Pax factors - regulatory targets and integration in the TRANSFAC database

Richard Ohnhäuser (rio@biobase.de), Edgar Wingender (ewi@biobase.de)
Biobase Biological Databases GmbH, Mascheroder Weg 1b, D-38124 Braunschweig

Introduction

Pax (paired box) factors are a highly conserved family of transcription factors belonging to the helix turn helix class. In higher vertebrates nine members of the Pax family have been isolated which are classified into four paralog groups. The expression of Pax genes is temporally and spatially restricted during development of CNS and various other organs. Their key role in cell fate, early patterning and organogenesis indicates that Pax genes are master control genes. Some Pax mutations are linked to certain human diseases (Waardenburg syndrome and Aniridia) and are also associated with cancer. Thus it is very importent to gain more knowledge about the regulatory targets of Pax factors to get a better insight in developmental processes and  formation of inherited diseases.

General features of Pax

Ø

large family of developmental regulatory genes that encode nuclear transcription factors

Ø

early expressed during embryogenesis

Ø

normally downregulated in adult stages

Ø

involved in organogenesis (mainly CNS)

Ø

role in morphological boundaries and early regionalisation

Ø

proto-oncogenes

General features of Pax

The paired domain is a DNA-binding domain of 128 aa originally described in the Drosophila segmentation gene paired (Bopp et al., 1986; Frigerio et al., 1986). The result of the paired mutation is a deletion of alternating segments (pair rule mutation) (figure 1). A paired domain has two subdomains (PAI domain and RED domain) which consists of three a-helixes each (figure 2). In addition to the paired domain some Pax proteins comprise a paired type homeo domain and a conserved octapeptide.

Figure 1: the paired mutation (Bopp et al. 1986)

Figure 2: structure of Pax genes

DNA binding of Pax proteins

The DNA binding is mainly confirmed by helix 3 of the PAI domain in the major groove whereas helix 1+2 are antiparallel (figure 3). Additionally the homeo and RED domain can be involved so various combinations of DNA binding are possible (Jun et al. 1996).

Figure 3: DNA binding of Pax (Jun et al. 1996)

Evolution of Pax

According to sequence similarity of the paired box and the existence of a homeobox and octapeptide, Pax genes were grouped into four paralog goups (table 1). Surprisingly in more primitive organisms certain precursors corresponding to these groups were found (Sun et al. 1997, figure 4). This indicates that gene duplication events lead to divergation of more members within one group based on one ancestral Pax gene.

Table 1: paralog groups of Pax genes

Figure 4: phyllogenetic tree of Pax genes  based on paired box similarities (Sun et al. 1997)

Regulatory targets of Pax and integration in TRANSFAC

Although Pax genes are universal mastergenes and should act upstream to many other genes there is little known about their natural targets. The binding sites and interacting factors known up to now were integrated in the TRANSFAC database and are summarised in figure 5. Interestingly all targets can grouped to genes concernig developmental processes like proliferation, differentiation, hormonal regulation, organogenesis, cell adhesion and migration (see legend figure 5). Pax genes were found ubiquitous in all known metazoan organisms (figure 4) although the number is decreased in more primitive animals. There are 78 Pax factors integrated in TRANSFAC to date which are all listed in table 2.
Due to the intensive reseach in developmental biology the number of binding sites and factors in different organisms is permanently increasing and TRANSFAC gains to update the database regularly in this important field.

Factor-site-links and interacting factors

Legend:

Figure 5: regulatory targets of Pax

Table 2: Pax factors in TRANSFAC

total number of entries: 78