J. Neurochem. (2011) 119, 262–274. Abstract Cone‐rod homeobox (Crx) encodes Crx, a transcription factor expressed selectively in retinal photoreceptors and pinealocytes, the major cell type of the pineal gland. In this study, the influence of Crx on the mammalian pineal gland was studied by light and electron microscopy and by use of microarray and qRTPCR technology, thereby extending previous studies on selected genes ( Furukawa et al. 1999 ). Deletion of Crx was not found to alter pineal morphology, but was found to broadly modulate the mouse pineal transcriptome, characterized by a > 2‐fold down‐regulation of 543 genes and a > 2‐fold up‐regulation of 745 genes (p < 0.05). Of these, one of the most highly up‐regulated (18‐fold) was Hoxc4, a member of the Hox gene family, members of which are known to control gene expression cascades. During a 24‐h period, a set of 51 genes exhibited differential day/night expression in pineal glands of wild‐type animals; only eight of these were also day/night expressed in the Crx^−/− pineal gland. However, in the Crx^−/− pineal gland 41 genes exhibited differential night/day expression that was not seen in wild‐type animals. These findings indicate that Crx broadly modulates the pineal transcriptome and also influences differential night/day gene expression in this tissue. Some effects of Crx deletion on the pineal transcriptome might be mediated by Hoxc4 up‐regulation.

Currently, membrane curvature is understood as an active mechanism to control cells spatial organization and activity.

Posttranslational regulation of protein function is an ubiquitous mechanism in eukaryotic cells. Here, we analyzed biological properties of nodes and edges of a human protein-protein interaction phosphorylation-based network, especially of those nodes critical for the network controllability. We found that the minimal number of critical nodes needed to control the whole network is 29%, which is considerably lower compared to other real networks. These critical nodes are more regulated by posttranslational modifications and contain more binding domains to these modifications than other kinds of nodes in the network, suggesting an intra-group fast regulation. Also, when we analyzed the edges characteristics that connect critical and non-critical nodes, we found that the former are enriched in domain-to-eukaryotic linear motif interactions, whereas the later are enriched in domain-domain interactions. Our findings suggest a possible structure for protein-protein interaction networks with a densely interconnected and self-regulated central core, composed of critical nodes with a high participation in the controllability of the full network and less regulated peripheral nodes. Our study offers a deeper understanding of complex network control and bridges the controllability theorems for complex networks and biological protein-protein interaction phosphorylation-based networked systems.