System-wide molecular evidence for phenotypic buffering in Arabidopsis

In this Brief Communication, in a very nice issue of Nature Genetics, the authors profiled several Arabidopsis lines for a system-wide genetical genomics¹ study of molecular variation, integrating transcript, protein and metabolite data from a population of recombinant inbred lines² (RILs) of this model plant.

It appears that most of the genome and transcript variation between the two parental accession used for generating the RILs (and of the RILs themselves), is 'buffered' and does not result on phenotypic variation.

The largest fraction of molecular variants is silent at the phenotypic level and only a few influential 'hot spots' regions cause a mejor phenotypic variation across a range of environmental conditions

Interestingly, and despite the generalized buffering, the authors described six QTL (quantitative trait loci) hotspots,

which seem to correspond to a few molecular 'breakpoints' of an otherwise robust regulatory system

This is somewhat related to an article I posted recently on phenotypic variation buffering in yeast. In general, the idea seems to be that most of the variation in a species remains silent phenotypically, but alterations in certain genes or sequences may unleash the accumulated genetic variation and it can then be manifested at the phenotypic level. In the case of the yeast article, the authors proposed these genes were network hubs.
Although the authors comment that most of the reported hotspots could be linked to well-studied genes, such as CRY2 or HUA2, it must me emphasized that due to the somewhat imperfect resolution of QTL mapping, maybe other genes or regions may be involved in the observed traits.
Buffering and release of variation may be widespread phenomenon; for example, in both Drosophila and Arabidopsis, altered Hsp90 function results in pleiotropic phenotypic effects dependent on the underlying genotype.

I leave you with the final sentence of this article

variation in a multitude of Arabidopsis complex traits can be explained to a considerable extent by only a few QTL hot spots

Here's the abstract and reference:

System-wide molecular evidence for phenotypic buffering in Arabidopsis

Fu J, Keurentjes JJ, Bouwmeester H, America T, Verstappen FW, Ward JL, Beale MH, de Vos RC, Dijkstra M, Scheltema RA, Johannes F, Koornneef M, Vreugdenhil D, Breitling R, Jansen RC.
[1] Groningen Bioinformatics Centre, University of Groningen, 9751NN Haren, The Netherlands. [2] Department of Genetics, University Medical Centre Groningen, University of Groningen, 9700RB Groningen, The Netherlands. [3] These authors contributed equally to this work.

We profiled 162 lines of Arabidopsis for variation in transcript, protein and metabolite abundance using mRNA microarrays, two-dimensional polyacrylamide gel electrophoresis, gas chromatography time-of-flight mass spectrometry, liquid chromatography quadrupole time-of-flight mass spectrometry, and proton nuclear magnetic resonance. We added all publicly available phenotypic data from the same lines and mapped quantitative trait loci (QTL) for 40,580 molecular and 139 phenotypic traits. We found six QTL hot spots with major, system-wide effects, suggesting there are six breakpoints in a system otherwise buffered against many of the 500,000 SNPs.

Jingyuan Fu, Joost J B Keurentjes, Harro Bouwmeester, Twan America, Francel W A Verstappen, Jane L Ward, Michael H Beale, Ric C H de Vos, Martijn Dijkstra, Richard A Scheltema, Frank Johannes, Maarten Koornneef, Dick Vreugdenhil, Rainer Breitling, Ritsert C Jansen (2009). System-wide molecular evidence for phenotypic buffering in Arabidopsis Nature Genetics, 41 (2), 166-167 DOI: 10.1038/ng.308

¹ I know this may be a new concept for many of you, and instead of explaining it myself, and give a possibly complex definition, I'd rather direct you to an article on Trends in Genetics (2001) 17:388-391.
²See Broman KW. Genetics. 2005 Feb;169(2):1133-46.