Deletion of
a
GC-Rich Region Flanking
the Enhancer Element
within the
Long
Terminal
Repeat
Sequences
Alters the Disease
Specificity of Moloney Murine
Leukemia Virus
RONNIE HANECAK,1* PAUL K. PATTENGALE,2 AND HUNG FAN'
Department ofMolecular Biology and Biochemistry, University of California, Irvine, California 92717,1 and Department of Pathology, Children's Hospital, Los Angeles, California 900272
Received 29 March 1991/Accepted16July 1991
Moloneymurine leukemia virus (M-MuLV) isareplication-competent retroviruswhich induces
T-lympho-blastic lymphoma2to4 monthsafter inoculation. Enhancersequences in the U3 region of theM-MuLV long
terminalrepeat, primarily the 75-bp tandemrepeats, stronglyinfluence the disease specificity and latencyof M-MuLV. We investigated the role of GC-rich sequences downstream ofthe tandem repeats in the disease
specificityof M-MuLV. A recombinant M-MuLV lacking 23 bases ofaGC-richsequence(-174to-151),Delta
27A M-MuLV, wastested for pathogenesis in neonatal NIH Swiss mice. Delta27AM-MuLV induced disease
withalonger latency than did M-MuLV (7versus3 months) in >85% ofinoculatedmice.Moreinterestingly,
this virus showed anexpanded repertoire of hematopoietic diseases. Molecularanalyses andhistopathologic
examinations indicated that while 39% of mice inoculated with Delta 27A M-MuLV developed T-cell
lymphoblastic lymphoma typical of wild-type M-MuLV, the majority developed acute myeloid leukemia,
erythroleukemia,orB-cell lymphoma. Viral DNA correspondingtoDelta 27A M-MuLVwasdetectableinmost ofthe tumors analyzed. These findings indicate that the GC-rich region significantly influences the disease specificity and latency ofM-MuLV.
Nonacuteretroviruses donotencodeoncogenes,and they
induce disease (usually leukemias) slowly. One important mechanism in leukemogenesis due to these viruses is tran-scriptional activation of cellular proto-oncogenes by
en-hancer or promoter sequences in the viral long terminal repeats(LTRs) (3,4,25, 26, 30, 32). The tissuespecificity of theLTRenhancers strongly influences the resulting disease fortheseviruses,presumablybecauseoftherequirement for
strong enhancer action in the differentiated target cell to causeproto-oncogene activation (2, 8, 9, 19, 27, 35).
Moloney (M) murine leukemia virus (MuLV) causes
T-lymphoblastic lymphoma when inoculated into neonatal mice, with a latency of 3 to 5 months (34). Upstream regulatory elements in the U3 region of the M-MuLV LTR includeaTATA elementat -30 bp, aCCAAT motifat-70 bp,andenhancersequences(-180to -340 bp) (12, 18, 22). TheM-MuLV enhancers consistoftwocopies ofatandemly
repeated 75-bp element; in addition, a GC-rich element
immediately downstream of the tandem repeats has been identified (18). Consistent with the T-lymphoid tumors causedby M-MuLV,the M-MuLV LTR has been shownto have strong promoter activity in T-lymphoid tumor lines, although italso shows considerable transcriptional activity in other cell types (27). Experiments with LTR chimeras between M-MuLVand Friend MuLV(which induces eryth-roleukemia) have implicated the tandemrepeats asprimary
determinants of disease specificity in the M-MuLV LTR
(8, 16, 21, 31). Binding sites for a multitude of nuclear transcription factors have been identified in the M-MuLV
*Correspondingauthor.
tandem repeats, and mutations in specific binding motifs
were found to increase the latency ofT-lymphoid disease (28, 29). Inaddition,mutations inoneenhancercoreelement
alteredthe M-MuLV disease specificity somewhat (29). We previously describeda series of infectious M-MuLVs
with internal deletions in the LTRs (13). One of these
mutants, Delta 27A M-MuLV, contained a small (23 bp) deletion that specifically removed the GC-rich sequences
downstream of the tandem repeats, leaving the tandem
repeats intact. In this study, the pathogenic behavior of Delta Mo27A M-MuLV was tested. This virus showed a
striking expansion in the types of resulting leukemias, in-cluding T and B lymphoid, acute myeloid, and erythroid leukemias. These resultsimplicatetheGC-richsequencesin the T-lymphoidspecificity of the wild-type M-MuLV LTR.
MATERIALS AND METHODS
Virus stocks. The Delta 27A deletion has been described previously (13). Molecular cloning of recombinant proviral plasmids carryingalterations in both the 5' and 3' LTRs has also been previously described (23). Infectious Delta 27A
M-MuLV virus stocks were produced by transfection of
NIH 3T3 cells with acomplete Delta 27A M-MuLV
recom-binant provirus. Twenty-four-hour viral supernatants were
harvested from confluently infected producer cells. Virus
titration and determination ofrelative infectivities weredone as describedpreviously (13).
Virus inoculation. One-to two-day-old NIH Swiss mouse neonateswereinoculatedintraperitoneallywith5x 103PFU
(as determinedbyXC plaquetitration [5, 24])of Delta 27A M-MuLV stock. Animalsweremonitoredfor the appearance
ofdisease, andmoribund animals were sacrificed. Portions oftumortissueswerefixed forhistopathologicexamination, 5357
0022-538X/91/105357-07$02.00/0
CopyrightC 1991, American Society for Microbiology
on November 10, 2019 by guest
http://jvi.asm.org/
Xba Sst Sma
U3
R
U5
i N,T
/r181 -151 -80 -35 0
/'/
\/
TCCCCAGATGCGGTCCAGCCCTCAGCAGTTTCTAGA
-174 -151
+30 +1'
< 27A >
Deletion
FIG. 1. M-MuLV LTR and Delta 27A deleted LTR. The U3, R, and U5 regions of the M-MuLV LTR are shown. Nucleotides are
numberedrelativetothestartsite for M-MuLV viraltranscriptionattheU3-Rjunction. CAT and TATA homologiesareindicated. Tandemly repeated enhancersareshownas shaded boxes.TheDelta27A deletionislocated between the enhancers andtheXbal site and lies within theregion denoted byasolid black bar. Thisregionisenlargedbelow the LTRdiagramtoshowthe 23nucleotides deletedfrom theM-MuLV LTR (delta -174 to -151). Deletion mutagenesis and the cloning strategy used to generate Delta 27A M-MuLV have been described previously (13). Restriction enzymeabbreviations: Xba, XbaI; Sst, Sstl; Sma, SmaI.
and high-molecular-weight DNA was extracted from the
remainders asdescribed previously (14).
Molecular andhistopathological characterizations. Restric-tion enzyme digests, agarose gel electrophoresis in
Tris-acetate buffer, and Southern blot hybridization procedures
wereperformedaspreviously described (14). Southern blot
analysisforgene rearrangementsof the T-cellreceptorbeta
gene
(TCR,)
and immunoglobulin heavy (IgH)-chain (mu)and kappa light-chain
(IgL,)
genes was performed asde-scribed previously (14). Tumors were classified as T
lym-phoid ifthey showed
TCRP
gene rearrangements; tumors wereB lymphoid if they lackedTCR,
generearrangements but contained IgH,, rearrangements. B-lymphoid tumors were further subclassified as B lymphoid if they showed rearrangements of bothIgL,
andIgH,,
genes orpre-B ifthey showedonlyIgH,,
generearrangements.Tumors lackingany gene rearrangements were indicative of either earlylym-phoidcells, myeloid cells,orerythroid cells. Routine
autop-sies and histopathologic examinations were performed as
describedpreviously (1, 14).
Hybridization probes.The
TCR1
probe, the JH probe, and theIgL,
probe have already been described (14). Agarose gel-purified DNAfragments were labeled with [oL-32P]dCTP(>3,000 Ci/mmol; Amersham) by using the random-primer method(10). Oligonucleotides were labeledat the5' OH by using[y-32P]ATP (>5,000 Ci/mmol; Amersham) and polynu-cleotide kinase (Boehringer Mannheim).
Assays for infectious virus. Levels of circulating virus in
serumweredetermined by XCplaquetitrationasdescribed
previously (5). The extents ofinfection in splenocytes and thymocytes weredetermined byinfectious-centerassays on
NIH3T3cellsaspreviously described (7, 8). Briefly, animals
inoculated with wild-type or Delta 27A M-MuLV were
sacrificed at 4 or 6 weeks postinoculation. Single cell
sus-pensions of thymocytes and splenocytes were serially
di-luted, and 102 to 106 cells were cocultivated with NIH 3T3
monolayers (5 x 104 cells per 5-cm-diameter dish) in
Dul-becco modified Eagle medium containing 10% calf serum
and 2 pug of Polybrene per ml for 24 h. The medium (containing nonadherent cells) was then aspirated, and the
monolayers were washed twice with phosphate-buffered
saline and allowed to grow to confluency. Upon reaching confluency,the cultureswereassayedforinfectiousvirusby
using the UV/XC syncytial plaque assay (24). Briefly, in-fected NIH 3T3 cell monolayers were irradiated with UV
light and overlaid with rat XC cells. The number of XC plaques was used to calculate the concentration of
spleno-cytes orthymocytes actingasinfectious centers.
RESULTS
Pathogenicity ofDelta 27A M-MuLV. The Delta 27A LTR deletion has been described previously and is illustrated in Fig.1(13).Itconsistsofa23-bpdeletion from -174to -151 bpinthe U3regionof the M-MuLV LTR. Transient
expres-sion assays using LTR-bacterial cat gene fusions indicated that thepromoteractivityof the Delta 27A LTR in NIH 3T3 cellswas 20to 30%of thatof the wildtypeM-MuLV LTR (13). In contrast, infectious M-MuLV driven by the Delta 27A LTR showed relatively normal specific infectivity-within70% of that of wild-type M-MuLV. Thus, theGC-rich regionwasnotessential forefficient virus replicationinNIH 3T3 cells.
The pathogenicity ofDelta 27A M-MuLV was tested by
intraperitoneal inoculation of neonatal NIH Swiss mice. Animalswere monitored fortheappearance of disease, and
moribundanimalswere sacrificed. Figure 2 showsmortality
plots for Delta 27A M-MuLV and wild-type M-MuLV (iso-late 43D). All mice inoculated with wild type M-MuLV developed typical T-lymphoblastic lymphoma with a mean
latency of 3 months. Incontrast, for mice inoculated with Delta 27A M-MuLV, 87% developed tumors with a mean
latency of7 months. Therefore, while Delta27A M-MuLV had a specific infectivity similar to that of wild-type M-MuLV in tissueculture,it had noticeably reduced patho-genicity inwhole animals.
Diseasesinduced by Delta 27A M-MuLV. Tumors induced byDelta 27A M-MuLVwere diagnosed bygrosspathologic
and histopathologic examinations, molecular analyses, and examination of peripheral blood smears. In contrast to wild-type M-MuLV, which induced T lymphoma exclu-sively,Delta 27A M-MuLVinducedtumorsfromavariety of
50 -340 5
on November 10, 2019 by guest
http://jvi.asm.org/
[image:2.612.130.493.78.241.2]460
20
~~ ~ ~ ~ ~ ~ ~ ~ 3
0.
0 10 20 30 40 50 60 70
Weeks After Inoculation
FIG. 2. Pathogenicity of wild-type M-MuLV and Delta 27A
M-MuLV. Two-day-old NIH Swiss mice wereinoculated
intraper-itoneally with 105 PFU of virus (as determined by XC assay) per
animal. Symbols: A, moribund Delta 27A M-MuLV-inoculated
mice;*, moribund M-MuLV-inoculated mice. Therateof appear-anceofmoribund animals is shown.
hematopoietic lineages. The gross pathologic changes of
moribund Delta 27A M-MuLV-inoculated animals revealed
an enlarged spleen and thymus insome cases(suggestive of
T lymphoma), an enlarged spleen and lymph nodes, or an
enlargedspleen andliver.Histopathologicalanalysisof fixed tumor tissues is summarized in Table 1. Lymphoblastic lymphomas (TcellorBcellderived)werediagnosedin 58%
ofthe cases. Five of these animals demonstrated two
mor-phologicallydistinct andseparate neoplasms: lymphoblastic lymphoma and erythroleukemia (Table 1). More interest-ingly, theremaining mice developedacutemyeloid or
eryth-roid leukemia, diseases rarely observed in mice inoculated with wild-type M-MuLV.
Molecular analysis of tumor DNAs for gene
rearrange-ment were carried out to confirm the histopathological diagnoses and distinguish between T-lymphoid and B-lym-phoidtumors.Rearrangements of T-cellreceptorgenes(with or without immunoglobulin gene rearrangements) are
char-acteristic ofT-lymphoid tumors. Tumor DNAs from Delta 27AM-MuLV-infected animalsweredigested withHpaIand
analyzed by Southern blot hybridization by using a
TCR,
gene constant-region probe (15). Thirty-nine percent of the tumors showedTCR,
gene rearrangements, indicating that they were T lymphoid (Table 1). Somewhat surprisingly,several tumors showing
TCR,3
gene rearrangements werediagnosed by histopathologic examination as having eryth-roleukemia and Tlymphoma in the same animal (Table 1).
For tumors that did not show TCR rearrangements,
analyses forrearrangements ofimmunogPobulin heavy-and light-chain loci were performed. Pre-B lymphomas show
rearrangements of the
IgH,,
gene without rearrangement of light-chain loci(predominantly IgLK), while more mature Blymphomas show rearrangements of both IgH,, and IgLK
genes.TumorDNAsweredigestedwith EcoRI andanalyzed
by Southern blot hybridization withan
IgH,,
Jregion probe ordigested withBamHIplusEcoRI and analyzedonSouth-ernblotswithanIgLKprobeasdescribedpreviously(20, 33). Eightpercentof the mice(3 of36) showed characteristics of
pre-B lymphomas, two animals showed two separate
neo-plasms (pre-B lymphoma and erythroleukemia), and 2 of 36
animals contained mature B lymphomas by this analysis. Thirty percent of the tumors (11 of 36) showed no immunq-globulin loci and were thus likely nonlymphatic ornull-cell lymphomas. The molecular analyses were generally in agree-ment with the histopathology; i.e., lymphoblastic lymphq-mas byhistopathologic examination generally showed gene rearrangements,whileerythroid andmyeloidtumorsdid not (with the exceptions mentioned above). A summary of the diagnoses combininghistopathologic and molecularanalyses is includedin Table 1 aswell.
Deletion of sequences lying within the GC-rich regiop resulted indevelopment of T-cell lymphoma, erythroleuke-mia, acute myeloid disease, and occasionally B-cell lym-phoma. A comparison of the latencies of the different kinds ofleukemiainduced byDelta 27A M-MuLVis shown inFig. 3. Delta 27A M-MuLV induced T-cell lymphoma in a pro-portion of inoculated mice but with a considerably longer latency than that observed for wild-type M-MuLV. Animals diagnosed with T-cell-derived tumors became moribund 4.5 to 9.5 months postinoculation instead of the 2 to 4 months observed for wild-type M-MuLV-inoculated mice. On the average, T-lymphoid and erythroid malignancies appeared more rapidly (mean time, ca. 6 months) than did myeloil leukemias and B lymphomas (mean time, 8 to 9 months). However, there was considerable overlap of the
timp
courses of the differenttumors.
Detection of Delta 27A M-MuLV in tumors. Given the relatively long latency and wide spectrum of leukemias induced by Delta 27A M-MuLV, it was important to test whether the resulting tumors contained the inputvirus. For instance, it was possible that additional deletions or rear-rangements within the LTR were necessary to induce some or all of the diseases. Therefore, tumor DNAs were analyzed for the presence of Delta 27A M-MuLV provirus by diges-tionwith PvuII and Southern blot hybridization. Digestion of wild-type or Delta 27A M-MuLV proviral DNA with PvuII yields a 1.7-kb fragment containing sequences from the upstream LTR (Fig. 4A). Since the Delta 27A deletion comprises only 23 bp, analyses on the basis of restriction fragment size differences were impractical. Therefore, a synthetic oligonucleotide probe spanning the Delta 27A deletion (and unique to this virus) was used to screen the blots. As shown in the top part of Fig. 4B and Table 1, virtually all of the mice examined were positive for Delta 27AM-MuLV. As expected, an M-MuLV-specific oligonu-cleotide probe corresponding to the sequences deleted from the Delta 27A LTR did not hybridize to any of the tumor DNAs, as shownin the bottom partofFig.4B. These results confirmed that the resulting tumors were infected with M-MuLV drivenby the Delta 27A LTR.
Virus infection in preleukemic mice. The extent and distri-bution of virus infection for Delta 27A M-MuLV wer,e compared with those of wild-type M-MuLV to determine whether they reflected the pathogenic behaviors. The levels ofcirculating virus were measuredby XC plaqueassays op seratakenfrom miceat4and 6weeks postinoculation
(Fig.
5a). At 4 weeks, Delta 27A M-MuLV-inoculated
micp
showed somewhat lower levels of viremia (ca. 3-fold less), and the difference was more pronounced at 6 weeks (more than 10-fold less). This might reflect a decreased ability qf Delta 27A M-MuLV to propagate over long times in ap animal (either dependent or independent ofimmunologic4l responses). This, in turn, may be related to the increased disease latency of Delta 27A M-MuLV in
comparison
wit,h wild-type M-MuLV.Inlight of thefact that Delta 27A M-MuLV causedamuc,h
on November 10, 2019 by guest
http://jvi.asm.org/
[image:3.612.60.297.77.250.2]TABLE 1. Histopathologic findingsanddiagnoses
Tumorno. Age(mo) 27A Histopathologic finding(s)
TCR13(
JH' JK{I Diagnosis171-1 3.5 + Erythroleukemia G G G Sameb
171-2 8.5 + LL' R ND ND T-cell LL
171-3 8.5 + Erythroleukemia G G G Same
171-4 10.0 + Erythroleukemia G G G Same
171-5 15.0 + Acutemyeloidleukemia G G G Same
174-1 6.5 + Erythroleukemiaand LL G R G Erythroleukemiaand pre-B
LL'd
174-2 6.5 + Erythroleukemia and LL R ND ND Erythroleukemiaand T-cellLLU
174-3 7.0 + LL R ND ND T-cellLL
174-4 8.0 + Erythroleukemia G G G Same
174-5 8.5 + LL G R R B-cell LL
174-6 9.0 + LL G R R B-cell LL
174-7 12.0 + Acutemyeloid leukemia G R G Same
207-2 3.0 + Erythroleukemiaand LL R ND ND ErythroleukemiaandT-cellLLd
207-4 4.5 + LL G R G Pre-B LL
207-5 5.5 ND Erythroleukemia G G G Same
207-6 6.0 + Erythroleukemiaand LL G R G Erythroleukemiaandpre-B LL"
207-7 7.0 + LL R ND ND T-cellLL
207-8 9.5 + LL R ND ND T-cell LL
207-9 9.5 + Acutemyeloid leukemia G ND G Same
207-10 10.0 + Acutemyeloid leukemia G ND ND Same
235-1 4.5 + LL R ND ND T-cellLL
235-3 5.0 + LL R ND ND T-cellLL
235-4 6.0 + LL R ND ND T-cellLL
235-5 6.5 + Acutemyeloid leukemia G G G Same
235-6 7.0 ND LL R ND ND T-cellLL
235-7 9.0 + LL G R G Pre-B LL
235-8 9.0 + LL R ND ND T-cell LL
235-9 12.0 + Acutemyeloid leukemia G G G Same
236-1 4.0 - Erythroleukemiaand LL R ND ND Erythroleukemia and T-cell LL'
236-2 4.0 + Erythroleukemia G G G Same
236-3 5.0 + LL R ND ND T-cell LL
236-4 5.0 + Acutemyeloid leukemia G R G Same
236-6 7.0 + LL R ND ND T-cell LL
236-7 7.5 - LL G G G Progenitor
236-8 10.0 + Acutemyeloid leukemia G G G Same
236-9 12.0 + LL G R G Pre-B LL
Tumorsshowinga germ line(G)configurationof theTCR3gene wereanalyzed forrearranged(R) JH and JK genes; ND, not done.
bSameasthehistopathologic finding. LL, lymphoblasticlymphoma.
dFourtumors initiallycharacterized by histopathologic examinationaserythroidshowed
TCR,3
generearrangements. Another tumor initiallycharacterizedaserythroid showed aJH rearrangement. The initial diagnosis oferythroleukemia was based onhistopathologic findings and evidenced by the presence of erythroblasts and associatedareas oferythroid differentiation(i.e., normoblasts)ingreatlyexpandedred pulp.Genomic analyses indicatedthatthesemice also had clonallymphoidcellproliferations(two pre-B and three T). Onhistologicreexamination, however, thesefiveanimalshadexpanded white pulp areas which contained neoplastic lymphoblasts. These fiveanimals were thereforediagnosed as having two separate anddistinct neoplasms (i.e., erythroleukemiaand
lymphoblasticlymphoma).
widerspectrum of tumor types than wild-type M-MuLV, it was interesting to test whether this was reflected in the distribution of viral infection among different hematopoietic cells. It might be expected that the relative infection in non-T-lymphoid cells for Delta 27A M-MuLV would be higher than for thewild-type virus, since disease specificity andtissue tropism ofinfection are generally correlated (17). Oneconvenientapproach was to compare the relative infec-tion levels ofthymocytes (virtually purely T lymphoid) and splenocytes (B lymphoid, T lymphoid, and myeloid). Infec-tious-center assays forsplenocytes and thymocytes at 4 and 6weeks postinoculation are shown in Fig. 5b. Contrary to our expectations, Delta 27A M-MuLV actually showed a relatively lower efficiency of splenocyte versus thymocyte infection than did wild-type M-MuLV, even though the formervirus induced both non-T-lymphoid and T-lymphoid tumorswhilethe latter caused only Tlymphomas. Thus, for Delta 27A M-MuLV, organ tropism of infection did not directly correlate with the resulting tumor type. Similar results from pathogenicity studies of chimeric M-MuLVs
suggest that thymotropism alone does not determine T-cell lymphoma-inducing potential (35).
DISCUSSION
In this report, the pathogenic behavior of an M-MuLV deletion mutant lacking the GC-rich sequencesdownstream of the enhancer tandem repeats was investigated. In com-parison with wild-type M-MuLV, Delta 27A M-MuLV showed two significant changes: a moderate increase in disease latency, and a striking expansion of leukemia types. This was particularly noteworthy because in tissue culture, Delta 27A M-MuLV has a relatively normal infectivity-to-particle ratio. These experiments indicate that the GC-rich sequences contribute to the T-lymphoma specificity of the M-MuLV LTR. While the importance of the GC-rich se-quences in promoter activity in NIH 3T3 cells has been reported by Laimins et al. (18), previous experiments have generally emphasized the tandem repeats as being
on November 10, 2019 by guest
http://jvi.asm.org/
16
14
12
c
0 = 10
co
'6
a
0.
0
4
2
0
[image:5.612.56.300.80.418.2]T-Cell LL B-Cell LL AML EL
FIG. 3. Latency of specific hematopoietic diseases induced by Delta 27AM-MuLV. Five animals each showedtwo separate and
distinctneoplasms: erythroleukemia and eitherpre-Bor T-lympho-blasticlymphoma (LL). Eachoneof these animalswasplotted as
twoseparatepoints. Most of the animals showedasingle neoplasm
ofT-cell,B-cell, erythroid (EL), ormyeloid (AML) origin, and all wereplottedasseparatepoints.
ble for the tissuespecificity of the M-MuLV LTR(2, 9, 16, 21, 31).
An attractive explanation for these results could be the interaction of nuclear factors binding to different DNA elements in the M-MuLV LTR. In thisregard,Golemisetal. studied LTR chimeras between Friend MuLV (erythroleu-kemogenic) and M-MuLV (11). Theyfound that combining elements of the Friend MuLV and M-MuLV enhancer regionscould reduce thepathogenic potentialinsomecases.
In particular, a chimeric LTR containing enhancer
se-quences from M-MuLV and GC-rich sequences (or their analogs)from Friend MuLVyielded avirus that induced T
lymphomas with reduced efficiency. They suggested that interaction between cellular factors that bindtothe
enhanc-ers and those that bind to the GC-rich sequences is
neces-sary formaximal tissue specificityandactivity of the LTR. Our results obtained with Delta 27A M-MuLV are
com-pletelyconsistent with thisinterpretation. Inthecaseofthe
Delta 27ALTR, the GC-rich sequenceswere deleted. As a
result, the cognate cellular factor wouldnotbind andcould
notinteract withproteins thatbindtothe enhancers,which
may have effectivelyreduced the T-lymphoid specificity of A
_
U3
|RU5F
i1.7kb B
kb 2.3
-
20-&) t~~~~~ -1 A- 14,ni
r) N- ~1 s
27A
01gonucleotide probe
23-
20-M--MUL'Ll
01igonucleotide
probe
FIG. 4. Detection of M-MuLVs in Delta 27A M-MuLV-induced
tumors. (A) The line drawing shows the 5' LTR and the adjacent viral genomic region of M-MuLV proviral DNA. The black box
denotes the tandem repeat elements. Digestion of high-molecular-weighttumorDNAsamples(prepared fromtumorsinducedin mice after inoculation withwild-type M-MuLV orDelta 27A M-MuLV) withrestrictionendonuclease PvuII yieldsadiagnostic 1.7-kb DNA fragment. (B) Hybridization of synthetic oligomers to genomic
DNAs prepared from mice inoculated with Delta 27A M-MuLV.
GenomicDNA(5,ug)wasdigested with Pvull and electrophoresed
ina 1.0%agarosegel. Duplicatefilterswere hybridizedtoeithera
25-bp 32P-labelled oligonucleotide probe specific for Delta 27A M-MuLV(ATGGTTCTCTAGAGCTGGGGACCAT)or a27-bp oli-gonucleotide probe specific forwild-type M-MuLV(TAGAAACT GCTGAGGGCTGGACCGCATCT). Abbreviations: NS, normal
spleen; 27A, DNA prepared from NIH 3T3 cells productively infected with Delta 27A M-MuLV; 43-D, DNA prepared from
wild-type M-MuLV-infected NIH 3T3 cells;S, spleen; T, thymus; LN, lymphnode. Positions ofHindIll-digestedlambda DNA
mark-ers areshown.
the M-MuLV LTR. The relaxed specificity may have, in turn, resulted in tumorsofmultiple lineages.
Onesomewhatsurprisingaspectof this workwasthe fact
thatDelta 27A M-MuLV didnotshowareductionin in vivo
thymocyteinfection relative to splenocyte infectionin par-allel with the expanded spectrum of leukemias induced. In
fact, theopposite wasobserved (Fig. Sb). Transient assays ofaT-lymphoidcell line also didnotshowdecreasedactivity of the Delta27A LTR relative towild-type M-MuLV(12a). Apartial explanationfor thisdiscrepancy mightbe related to
the fact that hematopoietic cells (e.g., splenocytes and thymocytes) can become infectedas uncommitted
progeni-torspriorto differentiation into particular lineages. Indeed,
we have shown that extensive infection of progenitors for
varioushematopoietic lineages (including myeloidand eryth-roidcells) occurs in mice infected withwild-type M-MuLV (6).
It is possible that the Delta 27A deletion altered the proximity of the enhancers to the promoter or that novel tissue-specificenhancersequenceswere generated by
juxta-posingthe two ends of the deletion.These may have affected thebiologicalbehaviorof Delta 27A M-MuLV.DNA-protein binding studies of the Delta 27A LTR and site-directed mutations oftheGC-richsequenceswouldmoredefinitively
testtheir involvement inexpanded pathogenic potential.
A
A AA
AA A
A A
A AA
A A A
A A
AAAA
A A A AA
A A A
A
AA
A A
A AA
A
A A
on November 10, 2019 by guest
http://jvi.asm.org/
[image:5.612.313.558.81.281.2]FIG. 5. AssayforinfectiousDelta 27AM-MuLVorwild-type M-MuLV in splenocytes, thymocytes, and sera of inoculated NIH Swiss mice. (a) Titrationofserapreparedfrom inoculated miceat4
and 6 weeks postinoculation. Symbols: K, serum titers in mice inoculated with43 D(wild-type) M-MuLV;A,sera preparedfrom
mice inoculated with Delta 27A M-MuLV. (b) Infectious-center
assays of splenocytes (S) and thymocytes (T) at 4 and 6 weeks postinoculation. One-day-old mice were inoculated with 104 PFU
(asdeterminedby theXCplaque assay) ofwild-typeorDelta 27A M-MuLV.Thymocytes andsplenocyteswerepreparedasdescribed
inMaterials and Methods andplatedonNIH3T3 monolayers.The
UV-XC syncytial plaque assay was used to quantitate levels of infectious virus per 105 lymphoid cells plated. Each datumpoint
represents a single inoculated animal. Symbols: K, virus titers in splenocytesfrom 43 D(wild-type)-inoculated mice;A, virus titers in
thymocytes from 43 D-inoculated mice. *, virus titers in
spleno-cytesfrom Delta 27A-inoculated mice. A,virus titers inthymocytes fromDelta27A-inoculated mice.
foundto cause leukemias of multiple types (29). It will be particularly interesting to examine proto-oncogene activa-tions in different tumor types induced by Delta 27A M-MuLV.
A
43
D 27A4
WEEKS
K> A
S T
43D
S T
27A
Serum
43 D
27A
6
WEEKS
A 0 A
A
0
IA
A K>
S T
43 D
Delta 27A M-MuLV will be a valuable tool for studying aspects ofM-MuLV pathogenesis, particularly tissue speci-ficityof disease. Inadditiontothe well-documentedprocess
of proto-oncogeneactivation, which probablyoccurs late in
leukemogenesis, we and others have described early
virus-induced preleukemic changes (7). It is possible that the expanded diseasespectrum involves earlyorlate eventsor
both. Delta 27A M-MuLV also isunusual in that it induces rhultipletypesofleukemia underoneexperimentalprotocol.
Mostothernonacuteretroviruses induceonetypeof disease
in agivenprotocol, althoughamutantofM-MuLV contain-ihg specific mutations inanenhancercoreelement has been
ACKNOWLEDGMENTS
WethankBertSemlerfor helpful discussionsandcriticalreading of the manuscript. The excellent technical assistance of Mark Castro, John Giezentanner, and David Treppisacknowledged. We are grateful to David Spodick and Suzanne Sandmeyerfor
assis-tancewith computergraphics.
This work was supported by U.S. Public Health Servicegrant
R01-CA32455from the NationalInstitutesof Healthto H.F.
REFERENCES
1. Brightman, B.K., K. G. Chandy, R. H. Spencer, S. Gupta, P. K. Pattengale, and H. Fan. 1988. Characterization of lymphoid
tumorsinducedbyarecombinant murine retroviruscarrying the
avian v-myconcogene. J. Immunol. 141:2844-2854.
2. Chatis, P. A., C. A. Holland, J. W.Hartley, W. P. Rowe, and N. Hopkins. 1983. Roleforthe 3' endofthe genome indetermining disease specificity of Friend and Moloney murine leukemia
viruses. Proc. Natl. Acad. Sci. USA 80:4408-4411.
3. Corcoran,L.M.,J.M.Adams, A. R. Dunn, and S. Cory. 1984.
Murine T cell lymphomas inwhich the cellularmyc oncogene has been activatedbyretroviralinsertion. Cell 37:113-122. 4. Cuypers, H. T., G.Selten, W.Quint, M. Zijlstra, E. R.
Maan-dag, W. Boelens, P. Van Wezenbeek, C. Melief, and A. Berns.
1984. Murine leukemia virus inducedTcell lymphomagenesis: integration ofproviruses inadistinct chromosomal region. Cell
37:141-150.
5. Davis, B., E. Linney, and H. Fan. 1985. Suppression of leu-kaemia viruspathogenicityby polyoma virusenhancers. Nature
(London) 314:550-553.
6. Davis,B.R., B. K.Brightman, K. G. Chandy, and H. Fan. 1987.
Characterization of a preleukemic state induced by Moloney
murine leukemia virus: evidencefor two infection events during
leukemogenesis. Proc. Natl. Acad.Sci. USA84:4875-4879.
7. Davis, B. R., G. Chandy, B. K. Brightman, S. Gupta, and H. Fan. 1986. Effects ofnonleukemogenic and wild-type Moloney murine leukemia virus onlymphoidcells invivo: identification
ofa preleukemic shift in thymocyte subpopulations. J. Virol.
60:423-430.
8. desGrossiellers, L., E. Rassart, and P. Jolicoeur. 1983.
Thymo-tropism of murine leukemia virus is conferred by its long terminal repeat. Proc. Nati. Acad. Sci. USA80:4203-4207. 9. desGrosseillers, L., E. Rassart, and P. Jolicoeur. 1984. The
tandemdirectrepeats within thelong terminal repeat of murine
leukemia viruses are the primary determinant oftheir leuke-mogenic potential. J.Virol. 52:945-952.
4
WEEKS
a
10
76
10
-5 10
6
WEEKS
E
XL
0
x
1043 10
2
b
lo54
_)
(a
@a) 10
.~
o 1
axI102
, 2i
10
on November 10, 2019 by guest
http://jvi.asm.org/
[image:6.612.62.300.75.593.2]10. Feinberg, A. P., and B. Vogelstein. 1983. A technique for radiolabeling DNA restriction endonuclease fragments to high specific activity. Anal. Biochem. 132:6-13.
11. Golemis, E., Y. Li, T. N. Fredrickson, J. W. Hartley, and N. Hopkins. 1989. Distinct segments within the enhancer region collaborate to specify the type of leukemia induced by
nonde-fective Friend and Moloney viruses. J. Virol. 63:328-337. 12. Graves, B. J., P. F. Johnson, and S. L. Mcknight. 1986.
Homologous recognition of a promoter domain common to the MSV LTR and the HSV tk gene. Cell 44:565-576.
12a.Hanecak, R., and H. Fan. Unpublished data.
13. Hanecak, R., S. Mittal, B. R. Davis, and H. Fan. 1986.
Gener-ation of infectious Moloney murine leukemia viruses with
deletions in the U3 portion of the long terminal repeat. Mol. Cell. Biol.6:4634-4640.
14. Hanecak, R., P. K. Pattengale, and H. Fan. 1988. Addition or
substitution of simian virus 40 enhancer sequences into the
Moloney murine leukemiavirus(M-MuLV) longterminal repeat
yields infectiousM-MuLVwith altered biological properties. J. Virol. 62:2427-2436.
15. Hedrick, S. M., E. A. Nielson, J. Kavaler, D. l. Cohen, and M. Davis. 1984. Sequence relationships between putative T cell receptorpolypeptides andimmunoglobulins. Nature (London)
308:153-158.
16. Ishimoto, A., M.Takimoto, A. Adachi, M. Kakuyama, S. Kato, K. Kakimi, K. Fukuoka, T. Ogiu, and M. Matsuyama. 1987.
Sequences responsible for erythroid andlymphoid leukemiain thelong terminalrepeatsof Friend-mink cellfocus-formingand
Moloney murine leukemiaviruses. J. Virol.61:1861-1866. 17. Jaenisch, R. 1979.Moloneyleukemiavirus gene expression and
gene amplification in preleukemic and leukemic BALB/Mo
mice. Virology93:80-90.
18. Laimins, L. A., P. Gruss, R. Pozzatti, and G. Khoury. 1984.
Characterization of enhancer elements in the long terminal repeat ofMoloney murinesarcomavirus.J. Virol. 49:183-189. 19. Lenz, J., D. Celander, R. L. Crowther, R. Patarca, D. W.
Perkins, and W. A. Haseltine. 1984. Determination of the
leukemogenicity ofamurine retrovirusbysequenceswithin the
long terminalrepeat. Nature (London)308:467-470.
20. Lewis, S., N. Rosenberg, F. Alt, and D. Baltimore. 1982.
Con-tinuing kappa-generearrangement inacell linetransformed by Abelsonmurineleukemia virus. Cell 30:807-816.
21. Li, Y., E. Golemis, J. W.Hartley, and N. Hopkins. 1987. Disease
specificity of nondefective Friend and Moloney murine leuke-mia viruses is controlledbya smallnumberofnucleotides. J. Virol.61:693-700.
22. Linney, E., B. Davis, J.Overhauser,E.Chao,and H. Fan.1984.
Non-function ofaMoloney murine leukemia virus regulatory
sequence in F9 embryonal carcinoma cells. Nature (London) 308:470-472.
23. Overhauser, J., and H. Fan. 1985. Generation of glucocorticoid-responsive Moloney murine leukemia virus by insertion of regulatory sequences from murine mammary tumor virus into the long terminal repeat. J. Virol. 54:133-144.
24. Rowe, W. P., W. E. Pugh,andJ. W. Hartley.1970.Plaque assay techniques for murine leukemiaviruses. Virology42:1136-1139. 25. Selten, G., H. T. Cuypers, and A. Berns. 1985. Proviral activa-tion ofthe putative oncogene pim-1 in MuLV-induced T cell lymphomas. EMBO J. 4:1793-1798.
26. Selten,G., H. T. Cuypers, M. Zijlstra, C. Melief, and A. Berns. 1984. Involvement of c-myc in MuLV-induced T cell lympho-masin mice:frequency and mechanisms of activation. EMBO J. 3:3215-3222.
27. Short, M. K., S. A.Okenquist, and J. Lenz.1987.Correlation of leukemogenic potential of murine retroviruses with transcrip-tional tissue preference of the viral long terminal repeat. J. Virol. 61:1067-1072.
28. Speck, N. A., and D. Baltimore. 1987. Six distinct nuclear factors interact with the75-base-pair direct repeat of the Molo-ney murine leukemia virus enhancer. Mol. Cell. Biol. 7:1101-1110.
29. Speck, N. A., B. Renjifo,E.Golemis, T. N. Fredrickson, J. W. Hartley, andN.Hopkins. 1990.Mutation of thecore oradjacent LVb elementsoftheMoloneymurine leukemiavirus enhancer altersdisease specificity. Genes Dev. 4:233-242.
30. Steffen, D. 1984. Proviruses are adjacent to c-myc in some murineleukemia virus-inducedlymphomas. Proc. Natl. Acad. Sci. USA 81:2097-2101.
31. Thiesen, H.-J., Z. Bosze, L. Henry, and P. Charnay. 1988. A DNAelementresponsiblefor thedifferent tissue specificitiesof
Friendand Moloneyretroviral enhancers. J. Virol. 62:614-618. 32. Tsichlis, P. M., M. A. Lohse, C. Szpirer, J. Szpirer, and F. Levan. 1985. Cellular DNAregionsinvolved in the induction of ratthymiclymphomas (mlvi-1, mlvi-2,mlvi-3,andc-myc) repre-sentindependent lociasdeterminedbytheir chromosomalmap locations in therat. J.Virol. 50:938-942.
33. Weaver,D., F. Constantini,T.Imanishi-Kari,andD. Baltimore. 1985. A transgenic immunoglobulin mu gene prevents rear-rangement ofendogenous genes. Cell 42:117-127.
34. Weiss,R.,N.Teich,H.Varmus,andJ.Coffin(ed.). 1982. RNA tumor viruses. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
35. Yuen, P. H., and P. F. Szurek. 1989. The reduced virulence of the thymotropic Moloney murine leukemia virus derivative MoMuLV-TB ismappedto11mutations within the U3regionof thelongterminal repeat. J. Virol. 63:471-480.
