These are genes that control other genes.  Networks of such genes can invoke development by a cascade effect, sending waves of new biochemistry through cells and tissues.  When researchers have looked for new genes that become active in senescence, regulatory genes are often well represented.

Here’s an example from the laboratory of Stefan Jansson at the University of Umeå in Sweden.  This group has been studying the timetable of events during autumnal senescence of the same field-grown aspen tree over several years.

Autumnal senescence in aspen is synchronised, with all the leaves on the tree senescing in a coordinated manner.  Gene chips were used to measure the activity of regulatory genes (transcription factors) in leaves at different times from the onset to the completion of yellowing.

The graph shows relative levels of the mRNAs coded by six different transcription factors in aspen leaves sampled between 17 August and 21 September (data from Andersson et al. 2004)

Patterns of expression estimated this way suggest a cause and effect relationship with senescence, though to confirm causation requires further challenging experimentation.  So far there have been surprisingly few published reports to show that expression of any such gene is absolutely required for senescence to occur.

Read more about aspen senescence in the following paper:

Andersson A, Keskitalo J, Sjödin A, Bhalerao R, Sterky F, Wissel K, Tandre K, Aspeborg H, Moyle R, Ohmiya Y, Bhalerao R, Brunner AM, Gustafsson P, Karlsson J, Lundeberg J, Nilsson O, Sandberg G, Strauss SH, Sundberg B, Uhlen M, Jansson S, Nilsson P (2004) A transcriptional timetable of autumn senescence. Genome Biology 5: R2.

The group of Vicky Buchanan-Wollaston at the University of Warwick is another leader in research on genes that regulate senescence.