Info

Figure 4-8. Variation in enhancer binding patterns in a A(C) chick, a A(m) mouse, and § 1(C) chick lens crystalline genes (that make the lens protein in these species). Symbols show the location of REs for various TFs. In particular, note the variable location and number of the Pax6 sites, in these genes that are each expressed in developing lens tissue. The broken arrows indicate RNA transcription start positions. Modifed from (Cvekl and Piatigorsky 1996).

Figure 4-8. Variation in enhancer binding patterns in a A(C) chick, a A(m) mouse, and § 1(C) chick lens crystalline genes (that make the lens protein in these species). Symbols show the location of REs for various TFs. In particular, note the variable location and number of the Pax6 sites, in these genes that are each expressed in developing lens tissue. The broken arrows indicate RNA transcription start positions. Modifed from (Cvekl and Piatigorsky 1996).

coexpressed in a given cell are phylogenetically or even functionally related (Caron et al. 2001; Spellman and Rubin 2002).

Similar REs may be, but are not necessarily, involved in coregulating genes in the same cells. For example, expression of Hox, immunoglobulin, and some color vision and olfactory genes is closely tied to their linkage relationships, but the alpha and beta hemoglobin genes are coregulated in erythrocytes by essentially unrelated mechanisms (Hardison 2000; 2001). Typically, only one alpha and one beta gene from each cluster of related genes is active in a given cell, and similar statements apply to opsin, immunoglobulin, olfactory, and other linked, related genes. Sometimes several, but not all, of the linked genes are expressed in a given context but others of the genes in other contexts; this is the case with the Hox developmental genes.

In addition to the mechanisms of cis-regulation by trans-acting TFs (that is, TFs whose own gene is on a chromosome other than that of genes it is regulating), there is also evidence for some true trans-regulation, that is, by sequence elements on one strand interacting with those on another. For example, there can be direct interac-

Figure 4-9. Schematic of variation in location of similar regulatory sequences. This is an alignment of the hindbrain r3/r5 segments enhancer elements of group 2 Hox genes from mouse, chicken, pufferfish, and human. Curves connect homologous blocks (solid squares) of sequence in the regulatory domains. The binding sites for the transcription factor Krox20 (indicated by a solid triangle) and the BoxA REs (indicated by a solid rectangle) are situated at similar locations with respect to the start codon ATGs in all six fragments and joined with a thicker curve. The solid boxes (blocks 1-13) represent short stretches of at least 70 percent sequence identity dispersed along the enhancer elements of all four species. Regions of the mouse and chicken genomic DNAs between the ATG s and the enhancers that had not been sequenced are depicted with a broken line. Broken arrows are RNA transcription start positions. Reprinted from Nonchev et al., 1996, with permission. Copyright 2000 National Academy of Sciences, U.S.A. For details consult this source. See Chapter 16 on hindbrain segmentation genes.

Figure 4-9. Schematic of variation in location of similar regulatory sequences. This is an alignment of the hindbrain r3/r5 segments enhancer elements of group 2 Hox genes from mouse, chicken, pufferfish, and human. Curves connect homologous blocks (solid squares) of sequence in the regulatory domains. The binding sites for the transcription factor Krox20 (indicated by a solid triangle) and the BoxA REs (indicated by a solid rectangle) are situated at similar locations with respect to the start codon ATGs in all six fragments and joined with a thicker curve. The solid boxes (blocks 1-13) represent short stretches of at least 70 percent sequence identity dispersed along the enhancer elements of all four species. Regions of the mouse and chicken genomic DNAs between the ATG s and the enhancers that had not been sequenced are depicted with a broken line. Broken arrows are RNA transcription start positions. Reprinted from Nonchev et al., 1996, with permission. Copyright 2000 National Academy of Sciences, U.S.A. For details consult this source. See Chapter 16 on hindbrain segmentation genes.

tion of mRNAs from different homologs or genes. Direct contact-based influence of a sequence on one chromosome on a gene on another chromosome occurs regularly in at least some plants (called paramutation), and similar phenomena have been seen in at least some animals (e.g., Hollick et al. 1997). Paramutation often is associated with inhibiting gene expression. Although these phenomena are rare, we will see peculiar instances of gene regulation (e.g., in olfaction) where they may have an important application.

Evolving, But Not Treelike

One of the keystones of evolution, in the usual model, is that it generates divergence among descendants relative to their common ancestor. We have seen how mutation generates this kind of relationship among gene sequences and how gene duplication will generate a tree of divergent sequence relationships among the members of a gene family. Interestingly, this is not quite how RE sequences evolve. A given TF recognizes a variety of RE sequences that are variations on a common theme (e.g., Table 4-4). They have relationships constrained perhaps by selection related to the binding efficiency, location near a regulated gene, and so on. But because they are at least largely generated by local mutation, not duplication and translocation, there need be no orderly tree of evolutionary relationships even among the enhancer sequences used by the same gene.

Was this article helpful?

0 0

Post a comment