Portal de Periódicos da CAPES

A New Gene-finding Tool

2004; Elsevier BV; Volume: 279; Issue: 8 Linguagem: Inglês

10.1074/jbc.m307347200

1083-351X

Sonja Eichmüller, Valeria Vezzoli, Claudia Bazzini, Markus Ritter, Johannes Fürst, Martin Jakab, Andrea Ravasio, Sabine Chwatal, Silvia Dossena, Guido Fernando Botta, G. Meyer, Brigitte Maier, Giovanna Valenti, Florian Läng, Markus Paulmichl,

Genetic Neurodegenerative Diseases

How can a large number of different phenotypes be generated by a limited number of genotypes? Promiscuity between different, structurally related and/or unrelated proteins seems to provide a plausible explanation to this pertinent question. Strategies able to predict such functional interrelations between different proteins are important to restrict the number of putative candidate proteins, which can then be subjected to time-consuming functional tests. Here we describe the use of the operon structure of the nematode genome to identify partner proteins in human cells. In this work we focus on ion channels proteins, which build an interface between the cell and the outside world and are responsible for a growing number of diseases in humans. However, the proposed strategy for the partner protein quest is not restricted to this scientific area but can be adopted in virtually every field of human biology where protein-protein interactions are assumed. How can a large number of different phenotypes be generated by a limited number of genotypes? Promiscuity between different, structurally related and/or unrelated proteins seems to provide a plausible explanation to this pertinent question. Strategies able to predict such functional interrelations between different proteins are important to restrict the number of putative candidate proteins, which can then be subjected to time-consuming functional tests. Here we describe the use of the operon structure of the nematode genome to identify partner proteins in human cells. In this work we focus on ion channels proteins, which build an interface between the cell and the outside world and are responsible for a growing number of diseases in humans. However, the proposed strategy for the partner protein quest is not restricted to this scientific area but can be adopted in virtually every field of human biology where protein-protein interactions are assumed. Functional interactions between different proteins provide a plausible explanation for the discrepancy between the limited number of individual genes and the huge number of well defined functional phenotypes in biological systems. Strategies, which allow for the identification of functional interactions between a target protein and unknown partners are of utmost importance to efficiently scrutinize the functional genome. A powerful approach for this is provided by the use of yeast at the level of orthologue search of human disease genes and as a genetic toolkit for hybrid-protein interactions but with the known limitations based on the archaic functional architecture of yeast if compared with animals. Caenorhabditis elegans is an animal system allowing for straightforward genetic approaches for the functional analysis of single genes, i.e. double-stranded RNA interference for knocking out single or multiple genes (1Kamath R.S. Fraser A.G. Dong Y. Poulin G. Durbin R. Gotta M. Kanapin A. Le Bot N. Moreno S. Sohrmann M. Welchman D.P. Zipperlen P. Ahringer J. Nature. 2003; 421: 231-237Crossref PubMed Scopus (2701) Google Scholar, 2Fire A. Xu S. Montgomery M.K. Kostas S.A. Driver S.E. Mello C.C. Nature. 1998; 391: 806-811Crossref PubMed Scopus (11654) Google Scholar), and possesses a genome that is partially structured in operons, unique for animal cells (3Nilsen T.W. Annu. Rev. Microbiol. 1993; 47: 413-440Crossref PubMed Scopus (141) Google Scholar, 4Spieth J. Brooke G. Kuersten S. Lea K. Blumenthal T. Cell. 1993; 73: 521-532Abstract Full Text PDF PubMed Scopus (250) Google Scholar). Operons are a special gene organization found in bacteria and Archaea, thronging together different genes encoding proteins that very often act in the same regulatory pathway (5Lawrence J.G. Cell. 2002; 110: 407-413Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar). We used this uniqueness of the nematode genome structure to predict a previously unknown functional relationship between the IClnN2 protein and the Nx protein (6Furst J. Ritter M. Rudzki J. Danzl J. Gschwentner M. Scandella E. Jakab M. Konig M. Oehl B. Lang F. Deetjen P. Paulmichl M. J. Biol. Chem. 2002; 277: 4435-4445Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). The cooperation between the two proteins led to the reduction of the voltage-dependent inactivation of the IClnN2-induced ion current when reconstituted in lipid bilayers (6Furst J. Ritter M. Rudzki J. Danzl J. Gschwentner M. Scandella E. Jakab M. Konig M. Oehl B. Lang F. Deetjen P. Paulmichl M. J. Biol. Chem. 2002; 277: 4435-4445Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). Now we will attempt to test the hypothesis that a functional interaction between operon-bound proteins on the level of nematode cells can also be extrapolated to the human system. Blumenthal et al. (7Blumenthal T. Evans D. Link C.D. Guffanti A. Lawson D. Thierry-Mieg J. Thierry-Mieg D. Chiu W.L. Duke K. Kiraly M. Kim S.K. Nature. 2002; 417: 851-854Crossref PubMed Scopus (271) Google Scholar) describe the nematode genome as embracing at least 1000 distinct operons containing between two and eight genes. Because we used the operon structure of the nematode genome as an efficient way to identify functional partner proteins, we examined the possible operon structure of genes coding for human orthologues of ion channels in C. elegans to provide a rational starting point for more in-depth analyses aiming to identify the functional entities of complex ion channels in human cells. Ion channels build an interface between the cell interior and the extracellular environment (8Hille B. Ionic Channels of Excitable Membranes. Sinauer Associates, Inc., Sunderland, MA1992Google Scholar) and are increasingly recognized to be responsible for numerous diseases in humans (9Ashcroft F.M. Ion Channels and Disease. Academic Press, Inc., New York2000Google Scholar, 10Hubner C.A. Jentsch T.J. Hum. Mol. Genet. 2002; 11: 2435-2445Crossref PubMed Scopus (170) Google Scholar). 276 different human ion channels were taken from the protein databases SwissProt, TrEMBL (us.expasy.org/sport), and GenBank™ (NCBI). The pairwise sequence alignment of the protein sequences was performed on WormBase release WS95 (WS95; total of 21,341 proteins; www.wormbase.org/db/searches/blat) using BLASTP.OMP-WashU provided from Washington University (St. Louis; blast.wustl.edu). A positive search result was assumed at a p value of 1500 bp, which does not exclude the existence of an operon structure.Table IGenomic structure of nematode orthologues of human channel proteinsView Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT)View Large Image Figure ViewerDownload Hi-res image Download (PPT) Open table in a new tab Table IINematode ion channel genes with an intercistronic distance of ≤300 bp and those structured in known operonsView Large Image Figure ViewerDownload Hi-res image Download (PPT) Open table in a new tab The nematode strain (N2) used in this work was provided by the Caenorhabditis Genetics Center. The nematodes were grown on Escherichia coli strain OP50, harvested, and homogenized. Total RNA was isolated according to standard protocols (13Maniatis T. Fritsch E. Sambrook J. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, New York1982Google Scholar). RT was performed according to standard protocols (13Maniatis T. Fritsch E. Sambrook J. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, New York1982Google Scholar) using avian myeloblastosis virus reverse transcriptase (Promega). For PCR amplifications standard PCR protocols using Taq or Pfu polymerases (Roche Applied Science and Stratagene) were applied according to the manufacturer's protocols. The primers used for RT-PCR were as follows: SL1, 5′-GGTTTAATTACCCAAGTTTGA-3′; SL2, 5′-GGTTTTAACCCAGTTACTCAAG-3′; acr-2, 5′-GAAGACGTTGGATTTGAAGG-3′ (position 479-498); acr-3, 5′-GGGTCCCATTGCATTTGATAA-3′ (first PCR; position 255-275) and 5′-CATTCCATTTCATTGTAGGC-3′ (second, or nested PCR; position 234-253); mrp-1, 5′-CATATTGCACGAGGTCAC-3′ (position 332-349); mrp-2, 5′-TCCGAATTGAACAAGGTCTCC-3′ (first PCR; position 331-351) and 5′-GTTGAGCAGTTCGGGTATGA-3′ (second, or nested PCR; position 177-196). The PCR products obtained were sequenced using an automatic sequencer (LiCor Gene ReadIR 4200) and SequiTherm EXEL-II DNA sequencing kit (Epicenter) with the protocols suggested by the manufacturer. Plasmids and Fusion Proteins—Standard procedures (13Maniatis T. Fritsch E. Sambrook J. Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory, New York1982Google Scholar) were used for DNA preparation, cloning, purification, and sequencing. A plasmid coding for a fusion protein of human ICln attached to the N terminus of ECFP (hICln-CFP) was prepared by cloning the open reading frame (ORF) coding for the human ICln in-frame into the vectors pECFP-N1 (Clontech) using the XhoI and BamH1 restriction sites. A plasmid coding for a fusion protein of HSPC038 attached to the C terminus of EYFP (YFP-HSPC038) was prepared by cloning the ORF coding for HSPC038 in-frame into the vector pEYFP-C1 (Clontech) using the XhoI and BamH1 restriction sites. All constructs were sequenced using an automated sequencer (Gene ReadIR 4200, LiCor). For all FRET experiments ECFP served as a donor, and EYFP as an acceptor (14Ritter M. Ravasio A. Jakab M. Chwatal S. Furst J. Laich A. Gschwentner M. Signorelli S. Burtscher C. Eichmuller S. Paulmichl M. J. Biol. Chem. 2003; 278: 50163-50174Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar). Cells were cotransfected with hICln-CFP and YFP-HSPC038. Transfected cells were superfused at room temperature (20-23 °C) with phosphate-buffered saline (136.9 mm NaCl, 2.7 mm KCl, 3.2 mm Na2HPO4, 1.47 mm KH2PO4, pH 7.4). Visualization of ECFP- and/or EYFP-expressing cells and detection of FRET was performed on an Olympus IX70 inverted microscope equipped with a monochromator (Polychrome 4, TILL Photonics) and a cooled CCD camera (TILL Imago SVGA) controlled by TILL Vision software (versions 4.01). Experiments were either performed by changing three separate Olympus BX cubes equipped with the appropriate filter combinations for ECFP, EYFP, and FRET measurements or by using a CFP/YFP dual-band polychroic mirror in combination with a real time dual color imaging device (Dual-View™, Optical Insights). All images were adjusted to pixel-by-pixel alignment and corrected for background autofluorescence, which was then clamped to zero (14Ritter M. Ravasio A. Jakab M. Chwatal S. Furst J. Laich A. Gschwentner M. Signorelli S. Burtscher C. Eichmuller S. Paulmichl M. J. Biol. Chem. 2003; 278: 50163-50174Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar). Acceptor Photobleaching—Light from a xenon lamp (XBO 75, Osram) was passed through a 515-nm cut-off filter toward the object. The FRET efficiency was calculated from the increase in ECFP intensity after partial bleaching of EYFP according to the formula calFRETeff=1-(ICFP-bg0/ICFP-bgt) In addition, FRET efficiency was extrapolated to the total bleaching of EYFP according to equations, exFRETeff=1-(αICFP-bg0/(ICFP-bgtβICFP-bg0)) α=1-β β=(IYFP-bgt/IYFP-bg0) where ICFP-bg and IYFP-bg are the background-corrected gray-value densities of ECFP and EYFP, respectively, measured before (suffix 0) and t min (suffix t) after bleaching of EYFP (14Ritter M. Ravasio A. Jakab M. Chwatal S. Furst J. Laich A. Gschwentner M. Signorelli S. Burtscher C. Eichmuller S. Paulmichl M. J. Biol. Chem. 2003; 278: 50163-50174Abstract Full Text Full Text PDF PubMed Scopus (55) Google Scholar, 15Jobin C.M. Chen H. Lin A.J. Yacono P.W. Igarashi J. Michel T. Golan D.E. Biochemistry. 2003; 42: 11716-11725Crossref PubMed Scopus (19) Google Scholar). The FRET efficiency image (Fig. 2c) was generated using Equation 2 on a pixel-to-pixel calculation from the images recorded at time zero and after 2 min. Homologues for the Human Ion Channels in C. elegans—In our study we included the following human channels: (i) sodium channels (epithelial, amiloride-sensitive channels, voltage dependent channels, and nicotinic acetylcholine-receptor-channels), (ii) potassium channels (voltage-gated channels of the Kv, eag, and KCNQ family, calcium-activated channels, inwardly rectifying and ATP-sensitive channels, and channels of the two pore domain family), (iii) calcium channels (voltage-dependent channels, transient receptor potential channel family (TRP), and ligand gated receptor channels), (iv) chloride channels (voltage-gated channels, intracellular channels, calcium-activated channels, channels from the ATP binding cassette (ABC) family, channels from the γ-aminobutyric acid, glutamate, and glycine receptor family, and other channels), (v) cyclic nucleotide-gated channels, (vi) aquaporines, (vii) gap-junction-forming channels, and (viii) other channels. These eight channel families consist of 276 different human ion channels listed in GenBank™ for which the respective nematode orthologues were identified using a pairwise sequence alignment of the protein sequences with a p value <10-2, as described in detail under “Experimental Procedures.” For the 10 voltage-dependent sodium channels (Nav1.1 to Nav1.9 and Nax) the only nematode orthologues that could be identified were egl-19 and cca-1, which are both identified as calcium channels and are, therefore, not included in our listing. For the remaining 266 human channel proteins a total number of 235 nematode orthologues were identified (88.3%) based on a p value of 10-40, and 5.1% demonstrate a p value of ≤10-2 and >10-10. These 235 nematode orthologues consist of 80 individual proteins, since different human ion channels correspond to the same nematode protein. A p value of <10-2 is no proof that the respective nematode orthologue is indeed a member of the same functional protein group. However, in 63.8% of the 80 nematode orthologues the respective human homologue can be obtained if the similarity search is performed by starting with the nematode protein and searching for the respective human orthologue. This value is slightly higher than the values of 44% reported for a random collection of human cDNA (16Wheelan S.J. Boguski M.S. Duret L. Makalowski W. Gene (Amst.). 1999; 238: 163-170Crossref PubMed Scopus (43) Google Scholar), the value of ≈42% reported for human disease genes (17Culetto E. Sattelle D.B. Hum. Mol. Genet. 2000; 9: 869-877Crossref PubMed Scopus (199) Google Scholar), or the value of 50% predicted by Ahringer (18Ahringer J. Curr. Opin. Genet. Dev. 1997; 7: 410-415Crossref PubMed Scopus (33) Google Scholar), pointing to the possibility that within the ion channel families in different species a higher sequence similarity is maintained with respect to other functional protein groups. Therefore, we presume that the respective nematode orthologues of human ion channel genes listed in Table I provide good support for the hypothesis that the corresponding proteins act in the same functional compartments. This is further supported by the fact that for 31.3% of the nematode orthologues (25 proteins marked accordingly in Table I) a biophysical characterization was done, demonstrating that the respective proteins are indeed involved in current formation. For the icl-1 protein, the nematode homologue of the ICln protein, known as a channel-forming protein involved in volume regulation in mammalian cells (19Paulmichl M. Li Y. Wickmann K. Ackerman M. Peralta E. Clapham D. Nature. 1992; 356: 238-241Crossref PubMed Scopus (310) Google Scholar), the direct channel-forming aptitude was proven by the reconstitution of the protein in lipid bilayer (6Furst J. Ritter M. Rudzki J. Danzl J. Gschwentner M. Scandella E. Jakab M. Konig M. Oehl B. Lang F. Deetjen P. Paulmichl M. J. Biol. Chem. 2002; 277: 4435-4445Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar), the only technique that unambiguously allows one to test for the channel-forming capacity of proteins. Ion Channels Organized in Nematode Operons—An operon is a gene cluster transcribed from a single promoter located upstream of its 5′-end. During transcription or shortly thereafter monocistronic mRNAs are processed from the polycistronic pre-mRNA and trimmed by trans-splicing (11Blumenthal T. Gleason K.S. Nat. Rev. Genet. 2003; 4: 112-120Crossref PubMed Scopus (21) Google Scholar). The latter process adds splice leaders (SL) at the respective 5′-ends of the monocistronic mRNAs. The SL sequences come in two forms, one termed SL1, which are short sequences of about 22 nucleotides added at the mRNA from the leading gene in the operon, and the second, termed SL2, added at the mRNAs of the downstream gene or genes of the same operon (20Nilsen T.W. Trends Genet. 2001; 17: 678-680Abstract Full Text Full Text PDF PubMed Scopus (67) Google Scholar). However, it was also described that SL1 sequences can lead the downstream gene (21Kato Y. Aizawa T. Hoshino H. Kawano K. Nitta K. Zhang H. Biochem. J. 2002; 361: 221-230Crossref PubMed Scopus (100) Google Scholar). The SL2 sequence was used to identify genes clustered in operons by probing whole-genome microarrays, leading to ≈900 operons estimated to contain about 90% of the operons in the nematode genome (7Blumenthal T. Evans D. Link C.D. Guffanti A. Lawson D. Thierry-Mieg J. Thierry-Mieg D. Chiu W.L. Duke K. Kiraly M. Kim S.K. Nature. 2002; 417: 851-854Crossref PubMed Scopus (271) Google Scholar). A total of 266 human ion channels were tested comprised of 84 potassium channels, 49 calcium channels, 23 sodium channels, 68 chloride channels, 9 cyclic nucleotide-gated channels, 11 aquaporin channels, 17 gap-junction channels, and 5 other channels (Table I). For these 266 human ion channels 235 nematode orthologues can be identified comprised of 75 potassium channels, 43 calcium channels, 23 sodium channels, 65 chloride channels, 9 cyclic nucleotide-gated channels, 11 aquaporin channels, 5 gap-junction channels, and 4 other channels (Table I). Scrutinizing the nematode genome for potassium channel homologues results in the identification of 51 individual potassium channels, which is well in agreement with the conjecture made by Bargmann (23Bargmann C.I. Science. 1998; 282: 2028-2033Crossref PubMed Scopus (718) Google Scholar) and Miller (22Miller C. Genome Biology. 2000; http://genomebiology.com/2000/1/4/REVIEWS/0004PubMed Google Scholar) for these channels in nematodes. However, from this huge family only 19 channels show a high similarity to human potassium channels and are included in Table I. A significant likelihood of an operon organization was assumed if in the nematode genome additional potentially expressed gene(s), organized in the same direction, could be identified within an intercistronic distance of ≤1500 bp. As the intercistronic distance we defined the bp located within the end of the 3′-untranslated region (UTR) of the leading gene and the start of the 5′-UTR of the following gene in the potential operon. Because the information on the UTRs was not available for all genes in question, we used (if the respective information was not available) the genomic sequence (bp distance) between the ORF of the leading gene and the following gene as the putative intercistronic distance. This circumstance is marked in Tables I and II as not known (#) or partially known (*). If a second nematode gene is located beyond 1500 bp the information on the respective nematode gene is not included in Table I. As described by Blumenthal et al. (7Blumenthal T. Evans D. Link C.D. Guffanti A. Lawson D. Thierry-Mieg J. Thierry-Mieg D. Chiu W.L. Duke K. Kiraly M. Kim S.K. Nature. 2002; 417: 851-854Crossref PubMed Scopus (271) Google Scholar) the intercistronic distances between the individual genes in an operon are typically clustered around 126 bp; however, several hundred bp are also described, as in the case of the known operons for the ion channels nmr-2 (944 bp) and the pannexin-1 homologue (1390 bp). For 40 different nematode ion channel genes listed in Table I, a second gene can be found within a distance of 1500 bp, and in 36 cases of these 40 ion channel genes (4 genes show overlapping UTRs, and therefore, no classical intercistronic sequence is defined), the intercistronic sequence is particularly rich in uridines (51.5% U content), which is indicative for intercistronic sequences (24Huang T. Kuersten S. Deshpande A.M. Spieth J. MacMorris M. Blumenthal T. Mol. Cell. Biol. 2001; 21: 1111-1120Crossref PubMed Scopus (45) Google Scholar). For some of the ion channel genes in nematode the operon structure has already been established, and the summary of these proteins is provided in Table II. In addition, in this table we also summarize those nematode ion channel genes with an intercistronic distance of ≤300 bp. The mere intercistronic distance of <1500 bp or even shorter distances of <300 bp are of course no proof that the respective genes are indeed organized in operons. For a fast screening for operons the best method is the identification of SL2 sequences attached to the downstream gene in a putative operon cluster (7Blumenthal T. Evans D. Link C.D. Guffanti A. Lawson D. Thierry-Mieg J. Thierry-Mieg D. Chiu W.L. Duke K. Kiraly M. Kim S.K. Nature. 2002; 417: 851-854Crossref PubMed Scopus (271) Google Scholar). We used RT-PCR to elucidate a possible operon structure of the two non-α subunits of the nicotinic acetylcholine receptor (sodium channel) acr-2/acr-3 and the two transmembrane proteins of the ATP binding cassette (ABC) superfamily of transport proteins mrp-1/mrp-2 gene-doublets. Indeed, we were able to identify SL2 sequences on the acr-3 as well as on the mrp-2 gene (Fig. 1). On the acr-2 and the mrp-1 genes the SL1 sequence was identified. No SL2 sequences could be found on these genes, demonstrating that the acr-2/acr-3 and the mrp-1/mrp-2 genes are indeed organized in operons. Functional Relation between the Operon-embraced Ion Channel Genes—As recently summarized by Blumenthal and Gleason (11Blumenthal T. Gleason K.S. Nat. Rev. Genet. 2003; 4: 112-120Crossref PubMed Scopus (21) Google Scholar), a number of bacterial proteins are clustered in operons, which indeed belong to the same functional pathway (see also Lawrence (5Lawrence J.G. Cell. 2002; 110: 407-413Abstract Full Text Full Text PDF PubMed Scopus (70) Google Scholar)). A much smaller correlation can be found in C. elegans (11Blumenthal T. Gleason K.S. Nat. Rev. Genet. 2003; 4: 112-120Crossref PubMed Scopus (21) Google Scholar). Thus far, as for ion channels, only two examples are known in which functionally related proteins are encoded within the same operon, i.e. two subunits of the acetylcholine receptor channel, i.e. des-2 and deg-3 (12Treinin M. Gillo B. Liebman L. Chalfie M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15492-15495Crossref PubMed Scopus (79) Google Scholar), and the ICln ion channel (icl-1; the three letter codes extended by a number are the gene names in the nematode genome), which functionally interacts with the Nx protein (6Furst J. Ritter M. Rudzki J. Danzl J. Gschwentner M. Scandella E. Jakab M. Konig M. Oehl B. Lang F. Deetjen P. Paulmichl M. J. Biol. Chem. 2002; 277: 4435-4445Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). Here we propose that scrutinizing the operon structure of the nematode genome could be used to identify partner proteins in an organism as distant from nematode as human. As stated above, we showed that IClnN2, the nematode homologue of the ICln channel, interacts with the nematode Nx protein, and that both proteins are coded within the same operon (6Furst J. Ritter M. Rudzki J. Danzl J. Gschwentner M. Scandella E. Jakab M. Konig M. Oehl B. Lang F. Deetjen P. Paulmichl M. J. Biol. Chem. 2002; 277: 4435-4445Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). According to Table II the human homologue of the nematode IClnN1 and its splice variant IClnN2 is a single gene named CLNS1A, and the human homologue of the nematode Nx is the HSPC038 gene, whose function in human cells is unknown. As shown in Fig. 2a, Nx and HSPC038 have a similar amino acid sequence. However, in contrast to the nematode system, both human genes are not in close genomic vicinity to each other. Whereas the CLNS1A gene in the human genome is located on chromosome 11 at the position 11q13.5, the HSPC038 gene is located on chromosome 8 at position 8q22.3 (Fig. 2b). Nevertheless, as shown in Fig. 2c, both human gene products are able to elicit FRET-based light emission, which indicates a functional interaction between both proteins. The bleaching of YFP-HSPC038 is paralleled by an increase of the hICln-CFP fluorescence to 107.6 ± 2.0% (n = 8) after 10 min. The calculated FRET efficiency is 6.8 ± 1.7% (n = 8) after 10 min and 9.1 ± 2.1% (n = 8) when extrapolated to total YFP bleaching (Equations 1 and 2, respectively, under “Experimental Procedures”). For the calculation of the FRET efficiency the entire area of the cells was taken, which, however, underestimates the local rise of the FRET efficiency. It is shown in Fig. 2c that the functional interaction leading to the FRET signal between the CLNS1A and the HSPC038 gene products is not distributed homogeneously throughout the cell but, in contrast, is most efficient in the areas of the membrane. Operons were first discovered in bacteria, and because it became obvious that genes encoding proteins from the translational as well as from the transcriptional machinery are clustered in operons in these organisms, it was speculated that the operon organization might be an efficient way to co-regulate genes in the same regulatory pathway (25Nomura M. Cell. 1976; 9: 633-644Abstract Full Text PDF PubMed Scopus (35) Google Scholar). This concept was sustained by further experimental evidences and furthermore discovered to also be adopted in higher species such as C. elegans (4Spieth J. Brooke G. Kuersten S. Lea K. Blumenthal T. Cell. 1993; 73: 521-532Abstract Full Text PDF PubMed Scopus (250) Google Scholar, 11Blumenthal T. Gleason K.S. Nat. Rev. Genet. 2003; 4: 112-120Crossref PubMed Scopus (21) Google Scholar). Thus far, for ion channels only two examples are known in which functionally related proteins are encoded within the same operon, i.e. two subunits of the acetylcholine receptor channel and the ICln ion channel (icl-1), which functionally interacts with the Nx protein. Treinin et al. (12Treinin M. Gillo B. Liebman L. Chalfie M. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 15492-15495Crossref PubMed Scopus (79) Google Scholar) describe two related subunits of the acetylcholine receptor channel (des-2 and deg-3) that are similar in sequence, encoded by the same operon, and in which only the simultaneous expression of both proteins generates a current in Xenopus laevis oocytes. We recently used the operon structure in nematodes to identify a protein (Nx), which is able to modify the current elicited by the IClnN2 ion channel protein (6Furst J. Ritter M. Rudzki J. Danzl J. Gschwentner M. Scandella E. Jakab M. Konig M. Oehl B. Lang F. Deetjen P. Paulmichl M. J. Biol. Chem. 2002; 277: 4435-4445Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar). It is important to note that these two proteins, unlike the two subunits of the acetylcholine receptor channel, are not related in terms of their sequence. Therefore, we provided the first evidence that by scrutinizing the operon structure of target genes, functional partner-proteins, which are not structurally related, can also be identified in eukaryotes. This finding led to the hypothesis that this approach could be used as a more general gene-finding tool, which is also able to predict functional partner-proteins in the human system. Therefore, we examined, first, the vast family of human ion channels to determine whether or not they could be organized in operons in the nematode genome and, second, if the predicted proteins in the human system really do functionally interact. As can be evidenced by the information given in Table II, for two ion channels (des-2/deg-3 and icl-1/Nx) an operon structure and functional interaction was already reported (11Blumenthal T. Gleason K.S. Nat. Rev. Genet. 2003; 4: 112-120Crossref PubMed Scopus (21) Google Scholar). For additional five ion channels, an operon structure can be assumed according to the data provided by the WormBase (see also Table II; the respective operon names are: CEOP5320, CEOPX084, CEOPX032, CEOP5436, and CEOP1108). Here we provide experimental evidence that for two additional candidate genes listed in Table II (the two non-α subunits of the nicotinic acetylcholine receptor acr-2/acr-3 and the two transmembrane proteins of the ATP binding cassette (ABC) superfamily of transport proteins mrp-1/mrp-2), an organization in operons can be assumed, proving that the information provided in the two tables can indeed help to identify new possible partner proteins for ion channels. In addition we showed that, indeed, the functional interaction of operon-packed genes in nematodes (icl-1/Nx) can be projected onto the human system, leading to the identification of partner proteins in human cells that indeed do functionally interact (CLNS1A/HSPC038). This indicates, that the “operon based partner protein quest (OBPQ),” suggested by us, could be a very efficient gene-finding strategy, which, apart from ion channels, could also be extended to other functional systems. We thank Dr. T. Blumenthal for reading the manuscript.