NORMAL GENOMIC

• Balaji S, Babu MM, Iyer LM, Aravind L
• Discovery of the principal specific transcription factors of Apicomplexaand their implication for the evolution of the AP2-integrase DNA bindingdomains.
• Nucleic Acids Res. 2005; 33: 3994-4006
• Display abstract

• The comparative genomics of apicomplexans, such as the malarial parasitePlasmodium, the cattle parasite Theileria and the emerging human parasiteCryptosporidium, have suggested an unexpected paucity of specifictranscription factors (TFs) with DNA binding domains that are closelyrelated to those found in the major families of TFs from other eukaryotes.This apparent lack of specific TFs is paradoxical, given that theapicomplexans show a complex developmental cycle in one or more hosts anda reproducible pattern of differential gene expression in course of thiscycle. Using sensitive sequence profile searches, we show that theapicomplexans possess a lineage-specific expansion of a novel family ofproteins with a version of the AP2 (Apetala2)-integrase DNA bindingdomain, which is present in numerous plant TFs. About 20-27 members ofthis apicomplexan AP2 (ApiAP2) family are encoded in differentapicomplexan genomes, with each protein containing one to four copies ofthe AP2 DNA binding domain. Using gene expression data from Plasmodiumfalciparum, we show that guilds of ApiAP2 genes are expressed in differentstages of intraerythrocytic development. By analogy to the plant AP2proteins and based on the expression patterns, we predict that the ApiAP2proteins are likely to function as previously unknown specific TFs in theapicomplexans and regulate the progression of their developmental cycle.In addition to the ApiAP2 family, we also identified two other novelfamilies of AP2 DNA binding domains in bacteria and transposons. Usingstructure similarity searches, we also identified divergent versions ofthe AP2-integrase DNA binding domain fold in the DNA binding region of thePI-SceI homing endonuclease and the C-terminal domain of the pleckstrinhomology (PH) domain-like modules of eukaryotes. Integrating thesefindings, we present a reconstruction of the evolutionary scenario of theAP2-integrase DNA binding domain fold, which suggests that it underwentmultiple independent combinations with different types of mobileendonucleases or recombinases. It appears that the eukaryotic versionshave emerged from versions of the domain associated with mobile elements,followed by independent lineage-specific expansions, which accompaniedtheir recruitment to transcription regulation functions.

• Makarova KS, Aravind L, Koonin EV
• SWIM, a novel Zn-chelating domain present in bacteria, archaea andeukaryotes.
• Trends Biochem Sci. 2002; 27: 384-6
• Display abstract

• A previously undetected domain with a CxCx(n)CxH pattern of predictedzinc-chelating residues was identified in a variety of prokaryotic andeukaryotic proteins. These include bacterial ATPases of the SWI2/SNF2family, plant MuDR transposases and transposase-derived Far1 nuclearproteins, and vertebrate MEK kinase-1. This domain was designated SWIMafter SWI2/SNF2 and MuDR, and is predicted to have DNA-binding andprotein-protein interaction functions in different contexts.

• Buurman ET, Jiang W, McCoy M, Averett DR, Thompson CM, Wobbe CR
• Validation of Cdc68p as a novel antifungal target.
• Arch Microbiol. 2002; 178: 428-36
• Display abstract

• Candida albicans is the main cause of systemic fungal infections for whichthere is an urgent need for novel antifungal drugs. The CP (Cdc68p-Pob3p)complex, which is involved in transcription elongation, was evaluated as aputative antifungal target. In order to predict the consequences ofinhibition of this complex, the largest CP subunit in Saccharomycescerevisiae, Cdc68p, was the first novel target to be tested in GATE, arecently described, quantitative target inactivation system. Depletion ofthe cell's pool of Cdc68p led to rapid cell death. Subsequently, the C.albicans orthologue of CDC68, CaCDC68, was cloned. Attempts to disruptboth alleles were unsuccessful, thus suggesting an essential role ofCaCDC68 in this fungus also. Furthermore, CDC68 was proven to be presentin Neurospora crassa and Aspergillus nidulans, thus suggesting that the CPcomplex is widespread among fungi and could serve as a broad rangeantifungal target. Analysis of Cdc68p and Pob3p sequences indicatedsignificant structural differences between fungal CP complexes and thosepresent in higher eukaryotes. These results predict that, in principle,fungal-specific ligands of CP complexes could be identified that couldsubsequently serve as chemical starting points towards the development ofnew antifungal therapeutic agents.

• Bork P, Gibson TJ
• Applying motif and profile searches.
• Methods Enzymol. 1996; 266: 162-84

• Rost B, Sander C
• Prediction of protein secondary structure at better than 70% accuracy.
• J Mol Biol. 1993; 232: 584-99
• Display abstract

• We have trained a two-layered feed-forward neural network on anon-redundant data base of 130 protein chains to predict the secondarystructure of water-soluble proteins. A new key aspect is the use ofevolutionary information in the form of multiple sequence alignments thatare used as input in place of single sequences. The inclusion of proteinfamily information in this form increases the prediction accuracy by sixto eight percentage points. A combination of three levels of networksresults in an overall three-state accuracy of 70.8% for globular proteins(sustained performance). If four membrane protein chains are included inthe evaluation, the overall accuracy drops to 70.2%. The prediction iswell balanced between alpha-helix, beta-strand and loop: 65% of theobserved strand residues are predicted correctly. The accuracy inpredicting the content of three secondary structure types is comparable tothat of circular dichroism spectroscopy. The performance accuracy isverified by a sevenfold cross-validation test, and an additional test on26 recently solved proteins. Of particular practical importance is thedefinition of a position-specific reliability index. For half of theresidues predicted with a high level of reliability the overall accuracyincreases to better than 82%. A further strength of the method is the morerealistic prediction of segment length. The protein family predictionmethod is available for testing by academic researchers via an electronicmail server.