R164H and V240H Replacements by Site-Directed Mutagenesis of TEM-149 Extended-Spectrum β-Lactamase: Kinetic Analysis of TEM-149^H240 and TEM-149^H164-H240 Laboratory Mutants

Abstract

Two laboratory mutant forms, TEM-149^H240 and TEM-149^H164-H240, of the TEM-149 extended-spectrum β-lactamase enzyme were constructed by site-directed mutagenesis. TEM-149^H240 and TEM-149^H164-H240 were similar in kinetic behavior, except with respect to benzylpenicillin and ceftazidime. Molecular modeling of the two mutant enzymes demonstrated the role of histidine at position 240 in the reduction of the affinity of the enzyme for ceftazidime.

TEXT

The TEM β-lactamases are the most-studied antibiotic resistance enzymes. To date, 200 mutants are known to differ from each other in their amino acid sequences and resistance phenotypes. The new variants have evolved from TEM-1/2 alleles as a result of the introduction of novel β-lactam antibiotics in clinical practice in the past 3 decades. As reported by Bush and Jacoby (http://lahey.org/studies), TEM variants are prevalently extended-spectrum β-lactamases (ESBLs). The main substitutions that modify the resistance phenotype found with high frequency in clinical isolates and in multiple experimental studies are E104K, R164C, R164H, R164S, A237T, G238S, and E240K (1). In several studies, it has been demonstrated that these amino acid replacements are involved in extending the resistance spectrum of TEM-1 (2). Moreover, there are some amino acid substitutions identified in laboratory studies by in vitro or in vivo evolution of TEM-1 (3) before their identification in clinical isolates (4). The success of their evolution is due to their localization on transferable plasmids capable of rapid horizontal spreading among different enterobacterial species (5).

(Part of this study was presented at the 11th β-Lactamase Meeting, 10 to 14 June 2011, Leonessa, Italy.)

During the last nationwide survey of ESBLs undertaken in Italy in 2003, a TEM-149 enzyme was isolated from Enterobacter aerogenes and Serratia marcescens (6). Compared to the TEM-1 sequence, TEM-149 showed an original array of amino acid changes: E104K, R164S, M182T, and E240V. The goal of this study was to investigate the role of the histidine at position 240 alone (TEM-149^H240, single mutant) and in combination with the histidine at position 164 (TEM-149^H164-H240, double mutant) in the TEM-149 enzyme in order to assess the contributions of these substitutions to the phenotypic resistance pattern of Escherichia coli and to kinetic parameters. A histidine at position 240 has never been found in clinical isolates producing TEM variants. To date, histidine 240 in TEM variants was found only in laboratory mutants, as reported by Salverda et al. (4). For this reason, two TEM-149 mutants were generated by site-directed mutagenesis using the overlap extension method (7). Recombinant plasmid pTEM-149 (6) was used as the template for site-directed mutagenesis experiments. Briefly, each mutation was introduced into a PCR amplicon using mutagenic primers in combination with external primers to generate two partially overlapping DNA fragments. These fragments were subsequently used in an overlap extension reaction coupled with amplification of the entire coding sequence with the TEM/F and TEM/R primers. All of the primers used for overlapping are listed in Table 1. Each fragment was sequenced using an ABI PRISM 310 monocapillary automated sequencer (Applied Biosystems, Life Technologies). The resulting amplicons were cloned into vector pBC-SK (Stratagene, Inc., La Jolla, CA) to obtain recombinant plasmids pBCTEM149^H240 and pBCTEM149^H164-H240. E. coli strain XL-1 was used as the host for recombinant plasmids. The authenticity of cloned mutant genes was verified by sequencing the recombinant plasmids on both strands. Each mutant enzyme was purified from overnight cultures of E. coli XL-1(pBCTEM149^H240) and XL-1(pBCTEM149^H164-H240) in brain heart infusion medium at 37°C. Enzymes were extracted by sonic disruption from bacterial cells suspended in 100 mM Tris-HCl (pH 8.0). The enzyme was purified from the clarified crude extract by three chromatography steps: anion-exchange chromatography on a Q-Sepharose FF column (GE Healthcare, Milan, Italy) equilibrated with 100 mM Tris-HCl (pH 8.0) and eluted with a linear NaCl gradient (0 to 1 M) in the same buffer; size exclusion chromatography on a Superdex 200 column equilibrated and eluted with 20 mM sodium phosphate buffer (pH 7.0) containing 0.15 M NaCl; and fast chromatofocusing on a MonoP HR 5/20 column (GE Healthcare, Milan, Italy) equilibrated with 25 mM Bis-Tris buffer (pH 7.1) and eluted with 25 ml of 10-fold-diluted Polybuffer 74 in a pH range of 7 to 5. During purification, β-lactamase activity was monitored by hydrolysis of 100 μM nitrocefin. The purity of the enzyme preparation was more than 95%, as determined by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. Steady-state kinetic parameters (K_m and k_cat) with β-lactam substrates were determined under initial-rate conditions as described previously (8). Inhibition by clavulanic acid and tazobactam was investigated using nitrocefin (200 μM) as the reporter substrate. In vitro susceptibility testing, performed by the conventional broth macrodilution procedure with a bacterial inoculum of 5 × 10⁵ CFU/ml, as recommended by the CLSI (9), showed similar resistance patterns for E. coli XL-1(pTEM-149), XL-1(pBCTEM-149^H240), and XL-1(pBCTEM-149^H164-H240), except that E. coli XL-1(pBCTEM-149^H240) had an aztreonam MIC of 4 mg/liter (Table 2). The kinetic parameters of the mutant enzymes were determined with several β-lactam substrates, and the results are shown in Table 3. The single and double mutants were similar in kinetic behavior, except with benzylpenicillin and ceftazidime. For instance, the single mutant TEM showed negligible hydrolysis of benzylpenicillin while the double mutant TEM showed the worst catalytic efficiency with respect to the wild type. Tazobactam and clavulanic acid behaved as competitive inhibitors, and the K_i values were very similar to those calculated for TEM-149 (Table 4).

Table 1.

Oligonucleotide primers used for site-directed mutagenesis

Primer	Sequence (5′–3′)a	Positionsb	Codon change
Amplification
TEMKPNNDE_for	GGTACCATATGAGTATTCAACATTTTCGT
TEMXHO_REV	CTCGAGTTACCAATGCTTAATCAGTG
Mutagenic for S164H
TEM149H164_for	TAACCCGCCTTGATCATTGGGAACCG	470–495	AGT→CAT
TEM149H164_rev	CGGTTCCCAATGATCAAGGCGGGTTA	495–470
TEM149H240_for	AATCTGGAGCCGGTCATCGTGGATCTCGC	695–723	GTG→CAT
TEM149H240_rev	GCGAGATCCACGATGACCGGCTCCAGATT	723–695

^aThe mutated positions are in bold. The KpnI/NdeI and XhoI restriction sites are underlined.

^bPositions are numbered in accordance with the bla_TEM-149 coding sequence.

Table 2.

Patterns of β-lactam resistance mediated by TEM-149, TEM-149^H240, and TEM-149^H164-H240 in E. coli XL-1

Antibiotic	MIC (μg/ml) for E. coli XL-1 carrying:
	pTEM-149	pBCTEM-149H240	pBCTEM-149H164-H240	pBC-SK
Piperacillin	>256	1	256	0.5
Piperacillin-tazobactam	1	1	0.125	0.5
Amoxicillin	>64	64	>64	2
Amoxicillin-clavulanate	8/4	2/1	4/2	2/1
Cefazolin	32	2	32	1
Cefotaxime	1	<0.0625	1	<0.0625
Ceftazidime	>64	>64	>64	0.5
Cefepime	8	0.0625	4	<0.0625
Imipenem	<0.625	0.25	0.25	0.125
Meropenem	<0.0625	<0.0625	<0.0625	<0.0625
Aztreonam	>64	4	>64	0.125

Table 3.

Kinetic parameters^a of mutant TEM-149 enzymes

Substrate	TEM-149b			TEM-149H240c			TEM-149H164-H240d
	Km (μM)	kcat (s−1)	kcat/Km (μM−1 s−1)	Km (μM)	kcat (s−1)	kcat/Km (μM−1 s−1)	Km (μM)	kcat (s−1)	kcat/Km (μM−1 s−1)
Benzylpenicillin	0.6e ± 0.05	3.5	5.8	4.7 ± 0.5	NDf	ND	87 ± 4	18	0.21
Cefazolin	68 ± 5	2.2	3.2 × 10−2	91 ± 4	1.7	1.9 × 10−2	205 ± 6	9.5	4.6 × 10−2
Cefotaxime	43 ± 2	0.07	0.16 × 10−2	12 ± 1	0.11	0.9 × 10−2	66 ± 4	0.37	0.56 × 10−2
Ceftazidime	19 ± 1	8.3	44 × 10−2	445 ± 5	96	21 × 10−2	150 ± 3	2.3	1.55 × 10−2
Cefepime	16 ± 1	2.4	15 × 10−2	186	1.6	0.86 × 10−2	53 ± 2	3.4	6.4 × 10−2
Aztreonam	51 ± 3	4.7	9.2 × 10−2	3 ± 0.05	0.81	27 × 10−2	55 ± 3	4.5	8.2 × 10−2
Nitrocefin	20 ± 2	7.4	37 × 10−2	136 ± 5	7.1	5.2 × 10−2	82 ± 4	20	24 × 10−2

^aEach kinetic value is the mean of five different measurements. The error was below 5%.

^bE104K/R164S/M182T/E240V.

^cE104K/R164S/M182T/E240H single mutant.

^dE104K/R164H/M182T/E240H double mutant.

^eK_m = K_i with 100 μM nitrocefin as the reporter substrate.

^fND, not determined.

Table 4.

β-Lactamase inhibitor analysis of TEM-149 mutant enzymes compared with wild-type TEM-149

Enzyme	Ki (μM)a for:
	Tazobactam	Clavulanic acid
TEM-149	0.018	0.26
TEM-149H240	0.039	0.15
TEM-149H164-H240	0.004	0.21

^aK_i values were calculated using 200 μM nitrocefin as the reporter substrate. The values are means of three measurements. The standard deviation was always lower than 5%.

Although the K_m and k_cat of the V240H mutant enzyme for ceftazidime were 23- and 11-fold higher, respectively, than those of wild-type TEM-149, its catalytic efficiency was comparable. The ceftazidime k_cat/K_m value of the R164H V240H double mutant was 29-fold lower than that of the TEM-149 enzyme and 13-fold lower than that of the single mutant. The decrease in the catalytic efficiency of the double mutant is related principally to the dramatic reduction of the k_cat value, whereas that of the single mutant is due to the increase in the K_m value.

Molecular modeling of TEM-149^H240 and TEM-149^H164-H240 was carried out using modeler 9.8 with TEM-1 as the template (1M40.pdb at a resolution of 0.85 Å). Each model was used for in silico docking with ceftazidime using AutoDock 4.0 imposing flexibility on the substrate and on residues S70, E166, K104, and T182; for the single mutant, S164 and H240, and for the double mutant, H164 and H240 (Fig. 1).

Fig 1.

Molecular modeling of the Michaelis complex of TEM-149 double mutant R164H E240H with ceftazidime, showing the residues discussed in the text. The histidine at position 164 is sufficiently close to the carboxylic group of the aspartate at position 179 to allow the formation of a hydrogen bond. The lysine side chain at position 104 is able to form an electrostatic bond with the carboxylic acid group of the oxyimino moiety of ceftazidime. The imidazolic group of histidine at position 240 can interact with the amino group of the aminothiazolic substituent of ceftazidime.

Residue H164.

Residue 164 is located in the omega loop of the TEM variants. The guanidinium side chain of the arginine usually found at this position makes an electrostatic bond with residue D179. In TEM-149, this amino acid is replaced with serine, which interacts with D179 by a hydrogen bond, making the omega loop more flexible. As demonstrated previously in TEM-134 (10), the presence of histidine at position 164 and the undetermined charged state of the imidazole group might result in a compromise between the constrained omega loop with arginine and the flexible loop with serine at the same position. The reduced flexibility of the omega loop has detrimental effects on the k_cat for ceftazidime, as also demonstrated by the dramatic reduction in the catalytic constant with respect to the TEM-149^H240 mutant enzyme.

Residue H240.

Residue 240 is located at the end of the B3 beta strand (2). In TEM-type enzymes, position 240 is generally taken up by a glutamate, which can interact with the amino group of the amino-thiazole substituent of cephalosporins. In wild-type TEM-149, this position is occupied by the nonpolar residue valine, which can facilitate the accommodation of the bulky oxyimino substituent of ceftazidime (6). This results in an increase in the affinity of the enzyme for the substrate. The presence of histidine at position 240 in both the single and double mutant enzymes generally decreases their affinity with respect to that of wild-type TEM-149 by increasing their K_m values to 445 and 150 μM, respectively. As shown by in silico analysis, H240 can interact with the amino group of the amino-thiazole substituent of ceftazidime by a hydrogen bond. Also, the lysine side chain at position 104 is able to form an electrostatic bond with the carboxylic acid group of the oxyimino moiety of ceftazidime. The combined effect of K104 and H240 might drastically reduce the affinity of the enzyme for the substrate.