Howard Sid Thomas | Senescence genomics

The molecular biology and omics of senescence

Senescence is programmed: it happens predictably during the plant life-cycle and is done with a purpose. The senescence programme consists of particular events turning on, others turning off and others carrying on unchanged. Each event is specified by one or more genes.

Cell biological representation of the leaf senescence program

Molecular biology gives us the tools to find out which genes are in the on or off state at any stage of development, including senescence. Although the principles of analysing patterns of gene activity are fairly simple and the technology is now pretty routine, it took a lot of Nobel Prize-winning research to get to this point.

When a gene is expressed, its DNA is transcribed into a messenger RNA (mRNA) copy. Each mRNA is then translated into a corresponding protein, each of which performs some kind of cell function such as participating as an enzyme in a metabolic reaction.

When a developmental program is activated, new genes are transcribed and proteins made while some existing genes are suppressed. We can identify the activation state of genes by looking at which mRNAs are present.

cDNA library - from http://fig.cox.miami.edu/~cmallery/150/gene/sf16x5.jpg This can be done by extracting RNA and back-copying it to DNA in a test-tube. This copy DNA (cDNA) can then be kept more or less indefinitely by introducing it to cells of a bacterium (usually E coli) where it is maintained and replicated as if it were the bacterium's own DNA. A collection of cDNAs representative of the mRNAs from a particular developmental stage and maintained in an E coli culture is called a cDNA library.

If we want to know what genes are switched when the senescence program is invoked, we make a cDNA library from tissue at the pre-senescent state and a cDNA library from senescing tissue. Then we ask which cDNAs are present in or absent from the two libraries.

In this way a picture has been built up of which genes contribute to the senescence program - SAGs (senescence-associated genes). Studies of senescence in many plant species have contributed to the compilation of a large list of SAGs. By grouping SAGs according to known biological function, the senescence program can be understood in terms of the coordinated transcription of particular gene classes.

Vicky Buchanan-Wollaston's functional classification of known SAGs

Perhaps it's not so surprising that genes encoding protein-degrading and amino acid metabolising enzymes are well represented amongst the up-regulated SAGs, though there's plenty we don't understand about the cell biology of protein recycling during senescence.

The SAGs also include an unexpectedly large group of genes for defence against pathogens. It's easy to imagine that a senescing leaf might represent a tempting point of invasion for fungal, bacterial or viral disease. Plants evidently invest heavily in activating defence SAGs to deal with this threat.

Prominent amongst the genes turned off during senescence are those encoding photosynthetic enzymes and structures. Some people think that senescence is a consequence of declining photosynthesis, while others believe induction of senescence comes first and down-regulation of photosynthesis follows.

The genome of the model plant species Arabidopsis was completely sequenced in 2000 (around the same time as, and using similar technology to, the human genome). Since then the rice genome has been sequenced and other major crop species are on the way.

The availability of complete DNA sequences has opened up approaches to SAG analysis that are more comprehensive and efficient than the original cDNA library screening methods.

DNA microarray A collection of every gene in the genome can be set out as a microarray (sometimes called a gene chip). The mRNA transcripts present at a particular developmental stage can be identified by hybridising them to the array and visualising their abundance by fluorescence imaging.

This kind of genome-wide screening approach is called genomics. The suffix omics now gets applied to analyses at all levels of gene expression, from transcriptomics to proteomics to metabolomics to cellomics and on and on.

A big issue for omics is the sheer volume of data every experiment yields. To make sense of the outputs requires bioinformatics - that is, application of the right computational and data-handling methods.

Large amounts of information from a microarray experiment - Zentgraf et al (2004)

The next step on from describing senescence in terms of omics is to put together all these patterns of gene activity and gene product abundance into a kind of control circuit diagram that models how they interact and are regulated in a way that would be familiar to a software engineer.

Such a cybernetic view of the way cells work is at the heart of a new discipline called Systems Biology.

Senescence is quite a good subject for developing Systems Biology approaches. There are already active projects in a number of laboratories aimed at taking omics information about senescence, applying various bioinformatics tools and building a Systems description of the senescence program.

It is, however, worth adding a note of caution about Systems Biology and the precise meaning of the term "program" in this context. This report of a recent meeting outlines the issues, and Stefan Jansson and I have recently aired some provocative opinions too.