| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, New Jersey 08543, USA
2 Bristol-Myers Squibb Pharmaceutical Research Institute, Pennington, New Jersey 08534, USA
3 Bristol-Myers Squibb Pharmaceutical Research Institute, Wallingford, Connecticut 06492, USA
Reprint requests to: Keith L. Constantine, Bristol-Myers Squibb Pharmaceutical Research Institute, P.O. Box 4000, Princeton, NJ 08543; e-mail: keith.constantine{at}bms.com; fax: (609) 252-6030.
(RECEIVED February 1, 2005; FINAL REVISION March 11, 2005; ACCEPTED March 12, 2005)
 |
Abstract |
|---|
 TOP Abstract IntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
Keywords: bioinformatics; biophysical methods; conserved essential bacterial gene; methyltransferase; mutagenesis; nuclear magnetic resonance; protein modeling; Streptococcus pneumoniae
Abbreviations: CEG, conserved essential gene • cfe, conserved essential gene for expression • HSQC, heteronuclear single-quantum coherence • ITC, isothermal titration calorimetry • Kd, dissociation constant • HMMs, hidden Markov models • NMR, nuclear magnetic resonance • NOE, nuclear Overhauser effect • NOESY, nuclear Overhauser effect spectroscopy • PDB, Protein Data Bank • RID, residue information data structure • rRNA, ribosomal ribonucleic acid • SAH, S-adenosyl- L-homocysteine • SAM, S-adenosyl-L-methionine • 
m, mixing time • Tm, melting temperature (transition midpoint) • 2D, two-dimensional • 3D, three-dimensional • 4D, four-dimensional.
Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.051389605.
|
Introduction |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
The functional annotations of the 113 CEGs from the above gene knock-out campaign encompassed a variety of cellular roles, with the most common being "unknown function." However, clues to the functional roles were obtained for some of the genes with "unknown function," such as a weak local similarity of cfe88 (CEG for Expression 88, the subject of this article) to S-adenosyl-L-methionine (SAM)–dependent methyltransferases. These enzymes are involved in methyl group transfers to a wide range of macromolecular substrates. A well-studied bacterial methyltransferase is ErmC', which confers antibiotic resistance against the macrolide-lincosamide-streptogramin type B antibiotics via methylation of an adenine residue in the bacterial 23S rRNA (Eady et al. 1990; Weisblum 1995). The genetic regulation of various erm genes has been investigated in a number of Gram-positive bacteria (for review, see Leclercq and Courvalin 2002), and structures of ErmC' methyltransferase have been determined (Bussiere et al. 1998; Schluckebier et al. 1999).
In this report, we provide a more detailed analysis of the cfe88 gene product, hereafter referred to as CFE88.CFE88 is a 227-residue protein with no significant (
30%)amino-acid sequence similarity to any proteins with publicly available 3D structures (Berman et al. 2000). Combining bioinformatics analysis, protein threading (computational fold recognition), and experimental NMR data has afforded a structure-guided sequence alignment between CFE88 and ErmC' methyltransferase. Subsequently, an initial unrefined model of CFE88 has been built by using this alignment. CFE88 is shown to bind to S-adenosyl-L-homocysteine (SAH) by NMR and other biophysical methods. The NMRdata show that critical conserved residues, located in or near the predicted active site, are perturbed by SAH binding. This result is consistent with the inferred methyltransferase function of CFE88. Genetic studies further suggest that CFE88 is an attractive target for antibiotic development.
|
Results |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
. Since only one related protein sequence was found in each of the examined bacterial genomes, all of the sequences shown in Figure 1 are likely to be orthologs of CFE88. The alignment demonstrates a few regions of conservation, most notably those around His26, Glu46, and Gly94. Based on nearly full-length sequences, the individual regions of conserved sequence in CFE88 and its orthologs match conserved segments in proteins with PFAM designations pfam04816, also known as DUF633 (domain of unknown function), as well as COG2384 (predicted SAM-dependent methyltransferase).
|
). Of the top 10 threading hits, four were from the same 3D fold: The SCOP superfamily is SAM-dependent methyltransferases, and in the CATH Protein Structure Classification Database the superfamily is numbered 3.40.50.150
[EC]
(transferase/methyltransferase). The protein fold filter was then applied, and the top 20 fold family members are shown in Table 1A. The best scoring protein has 19%identity over 150 aligned amino acids. The fold family contains amajority of methyltransferase proteins. To better understand the putative biochemical function of CFE88, the members of these fold super families were examined by using the enzyme nomenclature (E.C.) for methyltransferases, E.C. 2.1.1. (transferases, transferring one-carbon groups, methyltransferases). Table 1B shows the 10 families of methyltransferases that are members of the fold super family. All 10 family members use SAM to provide the methyl group for transfer. Conformational similarities between methyltransferase domains of superfamily members can be seen in Figure 2, which shows the top-ranked methyltransferase L-isoaspartate O-methyltransferase (green) overlayed with two other highly ranked family members, catechol O-methyltransferase (blue) and ErmC' methyltransferase (red). This fold superfamily has the Rossmann fold, which contains a core 
-sheet of seven strands surrounded by helices forming a three-layer (
) sandwich.
|
|
shows a 2D 1H-15N heteronuclear single quantum correlation (HSQC) spectrum of an ILV-labeled CFE88 sample. The peaks in the HSQC spectrum are well dispersed, indicative of a folded protein.
|
, and 13C assignments. Peak lists derived from 2D HSQC and 3D HNCO, HN(CA)CO, HNCA, HN(CO)CA, HN(CA)CB, and HN(COCA)CB spectra were supplied as input to the AUTOASSIGN program (Zimmerman et al. 1997), an expert system for automatically assigning protein backbone atoms based on tripleresonance NMR experiments. AUTOASSIGN assigned 210 out of 220 backbone amide resonances by using intraresidue and sequential correlations involving the backbone 13CO, 13C, and 13C resonances. In order to confirm and extend the backbone assignments, data were collected by using a sample of uniformly 1H/13C/15N-labeled CFE88. 2D HSQC and 3D HNCA, HN(CA)HA, CBCA(CO)NH, and HBHA(CO)NH spectra were acquired. These data were used to establish residue information data structures (RIDs) (Friedrichs et al. 1994; Constantine et al. 1997). The 13C assignments for uniformly 1H/13C/15N-labeled CFE88 were initially estimated by using the approach of Farmer and coworkers (Venters et al. 1996). Intraresidue and sequential 13C and 1H correlations were used to establish RID–RID links. All of the assignments obtained by AUTOASSIGN were confirmed, and additional backbone amide assignments were obtained. In summary, backbone amide assignments were obtained for all nonproline residues except Ile2, Glu81, and Ala146. 1H assignments were obtained for 218 out of 227 residues, and 1H assignments were obtained for 141 out of 210 nonglycine residues.
A partial set of side-chain assignments was obtained from data collected on uniformly 1H/13C/15N-labeled and ILV-labeled samples of CFE88. For the former samples, 3D H(C)(CC)(CO)NH, 3D HCCH-TOCSY, and 4D HCNH-NOESY (m= 80 msec) spectra were collected. For the latter samples, 3D H(C)(CC)(CO)NH, 3D (H)C(CC)(CO)NH, 4D HNNH-NOESY (m=200 msec), 4D HCNH-NOESY (m=90 and 300 msec), and 4D CHCH-NOESY (m=90 and 220 msec) spectra were recorded. All of these spectra were analyzed interactively. The methyl 1H and 13C resonances for all alanine, threonine, leucine, isoleucine, and valine residues have been assigned, and limited side-chain 1H, 13C, and 15N assignments have been obtained for other residue types. Tables of assignments for protonated CFE88 are supplied as Supplemental Material.
Secondary structure, alignment to ErmC', and modeling
13C and 13C NMR assignments immediately provide information on local protein backbone conformation (Spera and Bax 1991; Wishart and Sykes 1991). Differences are computed between the observed 13C and 13C chemical shifts and random-coil values to produce 
13C and 13C values, respectively. CC defined as CC=13C–13C (Metzler et al. 1993), is positive for residues in helical conformations and is negative for residues in extended (-strand) conformations. By taking the difference, biases introduced by any possible chemical-shift referencing errors are largely eliminated (Constantine et al. 1997). Figure 4 shows the CC values determined for CFE88. These data demonstrate that CFE88 has a mixed helical/extended fold for residues 1–163 and a primarily helical fold for residues 164–227. Most of the helices are verified by medium-range NOEs (see Supplemental Material). As discussed in more detail below, the NMR data on CFE88 are consistent with a 163-residue N-terminal domain that is similar to that observed in the Erm methyltransferases, and with a mainly helical 64-residue C-terminal domain that is similar in length to the C-terminal domains of the Erm methyltransferases.
|
m=200 msec) was recorded on a uniformly 2H/13C/15N-labeled sample, after replacement of 2H with 1H at the exchangeable sites. This spectrum contains 635 manually verified peaks. The 1HN-1HN NOEs that have been assigned with high confidence are reported in the Supplemental Material. Long-range NOE data identify a five-stranded parallel -sheet, as shown in Figure 5. This comprises the major portion of the seven-stranded -sheet expected for the fold superfamily.
|
Most of the helical and extended regions of CFE88 can be approximately matched to -helices and -strands in ErmC' (Fig. 6). The secondary structure correspondences were used to manually refine a sequence alignment between ErmC' and CFE88 (Fig. 6). This refined alignment served as input to the LOOK program (Levitt 1992) for generating a model of CFE88 (Fig. 7). In this alignment, 39 out of 227 CFE88 residues (17.2%) are aligned to identical residues in ErmC'. Based on secondary structure and protein threading results, CFE88 residues 163–164 appear to be located at a domain boundary. It should be noted that the initial CFE88 model has not been explicitly refined by using NMR-derived restraints, and that many of the details regarding local backbone conformations, starting and ending residue positions of the elements of secondary structure, etc., are not expected to be highly accurate. This approximate model of the overall fold of CFE88 is intended to facilitate the interpretation of additional data.
|
|
below) to CFE88 has yielded additional evidence for the biochemical function of CFE88. SAH is an endogenous inhibitor of methyltransferases that utilize SAM as a methyldonating substrate (cofactor). Enzymes that utilize SAM also include hydrolases and oxidoreductases; however, our results are consistent with a methyltransferase function.
|
. SAH was titrated into an 15N-labeled sample of CFE88 (see Materials and Methods), whereupon numerous peaks in the 2D 1H-15N HSQC spectrum have been observed to display extensive chemical-shift changes and/or line broadening effects. These perturbations, mapped onto the CFE88 model in Figure 8B, are consistent with site-specific binding of SAH. All of the perturbed residues are in the N-terminal domain of CFE88 (see Supplemental Material). Comparing panels A and B of Figure 8 clearly reveals that the SAH-induced perturbations of CFE88 residues are centered on the SAH binding site predicted by the ErmC'-based model. This result provides strong support both for the overall structural alignment between CFE88 and ErmC' and for identical cofactor specificities. Therefore, we conclude that CFE88 is a two-domain enzyme, with an N-terminal catalytic (methyltransferase) domain that utilizes SAM as a cofactor, and with a C-terminal domain that is probably involved in macromolecular recognition.
|
shows the results of titrating a concentrated solution of SAH (700 µM) into a dilute solution of CFE88 (initial concentration 28.8 µM). Fitting the experimental data to a single-site binding equation yielded the following parameters: Kd (dissociation constant)=7.3±0.2 µM, N (number of sites)=0.985±0.006, and H (observed enthalpy change)=–14.9±0.1 kcal/mol. These results (enthalpy-driven single-site binding) are consistent with the site-specific binding observed by NMR.
|
); (2) based on the initial CFE88 model, these residues are expected to be in or near the SAH binding groove (Fig. 7); (3) both residues are hydrophilic and likely to be solvent and ligand accessible; and (4) histidines, in particular, are often involved in the catalytic mechanisms of enzymes. His26 was changed to a tryptophan, producing the H26W mutant. Glu46 was changed to an arginine and to a tryptophan, producing the E46R and E46W mutants, respectively. Two of the three mutants (H26W and E46R) were characterized by NMR (data not shown). Both are properly folded. For the H26W mutant, SAH induces extensive chemical-shift changes and line-broadening effects similar to those observed for the wild-type protein. In contrast, the NMR data indicate extremely weak binding of SAH to the E46R mutant. At a ligand/protein concentration ratio of ~1.5:1, very minor chemical-shift changes are observed only for Leu6 and Leu29 of the E46R mutant in the presence of SAH. In the case of the wild-type protein, the 1H-15N HSQC peaks for these residues are broadened beyond detection by SAH binding.
Biophysical studies corroborate and extend the NMR findings. By using Thermofluor technology, SAH was observed to bind both to the wild-type CFE88 protein and to the H26W mutant, as reflected by changes in the protein denaturation midpoint (melting) temperature upon exposure to SAH (Table 2). SAH appears to bind with higher affinity to the H26W mutant than it does to the wild-type protein, based on a larger change in melting temperature (Table 2). SAH does not induce a significant change in the melting temperature for the E46R or E46W mutants (Table 2), consistent with the NMR data indicating minimal SAH binding to the E46R mutant.
|
). In both cases, the transformations did not result in viable colonies, while control transformations behaved as expected, indicating that His26 and Glu46 both make critical contributions to the essential function of CFE88. Polarity studies of the original CFE88 knock-out (Thanassi et al. 2002) indicate that it is nonpolar; i.e., it does not appear that a downstream gene was responsible for the essential phenotype. The H26W and E46R knock-in mutants confirm and extend these results.
|
|
Discussion |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
SAH is an analog of SAM, a common methyl-donating substrate (cofactor) for methyltransferase reactions. SAH binds to the predicted methyltransferase activesite cleft, supporting the hypothesis that CFE88 is a methyltransferase. In addition, mutations targeting the putative active-site cleft of CFE88 were designed, produced, and characterized. Mutating Glu46 to either tryptophan (E46W) or arginine (E46R) greatly reduces SAH binding affinity, indicating that this residue is important for SAH/SAM recognition. In the case of ErmC' methylytransferase, the side-chain of residue Glu59, which aligns to CFE88 Glu46 (Fig. 6), hydrogen bonds directly to the ribose hydroxyls of SAH (chart 1) and SAM (Schluckebier et al. 1999). Our results suggest that Glu46 plays a similar role in cofactor recognition by CFE88. An acidic residue at this position is conserved across various methyltransferase families.
Mutation of the CFE88 residue His26 to a tryptophan (H26W) does not hinder SAH binding. While cofactor binding does not appear to depend on His26, the genetic cross-in studies suggest that this residue may play an important and possibly unique role in the function of CFE88 and its orthologs. While strictly conserved in the orthologs of CFE88 (Fig. 1), His26 is not a conserved residue in other methyltransferase families. The corresponding residue in ErmC' (K41) is not a conserved site in the Erm methyltransferases; a histidine at position 41 does occur in at least one member of the Erm methyltrnasferase family (see Maravic et al. 2003, Fig. 1). We note that, based on the X-ray structures of ErmC' with bound SAH or SAM (Schluckebier et al. 1999), a tryptophan at position 41 would not be expected to block cofactor binding in the case of the ErmC' methyltransferases. His26 may play a role in the catalytic chemistry of CFE88 and its orthologs, or it may be involved in substrate binding.
NMR data, Thermofluor Tm values (Table 2), and thermal melting profiles (data not shown) indicate that all three mutants are properly folded. Importantly, the H26W and E46R mutations are both apparently lethal when crossed back into bacteria. (E46W was not examined by bacterial knock-in studies.) In summary, based on the CFE88 mutant studies, it appears that the essential activity of CFE88 requires a functional methyltransferase active site. As far as we are aware, this is the first indication of an essential methyltransferase activity in bacteria. It should be noted that the bacterial enzyme SAM:tRNA (uracil-5-) methyltransferase was found to be essential for bacterial survival (Persson et al. 1992). However, evidence was present indicating that the uracil methylation activity, localized to the C-terminal methyltransferase domain, is not the essential function of this protein. The investigators further propose that SAM:tRNA (uracil-5-) methyltransferase contains a second RNA-modifying activity localized to the N-terminal domain and that this activity is essential.
In the cases of CFE88 and the Erm methyltransferases, the methyltransferase activity is localized to the N-terminal domains, and the C-terminal domains of these enzymes are probably involved in noncovalent macromolecular recognition processes, rather than a second covalent modification activity. The observation that the C-terminal domains of CFE88 and its orthologs are not similar to any other known sequences suggests that the putative macromolecular recognition specificities of CFE88 and its orthologs are distinct from other methyltransferases.
Recently, the effects of single-site mutations on the in vivo function and in vitro activity of ErmC' methylytransferase have been reported (Maravic et al. 2003). Eight residues were selected for mutation to alanine. Surprisingly, of the eight residues mutated, only Tyr104 was found to be essential to the catalytic activity of ErmC'. This ErmC' residue aligns to Gly94 in CFE88 (Fig. 6). Gly94 and surrounding CFE88 residues are perturbed by SAH binding (see Supplemental Material), consistent with active site localization of these residues. ErmC' residues Lys41 and Glu59, which align to His26 and Glu46 of CFE88 (Fig. 6), were not mutated in the study by Maravic et al. (2003). In the Erm methyltransferase family, the glutamate residue at position 59 (ErmC' sequence) is strictly conserved (Maravic et al. 2003).
Overall, our results strongly support the identification of CFE88 as a methyltransferase with structural similarity to the Erm family of bacterial methyltransferases. It has previously been shown that CFE88 is essential for bacterial survival (Thanassi et al. 2002), and our mutational data suggest that the essential function of CFE88 involves methyltransferase activity.
Small-molecule inhibitors of Erm methyltransferases have been described (Hajduk et al. 1999; Hanessian and Sgarbi 2000), indicating that this is a class of enzymes for which small-molecule inhibitors can be developed The same is likely to be true for CFE88 and its orthologs. Therefore, CFE88 appears to be a potential target for the development of novel antibacterial compounds. Furthermore, identification of the natural substrate(s) of CFE88 should provide an understanding of the biological role played by this enzyme and may reveal a novel mechanism essential for bacterial survival.
|
Materials and methods |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
Protein threading was carried out by using ProHit [P]rofessional (version 1.8.1, ProCeryon BioSciences). The sequence of CFE88 was compared against all known protein folds from the PDB (Berman et al. 2000). The results were analyzed as previously described (Flockner et al. 1997). In order to add value to the threading results, the ranked threading hits were analyzed by protein fold/domain classifications using SCOP (Structural Classification of Proteins) (Murzin et al. 1995) and CATH Protein Structure Classification Database (Orengo et al. 1997). The World Wide Web version (http://www.chem.qmul.ac.uk/iubmb) of enzyme nomenclature was used to characterize the fold family members based upon enzymatic function.
Cloning, expression, and NMR sample preparation
Recombinant CFE88 proteins with a six-residue C-terminal His tag were prepared by first subcloning the cDNA sequence into the pET28b expression vector and then transforming Escherichia coli BL21(DE3) cells with that vector. The bacterial growth media contained [1,2-13C2, 99%]-sodium acetate (Isotec) and [15N, 99%] ammonium sulfate (Isotec) as the sole carbon and nitrogen sources, respectively. To produce a uniformly 2H/13C/15N-labeled sample, cells that can tolerate high levels of 2H2O (>90%) were selected by published methods (Venters et al. 1995). Samples with 1H/13C/15N leucine, isoluecine, and valine residues incorporated into an otherwise uniformly 2H/12C/15N-labeled protein were prepared using methods described previously (Metzler et al. 1996).
Cells were grown to an OD600 of ~1.0 and then induced with 0.4 mM isopropyl-, D-thiogalactopyranoside for 5 h at 37°C. After harvesting, the cells were resuspended in 30 mL (per L of culture) of 50 mM Tris-HCl and 150 mM NaCl (pH 8.0; disruption buffer), sonicated, and clarified in a Sorval centrifuge at 16,000 rpm for 20 min. The supernatant was applied on a Ni2+ resin (3 mL of gel per L of culture). After washing, the protein was eluted with 0.5 M imidazole in the disruption buffer. Concentrated eluted protein was then applied on to a Superdex 75 26/60 column equilibrated with the disruption buffer. The protein peak was then concentrated into NMR buffer containing 50 mM d4-imidazole, 150 mM KCl, 7% 2H2O (pH 7.0). The following samples of wild-type CFE88 were prepared for NMR studies: ILV-labeled (0.2 mM and 0.6 mM), uniformly 1H/13C/15N-labeled (1.0 mM), and uniformly 2H/13C/15N-labeled (0.5 mM). In addition, uniformly 15N-labeled samples (0.2 mM each) of the H26W and E46R mutants were prepared.
NMR spectroscopy
All NMR experiments were recorded at 33°C on 600-MHz Varian Inova or 600-MHz Varian Unity Plus spectrometers using 5 mm 1H-observe, 13C-15N triple resonance room temperature probes equipped with either triple- or single-axis (z) pulsed-field gradients. All spectra were processed and analyzed with a modified version of the FELIX program (Hare Research, Inc.; M.S. Friedrichs, unpubl.).
Backbone atom resonance assignments based on intraresidue and sequential 13CO, 13C, and 13C correlations were obtained from data collected on the 0.5 mM uniformly 2H/13C/15N-labeled sample. The 2D 1H-15N spectra were acquired with the sequence of Kay et al. (1992), modified to incorporate water flip-back. Modified versions (Constantine et al. 1997) of published 3D triple-resonance pulse sequences were used to collect data. The 3D HNCO spectrum was recorded with a modified version of the basic sequence of Kay et al. (1994). For the 3D HNCA, HN(CA)CB, HN(CO)CA, and HN(COCA)CB spectra, modified versions of the basic sequences of Shan et al. (1996) were used. The 3D HN(CA)CO spectrum was recorded with a modified version of the basic sequence of Matsuo et al. (1996).
Backbone assignment data were also recorded by using the 1.0 mM uniformly 1H/13C/15N-labeled sample. The 2D HSQC and 3D HNCA were recorded as described above. The 3D HN(CA)HA was recorded with a modified version of the basic sequence of Clubb et al. (1992). The CBCA(CO)NH and HBHA(CO)NH experiments were performed as described by Grzesiek and Bax (1992).
Side-chain assignment data were recorded by using the 1.0 mM uniformly 1H/13C/15N-labeled and 0.6 mM ILV-labeled samples. For both samples, 3D H(C)(CC)(CO)NH (Logan et al. 1992) and 4D HCNH-NOESY (Farmer and Mueller 1994) spectra were recorded by using the cited pulse sequences. For the 1.0 mM uniformly 1H/13C/15N-labeled sample, a gradientenhanced 3D HCCH-TOCSY was also recorded, as described (Kay et al. 1993). For the ILV-labeled sample, 3D (H)C(CC)(CO)NH (Farmer and Venters 1995), 4D HNNHNOESY (Farmer and Mueller 1994), 4D HCNH-NOESY (Farmer and Mueller 1994), and 4D CHCH-NOESY (Vuister and Bax 1993; B.T. Farmer II, unpubl.) were also acquired by using the cited methods.
Homology modeling
After preparing the manual alignment between CFE88 and ErmC' methylytransferase (Fig. 6), a model of CFE88 was constructed starting from the A chain of the 2.4 Å resolution X-ray structure of ErmC' methyltransferase (PDB code 1QAN
[PDB]
) by using the LOOK program (Molecular Applications Group). The homology modeling protocol within LOOK was used to generate the model; this protocol utilizes an automated segment-matching algorithm (Levitt 1992) that is implemented in LOOK’s SegMod module.
Biophysical measurements
ITC experiments were carried out at 25°C in a VP-ITC micro-calorimeter (Microcal Inc.) with CFE88 dialyzed extensively against 20 mM PIPES and 150 mM NaCl (pH 7.0). For the ITC measurements, 7 µL injections of SAH (700 µM), dissolved in the buffer above, were made into the calorimetric cell containing CFE88 (28.8 µM). Data were analyzed by using the single-site binding model in Origin 7 (Microcal) software.
The thermal stability enhancement effect of SAH on CFE88 was measured with a ThermoFluor instrument (Three-Dimensional Pharmaceuticals, Inc.) as described by Pantoliano et al. (2001). The technology measures thermal denaturation curves based on the increase in fluorescence due to preferential binding of the extrinsic, fluorescent, hydrophobic probe 1-anilino-8- napthalene sulfonate (ANS) to the denatured protein. Reactions containing CFE88 (4 µM) in the presence and absence of SAH (100µM)were analyzed in 20 mM PIPES (pH 7.0), 150 mM NaCl, 100 µM ANS, and 0.5%(v/v) DMSO. The 5 µL reaction wells were overlaid with 2.5µL of silicon oil (Sigma) to prevent evaporation. Reactions were monitored under UV illumination in black 384-well plates (Abgene) by increasing temperature in 1°C increments with 60-sec equilibration at each degree from 25°C to 80°C. Thermal denaturation curves were analyzed with ThermoFluor Analysis software (Three-Dimensional Pharmaceuticals, Inc.).
Genetic knock-in experiments
Genetic knock-in studies were performed by transformation of S. pneumoniae with plasmid constructs. These constructs consisted of 300- to 500-bp internal fragments (gene knock-out constructs) of full-length wild type or mutant CFE88 genes (knock-in constructs). Recombinant plasmids were constructed in pEVP3, transformed into S. pneumoniae, and assayed for cell viability as previously described (Thanassi et al. 2002).
|
Electronic supplemental material |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
|
Footnotes |
|---|
|
Acknowledgments |
|---|
|
References |
|---|
|
TOPAbstractIntroductionResultsDiscussionMaterials and methodsElectronic supplemental materialReferences |
|---|
Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T.N., Weissig, H., Shindyalov, I.N., and Bourne, P.E. 2000. The Protein Data Bank. Nucleic Acids Res. 28: 235–242.
Bussiere, D.E., Muchmore, S.W., Dealwis, C.G., Schluckebier, G., Nienaber, V.L., Edalji, R.P., Walter, K.A., Ladror, U.S., Holzman, T.F., and Abad-Zapatero, C. 1998. Crystal structure of ErmC', an rRNA methyltransferase which mediates antibiotic resistance in bacteria. Biochemistry 37: 7103–7112.[CrossRef][Medline]
Clubb, R.T., Thanabal, V., and Wagner, G. 1992. A new 3D HN(CA)HA experiment for obtaining fingerprint HN-H peaks in 15N- and 13C-labeled proteins J. Biomol. NMR 2: 203–210.[CrossRef][Medline]
Cohen, M.J. 1992. Epidemiology of drug resistance: Implications for a post-antimicrobial era. Science 257: 1050–1055.
Constantine, K.L., Mueller, L., Goldfarb, V., Wittekind, M., Metzler, W.J., Yanchunas Jr., J., Robertson, J.G., Malley, M.F., Friedrichs, M.S., and Farmer II, B.T. 1997. Characterization of NADP+ binding to perdeuterated MurB: Backbone atom NMR assignments and chemical-shift changes. J. Mol. Biol. 267: 1223–1246.[CrossRef][Medline]
Davies, J. 1994. Inactivation of antibiotics and the dissemination of resistance genes. Science 264: 375–382.
Eady, E.A., Ross, J.I., and Cove, J.H. 1990. Multiple mechanisms of erythromycin resistance. Antimicrob. Chemother. 26: 461–471.
Eddy, S.R. 1996. Hidden Markov models. Curr. Opin. Struct. Biol. 6: 361–365.[CrossRef][Medline]
Edelman, I., Faigler, S., Mintz, E., Natan, A., and Devereux, J. 1995. Frameshift: A rigorous alignment program for searching protein databases with nucleic acid queries. Genome Sequence and Analysis Conference. Hilton Head, SC.
Farmer II, B.T. and Mueller, L. 1994. Simultaneous acquisition of [13C, 15N] and [15N,15N]-separated 4D gradient-enhanced NOESY spectra in proteins. J. Biomol. NMR 4: 673–687.[CrossRef][Medline]
Farmer II, B.T. and Venters, R. 1995. Assignment of side chain 13C resonances in perdeuterated proteins. J. Am. Chem. Soc. 117: 4187–4188.[CrossRef]
Flockner, H., Domingues, F.S., and Sippl, M.J. 1997. Protein folds from pair interactions: A blind test in fold recognition. Proteins 1: 129–133.
Friedrichs, M.S., Mueller, L., and Wittekind, M. 1994. An automated procedure for the assignment of protein 1HN, 15N, 13C, 1H, 13C and 1H resonances. J. Biomol. NMR 4: 703–726.[CrossRef][Medline]
Grzesiek, S. and Bax, A. 1992. Correlating backbone amide and side-chain resonances in larger proteins by multiple relayed triple resonance NMR. J. Am. Chem. Soc. 114: 6291–6293.[CrossRef]
Hajduk, P.J., Dinges, J., Schkeryantz, J.M., Janowick, D., Kaminski, M., Tufano, M., Augeri, D.J., Petros, A., Nienaber, V., Zhong, P., et al. 1999. Novel inhibitors of Erm methyltransferase from NMR and parallel synthesis. J. Med. Chem. 42: 3852–3859.[CrossRef][Medline]
Hanessian, S. and Sgarbi, P. W. 2000. Design and synthesis of mimics of S-adenosyl-L-homocysteine as potential inhibitors of erythromycin methyltransferase. Bioorg. Med. Chem. Lett. 10: 433–437.[CrossRef][Medline]
Healy, M.D. 2003. Finding homologs to nucleic acid or protein sequences using the FrameSearch program. In Current Protocols in Bioinformatics. (eds. Baxevanis et al.) pp. 3.2.1–3.2.23. John Wiley & Sons, NY.
Higgins, D.G., Thompson, J.D., and Gibson, T.J. 1996. Using CLUSTAL for multiple sequence alignments. Methods Enzymol. 266: 383–402.[Medline]
Hillisch, A., Pineda, L.F., and Hilgenfeld, R. 2004. Utility of homology models in the drug discovery process. Drug Disc. Today 9: 659–669.[CrossRef][Medline]
Homans, S.W. 2004. NMR spectroscopy tools for structure-aided drug design. Angew. Chem. Int. Ed. 43: 290–300.[CrossRef]
Kay, L.E., Keifer, P., and Saarinen, T. 1992. Pure absorption gradient-enhanced heteronuclear single quantum correlation spectroscopy with improved sensitivity. J. Am. Chem. Soc. 114: 10663–10665.[CrossRef]
Kay, L.E., Xu, G.Y., Singer, A., Muhandiram, D.R., and Forman-Kay, J.D. 1993. A gradient-enhanced HCCH-TOCSY experiment for recording sidechain 1H and 13C correlations in H2O samples of proteins. J. Magn. Reson. B 101: 333–337.[CrossRef]
Kay, L.E., Xu, G.Y., and Yamazaki, T. 1994. Enhanced-sensitivity triple resonance spectroscopy with minimal H2O saturation. J. Magn. Reson. A 109: 129–134.[CrossRef]
Leclercq, R. and Courvalin, P. 2002. Resistance to macrolides and related antibiotics in Streptococcus pneumoniae. Antimicrob. Agents Chemother. 46: 2727–2734.
Levitt, M. 1992. Accurate modeling of protein conformation by automatic segment matching. J. Mol. Biol. 226: 507–533.[CrossRef][Medline]
Logan, T.M., Olejniczak, E.T., Xu, R.X., and Fesik, S.W. 1992. Side chain and backbone assignments in isotopically labeled proteins from two heteronuclear triple resonance experiments. FEBS Lett. 314: 413–418.[CrossRef][Medline]
Maravic, G., Feder, M., Pongor, S., Flogel, M., and Bujnicki, J.M. 2003. Mutational analysis defines the roles of conserved amino acid residues in the predicted catalytic pocket of the rRNA:m6A methyltransferase ErmC'. J. Mol. Biol. 332: 99–109.[CrossRef][Medline]
Matsuo, H., Kupce, E., Li, J., and Wagner, G. 1996. Use of selective C pulses for improvement of HN(CA)CO-D and HN(CACO)NH-D experiments. J. Magn. Reson. B 111: 194–198.[CrossRef][Medline]
Metzler, W.J., Constantine, K.L., Friedrichs, M.S., Bell, A.J., Ernst, E.I., Lavoie, T.B., and Mueller, L. 1993. Characterization of the three-dimensional solution structure of human profilin: 1H, 13C and 15N NMR assignments and global folding pattern. Biochemistry 32: 13818–13829.[CrossRef][Medline]
Metzler, W.J., Wittekind, M., Goldfarb, V., Mueller, L., and Farmer II, B.T. 1996. Incorporation of 1H/13C/15N-[Ile, Leu, Val] into a perdeuterated, 15N-labeled protein: Potential in structure determination of large proteins by NMR. J. Am. Chem. Soc. 118: 6800–6801.[CrossRef]
Murzin, A.G., Brenner, S.E., Hubbard, T., and Chothia, C. 1995. SCOP: A structural classification of proteins database for the investigation of sequences and structures. J. Mol. Biol. 247: 536–540.[CrossRef][Medline]
Nikaido, H. 1994. Prevention of drug access to bacterial targets: Permeability barriers and active efflux. Science 264: 382–388.
Orengo, C.A., Michie, A.D., Jones, S., Jones, D.T., Swindells, M.B., and Thornton, J.M. 1997. CATH: A hierarchic classification of protein domain structures. Structure 5: 1093–1108.[Medline]
Pantoliano, M.W., Petrella, E.C., Kwasnoski, J.D., Lobanov, V.S., Myslik, J., Graf, E., Carver, T., Asel, E., Springer, B.A., Lane, P., et al. 2001. High-density miniaturized thermal shift assays as a general strategy for drug discovery. J. Biomol. Screening 6: 429–440.[Abstract]
Persson, B.C., Gustafsson, C., Berg, D.E., and Bjork, G.R. 1992. The gene for a tRNA modifying enzyme, m5U54-methyltransferase, is essential for viability in Escherichia coli. Proc. Natl. Acad. Sci. 89: 3995–3998.
Pozzi, G., Masala, L., Iannelli, F., Manganelli, R., Havarstein, L.S., Piccoli, L., Simon, D., and Morrison, D.A. 1996. Competence for genetic transformation in encapsulated strains of Streptococcus pneumoniae: Two allelic variants of the peptide pheromone. J. Bacteriol. 178: 6087–6090.
Schluckebier, G., Zhong, P., Stewart, K.D., Kavanaugh, T.J., and Abad-Zapatero, C. 1999. The 2.2 Å structure of the rRNA methyltransfease ErmC' and its complexes with cofactor and cofactor analogs: Implications for the reaction mechanism. J. Mol. Biol. 289: 277–291.[CrossRef][Medline]
Shan, X., Gardner, K.H., Muhandiram, D.R., Rao, N.S., Arrowsmith, C.H., and Kay, L.E. 1996. Assignment of 15N, 13C, 13C, and HN resonances in an 15N, 13C, 2H labeled 64 kDa Trp repressor-operator complex using triple-resonance NMR spectroscopy and 2H-decoupling. J. Am. Chem. Soc. 118: 6570–6579.[CrossRef]
Smith, T.F. and Waterman, M.S. 1981. Identification of common molecular subsequences. J. Mol. Biol. 147: 195–197.[CrossRef][Medline]
Sonnhammer, E.L, Eddy, S.R., and Durbin, R. 1997. Pfam: A comprehensive database of protein domain families based on seed alignments. Proteins 28: 405–420.[CrossRef][Medline]
Spera, A. and Bax, A. 1991. Empirical correlation between protein backbone conformation and C and C 13C nuclear magnetic resonance chemical shifts. J. Am. Chem. Soc. 113: 5490–5492.[CrossRef]
Spratt, B.G. 1994. Resistance to antibiotics mediated by target alterations. Science 264: 388–393.
Thanassi, J.A., Hartman-Neumann, S.L., Dougherty, T.J., Dougherty, B.A., and Pucci, M.J. 2002. Identification of 113 conserved essential genes using a high-throughput gene disruption system in Streptococcus pneumoniae. Nucleic Acids Res. 30: 3152–3162.
Venters, R.A., Huang, C.-C., Farmer II, B.T., Trolard, R., Spicer, L.D., and Fierke, C.A. 1995. Hihg level 1H/13C/15N labeling of proteins for NMR studies. J. Biomol. NMR 5: 339–344.[Medline]
Venters, R.A., Farmer II, B.T., Fierke, C.A., and Spicer, L.D. 1996. Characterizing the use of perdeuteration in NMR studies of large proteins: 13C, 15N, and 1H assignments of human carbonic anhydrase II. J. Mol. Biol. 264: 1101–1116.[CrossRef][Medline]
Vuister, G.W. and Bax, A. 1993. Increased resolution and improved spectral quality in four-dimensional 13C/13C separated HMQC-NOESY-HMQC spectra using pulsed field gradients. J. Magn. Reson. ser. B 101: 210–213.[CrossRef]
Walsh, C. 2003. Where will new antibiotics come from? Nat. Rev. Microbiol. 1: 65–70.
CACB; see text) versus the CFE88 residues. 13C
