Logo of prosciprotein sciencecshl presssubscriptionsetoc alertsthe protein societyjournal home
Protein Sci. 2004 Jun; 13(6): 1651–1659.
PMCID: PMC2279982

New enzymes from environmental cassette arrays: Functional attributes of a phosphotransferase and an RNA-methyltransferase


By targeting gene cassettes by polymerase chain reaction (PCR) directly from environmentally derived DNA, we are able to amplify entire open reading frames (ORFs) independently of prior sequence knowledge. Approximately 10% of the mobile genes recovered by these means can be attributed to known protein families. Here we describe the characterization of two ORFs which show moderate homology to known proteins: (1) an aminoglycoside phosphotransferase displaying 25% sequence identity with APH(7″) from Streptomyces hygroscopicus, and (2) an RNA methyltransferase sharing 25%–28% identity with a group of recently defined bacterial RNA methyltransferases distinct from the SpoU enzyme family. Our novel genes were expressed as recombinant products and assayed for appropriate enzyme activity. The aminoglycoside phosphotransferase displayed ATPase activity, consistent with the presence of characteristic Mg2+-binding residues. Unlike related APH(4) or APH(7″) enzymes, however, this activity was not enhanced by hygromycin B or kanamycin, suggesting the normal substrate to be a different aminoglycoside. The RNA methyltransferase contains sequence motifs of the RNA methyltransferase superfamily, and our recombinant version showed methyltransferase activity with RNA. Our data confirm that gene cassettes present in the environment encode folded enzymes with novel sequence variation and demonstrable catalytic activity. Our PCR approach (cassette PCR) may be used to identify a diverse range of ORFs from any environmental sample, as well as to directly access the gene pool found in mobile gene cassettes commonly associated with integrons. This gene pool can be accessed from both cultured and uncultured microbial samples as a source of new enzymes and proteins.

Keywords: protein prospecting, gene cassette, APH, SpoU family, integron

High-throughput gene sequencing has resulted in the large sequence databases which today provide powerful resources for the detection of evolutionary relationships and understanding the organization of basic sets of functional genes (Gaasterland and Oprea 2001). The functional annotation of new genes in the databases often relies on homology, either at the sequence or protein fold levels, to detect orthologs that evolved from a common ancestral gene and retained a similar function (Henikoff et al. 1997). When a new protein has 30% or more sequence identity to a known protein in the structure database, bioinformatics tools allow the structure of the unknown protein to be modeled quite accurately, and useful functional information can be confidently inferred. Due to the reduced chance of success at lower sequence identity, structural genomics initiatives today have the aim of determining the structure of at least one member of all sequence families sharing over 30% sequence identity (Chance et al. 2002). However, these current attempts are particularly limited in scope by their target selection, which is focused on genomes of a few key organisms.

Given that bacteria are a major source of genetic novelty and that less than 1% have been cultured and identified (Pace 1997), a vast pool of genetic diversity remains to be accessed in the search for novel gene families and protein folds. Currently two methods are used to probe the microbial gene pool: (1) “shotgun” cloning of large fragments of DNA, generally of environmental origin, and screening of clones by sequencing or hybridization, and (2) the use of the polymerase chain reaction (PCR) with primers which anneal to conserved sequences within particular genes of interest.

We recently demonstrated a new strategy, cassette PCR, for recovering novel genes as whole open reading frames (ORFs) directly from environmental DNA (Stokes et al. 2001). Although based on PCR, the target for our primers is not the ORF itself, but rather the 59-base element (59-be) DNA recombination sites associated with mobile gene cassettes. Gene cassettes are the smallest mobile elements known and comprise only a gene and a 59-be. When inserted into integrons, cassettes are found as tandem arrays with the recombination sites flanking the genes contained within the cassettes (Hall et al. 1991). Integrons are natural gene acquisition and expression elements first described in relation to mechanisms of antibiotic resistance (Stokes and Hall 1989; Hall et al. 1999), but are now more generally identified as widespread in diverse bacteria within the environment (Nield et al. 2001; Holmes et al. 2003a; Rowe-Magnus et al. 2003).

During an investigation of the organization and distribution of gene cassettes in environmental DNA, we found that the vast majority of ORFs within cassettes did not match known sequences (Stokes et al. 2001), and therefore represented novel genes. However, a small subset of ORFs were found to display some homology (ranging from 20% to 30%) to genes in the sequence databases, allowing some designation of function. Metabolic activities of the potential gene products were varied, but included an aminoglycoside phosphotransferase, an enzyme family known to play a role in conferring antibiotic resistance.

As integrons appear to operate as general gene capture systems for bacterial adaptation (Hall et al. 1999; Holmes et al. 2003b), our method is a unique opportunity to recover novel proteins from within the unexplored mobile gene pool without prior gene sequence data. Here we describe the use of our prospecting tool to identify and isolate two new genes coding for putative members of the aminoglycoside phosphotransferase and RNA methyltransferase families, and their subsequent functional validation.


Isolation of genes of interest

DNA was successfully isolated directly from a variety of soil samples, including those taken from both pristine (Sturt Desert) and industrially contaminated (Balmain Power Station) sites. Using the pair of primers HS287 and HS286, which specifically target gene cassette arrays (Stokes et al. 2001), PCR amplification generated a number of products in the 150 to 1300 bp range. Two products of interest included ORFs coding for members of recognizable enzyme families. (The organization and diversity of cassette-associated gene products is more generally outlined in Stokes et al. [2001].)

One clone selected for study contained an environmental gene cassette, designated EGC034 (accession no. af356540) with an ORF of 346 codons (orf346_Pu5). This environmental cassette was cloned to create the plasmid pMAQ678. The translated protein from the encoded ORF has a theoretical mass of 39.1 kD and pI of 5.52, and was identified from sequence homology to be a likely member of the aminoglycoside phosphotransferase (APH) enzyme family. Several APH-like proteins (predominantly from bacterial sources) were found to have ~20% identity to this new sequence with low expectation (E) scores (E-value < 10−20) in PSI-BLAST (Altschul et al. 1997), consistent with them having a significant sequence relationship. The highest homology (25% identity) was found to hygromycin B phosphotransferase [or APH(7″)] from Streptomyces hygroscopicus (Zalacain et al. 1986). The sequence search results also included macrolide 2′-phosphotransferase I (an erythromycin resistance determinant) from Escherichia coli (Noguchi et al. 1995), with somewhat lower homology (23% identity). A search for protein folds consistent with the sequence revealed a strong match to the only known crystal structure of an APH enzyme, namely APH(3′)-IIIa from Enterococcus faecalis (Hon et al. 1997). The gene orf346_Pu5 isolated from a desert soil sample was therefore likely to code for a putative aminoglycoside phosphotransferase, and was named eAPH. A structure-based alignment for these related enzyme sequences constructed by threading each onto the tertiary fold of APH(3′)-IIIa (Russell and Torda 2002) is shown in Figure 1A.

Figure 1.Figure 1.
Structure-based alignments for environmentally derived APH (eAPH) and RNA 2-O-methyltransferase (eTrm) orthologs. (A) Alignment for eAPH includes homologous APH sequences spanning the evolutionary subfamilies identified by Wright and Thompson ...

A second PCR product of interest, derived from industrially contaminated soil and 1140 bp in length, was cloned to create the plasmid pMAQ715. This product contained two gene cassettes, one of which (EGC067; accession no. af349097) included an ORF of 208 codons. The predicted protein comprised 207 residues, with Mr 23 kD and pI 9.77. The sequence was immediately identified as a member of the 2-O-methyltransferase family (Bateman et al. 2000), showing 25%–29% identity with high significance (E-values < 10−30) to many genes annotated as hypothetical RNA methyltransferases. Several members of this enzymes class have been functionally characterized, including SpoU (or TrmH) from E. coli, and shown to use S-adenosyl-L-methionine (AdoMet) for the 2′-O-methylation of tRNA or rRNA residues (Persson et al. 1997). Identified signature sequence motifs are distinct from those of enzymes that N-methylate bases of RNA (Kagan and Clark 1994; Gustafsson et al. 1996). The gene product from this ORF was therefore named eTrm, as an environmentally sourced RNA ribose methyltransferase.

eTrm shows 24% sequence identity with SpoU, but PSI-BLAST also identified moderate (17%–21%) sequence relationship to three methyltransferases whose structures have been solved recently: the Ado-Met binding YibK from Haemophilus influenzae, the rRNA methyltransferase RlmB from E. coli, and a putative methyltransferase, RrmA, from Thermus thermophilus. Threading was used to derive structure-based alignments of eTrm against each of these three targets, aligning longer segments and achieving higher sequence identity. The 207-residue sequence aligns to the structures of YibK (Lim et al. 2003), the C-terminal domain of RlmB (Michel et al. 2002), and RrmA (Nureki et al. 2002), so as to yield identity matches of 28% (over 180 residues), 25% (over 200 residues), and 25% (over 198 residues), respectively. The pairwise alignments were used to group the enzyme sequences as a related family (Fig. 1B) and indicate equally significant homology of eTrm to the three structural folds defined to date. These folds are all based on α/β/α architecture, but differ in topology from a Rossmann fold.

Isolation of ORF sequences as recombinant proteins

Both ORFs of interest were cloned into bacterial expression vectors for production as recombinant proteins. The steps of preparation and purification of the material are seen in the polyacrylamide gels shown in Figures 2 and 3. eAPH was overexpressed as a glutathione-S-transferase (GST) fusion, purified as a 65-kD protein from inclusion bodies, and re-folded following urea solubilization. This process generated GST-eAPH at >98% purity (Fig. 2) and at sufficient concentration for use in the activity assay. eTrm was expressed as a His6-fusion protein (23.9 kD), and purified to >98% purity from the soluble cell fraction by Ni-NTA affinity (Fig. 3). The yield of product isolated was enhanced when ribonuclease A (RNase A) was present in the lysis buffer. The significantly lower retention of His6-tagged protein at the affinity step in the absence of this agent suggests that eTrm can rapidly form a complex with released bacterial RNA, rendering its binding site for the Ni-NTA affinity matrix inaccessible. However, it was necessary to omit RNase A from all steps during the preparation of His6-eTrm for use in methyltransferase activity assays.

Figure 2.
Production and phosphotransferase activity of recombinant GST-eAPH (GST-aminoglycoside phosphotransferase ortholog from orf346_Pu5). (A) SDS PAGE-gel (visualized with silver stain) shows: Lanes M, 10-kD MW ladder; 1, pre-induced BL21 E. coli cells; 2 ...
Figure 3.
Production and methyltransferase activity of recombinant His6-eTrm (tRNA methyltransferase ortholog from orf208_Bal31). (A) SDS PAGE-gel (visualized with Coomassie dye) shows: Lanes M, 10-kD MW ladder; 1, pre-induced BL21(De3)/pLysS E. coli cells; 2, ...

Determination of enzyme activity

In the absence of a known substrate for eAPH, a general and highly sensitive ATPase activity assay was chosen to probe for activity, rather than a specific phosphotransferase assay. ATPase activity was measured by monitoring the specific products (ADP and AMP) of enzyme-mediated hydrolysis of [8-14C]-ATP following 2-h incubation at 30°C. The resulting nucleotides were quantified following their separation by thin-layer chromatography, as shown in Figure 2B. GST-eAPH degraded ATP to ADP and AMP far in excess of natural ATP degradation observed in control experiments, indicating that the recombinant product has ATPase activity. That the proportion of these degradation products was not altered by the presence of either hygromycin B or kanamycin indicates that these aminoglycosides are not the specific substrates for eAPH.

Figure 3B shows the counts for [3H]-methyl group modification of tRNA in the presence of His6-eTrm as a function of substrate quantity. It can be seen that eTrm displays methyltransferase activity with tRNA. The reaction requires 50 μg of substrate before counts above background are seen, and becomes saturated at ~300 μg tRNA. These quantities of tRNA are relatively large, possibly indicating that some potential methylation sites are already substituted within the commercially sourced tRNA sample. The negative controls used for the reaction were (1) [3H]-AdoMet alone, (2) [3H]-AdoMet plus carrier RNA, (3) eTrm alone, and (4) tRNA alone, for which values ranged from 0.29 to 0.44 nmole [3H-CH3]-group incorporated/h/mg protein. Reactions to test methyltransferase activity of His6-eTrm in the presence of whole RNA were limited by the amount of RNA available, and methyl group incorporation was close to background. This result does not discount the possibility of eTrm also having specific rRNA methylation activity, as the number of free sites in the fraction of rRNA present within the whole RNA sample may have been too small. Kinetic analysis of the activity of eTrm would require measurement of parameters with a single substrate, rather than the complex mixture of potential substrates used here to demonstrate activity.


In this study we isolated two cassette-associated genes directly from environmental DNA and expressed them as fusion proteins in E. coli for functional study. Their putative identification by sequence homology to specific enzyme classes (rather than to individual proteins) was confirmed by biochemical assay. Both of these genes were amplified from whole DNA, and therefore the organism of origin is unknown. In this way, our search differs from conventional gene/protein discovery methods, which are limited to prospecting within culturable organisms.

The enzyme classes involved in this study, the APH family and the 2-O-ribose methyltransferases, are repeatedly associated with mechanisms of antibiotic resistance. Although integrons have been associated with the dissemination of aminoglycoside-modifying enzymes within clinical isolates (Shaw et al. 1993), our discovery of eAPH and eTrm within gene cassettes amplified from environmental DNA indicates that such enzymes may be widespread through all bacteria, and that the mobile genome is a reservoir for antibiotic resistance genes.

The aminoglycoside antibiotics interfere with bacterial protein synthesis by binding to the 16S rRNA of the bacterial ribosome (Davies 1994). The APH family of enzymes modify the aminoglycoside in a regioselective and ATP-dependent manner (Wright and Thompson 1999), thereby blocking the ribosomal binding site and inactivating the antibiotic. The specific site of modification made to the substrate has been used to classify different isoenzymes of the APH family (Shaw et al. 1993), which generally display low sequence homology. Most studied are the APH(3′) class, which carry out phosphorylation of kanamycin and other 4,6-disubstituted aminoglycosides. This class includes APH(3′)-IIIa, carried by Enterococci and Staphylococci pathogens. Phosphotransferases which modify aminoglycosides at positions other than the 3′-hydroxyl are also widespread; for example, resistance to hygromycin arises from the activity of either APH(4) or APH(7″) enzymes (Zalacain et al. 1986).

The environmental APH isolated in the present study clearly contains the known sequence characteristics of the APH family, although the alignment with well characterized relatives (Fig. 1A) highlights the moderate homology overall. Fold determinants are sufficient to allow good correlation of the sequence of eAPH with the tertiary structure of the APH-3′ class. Three-dimensional structures of APH(3′)-IIIa (Hon et al. 1997; Burk et al. 2001) show that conserved side chains of Lys44, Glu60, Asp190, Asn195, and Asp208 are essential in making contact with the nucleotide phosphate groups and cofactor metal ions in a binding cleft between N- and C-terminal lobes of the protein fold. Our structure-based alignment shows the corresponding residues in the sequence of eAPH to be Arg, Phe, Asp, Asn, and Asp, respectively. The conservative replacement of Lys44 by an Arg residue in the eAPH sequence is also observed in macrolide 2-phosphotransferase-1 (MPH-1; see Fig. 1A). The chelating carboxyl side chain of Glu60 is not directly conserved in eAPH, but might instead be provided from sequentially neighboring Glu residues, for example at positions 67 or 58 (APH-3′ numbering). With this close conservation of residues known to form the characteristic nucleotide and Mg2+-binding sites, our demonstration that eAPH is a functional phosphotransferase is readily rationalized.

Recent genome sequencing efforts have widened the diversity of potential APH genes and, in the absence of functional characterization, these have been distinguished by the sequence HGDx4N, taken to be a catalytic signature (e.g., APH-3′ residues 188–195; Shaw et al. 1993). In the sequence of eAPH, however, the His residue of this motif is replaced by Met. Similar minor variations of this “signature” sequence are also observed elsewhere in the family, for example, in MPH-I and MPH-2, erythromycin resistance determinants from E. coli (Noguchi et al. 1995).

The antibiotic binding pocket of the APH enzymes is thought to reside within the least conserved portion of the sequence, corresponding to variable helical elements and loops formed by residues 147–170 [APH(3′)-IIIa numbering]. Attempts were made in this work to identify an appropriate substrate for eAPH. The protein found to be most closely related in sequence to eAPH was a phosphotransferase previously classed as APH(7″) through its mode of modification of hygromycin B (Zalacain et al. 1986). Despite this, no change in phosphorylation activity was observed for eAPH in the presence of this specific aminoglycoside. This, together with the lack of any enhanced activity with kanamycin as substrate, indicates that eAPH cannot be placed in any of the more well known APH(3′), APH(4), or APH(7″) functional classes. Wright and Thompson (1999) postulated that these classes are scattered across three of four evolutionarily distinct clusters of APH sequences. With specific binding determinants for hygromycin B and kana-mycin absent in the sequence of eAPH, this protein appears to be distinct from previously described APH members.

The second enzyme characterized in this work, eTrm, shares moderate sequence homology with a large number of putative RNA 2-O-methyltransferases. The sequence clearly contains characteristics of the SPOUT enzyme class (Fig. 1B), recently shown to differ from the classical RNA methyltransferases (Anantharaman et al. 2002). Members of the SPOUT family display relatively low sequence conservation, but share several sequence motifs in the vicinity of the AdoMet binding site. Three methyltransferases are identified here as structural homologs to eTrm, indicating that all four are similarly (and remotely) related to each other. Output from PSI-BLAST showed that across the family of related eTrm-like sequences, residues corresponding to eTrm positions 38, 44, 140, 142, 145, and 170 are all strictly conserved. In the structure-based alignments shown in Figure 1B, these key sites as well as other highly conserved residues are seen to reside within the sequence motifs historically used to define the SPOUT enzyme class (Gustafsson et al. 1996): Asn38, Gly40, and Arg44 from “motif I” reside in a conserved helix; Gly140, Glu142, and Gly145 within “motif II” are likely to be functional loop residues; Ile160, Pro161, Met162, Ser168, and Asn170 appear to comprise the AdoMet-binding loop at “motif III.”

The enzyme family appears to utilize a mixed α/β/α architecture. The fold determined in the examples of RlmB, YibK, and RrmA was highlighted as a rare example of an unusually deep trefoil knot (Michel et al. 2002; Nureki et al. 2002; Lim et al. 2003). This feature includes the loop segments of motifs II and III, known to bind AdoMet cofactor, and places the family as evolutionarily distinct from the Rossmann-fold methyltransferases. RlmB and RrmA are multidomain proteins, with different RNA-recognition domains coupled at the N terminus as independent domains. The YibK protein, however, is found as a shorter single domain in H. influenzae and has been shown to bind AdoMet as a dimer (Lim et al. 2003). In the integron cassette, we also find eTrm coded as a single unit, allowing the possibility of a variety of substrate specificities.

We do find evidence for binding of RNA by eTrm; our recombinant form is able to methylate tRNA with AdoMet as cofactor (Fig. 3B). The unusually high amount of RNA needed for this activity to be detected is possibly explained by the absence of a specific RNA-binding element within eTrm. Some members of the SpoU family have been demonstrated to methylate ribose rings in tRNA (Persson et al. 1997), whereas others modify rRNA (Sirum-Connolly and Mason 1993). Given that methylation of rRNA is understood to be one of the molecular mechanisms underlying antibiotic resistance, it is interesting to speculate that eTrm may also be capable of such a functional role.

We have shown that cassette PCR can amplify gene cassettes from whole environmental DNA. Further, the cassette-associated genes amplified contain little noncoding DNA and are in a form easily amenable to functional investigation and characterization. This is an advance on current methods of molecular prospecting that only recover internal parts of a gene and require additional labor-intensive steps to recover the complete gene. Also, because our gene discovery method is able to isolate whole genes regardless of prior sequence knowledge, it provides access to pools of diversity that are unavailable to current discovery methods. Aminoglycoside-modifying enzymes have been previously detected using PCR techniques and Southern hybridization, predominantly in clinical isolates (see, e.g., Udo and Dashti 2000). Similarly, discovery of RNA methyl-transferases has been limited to native protein purification from whole cells (Constantinesco et al. 1999) and homology searching of databases and PCR (Gustafsson et al. 1996). Given the relatively low sequence homology between the cassette-associated genes and their functionally characterized orthologs, it is highly unlikely that eTrm and eAPH would have been isolated by PCR targeting of conserved sequences.

Recent studies suggest that up to 17% of bacterial genomes may comprise DNA acquired through lateral transfer (Ochman et al. 2000). If methods of protein discovery are limited to genomes or proteomes of single examples of a species, it is probable that a significant proportion of proteins will remain undiscovered. More importantly, such conventional approaches may be inefficient at uncovering protein families of truly novel function. We recently described the number and diversity of cassette-associated genes isolated from a number of different environments (Stokes et al. 2001). Approximately 90% of ORFs found did not contain identifiable homologs in the protein databases, indicating the unique nature of the mobile genome. The route of discovery that we have explored here, that is, via PCR-mediated prospecting of whole environmental DNA by targeting 59-be recombination sites, has the ability to sample a wider genetic pool than other methods, and also has the potential for recovering whole new families of proteins.

Materials and methods

Gene isolation and sequence analysis

Soil samples (400 mg) collected from Pulgamurtie (#Pu01213, Sturt National Park, NSW, Australia) and Balmain Power Station (Sydney, Australia) were treated by a FastPrep bead beating procedure (Yeates and Gillings 1998) to isolate DNA. Primers HS287 and HS286 for PCR were designed from conserved regions in the 59-be to target gene cassette arrays (Stokes et al. 2001). PCR, ligation of PCR products, transformation into JM109 E. coli competent cells, and plasmid isolation from insert-containing clones were carried out as described (Stokes et al. 2001). DNA sequences of all cloned inserts were verified (Macquarie University Sequencing Facility). Cassette-associated genes of interest in this study were orf346_Pu5 (from clone pMAQ682) and orf208_Bal31 (from pMAQ715).

Sequence analysis involved iterative similarity searches using PSI-BLAST against the nonredundant sequence database and the Protein Data Bank (PDB) of structures as of August 2003. An initial sequence profile was built using an acceptance threshold of E-value < 10−10 and iterated until convergence (6–11 rounds). This was then repeated, accepting sequences with E-value < 10−8, more than four orders of magnitude more conservative than the defaults. In all cases, this converged in three rounds or less. The resulting profile was used to search for homologs among PDB sequences. The reported structural homologs (1J7I, 1GZ0A, 1IPAA, 1J85A) are quite unambiguous, with E-value < 10−27. The sequences were aligned against the structures using WURST (Huber et al. 1999) with the parameters optimized for structural models. This method uses a structural term based on Bayesian statistics combined with a contribution from a conventional sequence-sequence alignment (Russell and Torda 2002).

Construct and clone preparation

Primers containing appropriate linkers were designed to amplify the coding region of eAPH from clone pMAQ682 (5′GAGGATCCATGAGTGGACATAGC3′, with BamH1 linker; 5′GGGAATTCCTATATCAGGG3′, with EcoR1 linker) and of eTrm from clone pMAQ715 (5′GAGGCAACATATGGAAGAAGCTAAG 3′, with Nde1 linker; 5′ ATGGATCCTTAGTCCACACG 3′, with BamH1 linker). PCR was performed using standard techniques. The thermocycling program was 94°C for 2.5 min for 1 cycle; 94°C for 30 sec, 55°C for 30 sec, 72°C for 1.5 min for 30 cycles; and 72°C for 3 min for 1 cycle. Each PCR product was gel-purified, ligated into the pGEM-T Easy Vector (Promega), transformed, and sequence-verified as above.

The expression vector pGEX-4T-2 (Amersham-Pharmacia Bio-tech), was selected to produce eAPH as a GST fusion protein. For recombinant eTrm, the expression vector pET15b (Novagen) was chosen to produce a His6-fusion. Inserts and vectors were prepared by sequential digest with appropriate restriction enzymes and buffers (Promega). Following gel purification, inserts and vectors were ligated with T4 DNA ligase (Promega) and transformed into JM109 E. coli cells. Recombinant plasmid was isolated and transformed by heat shock into E. coli strains BL21 (for eAPH) or BL21(De3)/pLysS (for eTrm).

Expression of recombinant enzymes

Transformed cells growing at 25°C in Luria Bertani broth (with antibiotic supplements) were induced with IPTG (0.2 mM) at an optical density of 0.5–0.6 (A600) and harvested at 16–18 h. The cells were pelleted by centrifugation at 4°C.

In the case of GST-eAPH, the recombinant product was localized to the insoluble cell fraction. The cell pellet was resuspended in ice-cold Tris (20 mM)/phosphate (50 mM) buffer (pH 8.0) (containing 100 mM NaCl, 0.5% Tween-20, 1 mM benzamidine HCl, 10 μg/mL RNase A) at 5 mL per gram of cells. The cells were lysed with a French press, and the inclusion bodies were collected by centrifugation. The pellet was washed (2 × 1% Triton-X-100, 2 × water) and resuspended in 8 M urea and Tris buffer (50 mM [pH 7.5]) at 10 mL per liter of original culture. After 20 min agitation at room temperature, the remaining insoluble material was pelleted. An aliquot of resuspended inclusion bodies, containing ~98% GST-eAPH, was rapidly stirred (3–5 min) with ice-cold Buffer A (50 mM Tris buffer [pH 7.0], containing 10% glycerol, 2 mM MgCl2, 1 mM ATP, 1 mM dithiothreitol) to a final concentration of 50 μg/mL, then incubated overnight at 4°C. For some samples, the refolded GST-eAPH was purified in a gravity column of glutathione-Sepharose (Amersham-Pharmacia Biotech) equilibrated with ice-cold Buffer A. Following loading, the affinity medium was washed with ice-cold Buffer A, and the protein was eluted with reduced glutathione (50 mM) in ice-cold Tris buffer (50 mM [pH 8.0]). Prior to the activity assay, the sample of GST-eAPH was desalted and concentrated.

His6-eTrm was expressed using the same conditions as above, but isolated from the soluble cell fraction. Following resuspension of the cell pellet in Tris (20 mM)/phosphate (50 mM) buffer, (pH 8.0) (containing 100 mM NaCl, 2 mM MgCl2, 0.5% Tween 20, 1 mM benzamidine HCl, 10 mM imidazole) and lysis by French press, the recombinant product was directly purified by affinity chromatography using Ni-NTA agarose (QIAGEN). The agarose, pre-equilibrated with ice-cold Buffer B (20 mM Tris/50 mM phosphate buffer [pH 8.0], containing 100 mM NaCl, 2mM MgCl2) + 10 mM imidazole, was agitated (0°C, 1 h) with the cell supernatant and then transferred to a gravity column. The medium was successively washed with Buffer B + 20 mM imidazole, Buffer B + 40 mM imidazole, and Buffer C (20 mM Tris/50 mM phosphate buffer [pH 8.0], containing 300 mM NaCl, 2 mM MgCl2) + 40 mM imidazole. Pure His6-eTrm was finally eluted from the column with a solution of Buffer C + 100 mM imidazole.

Activity assays

ATPase activity of GST-eAPH was determined by measuring hydrolysis of [8-14C]-ATP. All reactions contained ~1μg GST-eAPH (0.7 μM) and [8-14C]-ATP (50 nCi, 46 μM; Amersham-Pharmacia Biotech) and were made up to a final volume of 20 μL with 50 mM Tris (pH 7.0). Aminoglycoside substrates trialed were kana-mycin (2 nmole; Sigma) and hygromycin B (0.12 U; Calbiochem). Reactions were incubated at 30°C for 2 h and then stopped by cooling on ice (5 min), followed by addition of perchloric acid (5 μL, 5% v/v) and incubation at room temperature (5 min). Samples were centrifuged to pellet protein material, and the nucleotide species were separated by thin-layer chromatography (PEI-cellulose, Merck) with formic acid (1 M) and LiCl (1 M) as the mobile phase. The air-dried membrane was exposed to x-ray film for 48–72 h at room temperature, and band intensities were measured by densitometry.

RNA methyltransferase assays were based on the method of Gu et al. (1994) and monitored the [3H]-methyl group incorporation into either whole RNA (extracted from lab-grown cultures of E. coli and Candida albicans) or tRNA from E. coli strain W (Sigma). Each assay contained varying concentrations of RNA in the presence of His6-eTrm (4 μM), 20U RNA guard (Amersham-Pharmacia Biotech) and [3H]-AdoMet (1μ Ci, 40 μM; Amersham-Pharmacia Biotech) in a total volume of 50 μL Tris buffer (100 mM [pH 7.5], containing 40 mM NH4Cl, 2 mM MgCl2, and 5 mM DTT). Reactions were incubated at room temperature for 50 min, then stopped by addition of ice-cold trichloroacetic acid (TCA; to 5% w/v) and carrier RNA (20 μg degraded yeast RNA). All reactions were stored on ice before being transferred to nylon membrane filters (Biodyne B, Pall Gellman) prewashed with ice-cold 5% TCA. Following further washing (ice-cold 5% TCA, ice-cold 95% ethanol), the filters were air dried. Tritium counts from bound RNA were determined in a liquid scintillation counter (Tri-Carb, Model 2100TR; Packard Bioscience) for each filter in the presence of 2 mL scintillation fluid (ACS; Amersham-Pharmacia Biotech). Although there was high variation between assays, results for duplicates within each assay were highly consistent.


This work was funded by an Innovation Grant from Macquarie University.

The publication costs of this article were defrayed in part by payment of page charges. This article must therefore be hereby marked “advertisement” in accordance with 18 USC section 1734 solely to indicate this fact.


  • 59-be, 59-base element DNA recombination site
  • AdoMet, S-adenosyl-L-methionine
  • E-value, expectation value defined by homology search program
  • eAPH, aminoglycoside phosphotransferase from an environmental microbial source
  • eTrm, RNA 2-O-ribose methyl-transferase from an environmental microbial source
  • GST, glutathione-S-transferase
  • MPH, macrolide 2-phosphotransferase
  • ORF, open reading frame
  • PCR, polymerase chain reaction
  • PDB, Protein Data Bank
  • RlmB, methyltransferase from Escherichia coli
  • RrmA, putative methyltranferase from Thermus thermophilus
  • RNase A, ribonuclease A
  • TCA, trichloroacetic acid
  • tRNA, transfer RNA
  • YibK, methyltransferase from Haemophilus influenzae


Article and publication are at http://www.proteinscience.org/cgi/doi/10.1110/ps.04638704.


  • Altschul, S.F., Thomas, L.M., Schäffer, A.A., Zhang, J., Zhang, Z., Miller, W., and Lipman, D.J. 1997. Gapped BLAST and PSI-BLAST: A new generation of protein database search programs. Nucleic Acids Res. 25 3389–3402. [PMC free article] [PubMed]
  • Anantharaman, V., Koonin, E.V., and Aravind, L.J. 2002. SPOUT: A class of methyltransferases that includes spoU and trmD RNA methylase superfamilies, and novel superfamilies of predicted prokaryotic RNA methylases. J. Mol. Microbiol. Biotechnol. 4 71–75. [PubMed]
  • Bateman, A., Birney, E., Durbin, R., Eddy, S.R., Howe, K.L., and Sonnhammer, E.L.L. 2000. The Pfam protein families database. Nucleic Acids Res. 28 263–266. [PMC free article] [PubMed]
  • Burk, D.L., Hon, W.C., Leung, A.K., and Berghuis, A.M. 2001. Structural analyses of nucleotide binding to an aminoglycoside phosphotransferase. Biochemistry 40 8756–8764. [PubMed]
  • Chance, M.R., Bresnick, A.R., Burley, S.K., Jiang, J.S., Lima, C.D., Sali, A., Almo, S.C., Bonanno, J.B, Buglino, J.A., Boulton, S., et al. 2002. Structural genomics: A pipeline for providing structures for the biologist. Protein Sci. 11 723–738. [PMC free article] [PubMed]
  • Constantinesco, F., Motorin, Y., and Grosjean, H. 1999. Transfer RNA modification enzymes from Pyrococcus furiosus: Detection of the enzymatic activities in vitro. Nucleic Acids Res. 27 1308–1315. [PMC free article] [PubMed]
  • Davies, J. 1994. Inactivation of antibiotics and the dissemination of resistance genes. Science 264 375–382. [PubMed]
  • Gaasterland, T. and Oprea, M. 2001. Whole-genome analysis: Annotations and updates. Curr. Opin. Struct. Biol. 11 377–381. [PubMed]
  • Gu, X., Ofengand, J., and Santi, D.V. 1994. In vitro methylation of Escherichia coli 16S rRNA by tRNA (m5U54)-methyltransferase. Biochemistry 33 2255–2261. [PubMed]
  • Gustafsson, C., Reid, R., Greene, P.J., and Santi, D.V. 1996. Identification of new RNA modifying enzymes by iterative genome search using known modifying enzymes as probes. Nucleic Acids Res. 24 3756–3762. [PMC free article] [PubMed]
  • Hall, R.M., Brookes, D.E., and Stokes, H.W. 1991. Site-specific insertion of genes into integrons: Role of the 59-base element and determination of the recombination crossover point. Mol. Microbiol. 5 1941–1959. [PubMed]
  • Hall, R.M., Collis, C.M., Kim, M.-J., Partridge, S.R., Recchia, G.D., and Stokes, H.W. 1999. Mobile gene cassettes and integrons in evolution. Ann. N. Y. Acad. Sci. 870 68–80. [PubMed]
  • Henikoff, S., Greene, E.A., Pietrokovski, S., Bork, P., Attwood, T.K., and Hood, L. 1997. Gene families: The taxonomy of protein paralogs and chimeras. Science 278 609–614. [PubMed]
  • Holmes, A.J., Holley, M.P., Mahon, A., Nield, B.S., Gillings, M.R., and Stokes, H.W. 2003a. A distinctive and functional integron/gene cassette system present in soil bacterial communities associated with genomic diversity in Pseudomonas stutzeri. J. Bacteriol. 185 918–928. [PMC free article] [PubMed]
  • Holmes, A.J., Gillings, M.R., Nield, B.S., Mabbutt, B.C., Nevalainen, K.M.H., and Stokes, H.W. 2003b. The gene cassette metagenome is a basic resource for bacterial genome evolution. Environ. Microbiol. 5 383–394. [PubMed]
  • Hon, W.C., McKay, G.A., Thompson, P.R., Sweet, R.M., Yang, D.S., Wright, G.D., and Berghuis, A.M. 1997. Structure of an enzyme required for aminoglycoside antibiotic resistance reveals homology to eukaryotic protein kinases. Cell 89 887–895. [PubMed]
  • Huber, T., Russell, A.J., Ayers, D., and Torda, A.E. 1999. SAUSAGE: Protein threading with flexible force fields. Bioinformatics 15 1064–1065. [PubMed]
  • Kagan, R.M. and Clarke, S. 1994. Widespread occurrence of three sequence motifs in diverse S-adenosylmethionine-dependent methyltransferases suggests a common structure for these enzymes. Arch. Biochem. Biophys. 310 417–427. [PubMed]
  • Lim, K., Zhang, H., Tempczyk, A., Krajewski, W., Bonander, N., Toedt, J., Howard, A., Eisenstein, E., and Herzberg, O. 2003. Structure of the YibK methyltransferase from Haemophilus influenzae (HI0766): A cofactor bound at a site formed by a knot. Proteins 51 56–67. [PubMed]
  • Michel, G., Sauve, V., Larocque, R., Li, Y., Matte, A., and Cygler, M. 2002. The structure of the RlmB 23S rRNA methyltransferase reveals a new methyltransferase fold with a unique knot. Structure 10 1303–1315. [PubMed]
  • Nield, B.S., Holmes, A.J., Gillings, M.R., Recchia, G.D., Mabbutt, B.C., Nevalainen, K.M.H., and Stokes, H.W. 2001. Recovery of new integron classes from environmental DNA. FEMS Microbiol. Lett. 195 59–65. [PubMed]
  • Noguchi, N., Emura, A., Matsuyama, H., O’Hara, K., Sasatsu, M., and Kono, M. 1995. Nucleotide sequence and characterization of erythromycin resistance determinant that encodes macrolide 2′-phosphotransferase I in Escherichia coli. Antimicrob. Agents Chemother. 39 2359–2363. [PMC free article] [PubMed]
  • Nureki, O., Shirouzu, M., Hashimoto, K., Ishitani, R., Terada, T., Tamakoshi, M., Oshima, T., Chijimatsu, M., Takio, K., Vassylyev, D.G., et al. 2002. An enzyme with a deep trefoil knot for the active-site architecture. Acta Crystallogr. D 58 1129–1137. [PubMed]
  • Ochman, H., Lawrence, J.G., and Groisman, E.A. 2000. Lateral gene transfer and the nature of bacterial innovation. Nature 405 299–304. [PubMed]
  • Pace, N.R. 1997. A molecular view of microbial biodiversity and the biosphere. Science 276 734–740. [PubMed]
  • Persson, B.C., Jager, G., and Gustafsson, C. 1997. The spoU gene of Escherichia coli, the fourth gene of the spoT operon, is essential for tRNA (Gm18) 2′-O-methyltransferase activity. Nucleic Acids Res. 25 4093–4097. [PMC free article] [PubMed]
  • Rowe-Magnus, D.A., Guerot, A.-M., Biskri, L., Bouige, P., and Mazel, D. 2003. Comparative analysis of superintegrons: Engineering extensive genetic diversity in the Vibrionaceae. Genome Res. 13 428–442. [PMC free article] [PubMed]
  • Russell, A.J. and Torda, A.E. 2002. Protein sequence threading: Averaging over structures. Proteins 47 496–505. [PubMed]
  • Shaw, K.J., Rather, P.N., Hare, R.S., and Miller, G.H. 1993. Molecular genetics of aminoglycoside resistance genes and familial relationships of the aminoglycoside-modifying enzymes. Microbiol. Rev. 57 138–163. [PMC free article] [PubMed]
  • Sirum-Connolly, K. and Mason, T.L. 1993. Functional requirement of a site-specific ribose methylation in ribosomal RNA. Science 262 1886–1889. [PubMed]
  • Stokes, H.W. and Hall, R.M. 1989. A novel class of potentially mobile DNA elements encoding site-specific gene integration functions: Integrons. Mol. Microbiol. 3 1669–1683. [PubMed]
  • Stokes, H.W., Holmes, A.J., Nield, B.S., Holley, M.P., Nevalainen, K.M., Mab-butt, B.C., and Gillings, M.R. 2001. Gene cassette PCR: Sequence-independent recovery of entire genes from environmental DNA. Appl. Env. Microbiol. 67 5240–5246. [PMC free article] [PubMed]
  • Udo, E.E. and Dashti, A.A. 2000. Detection of genes encoding aminoglycoside-modifying enzymes in staphylococci by polymerase chain reaction and dot blot hybridization. Int. J. Antimicrob. Agents 13 273–279. [PubMed]
  • Wright, G.D. and Thompson, P.R. 1999. Aminoglycoside phosphotransferases: Proteins, structure, and mechanism. Frontiers Biosci. 4 D9–D21. [PubMed]
  • Yeates, C. and Gillings, M.R. 1998. Rapid purification of DNA from soil for molecular biodiversity analysis. Lett. Appl. Microbiol. 27 49–53.
  • Zalacain, M., Gonzalez, A., Guerrero, M.C., Mattaliano, R.J., Malpartida, F., and Jimenez, A. 1986. Nucleotide sequence of the hygromycin B phosphotransferase gene from Streptomyces hygroscopicus. Nucleic Acids Res. 14 1565–1581. [PMC free article] [PubMed]

Articles from Protein Science : A Publication of the Protein Society are provided here courtesy of The Protein Society
PubReader format: click here to try


Save items

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • MedGen
    Related information in MedGen
  • PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...