In the hunt for regulatory elements that control gene expression, much emphasis has been placed on the conservation of non-coding regions at the sequence level over large evolutionary distances. A recent study demands a rethink of this approach, showing that such similarity is not enough to identify some key regulators, which might only be detectable using functional assays.

The expression pattern of the RET receptor tyrosine kinase is highly conserved in zebrafish and humans. By contrast, only part of the coding sequence of the gene shows similarity between the two species at the sequence level. So, sequence comparisons between zebrafish and human provide no clue to how patterns of RET expression have been maintained during evolution.

To identify non-coding regions that regulate RET expression, Shannon Fisher and colleagues looked for sequences in the region around the gene that are conserved at shorter evolutionary distances. They identified 38 such sequences that are conserved either between zebrafish and pufferfish, or between human and other non-primate mammals.

Using a transposon-based reporter vector, the authors looked at the patterns of expression that were driven by these sequences in zebrafish embryos. Surprisingly, almost all of the non-coding sequences — whether they were derived from zebrafish or human — directed expression in cell types that were highly consistent with endogenous RET expression. So it seems that, unexpectedly, strong conservation of regulatory function does not necessarily correspond to sequence conservation.

This study also represents a technical advance: the transposon-based vector that was used overcomes the transgenic mosaicism that has limited the usefulness of previous zebrafish systems. Functional assays in the zebrafish therefore promise to provide an important tool in the search for other evolutionarily conserved regulators of gene expression.