Involvement of Sp1 and microsatellite repressor sequences in the transcriptional control of the human CD30 gene

Abstract

CD30, as a member of the tumor necrosis factor (TNF) receptor family, is expressed on the surface of activated lymphoid cells. CD30 overexpression is a characteristic of lymphoproliferative diseases such as Hodgkin's/non-Hodgkin's lymphomas, embryonal carcinoma, and a number of Th2-associated diseases. The CD30 gene has been mapped to a region of the murine genome that is involved in susceptibility to systemic lupus erythematosus. Functionally, CD30 may play a role in the deletion of autoreactive T cells. We were interested in determining the molecular nature of CD30 overexpression. Sequence comparison has revealed significant identity between the TATA-less human and murine CD30 promoters; they share a number of common consensus binding motifs. Transfection assays identified three regions of transcriptional importance; the region between position -1.2 kb and -336 bp, containing a CCAT microsatellite sequence, a conserved Sp1 site at positions -43 to -38, and a downstream promoter element (DPE) at positions +24 to +29. EMSA and DNase I footprinting showed specific DNA-protein interactions of the CD30 promoter with the Sp1 site and the CCAT repeat region. The DPE element was shown to be essential for start site selection. We conclude that the conserved Sp1 site at -43 to -38 is associated with maximum reporter gene activity, the DPE element is required for start site selection, and the CCAT tetranucleotide repeats act to repress transcription. We also have shown that the microsatellite is multiallelic, when we screened a random healthy population. Further studies are required to determine whether microsatellite instability in the repressor predisposes susceptible individuals to CD30 overexpression.

Figure 1.

Analysis of the promoter region of the human CD30 gene. A: Expression of CD30 5′ promoter deletion constructs in transiently transfected Jurkat cells. Constructs with different extents of CD30 promoter driving the firefly luciferase gene were cotransfected with pRL-TK and incubated for 32 hours before cells were assayed for luciferase activity. The normalized relative transcriptional activity is expressed relative to the full-length −3.5-kb construct. Error bars are standard errors from duplicate experiments. Possible transcription factor binding elements defined by searching the TRANSFAC database are represented on the full-length construct. Numbers (in bp) represent the position of the 5′ deletion end points in relation to the major transcription initiation site, shown as an arrow. B: Mutations were introduced into the full-length −3.5-kb CD30/luciferase construct at the five possible transcription factor binding elements and then transfected and assayed as in A. The normalized relative transcriptional activity is expressed relative to the full-length −3.5-kb construct. Error bars are standard errors from duplicate experiments.

Figure 2.

Interaction of Jurkat cell nuclear proteins with the core CD30 promoter. A: In vitro DNase I footprints of the region −311 to +152 of the CD30 gene. The probe was incubated with BSA or nuclear extracts in the presence of increasing amounts of DNase I, represented by triangles. Three regions of protection were identified at −159 to −149, −142 to −128, and −44 to −28. The −44 to −28 footprint region contains the consensus Sp1 site shown to be important for CD30 promoter activity by mutational analysis. A Maxam and Gilbert G+A sequencing ladder (G/A) of the same region serves as a size marker. B: EMSA using a double-stranded probe representing −57 to −22 of the CD30 promoter. Major complexes are indicated (a–f). Lanes 1–7: Probe plus Jurkat nuclear extracts (NE); lane 8: free probe. Lane 1: Probe alone. Lane 2: Probe plus 50-fold molar excess of unlabeled probe (self). Lane 3: Probe plus 50-fold molar excess of double-stranded consensus (GC) oligonucleotide specific for Sp1. Lane 4: Probe plus 50-fold molar excess of double-stranded consensus (GT) oligonucleotide. Lane 5: 50-fold molar excess of nonspecific GC-rich competitor (NI). Lane 6: Probe plus nuclear extract incubated with Sp1 antibody. Lane 7: Probe plus nuclear extract incubated with Sp3 antibody.

Figure 3.

Definition of binding sites within the −57 to −22 region of the CD30 promoter. A: Double-stranded oligonucleotide EMSA probe used with regions mutated by introduction of a SacI site, indicated by the boxed regions (M1, M2, M3). B: EMSA comparing wild-type probe with M1, M2, and M3 mutated probes. Lanes 1–7: The probe indicated plus Jurkat cell nuclear extract. Lanes 1–4: The wild-type (WT) or mutated probe indicated. (Note that multiple preparations of the M1 probe showed aberrant mobility, which presumably was due to secondary structure as a result of introduction of the mutation.) Lane 5−7: Wild-type probe plus 50-fold molar excess of mutant oligonucleotide as indicated. Lanes 8–11: Labeled probe only.

Figure 4.

Comparison of the human and murine microsatellite sequences within the CD30 promoter. A: Schematic representation of the relationship between the human and murine microsatellite region of the CD30 promoter. The 400-bp imperfect tetranucleotide repeat found in the human CD30 promoter is longer and less degenerate than the 150-bp murine equivalent. The microsatellites are represented by the filled boxes. B: Comparison of the 3′ part of the murine (GenBank accession AF069504) and human (GenBank accession no. AF065475) microsatellite sequences from −401 to −581 of the human CD30 promoter. Regions of identity are shown (*). Sequences were aligned using Clustal V.

Figure 5.

The CD30 promoter microsatellite can act to repress transcription. Luciferase reporter constructs were used to transfect Jurkat cells and were assayed for luciferase activity 32 hours later. Relative activities with respect to the full-length −3.5-kb construct are shown. Error bars are standard errors from duplicate experiments. ΔMS, deletion of microsatellite except (CCAT)₄; SV, pGL3 promoter containing the SV40 minimal promoter; SV-MS, pGL3 promoter plasmid with 400-bp microsatellite inserted upstream from the SV40 promoter.

Figure 6.

The CCAT repeat sequences in the CD30 microsatellite bind nuclear factors in vitro. A: Sequence of the wild-type EMSA probe representing the core microsatellite repeat (CCAT)₅CACCTTATGCAT(CCAT)₂. Three separate 6-bp mutations were introduced into the boxed regions indicated to produce the M1, M2, and M3 EMSA probes. B: EMSA comparing wild-type probe with M1, M2, and M3 mutated probes. Lanes 1–9: Probe plus Jurkat cell nuclear extracts. Lane 1: Wild-type probe. Lane 2: M1 probe. Lane 3: M2 probe. Lane 4: M3 probe. Lane 5: Wild-type probe plus 50-fold molar excess of M1 oligonucleotide. Lane 6: Wild-type probe plus M2 competitor. Lane 7: Wild-type probe plus M3 competitor. Lane 8: Wild-type probe plus 50-fold molar excess of (CCAT)₆ oligonucleotide. Lane 9: Wild-type probe plus 100-fold molar excess of (CCAT)₆ oligonucleotide. Lane 10: Free probe.

Figure 7.

Multiple alleles are found at the CD30 promoter CCAT microsatellite. PCR was used to amplify the CD30 CCAT microsatellite sequences. The resulting products were analyzed on 6% nondenaturing PAGE gels, followed by ethidium bromide staining. A: PAGE gel analysis of PCR products from seven randomly selected individuals showing multiple alleles. B: PCR products derived from six cell lines as indicated above the gel photograph. In both A and B pGEM DNA markers were used as a molecular weight standard (M. W. Std). The sizes shown are 460 bp (top), 396 bp, and 350 bp.

Figure 8.

Mutations in the downstream promoter element effect the position of transcription initiation. A: Transient transfection of Jurkat cells with CD30/luciferase downstream mutants. Mutations were introduced into the full-length −3.5-kb CD30/luciferase construct at the downstream AP1-like and DPE sites. Constructs were transfected into Jurkat T cells and assayed for luciferase activity 32 hours later. The transfection-normalized relative luciferase activity is shown. Error bars are standard errors from duplicate experiments. B: RPA analysis of CD30/luciferase mRNA expression from Jurkat cells transfected with either the −336 construct (lane 2) or the DPE mutant construct (lane 1). Molecular size standards (W) were HinfI-digested φX174.