Ror{gamma} (Rorc) Is a Common Integration Site in Type B Leukemogenic Virus-Induced T-Cell Lymphomas

The retrovirus type B leukemogenic virus (TBLV) causes T-celllymphomas in mice. We have identified the Ror ${gamma}$ locus as an integrationsite in 19% of TBLV-induced tumors. Overexpression of one ormore Ror ${gamma}$ isoforms in >77% of the tumors tested may complementapoptotic effects of c-myc overexpression.

INTRODUCTION

Top

Introduction

Type B leukemogenic virus (TBLV) is a retrovirus that is morethan 98% identical to mouse mammary tumor virus (MMTV) (1, 7).Differences between MMTV and TBLV include a 440-bp deletionof U3 sequences present within the MMTV long terminal repeat(LTR). This deletion removes negative regulatory elements thatinhibit viral transcription in many cell types, including lymphoidcells. LTR sequences flanking the deletion also are triplicatedin the TBLV U3 region to form a T-cell-specific enhancer (24).Our previous results have shown that cis-acting sequences fromthe TBLV LTR are sufficient to convert the disease tropism ofan infectious MMTV provirus from relatively long latency mammarytumors (6 to 9 months) to rapidly appearing T-cell lymphomas(2 to 3 months) (2).

Retroviruses that lack encoded oncogenes appear to induce cancerby insertional mutagenesis, leading to deregulation of nearbygenes. Because retroviral integration is relatively random,identification of viral insertions within or near the same genesin different tumors suggests that there has been selection foroutgrowth of cells carrying specific insertions. Such commonintegration sites (CISs) have been used as molecular tags toidentify oncogenes, tumor suppressor genes, and oncogenic pathways(5, 12, 17, 19, 25, 31). There are at least nine MMTV CISs,which generally fall into three categories (Wnt, Fgf, and Notchfamily genes [4, 16, 20, 21, 33]), whereas only two CISs, Tblvi1and c-myc, have been described for TBLV. The Tblvi1 CIS wasidentified in 20% of 55 TBLV-induced T-cell lymphomas examined(26) and mapped to the mouse X chromosome, but the target gene(s)remains unknown. We have detected integrations within or nearthe c-myc locus in 23% of TBLV-induced tumors (references 3and 28 and data not shown). However, unlike many other murineretroviral studies, our previous analysis of 35 TBLV-inducedtumors revealed only two tumors with detectable c-myc arrangementby Southern analysis, while PCR analysis confirmed that thosetwo tumors plus nine others had TBLV integrations near or withinthis locus (3). Surprisingly, one tumor (T623B) had at leastseven TBLV insertions at four sites within or near the c-myclocus. These studies suggested that TBLV-induced lymphomas arepolyclonal and that many of the integrations could not be detectedby Southern blotting due to the presence of multiple tumor cellclones.

Ror ${gamma}$ is a common TBLV integration site.

Top

Ror{gamma} is a common...

Using PCR analysis, we identified proviral insertions withinthe Ror ${gamma}$ (Rorc) locus, which encodes at least two protein isoforms.Ror ${gamma}$ and its thymus-specific isoform, Ror ${gamma}$ t, are members of thenuclear hormone receptor superfamily that includes ligand-regulatedtranscription factors and receptors for which a specific ligandhas not been identified (29). Ror ${gamma}$ also is known as RORC, RZR,thymus orphan receptor, and nuclear receptor 1F3 (22, 23, 27).Ror ${gamma}$ and Ror ${gamma}$ t are highly related proteins that use distinct promotersand differ only at their amino termini (11, 13, 34). Ror ${gamma}$ -knockoutmice, which lose expression of both RNA isoforms, lack lymphnodes and Peyer's patches, demonstrating a requirement for Ror ${gamma}$ / ${gamma}$ tin lymphoid organogenesis (10, 14, 30). These mice also showa 75% reduction of total T cells and greatly reduced expressionof the antiapoptotic gene Bcl-XL (14, 30). Exogenous expressionof either isoform in T-cell hybridomas inhibited interleukin-2and Fas ligand expression and blocked T-cell receptor-inducedcell proliferation and apoptosis (11, 18). Ror ${gamma}$ -null mice alsohave an increased rate of apoptosis and proliferation, resultingin rapid T-cell lymphoma formation (32). A recent report ofRor ${gamma}$ t-deficient mice has ascribed most of the gross anomaliesto the thymus-specific form (9).

In this study we screened a panel of 47 TBLV-induced tumorsby PCR analysis using 26 primer pairs consisting of a forwardor reverse TBLV-LTR primer and a Ror ${gamma}$ genomic primer (combinationsof those given in Table 1) as previously described (3). We detectedTBLV integrations within or near the Ror ${gamma}$ locus for 9 of the47 tumors tested (19%) (Fig. 1). Several of these tumors (T7,T9, T12, T623B, and T670) contained multiple TBLV integrationsin the same tumor but in different locations. Insertions occurrednear the beginning or end of the gene, consistent with an enhancerinsertion mechanism. No integrations interrupting the codingregions were detected since the two integrations in exon 11,T670 and T7, are located in the 3' untranslated region (UTR).The most common clustering of integrations occurred within intron2 of Ror ${gamma}$ , with six TBLV insertions detected in five differenttumors. TBLV integrations also were clustered near the 3' endof the Ror ${gamma}$ locus, which may alter promoter activity (36) ortranscript stability (insertions in the 3' UTR). None of theintegrations localized to the Ror ${gamma}$ locus by PCR could be detectedby Southern analysis with either a 4.3-kb genomic probe spanningRor ${gamma}$ exons 1 and 2, intron 1, and part of intron 2 through Ror ${gamma}$ texon 1(t) or a 3.9-kb probe including the 3' UTR of Ror ${gamma}$ / ${gamma}$ t exon11 and downstream region, suggesting that only a portion ofthe tumor population contained the TBLV integration. The majorityof the TBLV-induced tumors appeared to be at least semiclonalas judged by Southern blotting with T-cell receptor ßand ${gamma}$ probes (data not shown).

View this table:
[in this window]
[in a new window]
TABLE 1. Primers used to identify TBLV integrations within the Ror ${gamma}$ / ${gamma}$ t locus

View larger version (12K):
[in this window]
[in a new window]
FIG. 1. Location of TBLV insertions within the Ror ${gamma}$ locus in T-cell lymphomas as detected by PCR. Black arrows represent the location and orientation of TBLV proviruses. The tumors (designated T) containing the integrations are indicated closest to the arrow. Black bars represent Ror ${gamma}$ exons; 5' and 3' UTRs are indicated by hatched boxes.

The average size of the PCR products obtained (ca. 5 kb) maylimit detection of all integrations within the Ror ${gamma}$ locus. Weattempted to overcome such limitations by using sufficient primersets to completely scan the locus (Table 1). Nevertheless, 5kb represents the approximate region screened on either endof the gene, whereas retroviral insertions have been shown toaffect gene expression at distances over 200 kb (15). Furthermore,Ror ${gamma}$ locus intronic sequences include many single nucleotideruns and Alu repeats, which may have further diminished PCRproduct detection.

The Ror ${gamma}$ locus was recently identified as a Moloney murine leukemiavirus (MuLV) CIS in studies identifying p27^Kip1 collaboratingoncogenes (12). Complementarity between these two genes wassuggested by Winoto and Littman (35) and indicates the utilityof using numerous genetic and viral models to examine oncogenicpathways since other large-scale retroviral tagging studiesusing MuLV failed to detect the Ror ${gamma}$ locus as a CIS (19, 25).Furthermore, 8.5% of the TBLV-induced tumors showed integrationsin both c-myc and Ror ${gamma}$ (Table 2).

View this table:
[in this window]
[in a new window]
TABLE 2. TBLV tumors with integrations at multiple loci

Ror ${gamma}$ / ${gamma}$ t expression in developing thymocytes is tightly controlledand is necessary for T-cell maturation (10, 13). Two recentstudies using retroviral tagging identified the locus Sox4,encoding a transcription factor involved in B- and T-cell development,as a CIS for Moloney MuLV (25, 31). Although Sox4 was the mostfrequently targeted CIS in the study by Suzuki et al. (31) (55of 194 tumors), we were unable to detect any TBLV integrationsnear Sox4. As previously suggested (8), these studies indicatethat the unique TBLV enhancer is likely to identify additionalcellular genes that are not identified by MuLV insertional mutagenesis.

Ror ${gamma}$ and Ror ${gamma}$ t are overexpressed in TBLV-induced tumors.

Top

Ror{gamma} and ror{gamma}t are...

Ror ${gamma}$ , Ror ${gamma}$ t, c-myc and Gapdh mRNA levels in the TBLV-induced lymphomaswere analyzed and compared to those from normal thymus (Fig.2). Quantitative real-time reverse transcription-PCR (RT-PCR)was performed using an ABI PRISM 7700 sequence detection systemand SYBR Green Universal Master Mix according to the instructionsof the manufacturer (Applied Biosystems). (Primers are shownin Table 3.) Ror ${gamma}$ levels varied from 0.2- to 38-fold that observedin normal thymus, and 7 of the 18 tumors tested ( $~$ 39%) showedat least twofold overexpression (and significant differencesat the 95% confidence level by Student t tests). In contrastto previous reports (11), we routinely detected little or noRor ${gamma}$ expression in the thymus, yet expression was abundant inthe liver (33-fold higher than that of normal thymus). TBLVintegrations were identified in two of the tumors showing Ror ${gamma}$ overexpression (T12 and T14). Seven different integration siteswere detected in T12, four upstream of Ror ${gamma}$ exon 1 and one inintron 2 in the same transcriptional orientation, and one eachwithin introns 1 and 10, both in the reverse orientation. Ror ${gamma}$ texpression levels in the tumors tested ranged from 0.2- to 52-foldthat detected in normal thymus. Nine of the 18 lymphomas tested( $~$ 50%) showed at least twofold Ror ${gamma}$ t overexpression, which wassignificantly different from levels in normal thymus; together,more than 77% of the TBLV-induced tumors overexpressed one ormore Ror ${gamma}$ isoforms. We have not yet detected TBLV integrationsnear the Ror ${gamma}$ locus in any of the tumors showing Ror ${gamma}$ t overexpression.However, proviral insertions may occur at more distal locationsthan those that were examined in this study, or Ror ${gamma}$ t may beindirectly activated.

View larger version (24K):
[in this window]
[in a new window]
FIG. 2. Both c-myc and Ror ${gamma}$ / ${gamma}$ t are overexpressed in the majority of TBLV-induced tumors. c-myc (black bars), Ror ${gamma}$ t (gray bars), and Ror ${gamma}$ (white bars) expression levels from real-time RT-PCR analysis are shown relative to that from normal thymus. The standard errors for gene expression levels greater than 20-fold that of normal thymus are as follows: T13, Ror ${gamma}$ , 38 ± 0.4; T623B, c-myc, 27 ± 0.1; T703, c-myc, 22 ± 0.1; T708, c-myc, 21 ± 0.4; T708, Ror ${gamma}$ t, 52 ± 0.2; T709, c-myc, 21 ± 0.3; C3H liver, Ror ${gamma}$ , 33 ± 0.02. Gene expression experiments were performed in triplicate three to five times depending on the availability of tumor RNA. Real-time RT-PCR primers were used at a final concentration of 0.1 to 0.2 µM and had been previously determined to have similar amplification efficiencies (slopes of <0.1).

View this table:
[in this window]
[in a new window]
TABLE 3. Primers used for real-time RT-PCR analysis

Consistent with our previous analysis (28), c-myc overexpressionwas observed in 89% (16 of 18) of the TBLV-induced tumors tested.TBLV integrations have been detected in four of the primarytumors (T9, T10, T623B, and T700) with c-myc overexpression(Fig. 2 and data not shown). Elevated c-myc RNA levels alsowere detected in T16 and T17 cells passaged in syngeneic mice.Unfortunately, primary tumor RNA was available from only twoof the four tumors that contain TBLV insertions at both c-mycand Ror ${gamma}$ loci. Of these two tumors, T9 showed ca. fourfold c-mycoverexpression and did not show Ror ${gamma}$ or Ror ${gamma}$ t overexpression;T623B showed high levels of c-myc overexpression (27-fold) andseven detected integrations and modest (ca. twofold) Ror ${gamma}$ overexpression(with four detected integrations). We have previously demonstratedthat the proviral location and orientation relative to c-mycand the composition of the enhancer within the TBLV LTR allaffect target gene expression (3).

The observations that many tumors overexpressed both c-myc andRor ${gamma}$ / ${gamma}$ t and that $~$ 9% of tumors had detectable insertions in bothgenes suggest that these transcription factors are importantindividually in the progression toward disease and may collaborateduring T-cell lymphomagenesis. The idea that the antiapoptoticfactors Ror ${gamma}$ and Ror ${gamma}$ t (14, 30, 32) may antagonize the known proapoptoticeffects of c-Myc (6) is being explored.

ACKNOWLEDGMENTS

We thank members of the Dudley lab, particularly Jenny Mertz,for helpful suggestions.

This work was supported by NIH grant P01 77760.

FOOTNOTES

* Corresponding author. Mailing address: Section of Microbiology and Molecular Genetics, One University Station A5000, 24th Street and Speedway, ESB 226, Austin, TX 78712-0162. Phone: (512) 471-8415. Fax: (512) 471-7088. E-mail: jdudley{at}uts.cc.utexas.edu.

REFERENCES

Top