Circulating Anti-DNA Immune Complexes in Active Lupus Nephritis TAKESHISASAKI,M.D., TAI MURYOI, M.D., AKIRAHATAKEYAMA,M.D., MASAHIKOSUZUKI, M.D., HIROSHISATO,M.D., JIN SEINO,M.D., TAKAOSAITO,M.D., KAORUYOSHINAGA,M.D., Sendai, Japan
PURPOSE The role that circulating anti-DNA immune complexes play in autoimmunity has not yet been elucidated in humans. Theaimofthis study was to relate circulating anti-DNA immune complexes to a variety of renal histologic features and to immunoglobulin deposits in active lupus nephritis. PATIENTS AND METHODS: The study population consisted of 47 patients with active lupus nephritis, 28 with active systemic lupus erythematosus (SLE) in the absence of renal lesions, and 40 with other categories of the disease. All patients were e xamined for anti-DNA circulating immune complexes (CIC) and their anti-DNA idiotype expression by an isoelectrofocusing analysis. Patients with renal lesions were also examined for renal histologic and immunofluorescent findings in renal biopsy specimens. IWXJLXR Anti-DNA CIC expressing an antiDNA idiotype termed O-81 Id occurred in patients with active lupus nephritis but not in acute episodes lathing renal involvement or in remission. Positive test results for anti-DNA CIC were associated with the incidence of diffuse proliferative glomerulonephritis (DPGN). Patients with anti-DNA CIC were also found to have a statistically significant increase in the prevalence of immunoglobuIin immune deposits in the subendothelial area of the renal glomeruli. CONCLUSION: The fmdings suggest that antiDNA CIC preferentially occurred in lupus patients with DPGN. Examination for anti-DNA CIC may be a useful predictor of renal lesions, and therefore may contribute to the management of SIX The results also indicate that antiFrom the Second Department of Internal Medicine, Tohoku University School of Medicine, Sendai, Japan. This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science and Culture, and by a research grant for autoimmune diseases and progressive renal diseases from the Ministry of Health and Welfare, Japan. Requests for reprints should be addressed toTakeshi Sasaki, M.D., The Second Department of Internal Medicine, Tohoku University School of Medicine, Seiryocho 1-l. Sendai, 980, Japan. Manuscript submitted August 15, 1990, and accepted in revised form June 12, 1991.
DNA CIC may be associated with immunoglobulin deposition in the subendothelial area of the renal glomeruli.
upus nephritis is defined by the occurrence of a variety of histologic changes in renal glomeruli as classified by World Health Organization (WHO) criteria [l]. These histologic findings obtained from renal biopsy specimens provide important information about the management of lupus patients [2-4]. Renal biopsy specimens, however, are difficult to obtain in some cases. Consequently, it is important to find serologic markers to predict renal injury in patients. Systemic lupus erythematosus (SLE) is a representative immune complex disease, and circulating immune complexes (CIC) have frequently been found in sera from patients with lupus nephritis. It has not, however, been possible to discriminate between nonharmful and nephritogenic immune complexes through the use of currently available assays. Mounting data have revealed that anti-DNA antibodies and their immune complexes might play a role in the pathogenesis of lupus nephritis [5-lo]. Definite evidence demonstrating a causative role for anti-DNA CIC, however, is sparse and the results are controversial [ 1 l-141. Therefore, it is still not clear which characteristics of lupus nephritis are associated with anti-DNA CIC. We have demonstrated that it is possible to separate antibodies from CIC using isoelectrofocusing with 6 M urea [15]. The study also enabled us to identify specific clonotypes of anti-DNA antibodies in CIC of patients with active lupus nephritis. Many CIC-derived antibodies expressed the anti-DNA idiotypes termed O-81 Id and NE-l Id. These idiotypes were also detected on antibodies in immune deposits in renal glomeruli of patients with lupus [15,16]. Thus, anti-DNA antibodies expressing O-81 Id and NE-l Id may be nephritogenic. These studies allowed us to investigate the relationship between anti-DNA CIC and a variety of renal lesions in lupus patients. Here, we present evidence that anti-DNA CIC are closely associated with the occurrence of diffuse proliferative glomerulonephritis in patients with lupus.
L
October
1991 The American
Journal
of Medicine
Volume
91
355
ANTI-DNA
IMMUNE
COMPLEXES
IN LUPUS
NEPHRITIS
/ SASAKI ET AL
PATIENTS AND METHODS Patients All SLE patients in this study fulfilled the 1982 American Rheumatism Association criteria for the diagnosis of SLE. Disease activity was defined by clinical signs and symptoms (nephritis, arthritis, serositis, cerebral signs) according to the criteria previously described [17,18]. All of the patients with active lupus nephritis presented here showed abnormal findings on urinalysis, reduced total hemolytic complement (CH& levels, and active lupus nephritis as determined by renal biopsy [1,19]. Reagent Two types of monoclonal anti-idiotypic antibody (DlE2 and lF5) to human monoclonal anti-DNA antibody were obtained from mouse hybridoma using purified monoclonal anti-ssDNA antibody (O81) or monoclonal anti-dsDNA antibody (NE-l) [20]. The O-81 antibody binds preferentially to sshomopolymer with pyrimidine bases, whereas NE-1 reacts not only with ssDNA but also with dsDNA and zDNA [21,22]. DlE2 specifically binds to an idiotype found in the antigen-binding site of O-81 but not NE-l. lF5 reacts with NE-l, but not with O81 [23]. Isolation of immune Complexes from Plasma Each sample was precipitated by polyethylene glycol (at a final concentration of 3.5%), dissolved in 0.15 M phosphate-buffered saline (PBS), pH 7.2, and applied to an affinity-purified anti-C3 IgGcoupled Sepharose column [24]. After washing with PBS, the immune complexes were eluted with 0.15 M glycine-hydrochloric acid (HCl) buffer, pH 2.3, immediately neutralized with 3 M Tris-HCl, pH 10.0, and stored at -70°C until use. Measurement of Anti-DNA Antibodies Anti-DNA antibodies in sera were determined by a modified Farr assay [22]. In short, a 25-PL aliquot of sera was mixed with 100 rL (3 X lo4 cpm [counts per minute]) of 1251-labeled Escherichia coli plasmid DNA (1,000 bp [base pairs], obtained from Nippon DPC Corporation, Chiba, Japan). After incubation for 90 minutes at room temperature, an equal volume of saturated ammonium sulfate was added to the mixture. The immune complexes were precipitated by centrifugation (2,OOOXg) at 4OC for 15 minutes, and the radioactivity was assessed by a gamma-counter. The anti-DNA antibody titer was expressed as units calculated from a standard curve using serially diluted positive sera. Any result above 6 units was regarded as positive for anti-DNA antibody tests.
356
October
1991 The American
Journal
of Medicine
Volume
Isoelectricfocusing (IEF) and lmmunoblotting Flat-bed IEF was carried out in a 5% polyacrylamide gel (10 X 20 X 0.1 cm) containing 6.3% pharmalytes pH 3-11 (Pharmacia Fine Chemicals AB, Uppsala, Sweden), 13% glycerol, and 6 M urea using a Pharmacia FBE-3000 system [15]. The gel was prefocused at 8 W for 20 minutes at 4OC. Twenty micrograms of the sample, diluted to a final volume of 60 rL with 6 M urea, was applied to the gel surface using Whatman number 1 filter paper wicks (Whatman International Ltd., Maidstone, England). The gel was then focused at 30 W to a maximum of 3,000 V/hour at 4OC. The pH gradient was measured using a flat membrane pH electrode or IEF calibration kit. The focused gel was placed in a gel transfer holder, and blotted electrophoretically onto a nitrocellulose membrane in 0.7% acetic acid solution at 4OC for 30 minutes at 30 V, and then for an additional 30 minutes at 150 V. After blotting, free binding sites on the membrane were blocked by incubation in 100 mL of 1.5% bovine serum albumin (BSA) in PBS containing 0.1% Tween for 1 hour at 37°C on a rocker platform. The membrane was rinsed three times with PBS containing 1.5% BSA and 0.1% Tween 20 (washing buffer), overlayed with 10 mL (2 X lo6 cpm) of a 1251-labeled probe (ssDNA, dsDNA, DlE2, lF5, or mouse IgG), and then incubated for 90 minutes at room temperature on the rocker. The membrane was rinsed while being rocked for 16 hours at room temperature with several changes of washing buffer, then dried and exposed to x-ray film (Kodak, X-Omat, Eastman Kodak Company, Rochester, New York) at -7OOC. Finally, the band formation with each probe was regarded as positive [15]. Renal Histology and Detection of lmmunoglobulins in Biopsy Specimens of the Kidneys Histologic characteristics of renal biopsy specimens, all of which contained at least 10 glomeruli, were studied with light microscopy, immunofluorescence technique, and electron microscopy using standard methodologies. Tissues were stained with hematoxylin and eosin, periodic acid-Schiff, and azan Mallory. According to WHO criteria [ 11, the 47 biopsy specimens were classified into four groups as follows: 10 pure mesangial alteration, three focal segmental glomerulonephritis (FSGN), 22 diffuse proliferative glomerulonephritis (DPGN), and 12 diffuse membranous glomerulonephritis. The activity index or chronicity index was defined as the sum of individual scores of the items presented by Austin et al [2]. Portions of specimens obtained from renal biopsy were also snap-frozen and cut at 5 pm for immuno-
91
ANTI-DNA
IMMUNE
COMPLEXES
IN LUPUS
NEPHRITIS
/ SASAKI ET AL
fluorescent microscopy. Each section was stained with fluorescein-conjugated monospecific (FITC) antisera to human IgG, IgM, IgA, and C3 (Dako Corporation, Santa Barbara, California). Using light, electron, and immunofluorescent microscopy, the immune deposits were evaluated by three histology specialists according to their distribution: mesangial, subendothelial, or subepithelial area. The intensity of deposits was also semiquantitatively graded from 0 to 3+: negative (O), mild (l+), moderate (2+), and severe (3+). In order to detect anti-DNA idiotypes in glomeruli, the other sections in 24 cases were first incubated with F(ab’)a fragments of DlE2, lF5, or mouse IgG (1 mg/mL) at a dilution of 1:25 in a moist atmosphere for 1 hour at 37OC. After being washed for 20 minutes in PBS, the sections were incubated with FITC-conjugated rabbit antimouse IgG F(ab’)g (Jackson Immunoresearch Laboratories, Avondale, Pennsylvania) diluted 1:25 in PBS for 30 minutes. The sections were washed again in PBS for 20 minutes, rinsed in diluted buffer, and dried. The sections were mounted in buffered glycerol and examined under fluorescence microscope. Statistical Analysis The data were analyzed using Student’s t-test, and results with a p value of less than 0.05 were considered to be statistically significant.
RESULTS Case Reports PATIENT 3: A 15-year-old high school student was first hospitalized in July 1983 because of facial erythema and fever (Figure 1, top). Six months earlier, she had complained of transient finger joint pain. At that time, the results of urinalysis were normal despite the presence of high titers of antiDNA antibodies in her serum. Her clinical and laboratory findings were improved by treatment with prednisolone (40 mg/day). She remained clinically well until 3 years later, when she developed massive proteinuria necessitating hospitalization. A renal biopsy revealed DPGN, which included various types of immune deposits in the mesangial, subendothelial, and subepithelial areas (Figure 1, middle, and Patient 3 data in Table I). Immunofluorescent studies using anti-idiotypic antibodies (DlE2) revealed the deposition of immunoglobulin expressing the O-81 Id in the renal glomeruli (Figure 1, bottom). PATIENT 10: A 23-year-old woman experienced transient polyarthralgia without swelling of her joints in October 1977. Two months later, a butterfly rash developed with a slight fever. Urinary pro-
Figure 1. Top, serial studies in Patient 3. Ab = antibody; CY = cyclophosphamide. Mlddle, O-81 Id immune deposits in renal biopsy specimen (original magnification X250, reduced by 30%). Bottom, electron microscopic finding in renal glomeruIi.
tein was negative. Anti-DNA antibodies were positive but became undetectable with her clinical improvement after prednisolone therapy. She was well and experienced two pregnancies with successful births in 1982 and 1984. In August 1987, she had an episode of fever and facial erythema (Figure 2). Urinalysis disclosed numerous red blood cells under high-power microscopy, and urinary protein levels
October
1991 The American
Journal
of Medicine
Volume
91
357
ANTI-DNA IMMUNE COMPLEXES IN LUPUS NEPHRITIS / SASAKI ET AL TABLE I Anti-DNA CIC and Immune Depositsin Renal Glomeruli
Patient
Age
Sex
ssDNA
24 24
F
17
F F F F F
t t t t t t t t t t t t t t t -
33 23 32 28
24 46 37 17
F
F
F F F F F
22 30 25 M 46 F 41 14
22 31
F F F F
24
F
22 20
F F F M
19
31
CIC dsDNA O-81 Id
t t -
t t t t t t t t t t t t t t t
NE-1 Id
t -
Deposits in Glomerulit O-81 Id NE-1 Id
Renal Histologic Features* Pure mesangial alteration Pure mesangial alteration DPGN (IVc) DPGN (IVc) DPGN UVc) Membranous (Vb) DPGN (IVc) DPGN (IVc) DPGN (IVc) Membranous (Vd) DPGN (IVb) DPGN (IVb) DPGN (IVd) DPGN (IVb) DPGN (IVd) Pure mesangial alteration Pure mesangial alteration Pure mesangial alteration FSGN (Illb) FSGN (Illb) Membranous (Vc) Membranous (Va) Membranous (Va) Membranous (Vb)
t t t t t t t t t t t t
(Ilb) HIa)
Ig Deposits in Subendothelium Subepithelium
Mesangium
t t -
t t t t tt t t tt tt tt t t tt tt tt t t t t t tt
(Ilb) (Ila) (Ilb)
t t tt t t tt tt t tt t tt t tt t t t tt tt t t
t
Ig:= immunoglobulin;GN = giomerulonephritis. *WHO classification HI: Ila = pure mesangialalteration with mesangialwidening and/or mild hypercellularity; Ilb = pure mesangialalteration with moderate hyparcellularity; llla = FSGN with active necmtizing lesions; lllb = FSGNwith active and sclerosing lesions; lllc = FSGN with sclerosing lesions; IVa= diffuse GN without segmental lesions; IVb = diffuse GN with active necrotizinglesions; IVc = diftuseGN with activeand sclerosinglesions; IVd = diffuse GN with sclerosinglesions; Va = pure membranousglomerulonephritis;Vb = diffuse membranousGN associated with lesionsof category ha, b; Vc = dike
membranousGN associated with lesionsof category Illa, b, c; Vd = diffuse membranous GN associatedwith lesionsof category IVa, b, c, d.
tTheexpression of anti-DNA Id TO-81ld or NE-lid) on the deposits in renal glomeruliwas tested by an immunotlwrescence study as described in Patients and Methods, The distribution and the intensity of immunedepositswere determined by light, electron, and immunotluorescencemicroscopy as described in Patients and Methods.
were 3.5 to 5.0 g/day. Renal biopsy showed diffuse membranous glomerulonephritis (class Vd) including massive immune deposits in mesangial, subendothelial, and subepithelial areas (Patient 10 data in Table I). Combination therapy with prednisolone and cyclophosphamide improved her state, and the urinary protein level fell to lower than 0.5 g/day. Since then she has been well.
As shown in Figures 1 and 2, both cases were positive for anti-DNA CIC expressing the O-81 Id when analyzed by IEF during an acute episode with renal involvement. Anti-DNA CIC, however, could not be detected during remission or in acute episodes in which proteinuria was absent. Similar results were obtained with other SLE cases. Table II shows that anti-DNA CIC expressing O-81 Id or NE-1 Id specifically occur in patients with active lupus nephritis. On the other hand, anti-DNA CIC were less frequently found in patients without lupus nephritis during acute episodes. It was also noteworthy that O-81 Id-positive immune complexes were never found in the latter group. 1
TABLE II Incidence of Anti-DNA CIC
Categories
Figure
2. Serial cyclophosphamide.
358
studies
in Patient
10. Ab = antibody:
October 1991 The American Journal of Medicine
Active lupus nephritis Active SLE without lupus nephritis Other diseases including inactive SLE
CY =
Ab = antibody.
Volume 91
Percent Frequency of CIC with Number AntiAntiof Cases ssDNAAb dsDNA Ab O-81 Id NE-1 Id
58
2
17 0
58
ii
0
I?
40
0
0
0
0
ANTI-DNA IMMUNE COMPLEXES IN LUPUS NEPHRITIS / SASAKI ET AL
TABLE Ill
Between Renal Histologic Featuresand CIC
Relationship
Number of Cases
CIC Anti-ssDNA Ab t
II
Renal Histologic Features (WHO Classification) IV
Ill
6
:
20 2I
3:
i
i
1:
;i
z
:
20 2I
4;
1:
i
2:
zl
4*
Anti-dsDNA Ab
V
p to.01
1;
1
p CO.01
t
0-h
1;
Id
t NE-1 Id
t -
p tt
Anti-ssDNA Ab t Anti-dsDNA Ab t 0-i Id t I
Ab = antibody. *Immune deposits in mesangial,subendothelial,or subeoithelialarea of renal ~lomeruliwere gradedfrom - to 2 t t fNumkrsoicases. -
globulin deposits in this area (p
PURPOSE The role that circulating anti-DNA immune complexes play in autoimmunity has not yet been elucidated in humans. Theaimofthis study was to relate circulating anti-DNA immune complexes to a variety of renal histologic features and to immunoglobulin deposits in active lupus nephritis. PATIENTS AND METHODS: The study population consisted of 47 patients with active lupus nephritis, 28 with active systemic lupus erythematosus (SLE) in the absence of renal lesions, and 40 with other categories of the disease. All patients were e xamined for anti-DNA circulating immune complexes (CIC) and their anti-DNA idiotype expression by an isoelectrofocusing analysis. Patients with renal lesions were also examined for renal histologic and immunofluorescent findings in renal biopsy specimens. IWXJLXR Anti-DNA CIC expressing an antiDNA idiotype termed O-81 Id occurred in patients with active lupus nephritis but not in acute episodes lathing renal involvement or in remission. Positive test results for anti-DNA CIC were associated with the incidence of diffuse proliferative glomerulonephritis (DPGN). Patients with anti-DNA CIC were also found to have a statistically significant increase in the prevalence of immunoglobuIin immune deposits in the subendothelial area of the renal glomeruli. CONCLUSION: The fmdings suggest that antiDNA CIC preferentially occurred in lupus patients with DPGN. Examination for anti-DNA CIC may be a useful predictor of renal lesions, and therefore may contribute to the management of SIX The results also indicate that antiFrom the Second Department of Internal Medicine, Tohoku University School of Medicine, Sendai, Japan. This work was supported by a grant-in-aid for scientific research from the Ministry of Education, Science and Culture, and by a research grant for autoimmune diseases and progressive renal diseases from the Ministry of Health and Welfare, Japan. Requests for reprints should be addressed toTakeshi Sasaki, M.D., The Second Department of Internal Medicine, Tohoku University School of Medicine, Seiryocho 1-l. Sendai, 980, Japan. Manuscript submitted August 15, 1990, and accepted in revised form June 12, 1991.
DNA CIC may be associated with immunoglobulin deposition in the subendothelial area of the renal glomeruli.
upus nephritis is defined by the occurrence of a variety of histologic changes in renal glomeruli as classified by World Health Organization (WHO) criteria [l]. These histologic findings obtained from renal biopsy specimens provide important information about the management of lupus patients [2-4]. Renal biopsy specimens, however, are difficult to obtain in some cases. Consequently, it is important to find serologic markers to predict renal injury in patients. Systemic lupus erythematosus (SLE) is a representative immune complex disease, and circulating immune complexes (CIC) have frequently been found in sera from patients with lupus nephritis. It has not, however, been possible to discriminate between nonharmful and nephritogenic immune complexes through the use of currently available assays. Mounting data have revealed that anti-DNA antibodies and their immune complexes might play a role in the pathogenesis of lupus nephritis [5-lo]. Definite evidence demonstrating a causative role for anti-DNA CIC, however, is sparse and the results are controversial [ 1 l-141. Therefore, it is still not clear which characteristics of lupus nephritis are associated with anti-DNA CIC. We have demonstrated that it is possible to separate antibodies from CIC using isoelectrofocusing with 6 M urea [15]. The study also enabled us to identify specific clonotypes of anti-DNA antibodies in CIC of patients with active lupus nephritis. Many CIC-derived antibodies expressed the anti-DNA idiotypes termed O-81 Id and NE-l Id. These idiotypes were also detected on antibodies in immune deposits in renal glomeruli of patients with lupus [15,16]. Thus, anti-DNA antibodies expressing O-81 Id and NE-l Id may be nephritogenic. These studies allowed us to investigate the relationship between anti-DNA CIC and a variety of renal lesions in lupus patients. Here, we present evidence that anti-DNA CIC are closely associated with the occurrence of diffuse proliferative glomerulonephritis in patients with lupus.
L
October
1991 The American
Journal
of Medicine
Volume
91
355
ANTI-DNA
IMMUNE
COMPLEXES
IN LUPUS
NEPHRITIS
/ SASAKI ET AL
PATIENTS AND METHODS Patients All SLE patients in this study fulfilled the 1982 American Rheumatism Association criteria for the diagnosis of SLE. Disease activity was defined by clinical signs and symptoms (nephritis, arthritis, serositis, cerebral signs) according to the criteria previously described [17,18]. All of the patients with active lupus nephritis presented here showed abnormal findings on urinalysis, reduced total hemolytic complement (CH& levels, and active lupus nephritis as determined by renal biopsy [1,19]. Reagent Two types of monoclonal anti-idiotypic antibody (DlE2 and lF5) to human monoclonal anti-DNA antibody were obtained from mouse hybridoma using purified monoclonal anti-ssDNA antibody (O81) or monoclonal anti-dsDNA antibody (NE-l) [20]. The O-81 antibody binds preferentially to sshomopolymer with pyrimidine bases, whereas NE-1 reacts not only with ssDNA but also with dsDNA and zDNA [21,22]. DlE2 specifically binds to an idiotype found in the antigen-binding site of O-81 but not NE-l. lF5 reacts with NE-l, but not with O81 [23]. Isolation of immune Complexes from Plasma Each sample was precipitated by polyethylene glycol (at a final concentration of 3.5%), dissolved in 0.15 M phosphate-buffered saline (PBS), pH 7.2, and applied to an affinity-purified anti-C3 IgGcoupled Sepharose column [24]. After washing with PBS, the immune complexes were eluted with 0.15 M glycine-hydrochloric acid (HCl) buffer, pH 2.3, immediately neutralized with 3 M Tris-HCl, pH 10.0, and stored at -70°C until use. Measurement of Anti-DNA Antibodies Anti-DNA antibodies in sera were determined by a modified Farr assay [22]. In short, a 25-PL aliquot of sera was mixed with 100 rL (3 X lo4 cpm [counts per minute]) of 1251-labeled Escherichia coli plasmid DNA (1,000 bp [base pairs], obtained from Nippon DPC Corporation, Chiba, Japan). After incubation for 90 minutes at room temperature, an equal volume of saturated ammonium sulfate was added to the mixture. The immune complexes were precipitated by centrifugation (2,OOOXg) at 4OC for 15 minutes, and the radioactivity was assessed by a gamma-counter. The anti-DNA antibody titer was expressed as units calculated from a standard curve using serially diluted positive sera. Any result above 6 units was regarded as positive for anti-DNA antibody tests.
356
October
1991 The American
Journal
of Medicine
Volume
Isoelectricfocusing (IEF) and lmmunoblotting Flat-bed IEF was carried out in a 5% polyacrylamide gel (10 X 20 X 0.1 cm) containing 6.3% pharmalytes pH 3-11 (Pharmacia Fine Chemicals AB, Uppsala, Sweden), 13% glycerol, and 6 M urea using a Pharmacia FBE-3000 system [15]. The gel was prefocused at 8 W for 20 minutes at 4OC. Twenty micrograms of the sample, diluted to a final volume of 60 rL with 6 M urea, was applied to the gel surface using Whatman number 1 filter paper wicks (Whatman International Ltd., Maidstone, England). The gel was then focused at 30 W to a maximum of 3,000 V/hour at 4OC. The pH gradient was measured using a flat membrane pH electrode or IEF calibration kit. The focused gel was placed in a gel transfer holder, and blotted electrophoretically onto a nitrocellulose membrane in 0.7% acetic acid solution at 4OC for 30 minutes at 30 V, and then for an additional 30 minutes at 150 V. After blotting, free binding sites on the membrane were blocked by incubation in 100 mL of 1.5% bovine serum albumin (BSA) in PBS containing 0.1% Tween for 1 hour at 37°C on a rocker platform. The membrane was rinsed three times with PBS containing 1.5% BSA and 0.1% Tween 20 (washing buffer), overlayed with 10 mL (2 X lo6 cpm) of a 1251-labeled probe (ssDNA, dsDNA, DlE2, lF5, or mouse IgG), and then incubated for 90 minutes at room temperature on the rocker. The membrane was rinsed while being rocked for 16 hours at room temperature with several changes of washing buffer, then dried and exposed to x-ray film (Kodak, X-Omat, Eastman Kodak Company, Rochester, New York) at -7OOC. Finally, the band formation with each probe was regarded as positive [15]. Renal Histology and Detection of lmmunoglobulins in Biopsy Specimens of the Kidneys Histologic characteristics of renal biopsy specimens, all of which contained at least 10 glomeruli, were studied with light microscopy, immunofluorescence technique, and electron microscopy using standard methodologies. Tissues were stained with hematoxylin and eosin, periodic acid-Schiff, and azan Mallory. According to WHO criteria [ 11, the 47 biopsy specimens were classified into four groups as follows: 10 pure mesangial alteration, three focal segmental glomerulonephritis (FSGN), 22 diffuse proliferative glomerulonephritis (DPGN), and 12 diffuse membranous glomerulonephritis. The activity index or chronicity index was defined as the sum of individual scores of the items presented by Austin et al [2]. Portions of specimens obtained from renal biopsy were also snap-frozen and cut at 5 pm for immuno-
91
ANTI-DNA
IMMUNE
COMPLEXES
IN LUPUS
NEPHRITIS
/ SASAKI ET AL
fluorescent microscopy. Each section was stained with fluorescein-conjugated monospecific (FITC) antisera to human IgG, IgM, IgA, and C3 (Dako Corporation, Santa Barbara, California). Using light, electron, and immunofluorescent microscopy, the immune deposits were evaluated by three histology specialists according to their distribution: mesangial, subendothelial, or subepithelial area. The intensity of deposits was also semiquantitatively graded from 0 to 3+: negative (O), mild (l+), moderate (2+), and severe (3+). In order to detect anti-DNA idiotypes in glomeruli, the other sections in 24 cases were first incubated with F(ab’)a fragments of DlE2, lF5, or mouse IgG (1 mg/mL) at a dilution of 1:25 in a moist atmosphere for 1 hour at 37OC. After being washed for 20 minutes in PBS, the sections were incubated with FITC-conjugated rabbit antimouse IgG F(ab’)g (Jackson Immunoresearch Laboratories, Avondale, Pennsylvania) diluted 1:25 in PBS for 30 minutes. The sections were washed again in PBS for 20 minutes, rinsed in diluted buffer, and dried. The sections were mounted in buffered glycerol and examined under fluorescence microscope. Statistical Analysis The data were analyzed using Student’s t-test, and results with a p value of less than 0.05 were considered to be statistically significant.
RESULTS Case Reports PATIENT 3: A 15-year-old high school student was first hospitalized in July 1983 because of facial erythema and fever (Figure 1, top). Six months earlier, she had complained of transient finger joint pain. At that time, the results of urinalysis were normal despite the presence of high titers of antiDNA antibodies in her serum. Her clinical and laboratory findings were improved by treatment with prednisolone (40 mg/day). She remained clinically well until 3 years later, when she developed massive proteinuria necessitating hospitalization. A renal biopsy revealed DPGN, which included various types of immune deposits in the mesangial, subendothelial, and subepithelial areas (Figure 1, middle, and Patient 3 data in Table I). Immunofluorescent studies using anti-idiotypic antibodies (DlE2) revealed the deposition of immunoglobulin expressing the O-81 Id in the renal glomeruli (Figure 1, bottom). PATIENT 10: A 23-year-old woman experienced transient polyarthralgia without swelling of her joints in October 1977. Two months later, a butterfly rash developed with a slight fever. Urinary pro-
Figure 1. Top, serial studies in Patient 3. Ab = antibody; CY = cyclophosphamide. Mlddle, O-81 Id immune deposits in renal biopsy specimen (original magnification X250, reduced by 30%). Bottom, electron microscopic finding in renal glomeruIi.
tein was negative. Anti-DNA antibodies were positive but became undetectable with her clinical improvement after prednisolone therapy. She was well and experienced two pregnancies with successful births in 1982 and 1984. In August 1987, she had an episode of fever and facial erythema (Figure 2). Urinalysis disclosed numerous red blood cells under high-power microscopy, and urinary protein levels
October
1991 The American
Journal
of Medicine
Volume
91
357
ANTI-DNA IMMUNE COMPLEXES IN LUPUS NEPHRITIS / SASAKI ET AL TABLE I Anti-DNA CIC and Immune Depositsin Renal Glomeruli
Patient
Age
Sex
ssDNA
24 24
F
17
F F F F F
t t t t t t t t t t t t t t t -
33 23 32 28
24 46 37 17
F
F
F F F F F
22 30 25 M 46 F 41 14
22 31
F F F F
24
F
22 20
F F F M
19
31
CIC dsDNA O-81 Id
t t -
t t t t t t t t t t t t t t t
NE-1 Id
t -
Deposits in Glomerulit O-81 Id NE-1 Id
Renal Histologic Features* Pure mesangial alteration Pure mesangial alteration DPGN (IVc) DPGN (IVc) DPGN UVc) Membranous (Vb) DPGN (IVc) DPGN (IVc) DPGN (IVc) Membranous (Vd) DPGN (IVb) DPGN (IVb) DPGN (IVd) DPGN (IVb) DPGN (IVd) Pure mesangial alteration Pure mesangial alteration Pure mesangial alteration FSGN (Illb) FSGN (Illb) Membranous (Vc) Membranous (Va) Membranous (Va) Membranous (Vb)
t t t t t t t t t t t t
(Ilb) HIa)
Ig Deposits in Subendothelium Subepithelium
Mesangium
t t -
t t t t tt t t tt tt tt t t tt tt tt t t t t t tt
(Ilb) (Ila) (Ilb)
t t tt t t tt tt t tt t tt t tt t t t tt tt t t
t
Ig:= immunoglobulin;GN = giomerulonephritis. *WHO classification HI: Ila = pure mesangialalteration with mesangialwidening and/or mild hypercellularity; Ilb = pure mesangialalteration with moderate hyparcellularity; llla = FSGN with active necmtizing lesions; lllb = FSGNwith active and sclerosing lesions; lllc = FSGN with sclerosing lesions; IVa= diffuse GN without segmental lesions; IVb = diffuse GN with active necrotizinglesions; IVc = diftuseGN with activeand sclerosinglesions; IVd = diffuse GN with sclerosinglesions; Va = pure membranousglomerulonephritis;Vb = diffuse membranousGN associated with lesionsof category ha, b; Vc = dike
membranousGN associated with lesionsof category Illa, b, c; Vd = diffuse membranous GN associatedwith lesionsof category IVa, b, c, d.
tTheexpression of anti-DNA Id TO-81ld or NE-lid) on the deposits in renal glomeruliwas tested by an immunotlwrescence study as described in Patients and Methods, The distribution and the intensity of immunedepositswere determined by light, electron, and immunotluorescencemicroscopy as described in Patients and Methods.
were 3.5 to 5.0 g/day. Renal biopsy showed diffuse membranous glomerulonephritis (class Vd) including massive immune deposits in mesangial, subendothelial, and subepithelial areas (Patient 10 data in Table I). Combination therapy with prednisolone and cyclophosphamide improved her state, and the urinary protein level fell to lower than 0.5 g/day. Since then she has been well.
As shown in Figures 1 and 2, both cases were positive for anti-DNA CIC expressing the O-81 Id when analyzed by IEF during an acute episode with renal involvement. Anti-DNA CIC, however, could not be detected during remission or in acute episodes in which proteinuria was absent. Similar results were obtained with other SLE cases. Table II shows that anti-DNA CIC expressing O-81 Id or NE-1 Id specifically occur in patients with active lupus nephritis. On the other hand, anti-DNA CIC were less frequently found in patients without lupus nephritis during acute episodes. It was also noteworthy that O-81 Id-positive immune complexes were never found in the latter group. 1
TABLE II Incidence of Anti-DNA CIC
Categories
Figure
2. Serial cyclophosphamide.
358
studies
in Patient
10. Ab = antibody:
October 1991 The American Journal of Medicine
Active lupus nephritis Active SLE without lupus nephritis Other diseases including inactive SLE
CY =
Ab = antibody.
Volume 91
Percent Frequency of CIC with Number AntiAntiof Cases ssDNAAb dsDNA Ab O-81 Id NE-1 Id
58
2
17 0
58
ii
0
I?
40
0
0
0
0
ANTI-DNA IMMUNE COMPLEXES IN LUPUS NEPHRITIS / SASAKI ET AL
TABLE Ill
Between Renal Histologic Featuresand CIC
Relationship
Number of Cases
CIC Anti-ssDNA Ab t
II
Renal Histologic Features (WHO Classification) IV
Ill
6
:
20 2I
3:
i
i
1:
;i
z
:
20 2I
4;
1:
i
2:
zl
4*
Anti-dsDNA Ab
V
p to.01
1;
1
p CO.01
t
0-h
1;
Id
t NE-1 Id
t -
p tt
Anti-ssDNA Ab t Anti-dsDNA Ab t 0-i Id t I
Ab = antibody. *Immune deposits in mesangial,subendothelial,or subeoithelialarea of renal ~lomeruliwere gradedfrom - to 2 t t fNumkrsoicases. -
globulin deposits in this area (p
