Contributions of aspartate 49 and phenylalanine 142 residues of a tight binding inhibitory protein of beta-lactamases.

beta-Lactamases are bacterial enzymes that hydrolyze beta-lactam antibiotics to render them inactive. The beta-lactamase inhibitor protein (BLIP) of Streptomyces clavuligerus, is a potent inhibitor of several beta-lactamases, including the TEM-1 enzyme (Ki = 0.6 nM). Evidence from the TEM-1/BLIP co-crystal suggests that two BLIP residues, Asp-49 and Phe-142, mimic interactions made by penicillin G when bound in the active site of TEM-1. To determine the importance of these two residues, a heterologous expression system for BLIP was established in Escherichia coli. Site-directed mutagenesis was used to change Asp-49 and Phe-142 to alanine, and inhibition constants (Ki) for both mutants were determined. Each mutation increases the Ki for BLIP inhibition of TEM-1 beta-lactamase approximately 100-fold. To address how these two positions effect the specificity of beta-lactamase binding, Ki values were determined for the interaction of wild-type BLIP, as well as the D49A and F142A mutants, with two extended spectrum beta-lactamases (the G238S and the E104K TEM variants). Positions 104 and 238 are located in the BLIP/beta-lactamase interface. Interestingly, the three BLIP proteins inhibited the G238S beta-lactamase mutant to the same degree that they inhibited TEM-1. However, wild-type BLIP has a higher Ki for the E104K beta-lactamase mutant, suggesting that interactions between BLIP and beta-lactamase residue Glu-104 are important for wild-type levels of BLIP inhibition.

␤-Lactamases are bacterial enzymes that hydrolyze ␤-lactam antibiotics to render them inactive. The ␤-lactamase inhibitor protein (BLIP) of Streptomyces clavuligerus, is a potent inhibitor of several ␤-lactamases, including the TEM-1 enzyme (K i ‫؍‬ 0.6 nM). Evidence from the TEM-1/BLIP co-crystal suggests that two BLIP residues, Asp-49 and Phe-142, mimic interactions made by penicillin G when bound in the active site of TEM-1.
To determine the importance of these two residues, a heterologous expression system for BLIP was established in Escherichia coli. Site-directed mutagenesis was used to change Asp-49 and Phe-142 to alanine, and inhibition constants (K i ) for both mutants were determined. Each mutation increases the K i for BLIP inhibition of TEM-1 ␤-lactamase approximately 100-fold. To address how these two positions effect the specificity of ␤-lactamase binding, K i values were determined for the interaction of wild-type BLIP, as well as the D49A and F142A mutants, with two extended spectrum ␤-lactamases (the G238S and the E104K TEM variants). Positions 104 and 238 are located in the BLIP/␤-lactamase interface. Interestingly, the three BLIP proteins inhibited the G238S ␤-lactamase mutant to the same degree that they inhibited TEM-1. However, wild-type BLIP has a higher K i for the E104K ␤-lactamase mutant, suggesting that interactions between BLIP and ␤-lactamase residue Glu-104 are important for wild-type levels of BLIP inhibition.
␤-Lactamases are bacterial enzymes that confer resistance to ␤-lactam antibiotics, which include the penicillins and cephalosporins. These drugs act by inhibiting bacterial penicillin binding proteins (PBPs) 1 that are essential for the synthesis of the bacterial peptidoglycan layer (1). Several resistance mechanisms have evolved in bacteria to protect them from the lethal effects of ␤-lactam antibiotics.
␤-Lactamase imparts resistance to the ␤-lactams by hydrolyzing the amide bond in the four-membered ␤-lactam ring (2). Genes encoding ␤-lactamases may be found on plasmids, transposons, and on the bacterial chromosome (2)(3)(4)(5)(6). There are four classes of ␤-lactamases (classes A-D), categorized by their pri-mary sequence homology. ␤-Lactamase-mediated resistance has increased because of selective pressure from widespread use of ␤-lactam antibiotics and is now a serious threat to antibiotic therapy (7).
TEM-1 ␤-lactamase, encoded by the bla TEM-1 gene, is among the most prevalent plasmid-encoded ␤-lactamases found in Gram-negative bacteria (8). The name TEM is derived from the name of the patient carrying the pathogen from which the enzyme was isolated (9). It efficiently hydrolyzes penicillins and most cephalosporins (2). However, it is a poor catalyst for hydrolysis of the newer third-generation cephalosporins, such as ceftazidime, that were designed to circumvent TEM-1-mediated inactivation.
In addition to developing new drugs that are unable to be cleaved by ␤-lactamase, another method to counter the hydrolytic activity of this enzyme has been to administer ␤-lactamase inhibitors, such as sulbactam and clavulanic acid. Use of an inhibitor along with an existing ␤-lactam antibiotic is an effective means to treat various ␤-lactamase producing bacterial pathogens (10). Clavulanic acid was originally found as a metabolite of the Gram-positive soil bacterium Streptomyces clavuligerus and, today, is one of the most commonly used ␤-lactamase inhibitors.
␤-Lactamase-mediated resistance has been exacerbated by the fact that specific mutations in TEM-1 ␤-lactamase enable the enzyme to hydrolyze the newer third-generation antibiotics (11,12). These evolved ␤-lactamases, called extended spectrum ␤-lactamases (ESBLs), provide clinically relevant levels of resistance to even the most recently developed ␤-lactams. Furthermore, other mutants have been found with substitutions that allow ␤-lactamase to avoid inactivation by the ␤-lactamase inhibitors (13). A similar result has been observed with SHV-1 ␤-lactamase, another class A enzyme that is 68% identical to TEM-1 ␤-lactamase (7). Recently, a ␤-lactamase mutant (TEM-50) has been recovered from clinical isolates with both types of mutations, enabling ␤-lactamase to hydrolyze extended spectrum antibiotics and avoid inactivation by inhibitors (14).
In 1990, the ␤-lactamase inhibitor protein (BLIP) was isolated from S. clavuligerus culture supernatants (15). BLIP is a 165-amino acid protein in its secreted form and is a potent inhibitor of TEM-1 ␤-lactamase (K i ϭ 0.6 nM) (16). BLIP is able to inhibit several ␤-lactamases, as well as weakly inhibit a penicillin-binding protein (PBP) from Enterococcus faecalis (16). The DNA sequence and crystal structure of BLIP have been determined, as well as the co-crystal with TEM-1 ␤-lactamase (16,17).
The BLIP crystal structure shows that the protein has a novel fold with two similar domains. The BLIP mechanism of inhibition appears to be two-fold. At 2636 Å 2 , the surface area of the enzyme/inhibitor interface is one of the largest known for protein/protein complexes as BLIP essentially clamps over the active site of ␤-lactamase (17). In addition, an aspartic acid residue at position 49 of BLIP aligns itself in the active site pocket and forms strong hydrogen bonds with four catalytic residues of ␤-lactamase. Furthermore, a phenylalanine at position 142 of BLIP appears to mimic the benzyl group of the ␤-lactam antibiotic penicillin G (PenG) and further stabilizes the inhibitory complex (17). The result is potent inhibition of TEM-1 ␤-lactamase.
The ongoing problem of targeting bacteria with antimicrobial agents able to circumvent ESBL-mediated antibiotic inactivation creates a need for new potent inhibitors of ␤-lactamases. The ability to engineer BLIP to tightly bind the ESBLs and PBPs would aid in the design of effective new antimicrobials. New protein-protein interactions found to be effective in inhibiting ␤-lactamases and PBPs could be duplicated by a peptide inhibitor or synthetic compound designed to mimic the interactions. It would be easier to perform such studies with BLIP if the inhibitor could be expressed in a more manageable genetic and molecular system, such as Escherichia coli. This report describes the efficient expression of functional BLIP in E. coli and the use of this system to determine the importance of residues Asp-49 and Phe-142 for BLIP inhibition of TEM-1 ␤-lactamase, two extended spectrum TEM-1 mutants, and the SHV-1 ␤-lactamase.

EXPERIMENTAL PROCEDURES
Strains and Plasmids-E. coli XL1-Blue (recA1, endA1, gyrA96, thi-1, hsdR17, supE44, relA1, lac, (FЈ proAB, lacI q lacZDM15, Tn10(tet r ))) was used to propagate plasmid DNA (Stratagene, Inc.). E. coli RB791 (ϭ strain W3110 lacI Q L8) was used to express BLIP and the D49A and F142A BLIP mutants (18,19). Plasmid pTP123 is a cmp r amp s derivative of pTrc 99A (Amersham Pharmacia Biotech) (see Fig.  1). It was created by ligating the HincII cassette from pKRP10 into BsaI-digested pTrc 99A (20). The BsaI site was filled in using the Klenow fragment of DNA Polymerase I prior to ligation. This cloning step inserts a chloramphenicol acetyltransferase (cat) gene into the rrnBT 1 T 2 transcriptional terminators and part of the ␤-lactamase gene encoded by pTrc 99A. As a result, functional ␤-lactamase is not expressed, and potential difficulty in BLIP purification because of binding of endogenous ␤-lactamase is avoided. The cat gene in pTP123 is in the same orientation as the trc promoter.
BLIP Cloning and PCR Mutagenesis-A 6-histidine (6XHis) tag was first inserted between the ␤-lactamase signal sequence and the BLIP coding sequence of pG3-BLIP (21) by overlapping PCR mutagenesis (22). The internal mutagenic PCR primer sequences are: BLIPHIS-1 (a top strand primer), 5Ј-CACCACCACCACCACCACGCGGGGGTGATG-ACCGGGGCGAAG-3Ј; and BLIPHIS-2 (a bottom strand primer), 5Ј-C-GCGTGGTGGTGGTGGTGGTGTTCTGGGTGAGCAGCAAAAACAGG-AAGGCA-3Ј. The external PCR primers used to amplify the construct are: PD-bla1 (top strand, N-terminal), 5Ј-CGGGGAGCTCGTTTCTTA-GACGTCAGGTGGC-3Ј; and MALBLI-2 (bottom strand C-terminal), 5Ј-GGGAAATCTAGATTATACAAGGTCCCACTGCCG-3Ј. A SacI site in PD-bla1 and a XbaI site in MALBLI-2 allowed the PCR product to be cloned into SacI-and XbaI-digested pTP123 following treatment of the vector with calf intestinal alkaline phosphatase. The final SacI/XbaI fragment contains, from 5Ј to 3Ј, the TEM-1 ␤-lactamase constitutive promoter, the TEM-1 ␤-lactamase signal sequence, and a 6XHis tag at the N terminus of BLIP followed by the mature BLIP coding sequence. The sequence of this clone was confirmed by the dideoxy chain-termination method and was named pGR32 (Fig. 1). The positioning of this construct in pTP123 allows the N-terminal His-tagged BLIP to be expressed either under the TEM-1 ␤-lactamase constitutive promoter or by induction of the trc promoter with IPTG. The 6XHis tag facilitates the purification of BLIP using an appropriate nickel-or cobalt-based affinity column, while the ␤-lactamase signal sequence enables BLIP to be transported to the periplasmic space, thus eliminating the need to isolate whole cell extracts for BLIP purification.
BLIP and ␤-Lactamase Expression and Purification-Plasmids pGR32, pJP128, and pJP129 were transformed into E. coli RB791 by electroporation. An overnight culture of each was grown shaking in 40 ml of Luria-Bertani (LB) medium at 37°C in the presence of 12.5 g/ml chloramphenicol. The 40 ml of overnight culture were used to inoculate 2 liters of LB medium containing 12.5 g/ml chloramphenicol. The bacteria was then grown shaking at 25°C until A 600 ϭ 1.2. For induction of BLIP, 3 mM IPTG was added to each culture, and the cultures were then allowed to grow an additional 5 h.
Following the 5-h induction, the cells were pelleted and resuspended in 15 ml of sonication buffer (20 mM Tris-HCl (pH 8.0) and 500 mM NaCl). The cells were then sonicated in two batches, and insoluble material was pelleted by centrifugation. The soluble protein in the supernatant was purified over a 4-ml TALON column (CLONTECH) according to the manufacturer instructions. A 4 mM imidizole wash step was utilized to remove protein from the column which bound less tightly than the His-tagged BLIP. BLIP was eluted using an elution buffer consisting of 50 mM imidizole added to the sonication buffer (pH 8.0). Fractions were examined by SDS-PAGE to estimate purity and yield (Fig. 2). Approximately 500 g of Ͼ90% pure BLIP could be isolated for every two liters of culture using this strategy. Quantitative amino acid analysis was performed to calibrate a Bradford assay for determining BLIP and ␤-lactamase concentrations (23).
Wild-type ␤-lactamase and the G238S and E104K extended spectrum mutants were expressed and purified as described previously (24). The location of the two extended spectrum mutations, with respect to Asp-49 and Phe-142 of BLIP, is shown in Fig. 3.
BLIP Inhibition Assay-Varying concentrations of BLIP were incubated with 1 nM ␤-lactamase for 2 h at 25°C. 2 nM of the ␤-lactamase were used in the G238S studies. The enzyme-inhibitor incubation was done in 0.05 M phosphate buffer (pH 7.0) containing 1 mg/ml bovine serum albumin. Following the 2-h incubation, which is sufficient time to achieve binding equilibrium in a small volume, cephaloridine was added at a concentration of at least 10-fold lower than the K m of the ␤-lactamase being tested (e.g. wild-type TEM-1 ␤-lactamase has a K m of approximately 700 M for cephaloridine, therefore 70 M cephaloridine was added to the TEM-1/BLIP incubation). The final volume for the reaction was 0.5 ml. Hydrolysis of cephaloridine was monitored at A 260 on a Beckman DU70 spectrophotometer. The extinction coefficient used for cephaloridine was ⌬⑀ ϭ 10,200 M Ϫ1 cm Ϫ1 (24). Plots of the concentration of free ␤-lactamase versus inhibitor concentration were fit by nonlinear regression analysis to Equation 1, where [E free ] is the concentration of active ␤-lactamase calculated from the measured velocity and the activity and concentration of uninhibited ␤-lactamase, [E o ] is the total ␤-lactamase concentration, and [I o ] is the total inhibitor concentration (25). From the equation, apparent equilibrium dissociation constants (K i *) were determined.

RESULTS
Wild-type BLIP Binding-To determine whether the histidine tag affects BLIP inhibitory activity and to test the activity of the BLIP mutants against TEM-1, the ESBLs, and SHV-1 ␤-lactamase, an inhibitor assay was developed using the cephalosporin cephaloridine as a substrate. Wild-type or mutant BLIP was incubated with a target for 2 h, which is sufficient time to achieve equilibrium. After the 2-h incubation, cephaloridine (at a concentration 10-fold less than the cephaloradine K m for the ␤-lactamase being tested) was added. Monitoring the hydrolysis of cephaloridine at a concentration below K m allowed the concentration of uninhibited ␤-lactamase to be determined without shifting the binding equilibrium. The concentration of free ␤-lactamase was calculated from the cephaloridine activity in the presence of a given quantity of BLIP, the cephaloridine activity in the absence of BLIP, and the known molar concentration of ␤-lactamase being used. Fitting the data obtained when incubating varying concentrations of wildtype, His-tagged BLIP with 1 nM TEM-1 ␤-lactamase resulted in a K i of 0.11 nM (Fig. 4, Table I). This value compares reasonably well with the previously reported value of 0.6 nM found with BLIP purified from S. clavuligerus (16) and suggests that the N-terminal 6XHis tag has little effect on the binding of the inhibitor to the TEM-1 enzyme.
To determine whether wild-type BLIP has similar affinity for extended spectrum ␤-lactamases, the K i of BLIP for two representative ESBLs was determined. The G238S ␤-lactamase mutation is found in many extended spectrum enzymes (26). This single mutation increases the catalytic efficiency of ␤-lactamase for the third generation cephalosporins ceftazidime and cefotaxime approximately 70-and 40-fold, respectively (27). The E104K mutation, likewise, has been found in many extended spectrum ␤-lactamase variants (26). This mutation increases the catalytic efficiency of ␤-lactamase approximately 50-fold for ceftazidime and 10-fold for cefotaxime (28). Both Gly-238 and Glu-104 are located at the binding interface of BLIP and ␤-lactamase (17). Wild-type BLIP was found to have a K i of 0.07 nM for G238S, and a K i of 140 nM for E104K (Figs. 5 and 6, Table I). These values suggest that the G238S mutation has little effect on the binding of wild-type BLIP to ␤-lactamase, whereas the E104K mutation interferes with binding in such a way that the K i increases 1000-fold.
Mutant BLIP Binding-The x-ray structure of the BLIP⅐TEM-1 ␤-lactamase complex shows that BLIP residues Asp-49 and Phe-142 mimic portions of the ␤-lactam PenG when bound to ␤-lactamase (17). This structural mimicry suggests that these residues maintain important interactions in the inhibitory complex. To determine the contribution of these amino acids to inhibition of TEM-1 ␤-lactamase and extended spectrum-hydrolyzing ␤-lactamases, the D49A and F142A mutants were constructed. The K i of each was determined with TEM-1, E104K, and G238S ␤-lactamases (Figs. 4 -6; Table I).
Both the D49A and F142A mutants exhibited an approximately 100 -300-fold increase in K i compared with wild-type BLIP inhibition of TEM-1 ␤-lactamase. The D49A mutant inhibits TEM-1 with a K i of 8.3 nM, whereas the F142A mutant inhibits with a K i of 33 nM. To ascertain whether these mutations affect inhibition in an additive manner, the D49A/F142A double mutant was constructed. However, we were unable to express the double mutant under the same strategy used with wild-type BLIP and the other single amino acid mutants.
The interactions of the wild-type, D49A, and F142A BLIP inhibitors with the ESBLs show that BLIP binds G238S with similar strength as that of wild-type TEM-1, but binds E104K weakly. The K i values found for D49A and F142A mutants with G238S ␤-lactamase were similar to the K i values found for the BLIP mutants binding TEM-1. The D49A BLIP mutant inhibited G238S with a K i of 9.4 nM. Therefore, as with TEM-1, the D49A mutation reduced inhibition 100-fold. Likewise, the F142A BLIP bound G238S ␤-lactamase approximately 800-fold weaker, with a K i of 55 nM. The fact that these two substitutions in BLIP have a similar effect on the K i values for the TEM-1 and G238S ␤-lactamases suggests that BLIP inhibits G238S in much the same way it inhibits TEM-1.
Further experiments showed that Asp-49 and Phe-142 do not contribute to the binding of E104K ␤-lactamase as they do to the binding of TEM-1 and G238S. The K i of BLIP D49A with E104K ␤-lactamase is 1.5 M, which is only a 10-fold increase compared with that of the wild-type BLIP interaction with E104K. This suggests that BLIP residue Asp-49 is not as critical to inhibition of E104K ␤-lactamase as it is to the other enzymes tested. This observation is even more pronounced with the BLIP F142A substitution in that there was little change from the wild-type BLIP K i with the BLIP F142A mutant inhibiting E104K (K i (F142A) ϭ 240 nM). Therefore, in contrast to their effect on TEM-1 and G238S binding, the D49A and F142A substitutions do not have as detrimental an effect on the K i for E104K ␤-lactamase.
BLIP Binding to SHV-1 ␤-Lactamase-The amino acid sequence of SHV-1 is 68% identical to TEM-1 ␤-lactamase (7). How this similarity corresponds to structure homology is unknown as the crystal structure of SHV-1 ␤-lactamase has not yet been solved. Both the TEM-1 and SHV-1 enzymes hydrolyze a similar profile of penicillins and cephalosporins. It is not clear whether the homology between the two enzymes implies  that BLIP should inhibit both equally well. It may be that even slight differences in the three-dimensional structure of SHV-1 compared with TEM-1 would effect BLIP binding considerably. These issues were addressed by performing an additional inhibitory assay with wild-type BLIP and SHV-1 ␤-lactamase. SHV-1 was purified to greater than 90% homogeneity (data not shown) and was bound to increasing concentrations of wildtype BLIP. The K i of BLIP for SHV-1 was found to be 1.0 M, 9,000-fold higher than what was found for TEM-1 (Fig. 7, Table  I). Therefore, the high degree of identity between these enzymes does not translate to similar binding properties by BLIP.

DISCUSSION
The development of novel inhibitors for ␤-lactamases as well as penicillin-binding proteins would provide new options for the treatment of bacterial infections. As ␤-lactamases are be-lieved to have evolved from PBPs, it is conceivable that minor changes in the structure of BLIP could enable it to bind and inhibit PBPs (29). It has already been observed that wild-type BLIP weakly inhibits PBP5 from Enterococcus (16). Understanding how the amino acid sequence of BLIP encodes its tight binding affinity for certain ␤-lactamases, and its weaker affinity for PBPs, would facilitate the development of novel inhibitors with potent activity for the ESBLs and for the PBPs. The identification of amino acids most responsible for inhibition and those critical for binding specificity will pinpoint the residues that need to be targeted for engineering BLIP mutants with higher inhibitory activity for different ␤-lactamases and PBPs.
The crystal structure of the BLIP⅐TEM-1 complex suggests that two BLIP amino acids, Asp-49 and Phe-142, are critical for  (17). These residues are the primary candidates for mutational analysis to distinguish whether "hot spots" consisting of a subset of these amino acids are involved in specificity and binding or if all of these residues contribute to the binding of BLIP to TEM-1 ␤-lactamase. The first step in being able to identify amino acids important for BLIP specificity and inhibitory activity is to develop an expression system to produce functional BLIP. BLIP expressed in its native S. clavuligerus produces large quantities of protein, whereas expression in another Streptomyces species, Streptomyces lividans, produces limited quantities of BLIP (15,30). Successful expression of soluble BLIP in E. coli would facilitate the study of BLIP mutants and would also allow protein engineering techniques to be performed.
Histidine-tagged proteins are able to be purified in a relatively simple manner, while usually maintaining the native activity of the tagged protein. Therefore, an expression system centered around an N-terminal 6XHis-tagged BLIP was constructed. Expression is directed by the inducible trc promoter, and a cat gene is inserted into the ␤-lactamase gene of the plasmid to avoid possible complex formation during purification of BLIP (Fig. 1). This system enabled BLIP to be purified to Ͼ90% homogeneity in one step (Fig. 2). Wild-type His-tagged BLIP was found to have a K i of 0.11 nM. This value is slightly lower than the previously calculated value of 0.6 nM for BLIP isolated from S. clavuligerus (16). This difference could be attributed to the manner in which the K i was calculated, and confirms that the N-terminal His tag has no effect on BLIP binding.
Once expression of functional BLIP was achieved, the roles of BLIP residues in the inhibition of different targets could be determined. The crystal structure of BLIP with TEM-1 ␤-lactamase shows that Asp-49 of BLIP makes strong hydrogen bond contacts with four conserved residues in the TEM-1 active site pocket: Ser-130, Lys-234, Ser-235, and Arg-244 (17). These four amino acids are involved in the binding and catalysis of ␤-lactam antibiotics and are conserved in all class A ␤-lactamases. Mutation of the aspartic acid to an alanine removes the carboxylate moiety that serves as a hydrogen bond acceptor for the four active site TEM-1 residues. Elimination of the carboxylate reduced the inhibitory activity of BLIP approximately 100-fold, indicating residue Asp-49 does make an important contribution to BLIP inhibition of TEM-1 ␤-lactamase.
The crystal structure also leads to the prediction that Phe-142 is also important for inhibition of TEM-1 ␤-lactamase (17). Phe-142 is in contact with ␤-lactamase residues Glu-104, Tyr-105, Asn-170, Ala-237, Gly-238, and Glu-240 in the complex. As in the case of Asp-49, most of these residues are either conserved in class A ␤-lactamases or are involved in catalysis. It was also found that in other protein complexes, such as between human growth hormone and its receptor, the most critical interactions are hydrophobic (31). Therefore, the contribution of Phe-142 to binding and inhibition was examined. The F142A substitution removes the hydrophobic side chain that mimics the benzyl group in PenG from the TEM-1⅐PenG complex. This change also increases the K i approximately 100-fold, which suggests that the interactions mediated by Phe-142 are important for inhibitor binding, and that they are similar in magnitude to the contributions made by Asp-49. The D49A/ F142A double mutant was constructed to test additivity between these two residues, but the protein was not expressed.
The ability of wild-type BLIP and the D49A and F142A mutants to inhibit two extended spectrum ␤-lactamase mutants was also examined. G238S, the ␤-lactamase mutation found in TEM-19, and E104K, the mutation found in TEM-17, were the two representative extended spectrum mutants tested with BLIP and the BLIP mutants (26). The prevalence of the G238S and E104K substitutions in many of the extended spectrum ␤-lactamases makes these two single mutants ideal candidates for study. Interestingly, wild-type BLIP, D49A, and the F142A BLIP mutant each inhibited G238S at similar levels to that which they inhibited wild-type TEM-1 ␤-lactamase. According to the crystal structure, the only contact made with FIG. 7. Determination of wild-type BLIP K i for SHV-1 ␤-lactamase. BLIP inhibitory activity is expressed as the remaining concentration of free ␤-lactamase at varying inhibitor concentrations. SHV-1 concentration is 1 nM, and the cephaloridine concentration is 70 M for all experiments. The apparent K i was determined as in Fig. 4.
Gly-238 of TEM-1 is by Phe-142. The fact that no change in the inhibition profile was observed between the ␤-lactamase enzymes suggests that this contact between Gly-238 and Phe-142 is not critical for BLIP binding and inhibition. If this interaction played a role in BLIP inhibition, then replacement of the glycine side-chain at position 238 of TEM-1 would have resulted in an increased K i with wild-type BLIP. However, the possibility that the substituted serine at position 238 makes new interactions with BLIP, and corrects exactly for the loss of interaction with Phe-142, cannot be excluded.
In contrast to BLIP binding to G238S, significant changes in the inhibitory profile were observed when E104K was used as the target ␤-lactamase. Wild-type BLIP inhibited E104K approximately 1000-fold worse than for TEM-1, suggesting that the interactions made between BLIP and Glu-104 are critical for wild-type levels of activity. This result indicates that some, or all, of the BLIP residues interacting with Glu-104 (BLIP-Glu-73, Lys-74, Phe-142, and Tyr-143) are making important interactions. However, results from the BLIP F142A/␤-lactamase E104K binding experiment suggest that the lysine substitution at position 104 of ␤-lactamase disrupts critical interactions made by Phe-142 of BLIP. This disruption of the BLIP Phe-142 interaction with ␤-lactamase results in the pronounced loss of inhibition observed with the E104K enzyme.
It is apparent from this study that while BLIP residues Asp-49 and Phe-142 are critical for wild-type BLIP inhibitory levels, the D49A and F142A variants still bind TEM-1 ␤-lactamase with nanomolar affinity. Therefore, other interactions must also contribute to the strong levels of inhibition observed. The identification of the epitopes responsible for the remaining binding energy will facilitate the engineering of tighter, smaller inhibitors for these ␤-lactamases.
Determination of the K i of BLIP with SHV-1 ␤-lactamase shows that even though TEM-1 and SHV-1 are both class A ␤-lactamases and are 68% identical, the interactions which make BLIP a tight inhibitor of TEM-1 are not conserved with SHV-1. Although no crystal structure is available for SHV-1, the level of identity between TEM-1 and SHV-1 suggests that both enzymes share a similar protein fold. However, the fact that BLIP is a poorer inhibitor of SHV-1 shows that small differences between the ␤-lactamases are significant with respect to BLIP binding.