ARTHRITIS & RHEUMATISM Vol. 40, No. 4, April 1997, pp 627-638 0 1997, American College of Rheumatology 627

DIRECT EVIDENCE FOR THE EXPRESSION OF MULTIPLE ENDOGENOUS RETROVIRUSES IN THE SYNOVIAL COMPARTMENT

IN RHEUMATOID ARTHRITIS

KAKU NAKAGAWA, VLADIMIR BRUSIC, GEOFFREY McCOLL, and LEONARD C. HARRISON

Objective. Circumstantial evidence links retrovi- ruses (RVs) with human autoimmune diseases. The aim of the present study was to obtain direct evidence of RV gene expression in rheumatoid arthritis (RA).

Methods. Synovial samples were obtained from patients with RA, patients with osteoarthritis (OA), and normal control subjects. Reverse transcription- polymerase chain reaction (RT-PCR) was performed using synovial RNA and primers to conserved sequences in the polymerase (pol) genes of known RVs.

Results. PCR products (n = 857) were cloned and sequenced. Multiple pol transcripts, many with open reading frames, were expressed in every sample. Se- quences were aligned and classified into 6 families (Fl-F6) that contained 33 groups of known and un- known endogenous RVs (ERVs), each distinguished by a specific, deduced peptide motif. The frequency of se- quences in each family was similar between RA, OA, and normal synovial tissue, but differed significantly in RA synovial fluid cells. F1 sequences (undefined, but re- lated to murine and primate type C RVs) were lower in frequency, F2 (ERV-9-related), F4 (HERV-K-related), and F6 (HERV-Grelated) sequences were higher in frequency, and F3 (RTVL-H-related) sequences were not detected, in the RA synovial fluid cells compared with the RA synovial tissues.

Conclusion. Multiple ERVs are expressed in nor-

Supported by AMKAID Pty. Ltd. and the National Health and Medical Research Council of Australia.

Kaku Nakagawa, MSc, Vladimir Brusic, MSc, Geoffrey Mc- Coll, MB, BS, Leonard C. Harrison, MD, DSc: The Walter and Eliza Hall Institute of Medical Research, Royal Melbourne Hospital, Parkville, Australia. Kaku Nakagawa is a visiting scientist from Kaneka Corp., Japan.

Address reprint requests to Leonard C. Harrison, MD, DSc, The Walter and Eliza Hall Institute of Medical Research, Post Office, Royal Melbourne Hospital, Parkville 3050, Australia.

Submitted for publication August 19, 1996; accepted in revised form October 25, 1996.

ma1 and diseased synovial compartments, but specific transcripts can be differentially expressed in RA.

Several lines of evidence implicate retroviruses (RVs) in the etiology of chronic rheumatic diseases. Pathologic and clinical manifestations of these diseases are observed both in animals and in humans infected with certain species of RVs. For example, the lentivi- ruses, caprine arthritis-encephalomyelitis virus, and hu- man immunodeficiency virus type 1 (HIV-1) cause chronic joint inflammation similar to that found in rheumatoid arthritis (RA) in goats (1) and nonsymmet- ric polyarthritis in humans (2). Human T cell leukemia virus type I (HTLV-I), has also been implicated in chronic joint inflammation, termed HTLV-I-associated arthropathy (3). Mice transgenic for the tax gene, which encodes a strong transcriptional activator of the HTLV-I long terminal repeats (4) and of some cellular genes involved in T cell growth (5-7), are known to develop chronic arthritis characterized by synovial and periartic- ular inflammation (8).

However, attempts to identify known human infectious RVs, such as HIV and other lentiviruses or HTLV-I, in the synovial tissue of patients with RA have not been successful (9). Patients with multisystem auto- immune diseases, such as systemic lupus erythematosus (SLE) and Sjogren’s syndrome (SS), have circulating antibodies that react to HIV (10,ll) or to HTLV-I (12) and, possibly, to related antigens in diseased tissue (13-15). This phenomenon may represent immune cross-reactivity with unknown RVs that are antigenically related to HTLV-I or to HIV, because clinical or standard serologic evidence of infection by the latter is absent.

Endogenous retroviruses (ERVs) have structural and sequence similarities with exogenous RVs such as HIV or HTLV. ERVs are known to be associated with immune dysregulation and may be expressed in a tissue-

628 NAKAGAWA ET AL

specific or differentiation-dependent manner. A link between ERV and autoimmune disease was first sug- gested by the finding of antibodies to the envelope glycoprotein of murine leukemia virus in lupus-prone mice (16). Some patients with SLE or SS are reported to have circulating antibodies to a human ERV, the HTLV-1-related endogenous sequence (17,18). In fact, RV-like sequences and particles distinct from those of known exogenous RVs, and immune re- sponses to ERV proteins, have been observed in auto- immune disease (19-21).

ERV could initiate and/or drive autoimmune disease in several possible ways. Aberrant expression of a normally cryptic ERV, induced by environmental or endogenous factors, might initiate autoimmunity in re- sponse to novel antigenic epitopes. Alternatively, ERV products acting as cis- or trans-regulatory elements could modify the expression of cellular genes that are involved in immune regulation. An ERV product could also act directly as a superantigen to activate T cells, some of which might be self reactive. The tissue local- ization of autoimmune disease could be explained by tissue-specific expression of either an ERV or an endog- enous autoantigen that is cross-reactive with T cells elicited by the ERV product. These possible mecha- nisms for an association between ERV and autoimmune disease remain hypothetical, in the absence of direct evidence for disease-specific ERVs and their ability to elicit pathogenic immunity.

The polymerase (pol) genes contain sequences that are highly conserved among most RVs. Shih et al (22) used degenerate oligonucleotide primers to pol in reverse transcription-polymerase chain reaction (RT- PCR) analyses to demonstrate multiple, related pol genes in ERVs that were not identical to known RVs. By adopting this strategy, we have been able to demonstrate multiple, RV pol RNA transcripts and their differential expression in the RA synovial compartment.

PATIENTS AND METHODS

Patients and tissues. Synovial tissue was obtained from patients with RA or OA at the time of joint replacement or synovectomy. Synovial fluid was aspirated from the knee joints of other patients with active RA. Fluid samples were immedi- ately centrifuged at 2,000 revolutions per minute, and the cell pellet was recovered. All patients with RA satisfied the Amer- ican College of Rheumatology (formerly, the American Rheu- matism Association) diagnostic criteria for RA (23), and patients with OA had typical radiographic findings and no prior history of inflammatory arthritis.

Normal synovial tissue, confirmed histologically, was dissected from the knee joints of legs that had been amputated

Table 1. Clinical data on study subjects and number of pol tran- scripts cloncd and sequenced*

Numbcr of pol Subject Age Sex Diagnosis Specimenhite transcripts

ST-1 62 F RA (RF-) ST/elbow 50 ST-2 65 F RA (RF+) ST/elbow 49 ST-3 54 M RA (RF+) ST/hand 22 ST-4 62 M RA (RF+) ST/shoulder 57 ST-5 58 M RA (RF+) STknee 99 ST-6 71 F RA(RF+) STknee 56 SF-7 55 F RA (RF+) SFcellsknee 20 SF-8 68 F RA (RF+) SFcellsknee 44 SF-9 55 F RA (RF+) SFcellsknee 53 ST-10 56 M OA SThip 64 ST-11 71 M OA ST/hip 1 08 ST-12 65 F OA SThip 52 ST-13 26 M Normal STIelbow 47 ST-14 30 F Normal STknee 12 ST-15 44 M Normal STknee 78 PB-16 - M Normal PBMCblood 46

* PB-16 is a pool of equal numbers of peripheral blood mononuclear cells (PBMC) from 4 healthy male subjects, ages 22-35 years. ST = synovial tissue; RA = rheumatoid arthritis; RF = rheumatoid factor; SF = synovial fluid; OA = osteoarthritis.

for osteosarcoma or peripheral vascular disease, and 1 sample was dissected from the elbow joint of an arm amputated after brachial plexus trauma. Ten milliliters of venous blood from each of 4 healthy male subjects was obtained, and peripheral blood mononuclear cells (PBMC) were isolated from thc blood by Ficoll-Hypaque density gradient centrifugation. Cells were then washcd and pooled. A detailed description of the study subjects is given in Table 1. All subjects gave their informed consent, and the study was approved by the hospital and institute ethics committees.

Preparation of total RNA from tissues. Approximately 2 gm of synovial tissue or loR synovial fluid cells or PBMC were homogenizcd in 10 ml of denaturant solution containing 25 mM sodium citrate, pH 7.0, 4M guanidium thiocyanate, and 0.5% Sarcosyl, and immediately frozen at -80°C. Total RNA was later isolated by a modification of the acid guanidium t hiocyanate-phenol-chloroform extraction met hod (24). For every 10 ml of homogenate, 1 ml of Msodium acetate, pH 4.0, 10 ml of phenol saturated with water, and 2 ml of 24:l mixture of chloroform:isoamyl alcohol were sequentially added, mixed vigorously, left on ice for 15 minutes, and then centrifuged at 10,OOOg for 20 minutes. The aqueous layer was removed and precipitated with 1 volume of isopropanol. The pellct was resuspended in 3 ml of the denaturant solution and re- precipitated with 1 volume of isopropanol.

After 2 washes in 80% ethanol, the RNA pellet was suspended in 300 pl of dimethylpyrocarbonate-treated water containing 360 units of ribonuclease inhibitor (RNasin; Pro- mega Australia, Melbourne, Victoria, Australia). Subse- quently, 60 pI of the RNA solution was added to a solution containing 20 pI of transcription-optimized 5 x buffer (Pro- mega), 10 mM dithiothreitol, 160 units of RNasin, and 100 units of RNase-free DNase I (Promega), and incubated at 37°C for 2 hours. The quality of purified, DNase 1-treated

ENDOGENOUS SYNOVIAL RETROVIRUSES IN RA 629

total RNA was checked by electrophoresis, and by absorbance at 260/280 nm.

RT-PCR analysis. The degenerate primers used in the RT-PCR analysis, synthesized to amplify the conserved regions ofpol and bounded by restriction sites for Bum HI and Eco RI (designated with an underline), were as follows: sense primer, CTCGGATCCTGGA(A/C)(A/T)GT(A/G/C)(C/T)T(A/G/ T)CC(NC/T)CA(NG)GG; antisense primer, CCGGAATTC (NGIT) AG( AIGIT) ATGTCATCCAT( NG)TA.

In order to amplify as many RV-related sequences as possible from the RNA of each sample, RT-PCR was per- formed in multiple reaction tubes, each containing 1 pg of total RNA. Tubes were heated at 95°C for 5 minutes and chilled immediately on ice. RT was initiated with 200 units of Moloney murine leukemia virus reverse transcriptase (Gibco BRL, Gaithersburg, MD) in a total volume of 20 p1 containing 30 units of RNasin, 100 pmoles of random hexamers, and 1 mM of each dNTP at 23°C for 10 minutes, then 42°C for 1 hour. The reaction volume was then added to 80 pl of PCR mixture containing 20 mM Tris (pH 8.3), 30 mM KCl, 2 mM MgCl,, 2.5 units of Tuq DNA polymerase (Perkin-Elmer Cetus, Nonvalk, CT), and 50 pmoles of each degenerate primer. The PCR reaction was heat-started at 94°C for 4 minutes and run for 35 cycles of 30 seconds each at 94"C, 1 minute at 5S°C, and 1 minute at 72°C. Control reactions were always included with- out reverse transcriptase (control for chromosomal DNA contamination) and without RNA (control for contamination by previous PCR products). To avoid cross-contamination, experiments were performed on only 1 specimen at a time.

Cloning and sequencing of RT-PCR products. After ethanol precipitation, PCR products were digested with Bam HI and Eco RI and separated by electrophoresis in highly cross-linked precast 3-40% gradient polyacrylamide gels (Gra- dipore, Sydney, Australia). Amplification products visualized with ultraviolet light after ethidium bromide staining were excised and extracted in elution buffer containing 500 mM ammonium acetate, 10 mM MgCI,, 1 mM EDTA, and 0.1% sodium dodecyl sulfate, at 37°C overnight. After phenol- chloroform treatment and ethanol precipitation, PCR frag- ments were collected by centrifugation at 30,000 rpm for 30 minutes and then ligated into 200-400 ng of Bum HI- and Eco RI-restricted pUC19 in 10 pl of buffer containing 3 units of T4 DNA ligase (Promega), at 16°C overnight. DHSa competent cells (Gibco BRL) were transformed with the whole ligation mixture. Plasmid DNA from positive clones was purified by the alkali lysis method, and 2-5 pg of the purified DNA was sequenced by the Tag DyeDeoxy Terminator cycle sequencing method (Applied Biosystems, Foster City, CA).

Sequence analysis. Sequences were analyzed with GCG software (25). GCG/Pileup was used to align and classify multiple sequences for similarity. Homologies with known sequences were sought in public databases, including EMBL, GenBank, DDBJ, and NBRF, using BLAST programs (26). The Mann-Whitney U test (2-tailed) was used to statistically compare differences in the relative frequency of sequences between disease groups.

RESULTS

Cloning and sequencing of the pol region. Syno- vial tissue or synovial fluid was obtained from 9 patients

-540bp

-1 45bp

1 2 3 4 5 6 7 8 9 1 0 1 1 Figure 1. Reverse transcription-polymerase chain reaction (RT- PCR) products of RNA from the synovial fluid (SF-7) or synovial tissue (ST-2 and ST-4) of 3 patients with rheumatoid arthritis. Samples were electrophoresed in a highly cross-linked 3-40% gradient poly- acrylamide gel for 1 hour at 25 volts/cm and visualized by ethidium bromide staining. Lane 1, Molecular size markers (pUC19 digested by Hpa 11); lanes 2, 5, and 8, products of RT-PCR reactions with retrovirus pol-specific degenerate primer pairs; lanes 3, 6, and 9, products of RT-PCR reactions with a p-actin-specific primer pair; lanes 4, 7, and 10, products of RT-PCR reactions with retrovirus pol-specific degenerate primer pairs, but without reverse transcriptase; lane 11, products of the RT-PCR reaction without RNA.

with RA, 3 patients with OA, and 3 healthy subjects (Table 1). Using the RNA extracted from each sample, an expected 145-bp product was obtained by RT-PCR (for specific examples, see Figure 1). No products were detected in the absence of either reverse transcriptase or DNase 1-treated RNA, which demonstrated lack of contamination by chromosomal DNA or previous PCR products. The 145-bp fragment was excised, purified, and cloned into pUC19, and individual clones were sequenced. A total of 857 sequences were obtained, comprising 333 from the synovial tissue of 6 patients with RA, 117 from the synovial fluid cells of 3 patients with RA, 224 from the synovial tissue of 3 patients with OA, 137 from the synovial tissue of 3 subjects without joint disease, and 46 from a pool of PBMC from 4 healthy male subjects (Table 1).

Classification ofpol sequences. Amplified nucleic acid sequences were heterogeneous within and between subjects. However, analysis of 3 repeat samples in 1 subject yielded similar frequencies of individual se- quences, despite different total numbers of clones gen- erated each time. While this suggests that the within- subject pattern is reproducible, an indeterminant degree of heterogeneity may occur due to the semiquantitative nature of the PCR method and due to variable efficiency

NAKAGAWA ET AL 630

A.

30

40

60

B .

30

'I "

su

60

70

80

90

I I I ---

I ril

r l - 7 I I 1 1 - I I IA I

I , F6 F4 I * ' F I ' F2

Figure 2. Dendrographic representation of pol sequences, amplified by reverse transcription-polymerase chain reaction (RT-PCR), in samples obtained from 9 patients with rheumatoid arthritis (RA). As described in Patients and Methods, the GCGiPileup program was used for multiple alignment of 857 sequences amplified by RT-PCR. A, Dendrogram of 450 sequences from RA patients, aligned on the x-axis. The y-axis shows the percentage identity of the sequences. The sequences were classified into 6 major families, Fl-F6, and some minor families indicated by asterisks on the x-axis. Likewise, 407 sequences from control subjects were classified into the same corre- sponding 6 families (data not shown). B, Dendrogram of 117 se- quences from the synovial fluid cells of 3 patients with RA.

in cloning PCR products. Multiple sequence analysis (by GCG/Pileup) was used to align and classify nucleic acid sequences. Dendrographic representation of the 450 sequences obtained from the Y patients with RA re- vealed 6 major families (Fl-F6) (Figure 2A). Minor families were not analyzed because the number of sequences in each (< 12) was too small to allow compar- ison between RA and control subjects.

When sequences in each of the 6 major families were screened for homologies in the major databases, 5 of the 6 were found to be related to known human ERVs (HERVs): F1 was undefined, but related to murine and primate type C RVs; F2 comprised ERV-Y-related

sequences (27); F3, RTVL-H-related sequences (28); F4, HERV-K-related sequences (29); F5, RTVL-I- related sequences (30); and F6, HERV-L-related se- quences (31). Therefore, these RNA-amplified pol tran- scripts were probably derived from ERV sequences expressed in the synovial tissue or synovial fluid cells. Likewise, the 407 sequences obtained from the non-RA control subjects could also be classified into 6 major families, corresponding to the 6 families found in the RA patient samples (data not shown).

Dendrographic representations of sequences from the RA, OA, or normal synovial tissues, or from PBMC, revealed no major pattern differences. However, the pattern of sequences from RA synovial fluid cells was distinct (Figure 2B). Therefore, the relative fre- quency of the individual sequences in each family, defined as the percentage of the total number of se- quences found in each subject, was compared among RA, OA, and control samples (Table 2). The relative frequencies in the major family, F1, were lower (P = 0.009) in RA synovial fluid cells compared with synovial tissues, but were not significantly different among the RA, OA, and normal synovial tissues. The latter result is indirect evidence for the reproducibility of the RT-PCR strategy. F3 sequences were not detected in RA synovial fluid cells. However, the relative frequencies in the F2, F4, and F6 families were significantly higher in RA synovial fluid cells compared with RA synovial tissue (P = 0.02, P = 0.05, and P = 0.02, respectively), or compared with RA, OA, and normal synovial tissues combined (P = 0.04, P = 0.03, and P = 0.004, respectively).

Subclassification of pol sequences in each family. Although the sequences in F1 were at least 88% identi- cal, those in F2-F6 were more heterogeneous (Figure 2). For example, the most dissimilar sequences in F4, the least homogeneous family, were no more than 50% identical. This is not surprising, considering that many ERV families such as RTLV-H and HERV-K comprise up to 1,000 multigene copies per haploid genome (32), some of which could be transcriptionally or translation- ally active. This complexity, and the probability that only minor sequence differences in expressed ERV are likely to be responsible for potential pathogenicity, encour- aged us to subclassify the 6 families into more homoge- neous groups to identify potential disease-associated sequences.

A group (G) was defined as that containing sequences that were 290% identical and that were present in 2 2 subjects. The 6 major families were thus subclassified into 33 groups. When the deduced amino

ENDOGENOUS SYNOVIAL RETROVIRUSES IN RA

M Y

a,

a .- 4-

v1 .- I .- B 2

8 a, Y

w- * a,

a .- 4-

h

v 2

s Y z 2 e L

a, .- Y

3

2

L a, D ;

z a e

G- M

a, 2

2 Y - v

v ) 3 3

a,

U E E 5 E 3 .- m

B 6

.L

a, M

.+ Y

63 1

632 NAKAGAWA ET AL

Table 3. Deduced amino acid sequences of pol transcripts in 6 major families subclassified into groups*

Number of clones

Family-group, sequence RA RA OA Normal Normal Homology to ST SF ST ST PBMC Total ORF known ERV

F1-G1 FRDGPHLFRNALEQE*KELELEMG-VILQ FRDGPHLFRNALAQE*KELELEMG-VILQ FRDGPHLFRNALAQE* *ELELEMG-VILQ FRDGPHLFRNALAQE*KEL*LEMG-VILQ FRDGPHLFRNVLVQE*KEVELEMG-VILQ FRDGPHLFRNFLAQE*KELELEMG-VILQ FRDGPHLSRNAWEQE* KELELEMG-VILQ FRDGPHLFRNAWEQE*KELELEMG-VILQ FRDGPHLFRNAWEQE*KELELEMG-VISQ FRDGPHLFRNVLEQE*KELELEMG-VILQ FRDGPHLFRNVLAQE* KELELEMG-VILQ FRDGPHLFRHALAQE*KELELEMG-VILQ FRDGPHLFRNALEQE*KELELEMG- W L Q FRDGPHLFRNALARE*KELELEMG-VILQ FRDGPHLFRNALGRE*KELELEMG-VILQ FRDGPHLFRNDLAQE*KELELEMG-VILQ FRDGPHLFRNDLEQE*KELELEMG-VILQ FRDGPHLFRNVLGQE*KELELEMG-VILQ FRDGPHLFRNALGQE*KELELEMG-VILQ FRDGPHLFRNALVQE*KELELEMG-VILQ FRDGPHLFRNA* AQE*KELELEMG-VILQ FRDGPHLFRNAWAQE*KELELEMG-VILQ FRDGPHLFRNASAQE*KELELEMG-VILQ FRDGPHLFRNALAQE* KELEPEMG-VILQ FRDGPHLFRNTLEQE*KELELEMG-VILQ FRDGPHLFRNALEQE*RELELEMG-VILQ FRDGPHLFRNALTQE*KELELEMG-VILQ FRDGPHLFRNALQQE*KELELEMG-VILQ FRDGPHLFRNAMEQE*KELELEMG-VILQ FRDGPHLFRNALEQEYKELELEMG-VILQ FRDGPHLFRNALEQE* KELELEMR-VILQ FRDGPHLFRNALEQE* KELELEVG-VILQ FRDGPHLFRNALEQE* KELELETG-VILQ FRDGPHLFRNALEQE*KELELDMG-VILQ FRDGPHLFRNALAQK* KELELEMG-VILQ FKDGPQLFRNALAQE* KELELEMG-VILQ FRDGPHLFRNAVEQE*MELELDLG-VILQ

Total for Fl-GI F2-GI

FRD-SPHLFGQALAEDLSQFSYLDALVLW

LRDSPHLFGQALAQDLRQFSY LNSLVLW

FKDSPHLFGQALAQDLSQFSYLNTLIFW

FRDSsHLFGQALDDLSQFSY LDTLVLW FRDSmLFGQALmDLSQFSYVDTLVLW FRDSPHLLGQALmDLSQFSYLDTLVLW FRDSSIILLGQALmDLSQFSYLDTLVLG FRDSsHLFGQALmVSQFSYVDTLVLW

F2-G2

F2-G3

F2-G4

FRDS~LFGQALTDLSQFSYVDTLVLW FRDSmLFGQALBDLSQFSYLDTLVLW FRDSPHLFGQALmDLSQFSYLDTLVLW

Total for F2-G4 F2-G5

FRDSPYLFGQALAQDLSQFSYLDTHVLW FRDRELFGQALAQDLSQFSYLDTHVLW

Total for F2-G5 F2-G6

FRDSPHLFGQALAQDLSQFSHLDTLVLO

82 87 0 2 1 1 1 2 0 8 1 1 1 0 0 0 1 1 1 3 0 0 0 0 0 0 1 0 1 1 1 1 0 0 0 0 1

199

0

3

0

5 3 2 0 0 0 0 1

11

8 0 8

12

2 14 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 1 1 0 0 1 1 0 0 0 0 0 0 0 0 0 1 0 0

22

0

1

2

3 3 0 0 1 0 1 0 8

2 0 2

2

39 92 1 0 0 0 0 0 0 0 0 0 0 1 0 1 0 0 0 0 0 1 1 0 0 0 0 2 0 0 0 0 1 0 0 1 0

140

3

0

0

2 0 0 0 0 0 0 0 2

3 0 3

14

18 41 0 0 0 0 1 0 1 0 1 0 0 0 0 0 0 0 0 2 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0

65

1

1

0

0 1 0 1 0 0 0 0 2

2 1 3

3

8 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0

15

0

0

0

0 0 0 0 0 1 0 0 1

1 0 1

2

149 240

1 2 1 1 2 2 1 8 2 1 1 1 1 1 1 1 1 6 1 1 1 1 1 1 1 2 1 1 1 1 1 1 1 1 1

441

4

5

2

10 7 2 1 1 1 1 1

24

16 1

17

33

No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No No -

No

Yes

Yes

Yes Yes Yes Yes Yes No Yes Yes -

Yes Yes -

Yes

Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknonw Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown Unknown

-

ERV-9

ERV-9

ERV-9

ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 -

ERV-9 ERV-9 -

ERV-9

ENDOGENOUS SYNOVIAL RETROVIRUSES IN RA 633

Table 3. (Cont'd)

Number of clones

RA RA OA Normal Normal Homology to Family-group, sequence ST SF ST ST PBMC Total ORF knownERV

F2-G6 FRDSPHLFGQALAQDLSQFSYLDTLVLQ FRDSPHLFGQALAQDLSQFSYLDTLVLR FRDSPHLFGQALAQDLSQFSYLDTHVLQ

FRDSPHLFGQALAQDWSQFSHLDTLVLQ FRDSPHLFGQALAQDLSQLSHLDTLVLQ FRDSPHLFGQALDQDLSQFSHLDTLVLQ FRDSPHLLGQALAQDLSQFSHLDTLVLQ FRDSPHLFGQALAQDLSQFSYLDALVLQ FRDSLHLFGQALAQDLSQFSYLDTLVLQ FRDSPHLFGQVLAQDLSQFSHLDTLVLQ

- - - SPHLFGQALAQDLSQFSYLDTHVLQ

Total for F2-G6 F2-G7

FRDSPHLFGQALAQDLSHFLHPGTLILQ FRDSPHLFGQALAHDLSHFLHPGTLILQ FRDSPHLFGQALAQDLSHFSHPGTLILQ

Total for F2-G7 F2-G8

FRDRPHLFGQALAQDLGHFSSPGTLVLQ FRDEPHLLGQALAQDLGHFSSPGTLVLQ FRDBPHLFGQALDQDLGHFSSPGTLVLQ FKDBPHLFGQALAHDLGYFSSPGTLVLQ

Total for F2-G8 F3-G1

FRDSPHY FSQALSEDLLSFRPFVSHLIQ FRDSPHYFSQALSRDLFSFRPFVSHLIQ FRDSPHY FSQALSEDLFTFRPFVSHLIQ FRDSPHYFSQALSEDLLSFRPFVSHLIH

Total for F3-G1 F3-G2

FRDSPHYFSQALSHDLLYFHPSASHLIQ FRDSPHY FSQALSHDLLSFHPSASHLIQ FKDGPHYFSQALSHDLLSFHPSASHLIQ FRDSPHY FSQALSHDLLSCHPSASHLIQ FRDSPHY SSQALSHDLLSFHPSASHLI* FRDSPHYFSQALSHDLLSFHPSASHLI *

Total for F3-G2 F3-G3 FRDNPHYFS*ALSHDLLSFCPSVFHLIQ FRDNPHYFS *ALSHDLLSFCPPVFHLIQ FRDNPHYFSI ALSHDLLSFCPSVSHLIQ FRDNPHYFS*ALSHDLLSFCPSVCHLIQ

Total for F3-G3 F3-G4

FRDSPHY FSQALSHDLLSmSHLIQ FRDSPHYLSHDLSHDLLSmSHLIQ

Total for F3-G4 F4-G1 MLNS-TICQYYVGSILKPVRDQFPQFFIIH

MLNSPTICQLYVGOVLSPV' ARFPEAYILH MLNSPTICQFYVGOVLSPV* ARFPEAYILH MLNSPTICQLY VGQVLSPVGARFPEAY ILH MLNSPTICQFYVGOVLSPVGARFPEAY FLH MLNSPTICQLYVGOVLSPV* ARFPLAYILH

F4-G2

Total for F4-G2 F4-G3

MLNSPIIFQTFVGOAIEPTHKKFSQSY IIH MLNSFTICQSFYGOAIEPTPKKFSQSYIIH

Total for F4-G3

9 0 0 1 1 1 0 1 1 0 1

27

0 0 0 0

4 0 0 0 4

8 3 1 1

13

0 0 0 0 0 0 0

11 0 1 0

12

4 0 4

3

1 0 0 0 1 2

2 0 2

12 2 0 0 0 0 0 0 0 0 0

16

6 0 0 6

3 0 1 1 5

0 0 0 0 0

0 0 0 0 0 0 0

0 0 0 0 0

0 0 0

1

0 0 0 0 0 0

0 2 2

5 0 1 0 0 0 1 0 0 1 0

22

1 0 0 1

0 0 0 0 0

0 0 0 0 0

1 0 0 0 1 0 2

3 0 0 1 4

2 1 3

0

4 0 0 0 0 4

2 0 2

1 0 0 0 0 0 0 0 0 0 0 4

3 1 1 5

0 0 0 0 0

2 0 0 0 2

2 2 1 1 0 0 6

0 1 0 0 1

4 0 4

1

1 2 1 1 0 5

0 2 2

2 0 0 0 0 0 0 0 0 0 0 4

1 0 0 1

1 3 0 0 4

0 1 0 0 1

0 0 0 0 0 1 1

0 0 0 0 0

0 0 0

0

0 0 0 0 0 0

0 0 0

29 2 1 1 1 1 1 1 1 1 1

73

11 1 1

13

8 3 1 1

13

10 4 1 1

16

3 2 1 1 1 1 9

14 1 1 1

17

10 1

11

5

6 2 1 1 1

11

4 4 8

Yes Yes Yes No Yes Yes Yes Yes Yes Yes Yes -

Yes Yes Yes -

Yes Yes Yes Yes -

Yes Yes Yes Yes -

Yes Yes Yes Yes No No -

No No No N O -

No No -

No

No No Yes Yes NO -

Yes Yes -

ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9 ERV-9

-

ERV-9 ERV-9 ERV-9

-

ERV-9 ERV-9 ERV-9 ERV-9

-

RTVL-H RTVL-H RTVL-H RTVL-H

-

RTVL-H RTVL-H RTVL-H RTVL-H RTVL-H RTVL-H

-

RTVL-H RTVL-H RTVL-H RTVL-H

-

RTVL-H RTVL-H

-

HERV-K

HERV-K HERV-K HERV-K HERV-K HERV-K

-

HERV-K HERV-K

-

634 NAKAGAWA ET AL

Table 3. (Cont’d)

Number of clones

Family-group, sequence

F4-G4 MLNSPTICQTYVROAIEPTRKKFSQCY IIH MLNSPTICQTYVGOAIEPTSKKFSQCY IIH MLNSPTICQTYVGEAIEPTRKKFSQCYIIH MLNSPTICQTYVROTIEPTRKKFSQCYIIL MLNSPTICQTYVROAIEPTRKKFSLCYIIH ILNSPTICQTY VROAIEPTRKKFSQCY IIH MLNSPMTCQTYVROAIEPTRKKFSQCY IIH

Total for F4-G4 F4-G5

MLNNLPDICRASN*TYS* KN- QCYIIH F4-G6

MLNSPTICQTYVGKTIKPVREOF-KSY IIH MLNSPTICQTYVGKTIKPVREOFKKCYIIN MLNSPTICQTYVGKTIKPVRKOFKKCYIIN LLNSPTICQTYVGKTIKPVREOF-KSY IIH MLNSPTIGQTY VGKTIKPVREOFKKCY IIN

Total for F4-G6 F4-G7

MLNSPTICQTFVDRAIOTVKQFPDCYIIH MLNSPTICQTFVDRAIOTVKQFPDCY IIH MLNSPTICQTFVDRAIOTVKNQFPDRYIIH

Total for F4-G7 F4-G8 MLNSPTMCQYRVNKALLPSRKEFPNCKIIH MLNSPTMCQYRVNKALLPSRKEFPNFKIIH

Total for F4-G8 F4-G9

ILNSPTLCQHFVGRA*KEPWKFPTAY IIH ILNSPTLCQHFVGRA*KEPRKFPTAY IIH

Total for F4-G9 F5-G1

FTESPYLLSQILEWLEKFSLPSCICPLQ

FTDSPNLFGQILEaLEKMSIPKCISMLQ FTDSPNLFGQILEELLEKMSIPKCLSMLQ

F5-G2

Total for F5-G2 F5-G3

FTDSPNLFGQILEWLEKVVIPKQICLLQ FTDSPNLVGQILEWLEK-ICLLQ FTDSPNLLGQILEWLEK-ICLLQ STDSPNLFGQILEQJLEKVVIPKOICLLQ

Total for F5-G3 F5-G4 FTDSPNIFGEILEQALEKVFIPEOICLLQ FTDSPNIFGEILEHALEKVFIPEOICLLQ

Total for F5-G4 F5-G5

FIDSPNLLGQILEWLEKVVIPEOICLLQ FIDSPNLEGQILEWLEKVVIPEOICLLQ FIDSPNLVGQILEWLEKVVIPE~ICLLQ FIDSPNLFGQILEQJLEKVVIPEQICLLQ FIDSPNLCGQILEWLEKVVIPEQICLLQ FIDSPNLFGQILEQJPEKVVIPEQICLLQ FIDSPNLLGQILEQJLEKVVIPEOFCLLQ

Total for F5-G5 F6-Gl

Y ISSLALCH-FFHRVFDHISLPQDITLVH

YNNSPAPCHSLVHRDLDCLSLPQDITLVH Y NNSPAPWHSLVHRDLDCLSLPQDITLVH

F6-G2

Total for F6-G2

RA ST -

0 0 1 0 1 0 1 3

0

1 0 0 0 0 1

2 1 1 4

2 0 2

1 1 2

0

1 1 2

1 0 0 1 2

0 0 0

0 0 0 0 0 0 0 0

1

0 0 0

RA SF -

0 0 0 1 0 0 0 1

0

3 0 2 1 0 6

11 0 0

11

1 1 2

0 0 0

0

0 0 0

1 0 0 0 1

0 0 0

0 0 0 0 0 0 0 0

3

2 1 3

OA ST -

6 0 0 0 0 0 0 6

1

0 0 0 0 0 0

2 0 0 2

1 0 1

0 0 0

3

1 0 1

3 0 0 0 3

1 0 1

1 0 1 0 1 0 1 4

0

0 0 0

Normal ST

2 2 0 0 0 1 0 5

0

0 1 0 0 1 2

1 0 0 1

0 0 0

0 0 0

1

1 0 1

7 0 0 0 7

0 1 1

0 0 0 0 0 0 0 0

0

0 0 0

Normal Homology to PBMC Total ORF knownERV

1 0 0 0 0 0 0 1

3

0 2 0 0 0 2

0 0 0 0

0 0 0

0 0 0

0

0 0 0

0 1 1 0 2

0 0 0

0 1 0 1 0 1 0 3

0

1 0 1

9 2 1 1 1 1 1

16

4

4 3 2 1 1

11

16 1 1

18

4 1 5

1 1 2

4

3 1 4

12 1 1 1

15

1 1 2

1 1 1 1 1 1 1 7

4

3 1 4

Yes Yes Yes Yes Yes Yes Yes -

No

No Yes Yes No Yes -

Yes Yes Yes -

Yes Yes -

No No -

Yes

Yes Yes -

Yes Yes Yes Yes -

Yes Yes -

Yes Yes Yes Yes Yes Yes Yes -

No

Yes Yes -

HERV-K HERV-K HERV-K HERV-K HERV-K HERV-K HERV-K

-

HERV-K

HERV-K HERV-K HERV-K HERV-K HERV-K

-

HERV-K HERV-K HERV-K

-

HERV-K HERV-K

-

HERV-K HERV-K

-

RTVL-I

RTVL-I RTVL-I

-

RTVL-I RTVL-I RTVL-I RTVL-I

-

RTVL-I RTVL-I

-

RTVL-I RTVL-I RTVL-I RTVL-I RTVL-I RTVL-I RTVL-I

-

HERV-L

HERV-L HERV-L

-

ENDOGENOUS SYNOVIAL RETROVIRUSES IN RA 635

Table 3. (Cont’d)

Number of clones

RA RA OA Normal Normal Homology to Family-group, sequence ST SF ST ST PBMC Totdl ORF known ERV

F6-G3 Y INSLVLCHNLI * RDLDPFCFH-DITLVH YINSLVWCHNLI*RDLDPFGFH-DITLVH

Total for F6-G3 F6-G4

YINSPALCHNLIWRPLDRFSLPQDITLVH Y INSSALCHNLIWRPLDRFSLPQDITLVH Y INSPTLCHNLIWRPLDRFSLPQDITLVH Y INSPAWCHNLIWRPLDRFSLSQDITLVH YINSPALSHNLIWRPLDRFSLSQDITLVH YINSPGWCHNLIGRPLDRFSLSQDITLVH

Total for F6-G4 F6-G5

YMNY LALCHNLIORELDRFLAPKDITLVH

YINTLALCHNLVORDLEHFFLP*NITLVH F6-G6

1 1 1 0 2 1

0 0 0

0 0 0

0 0 0

2 1 3

No No

HERV-L HERV-L

-

4 3 2 0 0 0 0 1 0 1 0 0 6 5

0 0

1 0

0 0 0 0 0 0 0

0

8 2 1 1 1 1

14

2

Yes Yes Yes Yes Yes Yes

HERV-L HERV-L HERV-L HERV-L HERV-L HERV-L

-

Yes HERV-L 1

1 3 0 0 0 4 No HERV-L

* Each group was distinguished by a specific amino acid motif (indicated by underline). Although sequences in the same group shared 290% identity, minor differences in the nucleic acid sequence occasionally resulted in amino acid substitutions (indicated by boldface type). A hyphen in a sequence indicates an amino acid deletion due to a frameshift mutation. Termination codons are shown by asteriskr. ORF = open reading frame; ERV = endogenous retrovirus; see Table 2 for other definitions.

acid sequences in these groups were aligned, they were found to be distinguished by specific amino acid motifs (Table 3). All of the sequences in F1 contained termi- nation codons or frameshift mutations, but many of those in the F2-F6 groups had an open reading frame. Thirty-two sequences in the minor families (data not shown) did not belong to any of the groups in the 6 major families. Considering these findings and the se- quences in Table 3, at least 65 different pol transcripts were detected.

The increased relative frequency of individual sequences in F2, F4, and F6 in RA synovial fluid cells could be attributed, in some cases, to specific groups. Relative frequencies in the groups F2-G6, F2-G7, and F2-G8 combined, but not individually, were increased in RA synovial fluid cells compared with synovial tissue (P = 0.03). Sequence frequencies in F4-G6 were in- creased in RA synovial fluid cells compared with RA synovial tissue (P = 0.02) or compared with RA, OA, and normal synovial tissues (P = 0.01). The frequency in F6-G4 was increased in RA synovial fluid cells compared with RA synovial tissue, but was only significantly in- creased (P = 0.03) compared with RA, OA, and normal synovial tissues. Although caution is required in inter- preting such results from small groups, it is noteworthy that some of the sequence groups, namely F2-G1, F3- G2, F5-G1, F5-G4, F5-G5, and F6-G5, were not de-

tected in either the RA synovial tissue or RA synovial fluid cells.

DISCUSSION

RT-PCR analysis, using degenerate primers to conserved sequences in the pol genes of many known RVs, revealed a highly polymorphic expression of RV- related sequences, many of which were similar to known ERV families, in RA, OA, and normal synovial tissues. Overall, -40% of the sequences were interrupted by termination codons and/or frameshift mutations. We assume that the amplified sequences were derived from ERVs, but cannot exclude the possibility of unknown infectious RVs being the source, because some se- quences showed no more than 50% similarity to known ERV families.

Most of the 857 pol transcripts could be broadly classified into 6 major and several minor families. Five of the 6 major families were related to known ERV fami- lies. The 6 major families were then subclassified into 33 groups, each comprising sequences from 2 2 subjects and each with 290% identity. Thus, including the minor families, the 857 sequences from the 9 RA and 6 non-RA subjects comprised 65 groups, each with 290% identity. These were distinguished by specific amino acid motifs, although a few minor amino acid substitutions

636 NAKAGAWA ET AL

were occasionally observed within sequences in the same group.

Even by taking advantage of the highly conserved regions ofpol in most RVs, we did not necessarily detect all RV-related transcripts in synovial tissues. We did not, for example, detect h4-1 (33), which was found by Takeuchi et a1 (34) in synovial tissue. In addition, considering that the sequences that we amplified repre- sent a minor part of pol, a much larger variety of ERVs than we observed might be expressed in synovial tissues.

Multiple ERV sequences have been demon- strated in some cell lines and tissues, but only a re- stricted number have been analyzed, because detection was by Northern blot (34,35). In contrast, Franklin et al (36) cloned mouse mammary tumor virus (MMTV)- related gag-pol sequences from human breast cancer cell lines. Later, Medstrand and Blomberg (37) used RT- PCR with RNA from the PBMC of healthy individuals to amplify the pol region of MMTV-related sequences and showed that the multiple sequences generated could be classified into 6 groups with 275% nucleotide se- quence identity. Each of the 6 groups was related to MMTV, murine intracisternal A particles, or the HERV-K10 variant of HERV-K, which have at least 60% nucleotide sequence identity. Although Medstrand and Blomberg amplified a different region of pol than we did, our F4 family (HERV-K-related), whose se- quences showed no more than 50% identity, could include their sequences. The present study, therefore, not only confirms the expression of multiple ERVs in human tissues, but expands the findings of previous studies, because detection and classification of such a large variety of RV-related transcripts, especially in diseased tissue, has not been previously reported.

To address the question of disease-specific ex- pression of RV transcripts, we first compared the ex- pression patterns of the 6 major families among RA, OA, and normal tissues, by dendrographic and relative- frequency analysis. The patterns in normal and diseased synovial tissues were similar, but the pattern in RA synovial fluid cells was distinct. The pattern in normal PBMC was more similar to that in synovial tissue than to that in RA synovial fluid cells. In RA synovial fluid cells, compared with RA synovial tissue, we observed a signif- icant decrease in the sequence frequency of the unde- fined family, F1, a significant increase in the frequency of ERV-9, HERV-K, and HERV-L sequences, and the absence of RTVL-H sequences. In addition, the relative frequency of sequences in RTVL-I, a type C RV-related ERV, was 4 times lower in RA synovial tissue and 8

times lower in RA synovial fluid cells compared with normal synovial tissue.

There is a paucity of evidence for tissue-specific expression of human ERV (38). Our findings could reflect differences in cell types in synovial fluid and tissue. Neutrophils and activated lymphocytes predomi- nate in RA synovial fluid, and macrophages and fibroblast-like cells in synovial tissue (39). The cellular composition of synovial tissue from patients with RA is qualitatively different from that of healthy subjects, with more CD4+ T cells and a much higher proportion of macrophage-like cells than fibroblast-like cells (39). Despite this, the expression pattern of the 6 major families did not appear to differ among the RA, OA, and normal synovial tissues. Recently, we had the opportu- nity to analyze synovial fluid from a patient with spondyl- arthropathy. Although only 13 sequences were gener- ated, the expression pattern of the 6 major families was similar to that found in normal PBMC, but not to that found in RA synovial fluid cells (data not shown). In particular, no HERV-L sequences were observed. Fur- ther studies will be necessary to control for differences in the cell composition of RA synovial fluid.

We have not been able to obtain RNA from the small volumes of synovial fluid that can be aspirated from the knee joints of organ-donor subjects. Because synovial fluid is difficult to obtain even from other non-RA subjects, ERV expression could be compared in purified neutrophils and lymphocytes from synovial fluid and PBMC of RA subjects. In addition to differences in cellular composition, differential or absent expression of ERV in the synovial compartment in RA could reflect the influence of endogenous or exogenous factors. Microbial-derived B cell or T cell mitogens (40) and hormones or cytokines (15,26,41) can up-regulate ERV, and components of certain DNA viruses, such as SV40 and Herpes simplex, can transactivate the expression of human ERV (42,43). Variation in the expression of ERV could, therefore, be secondary to factors involved in the pathogenesis of RA.

Brodsky et a1 (43) performed RT-PCR on leuko- cyte RNA, with primers bounding a 90-bp region of pol in the HERV-K family, and found a single nucleic acid substitution that was restricted to myeloid leukemic subjects. Although they compared only 6 sequences from leukemic subjects with 8 from normal controls, their finding also provides preliminary evidence that differen- tial expression of ERV sequences is disease-related. Differential hybridization between PCR fragments from disease and control subjects may not necessarily detect such small sequence differences.

ENDOGENOUS SYNOVIAL RETROVIRUSES IN RA 637

Therefore, based on the hypothesis that minor, but unique, sequence differences may be potentially pathogenic, we sought disease-specific sequences in the 33 groups from the major families, each of which was distinguished by a specific amino acid motif. The fre- quencies in some groups, e.g., F4-G6 (HERV-K- related) and F6-G4 (HERV-L-related), were higher in RA synovial fluid cells than in normal synovial tissue. These 2 sequences were not observed in the spondyl- arthritis sample. It is noteworthy that HERV-K10 has a particle-forming ability (41,45), and related sequences were recently detected in RV particles isolated from the synovial fluid of 6 RA subjects (46). However, some groups from RTVL-H and RTVL-I were not observed in RA subjects. A subfamily of RTVL-H has been reported to encode the conserved motif of the envelope immuno- suppressive peptide p15E (47), present in all mammalian transforming type C RVs (48). Therefore, it is conceiv- able that pl5E-related ERV products could be gener- ated in humans and could play a role in immunologic homeostasis. If the expression of RTVL-H or RTVL-I was absent or reduced, this might contribute to the susceptibility to, or persistence of, diseases such as RA.

In conclusion, PCR-based screening demon- strates a complex expression pattern of multiple ERV transcripts in both normal and diseased synovial tissue. This complexity can be reduced to sequences defined by unique motifs, representing individual gene products. Specific ERV transcripts appear to be differentially expressed in the synovial compartment in RA. This direct evidence is a basis for further investigating the role of ERV in the pathogenesis of RA.

ACKNOWLEDGMENT

We wish to thank Margaret Thompson for providing secretarial assistance.

1.

2.

3.

4.

5.

REFERENCES

Crawford TB, A d a m DS, Cheevers WP, Cork LC: Chronic arthritis in goats caused by a retrovirus. Science 207:997-999, 1980 Johnson TM, Duvic M, Rapini RP, Rios A: AIDS exacerbates psoriasis (letter). N Engl J Med 313:1415, 1985 Nishioka K, Maruyama I, Sato K, Kitajima I, Nakajima Y, Osame M: Chronic inflammatory arthropathy associated with HTLV-1. Lancet i:441, 1989 Slamon DJ, Shimotono K, Cline MJ, Golde DW, Chen IS: Identification of the putative transforming protein of the human T-cell leukemia viruses HTLV-I and HTLV-11. Science 226:61-65, 1984 Maruyama M, Shibuya H, Harada H, Hatakeyama M, Seiki M, Fujita T, Inoue J, Yoshida M, Taniguchi T: Evidence for aberrant activation of the interleukin-2 autocrine loop by HTLV-1-encoded p40x and T3ITi complex triggering. Cell 48343-350, 1987

6.

7.

8.

9.

10.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

Cross SL, Feinberg MB, Wolf JB, Holbrook NJ, Wong-Staal F, Leonard WJ: Regulation of the human interleukin-2 receptor a chain promoter: activation of a nonfunctional promoter by the transactivator gene of HTLV-1. Cell 49:47-56, 1987 Fuji M, Sassone-Corsi P, Verma I: c-fos promoter trans-activation by the tax1 protein of human T-cell leukemia virus type I. Proc Natl Acad Sci U S A 8523526-8530, 1988 Iwakura Y, Tosu M, Yoshida E, Takiguchi M, Sat0 K, Kitajima I, Nishioka K, Yamamoto K, Takeda T, Hatanaka M, Yamamoto H, Sekiguchi T Induction of inflammatory arthropathy resembling rheumatoid arthritis in mice transgenic for HTLV-1. Science

Di Giovine FS, Bailly S, Bootman J, Almond N, Duff GW: Absence of lentiviral and human T cell leukemia viral sequences in patients with rheumatoid arthritis. Arthritis Rheum 37:349-358, 1994 Talal N, DauphinCe MJ, Dang H, Alexander SS, Hart DJ, Garry RF: Detection of serum antibodies to retroviral proteins in patients with primary Sjogren’s syndrome (autoimmune exocri- nopathy). Arthritis Rheum 33:774-781, 1990 Talal N, Garry RF, Schur PH, Alexander S, Dauphinee MJ, Livas IH, Ballester A, Takei M, Dang H: A conserved idiotype and antibodies to retroviral proteins in systemic lupus erythematosus. J Clin Invest 85:1866-1871, 1990 Brookes SM, Pandolfino YA, Mitchell TJ, Venables PJW, Shattles WG, Clark DA, Entwistle A, Maini RW: The immune response to and expression of cross-reactive retroviral gag sequences in auto- immune disease. Br J Rheumatol 31:735-742, 1992 Ziegler B, Gay RE, Huang G, Fassbender H-G, Gay S: Immuno- histochemical localization of HTLV-1 p19- and p24-related anti- gens in synovial joints of patients with rheumatoid arthritis. Am J Pathol 135:l-5, 1989 Ziegler BL, Thomas C A Detection of HIV-I related antigens in rheumatoid synovial tissues. Acta Histochem Cytochem 24:l-9, 1991 Shattles WG, Brookes SM, Venables PJW, Clark DA, Maini R: Expression of antigen reactive with a monoclonal antibody to HTLV-1 p19 in salivary glands in Sjogren’s syndrome. Clin Exp Immunol 89:46-51, 1992 Yoshiki T, Mellors RC, Strand M, August J T The viral envelope glycoprotein of murine leukemia virus and the pathogenesis of immune complex glomerulonephritis of New Zealand mice. J Exp Med 140:1011-1027, 1974 Banki K, Maceda J, Hurley E, Ablonczy E, Mattson DH, Szegedy L, Hung C, Per1 A Human T-cell lymphotropic virus (HTLV)- related endogenous sequence, HREV-1, encodes a 28-kDa pro- tein: a possible autoantigen for HTLV-1 gag-reactive autoantibod- ies. Proc Natl Acad Sci U S A 89:1939-1943, 1992 Per1 A, Colombo E, Dai H, Aganval R, Mark KA, Banki K, Poiesz BJ, Phillips PE, Hock SO, Reveille JD, Arnett FC: Antibody reactivity to the HRES-1 endogenous retroviral element identifies a subset of patients with systemic lupus erythematosus and overlap syndromes: correlation with antinuclear antibodies and HLA class I1 alleles. Arthritis Rheum 38:1660-1671, 1995 Iversen 0-J: Isolation of virus-like particles in urine from a psoriatic patient. Acta Pathol Microbiol Immunol Scand Sect B 91:407-412, 1983 Perron H, Lalande B, Gratacap B, Laurent A, Genoulaz 0, Geny C, Mallaret M, Schuller E, Stoebner P, Seigneurin JM: Isolation of retrovirus from patients with multiple sclerosis. Lancet 337362- 863, 1991 Garry RF, Fermin CD, Hart DJ, Alexander SS, Donehower LA, Hong L-Z: Detection of a human intracisternal A-type retroviral particle antigenically related to HIV. Science 250:1127-1129, 1990 Shih A, Misra R, Rush MG: Detection of multiple, novel reverse transcriptase coding sequences in human nucleic acids: relation to primate retroviruses. J Virol 63:64-75, 1989

235: 1026-1 028, 1991

NAKAGAWA ET AL

23. Arnett FC, Edworthy SM, Bloch DA, McShane DJ, Fries JF, Cooper NS, Healey LA, Kaplan SR, Liang MH, Luthra HS, Medsger TA Jr, Mitchell DM, Neustadt DH, Pinals RS, Schaller JG, Sharp JT, Wilder RL, Hunder GG: The American Rheuma- tism Association 1987 revised criteria for the classification of rheumatoid arthritis. Arthritis Rheum 31:315-324, 1988

24. Chomczynski P, Sacchi N: Single-step method of RNA isolation by acid guanidium thiocyanate-phenol-chloroform extraction. Anal Biochem 162:156-159, 1987

25. Program Manual for the Wisconsin Package. Version 8. Madison, WI, Genetics Computer Group, September 1994

26. Gish W, States DJ: Identification of protein coding regions by database similarity search. Nature Genet 3:266-272, 1993

27. Mantia GL, Maglione D, Pengue G, Cristofano AD, Simeone A, Lanfrancone L, Lania L Identification and characterization of novel human endogenous retroviral sequences preferentially ex- pressed in undifferentiated embryonal carcinoma cells. Nucleic Acids Res 19:1513-1520, 1991

28. Mager DL, Henthorn PS: Identification of a retrovirus-like repet- itive element in human DNA. Proc Natl Acad Sci U S A 81:7510-7514, 1984

29. Ono M, Kawakami M, Ushikubo H: Stimulation of expression of the human endogenous retrovirus genome by female steroid hormones in human breast cancer cell line T47D. J Virol61:2059- 2062, 1987

30. Maeda N, Kim H-S: Three independent insertions of retrovirus- like sequences in the haptoglobin gene cluster of primates. Genomics 8:671-683, 1990

31. Cordonnier A, Casella J-F, Heidmann T: Isolation of novel human endogenous retrovirus-like elements with foamy virus-related pol sequence. J Virol 69:5890-5897, 1995

32. Krieg AM, Gourley MF, Perl A. Endogenous retroviruses: poten- tial etiologic agents in autoimmunity. FASEB J 6:2537-2544, 1992

33. Repaske R, Steele PE, O’Neill RR, Rabson AB, Martin M A Nucleotide sequence of a full-length human endogenous retroviral segment. J Virol 54:764-772, 1985

34. Takeuchi K, Katsumata K, Ikeda H, Minami M, Wakisaka A, Yoshiki T: Expression of endogenous retroviruses, ERV3 and h4-1, in synovial tissues from patients with rheumatoid arthritis. Clin Exp Immunol99:338-344, 1995

35. Krieg AM, Gourley MF, Klinman DM, Perl A, Steinberg AD: Heterogeneous expression and coordinate regulation of endoge- nous retroviral sequences in human peripheral blood mononuclear cells. AIDS Res Hum Retroviruses 8:1991-1998, 1992

36. Franklin GC, Chretien S, Hanson IM, Rochefort H, May FEB, Westley BR: Expression of human sequences related to those of mouse mammary tumor virus. J Virol 62:1203-1210, 1988

37. Medstrand P, Blomberg J: Characterization of novel reverse transcriptase encoding human endogenous retroviral sequences similar to type A and type B retroviruses: differential transcription in normal human tissues. J Virol 67:6778-6787, 1993

38. Leib-Mosch C, Bachmann M, Brack-Werner R, Werner T, Erfle V, Hehlmann R: Expression and biological significance of human endogenous retroviral sequences. Leukemia 6 (suppl 3):72-75, 1992

39. Harris ED: Etiology and pathogenesis of rheumatoid arthritis. In, Textbook of Rheumatology. Vol. 1. Edited by WN Kelley, ED Harris Jr, S Ruddy, CB Sledge. Philadelphia, WB Saunders, 1994

40. Krieg AM, Khan AS, Steinberg AD: Multiple endogenous xeno- tropic and mink cell focus-forming murine leukemia virus-related transcripts are induced by polyclonal immune activators. J Virol 62:3545-3550, 1988

41. Seifarth W, Skladny H, Krieg-Schneidcr F, Reichert A, Hehlmann R, Leib-Mosch C: Retrovirus-like particles released from the human breast cancer cell line T47-D display type B- and C-related endogenous retroviral sequences. J Virol 69:6408-6416, 1995

42. Feuchter AE, Mager DL: SV40 large T antigen trans-activates the long terminal repeats of a large family of human endogenous retrovirus-like sequences. Virology 187:242-250, 1992

43. Brodsky I, Foley B, Haines D, Johnston J, Cuddy K, Gillespie D: Expression of HERV-K proviruses in human leukocytes. Blood

44. Boller K, Konig H, Sauter M, Mueller-Lantzsch N, Lower R, Lower J, Kurth R: Evidence that HERV-K is the endogenous retrovirus sequence that codes for the human teratocarcinoma- derived retrovirus HTDV. Virology 196:349-353, 1993

45. Perron H, Suh M, Lalande B, Gratacap B, Laurent A, Stoebner P, Seigneurin JM: Herpes simplex virus ICPO and ICP4 immediate early proteins strongly enhance expression of a retrovirus har- boured by a leptomeningeal cell line from a patient with multiple sclerosis. J Gen Virol 74:65-72, 1993

46. Roberts CR, Herrmann M, Aicher WK, Miiller-Ladner U, Matsu- moto S, Gay RE, Kalden JR, Gay S: Sequences derived from PCR amplification of virus-like particles present in synovial fluid of patients with rheumatoid arthritis (abstract). Arthritis Rheum 38 (suppl 9):S217, 1995

47. Hirose Y, Takamatsu M, Harada F: Presence of env gene in members of the RTVL-H family of human endogenous retrovirus- like elements. Virology 192:52-61, 1993

48. Mathes LE, Olsen RG, Hebebrand LC, Hoover EA, Schaller JP: Abrogation of lymphocyte blastogenesis by a feline leukemia virus protein. Nature 274:687-689, 1978

81:2369-2374, 1993