- Email: [email protected]
Protooncogene expression in peripheral blood mononuclear cells from patients with systemic lupus erythematosus as an indicator of the disease activity
Protooncogene Expression in Peripheral Blood Mononuclear Cells from Patients with Systemic Lupus Erythematosus as an Indicator of the Disease Activity YASUHIRO DEGUCHI,"~? HIDEKI HARA,~ SHICERU NEGORO,?.~ TAKEO KAKUNAGA,* ANDSUSUMU KISHIMOTO~
In the present study. we examined the various protooncogene expressions in PBMC (peripheral blood mononuclear cell) of systemic lupus erythematosus (SLE) patients to determine if they could be an indicator for the disease activity. We divided SLE patients into “very active.” “active,” and “remitting” states according to the clinical symptoms in addition to the laboratory data peculiar to SLE. In addition. we determined the amount of circulating immune complex (ICY)as one of the representative laboratory indicators for the disease activity. We found a positive correlation with either c-~IJY or c-q17 expression and the amounts of IC and clinical disease activity. The degree of c-myc and c-myh expression was significantly reduced along with or prior to the amelioration of clinical symptoms and improvement as determined by laboratory data under treatment with prednisolone and/or azathioprine administration. The degree of c-my and c-m$~ gene expression had no direct relation to the presence of particular clinical sign(s) or autoantibody. The expression of the c-n!fgene was found in SLE and other systemic autoallergic patients although it showed no correlation with the disease activity. No significant expression of C-JI’(, c-r(/.t, c:fil.v. c:fiqr. c:f).\. c,fe.x. c+f.\. c-vc’.t, c-r-e/, c-oh/. c-mos, c-sis. and c-et% B genes was found in the patients. c-rnyc’ and c-m.vh expression as having pathogenic and clinical significance is discussed. IYXi ~\L.Kkrnlilk\\. llli
INTRODUCTION
Autoantibody production and allergic reaction against the host are the characteristic features of autoimmune diseases. In systemic autoallergic diseases such as systemic lupus erythematosus (SLE), connective tissue inflammation and microvascular damages which result in multiorgan system involvement are the most prominent pathologic changes. To these pathologic processes, elevation of the amount of circulating immune complex (IQ, their interaction with cell receptors, deposition to the tissue, and activation of various effector systems are thought to have the important relevance. Elevation of the amount of IC and reduction of complement titer have been considered to be one of the important indicators of disease activity (1, 2). although not necessarily indicators of activity in some cases. Recently, immunology has gained a great advance in the understanding of the mechanisms of immune regulation, but most of the fundamental causes of au’ To whom correspondence should be addressed. 424 3090-1229187 $1.50
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
425
toimmune disorders are still unknown. Williams et al. (3) claimed that most autoimmune disorders could be considered to be benign proliferative diseases. Mountz et al. (4) reported that the spleen cells of autoimmune NZB and BXSB mice express two to three times more c-myc RNA than those of nonautoimmune mice and that the lymphoid organs of MRL lprllpr mice or bone marrow cells of patients with immunoblastic lymphadenopathy with dysproteinemia (AILD) contained striking and consistently elevated amounts of c-myb RNA. Boumpas et (11. (5) also found that T and B lymphocytes from patients with an autoimmune disease such as SLE showed significantly increased expression of c-myc, c-myb, and c-ruf RNA when compared with that of normal individuals. In the present study, we have tried to examine the relations among the levels of these protooncogene expressions, amounts of circulating IC, and clinical disease activity. The results indicate that there is a good correlation among these parameters. The importance of these parameters for the disease activity is discussed. MATERIALS
AND METHODS
Subjects. We examined nine SLE patients, one dermatomyositis (DM) patient, and one progressive systemic sclerosis (PSS) patient. All received prednisolone and/or azathioprine; all SLE patients met ARA criteria for SLE (6). The diagnoses of DM and PSS were clinically made in addition to biopsy findings of the lesions. Clinical activity of SLE was divided into “very active,” “active,” and “remitting,” according to the UCH/Middlesex criteria (7) with minor modifications. In brief, the patients were judged to be very active when three or more of the following clinical manifestations were evident: malar rash, oral ulcer, arthritis, serositis, renal disorder, neurologic disorder, and hematologic disorders in addition to the laboratory data peculiar to SLE. Usually, these patients were febrile. When the patients had only one or two of the above-mentioned clinical manifestations, they were considered to be clinically “active.” We considered the patients to be clinically in “remitting” condition when they became free from all of the above-mentioned clinical signs even though some abnormal laboratory data persisted. All of the subjects were examined at our clinic (The Third Department of Internal Medicine, Osaka University Hospital). We also examined five patients with bronchial asthma (BA) who had been receiving 20 mg or more of prednisolone per day for the treatment of asthmatic attack and six healthy subjects who were laboratory or hospital personnel and had no history of use of any drug known to affect immune functions. Preparation of test serum. Peripheral venous blood was drawn, allowed to coagulate at room temperature for 2 hr in glass tubes, and centrifuged at 1500 rpm for 10 min. Sera were collected, centrifuged again at 2500 rpm for 10 min, and stored at - 20°C. Antisera reugents and buffers. Clq was purified as described by Yonemasu and Stroud (8). C3 was partially purified by ammonium sulphate precipitation, fractionation through a DEAE-cellulose column, and absorption of contaminating IgG with an anti-human IgG column. Goat anti-human Clq serum was obtained from Cappel Laboratories, Inc. (Malvern, PA). Sheep anti-human C3d serum was obtained from Japan Tanner,
436
I>EGUCHI
ET
A1
Inc. (Japan). Gamma globulin fractions were prepared by ammonium sulfate precipitation methods and digested with pepsin. F(ab’)z fractions of specific antibody to C3d or Clq were purified by affinity column-binding partially purified human C3 or Clq, respectively. Anti-human IgG alkaline phosphatase (ALP) conjugate (Sigma Chemical Co.. Cat. No. A-3150, Lot. No. 114F-8836) and ALP substrate (p-nitrophenyl phosphate disodium, Sigma Cat. No. 104) were purchased from Wako Chemical Co., Ltd. Aggreguted human gamma globulin (AHG). Human IgG was purchased from TAGO, Inc. IgG was dissolved in PBS at the final concentration of I mgiml and incubated in a water bath at 63°C for 30 min. After centrifugation at 10,000 rpm for 10 min to remove large insoluble aggregates, it was passed through a 0.22~km Millipore filter (Millex-GS, Millipore Corp., Bedford, MA). The preparation was stored in small aliquots at -70°C until use. Fresh normal human sera as a complement source were stored in small aliquots at -70°C until use. An equal volume of serially diluted AHG was mixed with the sera, incubated at 37°C for 30 min, and used as the standard for IC. Measurement of the amount of immune complex (IC). The solid-phase anti-C3d and anti-Clq ELISA for circulating IC were performed as follows. Wells of NUNC-Immunoplate II (Nippon Intermed Co., Japan) were coated with 50 ~1 of either F(ab’), anti-C3d or F(ab’), anti-Clq solution in 20 mM PBS (pH 7.2). After overnight incubation, the antibody solution was discarded and the wells were washed twice with PBS. To these wells, 400 ~1 of 1% BSA-PBS, pH 7.2. was added to block unreacted sites. Test sera were diluted 50-fold by sample buffer (PBS containing 1% BSA, 0.05% Tween 20). After standing for 2 hr at 37”C, the 1% BSA-PBS was discarded and 50+1 samples were added in duplicate to each well. After incubation for 90 min at room temperature, the sample solution was discarded and the wells were washed five times with washing buffer (PBS containing 0.1% BSA. 0.05% Tween 20). Then, 50 ~1 of ALP-conjugated anti-human gamma heavy chain-specific antibody (diluted lOOO-fold in sample buffer) was added to each well. After another incubation for 90 min at room temperature, the wells were washed with washing buffer five times and 100 ~1 of the ALP substrate solution was added to wells. The enzymatic reaction was allowed to proceed for about 30 to 40 min at room temperature and the absorbance values were recorded with an NJ-2000 Nippon Intermed immunoreader (Nippon Intermed Co.). Preparation of peripheral blood mononuclear cells. Heparinized peripheral blood was diluted twofold with HBSS, layered on Ficoll-Paque, and centrifuged at 1800 rpm for 20 min. Peripheral blood mononuclear cells (PBMC) were collected from the interface and washed with HBSS three times. The number of cells was counted by a hemocytometer and cell pellets of I x 10’ PBMC were stored in liquid N, until the assay of expression of protooncogenes. Preparation ofZ?NA and DNA. RNA from PBMC was isolated using the guanidine isothiocyanate method (9). RNA was prepared by centrifugation through cesium chloride. Integrity of the isolated RNA was determined by agarose gel electrophoresis of glyoxal-denatured samples (10). Poly(A)-tailed RNA was obtained by passing total RNA over an oligo(dT)-cellulose column and eluting with proper
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
427
buffer (11). DNA from PBMC was prepared using SDS, proteinase K. and phenol/chloroform. After treatment with these chemicals, DNA was recovered by ethanol precipitation and spooling (12). Oncogene probe. We used the following fragments as the oncogene probes: v-myc (1.5kb PstI fragment), v-myb (4.6-kb BamHI fragment), v-ruf (4.2-kb EcoRI + Hind111 fragment), v-K-ras (1 .O-kb EcoRI fragment), v-fos (1.3-kb PstI fragment), v-SYC’(0.8-kb PvuII fragment), v-fps (0.4-kb BamHI fragment), v-&s (0.4-kb PstI fragment), v-fms (1.4-kb PstI fragment), v-yes (1.7-kb Sal1 fragment), v-w1 (1 .O-kb EcoRI fragment), v-abl(2.3-kb BumHI + Hind111 fragment), v-mos (2.1-kb Hind111 fragment), v-s& (1.3-kb PstI fragment). and v-erb B (0.5kb BarnHI + EcoRI fragment). All of the oncogene probes were gifts from the Japanese Cancer Research Resources Bank (JCRB). i.Tfgr (0.93kb y-actin-free EcoRI + Hind111 fragment) was prepared by ourselves from cloned proviral genome of G.R.FeSV. Dot blot hybridization. Total cellular RNA was prepared from PBMC and applied to nitrocellulose filters at the indicated amount. Filters were air-dried and baked at 80°C for 2 hr. The prehybridization step was performed using 5 x SSC (sodium citrate buffer), 50 mM sodium phosphate buffer (pH 7.0), 5 x Denhardt’s solution, 0.1% SDS, 50% formamide, and 250 p.g/ml t-RNA at 42°C. The hybridization step was performed in the same buffer containing 10% dextran and 350 pg/ml t-RNA at 42”C, using the respective 32P-nick-translated oncogene DNA probes and control actin probe (approximately 2 x IO* cpm/pg). Filters were washed live times in 0.2 x SSC with 0.15% SDS and subjected to autoradiogram (13-15). Northern blot and Southern blot. Glyoxal-denatured samples of 10 pg poly(A)tailed RNA per lane were electrophoresed on 1.2% agarose gel containing formamide. RNA was blotted onto nitrocellulose filters and baked. They were prehybridized and then hybridized with a v-myc probe under the same conditions as dot hybridization (10). DNA was digested using Hind111 or EcoRI and then electrophoresed on 0.8% agarose gels. The size and concentration of oncogene-containing bands were detected by Southern analysis (10, 16). The analysis of individual subjects was repeated several times on different gels with the same results. RESULTS
Correlation
between Protooncogzne
Expression
and Disease Activity
As shown in Tables 1 and 2, patients Nos. l-a, 2-a, and 3-a were clinically in the very active state with accompaning high fever, a variety of clinical manifestations peculiar to active SLE, a lower level of complement, a high titer of autoantibodies, and a high level of IC. As shown in Fig. 1, all of them (columns 3, 8, and 12 in Fig. 1) showed the highest level of c-myc and c-myb gene expression, but only one of them showed higher c-raf gene expression than healthy persons. Other two active SLE patients (Nos. 4 and 5 in Tables 1 and 2) whose clinical manifestations were beginning to ameliorate but still in the active state showed a high level of either c-myc or c-myb gene expression (columns 2 and 5 in Fig. 1). One of them showed c-rufgene expression. Four remitting SLE patients (Nos. 6, 7, 8, and 9 in Tables 1 and 2) whose disease states were clinically inactive but still
SLE
F
F F F F F F F M F F M F F M F
3x
48 46 41 32 48 23 40 41 32 28 26 32 42 32 29
SLE SLE SLE SLE SLE SLE DM PSS BA Healthy Healthy Healthy Healthy Healthy Healthy
SLE
M
27
SLE
Diagnosi@
F
Sex
42
Age Sk, R. N, H, F, (Sk R, N, H. F,) Sk.O,A,R,H.F, (Sk, 0, A. R, H. F.) Se. A, H. F. (Se, A, H. F-1 A, R, (H, FJ Sk R, (A. F,) (Sk, R, H, F,) (Sk. A. H, F,) (Sk, A. N, F.) (Sk. A, H, F,) Skin lesion, myositis. Sclerotic skin, etc. Asthma attack -
-
Clinical manifestationsd
SUBJECTS
Very active Remitting Very active Remitting Very active Remitting Active Active Remitting Remitting Remitting Remitting Active Active -
Clinical activityc
1
-
P, AZ P. AZ P. AZ P. AZ P P. AZ P P P. Az P P. AL P P P P -
Treatments 3 in Fig. b in Fig. 8 in Fig. d in Fig. I2inFig. e in Fig. 2 in Fig. 5 in Fig. -1 in Fig. 7 in Fig. 13 in Fig. a in Fig. IO in Fig. 14 in Fig. I I in Fig. I in Fig. h in Fig. 9 in Fig. c‘ in Fig. fin Fig. g in Fig.
I 5 I 5 I 5 1 I I I I 5 I 1 I I I I 5 5 5
Designation
I’ Each number designates each of the different persons. NW. l-a and I-b, 2-a and 2-b. and 3-a and 3-b designate the same patient. ~~r~prctivrl!. hsforc (a) and after (b) clinical remission with treatment. h The diagnoses of systemic lupus erythematosus (SLE) were made by the ARA criteria for SLE. The diagnose\ of demmatomywitis tDM1 dnd progressive systemic sclerosis (PSS) were performed by the biopsies of the Iesionh. The patient with bronchial asthma (B.4) received predniwlonr for asthmatic attacks. c Clinical activity wab determined according to the UCHiMiddlesex criteria. rl Abbreviations used: Sk. malar rd\h: R. renal disorder: N. neurological disorder: H. hematological diwrdrr: Se. srrositis. 0. oral ulcer .\. .Irthrltt\ F. high fever. e Pq prednisolone; AZ. azathiopurine. f The designation shown here corresponds to the one in Fig. I or Fig. 5
I-a I-b 2-a 2-b 3-a 3-b 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Subject No.”
TABLE CLINICALEVALUATIONOFTHE
SLE SLE SLE SLE SLE SLE DM PSS BA Healthy Healthy Healthy Healthy Healthy Healthy
SLE
SLE
SLE
Clinical diagnosis
ESRb
0 0
1
0
L
0 0
WBC’
0
0 0 0 0
t t 0 0
1 0 0
A 0
i 4 0 0
0
J
Complementd 2’0 26 2’0 2s 2’2 26 28 16 24 214 26 26 24 26 <2’ <2’ <2’ <2’ <2’ <2’ <2’
ANF 122 23 68 32 14 10 32 18 13 14 15 21 16 10
AntiDNA 0 0 16 4 0 0 4 0 0 32 0 0 0 0 0 0 0 0 0 0 0
AntiRNP 0 0 2 1 0 0 0 0 0 8 0 0 0 0 0 0 0 0 0 0 0
AntiSm RF <30 <30 <30 <30 48 <30 180 <30 <30 <30 32 <30 (30 146 <30 <30 <30 <30 (30 <30 X30
~ 132 20 276 49 243 35 63 56 20 37 11 30 52 128 5 8 6 7 5 6 5
Anti-C3d 113 14 320 30 152 64 18 62 6 40 9 18 42 36 4 5 2 6 4 5 6
Anti-Clq
lmmune complexJ
SUBJECTS
-
+++ + ++ ++ +++ + ++ ++ +++ + +++ + +++ + + +++ + ++ + +++ + +++ + +++ + ++ ++ + + + ++ ++ -
f-f
wb
Oncogene activity’ tnyc’
p
” Each subject number corresponds to the one in Table 1. b Erythrocyte sedimentation rate (ESR) was judged increased ( 1 ) when the rate was more than 20 mm/hr and normal (0) when the rate was less than that. c White blood cell count (WBC) less than 3000/mms was judged as decreased f 5 ): that more than 3,000/mms was judged as normal (0). d The amount of complement was determined by either CHSO hemolytic titer or immunodiffusion assay of the C3 and C4. The amount of complement was judged as decreased ( 1 ) when the amount of complement decreased as determined by either method. Normal value is shown as (0). r The following autoantibodies were determined: titer of ANF (antinuclear factor) was less than 2’ in the normal; anti-DNA (anti-double-stranded DNA antibody) was less than 10 u/ml in the normal: the titers of anti-RNP (antiribonuclear protein antibody) and anti-Sm antibody was less than 0 in the normal: the titer of RF (rheumatoid factor) was less than 30 u/ml in the normal. f The amount of immune complex (1C) was determined by either solid-phase anti-C3d or solid-phase anti-Clq assay methods as described under Materials and Methods. Y The degree of oncogene expression was expressed as the ~or-e determined by the ratio of the intensity of the dot blot by respective oncogene probe to that of the dot blot by actin probe as shown in Figs. I and 5.
I-a l-b 2-a 2-b 3-a 3-b 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sub.ject Ni.”
Autoantibodye
TABLE 2 LABORATORYEVALUATIONOFTHE
430
:a* ()
ooa
0
a, c m y c
0‘033 011
oe** c
0 F‘
i
0 (I
'O
0 33 011
0 0
1.0
0 33 011
*o d)
c-raf
e)
actin
**
*
0
0
lo 0.33 0.11
.
II
eoeeo
().I)*
*
0.5
FIG. 1. The protooncogene expression in PBMC of the patients and normal control subjects. Total RNA extracts were dotted onto the nitrocellulose filter. The amounts of RNA dotted (kg/dot) are shown along the right-hand border of the figure. As shown on the left-hand border, each RNA dot on the filter was hybridized with the “P-nick-labeled oncogene DNA probes as follows: (a) v-myc, (b) myb, (c) v-src, and (d) v-raf. As shown at (e), 0.5 kg of RNA was dotted and hybridized with a P-actin DNA probe (Wako Chemical Company. Japan) as a control. The expression of actin mRNA did not show any significant variation during cell activation. The number on each column, from left to right, corresponds to subject Nos. 13, 4, I-a, 6. 5. 14, 7. 2-a. IS. 10. I 3. 3-a. 8. and II in Table 1, respectively. C shows positive control hybridizations of each of the cold denatured v-oncogene DNA with respective 32P-labeled probe. N show\ a negative control of denatured A phage DNA.
sustaining ANF positivity and significantly higher amounts of circulating IC than normal healthy persons showed lower levels of c-myc and c-myh gene expression than the patients in active states but significantly higher levels of expression of the other genes than normal healthy persons (columns 4, 7, and 13 in Fig. 1; column a in Fig. 5). One of them showed c-rqf activity. One active DM and one active PSS patient showed high levels of c-m~~ and c-rufactivity but no c-myh activity (columns 10 and 14 in Fig. I). All of the patients already had been kept on prednisolone and/or azathioprine when examined. To examine the effect of glucocorticoid administration to the expression of the protooncogenes, we included five asthmatic patients who had been taking 20 mg or more of prednisolone per day for the treatment of asthmatic attack. None of the oncogene expression was found in these patients. Representative data (No. 12) is given in Tables 1 and 2; results of protooncogene expression are shown in column 11 in Fig. 1. None of the six normal healthy persons showed expression of c-myc, c-myb, and c-rufgenes (columns 1, 6, and 9 in Figs. 1.~; f and g in Fig. 5). None of the patients and normal healthy persons showed expression of C-SK (Fig. 1) and c-YUS genes (Fig. 5). We also examined for the expression of c-fos, c-xqr. c:fps. c:fes, c-fms, c-yes, c-rel, c-ubl, c-mos. c-sis, and
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
431
OF SLE
c-erb B genes in some diseased and healthy control subjects. We found no significant change of these gene expressions among the subjects examined (data not shown). As the amount of actin mRNA is known to be constant during cell activation, we used an actin probe for the reference of the mRNA content in RNA extracts of each sample in these experiments. There was no significant difference in the level of actin gene expression between the patients and the control subjects. Correlution between Protooncogene of Circulating IC
Expression and the Amount
We measured the amount of IC in the sera of the subjects with use of a solidphase anti-C3d or anti-Clq method. The anti-C3d method was expected to be suitable for the detection of large-sized IC whereas the anti-Clq method was chosen for the detection of smaller sized ones. As shown in Table 2, with either method, the amount of IC was highest in very active SLE patients and significantly higher even in remitting SLE patients than in healthy control subjects. There was an apparent correlation between the amount of IC and the clinical disease activity. As shown in Table 2, the oncogene expression was rated as the score determined by the ratio of the intensity of dot blot by the respective oncogene probe to that of the dot blot by the reference actin probe. As shown in Fig. 2, there was a direct correlation between the score of c-myc or c-myb expression and the amount of IC in the sera of SLE patients. However, no apparent correlation was found between the c-rufexpression and the amount of IC. No Relation of the Oncogene Expression Signs and Autoantibodies
to the Presence of Particular
Clinical
As can be seen in Table 1, of the five active SLE patients who showed high levels of c-myc gene expression, four showed renal disorder, four hematologic disorder, three malar rash, three neurologic disorder, two arthritis, one oral ulcer, and one serositis. Of the four active SLE patients who showed high level of c-myb expression, three showed malar rash, three renal disorder, three hematologic disorder, two neurologic disorder, two arthritis, one oral ulcer, and one serositis. Of the two active SLE patients who showed high levels of c-rufexpression, two showed malar rash, two renal disorder, two arthritis, one oral ulcer, one neurologic disorder, and one hemotologic disorder. However, one active SLE patient who had malar rash, renal disorder, neurologic disorder, hematologic disorder, and high fever showed no c-ruf gene expression. One patient who had arthritis and renal disorder also showed no c-rufgene expression. Thus, the high levels of c-myc, c-myb, or c-rufexpression had no direct relation to the presence of the particular clinical sign. As seen in Table 2, all of the patients with high levels of c-myc and c-myb expression showed a high titer of antinuclear factor (ANF) but other patients with low c-myc or c-myb expression also showed high titers of ANF. The levels of c-myc and c-myb gene expression had no direct relation to the presence of anti-DNA antibody, anti-RNP antibody, anti-Sm antibody, and RF. None of our SLE patients showed anti-SS-A and anti-SS-B positivity.
432
L)EGUCHI
El
AL
FIG. 2. Correlation between protooncogene expression and the amount of IC in SLE patients. The level of expression of the protooncogenes is expressed as the score determined by the ratio of densitometric intensity of the each dot to that of control actin dot in Fig. 1 and is shown on the abscissa. The amount of IC determined by solid-phase antiC3d methods is shown on the ordinate. The following marks represent the respective protooncogene expression: 0. c-myc: C,. c-myb; 3, c-raf.
Northern Blot Analysis of c-myc mRNA Gene in SLE Patients
and Southern Blot Analysis oj’c-myc
We examined by Northern blot analysis whether there was an abnormal c-myc gene transcript in SLE patients. A representative result of poly(A)-tailed mRNA from an active SLE patient (No. 2-a in Table 1) is shown in Fig. 3. A normal-sized 2.3-kb band was seen. The same pattern was seen in other active SLE patients. We found no truncated or abnormal-sized c-myc gene transcript. We next examined for the presence of a c-myc gene abnormality. DNA extracts from each subject were digested with Hind111 and subjected to Southern blot analysis. As shown in Fig. 4, we found no amplification, deletion, or rearrangement of the c-myc gene in PBMC of SLE patients. Similar results were obtained with EcoRi digestion of DNA extracts.
FIG. 3. The Northern blot hybridization of the c-myc mRNA in PBMC from SLE patient. RNA was extracted from PBMC of SLE patient No. 2-a in Table I (8 in Fig. 1). poly(A)-tailed RNA was obtained by oligo(dT) column, electrophoresed on 1.2% agarose gel. blotted to the nitrocellulose filter, and hybridized with 32P-nick-labeled v-myc DNA (1.5.kb Psrl fragment. 0.2 pg). The arrow at the lefthand border designates the presence of a normal-sized 2.3kb band.
ONCOGENE
EXPRESSION
AND DISEASE ACTIVITY
Depression of c-myc and c-myb Activity SLE Patients
OF SLE
after Immunosuppressive
Treatment
433 in
We reexamined the protooncogene expression in PBMC of the three SLE patients (patients Nos. 1, 2, and 3 in Tables 1 and 2) who had been clinically in very active state at the first examination but were in the remitting state after a 6-month intensive therapy with prednisolone and azathioprine. We also included another SLE patient (a in Fig. 5) who was clinically in the remitting state. Figure 5 shows all of them (b, d, and e in Fig. 5) to have levels of c-myc and c-myb gene activity similar to that of an inactive patient (a in Fig. 5) but still significantly higher than those of normal healthy subjects (c and fin Fig. 5). Patient No. 2 (d in Fig. 5) showed reduced c-raf gene expression but patient No. 3 (e in Fig. 5) became positive for c-rafgene expression after treatment. Again, expression of c-myc and c-myb had a positive correlation with disease activity but expression of the c-ruf gene had nothing to do with disease activity. As shown in Fig. 6, a decrease in the amount of circulating IC paralleled a decrease in the levels of c-myc or c-myb but not c-rafgene expression. DISCUSSION
Titers of autoantibodies (anti-DNA, ANF), levels of complement (CH50, C3, or C4), erythrocyte sedimentation rate (ESR), and lymphocyte and platelet counts have been clinically used as the laboratory indicators of SLE disease activity (17). Also biopsies of the kidney, skin, or other organs are most excellent and specific methods to obtain information of disease activity but they are invasive and sometimes dangerous to the patients. Recently, the number of immunoglobulin-secreting cells (ISC) was shown to be increased in the peripheral blood and bone marrow of active SLE patients (18). The number of ISC in patients was reported to correlate directly with disease activity (19). An increase in the 3
4
6
6
91112
FIG. 4. The Southern blot hybridization of DNA in PBMC from SLE and normal healthy subjects. DNA extracts from SLE and normal healthy subjects were digested with Hind111 and subjected to electrophoresis on a 0.8% agarose gel (20 pg DNA/lane). Blot hybridization analysis with a v-myc probe was carried out under stringent conditions (50% formamide, 5 x SSC. 42°C: washing: 2 x SSC, 0.1% SDS at 65°C. 0. I x SSC at 42°C). The number shown on each lane corresponds to the number of the subjects shown in Fig. 1.
DEGUCHI
434
ET Al..
myc
0 33
1 0
“l .)
I
mYb
0.33
1 .o
1. ‘W
raf
0.33
1 .o
ras
actin
* a
*
e
b
e
0
*
*
o.33
C
d e t 9 FIG. 5. Expression of the protooncogeneh m PBMC trom SLE patients after treatment with prednisolone and azathioprine with clinical remirGon. The RNA extracts from PBMC of each subject were dotted to the nitrocellulose filter. The amounts of KNA dotted (pgidot) are shown along the right-hand border of the figure. As shown on the left-hand border. each RNA dot on the filter was hybridized with the 32P-nick-labeled oncogene DNA probes (2 Y IOx cpm/kg) as follows: v-myc, v-myb, v-raf, and V-K-ras. from top to bottom. As shown at the bottom. 0.33 pg of RNA from each subject was dotted and hybridized with actin DNA as a reference of each subject. The figure shown on the bottom border corresponds to subject Nos. 9. I-h. Ih. 2-h. 3-b. 17. and 18 in Table I from left to right.
number of spontaneous
ISC is thought to reflect the activated state of PBMC
in
vivo (20).
The amount of circulating immune complex has also been known to be an excellent marker of the disease activity (2, 17). There are many methods for the determination of IC which are based on different principles. In studies employing polyethylene glycol precipitation and the Clq-binding assay, the correlation with disease activity was reported to be poor by some investigators (21). This may be due to many variables involved in the assay. Recently. solid-phase anti-C3d and anti-Clq assays with ELISA have been shown to be excellent and convenient methods for the determination of IC (2. 22). The presence of pepsin agglutinator may modify the results with these methods (23). In our experimental system, we used 50-fold diluted samples and did not find any of these activities. The anti-Clq method was selected as suitable for the detection of smaller sized IC which could drive the reaction with complement system only up to the first component, whereas the anti-C3d method was chosen as suitable for the detection of larger sized ones which could drive the reaction with complement system up to C3. In
ONCOGENE
EXPRESSION
AND DISEASE ACTIVITY
OF SLE
435
300..
i ‘M s B e ; g g
250 200.. 150.. loo50-e
0 scores
+ -II-k of onccgetle.5 actw4ty
FIG. 6. The level of oncogene expression and the amount of IC in SLE patients before and after clinical remission with treatment. The score of the level of oncogene expression was determined as shown in Fig. 2 and is shown on the abscissa. The amount of IC determined by solid-phase anti-C3d methods is shown on the ordinate. The following mark shows each oncogene expression: 0. c-myc; ~1. c-myb: 0. c-raf.
our patients, we found a good correlation between the amount of IC determined by either method and the disease activity (Table 2). We also found a direct correlation between the amount of IC and the level of c-myc or c-myb expression (Fig. 2). Kelly et al. (24) were the first to report that B-cell-specific mitogen lipopolysaccharide (LPS) and T-cell-specific mitogen concanavalin A (Con A) could induce c-myc RNA in B and T of murine spleen cells in vitro, respectively. The c-myc gene, identified originally as the cellular homolog of the transforming determinant carried by avian myelocytomatosis virus MC29 (v-myc), is altered in association with a broad spectrum of neoplasms. The enhanced expression of c-myc mRNA was reported in association with retroviral insertion, gene amplification, and translocation of the c-myc gene to the various immunoglobulin loci (25). In the present experiment, the results by Southern and Northern blot analyses showed no evidence of either amplification or translocation of the c-myc gene in PBMC of SLE patients (Figs. 3 and 4). The amount of c-myc mRNA in lymphoid cells can be increased by either enhancement of the transcription rate or stabilization of mRNA formed in the cell. The c-myc mRNA was known to be “superinduced” by mitogen in the presence of cycloheximide (24). In the present experiment, we found no direct effect of prednisolone or azathioprine on the superinduction of c-myc mRNA (Fig. l), because none of the five asthmatic patients who were taking 20 mg or more of prednisolone per day for treatment of asthmatic attack showed enhanced c-myc mRNA expression and one “very active,” two “active,” and two “inactive” SLE patients were not taking azathioprine at examination of oncogene expressions in their PBMC. The c-myc protein (MW 48,000 Da) is found predominantly in the cell nucleus
436
DEGUCHI
ET Al..
and has the ability to bind to double- and single-stranded DNA (26). The syn thesis of this protein was reported to be increased in various tumor cell lines and thought to have some relevance to immortalization of the cell. After stimulation with LPS, murine normal B cells also contained elevated amounts of c-nr~c~ pro tein (26). Human PBMC were induced to produce c-,n.vc’ protein after stimulation with PHA (27). It is well known that the actively proliferating normal and tumor cells showed comparable levels of c-nzvc mRNA an protein throughout GI. S, G2, and M rather than at a particular “cycling” phase (25). Recently, the shut-oft of c-r)ryc transcription was shown to be an integral part of the differentiation process and cessation of proliferation. Deregulation of c-rnyc transcription or translation may disrupt the normal cell control (28). In our experiment, the levels of c-myc expression were increased directly relative to the disease activity of SLE patients (Fig. I). Klinman et 111.(29) also reported that the levels of c-rnyc, expression were highest in very active SLE patients and higher in less active SLE patients than normal persons. Boumpas rt ~1. (5) reported that the expression of the c-myc* gene was increased in both T and B cells of SLE patients. As yet, the mechanisms for the enhanced levels of c-myc’ mRNA expression in PBMC of SLE patients are still unknown. The spleen cells of autoimmune NZB and BXSB mice also contained two to three times more c-myc. RNA than those of nonautoimmune mice, and Southern blot hybridization showed no evidence of nrvc’ gene rearrangement or amplification (4). A forbidden clone theory of autoimmune disease has predicted the presence of autoreactive competent cells in diseased subjects. Either the presence of intrinsic or extrinsic factors which enhance the levels of competence of lymphoid cells or the absence of those which repress the levels of competence of the cells may explain the mechanisms. PDGF, EGF. IL-2, BCGF-I, or BCGF-2 was shown to increase the level of c-rnvc, mRNA in the cell whereas IFN-(w or -p was shown to induce down-regulation of c-rnyc expression (25, 30). Production of IFN-1/ was reported to be depressed in PBMC of SLE patients (31). Recently, the state of methylation of the regulatory noncoding region of cellular protooncogenes was shown to influence the level of transcription of these genes. We found the demethylation of the region of the c-mvc’ gene was more pronounced in PBMC of all our SLE patients examined than that of normal healthy persons even though the clinical signs or laboratory data seemed near normal. This indicates that demethylation of the region of the c-m.vc gene was a more fundamental event in the pathogenesis of SLE. We also found an association among demethylation of the region of the c-rnyc gene, a high level of c-tnyc’ mRNA in lymphoid cells. and a high level of immune complexes in Castleman disease or giant lymphnode hyperplasia patients (manuscript in preparation). The c-myb gene is homologous to the transforming gene of avian myeloblastosis virus (32). The c-myb protein is also located in the cell nucleus and has a DNA-binding ability. The amount of c-myb mRNA was reported to be high in hematopoietic cells of the chicken or thymus cells in mice. Klinman et ~1. (29) reported that the purified T cells from healthy persons could be induced to express both the c-mvc and the c-myb genes by Con A stimulation, but purified B cells could be induced to express the c-rn?,c gene but not the c-myb gene by LPS stimulation. Thus, one can assume that the levels of c-myc’ expression in PBMC
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
437
may reflect the activity of both T and B cells, and those of the c-tnyb may represent the activity of T cells in PBMC. Mountz et al. (4, 33) have shown that the c-myb gene expression was increased in the spleen cells of MRL-lpr/lpr and C57BL/6-lpr/lpr mice which carry a single recessive gene (fpr gene) and develop autoantibody production and massive lymphocyte proliferation. The MLR fprllpr mice have very large amounts of myb RNA in the lymph node (LN) cells, but abnormally low levels of myb RNA in the thymus. The c-tnyb expression was also enhanced in the PBMC and bone marrow cells of the patients with angioimmunoblastic lymphadenopathy (AIBL or AILD), which is characterized by massive lymphadenopathy, hepatosplenomegaly, polyclonal. gammopathy, and autoantibody production. Mice with the gld/gld genotype, the genes of which are located on chromosome 1 and determine the development of an early onset of a lymph-proliferative disorder with autoimmunity. have also high myb RNA levels in peripheral LN cells but normal thymic myb RNA levels (33, 34). In the present study, ail of the active SLE patients showed elevated levels of c-myb gene expression in PBMC and the levels of the gene expression had a direct relationship between the disease activity and the amount of immune complexes (Figs. I and 2). Boumpas et al. (5) established the elevation of c-tnyb expression in both T and B cells of active SLE patients. In contrast, Klinman et ul. (29) reported that none of the active SLE patients examined showed elevated c-my6 expression. The reason for this discrepancy may be a difference in the sensitivity of the methods used (dot-blot hybridization vs Northern hybridization). According to Mountz et al. (33), an administration of a single large dose of cyclophosphamide to gld/gld mice led to marked regression of the lymphadenopathy and long-term amelioration of their autoimmune syndrome. Furthermore, the levels of c-tnyb expression in LN was normalized in accordance with the clinical improvement. In contrast, MRL lprilpr mice needed more cyclophosphamide administrations to obtain the amelioration of the disease. In our SLE patients, the levels of c-myb and c-myc expression were reduced along with clinical remission and reduction in amounts of circulating immune complexes after treatment (Figs. 5 and 6). The c-ruf gene is the cellular homolog of the transforming gene of 3611 MSC (v-uuf) (35). It is one of the members of the src subfamily with serine/threoninespecific protein kinase activity (36). Two kinds of genes with homology to v-ruj are present in human genomic DNA: the one on chromosome 4 is a pseudogene and the other on the short arm (3~25) of chromosome 3 is an active one. The transcripts of an active gene consist of a major 3.5-kb mRNA and a minor 5.5-kb mRNA. A 74- and a 60- to 65-kDa protein are the major c-rufproducts (37). They are located predominantly in cytoplasm. Balow et al. (38) reported that c-rufexpression was increased at 8 to 10 hr after mitogenic stimulation of PBMC from healthy persons. They showed a significant increase in c-rufexpression in both T and B cells from SLE patients. In the present study, the levels of c-ruj’ were increased in PBMC of some SLE patients and other systemic autoimmune patients but not in those of the healthy subjects. Thus, activation of this gene may have some relevance to the disease process but we found no direct relation to the clinical disease activity and the amount of immune complexes. Although Klinman
ct trl. (39) have shown that the levels of the N-1.rr.j gene were Increased In vcrk active SLE patients. we found no evidence of activity of the N-rtr.\ gent in an! ot our patients who already had received prednisolonc and/or aL.athioprine prior to examination. This may account for the discrepancy. As to the pathogenic and clinical meaning of the expression of these oncogenes in PBMC in SLE patients, wc found that increased c-myc’ gene activation prcceded a rise in serum immune complexes and clinical exacerbation of SLE folowing a diminished amount of prednisolone and that decreased oncogene expression preceded a fall in immune complex levels and amelioration of clinical signs with an increased amount of the drug. Thus. we expect examination of oncogene expression can give us the clue to evaluate the effectiveness of the immunosuppressive therapy for the SLE patients. As the number of patients was restricted in the present experiments. we need a more extensive experiment to evaluate the clinical value of this research.
REFERENCES I. Kerr, L. D.. Adelsberg. B. R.. and Spiera. H.. J. Klx~llr)~rrtol. 13, 313. 1986. 2. Sekita. K., Doi. T.. Muso. E.. Yoshida. H.. Kanatsu. K.. and Hamashima, Y., C‘lirr. E.rp. Irrrmrtnof. 55, 487. 1984. 3. Williams, R. C.. Jr.. Sibbitt. W. L., Jr.. and Husby. G.. Anw. J. Med. 80, 1011. 1986. 4. Mountz. J. D.. Steinberg, A. D., Klinman. D. M.. Smith. H. R.. and Mushinski. J. F., .s‘&~z(~( 226, 1087. 1984. 5. Boumpas. D. T.. Tsokos. G. C.. Mann. D. L.. Eleftheriades. E. G.. Harris, C. C.. and Mark. G. E.. Arfhviti.s Rheum. 29, 755. 1986. 6. Tan, E. M.. Cohen. A. S.. Fries. J. F.. Masi. A. T.. McShane. D. J.. Rothfield. N. F.. Shaller. J. G.. Talal. N.. and Winchester, R. J.. Arfhrifis Rheum. 25. 1271, 1981_. 7. Morrow. W. J. W., Isenberg. D. A., Parry, H. E. and Snaith. M. I,.. J. Khrrrrwrol. 8, 599. lY81. 8. Yonemasu. K., and Stroud. R. M.. J. Ifnmlrnol. 106, 303. 1971. 9. Chirgwin. J., Aeyble. A., McDonald, R.. and Rutter. W.. Biwhemisrp 18, 5294, 1979. IO. McGookin. R., Mathew. C. G. P.. Gaastra. W., Jbrgensen. P. L.. Gurney, T., and Slater, R. J., In “Methods in Molecular Biology” (J. M. Walker. Ed.). Vol. 2. pp. 55-120. Humana Press. Clifton. NJ, 1984. I I. Aviv. H.. and Leder, P.. Proc. N[I~/. Acud. Sci. USA 69, 1408. 1972. 12. Maniatis. T.. Fritsch. E. F.. and Sambrook. J.. In “Molecular Cloning: A Laboratory Manual” pp. 280-285. Cold Spring Habor Laboratory, Cold Spring Harbor. NY. 1982. 13. David. I. B.. Biochim. Bioph.v.\. Ac,fci 477, 191. 1977. 14. Jeffrey, A. J.. and Flavell. R. A. A.. Cell 12. 429. 1977. IS. Denhert. D. T. A., Bioclleln. Biophy.\. Rc~s. Commw~. 23, 641. 1966. 16. Southern, E. M.. J. Mol. Biol. 98, 503. 1975. 17. Morrow. W. J. W., Isenberg. D. A.. Todd-Pokropek. A.. Parry, H. F., and Snaith. M. L.. Q. J. Med. 202, 135. 1982. 18. Budman. D. R.. Merchant, E. B.. Steinberg, A. D.. Daft, B.. Gershwin. M. E., Lizzio. E.. and Reeves. J. P.. Arthritis Rheurn. 20, 829, 1977. 19. Blaese. R. M., Crayson. J.. and Steinberg, A. D.. Anwr. J. Med. 69. 345. 1980. 20. Saiki. 0.. Saeki. Y.. and Kishimoto, S.. J. Clin. Inwst. 76. 1865. 1985. 21. Inman. R. 0.. Fang. K. K. J.. Pussell. B. A.. Ryan. P. J.. and Hughes, G. R. V.. Atrhrifi.\ Rherrrn.
22. 23. 24. 25. 26.
23, 1282. 1980.
Day. N. K.. Good. R. A., and Wahn. V.. AVIPI.. J. McTd. 76(A). 35. 1984. Davey. M. P.. and Korngold, L., Irrt. AK/I. Allergy Appl. Immune/. 67, 278. 1982. Kelly. K.. Cochran. B. H., Stiles, C. D.. and Leder. P.. Cell 35, 603. 1983. Kelly. K.. and Siebenlist. U., Atmrc. Ru. In7mrnol. 4. 317. 1986. Giallongo. A.. Appella. E.. Ricciardi. R.. Rovera. G.. and Croce. C. M.. Sc,icnce 222. 430, 1983.
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
439
27. Persson. H., Hennighausen. L., Taub, R., DeGrado, W., and Leder, P., Science 225, 687, 1984. 28. Prochownik, E. V., and Kukowska, J., Nature (London) 322, 848, 1986. 29. Klinman, D. M., Mushinski, J. F., Honda, M., Ishigatsubo. Y.. Mountz. J. D.. Raveche. E. S.. and Steinberg, A. D., .I. Exp. Med. 163, 1292, 1986. 30. Resnitzky, D., Yarden, A., Zipori, D., and Kimchi. A., Cell 46, 31, 1986. 31. Tsokos, G. C., Boumpas. D. T., Smith, P. L.. Djeu, J. Y.. Balow. J. E., and Rook, A. H., Arthritis Rheum. 29, 1210, 1986. 32. Gonda, T. J.. and Bishop, J. M., J. Viral. 46, 212. 1983. 33. Mountz. J. D., Mushinski. J. F.. Smith, H. R.. Klinman. D. M., and Steinberg, A. D., J. /mmunol. 135, 2417, 1985. 34. Roths. J. B.. Murphy, E. D., and Either, E. M., J. Exp. Med. 159, 1. 1984. 35. Kozak, C.. Gunnell, M. A., and Rapp, U. R., 1. Viral. 49, 297, 1984. 36. Mark, G. E.. and Rapp, U. R., Science 224, 285, 1984. 37. Banner. T. B.. Kerby, S. B.. Sutrave. P., Gunnell, M. A.. Mark. G.. and Rapp, IJ. R., MO/. C’e// Bid. 5, 1400. 1985. 38. Balow. J. E., Austin, H. A., HI, Tsokos, G. C., Antonovych, T. T.. Steinberg, A. D., and Kippel. J. N.. Amt. In?. Med 106, 79. 1987. Received April IO, 1987: accepted with revision July 7, 1987
In the present study. we examined the various protooncogene expressions in PBMC (peripheral blood mononuclear cell) of systemic lupus erythematosus (SLE) patients to determine if they could be an indicator for the disease activity. We divided SLE patients into “very active.” “active,” and “remitting” states according to the clinical symptoms in addition to the laboratory data peculiar to SLE. In addition. we determined the amount of circulating immune complex (ICY)as one of the representative laboratory indicators for the disease activity. We found a positive correlation with either c-~IJY or c-q17 expression and the amounts of IC and clinical disease activity. The degree of c-myc and c-myh expression was significantly reduced along with or prior to the amelioration of clinical symptoms and improvement as determined by laboratory data under treatment with prednisolone and/or azathioprine administration. The degree of c-my and c-m$~ gene expression had no direct relation to the presence of particular clinical sign(s) or autoantibody. The expression of the c-n!fgene was found in SLE and other systemic autoallergic patients although it showed no correlation with the disease activity. No significant expression of C-JI’(, c-r(/.t, c:fil.v. c:fiqr. c:f).\. c,fe.x. c+f.\. c-vc’.t, c-r-e/, c-oh/. c-mos, c-sis. and c-et% B genes was found in the patients. c-rnyc’ and c-m.vh expression as having pathogenic and clinical significance is discussed. IYXi ~\L.Kkrnlilk\\. llli
INTRODUCTION
Autoantibody production and allergic reaction against the host are the characteristic features of autoimmune diseases. In systemic autoallergic diseases such as systemic lupus erythematosus (SLE), connective tissue inflammation and microvascular damages which result in multiorgan system involvement are the most prominent pathologic changes. To these pathologic processes, elevation of the amount of circulating immune complex (IQ, their interaction with cell receptors, deposition to the tissue, and activation of various effector systems are thought to have the important relevance. Elevation of the amount of IC and reduction of complement titer have been considered to be one of the important indicators of disease activity (1, 2). although not necessarily indicators of activity in some cases. Recently, immunology has gained a great advance in the understanding of the mechanisms of immune regulation, but most of the fundamental causes of au’ To whom correspondence should be addressed. 424 3090-1229187 $1.50
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
425
toimmune disorders are still unknown. Williams et al. (3) claimed that most autoimmune disorders could be considered to be benign proliferative diseases. Mountz et al. (4) reported that the spleen cells of autoimmune NZB and BXSB mice express two to three times more c-myc RNA than those of nonautoimmune mice and that the lymphoid organs of MRL lprllpr mice or bone marrow cells of patients with immunoblastic lymphadenopathy with dysproteinemia (AILD) contained striking and consistently elevated amounts of c-myb RNA. Boumpas et (11. (5) also found that T and B lymphocytes from patients with an autoimmune disease such as SLE showed significantly increased expression of c-myc, c-myb, and c-ruf RNA when compared with that of normal individuals. In the present study, we have tried to examine the relations among the levels of these protooncogene expressions, amounts of circulating IC, and clinical disease activity. The results indicate that there is a good correlation among these parameters. The importance of these parameters for the disease activity is discussed. MATERIALS
AND METHODS
Subjects. We examined nine SLE patients, one dermatomyositis (DM) patient, and one progressive systemic sclerosis (PSS) patient. All received prednisolone and/or azathioprine; all SLE patients met ARA criteria for SLE (6). The diagnoses of DM and PSS were clinically made in addition to biopsy findings of the lesions. Clinical activity of SLE was divided into “very active,” “active,” and “remitting,” according to the UCH/Middlesex criteria (7) with minor modifications. In brief, the patients were judged to be very active when three or more of the following clinical manifestations were evident: malar rash, oral ulcer, arthritis, serositis, renal disorder, neurologic disorder, and hematologic disorders in addition to the laboratory data peculiar to SLE. Usually, these patients were febrile. When the patients had only one or two of the above-mentioned clinical manifestations, they were considered to be clinically “active.” We considered the patients to be clinically in “remitting” condition when they became free from all of the above-mentioned clinical signs even though some abnormal laboratory data persisted. All of the subjects were examined at our clinic (The Third Department of Internal Medicine, Osaka University Hospital). We also examined five patients with bronchial asthma (BA) who had been receiving 20 mg or more of prednisolone per day for the treatment of asthmatic attack and six healthy subjects who were laboratory or hospital personnel and had no history of use of any drug known to affect immune functions. Preparation of test serum. Peripheral venous blood was drawn, allowed to coagulate at room temperature for 2 hr in glass tubes, and centrifuged at 1500 rpm for 10 min. Sera were collected, centrifuged again at 2500 rpm for 10 min, and stored at - 20°C. Antisera reugents and buffers. Clq was purified as described by Yonemasu and Stroud (8). C3 was partially purified by ammonium sulphate precipitation, fractionation through a DEAE-cellulose column, and absorption of contaminating IgG with an anti-human IgG column. Goat anti-human Clq serum was obtained from Cappel Laboratories, Inc. (Malvern, PA). Sheep anti-human C3d serum was obtained from Japan Tanner,
436
I>EGUCHI
ET
A1
Inc. (Japan). Gamma globulin fractions were prepared by ammonium sulfate precipitation methods and digested with pepsin. F(ab’)z fractions of specific antibody to C3d or Clq were purified by affinity column-binding partially purified human C3 or Clq, respectively. Anti-human IgG alkaline phosphatase (ALP) conjugate (Sigma Chemical Co.. Cat. No. A-3150, Lot. No. 114F-8836) and ALP substrate (p-nitrophenyl phosphate disodium, Sigma Cat. No. 104) were purchased from Wako Chemical Co., Ltd. Aggreguted human gamma globulin (AHG). Human IgG was purchased from TAGO, Inc. IgG was dissolved in PBS at the final concentration of I mgiml and incubated in a water bath at 63°C for 30 min. After centrifugation at 10,000 rpm for 10 min to remove large insoluble aggregates, it was passed through a 0.22~km Millipore filter (Millex-GS, Millipore Corp., Bedford, MA). The preparation was stored in small aliquots at -70°C until use. Fresh normal human sera as a complement source were stored in small aliquots at -70°C until use. An equal volume of serially diluted AHG was mixed with the sera, incubated at 37°C for 30 min, and used as the standard for IC. Measurement of the amount of immune complex (IC). The solid-phase anti-C3d and anti-Clq ELISA for circulating IC were performed as follows. Wells of NUNC-Immunoplate II (Nippon Intermed Co., Japan) were coated with 50 ~1 of either F(ab’), anti-C3d or F(ab’), anti-Clq solution in 20 mM PBS (pH 7.2). After overnight incubation, the antibody solution was discarded and the wells were washed twice with PBS. To these wells, 400 ~1 of 1% BSA-PBS, pH 7.2. was added to block unreacted sites. Test sera were diluted 50-fold by sample buffer (PBS containing 1% BSA, 0.05% Tween 20). After standing for 2 hr at 37”C, the 1% BSA-PBS was discarded and 50+1 samples were added in duplicate to each well. After incubation for 90 min at room temperature, the sample solution was discarded and the wells were washed five times with washing buffer (PBS containing 0.1% BSA. 0.05% Tween 20). Then, 50 ~1 of ALP-conjugated anti-human gamma heavy chain-specific antibody (diluted lOOO-fold in sample buffer) was added to each well. After another incubation for 90 min at room temperature, the wells were washed with washing buffer five times and 100 ~1 of the ALP substrate solution was added to wells. The enzymatic reaction was allowed to proceed for about 30 to 40 min at room temperature and the absorbance values were recorded with an NJ-2000 Nippon Intermed immunoreader (Nippon Intermed Co.). Preparation of peripheral blood mononuclear cells. Heparinized peripheral blood was diluted twofold with HBSS, layered on Ficoll-Paque, and centrifuged at 1800 rpm for 20 min. Peripheral blood mononuclear cells (PBMC) were collected from the interface and washed with HBSS three times. The number of cells was counted by a hemocytometer and cell pellets of I x 10’ PBMC were stored in liquid N, until the assay of expression of protooncogenes. Preparation ofZ?NA and DNA. RNA from PBMC was isolated using the guanidine isothiocyanate method (9). RNA was prepared by centrifugation through cesium chloride. Integrity of the isolated RNA was determined by agarose gel electrophoresis of glyoxal-denatured samples (10). Poly(A)-tailed RNA was obtained by passing total RNA over an oligo(dT)-cellulose column and eluting with proper
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
427
buffer (11). DNA from PBMC was prepared using SDS, proteinase K. and phenol/chloroform. After treatment with these chemicals, DNA was recovered by ethanol precipitation and spooling (12). Oncogene probe. We used the following fragments as the oncogene probes: v-myc (1.5kb PstI fragment), v-myb (4.6-kb BamHI fragment), v-ruf (4.2-kb EcoRI + Hind111 fragment), v-K-ras (1 .O-kb EcoRI fragment), v-fos (1.3-kb PstI fragment), v-SYC’(0.8-kb PvuII fragment), v-fps (0.4-kb BamHI fragment), v-&s (0.4-kb PstI fragment), v-fms (1.4-kb PstI fragment), v-yes (1.7-kb Sal1 fragment), v-w1 (1 .O-kb EcoRI fragment), v-abl(2.3-kb BumHI + Hind111 fragment), v-mos (2.1-kb Hind111 fragment), v-s& (1.3-kb PstI fragment). and v-erb B (0.5kb BarnHI + EcoRI fragment). All of the oncogene probes were gifts from the Japanese Cancer Research Resources Bank (JCRB). i.Tfgr (0.93kb y-actin-free EcoRI + Hind111 fragment) was prepared by ourselves from cloned proviral genome of G.R.FeSV. Dot blot hybridization. Total cellular RNA was prepared from PBMC and applied to nitrocellulose filters at the indicated amount. Filters were air-dried and baked at 80°C for 2 hr. The prehybridization step was performed using 5 x SSC (sodium citrate buffer), 50 mM sodium phosphate buffer (pH 7.0), 5 x Denhardt’s solution, 0.1% SDS, 50% formamide, and 250 p.g/ml t-RNA at 42°C. The hybridization step was performed in the same buffer containing 10% dextran and 350 pg/ml t-RNA at 42”C, using the respective 32P-nick-translated oncogene DNA probes and control actin probe (approximately 2 x IO* cpm/pg). Filters were washed live times in 0.2 x SSC with 0.15% SDS and subjected to autoradiogram (13-15). Northern blot and Southern blot. Glyoxal-denatured samples of 10 pg poly(A)tailed RNA per lane were electrophoresed on 1.2% agarose gel containing formamide. RNA was blotted onto nitrocellulose filters and baked. They were prehybridized and then hybridized with a v-myc probe under the same conditions as dot hybridization (10). DNA was digested using Hind111 or EcoRI and then electrophoresed on 0.8% agarose gels. The size and concentration of oncogene-containing bands were detected by Southern analysis (10, 16). The analysis of individual subjects was repeated several times on different gels with the same results. RESULTS
Correlation
between Protooncogzne
Expression
and Disease Activity
As shown in Tables 1 and 2, patients Nos. l-a, 2-a, and 3-a were clinically in the very active state with accompaning high fever, a variety of clinical manifestations peculiar to active SLE, a lower level of complement, a high titer of autoantibodies, and a high level of IC. As shown in Fig. 1, all of them (columns 3, 8, and 12 in Fig. 1) showed the highest level of c-myc and c-myb gene expression, but only one of them showed higher c-raf gene expression than healthy persons. Other two active SLE patients (Nos. 4 and 5 in Tables 1 and 2) whose clinical manifestations were beginning to ameliorate but still in the active state showed a high level of either c-myc or c-myb gene expression (columns 2 and 5 in Fig. 1). One of them showed c-rufgene expression. Four remitting SLE patients (Nos. 6, 7, 8, and 9 in Tables 1 and 2) whose disease states were clinically inactive but still
SLE
F
F F F F F F F M F F M F F M F
3x
48 46 41 32 48 23 40 41 32 28 26 32 42 32 29
SLE SLE SLE SLE SLE SLE DM PSS BA Healthy Healthy Healthy Healthy Healthy Healthy
SLE
M
27
SLE
Diagnosi@
F
Sex
42
Age Sk, R. N, H, F, (Sk R, N, H. F,) Sk.O,A,R,H.F, (Sk, 0, A. R, H. F.) Se. A, H. F. (Se, A, H. F-1 A, R, (H, FJ Sk R, (A. F,) (Sk, R, H, F,) (Sk. A. H, F,) (Sk, A. N, F.) (Sk. A, H, F,) Skin lesion, myositis. Sclerotic skin, etc. Asthma attack -
-
Clinical manifestationsd
SUBJECTS
Very active Remitting Very active Remitting Very active Remitting Active Active Remitting Remitting Remitting Remitting Active Active -
Clinical activityc
1
-
P, AZ P. AZ P. AZ P. AZ P P. AZ P P P. Az P P. AL P P P P -
Treatments 3 in Fig. b in Fig. 8 in Fig. d in Fig. I2inFig. e in Fig. 2 in Fig. 5 in Fig. -1 in Fig. 7 in Fig. 13 in Fig. a in Fig. IO in Fig. 14 in Fig. I I in Fig. I in Fig. h in Fig. 9 in Fig. c‘ in Fig. fin Fig. g in Fig.
I 5 I 5 I 5 1 I I I I 5 I 1 I I I I 5 5 5
Designation
I’ Each number designates each of the different persons. NW. l-a and I-b, 2-a and 2-b. and 3-a and 3-b designate the same patient. ~~r~prctivrl!. hsforc (a) and after (b) clinical remission with treatment. h The diagnoses of systemic lupus erythematosus (SLE) were made by the ARA criteria for SLE. The diagnose\ of demmatomywitis tDM1 dnd progressive systemic sclerosis (PSS) were performed by the biopsies of the Iesionh. The patient with bronchial asthma (B.4) received predniwlonr for asthmatic attacks. c Clinical activity wab determined according to the UCHiMiddlesex criteria. rl Abbreviations used: Sk. malar rd\h: R. renal disorder: N. neurological disorder: H. hematological diwrdrr: Se. srrositis. 0. oral ulcer .\. .Irthrltt\ F. high fever. e Pq prednisolone; AZ. azathiopurine. f The designation shown here corresponds to the one in Fig. I or Fig. 5
I-a I-b 2-a 2-b 3-a 3-b 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Subject No.”
TABLE CLINICALEVALUATIONOFTHE
SLE SLE SLE SLE SLE SLE DM PSS BA Healthy Healthy Healthy Healthy Healthy Healthy
SLE
SLE
SLE
Clinical diagnosis
ESRb
0 0
1
0
L
0 0
WBC’
0
0 0 0 0
t t 0 0
1 0 0
A 0
i 4 0 0
0
J
Complementd 2’0 26 2’0 2s 2’2 26 28 16 24 214 26 26 24 26 <2’ <2’ <2’ <2’ <2’ <2’ <2’
ANF 122 23 68 32 14 10 32 18 13 14 15 21 16 10
AntiDNA 0 0 16 4 0 0 4 0 0 32 0 0 0 0 0 0 0 0 0 0 0
AntiRNP 0 0 2 1 0 0 0 0 0 8 0 0 0 0 0 0 0 0 0 0 0
AntiSm RF <30 <30 <30 <30 48 <30 180 <30 <30 <30 32 <30 (30 146 <30 <30 <30 <30 (30 <30 X30
~ 132 20 276 49 243 35 63 56 20 37 11 30 52 128 5 8 6 7 5 6 5
Anti-C3d 113 14 320 30 152 64 18 62 6 40 9 18 42 36 4 5 2 6 4 5 6
Anti-Clq
lmmune complexJ
SUBJECTS
-
+++ + ++ ++ +++ + ++ ++ +++ + +++ + +++ + + +++ + ++ + +++ + +++ + +++ + ++ ++ + + + ++ ++ -
f-f
wb
Oncogene activity’ tnyc’
p
” Each subject number corresponds to the one in Table 1. b Erythrocyte sedimentation rate (ESR) was judged increased ( 1 ) when the rate was more than 20 mm/hr and normal (0) when the rate was less than that. c White blood cell count (WBC) less than 3000/mms was judged as decreased f 5 ): that more than 3,000/mms was judged as normal (0). d The amount of complement was determined by either CHSO hemolytic titer or immunodiffusion assay of the C3 and C4. The amount of complement was judged as decreased ( 1 ) when the amount of complement decreased as determined by either method. Normal value is shown as (0). r The following autoantibodies were determined: titer of ANF (antinuclear factor) was less than 2’ in the normal; anti-DNA (anti-double-stranded DNA antibody) was less than 10 u/ml in the normal: the titers of anti-RNP (antiribonuclear protein antibody) and anti-Sm antibody was less than 0 in the normal: the titer of RF (rheumatoid factor) was less than 30 u/ml in the normal. f The amount of immune complex (1C) was determined by either solid-phase anti-C3d or solid-phase anti-Clq assay methods as described under Materials and Methods. Y The degree of oncogene expression was expressed as the ~or-e determined by the ratio of the intensity of the dot blot by respective oncogene probe to that of the dot blot by actin probe as shown in Figs. I and 5.
I-a l-b 2-a 2-b 3-a 3-b 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Sub.ject Ni.”
Autoantibodye
TABLE 2 LABORATORYEVALUATIONOFTHE
430
:a* ()
ooa
0
a, c m y c
0‘033 011
oe** c
0 F‘
i
0 (I
'O
0 33 011
0 0
1.0
0 33 011
*o d)
c-raf
e)
actin
**
*
0
0
lo 0.33 0.11
.
II
eoeeo
().I)*
*
0.5
FIG. 1. The protooncogene expression in PBMC of the patients and normal control subjects. Total RNA extracts were dotted onto the nitrocellulose filter. The amounts of RNA dotted (kg/dot) are shown along the right-hand border of the figure. As shown on the left-hand border, each RNA dot on the filter was hybridized with the “P-nick-labeled oncogene DNA probes as follows: (a) v-myc, (b) myb, (c) v-src, and (d) v-raf. As shown at (e), 0.5 kg of RNA was dotted and hybridized with a P-actin DNA probe (Wako Chemical Company. Japan) as a control. The expression of actin mRNA did not show any significant variation during cell activation. The number on each column, from left to right, corresponds to subject Nos. 13, 4, I-a, 6. 5. 14, 7. 2-a. IS. 10. I 3. 3-a. 8. and II in Table 1, respectively. C shows positive control hybridizations of each of the cold denatured v-oncogene DNA with respective 32P-labeled probe. N show\ a negative control of denatured A phage DNA.
sustaining ANF positivity and significantly higher amounts of circulating IC than normal healthy persons showed lower levels of c-myc and c-myh gene expression than the patients in active states but significantly higher levels of expression of the other genes than normal healthy persons (columns 4, 7, and 13 in Fig. 1; column a in Fig. 5). One of them showed c-rqf activity. One active DM and one active PSS patient showed high levels of c-m~~ and c-rufactivity but no c-myh activity (columns 10 and 14 in Fig. I). All of the patients already had been kept on prednisolone and/or azathioprine when examined. To examine the effect of glucocorticoid administration to the expression of the protooncogenes, we included five asthmatic patients who had been taking 20 mg or more of prednisolone per day for the treatment of asthmatic attack. None of the oncogene expression was found in these patients. Representative data (No. 12) is given in Tables 1 and 2; results of protooncogene expression are shown in column 11 in Fig. 1. None of the six normal healthy persons showed expression of c-myc, c-myb, and c-rufgenes (columns 1, 6, and 9 in Figs. 1.~; f and g in Fig. 5). None of the patients and normal healthy persons showed expression of C-SK (Fig. 1) and c-YUS genes (Fig. 5). We also examined for the expression of c-fos, c-xqr. c:fps. c:fes, c-fms, c-yes, c-rel, c-ubl, c-mos. c-sis, and
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
431
OF SLE
c-erb B genes in some diseased and healthy control subjects. We found no significant change of these gene expressions among the subjects examined (data not shown). As the amount of actin mRNA is known to be constant during cell activation, we used an actin probe for the reference of the mRNA content in RNA extracts of each sample in these experiments. There was no significant difference in the level of actin gene expression between the patients and the control subjects. Correlution between Protooncogene of Circulating IC
Expression and the Amount
We measured the amount of IC in the sera of the subjects with use of a solidphase anti-C3d or anti-Clq method. The anti-C3d method was expected to be suitable for the detection of large-sized IC whereas the anti-Clq method was chosen for the detection of smaller sized ones. As shown in Table 2, with either method, the amount of IC was highest in very active SLE patients and significantly higher even in remitting SLE patients than in healthy control subjects. There was an apparent correlation between the amount of IC and the clinical disease activity. As shown in Table 2, the oncogene expression was rated as the score determined by the ratio of the intensity of dot blot by the respective oncogene probe to that of the dot blot by the reference actin probe. As shown in Fig. 2, there was a direct correlation between the score of c-myc or c-myb expression and the amount of IC in the sera of SLE patients. However, no apparent correlation was found between the c-rufexpression and the amount of IC. No Relation of the Oncogene Expression Signs and Autoantibodies
to the Presence of Particular
Clinical
As can be seen in Table 1, of the five active SLE patients who showed high levels of c-myc gene expression, four showed renal disorder, four hematologic disorder, three malar rash, three neurologic disorder, two arthritis, one oral ulcer, and one serositis. Of the four active SLE patients who showed high level of c-myb expression, three showed malar rash, three renal disorder, three hematologic disorder, two neurologic disorder, two arthritis, one oral ulcer, and one serositis. Of the two active SLE patients who showed high levels of c-rufexpression, two showed malar rash, two renal disorder, two arthritis, one oral ulcer, one neurologic disorder, and one hemotologic disorder. However, one active SLE patient who had malar rash, renal disorder, neurologic disorder, hematologic disorder, and high fever showed no c-ruf gene expression. One patient who had arthritis and renal disorder also showed no c-rufgene expression. Thus, the high levels of c-myc, c-myb, or c-rufexpression had no direct relation to the presence of the particular clinical sign. As seen in Table 2, all of the patients with high levels of c-myc and c-myb expression showed a high titer of antinuclear factor (ANF) but other patients with low c-myc or c-myb expression also showed high titers of ANF. The levels of c-myc and c-myb gene expression had no direct relation to the presence of anti-DNA antibody, anti-RNP antibody, anti-Sm antibody, and RF. None of our SLE patients showed anti-SS-A and anti-SS-B positivity.
432
L)EGUCHI
El
AL
FIG. 2. Correlation between protooncogene expression and the amount of IC in SLE patients. The level of expression of the protooncogenes is expressed as the score determined by the ratio of densitometric intensity of the each dot to that of control actin dot in Fig. 1 and is shown on the abscissa. The amount of IC determined by solid-phase antiC3d methods is shown on the ordinate. The following marks represent the respective protooncogene expression: 0. c-myc: C,. c-myb; 3, c-raf.
Northern Blot Analysis of c-myc mRNA Gene in SLE Patients
and Southern Blot Analysis oj’c-myc
We examined by Northern blot analysis whether there was an abnormal c-myc gene transcript in SLE patients. A representative result of poly(A)-tailed mRNA from an active SLE patient (No. 2-a in Table 1) is shown in Fig. 3. A normal-sized 2.3-kb band was seen. The same pattern was seen in other active SLE patients. We found no truncated or abnormal-sized c-myc gene transcript. We next examined for the presence of a c-myc gene abnormality. DNA extracts from each subject were digested with Hind111 and subjected to Southern blot analysis. As shown in Fig. 4, we found no amplification, deletion, or rearrangement of the c-myc gene in PBMC of SLE patients. Similar results were obtained with EcoRi digestion of DNA extracts.
FIG. 3. The Northern blot hybridization of the c-myc mRNA in PBMC from SLE patient. RNA was extracted from PBMC of SLE patient No. 2-a in Table I (8 in Fig. 1). poly(A)-tailed RNA was obtained by oligo(dT) column, electrophoresed on 1.2% agarose gel. blotted to the nitrocellulose filter, and hybridized with 32P-nick-labeled v-myc DNA (1.5.kb Psrl fragment. 0.2 pg). The arrow at the lefthand border designates the presence of a normal-sized 2.3kb band.
ONCOGENE
EXPRESSION
AND DISEASE ACTIVITY
Depression of c-myc and c-myb Activity SLE Patients
OF SLE
after Immunosuppressive
Treatment
433 in
We reexamined the protooncogene expression in PBMC of the three SLE patients (patients Nos. 1, 2, and 3 in Tables 1 and 2) who had been clinically in very active state at the first examination but were in the remitting state after a 6-month intensive therapy with prednisolone and azathioprine. We also included another SLE patient (a in Fig. 5) who was clinically in the remitting state. Figure 5 shows all of them (b, d, and e in Fig. 5) to have levels of c-myc and c-myb gene activity similar to that of an inactive patient (a in Fig. 5) but still significantly higher than those of normal healthy subjects (c and fin Fig. 5). Patient No. 2 (d in Fig. 5) showed reduced c-raf gene expression but patient No. 3 (e in Fig. 5) became positive for c-rafgene expression after treatment. Again, expression of c-myc and c-myb had a positive correlation with disease activity but expression of the c-ruf gene had nothing to do with disease activity. As shown in Fig. 6, a decrease in the amount of circulating IC paralleled a decrease in the levels of c-myc or c-myb but not c-rafgene expression. DISCUSSION
Titers of autoantibodies (anti-DNA, ANF), levels of complement (CH50, C3, or C4), erythrocyte sedimentation rate (ESR), and lymphocyte and platelet counts have been clinically used as the laboratory indicators of SLE disease activity (17). Also biopsies of the kidney, skin, or other organs are most excellent and specific methods to obtain information of disease activity but they are invasive and sometimes dangerous to the patients. Recently, the number of immunoglobulin-secreting cells (ISC) was shown to be increased in the peripheral blood and bone marrow of active SLE patients (18). The number of ISC in patients was reported to correlate directly with disease activity (19). An increase in the 3
4
6
6
91112
FIG. 4. The Southern blot hybridization of DNA in PBMC from SLE and normal healthy subjects. DNA extracts from SLE and normal healthy subjects were digested with Hind111 and subjected to electrophoresis on a 0.8% agarose gel (20 pg DNA/lane). Blot hybridization analysis with a v-myc probe was carried out under stringent conditions (50% formamide, 5 x SSC. 42°C: washing: 2 x SSC, 0.1% SDS at 65°C. 0. I x SSC at 42°C). The number shown on each lane corresponds to the number of the subjects shown in Fig. 1.
DEGUCHI
434
ET Al..
myc
0 33
1 0
“l .)
I
mYb
0.33
1 .o
1. ‘W
raf
0.33
1 .o
ras
actin
* a
*
e
b
e
0
*
*
o.33
C
d e t 9 FIG. 5. Expression of the protooncogeneh m PBMC trom SLE patients after treatment with prednisolone and azathioprine with clinical remirGon. The RNA extracts from PBMC of each subject were dotted to the nitrocellulose filter. The amounts of KNA dotted (pgidot) are shown along the right-hand border of the figure. As shown on the left-hand border. each RNA dot on the filter was hybridized with the 32P-nick-labeled oncogene DNA probes (2 Y IOx cpm/kg) as follows: v-myc, v-myb, v-raf, and V-K-ras. from top to bottom. As shown at the bottom. 0.33 pg of RNA from each subject was dotted and hybridized with actin DNA as a reference of each subject. The figure shown on the bottom border corresponds to subject Nos. 9. I-h. Ih. 2-h. 3-b. 17. and 18 in Table I from left to right.
number of spontaneous
ISC is thought to reflect the activated state of PBMC
in
vivo (20).
The amount of circulating immune complex has also been known to be an excellent marker of the disease activity (2, 17). There are many methods for the determination of IC which are based on different principles. In studies employing polyethylene glycol precipitation and the Clq-binding assay, the correlation with disease activity was reported to be poor by some investigators (21). This may be due to many variables involved in the assay. Recently. solid-phase anti-C3d and anti-Clq assays with ELISA have been shown to be excellent and convenient methods for the determination of IC (2. 22). The presence of pepsin agglutinator may modify the results with these methods (23). In our experimental system, we used 50-fold diluted samples and did not find any of these activities. The anti-Clq method was selected as suitable for the detection of smaller sized IC which could drive the reaction with complement system only up to the first component, whereas the anti-C3d method was chosen as suitable for the detection of larger sized ones which could drive the reaction with complement system up to C3. In
ONCOGENE
EXPRESSION
AND DISEASE ACTIVITY
OF SLE
435
300..
i ‘M s B e ; g g
250 200.. 150.. loo50-e
0 scores
+ -II-k of onccgetle.5 actw4ty
FIG. 6. The level of oncogene expression and the amount of IC in SLE patients before and after clinical remission with treatment. The score of the level of oncogene expression was determined as shown in Fig. 2 and is shown on the abscissa. The amount of IC determined by solid-phase anti-C3d methods is shown on the ordinate. The following mark shows each oncogene expression: 0. c-myc; ~1. c-myb: 0. c-raf.
our patients, we found a good correlation between the amount of IC determined by either method and the disease activity (Table 2). We also found a direct correlation between the amount of IC and the level of c-myc or c-myb expression (Fig. 2). Kelly et al. (24) were the first to report that B-cell-specific mitogen lipopolysaccharide (LPS) and T-cell-specific mitogen concanavalin A (Con A) could induce c-myc RNA in B and T of murine spleen cells in vitro, respectively. The c-myc gene, identified originally as the cellular homolog of the transforming determinant carried by avian myelocytomatosis virus MC29 (v-myc), is altered in association with a broad spectrum of neoplasms. The enhanced expression of c-myc mRNA was reported in association with retroviral insertion, gene amplification, and translocation of the c-myc gene to the various immunoglobulin loci (25). In the present experiment, the results by Southern and Northern blot analyses showed no evidence of either amplification or translocation of the c-myc gene in PBMC of SLE patients (Figs. 3 and 4). The amount of c-myc mRNA in lymphoid cells can be increased by either enhancement of the transcription rate or stabilization of mRNA formed in the cell. The c-myc mRNA was known to be “superinduced” by mitogen in the presence of cycloheximide (24). In the present experiment, we found no direct effect of prednisolone or azathioprine on the superinduction of c-myc mRNA (Fig. l), because none of the five asthmatic patients who were taking 20 mg or more of prednisolone per day for treatment of asthmatic attack showed enhanced c-myc mRNA expression and one “very active,” two “active,” and two “inactive” SLE patients were not taking azathioprine at examination of oncogene expressions in their PBMC. The c-myc protein (MW 48,000 Da) is found predominantly in the cell nucleus
436
DEGUCHI
ET Al..
and has the ability to bind to double- and single-stranded DNA (26). The syn thesis of this protein was reported to be increased in various tumor cell lines and thought to have some relevance to immortalization of the cell. After stimulation with LPS, murine normal B cells also contained elevated amounts of c-nr~c~ pro tein (26). Human PBMC were induced to produce c-,n.vc’ protein after stimulation with PHA (27). It is well known that the actively proliferating normal and tumor cells showed comparable levels of c-nzvc mRNA an protein throughout GI. S, G2, and M rather than at a particular “cycling” phase (25). Recently, the shut-oft of c-r)ryc transcription was shown to be an integral part of the differentiation process and cessation of proliferation. Deregulation of c-rnyc transcription or translation may disrupt the normal cell control (28). In our experiment, the levels of c-myc expression were increased directly relative to the disease activity of SLE patients (Fig. I). Klinman et 111.(29) also reported that the levels of c-rnyc, expression were highest in very active SLE patients and higher in less active SLE patients than normal persons. Boumpas rt ~1. (5) reported that the expression of the c-myc* gene was increased in both T and B cells of SLE patients. As yet, the mechanisms for the enhanced levels of c-myc’ mRNA expression in PBMC of SLE patients are still unknown. The spleen cells of autoimmune NZB and BXSB mice also contained two to three times more c-myc. RNA than those of nonautoimmune mice, and Southern blot hybridization showed no evidence of nrvc’ gene rearrangement or amplification (4). A forbidden clone theory of autoimmune disease has predicted the presence of autoreactive competent cells in diseased subjects. Either the presence of intrinsic or extrinsic factors which enhance the levels of competence of lymphoid cells or the absence of those which repress the levels of competence of the cells may explain the mechanisms. PDGF, EGF. IL-2, BCGF-I, or BCGF-2 was shown to increase the level of c-rnvc, mRNA in the cell whereas IFN-(w or -p was shown to induce down-regulation of c-rnyc expression (25, 30). Production of IFN-1/ was reported to be depressed in PBMC of SLE patients (31). Recently, the state of methylation of the regulatory noncoding region of cellular protooncogenes was shown to influence the level of transcription of these genes. We found the demethylation of the region of the c-mvc’ gene was more pronounced in PBMC of all our SLE patients examined than that of normal healthy persons even though the clinical signs or laboratory data seemed near normal. This indicates that demethylation of the region of the c-m.vc gene was a more fundamental event in the pathogenesis of SLE. We also found an association among demethylation of the region of the c-rnyc gene, a high level of c-tnyc’ mRNA in lymphoid cells. and a high level of immune complexes in Castleman disease or giant lymphnode hyperplasia patients (manuscript in preparation). The c-myb gene is homologous to the transforming gene of avian myeloblastosis virus (32). The c-myb protein is also located in the cell nucleus and has a DNA-binding ability. The amount of c-myb mRNA was reported to be high in hematopoietic cells of the chicken or thymus cells in mice. Klinman et ~1. (29) reported that the purified T cells from healthy persons could be induced to express both the c-mvc and the c-myb genes by Con A stimulation, but purified B cells could be induced to express the c-rn?,c gene but not the c-myb gene by LPS stimulation. Thus, one can assume that the levels of c-myc’ expression in PBMC
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
437
may reflect the activity of both T and B cells, and those of the c-tnyb may represent the activity of T cells in PBMC. Mountz et al. (4, 33) have shown that the c-myb gene expression was increased in the spleen cells of MRL-lpr/lpr and C57BL/6-lpr/lpr mice which carry a single recessive gene (fpr gene) and develop autoantibody production and massive lymphocyte proliferation. The MLR fprllpr mice have very large amounts of myb RNA in the lymph node (LN) cells, but abnormally low levels of myb RNA in the thymus. The c-tnyb expression was also enhanced in the PBMC and bone marrow cells of the patients with angioimmunoblastic lymphadenopathy (AIBL or AILD), which is characterized by massive lymphadenopathy, hepatosplenomegaly, polyclonal. gammopathy, and autoantibody production. Mice with the gld/gld genotype, the genes of which are located on chromosome 1 and determine the development of an early onset of a lymph-proliferative disorder with autoimmunity. have also high myb RNA levels in peripheral LN cells but normal thymic myb RNA levels (33, 34). In the present study, ail of the active SLE patients showed elevated levels of c-myb gene expression in PBMC and the levels of the gene expression had a direct relationship between the disease activity and the amount of immune complexes (Figs. I and 2). Boumpas et al. (5) established the elevation of c-tnyb expression in both T and B cells of active SLE patients. In contrast, Klinman et ul. (29) reported that none of the active SLE patients examined showed elevated c-my6 expression. The reason for this discrepancy may be a difference in the sensitivity of the methods used (dot-blot hybridization vs Northern hybridization). According to Mountz et al. (33), an administration of a single large dose of cyclophosphamide to gld/gld mice led to marked regression of the lymphadenopathy and long-term amelioration of their autoimmune syndrome. Furthermore, the levels of c-tnyb expression in LN was normalized in accordance with the clinical improvement. In contrast, MRL lprilpr mice needed more cyclophosphamide administrations to obtain the amelioration of the disease. In our SLE patients, the levels of c-myb and c-myc expression were reduced along with clinical remission and reduction in amounts of circulating immune complexes after treatment (Figs. 5 and 6). The c-ruf gene is the cellular homolog of the transforming gene of 3611 MSC (v-uuf) (35). It is one of the members of the src subfamily with serine/threoninespecific protein kinase activity (36). Two kinds of genes with homology to v-ruj are present in human genomic DNA: the one on chromosome 4 is a pseudogene and the other on the short arm (3~25) of chromosome 3 is an active one. The transcripts of an active gene consist of a major 3.5-kb mRNA and a minor 5.5-kb mRNA. A 74- and a 60- to 65-kDa protein are the major c-rufproducts (37). They are located predominantly in cytoplasm. Balow et al. (38) reported that c-rufexpression was increased at 8 to 10 hr after mitogenic stimulation of PBMC from healthy persons. They showed a significant increase in c-rufexpression in both T and B cells from SLE patients. In the present study, the levels of c-ruj’ were increased in PBMC of some SLE patients and other systemic autoimmune patients but not in those of the healthy subjects. Thus, activation of this gene may have some relevance to the disease process but we found no direct relation to the clinical disease activity and the amount of immune complexes. Although Klinman
ct trl. (39) have shown that the levels of the N-1.rr.j gene were Increased In vcrk active SLE patients. we found no evidence of activity of the N-rtr.\ gent in an! ot our patients who already had received prednisolonc and/or aL.athioprine prior to examination. This may account for the discrepancy. As to the pathogenic and clinical meaning of the expression of these oncogenes in PBMC in SLE patients, wc found that increased c-myc’ gene activation prcceded a rise in serum immune complexes and clinical exacerbation of SLE folowing a diminished amount of prednisolone and that decreased oncogene expression preceded a fall in immune complex levels and amelioration of clinical signs with an increased amount of the drug. Thus. we expect examination of oncogene expression can give us the clue to evaluate the effectiveness of the immunosuppressive therapy for the SLE patients. As the number of patients was restricted in the present experiments. we need a more extensive experiment to evaluate the clinical value of this research.
REFERENCES I. Kerr, L. D.. Adelsberg. B. R.. and Spiera. H.. J. Klx~llr)~rrtol. 13, 313. 1986. 2. Sekita. K., Doi. T.. Muso. E.. Yoshida. H.. Kanatsu. K.. and Hamashima, Y., C‘lirr. E.rp. Irrrmrtnof. 55, 487. 1984. 3. Williams, R. C.. Jr.. Sibbitt. W. L., Jr.. and Husby. G.. Anw. J. Med. 80, 1011. 1986. 4. Mountz. J. D.. Steinberg, A. D., Klinman. D. M.. Smith. H. R.. and Mushinski. J. F., .s‘&~z(~( 226, 1087. 1984. 5. Boumpas. D. T.. Tsokos. G. C.. Mann. D. L.. Eleftheriades. E. G.. Harris, C. C.. and Mark. G. E.. Arfhviti.s Rheum. 29, 755. 1986. 6. Tan, E. M.. Cohen. A. S.. Fries. J. F.. Masi. A. T.. McShane. D. J.. Rothfield. N. F.. Shaller. J. G.. Talal. N.. and Winchester, R. J.. Arfhrifis Rheum. 25. 1271, 1981_. 7. Morrow. W. J. W., Isenberg. D. A., Parry, H. E. and Snaith. M. I,.. J. Khrrrrwrol. 8, 599. lY81. 8. Yonemasu. K., and Stroud. R. M.. J. Ifnmlrnol. 106, 303. 1971. 9. Chirgwin. J., Aeyble. A., McDonald, R.. and Rutter. W.. Biwhemisrp 18, 5294, 1979. IO. McGookin. R., Mathew. C. G. P.. Gaastra. W., Jbrgensen. P. L.. Gurney, T., and Slater, R. J., In “Methods in Molecular Biology” (J. M. Walker. Ed.). Vol. 2. pp. 55-120. Humana Press. Clifton. NJ, 1984. I I. Aviv. H.. and Leder, P.. Proc. N[I~/. Acud. Sci. USA 69, 1408. 1972. 12. Maniatis. T.. Fritsch. E. F.. and Sambrook. J.. In “Molecular Cloning: A Laboratory Manual” pp. 280-285. Cold Spring Habor Laboratory, Cold Spring Harbor. NY. 1982. 13. David. I. B.. Biochim. Bioph.v.\. Ac,fci 477, 191. 1977. 14. Jeffrey, A. J.. and Flavell. R. A. A.. Cell 12. 429. 1977. IS. Denhert. D. T. A., Bioclleln. Biophy.\. Rc~s. Commw~. 23, 641. 1966. 16. Southern, E. M.. J. Mol. Biol. 98, 503. 1975. 17. Morrow. W. J. W., Isenberg. D. A.. Todd-Pokropek. A.. Parry, H. F., and Snaith. M. L.. Q. J. Med. 202, 135. 1982. 18. Budman. D. R.. Merchant, E. B.. Steinberg, A. D.. Daft, B.. Gershwin. M. E., Lizzio. E.. and Reeves. J. P.. Arthritis Rheurn. 20, 829, 1977. 19. Blaese. R. M., Crayson. J.. and Steinberg, A. D.. Anwr. J. Med. 69. 345. 1980. 20. Saiki. 0.. Saeki. Y.. and Kishimoto, S.. J. Clin. Inwst. 76. 1865. 1985. 21. Inman. R. 0.. Fang. K. K. J.. Pussell. B. A.. Ryan. P. J.. and Hughes, G. R. V.. Atrhrifi.\ Rherrrn.
22. 23. 24. 25. 26.
23, 1282. 1980.
Day. N. K.. Good. R. A., and Wahn. V.. AVIPI.. J. McTd. 76(A). 35. 1984. Davey. M. P.. and Korngold, L., Irrt. AK/I. Allergy Appl. Immune/. 67, 278. 1982. Kelly. K.. Cochran. B. H., Stiles, C. D.. and Leder. P.. Cell 35, 603. 1983. Kelly. K.. and Siebenlist. U., Atmrc. Ru. In7mrnol. 4. 317. 1986. Giallongo. A.. Appella. E.. Ricciardi. R.. Rovera. G.. and Croce. C. M.. Sc,icnce 222. 430, 1983.
ONCOGENE
EXPRESSION
AND
DISEASE
ACTIVITY
OF SLE
439
27. Persson. H., Hennighausen. L., Taub, R., DeGrado, W., and Leder, P., Science 225, 687, 1984. 28. Prochownik, E. V., and Kukowska, J., Nature (London) 322, 848, 1986. 29. Klinman, D. M., Mushinski, J. F., Honda, M., Ishigatsubo. Y.. Mountz. J. D.. Raveche. E. S.. and Steinberg, A. D., .I. Exp. Med. 163, 1292, 1986. 30. Resnitzky, D., Yarden, A., Zipori, D., and Kimchi. A., Cell 46, 31, 1986. 31. Tsokos, G. C., Boumpas. D. T., Smith, P. L.. Djeu, J. Y.. Balow. J. E., and Rook, A. H., Arthritis Rheum. 29, 1210, 1986. 32. Gonda, T. J.. and Bishop, J. M., J. Viral. 46, 212. 1983. 33. Mountz. J. D., Mushinski. J. F.. Smith, H. R.. Klinman. D. M., and Steinberg, A. D., J. /mmunol. 135, 2417, 1985. 34. Roths. J. B.. Murphy, E. D., and Either, E. M., J. Exp. Med. 159, 1. 1984. 35. Kozak, C.. Gunnell, M. A., and Rapp, U. R., 1. Viral. 49, 297, 1984. 36. Mark, G. E.. and Rapp, U. R., Science 224, 285, 1984. 37. Banner. T. B.. Kerby, S. B.. Sutrave. P., Gunnell, M. A.. Mark. G.. and Rapp, IJ. R., MO/. C’e// Bid. 5, 1400. 1985. 38. Balow. J. E., Austin, H. A., HI, Tsokos, G. C., Antonovych, T. T.. Steinberg, A. D., and Kippel. J. N.. Amt. In?. Med 106, 79. 1987. Received April IO, 1987: accepted with revision July 7, 1987