Every enhancer works with every promoter for all the combinations tested: could new regulatory pathways evolve by enhancer shuffling?

Abstract

The promoters and enhancers of cell type-specific genes are often conserved in evolution, and hence one might expect that a given enhancer has evolved to work best with its own promoter. While this expectation may be realized in some cases, we have not found evidence for it. A total of 27 combinations of different promoters and enhancers were tested by transfection into cultured cells. We found that the relative efficiency of the enhancers is approximately the same, irrespective of the type of promoter used, i.e., there was no strong preference for any given enhancer/promoter combination. Notably, we do not see particularly strong transcription when the immunoglobulin kappa enhancer (or the immunoglobulin heavy chain enhancer) is used to activate a kappa gene promoter. We propose that a generally permissive enhancer/promoter interaction is of evolutionary benefit for higher eukaryotes: by enhancer shuffling, genes could be easily brought under a new type of inducibility/cell type specificity.

Figure 1

Schematic representation of the recombinant DNAs. A. DNA constructs with the immunoglobulin kappa light chain promoter and its derivatives. A 202 bp fragment of a mouse k chain promoter or its derivatives (black region) was fused to a truncated rabbit β-globin gene (white region). Promoter and enhancer sequences used in these constructs are designated with the letters P and E, respectively. Plasmid DNA is shown by a wavy line. The constructions are not drawn to scale. B. DNA constructs with the SV40 promoter. The 21 bp repeats from the early promoter region of SV40 containing 5 functional Spl binding sites (stippled boxes) were placed in front of the β-globin gene (white region). The enhancers used are identical to some of those shown in A. C. DNA constructs containing the OVEC reporter gene. OVEC is a reporter gene containing the coding sequence of the rabbit β-globin gene (as are the constructs in A and B) and convenient cloning sites for promoter and enhancer elements (Westin et al., 1987). DNA oligonucleotides containing a single or a duplicated binding site for different transcription factors were inserted upstream of the β-globin TATA box (white). These promoter constructions were tested with the SV40 enhancer or a multimeric NF-κB enhancer.

Figure 2

Transcriptional activation of two genuine promoters by different enhancers. A. S1 nuclease analysis of Igκ promoter activity. Human B-type lymphoblastoid cells (BJA-B) were transfected with seven constructs including the κ-light chain promoter alone (designated “0,” lane 1) or in combination with the six enhancers shown in Figure 1A (lanes 2–7, designation according to Fig. 1A). The products of S1 nuclease protection were analyzed on a denaturing 6% polyacrylamide gel in parallel with size markers (M) of HpaII digested pBR322 DNA. The bands corresponding to the correctly initiated and the reference transcripts are indicated with “ct” and “ref,” respectively. B. Activity of the SV40 promoter. The same procedure as in A was used for transcriptional analysis of the SV40 promoter constructs depicted in B. The activation of this promoter by the SV40-, IgH (heavy)- and IgL(κ)-enhancers is shown in lanes 3, 4 and 5, respectively. “0,” SV40 promoter construct without enhancer. All other designations are as in A.

Figure 3

Transcriptional activation of the Igk promoter and its derivatives by different enhancers. The transcriptional activity of four series of constructs including the wild-type promoter (KapP) and its derivatives (SplP or Kap(-)P, see Figure 1A) in combination with the enhancers from IgL(K) (A); IgH (heavy chain) (B); SV40 (C) or MoMSV (D) were analyzed by SI nuclease mapping. The construct designations are according to Figure 1A. “0,” IgK promoter without enhancer. All other designations areas in Figure 2. Mappings shown in A and B were run on the same gel along with the same size marker lane. The bands between the “ref”- and “ct” bands are from read-through transcription of the reference plasmid. To improve presentation of the data, different autoradiogram exposures are shown for A and B, and the size marker is now shown at the sides of both A and B.

Figure 4

Activity of β-globin promoter derivatives with either of two enhancers. The level of transcription was measured from different promoters without an enhancer (A), with the SV40 enhancer (B), or with eight NF-κB sites (C). The constructs used are schematically drawn in Figure 1C. The following promoters were tested: TATA box alone (lane 1); the octamer motif (lane 2); the NF-kB binding site (lane 3); the Pu box (lane 4); the GC box, recognized by the Spl transcription factor (lane 5); and the entire β-globin promoter (lane 6). The constructs were transfected into X63Ag8 myeloma cells and the amount of β-globin mRNA quantitated by RNase mapping, ct indicates correctly initiated transcripts, and ref indicates the transcripts from OVEC-REF (Westin et al., 1987) which was included as an internal reference for transfection efficiency. HpaII digested pBR322 DNA was used as size marker (lane 7).