Information on the secondary signals that propagate and diversify senescence responses inside the
cell acting at the gene expression (transcriptional) level may be gathered, in principle, from the analysis
of the SAG promoters. However, the nonexistence of mutants lacking the whole senescence syndrome
and the diversity of expression patterns among SAGs indicate an intrinsic redundancy and/or complexity
of this process. Indeed, SAG promoters have proved difficult to analyze. Some of them display several
regulatory boxes responding to developmental stage, stress, or/and hormones. Among those involved in
natural senescence, SAG12 from Arabidopsis thalianais one of the best senescence markers found to date
[40]. The dissection of this promoter has allowed the identification of a senescence-specific region re-
sponsible for its expression [217]. This sequence remains functional when expressed in heterologous sys-
tems such as tobacco [17], and it is conserved in orthologue genes from Brassica[218], suggesting that
some senescence regulatory mechanism may be conserved among higher plant species. This region has
been shown to bind nuclear proteins, perhaps senescence-specific transcription factors, which remain to
be characterized.
On the other hand, some habitual components of cellular signal transduction pathways have also
been reported to participate in plant senescence. For example, active MAP kinases have been identified
in senescent maize leaves [219], GTP binding and protein phosphorylation are present in senescing Ara-
bidopsis thaliana[220], Ca^2 chelators have been shown to prevent the senescence syndrome in victorin-
treated oat leaves [207], and overexpression of the transcription factor AmMYB308, which inhibits phe-
nolic acid metabolism, is known to induce premature senescence in tobacco [221]. Moreover,
homologues of the prohibitin family, which are mitochondrial proteins that regulate the replicative life
span, have also been found in plants [222], and the plant homologue of the defender against apoptotic
death gene is known to be down-regulated during senescence of flower petals [223].
In summary, current knowledge indicates that senescence regulation is a highly complex process in-
volving a multitude of signals that propagate and diversify the cellular responses in a dense network,
where even the main pathways remain poorly understood.
VII. MONITORING SENESCENCE
To follow senescence, it is necessary to find the appropriate parameters to measure the evolution of the
process. No single measurement is definitive, although in some cases certain parameters may be adequate
for particular tissues. However, single measurements should be checked against other parameters when-
ever possible.
The loss of chlorophyll is one of the most obvious changes during senescence of green organs, al-
though the existence of mutants in which senescence proceeds without chlorophyll loss indicates that this
change is not crucial to the process [5,46]. Precautions should be taken during the extraction of chloro-
phyll because some protocols (e.g., involving acetone) may lead to its degradation [224]. Radiolabeling
of chlorophyll has allowed a more sensitive measurement of its disappearance and the identification of
the degradation products [225].
Biochemical changes that precede chlorophyll loss include a decline in photosynthetic capacity [55]
and lowering of protein content. The progressive impairment of photosystem II may be evaluated by mon-
itoring its photochemical efficiency from chlorophyll fluorescence quenching curves [226,227]. Besides,
laser-induced fluorescence imaging is a sensitive and noninvasive technique that can be used to assess the
in vivo photosynthetic activity of green tissues [228], although it requires sophisticated instrumentation.
In contrast, the degradation of the CO 2 -fixing enzyme (Rubisco), a preferent target of proteases, is an easy
and widely used parameter to follow senescence in photosynthetically active organs [229]. The decrease
in total protein levels may also be measured. To avoid the interferences inherent in some methods, they
may be assayed through dye binding to protein adsorbed on washed paper disks [230] or by nitrogen de-
termination in digests [231].
Senescence may also be followed through measurements related to oxidative damage, such as extent
of lipid peroxidation [232] or levels of protective enzymatic activities (typically superoxide dismutase
and/or ascorbate peroxidase) [157,233]. Furthermore, specific assays have been developed to measure to-
tal antioxidant power (e.g., the FRAP assay) [234]. Leakage of the cell membranes may be also a signif-
icant parameter but occurs late in the senescence process.
Besides, in cases in which a hormone is directly implicated in senescence, measurement of the hor-
monal levels can be a good approach to detect early symptoms of decay. This is the case for some fruits,
such as tomato or avocado, where a burst of ethylene precedes the onset of ripening [235].
194 PEÑARRUBIA AND MORENO