, but not the latter, reactivity was specific for microbial antigens. Preincubation of tissue with unlabeled antiserum diminished subsequent staining with conjugated antiserum by 2-4 titer steps (doubling dilutions). Deposits of immunoglobulins, C3 and fibrinogen were also seen in Bowman's capsule in focal and segmental distribution. Frequently, granular, linear, or patchy staining for IgG, IgA, or C3 was seen in structures of the interstitium. All kidney specimens studied were reacted with thioflavine for amyloid but showed no staining. Electron Microscopy (Table 4)

The most frequent glomerular changes occurring in these rabbits are illustrated in Figure 8. The capillary lumen is occluded by extensive proliferation of en-

Figure 8-Rabbit 255. Glomerular capillary. Proliferating endothelial cells (EN); PMN, degenerated epithelial cells (EP) with fused foot processes; subendo-

thelial (arrows) subepithelial (humps, H) electron-dense deposits. Between the arrowheads, minute electron-dense deposits are seen in the basement membrane. (x 16,500) Inset-Rabbit 266. FITC-conjugated antiserum to rabbit IgG. PMNs and small immune deposits (arrows). (x 160)

414

ALBINI ET AL

AJP * March 1985

Table 4-Electron Microscopy

Pathologic lesion Glomeruli Deposits in Laminae rarae* Subendothelium Subepitheliumt Humps Lamina densa Capsular Mesangial Podocyte fusion Podocyte degeneration Endothelial cell proliferation Urinary space Fibrin

Collagen PMNs Erythrocytes Sclerosis GBM Mesangium

Interstitium Accumulation of collagen Mild or moderate Extensive *

The electron-dense deposits seen in or along the GBM had a variety of patterns. In mildly involved glomerular loops with patent lumens, minute polygonal electron-dense deposits were seen predominantly in the

Group A

Group B

Control

3/6 5/6

2/4 3/4

0/2 0/2

5/6

4/4

0/2

3/6 1/6

4/4 1/4

0/2 0/2

(Figure 10) but in some instances presented as large con-

5/6

4/4

0/2

6/6

4/4

0/2

4/6 5/6

4/4 2/4

0/2 0/2

tinuous deposits reminiscent of "wire-loop" lesions. Intramembranous electron-dense material was seen infrequently and limited to short stretches of GBM.

laminae seemed

rarae to

(Figure 9). These deposits sometimes

aggregate

just below the podocytes (arrow

in Figure 9). Subendothelial deposits were often small

Subepithelial electron-dense deposits were observed in

a variety of patterns: as minimal accumulations

3/6

2/4

0/2

granular material (Figure 9),

as

of finely

dense deposits reach-

4/6

4/4

0/2

2/6

1/4

0/2

ing into the slit pores, or forming huge aggregates of granular and fibrillar material ("humps"; Figure 11). A rim of osmophilic cytoplasmic fibrils was seen in the

2/6 5/6

2/4 4/4

0/2 0/2

podocyte surrounding the hump. Humps

5/6 3/6

4/4 2/4

0/2 0/2

Finely granular deposits (see FFigure 9).

t Small to moderately large deposits, excluding "humps."

dothelial cells. A PMN is se-en between endothelial cells. The basement membrane shows some wrinkling and variable thickness of the la mina densa. Electron-dense material is seen deposited in or along the GBM. The podocytes show fusion of foot processes and degenerative changes, including accumulation of fibrillar intracytoplasmic material, formation of villi, and cloudy swelling of mitochondria. Some glomerular loops with patent lumens and thin endothelium showed fusion of foot processes (eg, Figure 9). In some glomerular capillaries, monocytes could be seen. At other sites, degenerative, necrotic, and sclerotic processes determined the morphologic character of glomerular structures (Figure 12 A and B). Bowman's capsule often showed irregular thickening and splitting of basement membrane, apposition of electrondense material, and degenerative changes of parietal epithelial cells. In some instances, fibrin accumulated in the urinary space (Figure 12A).

also in

areas

were

present

with extensive fibrin deposition (Figure

12A) and sometimes were replaced by basement membrane-like material in sclerotic glomerular loops (Figure 12B). PMNs or mononuclear cells,

lymphocytes,

were

also

seen

predominantly

small

in the tubular epithelium

and lumen. Mononuclear cells, predominantly plasma cells with Russell bodies, and granulocytes were seen in edematous interstitium or in close apposition to interstitial collagen fibers.

Discussion The viridans group of streptococci has been known to be associated with SBE since early in this century.34

Many patients with SBE show "peripheral manifestation," including involvement of the kidney. The pathogenesis of the SBEGN was explained first by bacterial emboli arising from L._ endocardial lesions.35 As early as 1916, Longcope36 speculated on an allergic pathogenesis for SBEGN. The probability of involvement of immunologic mechanisms was further stressed by more recent reports.37-39 S mutans has been implicated in SBE,20 2' and a study by Neefe et al'9 suggests that this microbe may be involved much more often in SBE than reported because of possible misdiagnosis by confusion

-~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Figure 9-Rabbit 256. Glomerulus. The endothelium is normal; podocytes (EP) show fusions. The lamina densa of the GBM is not altered, but the laminae rarae contain minute electron-dense granules (small arrows). These granules accumulate focally under the foot processes (large arrow), possibly leading to the larger, more homogeneous subepithelial deposits (double arrow). (x 40,000) Figure 10-Rabbit 267. The endothelial fenestrae are not visible; podocytes show segmental fusion. Deposits of electron-dense material (arrowheads) are seen between the endothelium and lamina densa. (x 42,000) Figure 11-Rabbit 255. Proliferating endothelial cells (EN) fill the capillary lumen. There is no visible lamina rara interna. Humps (asterisk) are seen on the subepithelial side of the GBM. The humps are delineated by a thin lucent space against the podocytes and, in some areas, against the lamina densa. Podocyte cytoplasm shows accumulation of osmiophilic fibrils surrounding the humps (arrowheads). The foot processes show extensive fusion, and the cell body of the podocytes (EP) is swollen. (x 22,750) Figure 12A-Rabbit 255. Swollen endothelial cells, degenerated podocytes, and humps (asterisk) are seen. Large amounts of fibrin (double arrows) are present in the urinary space. (x 22,750) B-Rabbit 261. Sclerotic glomerulus. Basement membrane-like material is at the possible site of a resolved hump (asterisk). (x 38,000)

STREPTOCOCCAL NEPHRITIS IN RABBITS

Vol. 118 * No. 3

4*

Wll". .i..:

piTh

v i

. --

:.¼. ;-4

415

416

ALBINI ET AL

with enterococci. The histopathology is characterized by a segmental-focal or diffuse GN.28'40 Animal models of SBE-glomerulonephritis have been established using staphylococci41 and Streptococcus salivarius. 4 In the rabbit model developed by Arnold et al, 14 streptococci were inoculated into the heart. Sindrey et al,42 using this model, stressed the pathogenic role of antiglobulins. However, a comprehensive morphologic analysis of SBE-associated kidney pathology seen in long-lasting experimental disease has not yet been published. The histopathologic features of the nephritis reported here encompass exudative, proliferative, and sclerosing lesions of the glomerulus as well as interstitial involvement characterized by edema, inflammatory cell infiltrates with prominent PMNs and plasma cell populations, disease of the tubular basement membrane (TBM), and focal to diffuse sclerosis and fibrosis of the interstitium. The predominant picture was that of diffuse glomerular involvement. Diffuse forms of glomerulonephritis associated with various streptococci and staphylococci share many morphologic characteristics. Many similarities are seen between the S mutans-induced experimental nephritis and SBEGN and, to a certain extent, also APSGN in man. Endothelial proliferation and increased number of PMNs are salient features of both S mutans-induced GN and ASPGN. Electron-dense deposits in the form of humps are seen in APSGN and SBEGN40'43 and the S mutans-induced GN in rabbits. Humps are considered hallmarks of the lesion of ASPGN, albeit they may occur also in other glomerular lesions and are not seen in all kidneys of patients with ASPGN.28 As in both SBEGN and APSGN,28 subendothelial, intramembranous, and "classical" subepithelial deposits were seen in the rabbit GN. Analysis of immunologically reactive components in the glomerular deposits (Table 3) showed C3 to be the most frequently encountered moiety, followed closely by IgG. The mesangium showed immune and electron-dense deposits and, in many of the rabbits, impressive increase of mesangial matrix, including axial and paramesangial involvement. This is also seen in SBEGN and APSGN in man. Epithelial cell crescents, seen in over one-third of animals studied, are also seen in a subpopulation of patients with SAN.28 The similarities in histopathologic features between SBEGN, APSGN, and S mutans-induced nephritis in rabbits have been so convincing that we could use the grading of glomerular lesions of APSGN as proposed by Baldwin et a130 in reporting S mutans-induced GN of rabbits (Table 2). APSGN in man is clinically an acute disease, widely considered to have a "one-shot" pathogenesis comparable to that of experimental acute serum sickness, which

AJP * March 1985

has also been proposed as its animal model. 13 Because the S mutans-associated kidney disease of rabbits was not induced by a single dose of antigen, it seems to be comparable to APSGN only with respect to the morphologic and immunopathologic features, but not to the mode of disease induction and pathogenesis. This one-shot hypothesis indeed may prove to be true, but not many data are available on the early development of human APSGN in man. Immunity to M-proteins of P-hemolytic streptococci seems to prevent the reoccurrence of clinically manifest disease caused by the same microorganism; however, colonization of the nasopharynx or the skin is not prevented by development of immunity. 16 In addition, repeated injury to kidney may occur in the wake of several infections with a variety of different species, strains, or serotypes. Because APSGN may not lead to overt clinical symptoms but may only be documented upon careful urinalysis,44 it seems likely that mild episodes of the disease may go unnoticed by the patient. Therefore, repeated showers of streptococcal antigen may indeed occur in ASPGN. The pathogenesis of APSGN and SBEGN remains enigmatic despite thorough studies of immunologic involvement of the clinical disease in man. 112,28,45-49 Understanding is further complicated by controversies over the possible development of chronic nephritis from APSGN or the incidence of APSGN in the course of other unrelated chronic kidney diseases.45'46 Several immunologic effectors have been described in patients with APSGN and SBEGN, including antibodies to microbial components, antibodies crossreacting with tissue and bacteria, antibodies to immunoglobulin,49 and circulating immune complexes.50 In the rabbit GN reported here, a serologic pattern comparable to that described in SBEGN and APSGN in man was encountered.22 Direct involvement of S mutans in the pathogenesis of the rabbit glomerulonephritis was suggested by the demonstration of S mutans antigens in the immune deposits of diseased kidneys by immunofluorescence (Table 3). The staining pattern varied from homogeneous to granular. The control tests support specificity of the glomerular peripheral staining, but additional analyses are necessary to substantiate and extend these observations. The reactivity with antiserum to S mutans antigens in rabbit glomeruli is in agreement with results published by Andres et all' and Michael et a148 demonstrating streptococcal antigens in glomeruli of patients with APSGN. One may speculate that the fine-granular electrondense deposits seen in some rabbits in the laminae rarae (Figure 9) may represent S mutans components interacting with polyanionic sites of basement membranes.52

STREPTOCOCCAL NEPHRITIS IN RABBITS

Vol. 118 * No. 3

Similar patterns have been observed in GN seen in filaria-infested dogs.53 Spyogenes binding components react strongly with human and monkey GBM54; in indirect immunofluorescence, "double-track" patterns of staining have been observed that suggest binding to the laminae rarae (unpublished data). Streptococcal binding components include both lipoteichoic acid and proteins.55 S mutans also have proteins that bind to various basement membranes.56 These proteins have been identified by Western blot gel transfer techniques. Interestingly, two of the bands characterizing binding proteins have been recovered from eluates of diseased rabbit kidneys.56 The S mutans-induced nephritis reported here seems more severe and to occur more frequently than nephritis reported in other animal models of SBEGN and APSGN.13.14'42 Although the reasons for this still are not clear, there are three possible explanations. First, S mutans may have more affinity to rabbit renal basement membranes than other streptococci. With few exceptions,57 P-hemolytic streptococci do not infect experimental animals spontaneously; it has been suggested that P-hemolytic streptococci should not be expected to induce a disease in animals reproducing adequately the disease seen in man.12 Second, use of disrupted streptococci for immunization may have fulfilled the requirements for optimal reactivity with basement membrane components, as determined for the size of cationized ferritin.58 Third, severe disease may develop only after very long periods of administration of microbial components. Indeed, our rabbits were given S mutans longer than reported by other researchers for their experiments.13 The data reported in this paper encourage us to propose S mutans-induced nephritis in rabbits as a model for the diffuse form of SBEGN and, to some extent, also for APSGN in man and, at the same time, cause us to caution against the use of crude S mutans preparations as vaccines against dental caries. Further analyses of the immunopathology and the involvement of immunologic and microbial components in this disease should lead to a better understanding of the pathogenesis of SAN of man and should make it possible to identify and purify nonpathogenic but antigenic components of S mutans for use as a dental caries vaccine.

4. 5.

6. 7.

8. 9. 10.

11.

12.

13.

14.

15.

16. 17.

18. 19.

20.

References 1. Read SE, Zabriskie JB: Streptococcal disease and the immune response. New York, Academic Press, 1980 2. Paterson PY: Bacterial endocarditis, Biologic and Clinical Basis of Infectious Diseases. Edited by GP Youmans, PY Paterson, M Sommers. Philadelphia, W. B. Saunders Co., 1980, pp 692-706 3. Jennings RB, Earle DP: Poststreptococcal glomerulone-

21. 22.

23.

417

phritis: Histopathologic and clinical studies of the acute, subsiding acute and early chronic latent phases. J Clin Invest 1961, 40:1525-1595 Rammelkamp CH Jr: Acute hemorrhagic glomerulonephritis, Streptococcal Infections. Edited by M McCarty, New York, Columbia University Press, 1954, pp 197-207 Markowitz AS, Lange CF Jr: Streptococcal related glomerulonephritis: I. Isolation, immunochemistry, and comparative chemistry of soluble fractions from type 12 nephritogenic streptococci and human glomeruli. J Immunol 1964, 92:565-575 Holm SE: Precipitinogens in beta-hemolytic streptococci and some related human kidney antigens. Acta Pathol Microbiol Scand 1967, 70:79-94 Maxter WR, Potter EV: The presence of type 12 M protein in group G streptococci. J Gen Microbiol 1967, 49:119-125 Appel GB, Kunis CL: Acute tubulo-interstitial nephritis, Tubulo-interstitial Nephropathies. Edited by RS Cotran. New York, Churchill Livingstone, 1983, pp 175-185 Meadow SR: Post-streptococcal nephritis -A rare disease? Arch Dis Child 1975, 50:379-382 Koshi G, Webb JKG, Myers RM: Association of preceding streptococcal skin infection and acute glomerulonephritis in children in South India. Ind J Med Res 1968, 56:951-955 Koshi G, Sridharanm G, Thangavelu CP, Shastry JCM: Streptococcal antibodies and complement components in tropical post-streptococcal glomerulonephritis. Trans R Soc Trop Med Hyg 1983, 77:189-191 Heymer B, Krause RM: Streptococcal infections and nonsuppurative complications, Immunological Diseases. Edited by M Sampter. Boston, Little, Brown, 1978, pp 592-612 Michael AF, Kim Y: Pathogenesis of acute poststreptococcal glomerulonephritis, Streptococcal Diseases and the Immune Response. Edited by SE Read, JB Zabriskie. New York, Academic Press, 1980, pp 53-77 Arnold SB, Valone JA, Askenase PW, Kashgarian M, Freedman LR: Diffuse glomerulonephritis in rabbits with Streptococcus viridans endocarditis. Lab Invest 1975, 32:681-689 Becker CG: On the pathology of poststreptococcal glomerulonephritis in man and induced experimentally in rabbits, Streptococcal Diseases and the Immune Response. Edited by SE Read, JB Zabriskie. New York, Academic Press, 1980, pp 79-92 Ginsburg I: Streptococci, Medical Microbiology and Infectious Diseases. Edited by Al Braude. Philadelphia, W. B. Saunders Co., 1981, pp 281-295 Clarke JK: On the bacterial factor in the aetiology of dental caries. Br J Exp Pathol 1924, 5:141-151 Lehner T, Challacombe SJ, Caldwell J: Immunologic basis for vaccination against dental caries in rhesus monkeys. J Dent Res 1976, 55:c166-c180 Neefe LI, Chretien JH, Delaha EC, Garagusi VF: Streptococcus mutans endocarditis: Confusion with enterococcal endocarditis by routine laboratory testing. JAMA 1974, 230:1298-1299 Lockwood WR, Lawson LA, Smith DL, McNeill KM, Morrison FS: Streptococcus mutans endocarditis: Report of a case. Ann Intern Med 1974, 80:369-370 Harder EJ, Wilkowske CJ, Washington JA II, Geraci JE: Streptococcus mutans endocarditis. Ann Intern Med 1974, 80:364-368 Stinson MW, Nisengard RJ, Neiders M, Albini B: Serology and tissue lesions in rabbits immunized with Streptococcus mutans. J Immunol 1983, 131:3021-3027 van de Rijn I, Bleiweiss AS: Antigens of Streptococcus mutans: I. Characterization of a serotype-specific deter-

418

ALBINI ET AL

minant from Streptococcus mutans. Infect Immun 1973, 7:795-804 24. Sunderman FW, Sunderman FW Jr: Creatinine. Proficiency Test Service. Institute for Clinical Service, Philadelphia, February 1973 25. Shevky MC, Stafford DD: Clinical methods for estimation of protein in urine and other body fluids. Arch Intern Med 1923, 32:222-224 26. Karnovsky MJ: A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 1965, 27:137A-138A 27. Lillie RD: Histopathologic technique and practical histochemistry. 3rd edition. New York, McGraw-Hill Book Co., 1965 28. Zollinger HU, Mihatsch MJ: Renal pathology in biopsy. New York, Springer, 1978 29. Wright J, Calkins E, Humphrey R: Potassium permanganate reaction in amyloidosis. Lab Invest 1977, 36:274-281 30. Baldwin DS, Gluck MC, Schacht RG, Gallo G: The longterm course of poststreptococcal glomerulonephritis. Ann Inter Med 1974, 80:342-358 31. Holubar K, Stingl G, Albini B: Praxis und Methodologie der definierten Imunfluoreszenztechnik [Techniques and methodology of defined immunofluorescence]. Hautarzt 1976, 27:78-89 32. Beutner EH, Sepulveda MR, Barnett EV: Quantitation studies of immunofluorescent staining. Bull WHO 1968, 39:587-606 33. Vassar PS, Culling CFA: Fluorescent stains with special reference to amyloid and connective tissue. Arch Pathol 1959, 68:487-498 34. Horder J: Infective endocarditis with an analysis of 150 cases and with special reference to the chronic form of the disease. Q J Medicine 1908, 2:289-293 35. Lhlein M: Uber hamorrhagische Nierenaffektionen bei chronischer ulzeroser Endokarditis [On hemorrhagic kidney involvement in chronic ulcerative endocarditis]. Med Klin 1910, 6:375-379 36. Longcope WT: The susceptibility of man to foreign proteins. Am J Med Sci 1916, 152:625-637 37. Bain RC, Edwards JE, Scheifley CH, Geraci JE: Right sided bacterial endocarditis and endarteritis: A clinical and pathological study. Am J Med 1958, 24:98-110 38. Williams RC Jr, Kunkel HG: Rheumatoid factor, complement, and conglutinin abberations in patients with subacute bacterial endocarditis. J Clin Invest 1962, 41:666-674 39. Levy RL, Hong R: The immune nature of subacute bacterial endocarditis (SBE) nephritis. Am J Med 1973, 54:645-652 40. Gutman RA, Striker GE, Gilliland BC, Cutler RE: The immune complex glomerulonephritis of bacterial endocarditis. Medicine 1972, 51:1-25 41. Garrison PK, Freedman LR: Experimental endocarditis: I. Staphylococcal endocarditis in rabbits resulting from placement of a polyethylene catheter in the right side of the heart. Yale J Biol Med 1970, 42:394-410 42. Sindrey M, Barratt J, Hewitt J, Naish P: Infective endocarditis-associated glomerulonephritis in rabbits: Evidence of a pathogenetic role for antiglobulins. Clin Exp Immunol 1981, 45:253-260 43. Movat HZ, Steiner JW, Herhn J: The fine structure of

AJP * March 1985

44.

45. 46. 47.

48. 49. 50.

51.

52.

53.

54.

55. 56.

57.

58.

the glomerulus in acute glomerulonephritis. Lab Invest 1962, 11:117-135 Dodge WF, Spargo BH, Travis LB, Srivastava RN, Carvajal H, De Benkelaer MR, Longley MP, Menchaca JA: Poststreptococcal glomerulonephritis, a prospective study in children. N Engl J Med 1972, 286:273-278 Heptinstall RH: Pathology of the Kidney. 2nd edition. Boston, Little, Brown, 1974 Glassock RJ, Bennett CM: The glomerulopathies, The Kidney. Edited by BM Brenner and FC Rector. Philadelphia, W. B. Saunders Co., pp 941-1078 Hinglais N, Grunfeld JP, Garcia-Torres R: Acute glomerulonephritis. In Nephrology. Edited by H Hamburger, J Crosnier, JP Grunfeld. New York, Wiley-Flammarion, 1979, pp 455-473 Michael AF Jr, Drummond KN, Good RA, Vernier RL: Acute poststreptococcal glomerulonephritis: Immune deposit disease. J Clin Invest 1966, 45:237-248 McIntosh RM: Cryoprotein in poststreptococcal glomerulonephritis. Ann Intern Med 1970, 73:857 Bayer AS, Theofilopoulos AN, Dixon FJ, Guze LB: Circulating immune complexes in experimental streptococcal endocarditis: A monitor of therapeutic efficacy. J Infect Dis 1979, 139:1-8 Andres GA, Accinni L, Hsu KC, Zabriskie JB, Seegal BC: Electron microscopic studies of human glomerulonephritis with ferritin-conjugated antibody: Localization of antigen-antibody complexes in glomerular structures of patients with acute glomerulonephritis. J Exp Med 1966, 123:399-412 Caulfield JP, Farquhar MG: Distribution of anionic sites in glomerular basement membranes: Their possible role in filtration and attachment. Proc Natl Acad Sci USA, 1976, 73:1646-1650 Aikawa M, Abramowsky C, Powers KG, Furrow R: Dirofilariasis: IV. Glomerulonephropathy induced by Dirofilaria immitis infection. Am J Trop Med Hyg 1981, 30:84-91 Stinson MW, Bergey EJ: Isolation of a heart- and a kidney-binding protein from Group A streptococci. Infect Immun 1982, 35:335-342 Beachey EH: Bacterial adherence: Adhesin-receptor interactions mediating the attachment of bacteria to mucosal surfaces. J Infect Dis 1981, 143:325-345 Stinson MW, Barua K, Nisengard R, Neiders M, Albini B: Binding of Streptococcus mutans antigen to heart and kidney basement membranes. Infect Immun 1984, 46: 145-151 Lancefield RC: Group A streptococcal infections in animals -Natural and experimental, Streptococci and Streptococcal Diseases. Edited by LW Wannemaker, JM Matsen. New York, Academic Press, 1972, pp 313-326 Vogt A, Schmidt HV, Takamiya H, Batsford S: "In situ" immune complex nephritis and basic proteins. Proc EDTA 1980, 17:613-619

Acknowledgments We thank Dr. J. R. Wright for assisting us in the interpretation of Congo red stain for amyloid and Drs. G. Andres and F. Milgrom for helpful discussion in the course of the studies reported here.