Agriculture Reference
In-Depth Information
different questions their adaptive value; (iii) their low heritability makes
them appear stochastic and thus they were considered mostly to be
genetic noise in Mendelian inheritance studies (the famous environ-
ment
genetic interaction). All of these issues have been rigorously
addressed and some maternal environment effects appear likely to be
nonstochastic epigenetic transgenerational events (Donohue and
Schmitt 1998; Donohue 2009; Herrera and Bazaga 2010; Paun et al.
2010). The greatest issue that remains here is how adaptive and stable are
transgenerational epigenetic traits like maternal environment effects and
if the environment has more than just a stochastic effect (Donohue and
Schmitt 1998; Richards 2006; Swiezewski et al. 2007).
The metastable inheritance of transgenerational epigenetic pheno-
types underlies our fragile understanding of a hereditary system that
is highly plastic but not without boundaries. Until we more
×
firmly
understand the rules of the informational biochemistry of epigenetics
as we have accomplished with the four-letter genetic code, we will not
have the tools to put epigenetics in its proper place of signi
cance to
biology. Thus, whatever molecular processes are involved, understand-
ing this epigenetic plasticity will not only help explain transgenerational
epigenetics but also elucidate the mysterious and highly debated role of
the environment in evolution. Weisman
'
s paradoxical window of oppor-
tunity for the environment may be small, but nonetheless it may still
effectively allow the environment to sneak in and help direct evolution.
B. Extreme Environments and Emergent Genetics
It has long been professed that many important traits, such as yield and
stress tolerance, are multigenic or quantitative. It has been argued that it
will be exceedingly dif
cult to identify the heritable bases of these
complex characteristics. However, it has been recently realized that
the impressive yield gains that have been achieved through domestica-
tion have resulted from remarkably few genetic changes that likely are
altered heterochronic developmental modules (Tang et al. 2009). The
dif
culty in accepting the possibility that very few loci may dramatically
affect very complex traits has resulted from our historical bias for
genotype versus phenotype inheritance originating from Johannsen
s
theories (Johannsen 1911). The concept of an emergent property reveals
this biased view. By de
'
nition, no single or multiple sets of components,
short of the entire set, can be suf
cient for an emergent trait. By contrast,
any one of such components is much more likely to be required for
quantitative or emergent phenomena and thereby would be possible to
identify by any component subtraction approach, that is, loss of function
