ASSAYS FOR IMPROVING AUTOMATED ANTIMICROBIAL SUSCEPTIBILITY TESTING ACCURACY

Abstract

Phenotypic antimicrobial susceptibility testing (AST), the gold-standard diagnostic that indicates whether an antimicrobial will be clinically effective, often suffer the slowest times-to-result for the most resistant pathogens. Here we introduce novel assays to be performed in parallel with standard AST assays that enable rapid, same-shift reporting of AST results for a plurality of pathogens. The assays developed here are further capable of detecting resistance to carbapenems, the most powerful class of beta-lactams commonly used as “last-resort” antimicrobials.

Claims

1. A method for performing automated antimicrobial susceptibility testing of a carbapenem antimicrobial comprising (a) performing a dilution assay comprising inoculating a microbial inoculum of concentration C.sub.0 into a plurality of fluid wells defining a dilution series for the carbapenem antimicrobial; and measuring in each of the plurality of fluid wells a signal associated with microbial growth; and (b) in parallel with the dilution assay, performing a carbapenemase assay comprising measuring a signal associated with carbapenem degradation in a first well comprising the carbapenem antimicrobial and a microbial inoculum with concentration C.sub.R, wherein C.sub.R>C.sub.0; measuring a signal associated with carbapenem degradation in one or more second wells comprising different contents than the first well and thereby acting as controls for the first well; and (c) comparing the signals measured from the carbapenemase assay wells; and (d) combining the data derived from the dilution assay with that derived from the carbapenemase assay to define and label the microorganism as carbapenem susceptible or carbapenem resistant.

2. The method of claim 1, further comprising comparing the signal measured in each of the plurality of fluid wells to define a minimum inhibitory concentration (MIC) of the carbapenem antimicrobial.

3. The method of claim 1, wherein C.sub.R>10×C.sub.0.

4. The method of claim 1, wherein carbapenem degradation is determined by a signal associated with an indicator.

5. The method of claim 4, wherein the indicator is a fluorescent or optical pH indicator.

6. The method of claim 5, wherein the pH indicator is one or more of fluorescein, pyranine, tinopal, or a derivative thereof.

7. The method of claim 1, wherein the carbapenem is selected from imipenem and biapenem and their derivatives.

8. The method of claim 1, wherein the first well further comprises ionic zinc.

9. The method of claim 8, wherein ionic zinc is one or more of zinc sulfate, zinc chloride or zinc hydroxide.

10. The method of claim 1, wherein the assay wells are optically interrogated after less than or equal to 3, 6, 8, 10 hours of incubation under conditions that promote microbial growth.

11. The method of claim 1, wherein the one or more second wells comprise (a) a first control well comprising a microbial inoculum at C.sub.R, a pH indicator and ionic zinc; (b) a second control well comprising a pH indicator and ionic zinc; and (c) a third control well comprising the carbapenem, a pH indicator and ionic zinc.

12. The method of claim 1, wherein the carbapenemase assay comprises one or more more additional wells comprising a beta-lactamase inhibitor.

13. An automated antimicrobial susceptibility test assay cartridge for a beta-lactam/beta-lactamase inhibitor antimicrobial comprising: a. two or more assay wells comprising an antimicrobial dilution series for a beta-lactam antimicrobial inoculated at a concentration C.sub.0 with a microbial inoculum; and b. a series of assay wells for a beta-lactamase inhibitor assay probe inoculated at a concentration C.sub.R from the same microbial inoculum comprising: i. one or more assay wells comprising a beta-lactamase probe; and ii. two or more assay wells comprising the same beta-lactamase probe and a beta-lactamase inhibitor at one or more concentrations.

14. The cartridge of claim 13, wherein the probe is chromogenic or fluorometric.

15. The cartridge of claim 13, wherein the probe is nitrocefin.

16. The cartridge of claim 13, wherein the inhibitor is one or more of clavulanate, cloxacillin, tazobactam, avibactam, relebactam and vaborbactam.

17. An automated antimicrobial susceptibility test assay cartridge for one or more carbapenems comprising: a. two or more assay wells comprising an antimicrobial dilution series for a carbapenem antimicrobial; b. a series of assay wells for a carbapenemase assay comprising: i. one or more assay wells comprising a beta-lactamase probe; and ii. two or more assay wells comprising the same beta-lactamase probe and two or more inhibitors selected from the list of clavulanate, cloxacillin, and tazobactam; and adapted for inoculation with microorganisms from a single microbial inoculum, wherein the results of the carbapenemase assay influence algorithmic MIC determinations of one or more carbapenems.

18. An automated antimicrobial susceptibility test assay cartridge for one or more carbapenems and/or beta-lactam/beta-lactamase inhibitors comprising: a. two or more assay wells comprising an antimicrobial dilution series for a carbapenem antimicrobial; b. a series of assay wells for a carbapenemase assay comprising: i. one or more assay wells comprising saline, a carbapenem, ionic zinc, and a pH indictor; ii. one or more assay wells comprising saline, a carbapenem, ionic zinc, a pH indictor, and a microbe sample comprising <1×10.sup.8 CFU intact microbes; iii. one or more assay wells comprising saline, ionic zinc, and a pH indictor; iv. one or more assay wells comprising saline, ionic zinc, a pH indictor, and a microbe sample comprising <1×10.sup.8 CFU intact microbes; and adapted for inoculation with microorganisms from a single microbial inoculum and wherein the results of the carbapenemase assay influence algorithmic MIC determinations of one or more carbapenems.

19. The cartridge of claim 18, wherein the carbapenem is imipenem or biapenem.

20. The cartridge of claim 18, wherein the pH probe is fluorescein.

21. The cartridge of claim 18, wherein the ionic zinc is selected from zinc sulfate or zinc chloride.

22. The cartridge of claim 18, wherein one or more additional wells comprise beta lactamase inhibitors.

23. A method for performing multi-assay rapid antimicrobial susceptibility testing sequences, the method comprising (a) inoculating two or more different concentrations of a sample comprising a microorganism derived from a clinical sample into a plurality of wells of a test panel, at least a portion of the plurality of wells containing one or more antimicrobials of a plurality of antimicrobials for inoculation of the sample; (b) loading the test panel into an automated rapid antimicrobial susceptibility testing system for performing a multi-assay testing sequence; and (c) operating the testing system to move the loaded test panel to an incubation assembly comprising a nest assembly adapted to: i) house at least one test panel, and ii) facilitate incubation of one or more test panels in order to undergo the multi-assay testing sequence the incubation assembly comprising: an agitation system configured to generate a repeated motion of the nest assembly; incubate and agitate the inoculated sample in the incubation assembly; at least once, periodically measure an amount of sample growth in a plurality of control wells of the plurality of wells; responsive to determining that a level of growth in the control wells meets or exceeds a threshold level of growth, stop incubation; perform a carbapenemase and/or beta-lactamase assay for one or more carbapenems and/or beta-lactam/beta-lactamase inhibitors on incubated samples in the test panel; perform one or more end point assays on incubated samples in the test panel; measure an optical output from the sample in the plurality of wells of the test panel, the optical output corresponding to an amount of the microorganism remaining in each of the plurality of wells; and report at least one of: a minimum inhibitory concentration of and/or a qualitative susceptibility interpretation for the microorganism remaining in each of the plurality of wells and the plurality of antimicrobials, such that the results of the carbapenemase and/or beta-lactamase assays influence algorithmic MIC determinations of one or more carbapenems and/or beta lactams.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0099] FIG. 1 depicts exemplary data from beta-lactamase inhibitor assay tests of clavulanic acid and tazobactam with three microbial strains.

DETAILED DESCRIPTION

Overview

[0100] Generally, this disclosure focuses on beta lactamase and/or carbapenemase assays that are performed in parallel with standard phenotypic AST assays on an automated analyzer, e.g., as described in U.S. Pat. No. 10,161,948 to Vacic et al, which is incorporated by reference herein for all purposes. In several exemplary embodiments described herein, both types of assays are run on the same cartridge such that only a single cartridge is required for all tests for the automated analyzer.

[0101] Those of skill in the art will appreciate that the integration of these two assays into a single automated process involves several departures from existing industry standard processes. First, commercially available AST systems generally require all wells of the multi-well AST testing vessel (e.g., microtiter plate, multi-well cassette, etc.) be loaded with substantially equivalent quantities of microbial sample, so as to ensure that differences in growth observed across antimicrobial dilution series are not due to differences in initial loading of the wells. By contrast, in some of the embodiments of this disclosure, the time-to-result for carbapenemase and beta-lactamase assays is reduced by increasing the quantity of microorganisms in the reservoirs for these tests relative to the lower microorganism concentration present in the plurality of wells in the panel that are used for MIC determinations, as is discussed in greater detail below.

[0102] Second, existing automated AST systems generally take one of two approaches to carbapenemase or beta-lactamase testing: first, in an automated sample in which beta lactam or carbapenem resistance is suspected based upon algorithmic interpretations of growth curve results, the incubation period will be continued until the algorithmic interpretation of the growth curves is confident the resistance either is or is not present. This may add significant time before AST results are available due to the inducible nature of some resistance mechanisms, as known to those skilled in the art. This growth curve-based approach has the advantage of automation, but it can extend the time-to-result by hours, and in laboratories staffed in discontinuous shifts (e.g., without an overnight shift), extended assay times may shift the result from the same day to the following morning. In a second approach, a carbapenemase or beta lactamase assay is performed offline, independently of the automated AST assay. This approach may allow a result to be obtained and reported sooner, but it may also be more resource intensive than a fully automated process, and the separation of the AST and resistance assays raises a risk that data from both assays will not be associated with the same patient record during the period where the AST results are interpreted and prescribing decisions are made. The systems and methods presented herein leverage the advantages of both approaches yet can reduce or eliminate delays in delivery of results and integrate data from both assays into a single output. In the preferred embodiment, the results from the carbapenemase and/or beta-lactamase assays are used to influence the algorithmic MIC determinations of results from the appropriate carbapenem and beta lactam dilution series. In additional embodiments, the results from the carbapenemase and/or beta-lactamase assays may be provided to the user.

[0103] In some embodiments of this disclosure, beta-lactamase or carbapenemase activity is measured indirectly, for example by detecting a change in pH caused by the liberation of free acids by degradation of the antimicrobial. Those of skill in the art will appreciate that pH in particular may be affected by factors other than enzyme-catalyzed degradation due to microbial resistance. The potential for these other factors to confound the results of AST and carbapenemase/beta-lactamase probe-based assays can be reduced by the inclusion of one or more carbapenemase or beta-lactamase inhibitors. By comparing the signal from an “inhibited” well to one containing the probe and microbes alone, the efficacy of the inhibitor may be tested. For example, tazobactam's inhibitory effects on a clinical Escherichia coli sample may be determined by setting up two wells containing E. coli and nitrocefin and adding tazobactam to the second well. By extending the dilution series, minimum inhibition concentrations may be obtained.

[0104] In some embodiments, such tests would be performed by an automated AST analyzer, in parallel with AST tests, and the combined results would add increased accuracy to the tests. Since some beta-lactamases may be induced by the presence of beta-lactams, it may be advantageous to perform beta-lactamase inhibitory testing after microbial growth in the presence a beta-lactam. In the case where the presence of the beta-lactam inhibits microbial growth, no useful data from beta-lactamase inhibitory testing may be obtained.

[0105] Any beta-lactamase probe may be utilized for these assays. Engineered probes with less-broad activity may also be used. However, the key to the assay concept is that the probe is sufficiently broad to ensure a plurality of potential beta-lactamase variants that can be inhibited by the inhibitor will degrade the probe for which the assay results will influence the MIC-determining algorithm of the AST platform.

[0106] By combining inhibitors in a single test well, information about the presence of beta-lactamase classes may be obtained. In this case, inhibitors may include drugs and potential drugs as well as agents with unacceptable toxicities for human use, such as EDTA, which is well-known to inhibit metallo-beta-lactamases (MBLs) but is toxic. An assay for carbapenemase activity may be developed by including inhibitors of non-carbapenemase ESBLs in one well, together with the clinical sample and nitrocefin, and comparing this to a sample comprising the clinical sample and nitrocefin without the inhibitors. Such inhibitors include, but are not limited to, clavulanate, cloxacillin, and tazobactam. Additional accuracy may be obtained by adding a third well comprising known inhibitors of carbapenems, including but not limited to avibactam, vaborbactam, and EDTA.

[0107] Information on carbapenemase presence may also be obtained by utilizing a carbapenem antimicrobial as a competitor to a beta-lactamase probe. In this case, one or more inhibitors may be added to inhibit non-carbapenemase beta-lactamase activity in all assay wells, in addition to a clinical microbe sample and a broad-spectrum probe, such as nitrocefin. Note a more narrow-spectrum probe may be advantageous here.

[0108] One well would then comprise a high concentration of a carbapenem, which would not be present in the second well. By comparing the rate of nitrocefin degradation in each well, the presence of a carbapenemase may be inferred: if the carbapenem retards the generation of the signal resulting from nitrocefin degradation, a carbapenemase is likely present. In alternative embodiments, oxacillin may be used in one or more additional wells as a competitor, specifically for the detection of “OXA” carbapenemases.

[0109] Exemplary carbapenemase assays may also utilize the acidimetric technique. Currently disclosed acidimetric carbapenemase assays utilize >10.sup.9 CFU/mL microbes, complicating their use in automated AST testing following standard laboratory workflows, where <5×10.sup.7 CFU/mL microbes are commonly available. Furthermore, standard laboratory workflows for AST testing utilize intact microbes of which a plurality are viable. These standard workflow requirements challenge previously-disclosed carbapenemase assays, which rely on greater microbial quantities and lysed microbes. These requirements are due to the fact that many carbapenemases may be intracellular and may be expressed in low concentration at early timepoints due to their inducible nature.

[0110] Here we demonstrate the surprising finding that acidimetric carbapenemase assays may be performed using less than one-tenth ( 1/10.sup.th) of the number of intact microbes by utilizing four assay wells to provide an appropriately normalized result. Additionally, the assay utilizes a fluorometric, rather than a colorimetric, pH probe, increases assay incubation times to 4 hours, and increases carbapenem concentrations to >1 mg/mL.

[0111] Here we further demonstrate the counter-intuitive finding that faster AST results may be made available by utilizing different microorganism concentrations in different test reservoirs of an automated AST panel. This is particularly surprising because existing automated AST platforms, including the Vitek2® (bioMérieux), MicroScan™ (Danaher/Beckman-Coulter), Phoenix® (Becton-Dickinson), and SensiTitire™ (ThermoFisher) provide dedicated hardware specifically to ensure all reservoirs on the panel inoculated with microorganisms (e.g. all reservoirs except the contamination control that is inoculated with zero microorganisms) receive a substantially equivalent quantity of microorganisms. This is the case because MICs are relative measurements, in which the growth in each reservoir of a dilution series of an antimicrobial is compared with the growth in other reservoirs of the same dilution series (and may be compared with that for other antimicrobials).

[0112] In contrast, here we demonstrate that AST result accuracy may be enhanced through the inoculation of two or more different quantities of microorganisms in different automated AST panel reservoirs. In particular, microorganisms inoculated into reservoirs comprising dilution series of antimicrobials may be of a substantially equivalent concentration, C.sub.0, and microorganisms inoculated into specific resistance mechanism-determining assays may be of a different concentration, C.sub.R. In particular, the concentrations may be such that C.sub.R=αC.sub.0, where α is preferably greater than 1 and most preferably greater than 10.

[0113] In an exemplary embodiment, the assay comprises a minimum of 4 wells and utilizes a sample comprising intact microbes. The first well (Well.sub.1) comprises a medium such as a buffered saline solution or a nutrient broth, a carbapenem, ionic zinc, and a pH indictor. The second well (Well.sub.2) comprises the medium, a carbapenem, ionic zinc, a pH indictor, and a microbe sample comprising <1×10.sup.8 CFU microbes. The third well (Well.sub.3) comprises the medium, ionic zinc, and a pH indictor. The fourth well (Well.sub.4) comprises the medium, ionic zinc, a pH indictor, and a microbe sample comprising <1×10.sup.8 CFU microbes. The following formula can then be utilized to normalize the result:

Normalized Activity=(Well.sub.1−Well.sub.2)−(Well.sub.3−Well.sub.4)   [Formula 1]

[0114] The above says that imipenem degradation is the change in pH (e.g. loss of signal) due to the presence of microbes less the change in pH (e.g. loss of signal) due to microbes in solution only.

[0115] In additional embodiments, acidimetric methods may be utilized for testing inhibitors. For example, the addition of an inhibitor with known carbapenemase activity, such as avibactam or vaborbactam, to an additional well of the carbapenemase assay described above may serve as an assay for inhibitor activity. Typically, high sensitivity assays that are based on amplification (e.g., catalytic) can be performed only once since chemistries necessary for those assays usually destroy the target microorganism. Thus, the systems and methods described herein typically use two types of assays to address this issue. In some cases, a preliminary (e.g., checkpoint) assay can be performed first and can be repeated periodically to interrogate growth of uninhibited microorganisms (i.e., without antimicrobial presence). These checkpoint assays can be performed in wells referred to herein as control wells. Examples of typical control wells are a growth well containing microorganisms in nutrient broth and a contamination control well containing nutrient broth only. The system interrogates growth/no growth optically (e.g., absorbance, fluorescence metabolic dye, etc.) and once a particular ratio and/or kinetic change between the control wells is achieved and detected, one or more end point assays (e.g., an amplification assay or growth assay) can be initiated on samples disposed in other portions of the test panel (e.g., the rest of, or the entire, test panel). The samples, for example, can include microorganisms originating from a clinical sample. Additional wells, such as wells containing microorganisms in saline or other media that does not promote growth of microorganisms (i.e., due to lack of nutrients) can be utilized for growth check and MIC determination. These wells can contain concentrations of microorganisms that are similar to the starting sample and referred to as “frozen in time” (e.g., FTT) control.

[0116] In some cases, the systems and methods described herein can be implemented to provide faster testing than some conventional systems. For example, though some automated systems may speed time to obtain results, the time-to-results for carbapenemase-expressing strains rarely meet the ˜6-hour definition of “same-shift” results for many clinical laboratories. Because of this slow time-to-results and because AST results are complex and may utilize expert interpretation for clinical action, such conventional systems can result in a day delay between the onset of susceptibility testing and clinical action for patients infected with these difficult-to-treat samples.

TABLE-US-00001 TABLE 1 Carbapenemase assay results for 26 microbial strains within 4 hours. Known Other Carba- Carba- known penemase Species Strain penemase resistance Activity E. coli CDC CNP 73 KPC 3480.1845 E. coli CDC CRE 26 NDM 41326.61 E. coli CDC CRE 03 KPC-3 7149.729 E. coli CDC CRE 07 NDM 27447.359 E. coli CDC BIT 01 KPC 8803.9255 E. coli CDC CNP 38 NDM 19534.8825 E. coli CDC CRE 17 NDM 36400.9235 E. coli CDC CNP 27 none ESBL + −2401.2335 E. coli CDC CNP 54 none Amp beta −1950.443 lactamase E. coli SML 19 none −2540.0545 E. coli SML20 none −2734.3825 E. coli AR0346 none ESBL + −2560.861 E. coli AR0348 none −2960.1665 E. coli AR0349 none ESBL + −2861.008 E. coli AEL18 none −1551.4885 E. coli AEL19 none −696.4825 E. coli AEL 20 none −882.3165 E. coli TRICORE 71 none −1138.2225 E. coli TRICORE 72 none −2554.797 E. coli 25922 none −2106.4115 K. CDC CNP 03 IMP 3662.106 pneumoniae K. CDC CNP 08 OXA-181 1323.2385 pneumoniae K. CDC CNP 09 VIM 2315.76 pneumoniae K. 700603 none −2659.8725 pneumoniae K. SML86 none −1861.1485 pneumoniae K. SML87 none −3075.4205 pneumoniae

TABLE-US-00002 TABLE 2 Carbapenemase assay results for various concentrations of imipenem within 4 hours. Known Carbapenemase Activity in various concentrations of imipenem Species Strain Carbapenemase 50 μg/mL 125 μg/mL 250 μg/mL 500 μg/mL 1250 μg/mL 2500 μg/mL E coli ATCC 25922 none −70.5 −522 −599.5 −558.5 −463.5 −282 K pneumoniae ATCC BAA-2342 KPC 1817 4702.5 4270.5 3708 3251 3114.5 E coli AR 114 KPC3 930.5 3793.5 3342.5 2784.5 2173.5 1825.5 E coli AR 69 NDM 1111 3367 2827 2300.5 1908.5 1902.5 K pneumoniae AR 34 IMP 1034.5 2776.5 2456.5 1807.5 1148 637.5

TABLE-US-00003 TABLE 3 Carbapenemase assay performed with dried reagents. Known Carbapenemase Species Strain Carbapenemase Activity E coli ATCC 25922 none −6116 K pneumoniae ATCC BAA-1705 KPC2 19742 K pneumoniae ATCC BAA-2814 KPC3 18057 K pneumoniae ATCC 700603 none −2026

EXAMPLES

[0117] Example 1. Microbes were prepared by diluting colonies into saline to reach a McFarland value of 0.5, which was verified using a spectrophotometer. This was diluted 1:20 into saline and 50 μI of inoculum was added to wells of a 384-well plate. For each strain, 25 μI of solutions containing either 0.25 mg/mL nitrocefin alone (Nitro), 0.25 mg/mL nitrocefin and 4 μg/mL clauvulanate (Nitro Clay), or 0.25 mg/mL nitrocefin and 4 μg/mL tazobactam (Nitro Tazo) were added to duplicate wells. Inoculated plates were incubated at 35° C., shaking at 150 rpm for 4 hours. Absorbance was read at 490 nm and values shown in FIG. 1 are the absorbance in the indicated condition minus the absorbance from microbial inoculum alone without nitrocefin. A greater value indicates microbial ability to degrade nitrocefin, and thus, the presence of a beta lactamase. A reduction in absorbance value in the presence of an inhibitor indicates the beta lactamase present is sensitive to the inhibitor. For the data presented here, EC 25922 does not contain a beta lactamase, but the other two organisms do. Further, the inhibitors are both active against each of the strains with known beta lactamases.

[0118] Example 2. Microbes were prepared by diluting colonies into saline to reach a McFarland value of 0.5, which was verified using a spectrophotometer. 100 μL of the microbe solution was added to each of 2 wells in a 96-well plate. 100 μL of saline was added to 2 separate wells on the plate. 100 μL of an imipenem solution containing 5 mg/mL imipenem, 0.1 mM ZnSO.sub.4, and 10 μM Fluorescein Na salt, pH 8, was added to one of the microbe-containing wells (Well.sub.2) and one of the saline wells (Well.sub.1). 100 μL of a negative control solution containing 0.1 mM ZnSO.sub.4, and 10 μM Fluorescein Na salt pH 7.5, was added to one of the microbe-containing wells (Well.sub.4) and one of the saline wells (Well.sub.3). Plates were incubated, shaking at 35° C. for 4 hours and fluorescence of the wells was read at Em=490 nm/Ex=510 nm. The values reported were calculated to determine the hydrolysis of imipenem by the microbes using Equation 1. These data are tabulated in Table 1.

[0119] Example 3. Microbes were prepared by diluting colonies into saline to reach a McFarland value of 0.5, which was verified using a spectrophotometer. 25 μL of the microbe solution was added to each of 2 wells in a 384-well plate. 25 μL of saline was added to 2 separate wells on the plate. 25 μL of an imipenem solution containing imipenem, 0.1 mM ZnSO.sub.4, and 10 μm Fluorescein Na Salt, pH 8, was added to one of the microbe-containing wells (Well.sub.2) and one of the saline wells (Well.sub.2). 25 μL of a negative control solution containing 0.1 mM ZnSO.sub.4, and 10 μM Fluorescein Na Salt, pH 7.5, was added to one of the microbe containing wells (Well.sub.4) and one of the saline wells (Well.sub.3). Plates were incubated, shaking at 35° C. for 3 hours and fluorescence of the wells was read at Em=490/Ex=510. This organization of the assay was performed multiple times, varying the concentration of imipenem in solution. The assay was run with final imipenem concentrations of 50 μg/mL, 125 μg/mL, 250 μg/mL, 500 μg/mL, 1250 μg/mL, and 2500 μg/mL. The values reported were calculated to determine the hydrolysis of imipenem by the microbes using Equation 1. These data are tabulated in Table 2.

[0120] Example 4. Microbes were prepared by diluting colonies into saline to reach a McFarland value of 0.5, which was verified using a spectrophotometer. 100 μL of the microbe solution was added to each of 2 wells in a vacuum-dried 96-well plate (Well.sub.2, Well.sub.4). 100 μL of saline was added to 2 separate wells on the plate (Well.sub.1 and Well.sub.3). Following addition of microbial or saline solutions, the final concentrations of the contents in each well were: 300 μg/mL imipenem, 0.05 mM ZnSO.sub.4, and 10 μM Fluorescein Na Salt, pH 7.8, (Well.sub.1, Well.sub.2); only 0.05 mM ZnSO.sub.4 and 10 μM Fluorescein Na Salt, pH 7.8, were present in the third and four wells (Well.sub.3, Well.sub.4). Plates were incubated, shaking at 35° C. for 4 hours and fluorescence of the wells was read at Em=490/Ex=510. The values reported were calculated to determine the hydrolysis of imipenem by the microbes using Equation 1. These data are tabulated in Table 3.