hlllllllllO-

Immunogenetics 36: 3-14, 1992

genetics

© Springer-Verlag 1992

Original articles Activation of Lyt-2 associated with distant upstream insertion of an SL3-3 provirus Donald S. Anson, Kristie Clarkin, and Robert Hyman Department of Cancer Biology, The Salk Institute, P.O. Box 85800, San Diego, CA 92186-5800, USA Received September 2, 1991; revised version received September 16, 1991

Abstract. Two Lyt-2 + mutants of the T-cell lymphoma SL12.4.10 were selected by fluorescence activated cell sorting. Both mutants expressed Lyt-2 (CD8 c~-chain) but not Lyt-3 (CD8/3-chain). Derivatives of one Lyt-2 ÷ mutant that expressed Lyt-3 could be isolated by sorting for Lyt-3 + cells. Southern blotting analysis indicated that both mutants had structural rearrangements within or immediately 3' of the Lyt-3 gene, accompanied by demethylation of at least one Hpa II site within the Lyt-2 gene. Gene cloning analysis of one mutant demonstrated that the structural rearrangement was due to insertion of an SL3-3 provirus 35 kb 5' to the Lyt-2 gene. It is likely that Lyt-2 gene activation is a direct or indirect consequence of proviral insertion at this site.

Introduction The activation of cellular genes by the integration of a provirus adjacent to the gene in question is now well documented (Varmus 1984; Nusse 1986; Keshet et al. 1991). The integration event can give rise to transcription of the cellular gene either by the provirus acting as a promoter element or as a transcriptional enhancer for the cellular gene's own promoter. In the latter case, the provirus may be found either upstream or downstream of the transcribed sequence. In most cases, the provirus is found within a few kilobases (kb) of the promoter. In some instances, however, the site of proviral integration may be 17-270 kb from the activated cellular gene (Corcoran et al. 1984; Peters et al. 1989; Lazo et al. 1990). This is of interest with respect to the normal control of transcription via enhancer elements. Generally, enhancers act in concert with the closest promoter, although the enhancer/ promoter orientation and distance are variable (Serfling 1985; Ptashne 1988). However, the activation of a par-

Address correspondence and offprint requests to: R. Hyman.

ticular promoter by an enhancer may depend on additional specific regulatory elements (Choi and Engel 1988). Endogenous proviruses are dispersed throughout the mouse genome (Frankel et al. 1989). Some are integrated near genes coding for cell surface molecules (Meruelo et al. 1983; Wejman et al. 1984; Frankel et al. 1989). The significance of proviral integration for expression of these genes in vivo is uncertain. Examples of the alteration of gene expression caused by retroviral integration in vivo are known (Stoye et al. 1988; Keshet et al. 1991). Spontaneous or induced T-cell lymphomas show complex patterns of proviral integrations (Canaani and Aaronson 1979; Quint et al. 1981; van der Putten et al. 1981; Herr and Gilbert 1983; Mucenski et al. 1988; Ihle et al. 1989). It is thought that the activation or repression of cellular genes may be one consequence of these proviral integrations (Cooper and Lane 1984; Nusse 1986; Bear et al. 1989; van Lohuizen et al. 1989). Here, we report the isolation of somatic cell mutants of a mouse T-cell lymphoma that have activated expression of the gene coding for the cell surface molecule Lyt-2 (CD8 c~-chain). We present evidence that the activation event is associated with a proviral insertion 35 kb 5' to the Lyt-2 gene and with the demethylation of one or more sites within the Lyt-2 gene.

Materials and methods Cell lines. SL12.4.10 is an Lyt-2- clone of the tissue culture adapted AKR/J lymphomaSL12.4 which was obtained from a spontaneoustumor of an AKR/J mouse (MacLeod et al. 1984). Lyt-2+ mutants of SL12.4.10 were obtainedby fluorescenceactivated cell sorting (FACS) for Lyt-2+ cells (Hyman et al. 1982). Mutant 1 was isolated after mutagenesis with ethylmethane sulfonate. Cells (5 × 106) were treated with 0.7 mg/ml ethylmethanesulfonatefor 18 h, grown for 7 days and then sorted for the most fluorescent 0.3% of cells. These cells were grown and the sortingprocedure repeated. A detectable Lyt-2+ population was seen when the cells were sorted a third time. An Lyt-2+ clone was isolated by limit dilution cloning. Mutant 2 was isolated by an

4 analogous sorting protocol except that no mutagenesis was used. A detectable Lyt-2 + population was again seen at the third sorting cycle, and an Lyt-2 + clone was isolated by limit dilution cloning. Lyt-2revertant clones were isolated by antibody-complement mediated selection (Hyman et al. 1980) using an Lyt-2-speciflc monoclonal antibody (mAb), followed by limit dilution cloning. Individual revertant clones were isolated from replicate selections on mutant 1. Lyt-3 + derivatives of mutant 1 were isolated by fluorescence activated cell sorting (Hyman et al. 1982) for a subline into which the Lyt 2b gene had been introduced. A 5.5 kb Hind III fragment containing the Lyt 2b gene (Nakauchi et al. 1987b) was ligated into the Hind IU site of pSV2-gpt (Mulligan and Berg 1980) and this construct was inroduced into mutant 1 by lipofection (Felgner and Holm 1989). Transfectants were isolated by growth in selective medium (Mulligan and Berg 1980) and were screened for expression of Lyt 2.2 by flow cytometry. An Lyt2.2 + transfectant (TFX 20) was used to isolate Lyt-3 derivatives by FACS. Detectable positive ceils (13 %) were seen after two sorting cycles. This Lyt-3. i + positive population was resorted and then cloned by limit dilution. Somatic cell hybrids were obtained by polyethylene glycol-mediated cell fusion as described by Hyman and co-workers (1980). DNA content/cell was determined as described by Crissman and Steinkamp (1982). The origin and properties of the other cell lines used are as described by Hyman and co-workers (1980). All cell lines are available to interested investigators for noncommercial use upon request.

Flow cytometry. Flow cytometry was performed as described by Lesley and co-workers (1984). Cells were stained by indirect immunofluorescence (IDIF) using the appropriate hybridoma supernatant as the first reagent and fluorescein-conjugated goat antibody specific for rat or mouse immunoglobulin as the second reagent.

Molecular biological techniques. Cytoplasmic RNA was isolated as described by Hyman and Cunningham (1986) and analyzed on formaldehyde gels as described by Sambrook and co-workers (1989). Ethidium bromide was added to the RNA samples to a final concentration of 0.05 gg/~tl before denaturing and loading on 1.0% agarose gels. Cellular DNA was isolated as described by Little (1987). Digestion with restriction endonucleases, gel electrophoresis, transfer to nitrocellulose membranes, hybridization, and washing were carried out as described by Evans and co-workers (1987). Genomic libraries were prepared and analyzed as described by Anson and co-workers (1984). For DNA sequencing, restriction fragments were cloned into m13 mpl8 or 19 (Messing 1983) and sequenced using a sequenase kit (US Biochemicals, Cleveland, OH) according to the manufacturer's recommendations. The Eco RI-Eco RI fragment of the cDNA clone pLy2C-1 (Nakauchi et al. 1985) was used as a probe for the Lyt-2 gene. For the experirfient described in Figure 9, an approximately 450 base pair (bp) Hind III-Nco I fragment from the clone pBSM 13. Lyt- 15.1 (provided by G. Evans, The Salk Institute, San Diego) was used as a probe for the 5' region of the Lyt-2 gene. This probe spans a region extending approximately 450 bp upstream from the first initiation codon (Fig. 9A). The 3' Eco RI-Eco RI fragment of the cDNA clone pLyC23 (Nakauchi et al. 1987a) was used as a probe for sequences at the 3' end of the Lyt-3 gene. A 300 bp Bam HI-Sal I fragment from the Lyt-3 proximal portion of the genomic clone XKB0.45.12.1 (D. S. Anson, K. Clarkin, R. Hyman, unpublished data) was used as a probe for the 4.4 kb Barn HI fragment immediately 3' of the 16 kb Bam HI fragment encompassing the Lyt-3 gene and is termed probe 2.

Determination ofgene copy number. Gene copy number was determined as described in Nelson and co-workers (1989). Autoradiograms from blots hybridized with an Lyt-2 probe were scanned using an LKB Ultroscan XL Enhanced Laser Densitometer (LKB Instruments, Bromma, Sweden). To standardize the amount of DNA/lane, the blots were rehybridized to a probe detecting the Thy-1 gene. Since all cell lines should contain equal copy numbers of the Thy-1 gene, which is present

D.S. Anson et al. : Proviral enhancement of Lyt-2 expression on a different chromosome than the gene coding for Lyt-2, the relative intensities of the Thy-1 bands should serve as a measure of the amount of DNA loaded/lane and as a control for relative differences between lanes in the amount of DNA transferred. Values are expressed relative to the parental cell line as 1.0.

Results Isolation o f Lyt-2 + mutants. The S L 1 2 . 4 . 1 0 parental cell line expresses neither Lyt-2 ( C D 8 a chain) nor Lyt-3 (CD8/3 chain) on its cell surface (Fig. 1) and accumulates neither Lyt-2 nor Lyt-3 m R N A ' s (Fig. 2). T w o Lyt-2 + mutants w e r e isolated by F A C S as described in Materials and methods. Mutant 1 was isolated after treatment o f S L 1 2 . 4 . 1 0 with ethylmethane-sulfonate, w h i l e mutant 2 was isolated f r o m an independent selection o f S L 1 2 . 4 . 1 0 not subjected to mutagenesis. By f l o w cytometry, both mutants e x p r e s s e d Lyt-2 antigen on their cell surface at 3 0 - 1 0 0 x b a c k g r o u n d levels (Fig. 1). Lyt-3 expression on the cell surface o f the mutants ranged f r o m undetectable to 2 x b a c k g r o u n d (Fig. 1). The l e v e l o f accumulation o f Lyt-2 and Lyt-3 m R N A as d e t e r m i n e d by northern blotting reflected the level of cell surface expression (Fig. 2). N u c l e a r run-on transcription analysis c o n f i r m e d that the level o f m R N A as d e t e r m i n e d by northern blotting reflected gene transcription (data not shown). Analysis o f the D N A content/cell o f the parental and mutant cell lines (Crissman and Steinkamp 1982) indicated that the parental cell line and mutant 1 w e r e pseudodiploid, w h i l e mutant 2 was pseudotetraploid. The mutations defined by these two cell lines might act in trans position or in cis position to the Lyt-2 gene. Somatic cell hybrids b e t w e e n mutant 1 (Lyt 2.1 +) and the B A L B / c (Lyt 2 . 2 - ) T-cell l y m p h o m a $49.1 w e r e exa m i n e d for expression of antigens coded for by the respective Lyt-2 alleles (Table 1). All hybrids maintained expression o f the Lyt 2 a allele characteristic o f mutant 1 (although at a s o m e w h a t r e d u c e d level) and no hybrid s h o w e d e x p r e s s i o n o f the Lyt 2 b allele. N o restriction f r a g m e n t length p o l y m o r p h i s m ( R F L P ) that differentiates b e t w e e n Lyt-2 alleles is k n o w n (Liaw et al. 1986; Y o u n et al. 1988a) and we did not o b s e r v e any structural p o l y m o r p h i s m s for Lyt-2 b e t w e e n A K R / J and B A L B / c m i c e w h e n a n u m b e r o f restriction endonucleases w e r e surveyed. W e w e r e able to demonstrate that all hybrids retain the closely linked Lyt-3 g e n e o f each parent (Table 1). Thus, it is v e r y unlikely that the failure o f these hybrids to express Lyt 2.2 reflects selective loss o f the Lyt 2 b gene. W e conclude, therefore, that the mutation defined by mutant 1 acts in cis position to activate Lyt-2 expression.

Gene structure o f mutant cell lines. T h e gene structure o f the Lyt-2 and Lyt-3 genes in the parental and mutant

D.S. Anson et al.: Proviral enhancement of Lyt-2 expression Lyt-2

5

Lyt-3

CD45

AKR1

PARENT

MUTANT 1 (EMS-induced) ...........

MUTANT 2 (spontaneous)

lO

lO0

1 10 100 1 RELATIVE FLUORESCENCE

Fig. 2. Lyt-2 and Lyt-3 mRNA accumulation by SL12.4.10 (P) and cloned Lyt-2 + mutant 1 (M1) and mutant 2 (M2) cell lines determined by northern blotting. AKR1 is an Lyt-2,3 + T-cell lymphoma. The ethidium bromide staining of the gels before blotting is shown below the respective blots.

10

100

Fig. 1. Flow cytometric analysis of SL12.4.10 (parent) and cloned Lyt-2 + mutant cell lines (mutant 1, mutant 2). Dotted lines indicate background fluorescence in which Dulbecco's modified Eagle's medium containing 10% horse serum was used as the first step staining reagent in place of hybridoma supernatant. Solid lines indicate specific fluorescence for the indicated antigen. AKR1 is an Lyt-2,3 + T-cell lymphoma. Staining for CD45 is shown as a positive control.

cell lines was examined by Southern blotting (Fig. 3). No detectable rearrangements in the Lyt-2 gene of either mutant were seen (Figure 3, and data not shown). However, structural rearrangements at or near the 3' border of the Lyt-3 gene were found in both mutant cell lines (Fig. 3). The structural rearrangements in the respective mutants were not identical and in each mutant only one gene copy appeared structurally altered. When a probe (probe 2) complementary to sequences immediately 3' of the Lyt-3 gene was used, structural rearrangements were detected in mutant 1, but not mutant 2. Comparison of these results with the known restriction map of the Lyt-3 gene (Youn et al. 1988b; Nakayama et al. 1989) suggested the presence of a structural rearrangement between 1 and 2.5 kb 3' of the Barn HI site at the 3' border of the Lyt-3 gene in mutant 1 and between 0.5-1 kb 5' of this site in mutant 2. No other structural rearrangements within or several kb 5' of the Lyt-3 gene were detected in either mutant (data not shown). To determine the basis of the rearrangement in mutant 1, a genomic library of this cell line was prepared in the X vector EMBL3 and screened with Lyt-2 and Lyt-3 gene probes. Several clones were isolated and analyzed by restriction mapping and Southern blotting. New probes were made from these initial clones to allow the cloning of the region between the two genes (Fig. 4a). Analysis of these clones revealed the presence of an unaltered Lyt-2/3 gene region as well as clones indicative of the rearrangement immediately 3' of the Lyt-3 gene detected by

D.S. Anson et al.: Proviral enhancement of Lyt-2 expression

6

Southern analysis. Comparison of the two sets of clones showed that the detected rearrangement was due to an insertion of 8.5 kb about 2.5 kb 3' of the Lyt-3 gene polyadenylation site (35 kb 5' of the Lyt-2 gene).

Table 1. The mutation defined by mutant 1 acts in cis position to activate

Lyt-2. Cell line*

Mutant 1 $49.1 T1M1.7 II F4530.1 F4531.1 F4532.1 F4533.2 F4535.1 F4536.1

DNA content/ Lyt-3 RFLP* cell t 7.5 kb 6.9 kb

Expression§ of Lyt 2.1

Lyt 2.2

0.49 0.51 Not done 1.01 0.94 0.96 1.01 1.00 0.98

33.1 0.9 0.8 6.4 7.9 8.3 9.3 9.6 6.4

1.1 1.0 3.9 1.0 1.0 1.1 1.0 1.1 1.0

+ + Not done + + + + + + + + + + + +

* F4530.1-F4536.1 are clones from somatic cell hybrids between mutant 1 and $49.1. Determined by staining with propidium iodide as described in Crissman and Steinkamp (1982). The sum of the two parental cell lines was set at 1.00. * Determined by Southern blot analysis of DNA digested with restriction enzyme Bgl II. The 5' and 3' Eco RI fragments of the cDNA clone pLyC23 (Nakauchi et al. 1987a) was used as a hybridization probe for the Lyt-3 gene. § Determined by flow cytometry. Values are geometric means of log fluorescence relative to the background fluorescence of each cell line which was set at 1.0. II T1M1.7 is an Lyt-2.2 + control cell line.

To determine the exact nature of the insertion, restriction fragments containing the insertion boundaries and the corresponding site of insertion in the normal allelic region were subcloned, mapped in detail (Fig. 4b), and sequenced. The sequence of the insertion (Fig. 5) shows the 4 bp duplication of genomic DNA at the site of insertion that is characteristic of retroviral integration. The sequence corresponding to each end of the insertion was used to search the GenBank database. This analysis showed that the insertion was of an MLV provirus in an opposite transcriptional orientation to the Lyt-2 and Lyt-3 genes. The complete sequence of the proviral long terminal repeat (LTR) was determined by combining sequence derived from both ends of the provirus (Figure 5, and data not shown). This sequence is identical to the sequence of the endogenous AKR retrovirus SL3-3 (Lenz et al. 1984). Outside of the proviral insertion, only one nucleotide difference between the wild type and mutant sequences was found, this being the deletion of a single guanine residue in the mutant sequence, resulting in a stretch of 11 guanines, compared with 12 in the wild type sequence (Fig. 5, nucleotides 15-26).

Reversion analysis. To examine the linkage of the activation of Lyt-2 gene expression with the proviral insertion in mutant 1, antibody-complement mediated cytotoxic selection against Lyt-2 was used to isolate a series of Lyt-2- revertant cell lines from mutant 1. In one series, revertants were isolated from replicate samples of mutant 1. In another series, revertants were isolated from independent subclones derived from the original clone of

Fig. 3. Southern blot analysis of DNA from SL12.4.10 parent (P) and cloned Lyt-2 + mutant 1 (M1) and mutant 2 (M2) cell lines. Numerals indicate the position of molecular size standards in kb. Restriction endonucleases used for digestion and probes used (see Materials and methods) are indicated on the figure.

D. S. Anson et al.: Proviral enhancement of Lyt-2 expression

7

(a) Normal

Kpn,

BmH,

0

I0

20

30

I

I

I

I

I

I

Ly't-3' f

(b) I

2

Lyt-Yboundary of Insertion

Normal sequence (site of Insertion)

Lyt-2'boundary of insertion

60

I

I

S

-"

f

BX

X

HA

II

I

II /.

K v

V X

K B

,,

,,

Insertion BX

X

H A

X

II

I

II

I

B

3

50

I

Ill I

"1 21 s

Insertion

40

K Bs

K V

BsH B

IF

X

Bs H B

I

I1(

A B Bs H K V

Accl BamHI BstEII Hlncll Kpnl EcoRV

X

Xbal

Ikb

Insertion

Fig. 4a, b. a The Bam HI and Kpn I restriction maps of th normal Lyt-2/3 region is shown with the scale, in kb, and the position of representative X clones (a-t) indicated above. The position and orientation of the Lyt-2 and Lyt-3 genes are indicated by arrows. The map of the insertion and adjacent regions is shown below with the position of representative X clones (y and z) indicated at the bottom of the figure. Subcloned Barn HI fragments used for detailed restriction analysis are numbered 1-3. b Restriction maps of the subcloned fragments indicated in a. The key to the restriction enzymes, and the scale used, are shown on the right. The sequences corresponding to the proviral insertion in snbclones 1 and 3 are identified by shaded boxes. Lyt-3'

i0 20 30 40 50 TTGAAAACCATAATGGGGGGGGGGG TGTAAGTGGGGAGGTTCAGATGGT

Normal

TTGAAAACCATAATGGGGGGGGGGGGTGTAAGTGGGGAGGTTCAGATGGT

Lyt-2'

TGGTATTTTTCCCATGCCTTGCAAAATGGCGTTACTGCAGTTAGCTGGCT

Lyt-3'

LTR (3'- 5 ' ) ~ 60 70 80 90 i00 AAATCCCTTACTTAAGTGACATGCACATGAAAGACCCCCAGGCTGGGCAG

Normal

AAATCCCTTACTTAAGTGACATGCACAAAACTCACGAACCCAGGGTGGTG

Lyt-2'

AAGCCTTATGAAGGGGTCTTTCACACAAAACTCACGAACCCAGGGTGGTG ~9----LTR (3'- 5')

Lyt- 3 '

ll0 120 130 140 150 TCAATCACTCTGAGGAGACCCTCCCAAGGATCAGCGAGACCACGATTCGG

Normal

GTGCTTGCCTGTGATCACAGCAGTCAAGAGATAGAAGAAGCAGGAGGATC

Lyt-2 '

GTGCTTGCCTGTGATCACAGCAGTCAAGAGATAGAAGAAGCAGGAGGATC

Fig. 5. The sequence of the two insertion junctions and the sequence corresponding to the site of integration in the normal allele. The orientation of the sequence is the same as Figure 4, as the provirus is orientated in the opposite direction to the Lyt-2 and Lyt-3 genes, the proviral sequence shown is antisense. Lyt-3' demarks the sequence of the Lyt-3 proximal insertion junction, Lyt-2' the sequence of the Lyt-2 proximal junction, and normal the sequence of the unaltered allele. The insertion sequences are shown in italics and the 4 bp sequence duplicated at the site of insertion is underlined.

mutant 1. Since the results o f both series w e r e similar, only the first will be presented. Southern blotting data on r e v e r t a n t clones B - F d e r i v e d f r o m mutant 1 is shown in F i g u r e 6. S o m e clones (B, F, G) both lost the structural alteration i m m e d i a t e l y 3' o f the Lyt-3 g e n e and s h o w e d a reduction in Lyt-2 g e n e copy n u m b e r (Table 2). Clones B, F, and G h a v e apparently lost a large region spanning both the structural r e a r r a n g e m e n t and the activated Lyt-2 gene. O t h e r clones (C, E) showed structural r e a r r a n g e m e n t s that i n v o l v e d one copy o f the Lyt-2 gene either a c c o m p a n i e d (clone E) or not (clone C) by loss o f the structural alteration near the Lyt-3 gene. P r e s u m a b l y , these structural alterations i n v o l v e the activated Lyt-2 g e n e and c o n v e r t it to a nonfunctional form. C l o n e D is o f particular interest. In this clone, neither the Lyt-3 nor the Lyt-2 genes had detectable structural alterations and the g e n e c o p y n u m b e r o f both genes was similar to that o f the S L 1 2 . 4 . 1 0 parent. This result is compatible either with loss o f the structural alteration by r e c o m b i n a t i o n , restoring the parental gene structure, or with deletion o f the entire c h r o m o s o m e bearing the structural r e a r r a n g e m e n t and duplication o f the n o r m a l

D.S. Anson et al.: Proviral enhancement of Lyt-2 expression

8

Fig. 6. Southern blot analysis of DNA from the SL12.4.10 parent (P), cloned Lyt-2 + mutant 1 (M), and Lyt-2 revertant clones (B-G) isolated from mutant 1. Clones B-G were isolated from replicate cytotoxic selections on independent samples of the cloned Lyt-2 + mutant 1 cell line and were cloned by limit dilution. All DNA samples were digested with Barn HI. Probe 2 was used to examine structural rearrangement within the approximately 4.4 kb region immediately 3' of the Lyt-3 gene (left-hand panel, Lyt-3).

Table 2. Lyt-2 gene copy number of revertant clones*. Lyt-2 band t

~per

Normalized* amount of DNA for indicated cell line P 1.0

M 0.9

B 0.6

1.0

0.9

0.6

Middle

~wer

C 0.6

D 0.9

E 0.6

0.9

0.6

0.5

0.6

F 0.3

G 0.2

0.3

0.2

0.5

* Determined as described in Materials and methods. t Bands refer to those shown in Figure 6, Lyt-2 panel. * Corrected for the amount of DNA/lane as determined by rehybridization with a probe recognizing the unlinked Thy-1 gene (see Materials and methods). For unknown (presumably technical) reasons, absolute values for clones F and G are somewhat lower than expected.

homologue (Campbell and Worton 1981). It is not possible to distinguish unambiguously between these two explanations in this homozygous cell line.

Can the Lyt-3 gene of the mutant be activated? It was noteworthy that the Lyt-3 gene of both mutants 1 and 2 was not expressed even though the structural rearrangement seen in both mutants was close to or within the Lyt-3 gene and 35 kb or more from the active Lyt-2 gene. It was possible that the Lyt-3 gene was defective in these mutant cell lines and could not be activated. A more interesting possibility was that steric or other factors prevented interaction of the inserted proviral element with the Lyt-3 promoter. It was also possible that the Lyt-2 and Lyt-3 promoters could only be alternatively, but not coordinately, activated.

To examine these alternatives, we used F A c s to isolate Lyt-3 + derivatives from mutant 1. Lyt-3 cannot be expressed on the cell surface in the absence of a functional Lyt-2 gene product (Blanc et al. 1988; Gorman et al. 1988; Youn et al. 1988b). Therefore, we first transfected mutant 1 (Lyt 2 a) with an Lyt 2 b gene. Use of this derivative (TFX 20) would allow isolation of Lyt-3 + cells even if the endogenous Lyt-2 gene could no longer be expressed. Lyt-3 + derivatives of TFX 20 could be isolated after sorting TFX 20 for Lyt-3 + cells, indicating that the Lyt-3 gene in mutant 1 was intact. These Lyt-3 + derivatives fell into two classes (Fig. 7). Lyt-3 + clones, such as 1 and 6, maintained expression of the endogenous Lyt 2 a gene. Other clones, such as 9, however, expressed Lyt 3.1 but no longer expressed the endogenous Lyt-2 a gene. The gene structure around the Lyt-2 and Lyt-3 genes in TFX 20 and the Lyt-3 ÷ clones derived from it was examined by Southern blotting (Fig. 8). The structure of the two Bam HI fragments spanning the Lyt-2 gene appeared normal in TFX 20 and all Lyt-3 + derivatives (Fig. 8B; differences in the intensity of bands in this blot were not reproducible in other blots). Similarly, the structure of the Lyt-3 gene appeared normal (Fig. 8C). When DNA was digested with Kpn I and probed for sequences at the 3' end of the Lyt-3 gene, the 8 kb band characteristic for the proviral insertion of mutant 1 was seen in TFX 20 and all Lyt-3 + derivatives (Fig. 8D, diagrammed in Fig. 8A). A different result was seen, however, when DNA from TFX 20 and its Lyt-3 + derivatives was digested with Barn HI and probed using "probe 2 " , complementary to genomic sequences immediately 3' of the proviral

D.S.

Anson et al.: Proviral enhancement of Lyt-2 expression Lyt 2.1

Lyt 2.2

Lyt 3.1 F

TFX20

TFX20 L~-3 + CLONE1

17"

i

w

m

,/f~.~

[ ,[ ....._ ~ ....

z

i

TFX20 Lyt-3 +

CLONE 6

TFX20 Lyt-3+ CLONE 9

.... ............ I

10

100

I

10

i 100

RELATIVE FLUORESCENCE

insertion (Fig. 8A). TFX 20 showed the 3.5 kb band characteristic for the proviral insertion of mutant 1 (compare Fig. 3), as well as the 4.4 kb band characteristic for the normal allele. In all of the Lyt-3 + clones derived from TFX 20, the strong 3.5 kb band characteristic of mutant 1 was not seen. In clones 1 and 6, a unique band of about 5 kb was seen; this band was deleted in clone 9, which showed only the 4.4 kb band characteristic of the normal allele. The 5 kb band present in clones 1 and 6 is consistent with loss of the Bam HI site at the 5' end of the 3.5 kb fragment Barn HI fragment characteristic of mutant 1. Loss of the Barn HI site may have occurred via a point mutation or small deletion, since clones 1 and 6 retain the 9.0 kb Kpn I fragment characteristic of mutant 1 (Figure 8A, and data not shown). This result suggests that a large scale rearrangement has not occurred in these two clones. Clone 9, in contrast, appears to have undergone loss of a large region immediately 3' of the inserted provirus, possibly including the 3' portion of the proviral sequences.

Mutants 1 and 2 show demethylation of a site within the Lyt-2 gene. Although no structural difference between the Lyt-2 gene of the parental SL12.4.10 cell line and mutants 1 and 2 was detected when DNA from these cell lines was digested with various restriction enzymes (Figure 3, and data not shown), a difference between parental and mutant

I

10

100

Fig. 7. Flow cytometric analysis of Lyt-3 + clones isolated from the TFX 20 derivative of mutant 1 by FACS. Dotted lines indicate background fluorescence in which Dulbecco's modified Eagle's medium containing 10% horse serum was used as the first step staining reagent in place of hybridoma supernatant. Solid lines indicate specific fluorescence for the indicated antigen.

cell lines was seen when the methylation status of the Lyt-2 gene was examined. The two Bam HI fragments spanning the Lyt-2 gene contain a number of Cfo I (Hha I) and Hpa II restriction sites that cannot be cleaved if the DNA is methylated (Carbone et al. 1988a, b; Liaw et al. 1986; Nakauchi et al. 1987b; Youn et al. 1988a). No sites that could be cleaved with either Cfo I or Hpa II were detected in either parental or mutant cells when the Bam HI fragment spanning the 3' portion of the Lyt-2 gene was examined (data not shown). A difference between the parental and mutant cells was observed, however, when methylation sensitive sites within the 8.6-8.8 kb Bam HI fragment spanning the 5' portion of the Lyt-2 gene were examined. Relevant restriction sites are shown in Figure 9A. A single Cfo I site approximately 4 kb 5' of the transcriptional start site of the Lyt-2 gene is methylated in the parent and in both mutants (Fig. 9C), although this site is demethylated in the Lyt-2,3 + cell line AKR1 (Fig. 9B). The parental cell line shows a single major band of about 5.6 kb when DNA from this cell line is digested with Barn HI + Hpa II (Fig. 9D). A minor proportion of the DNA appears fully methylated. The 3' terminus of this major fragment could be at the 3' Barn HI site or one of the four Hpa II sites located in the region of exons II and III. A 5 kb Bgl II fragment spans this region (Fig. 9A). When parental DNA was digested with Bgl II + Hpa II, no demethy!ated Hpa II sites were detected. This analysis

Fig. 9A-D. Determination of methylation sensitive sites in SL12.4.10 and Lyt-2 + mutants 1 and 2 by Southern blotting. A Diagram of relevant restriction sites mapped from data according to Youn and coworkers (1988a) and from blots shown in this figure and similar blots. Open boxes and Roman numerals indicate position of exons (Youn et al. 1988a; Nakauchi et al. 1987b). The region spanned by the probe used is indicated by the hatched box. B=Bam HI, Bg=Bgl II, C=Cfo I, H =Hpa II. B-D Numerals indicate position of molecular size standards in kb. Abbreviations for restriction endonucleases as in A. AKR1 is an Lyt-2,3 + control line.

Fig. 8A-E. Southern blot analysis of DNA from the SL12.4.10 parent (P), the TFX 20 derivative of mutant 1 (TFX 20), and Lyt-3 + clones isolated from TFX 20 (see Fig. 7). A Diagram of the region spanning and immediately 3' of the Lyt-3 gene. The top two bars diagram the normal allele and the bottom two bars the mutant allele, showing the proviral insertion. The location of sequences complementary to the probes used are shown as solid boxes. B =Bam HI, K =Kpn I. B-E Numerals indicate the position of molecular size standards in kb. Restriction endonucleases used for digestion and probes used are indicated above each panel. Mutant 1 (M) is shown in panel E.

places the 3' t e r m i n u s o f the f r a g m e n t d e t e c t e d in t h e Bam H I + Hpa II d i g e s t at the 3' Bam H I site a n d p l a c e s the d e m e t h y l a t e d Hpa II site a p p r o x i m a t e l y 4 k b 5' o f the t r a n s c r i p t i o n a l start site, close to t h e Cfo I site m a p p e d . ( T h e a n a l y s i s is n o t sufficiently p r e c i s e to d e t e r m i n e t h e e x a c t o r i e n t a t i o n o f t h e Hpa II a n d Cfo I sites.) When DNA from both mutants 1 and 2 were digested w i t h Barn H I a n d Hpa II, t w o m a j o r b a n d s o f a p p r o x i m a t e -

D. S. Anson et al. : Proviral enhancement of Lyt-2 expression ly 5.6 and 5.0 kb were seen (Fig. 9D). A minor proportion of the DNA was fully methylated. When DNA from both mutant 1 and mutant 2 was digested with Bgl II and Hpa II, a band of 2.5 kb is seen, in addition to the 5.0 kb hand seen when the DNA was digested with Bgl II alone (Fig. 9D). This observation indicates that one (or more) of the Hpa II sites in the region of exons 2 and 3 has been demethylated in the mutant cell lines. From the size of this band, it is likely that the site detected is one or the other of the two Hpa II sites spanning exon 2, but the analysis is not sufficiently precise to determine which. The relative intensity of the bands suggests that this demethylation has occurred on only a single copy of the Lyt-2 gene (note that mutant 2 is pseudotetraploid). In both parental and mutant cell lines, other Hpa II sites, which are demethylated in other Lyt 2,3 + cell lines such as AKR1, appear fully methylated (compare Figures 9B and 9D, and see Carbone et al. 1988a, b).

Discussion We have shown that activation of the Lyt-2 gene by a mechanism acting in cis position is correlated with structural rearrangements > 35 kb 5' of this gene and with demethylation of one or more sites within the Lyt-2 gene. Since both events occur in what appear to be two independent mutants, we suggest that both may contribute to Lyt-2 gene activation. One event may be a consequence of the other. In one Lyt-2 + mutant clone, the structural rearrangement has been shown to be due to the de novo insertion of an SL3-3 type provirus 35 kb 5' to the Lyt-2 gene. It is likely, although unproven, that the structural rearrangement in mutant 2 has a similar basis. The potent and T cell-specific leukemogenicity of the SL3-3 provirus is due to the nature of its enhancer activity. Comparison of SL3-3 with the closely related, but nonleukemogenic, Akv virus shows that the major determinant of leukemogenicity in SL3-3 is a single nucleotide change from the Akv sequence in the enhancer core region of the viral LTR (Lenz et al. 1984). This change results in a strong T cellspecific enhancer activity in SL3-3 (Boral et al. 1989). As the provirus is transcriptionally orientated in the opposite direction to the Lyt-2 gene and the activation is at the mRNA level (with the gene transcript appearing normal in size), we conclude that a reasonable explanation for Lyt-2 gene activation in mutant 1 is transcriptional enhancement as a consequence of viral integration. While this explanation for Lyt-2 gene activation seems the most straightforward, it is difficult to prove a direct cause and effect relationship. The fact that Lyt-2 + mutant isolates obtained both with and without prior mutagenesis occurred with a similar frequency and that each isolate contained a (nonidentical) structural rear-

11 rangement in the vicinity of Lyt-3 argues that the Lyt-2 + cells carrying these rearrangements probably pre-existed in the parental population and provides strong correlative evidence regarding the relationship between presence of a structural rearrangement near Lyt-3 and Lyt-2 gene activation. In the case of mutant 1, the structural rearrangement is due to proviral insertion at this site. As with other experimental examples of activation of distant genes by retroviruses (Corcoran et al. 1984; Peters et al. 1989; Lazo et al. 1990), it cannot be excluded that some other, undefined, event is responsible for Lyt-2 gene activation. Since most revertant clones isolated from mutant 1 that showed loss of the integrated provirus have lost or show structural alterations in that copy of the Lyt-2 gene presumably linked to the provirus (Fig. 6), we cannot conclude with certainty that loss of the provirus, per se, is sufficient for loss of Lyt-2 gene expression. One revertant clone (clone D in Figure 6) is compatible with such an explanation, but it is formally possible that this clone may have arisen by loss of the entire chromosome bearing the inserted provirus and duplication of the remaining normal homologue, rather than by mitotic recombination with loss of the integrated provirus (Campbell and Worton 1981). The isolation of similar mutants and revertants in a heterozygous cell line (Kipps and Herzenberg 1986; Nelson et al. 1989) or the use of homologous recombination to excise the viral sequences specifically would be required to prove the point definitively. How the putative proviral enhancer acts to stimulate Lyt-2 gene transcription is also not known. Activation of the Lyt-2 gene has been correlated with demethylation at several sites, some of which are within the 5' portion of the Lyt-2 gene, while others are contained in a region extending several kb 5' of the site of initiation of transcription of the Lyt-2 gene (Carbone et al. 1988a, b). The parental SL12.4.10 cell line shows demethylation at a single Hpa II site approximately 4 kb 5' of the transcriptional start site of the Lyt-2 gene but does not accumulate Lyt-2 mRNA. In both mutants, Lyt-2 mRNA accumulation is correlated with demethylation of one or more Hpa II sites within the Lyt-2 gene (Fig. 9). Other methylationsensitive sites, however, remain fully methylated in the parental cell line and in both mutants. The role of methylation in regulating gene transcription is uncertain, but in many cases, the degree ofmethylation shows an inverse correlation with the degree of gene transcription (Ceder 1988). There is evidence from transient transfection assays that demethylation is required before transcription can occur (Paroush et al. 1990). Repression of transcription may be regulated by specific nuclear factors that recognize methylated CpG residues (Boyes and Bird 1991). Demethylation occurs under the control ofcis-acting elements (Paroush et al. 1990) which themselves may be subject to regulation. There is evidence that the failure of methylated genes to be

12

transcribed depends on their integration into inactive regions of chromatin (Buschhausen et al. 1987; Antequera et al. 1990; Thompson et al. 1991), which may also indicate that protein factors determine whether or not methylated DNA is transcribed. It is possible that a specific demethylation of the Lyt-2 gene is one consequence of transcriptional enhancement due to proviral integration, although this interpretation is speculative. There are several examples oftrans activation of methylated genes (Thompson et al. 1986; Weisshaar et al. 1988; Bednarik et al. 1990), but in none of these cases was demethylation of the target sequences observed. The fact that only one gene copy shows demethylation suggests that the mechanism responsible for demethylation in these cell lines acts in cis position. There is evidence for maintenance of allele-specific methylation in cell lines (Chandler et al. 1987), but the mechanisms involved are unknown. The proviral insertion may act to alter DNA conformation such that the binding of specific" cis-acting factors to one or more methylation-sensitive sites is altered, resulting in demethylation. Whether demethylation is required for Lyt-2 gene transcription, or whether the demethylation observed is a secondary consequence of enhancement of transcription by the provirus is unknown. The most interesting feature of this putative example of gene activation by retroviral insertion/transcriptional enhancement is the position of the provirus. Not only is it 35 kb from the site of initiation of transcription of the Lyt-2 gene, but it is actually closer (18 kb in mutant 1) to the Lyt-3 gene promoter which is not transcriptionally activated in the mutant to any significant extent, as judged by mRNA content determined by northern blotting. Why the proviral enhancer does not act on the Lyt-3 promoter is not certain. Lyt-3 + derivatives of mutant 1 could be isolated, thus ruling out the trivial explanation that the Lyt-3 gene is defective in this mutant. While it is not certain that the observed activation of the Lyt-3 gene in Lyt-3 + derivatives is a consequence of proviral action, it may be significant that those clones that express both Lyt-3 and Lyt-2 show a structural rearrangement, which we presume involves proviral sequences (although there is not direct evidence for this point). Steric or other factors may prevent the simultaneous interaction of the provirus with two different promoters, Such a limitation could result from the requirement to bring promoter and enhancer elements in close proximity by the looping out of the intervening DNA (Ptashne 1988). It may be that the structural rearrangements observed in clones which have activated both the endogenous Lyt-2 and Lyt-3 genes are such that they allow interaction of proviral enhancer elements with both the Lyt-3 and the Lyt-2 promoters. Deletion of all or a portion of this sequence (as in clone 9, Figure 8E) might change the orientation of the provirus such that the Lyt-2 gene can no longer be activated. Alter-

D.S. Anson et al. : Proviral enhancement of Lyt-2 expression

natively, it is possible that sequences present in the deleted region are required for Lyt-2 gene activation. Acknowledgments. We would like to thank Glen Evans, Pauline Johnson, and Hiromitsu Nakauchi for DNA probes, and Angelica Bayardo for assistance with the analysis of the revertant clones. This work was supported by National Cancer Institute grant CA-13287 to R.H. and by core grant CA-14195. D.A. was a Special Fellow of the Leukemia Society of America.

References Anson, D. S., Choo, K.H., Rees, D . J . G . , Gianelli, F., Gould, K., Huddleston, J. A., and Brownlee, G. G.: The gene structure of antihaemophilic factor IX. EMBO J 3: 1053-1060, 1984 Antequera, F., Boyes, J., and Bird, A.: High levels of de novo methylation and altered chromatin structure at CpG islands in cell lines. Cell 62: 503-514, 1990 Bear, S., Bellacosa, A., Lazo, P., Jenkins, N., Copland, N., Hanson, L., Levan, G., and Tsichlis, P.: Provirus insertion in Tpl-1, an Ets-l-related oncogene, is associated with tumor progression in Moloney murine leukemia virus-induced rat thymic lymphoma. Proc Natl Acad Sci USA 86: 7495-7499, 1989 Bednarik, D., Cook, J., and Pitha, P.: Inactivation of the HIV LTR by DNA CpG methylation: evidence for a role in latency. EMBO J 9: 1157-1164, 1990 Blanc, D., Bron, C., Gabert, J., Letourneur, F., MacDonald, H.R., and Malissen, B.: Gene transfer of the Ly-3 chain of the mouse CD8 molecular complex: co-transfer with the Ly-2 polypeptide gene results in detectable cell surface expression of the Ly-3 antigenic determinants. Eur J Immunol 18: 613-619, 1988 Boral, A. L., Okenquist, S. A., and Lenz, J. : Identification of the SL3-3 virus enhancer core as a T-lymphoma cell specific element. J Firol 63: 76-84, 1989 Boyes, J. and Bird, A. : DNA methylation inhibits transcription indirectly via a methyl-epG binding protein. Cell 64: 1123-1134, 1991 Buschhausen, B., Wittig, M., Graessmann, M., and Graessmann, A.: Chromatin structure is required to block transcription of the methylated herpes simplex virus thymidine kinase gene. Proc Natl Acad Sci USA 84: 1177-1181, 1987 Campbell, C. and Worton, R.: Segregation of recessive phenotypes in somatic cell hybrids: Role of mitotic recombination, gene activation, and chromosome nondisjunction. Mol Cell Biol 1: 336-346, 1981 Canaani, E. and Aaronson, S.: Restriction enzyme analysis of mouse cellular type C viral DNA: Emergence of new viral sequences in spontaneous AKR/J lymphomas. Proc Natl Acad Sci USA 76: 1677-1681, 1979 Carbone, A., Marrack, P., and Kappler, J.: Remethylation at sites 5' of the murine Lyt-2 gene in association with shutdown of Lyt-2 expression. J Immunol 141: 1369-1375, 1988a Carbone, A., Marrack, P., and Kappler, J.: Demethylated CD8 gene in CD4 + T cells suggests that CD4 + cells develop from CD8 + precursors. Science 242: 1174-1176, 1988b Ceder, H. : DNA methylation and gene activity. Cell 53: 3-4, 1988 Chandler, L., Ghazi, H., Jones, P., Boukamp, P., and Fusenig, N.: Allele-specific methylation of the human c-Ha-ras-1 gene. Cell 50: 711-717, 1987 Choi, O.-R. B. and Engel, J. D.: Developmental regulation of/3-globin gene switching. Cell 55: 17-26, 1988 Cooper, G. and Lane, M 7: Cellular transforming genes and oncogenesis. Biochim Biophys Acta 738: 9-20, 1984 Corcoran, L. M., Adams, J.M., Dunn, A. R., and Cory., S.: Murine T lymphomas in which the cellular myc oncogene has been activated by retroviral insertion. Cell 37. 113-122, 1984

D.S. Anson et al.: Proviral enhancement of Lyt-2 expression Crissman, H. and Steinkamp, J. : Rapid one step staining procedures for analysis of cellular DNA and protein by single and dual laser flow cytometry. Cytometry 3: 84-90, 1982 Evans, G., Hyman, R., and Lewis, K. : A mutant lymphoma cell line with a defective Thy-1 glycoprotein gene. Immunogenetics 25: 28-34, 1987 Felgner, P. and Holm, M.: Cationic liposome mediated transfection. Focus 11: 21-25, 1989 Frankel, W., Stoye, J., Taylor, B., and Coffin, J.: Genetic analysis of endogenous murine leukemia viruses: association with two common mouse mutations and the viral restriction locus Fu-l. J Virol 63: 1763-1774, 1989 Gorman, S.D., Sun, Y.H., Zamoyska, R., and Parnes, J.R.: Molecular linkage of the Ly-3 and Ly-2 genes. J Immunol 140: 3646-3653, 1988 Herr, W. and Gilbert, W.: Somatically acquired recombinant murine leukemia proviruses in thymic leukemias of AKR/J mice. J Virol 46: 70-82, 1983 Hyman, R. and Cunningham, K. : A Thy-1- mutant defining a gene acting in trans position to regulate cell-surface Thy-1 glycoprotein expression and Thy-1 messenger RNA content, lmmunogenetics 23: 312-321, 1986 Hyman, R., Cunningham, K., and Stallings, V. : Evidence for a genetic basis for the class A Thy-1 - defect. Immunogenetics 10: 261-271, 1980 Hyman, R., Trowhridge, I., Stallings, V., and Trotter, J.: Revertant expressing a structural variant of T200 glycoprotein. Immunogenetics 15: 413-420, 1982 Ihle, J., Smith-White, B., Sisson, B., Parker, D., Blair, D., Schultz, A., Kozak, C., Lunsford, R., Askew, D., Weistein, Y., and Isfort, R. : Activation of the c-H-ras proto-oncogene by retrovirus insertion and chromosomal rearrangement in a Moloney leukemia virus-induced T-ceil leukemia. J Virol 63: 2959-2966, 1989 Keshet, E., Schiff, R., and Itin, A. : Mouse retrotransposons: a cellular reservoir of long terminal repeat elements with diverse transcriptional specificities. Adv Cancer Res 56." 215-251, 1991 Kipps, T. and Herzenberg, L. : Homologous chromosome recombination generating immunoglobulin allotype and isotype switch variants. EMBO J 5: 263-268, 1986 Lazo, P., Lee, J., and Tsichlis, P. : Long distance activation of the Myc protooncogene by provirus insertion in Mlvi-1 or Mlvi-4 in rat Tcell lympbomas. Proc Natl Acad Sci USA 87: 170-173, 1990 Lenz, J., Celander, D., Crowther, R. L., Patarca, R., Perkins, D. W., and Haseltine, W. A. : Determination of the leukaemogenicity of a murine retrovirus by sequences within the long terminal repeat. Nature 308: 467-470, 1984 Lesley, J., Hyman, R., Schulte, R., and Trotter, J.: Expression of transferrin receptor on murine hematopoietic progenitors. Cell Immunol 83: 14-25, 1984 Liaw, C., Zamoyska, R., and Parnes, J.: Structure, sequence, and polymorphism of the Lyt-2 T-cell differentiation gene. J lmmunol 137: 1037-1043, 1986 Little, P.: Choice and use of cosmid vectors. In D. M. Glover (ed.): DNA Cloning, Vol. 3, pp. 19-42, IRL Press, Oxford, 1987 MacLeod, C. L., Hays, E.F., Hyman, R., and Bourgeois, S.: A new murine model system for the in vitro development of thymoma cell heterogeneity. Cancer Res 44: 1784-1790, 1984 Meruelo, D., Rossomando, A., Offer, M., Buxbaum, J., and Pellicer, A.: Association of endogenous viral loci with genes encoding murine histocompatibility and lymphocyte differentiation antigens. Proc Natl Acad Sci USA 80: 5032-5036, 1983 Messing, J. : New m13 vectors for cloning methods. Methods Enzymol 101: 20-78, 1983 Mucenski, M., Bedigian, H., Shull, M., Copeland, N., and Jenkins, N.: Comparative molecular genetic analysis of lymphomas from six inbred mouse strains. J Virol 62: 839-846, 1988

13 Mulligan, R. and Berg, P. : Expression of a bacterial gene in mammalian cells. Science 209: 1422-1427, 1980 Nakauchi, H., Nolan, G., Hsu, C., Huang, H., Kavathas, P., and Herzenberg, L.: Molecular cloning of Lyt-2, a membrane glycoprotein marking a subset of mouse T lymphomas: Molecular homology to its counterpart, Leu-2/T8; and to immunoglobulin variable regions. Proc Natl Acad Sci USA 82: 5126-5130, 1985 Nakanchi, H., Shinkai, Y.-I., and Okumura, K.: Molecular cloning of Lyt-3, a membrane glycoprotein marking a subset of mouse T-lymphocytes: Molecular homology to immunoglobulin and T-cell receptor variable and joining regions. Proc Natl Acad Sci USA 84: 4210-4214, 1987a Nakauchi, H., Tagawa, M., Nolan, G., and Herzenberg, L.: Isolation and characterization of the gene for the murine T-cell differentiation antigen and immunoglobnlin related molecule Lyt-2. Nucleic Acids Res 15: 4337-4347, 1987b Nakayama, K.-I., Shinkai, Y.-I., Okumura, K., and Nakanchi, H.: Isolation and characterization of the mouse CD8 13-chain (Ly-3) genes. Absence of an intervening sequence between V- and J-like gene segments. J Immunol 142: 2540- 2546, 1989 Nelson, F., Frankel, W., and Rajan, T.: Mitotic recombination is responsible for the loss of heterozygosity in cultured murine cell lines. Mol Cell Biol 9: 1284-1288, 1989 Nusse, R.: The activation of cellular oncogenes by retroviral insertion. Trends Genet 2: 244-247, 1986 Paroush, Z., Keshet, I., Yisraeli, J., and Ceder, H.: Dynamics of demethylation and activation of the a-actin gene in myoblasts. Cell 63:1229-1237, 1990 Peters, G., Brooks, S., Smith, R., Placzek, M., and Dickson, C.: The mouse homolog of the hst/K-FGF gene is adjacent to int-2 and is activated by proviral insertion in some virally induced mammary tumors. Proc Natl Acad Sci USA 86: 5678-5682, 1989 Ptasbne, M.: How eukaryotic transcriptional activators work. Nature 335: 683-689, 1988 Quint, W., Quax, W., van der Putten, H., and Berns, A.: Characterization of AKR murine leukemia virus sequences in AKR substrains and structure of integrated recombinant genomes in tumor tissues. J Virol 39: 1-10, 1981 Sambrook, J., Fritsche, R., and Maniatis, T. : Molecular Cloning. A Laboratory Manual, 2nd Ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1989 Setting, E., Jasin, M., and Schaffner, W.: Enhancers and eukaryotic gene transcription. Trends Genet 1: 224-230, 1985 Stoye, J., Fenner, S., Greenoak, G., Moran, C., and Coffin, J.: Role of endogenous retroviruses as mutagens: the hairless mutation of mice. Cell 54: 383-391, 1988 Thompson, J., Granoff, A., and Willis, D. : Trans-activation of a methylated adenovirus promoter by a frog virus 3 protein. Proc Natl Acad Sci USA 83: 7688-7692, 1986 Thompson, J., Simkevich, C., Holness, M., Kang, A., and Raghow, R. : In vitro methylation of the promoter and enhancer of pro cd (I) collagen gene leads to its transcriptional inactivation. J Biol Chem 266: 2549-2556, 1991 Van der Putten, H., Quint, W., van Raaij, J., Maandag, E., Verma, I., and Berns, A.: M-MuLV-induced leukemogenesis: Integration and structure of recombinant proviruses in tumors. Cell 24: 729-739, 1981 Van Lohuizen, M., Breuer, M., and Berns, A.: N-myc is frequently activated by proviral insertion in MuLV-induced T-cell lymphomas. EMBO J 8: 133-136, 1989 Varmns, H. E.: The molecular genetics of cellular oncogenes. Annu Rev Genet 18: 553-612, 1984 Weisshaar, B., Langner, K.-D., Juttermann, R., Muller, U., Zock, C., Klimait, T., and Doerfler, W. : Reactivation of the methylationinactivated late E2A promoter of adenovirus type 2 by EIA (13S) functions. J Mol Biol 202: 255-270, 1988

14 Wejman, J., Taylor, B., Jenkins, N., and Copland, N.: Endogenous xenotropic murine leukemia virus-related sequences map to chromosomal regions encoding mouse lymphocyte antigens. J Virol 50: 237-247, 1984 Youn, H., Harris, J., and Gotflieb, P.: Nucleotide sequence analysis

D.S. Anson et al.: Proviral enhancement of Lyt-2 expression of the C.AKR Lyt-2~ gene: structural polymorphxsm in alleles encoding the Lyt 2.1 T-ceil surface alloantigen. Immunogenetics 28: 345-352, 1988a Youn, H., Harris, J., and Gottlieb, P.: Structure and expression of the L y t - f gene of C.AKR mice. Immunogenetics 28: 353-361, 1988b