Home Research Lab Staff Contact Info Web Links Publications Search Tools Journals Protocols Courses Intranet
   You are here: Home | Research | AtRAC

General information on Rac and Rho proteins

Rac and Rho proteins belong to a large superfamily of small GTPases, often but misleadingly called the Ras-superfamily. The true Ras proteins evolved relatively late and are probably derived from the abundant Rab GTPases. Rac and Rho-like proteins evolved early during the evolution of eukaryots and they have been found in several primitive protists as well as in plants, fungi and animals. How the first ancestral Rac/Rho proteins looked like is a open question, but a study of protist Rac/Rho homologs suggest that they had more in common with present day Rac proteins than with Rho proteins, Winge unpublished observations. This means that the Rho proteins evolved from a Rac-like ancestor and thereafter went through a period of more relaxed selection. That is the Rho proteins evolved faster than the Rac proteins and became a distinct family when they aquired a specific cellular function. So far just one Rho protein has been found in a protist, (Entamoeba histolytica Rho), and it appear as if the gene duplication that gave rise to the Rho proteins happened after the major division between plants/alveolates and Mycetozoa/fungi/animals, Baldauf S. L. et al. 2000. Distinct Rac-like proteins are on the other hand found in Entamoeba histolytica, Dictyostelium discoideum, Paramecium tetraurelia, Giardia intestinalis and in higher plants.

In contrast to the highly variable Rac proteins in the slime mold Dictyostelium discoideum Rivero F. et al. 2001, the plant RAC proteins are very distinct and show much less variability. This is a good indication that the plant RAC proteins have evolved from a single gene relatively late in evolution, probably as multicellularity and the first terrestrial plants appeared some 480 million years ago. This is also supported by the fact that RAC proteins from bryophyta are very similar to RAC proteins in vascular plants Winge et al. 2000.

In Arabidopsis thaliana this gene family consists of 11 members, Winge et. al. 1997. Based on phylogenetic analyzes and protein comparisons we decided to call these proteins ARAC (Arabidopsis rac like proteins) or alternatively AtRACs.

The RAC gene family are divided into 2 major groups as shown on Figure 1. below. The largest group, group I, includes the proteins: AtRAC1, AtRAC2, AtRAC3, AtRAC4, AtRAC5, AtRAC6, AtRAC9 and AtRAC11. Members of this group all share the chAtRACteristic C-terminal geranylgeranylation signal CXXL which is common among the known Rac proteins. AtRAC7, AtRAC8 and AtRAC10 forms a another distinct group, group II, where the members have certain key amino acids differences in both the effector and insert region. In addition AtRAC7, 8 and 10 do not have the C-terminal geranylgeranylation motif, but have another conserved Cys motif (aaCG, a=aliphatic amino acid) with resemblance to palmitoylation signals found in Ras proteins, in addition AtRAC7 have a C-terminal consensus signal for farnesylation. Proteins belonging to RAC group II are the newcomers and they probably emerged some 200-400 million years ago as a result of the insertion of an extra intron in the extreme 3' end of an ancestral rac gene. All seed plants / flowering plants have rac multigene families that include proteins from group I and II. Even non vascular plants have several RAC proteins.

The function of the plant RAC proteins is varied and they probably have functions similar to RAC proteins found in other eukaryotic organisms. That is: they are instrumental in regulating the actin cytoskeleton, they regulate the production on superoxide through a membrane bound NADPH dependent oxidase and they probably have important regulatory functions downstream of various membrane bound receptors.

We have also found a number of interesting genes flanking the AtRAC genes during the mapping and sequencing process. What we find intriguing is that a number of the neighboring genes encodes proteins which one might suspect are tied to the function of the RAC proteins. These includes proteins involved in vesicle transport, components involved in the regulation of the actin cytoskeleton, cell signaling (for example: a histidine kinase, a receptor like kinase and more), cell cycle regulation, proteins suspected to be involved in polar cell growth and stem elongation, and proteins involved in the regulation of H2O2 levels in the plant (ascorbate peroxidases and catalases). Even though the Arabidopsis genome is highly duplicated and extensively reshuffled, it is unlikely that these genes are arranged next to the AtRAC genes by pure coincidence. A similar clustering of genes is also observed for genes involved in DNA repair, replication, and other pathways/processes.

FIG. 1 Abbreviations: As; Populus tremuloides (Aspen), Br; Brassica rapa, Bv; Beta vulgaris, Ca; Cicer arietinum, Gh; Gossypium hirsutum, Gm; Glycine max, Le; Lycopersicon esculentum, Lj; Lotus japonicus, Ms; Medicago sativa, Nt; Nicotiana tabacum, Os; Oryza sativa, Ph; Physcomitrella patens, Pm; Picea mariana, Ps; Pisum sativum, Pt; Pinus taeda, Zm; Zea mays.

EST1: AI730323, EST2: AI727570, EST3: AI731040, EST4: AI775563, EST5: D41104 and C26233, EST6: AW039993, EST7: AI937960, EST8: AW218480, EST9: AI759954, EST10: AW102025 and AI900160, EST11: AI162198, EST12: AU029919 and C73805, EST13: AI164960 and AI161509, EST14: AI901151, EST15: AW225989, EST16: AW056772, EST17: AI812534.

AtRAC/AtRAC alignment

Duplications of AtRAC genes

The RAC genes in plants have undergone several duplications during the last 200-400 million years, Winge et al., 2000. With the complete genome sequence of Arabidopsis thaliana a picture is emerging which can partly explain why certain gene families have undergone large expansions in numbers. In a recent paper by Vision et al., 2000 they have tried to divide the genomic duplications in Arabidopsis into age classes, A --> F. Classes A and B are recent large duplications (0 - 50 million years ago), whereas the classes C, D, E, represent older duplications (100 - 170 million years ago). The oldest class F includes duplications that is 200 million years or older. Some of the AtRAC gene duplications can be assigned to these age classes but not all. The youngest of the AtRAC gene duplications which can be assigned to age class A and B involves the AtRAC1/AtRAC6 and AtRAC4/AtRAC5 gene pairs. This means that homologs (othologs) of these genes exist probably just in Brassicales or closely related orders. The AtRAC1 and AtRAC6 genes are closest related to ARAC11, and the duplication which resulted in the AtRAC11 (AtRAC1/AtRAC6) gene pair probably belongs to age class C. This is supported by phylogenetic studies which show that homologs of the AtRAC1, AtRAC6, AtRAC11 sub-group is found in both eurosid I and eurosid II. The division of eurosid I and euosid II groups dates back to 90-100 million years and fit well with age class C.

The relationship between AtRAC1, AtRAC6 and AtRAC11 sub-group and the other AtRAC genes is further "illuminated" when the genes flanking the AtRAC genes are compared. This reveal a complex genome with many gene rearrangements, small insertions and deletions. See figure below.

From this figure it is possible to see the results of some of the large duplications that has given rise to of the AtRAC genes. The close relationship between AtRAC1, AtRAC6 and AtRAC11 is evident and suggest that the founding member of this sub-group was created by a large duplication preceding the creation of the eurosid I and eurosid II groups. The ancestor of the AtRAC1, 6 and 11 sub-group, may be the result of a duplication involving the AtRAC4/AtRAC5 ancestor, but this group may also be a direct descendant from AtRAC3. In any case this duplication must have taken place more than 120 million years ago. A phylogenetic analysis suggests that it dates back to a time well before the division between asterids and rosids and this duplication may be assigned to age class D (approximately 140 million years ago). The phylogenetic tree at the beginning of this document shows the deduced evolutionary relationships between the various plant RAC genes and division into distinct groups.

Interestingly no genes closely related to AtRAC4 and AtRAC5 (orthologs) have been found in plants other than the Brassicaceae. This may indicate it is a fast evolving sub-group or that it has been deleted in number of other plant families such as; Fabales, Malvales and Solanales.

The AtRAC3 gene is a "wild card" that probably recently was inserted at its current location on Chr IV. There are some indications that it was previously situated in the middle part of Chr II. Whether AtRAC3 was created as the result of a duplication involving the AtRAC4/AtRAC5 ancestor or the other way around is at present unknown. Anyway, this gene duplication, which resulted in the AtRAC3 and AtRAC4/AtRAC5 genes, may be as old as 170 million years, that is equivalent to age class E. Furthermore, an ancestral protein with high similarity to the present day AtRAC3, AtRAC4 and AtRAC5 proteins must have existed prior to the division of mono-cotyledons and dicotyledons (170-200 million years ago). Evidence for this is found in a small but distinct group of RAC proteins from monocots, RAC group Ic (monocot racDs), which are similar to rac proteins belonging to RAC group Ia (see phylogenetic tree). The common ancestor of RAC group Ia and monocot racDs must therefore have been created by a ancient gene duplication dating back to maybe 250 million years or older, that is it belongs to age class F or older. (There is some evidence that RAC genes related to monocot racDs and their dicot group Ia relatives exist in gymnosperms). So which one of the RAC genes are the "primordial" RAC gene, the ancestor for group Ia and Ic.

From the figure of the AtRAC gene duplications a clear relationship between AtRAC2 and AtRAC11 and AtRAC4/AtRAC5 genes is evident. Two copies of the AtCrn gene exist in the Arabidopsis genome, both are next to a AtRAC gene (5 Kb downstream of AtRAC2 and approximately 40 Kb upstream of AtRAC11). In addition upstream of AtRAC2 a number of anther specific proline rich proteins (APGs) are located, whereas similar APGs are located 5-10 Kb downstream of AtRAC4 and AtRAC5. These are the relicts of an old duplication which probably happened more than 250 million years ago. This suggest that RAC group Ia (dicots) and Ic (monocots) are derived from RAC group Ib, the AtRAC2 sub-group.

AtRAC2 and its cousins in RAC group Ib are probably some of the "oldest" members of the plant RAC gene family, and there are indications that proteins related to AtRAC2 existed as far back as gymnosperms, 250 million years ago. Members of RAC group Ib are also similar to RAC proteins in conifers and mosses (bryophytes).

AtRAC9 is the most divergent of the AtRAC proteins and phylogenetic analysis of plant RAC proteins often place AtRAC9 into a separate group at the base of the tree. This may suggest that AtRAC9 is the "oldest" of the RAC genes, or an alternative view, that it has not been duplicated in a very long time. Studies of the AtRAC gene duplications show that AtRAC1 and AtRAC9 have a common neighbor gene, encoding a protein with a phorbol ester / diacylglycerol binding domain, T17A5.16 and F4I1.18 / F4I1.19 (a tandem duplication). Thus, both AtRAC2 and AtRAC9 has gene neighbors that provide a link with the other members of RAC group I. While several genes related to AtRAC2 has been found in other plants, AtRAC9 remains more elusive and has so far just been detected in Arabidopsis. This is most likely due to its low expression.

RAC group II include members both from monocots and dicots, and their ancestral gene probably emerged some 200-400 million years ago as a result of the insertion of an extra intron/exon in the extreme 3' end of a RAC gene. This insertion destroyed the C-terminal prenylation motif in these genes, but apparently some conserved C-terminal Cys residues in the last exon (exon 8) may serve as similar motifs directing membrane association through other modifications, by lipids or others.