Downloaded from genome.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press Selective AmpliGfi atLon of DNA Fragments Coding for the n of Factors IF2 and EF-Tu, Two G Proteins from the Myxobacterium Stigmatella aurantiaca L. Bremaud, B. Derijard, 1 and Y. Cenatiempo Institut de Biologie Mol~culaire et d'Ing~nierie G6n~tique, Centre National de la Recherche Scientifique (CNRS) URA1172, Universit~ de Poitiers, 86022 Poitiers Cedex, France Two DNA fragments of the genome of Selective amplification of DNA frag- bacterial species, because under starva- the myxobacterlum $tigmotella au- ments is desired as the first step in gene tion these prokaryotic cells undergo a rantlaca were selectively amplified by cloning experiments, especially when developmental cycle leading to the for- PCR. These fragments encode a seg- the use of homologous probe is required. mation of particular structures termed ment of the G domain of transla- This situation occurs when dealing with fruiting bodies. (7'8) Multicellular devel- tional initiation factor IF2 and elon- a family of proteins with members that opment of myxobacteria is sustained by gation factor EF-Tu, two GTP-binding display strong similarities in their pri- signal exchanges presumably transduced proteins. This was made possible by mary structure. In this respect, an inter- by G proteins, as in most eukaryotes. (9'1~ carefully designing the primers for esting example is provided by the widely Two soluble G proteins, translational ini- this reaction to avoid the amplifica- spread GTP-binding proteins (G pro- tiation IF2 and elongation factor EF-Tu, tion of every G-domain-encoding re- teins). may also play a role, directly or indi- gion of the genome. The sequence of The members of the G protein super- rectly, in this developmental process, via two pairs of primers was deduced family (1) are involved in diverse biolog- the selection and translation of particu- from highly conserved regions, ical processes such as transmembrane lar mRNAs. namely G1 and G 3 and/or their viol- signaling, control of proliferation and IF2 and EF-Tu are present in different hal amino acids, within each subfam- differentiation, and initiation and elon- forms in bacteria, and they arise from ily (initiation and elongation factors, gation of protein synthesis in prokary- different processes. In Escherichia coli, respectively) of GTP-binding pro- otes and eukaryotes. (z--s) In spite of these one gene, infg, directs the synthesis of teins. On the basis of the expected different roles, similar molecular mech- the various forms of IF2, caused by the size, one band was selected in each anisms occur for GTP binding (active presence of several translational initia- experiment, cloned into a vector, and form) followed by its hydrolysis into tion sites on infB mRNA, (11) whereas EF- sequenced. This showed unambigu- GDP (inactive form). Conserved primary Tu appears as two polypeptides bearing a ously after comparison analysis that and presumably tertiary protein struc- different amino acid at the carboxyl ter- they belong to the IF2 and EF-Tu tures were identified, especially within minus. The latter are coded by two genes, respectively. This strategy the G domain bearing the most con- genes, tufA and tufB, differing only in 13 seems suitable for the amplification served motifs. (L6) positions out of a total of 1185 nucle- of a segment of any gene coding for a It is now accepted that at least four otides; 8 of the 13 positions are located G protein from any origin. conserved regions, termed G~ to G4, are within the region encoding the G do- found throughout all the superfamily, main.(12,13) but with an even higher homology In this report we show that it is pos- within subgroups such as translational sible to amplify specifically a segment of initiation factors, elongation factors, the genes for IF2 and for EF-Tu. This was small eukaryotic GTPases like Ras pro- performed by designing different pairs teins, and subunits of signal-transducing of primers, taking into account the most G proteins. Consensus sequence motifs conserved sequences, but also the main have been proposed: GXXXXGKS#I', differences among subfamilies, within D-Xn-T, DXXG, and NKXD within G~, and at the border of the consensus re- 1Present address: Howard Hughes Medical Institute G2, G3, and G4, respectively. (~) gions G~ and G3. Research Laboratories, University of Massachusetts Medical Center, II Biotech Park, Worcester, Massachu- We began studying GTP-binding pro- The amplified DNA fragments, which setts 01605 USA. teins in Stigmatella aurantiaca, a myxo- were identified unambiguously by se- 3:195-1999 by Cold Spring Harbor Laboratory Press ISSN 1054-9803/93 $5.00 PCR Methods and Applications 195 Downloaded from genome.cshlp.org on October 3, 2021 - Published by Cold Spring Harbor Laboratory Press quencing, will now be used as homolo- appended with EcoRI and BamHI restric- fication with Tu primers (tuf G1 and gous probes to clone the corresponding tion sites at their 5' terminus. G'3a ) was almost identical, except that genes of S. aurantiaca. the annealing temperature was dropped to 56~ PCR Amplification Twenty microliters of each reaction MATERIALS AND METHODS PCR was performed with a Gene ATAQ mixture was loaded onto a 2% agarose Bacterial Strains and Pharmacia Thermocycler. Amplification gel. After electrophoresis, amplification Growth Conditions of the desired segments was performed products were visualized by ethidium in a 50-p4 reaction mixture containing 1 bromide staining. PCR products were S. aurantiaca DW4 was obtained from D. ~g of S. aurantiaca genomic DNA; 50 purified using the freeze-squeeze White (Indiana University, Blooming- pmoles of each primer; 200 p.M each of method. (~7) After extraction and ethanol ton). This strain was grown at 30~ in 1% dATP, dTTP, dCTP, and dGTP; Taq DNA precipitation, amplification products Bactocasitone (Difco), with 8 mM polymerase buffer [10 mM Tris-HCl (pH were digested with BamHI (BRL) and MgSO4, and harvested at -4.108 cells/ 9), 50 mM KC1, 1.5 mM MgCl2, 0.1% Tri- EcoRI (BRL). DNA fragments were sub- ml. E. coli TG1 was grown at 37~ in 2x ton X-100; and 1.25 units of Taq DNA cloned into M13mp19, digested previ- YT medium. (~4) polymerase. ously with the same restriction enzymes, Phenol-extracted S. aurantiaca ge- and transformed into E. coli TG1. The DNA Preparation nomic DNA was denatured for 5 min at single-stranded template was sequenced 100~ in the presence of both primers, by the dideoxy chain termination Genomic DNA of S. aurantiaca DW4 was overlaid with 50 ~1 of mineral oil method, (18) using a fluorescent universal extracted by the procedure described by (Sigma), and then cooled rapidly on ice. M13 primer and T7 DNA polymerase. Se- Starich and Zissler. (~s) Next, dNTPs, Taq DNA polymerase quencing was performed with an auto- buffer, water, and Taq DNA polymerase mated sequencer (A.L.F., Pharmacia). were added to the mineral oil and Primer Synthesis dropped by a brief centrifugation. Degenerate oligonucleotides used for Amplification with IF2 primers (infB Computer Analysis of PCR were produced by the phosphora- G1 and G3a) was carried out as follows: DNA Sequences midite method as described previ- Each cycle consisted of denaturation at Computer analysis of PCR fragments was ously, (16) using a Gene Assembler Syn- 94~ for 30 sec, annealing at 65~ for 30 performed using DNAsis software (Hita- thesizer (Pharmacia LKB). Upstream and sec, and extension at 72~ for 1 rain. chi). downstream primers, respectively, were This cycle was repeated 30 times. Ampli- RESULTS Primer Design The first stage of this work was to design G 1 G 3 two sets of primers to amplify specifi- GHVDHGK, .DxPG .......... VVTIMGH .................. ITFLDTP .............. cally a segment of the genes infB and tuf of S. aurantiaca coding for two GTP- I 162 bp binding proteins, repectively, IF2 and .......... NIGTIGHV .................... ~ ..... NMITGAAQ .... EF-Tu. As shown in Figure 1A, to obtain --tuf G-~, <--tuf G~- i 255 bp an amplification of a G-protein-encod- ing gene segment, we took advantage of (1) conserved amino acid sequences G 1 infB G1 5' CGGAATTC GTG GTG ACC ATC ATG GG~ CAC 3" and G3 to design the primers inf8 G1, C C G T infB G3a, and tufG 1, which overlap those consensus sequences and upstream tuf GI 5' CGGAATTC AAC ATC GGC ACG ATC GGC CAC GTT 3' C amino acids; and (2) a conserved stretch of 8 amino acids, located downstream infB G3a 5' GCGGATCC GGG GGT GTC ~AG GAA GTT GAT 3' c c ~ G C from G3, in the Tu subgroup of the G-protein family to design tuf G'3a. tuf G'3a 5' GCGGATCC CTG ~GC ~GC GCC ~GT GAT CAT GTT 3' G G G A We had to take into account the GC content (-70%) of the myxobacterial ge- nome, especially at the third position of FIGURE 1 Conserved amino acid sequences in the G domain of IF2 and EF-Tu (A) and nucleotide codons where it can be as high as sequence of primers used in PCR (B). (A) G1 and G 3 represent two consensus motifs found 90%. (19) Moreover, the high GC content throughout G proteins; broken lines correspond to IF2 and EF-Tu, respectively. Only those amino of the primers could theoretically lead to acids taken into account to design the primers (infB G1, infB G 3, tuf G1, and tuf G'3a) are dis- played. The size (in bp) of the expected PCR fragments is also indicated. (B) The nucleotide secondary structures, because of self-an- sequence of degenerate primers is described. At degenerated positions, in the absence of any nealing, and to primer-dimer forma- other indication, each base represents 50% of the mixture, whereas underlined or doubly un- tion.