ENMs were based on the 19 bioclimatic variables in the WorldClim data set [57]. These variables represent summaries of means and variation in temperature and precipitation, and likely summarize dimensions of climate particularly relevant in determining species distributions (Text S1 in Supporting Information).
http://plosone.org/article/info:doi/10.1371/journal.pone.0000563#article1.body1.sec4.sec1.p3

The PLoSONE discussion issue of using all Worldclim variables or just one set of summary variables (Bio1-19) refers to this section. The authors of the paper did discuss environmental variable choice quite a bit and agreed that bioclim variables were the best choice because they do represent a variety of climatic measurements that are likely explanatory in species distributions - bioclim includes not just yearly averages but also seasonal" measurements, like "precipitation of the warmest quarter", "mean temperature of wettest quarter" along with "highest/lowest" measurements like "max temperature of the warmest month". It might be interesting to see the magnitude of differences in the results using all the climatic variables --- my guess is that the other layers are highly correlated with the bioclim layers and results would be only marginally different.