WO2012113699A1

WO2012113699A1 - Mass spectrometric determination of the inhibitory effect of substances on beta-lactamases

Info

Publication number: WO2012113699A1
Application number: PCT/EP2012/052662
Authority: WO
Inventors: Markus Kostrzewa; Ulrich Weller
Original assignee: Bruker Daltonik GmbH
Current assignee: Bruker Daltonics GmbH and Co KG
Priority date: 2011-02-23
Filing date: 2012-02-16
Publication date: 2012-08-30
Anticipated expiration: 2013-08-23
Also published as: DE102011012060A1

Abstract

The invention provides a method for determining the inhibitory effect of a substance on beta- lactamases, where the substance is mixed with a beta-lactamase and a substrate which has a beta-lactam ring, and after a given reaction time span a mass spectrum is acquired and evaluated as to whether the substrate has been enzymatically modified by the beta-lactamases.

Description

Mass Spectrometric Determination of the Inhibitory Effect of Substances on Beta- Lactamases

Field of Invention

[0001] The invention relates to a method to determine the inhibitory effect of test substances on beta-lactamases and particularly the inhibitory efficiency of the inhibitor of an antibiotic combination drug comprising a beta-lactam antibiotic and an inhibitor.

Prior Art

[0002] In general parlance, the term antibiotic usually means a pharmacologically active substance for the treatment of bacterial infectious diseases. The successes of penicillin, but also the appearance of the first resistances, led researchers to search for and discover many more antibiotics.

[0003] Penicillin is a beta-lactam antibiotic. The beta-lactam antibiotics are a group of antibiotics which have a four-membered beta-lactam ring in their structural formula. They bond to the penicillin binding protein (PBP), a peptidoglycan transpeptidase which is present in bacteria in different variants. This enzyme is necessary for the formation and renewing of a rigid cell wall in the division or growth phase of bacteria. The bonding of a beta-lactam antibiotic to the PBP causes the PBP to become ineffective. The beta-lactam ring of the antibiotic represents the attachment motif. The attachment prevents the re-synthesis of the cell wall; the bacteria are therefore unable to divide, but live on initially until their growth leads to a sufficiently large number of cell wall lesions to cause the death of the cell. The division and growth of human cells are not impeded, however, because human cells have only a cell membrane, but no cell wall and have therefore no corresponding transpeptidase.

[0004] Ever since penicillin has been used as a pharmacologically active substance, bacterial strains have increasingly developed various types of resistance, i.e. they have acquired charac- teristics which allow them to weaken or completely neutralize the effect of antibiotic substances. Resistances are now widespread; in the USA, around 70% of the infectious pathogens acquired in hospitals are resistant to at least one antibiotic. Patients are often infected with bacterial strains which are resistant to several antibiotics (multi-resistance). Problematic pathogens are mainly the methicillin-resistant Staphylococcus aureus (MRSA), Pseudomonas spec, Escherichia coli with ESBL resistance and Mycobacterium tuberculosis. Estimates by the CDC (Center for Disease Control and Prevention) assume two million hospital-acquired infections in the USA in 2004, with around 90,000 deaths. The reasons for the increase in the resistances are manifold: irresponsible prescription of antibiotics, even when not necessary; courses of treatment which are irresponsibly broken off, irresponsible, often purely preventative usage in agriculture and animal husbandry. All these types of behavior assist in the selection and growth of the resi stant bacterial strains compared to the non-resistant strains. [0005J The increase in resistances means that the success of a therapy for bacterial infections often depends on an effective first administration of an antibiotic. Acute situations, such as a sepsis or a secondary infection during an already existing primary illness (or primary infection), can be life- threatening. Targeted administration requires the pathogen or pathogens to be identified as fast as possible and as correctly as possible, as well as the rapid detemiination of their resistances to the different antibiotics.

[0006] In routine microbiological work, the resistance of bacteria is nowadays determined by culturing the bacteria in vitro on a nutrient medium (e.g. an agar plate) or in a nutrient medium (e.g. a culture broth), to which an antibiotic is added. Whether the bacteria grow under the influence of the antibiotic, i.e. are resistant to the antibiotic, is determined visually, either by eye or automated with optical devices. The usual procedure is to test cultures with graduated concentrations of the antibiotic in order to determine the minimum inhibitory concentration (MIC).

[0007] The minimum inhibitory concentration designates the lowest concentration of a sub- stance at which the multiplication of a microorganism can no longer be perceived visually. Usually, that concentration of an antibiotic is detennined which just still inhibits the growth of a colony. Antibiotics can also be characterized via the minimum bactericidal concentration (MBC) at which just about 99.9% of the pathogens are killed within a given period of time. While the MIC can be determined in principle for every antibiotic, the MBC only makes sense for those which can develop not only an inhibitory, but also a destructive (bactericidal) effect. These are, for example, aminoglycosides, gyrase inhibitors, and penicillins. According to the strength of the resistance determined, the pathogens detected are termed S (sensitive), I (intermediate) or R (resistant).

[0008] The routine method based on culturing allows a simple but time-consuming determi- nation of the resistance, because the duration of the analysis is dictated by the growth rate of the bacteria. An additional advantage of the routine method consists in the fact that the effi- ciency or inefficiency of an antibiotic against the bacteria of a sample is measured directly (functional test).

[0009] In addition to culturing in the presence of antibiotics, there are also genetic methods to determine resistances, by detecting known resistance genes in the genome of the pathogen in question. An advantage of the genetic methods consists in the fact that the resistance genes can be multiplied by techniques such as polymerase chain reaction (PCR), and thus the time needed for the analysis is no longer determined by the growth rate of the bacteria. The disadvantages are that they are more expensive than routine methods and do not represent functional tests. A resistance gene may be present, but is not expressed, which makes the bacterial strain under investigation not resistant, but the method detects it as being resistant.

[0010] Many types of microorganisms, particularly bacteria and unicellular fungi, can be identified mass spectrometrically nowadays - quickly and with low error rates. The term "identification" here is deemed to mean the taxonomic classification, i.e. the determination of family, genus and species. The identification of bacteria by mass spectrometric methods is de- scribed in detail in a scientific review article by van Baar, for example (FEMS Microbiology Reviews, 24, 2000, 193-219: "Characterization of bacteria by matrix-assisted laser desorp- tion/ionization and electrospray mass spectrometry"). In routine laboratory work, the identification is achieved by similarity analyses between a MALDI mass spectrum (MALDI = Matrix Assisted Laser Desorption/Ionization) of the sample under investigation and MALDI reference spectra of known microorganisms. The similarity analysis involves assigning each reference spectrum an indicator, which is a measure of the agreement between the corresponding reference spectrum and the mass spectrum of the sample (see for example Jarman et al., Analytical Chemistry, 72[6], 1217-1223, 2000: "An algorithm for automated bacterial identification using matrix-assisted laser desorption/ionization mass spectrometry"). This method of identifying microorganisms has proven to be extraordinarily successful both in studies as well as in the daily routine in many microbiological laboratories. It is fast, low-cost and has very low error rates, far lower than conventional microbiological identification methods.

[0011] Attempts have been made to extend the mass spectrometric identification of microorganisms to a mass spectrometric determination of their resistances. However, it has been found that, until now, resistances can only be determined directly from a mass spectrum in exceptional cases, even though the resistances should really consist in the expression of new or modified proteins. The patent specification DE 10 2006 021 493 B4 (Govorun et al.) elucidates a mass spectrometric method to determine the resistance of bacteria, wherein protein profiles of the bacteria are mass spectrometrically measured before and after culturing, and where the cultur- ing is done with the addition of an antibiotic. The method of Govorun et al. is a functional test in which the duration of the analysis is dictated by the growth rate of the bacteria, just like the routine method described above.

[0012] One type of bacterial resistance to beta-lactam antibiotics consists in the formation of enzymes (beta-lactamases) which enzymatically modify the beta-lactam antibiotics, often by a hydrolytic cleaving of the β-lactam ring, and thus render them ineffective. The catalytic effect means that a small quantity of the beta-lactamases is sufficient to render large quantities of beta-lactam antibiotics ineffective. The genetic information for the synthesis of the enzyme, which originated from corresponding mutations, is passed on chromosomally or plasmid- encoded. The plasmidic information can be exchanged between bacteria by various mechanisms, even between bacteria of different species and genera ("horizontal exchange"). More than 340 variants of beta-lactamases are currently known, and they are formed by a large number of bacterial species. They are grouped into four classes (A to D) according to their molecular structure, and into groups and subgroups depending on how they act. Today there are a large number of different beta-lactam antibiotics, including several penicillins (benzylpenicillins, oral penicillins, aminopenicillins, isoxazolyl penicillins, acylaminopenicillins), cephalosporins, monobactams and carbapenems. These usually have larger chemical groups in order to sterical- ly inhibit the beta-lactamases.

[001 ] A subgroup of the beta-lactamases (and the bacteria that produce them) comprises the so-called extended spectrum beta-lactamases (ESBL or ESBL bacterial strains), which have originated from beta-lactamase encoding genes in the plasmids by point mutations and can cleave a broad spectrum of beta-lactam antibiotics. ESBL bacterial strains are resistant to penicillins, cephalosporins, (generations 1 -4) and monobactams. It is mainly E. coli and leb- siellae (Gram-negative bacteria) which carry ESBL encoding genes. Microbiologists are watching the rapid spread of this ESBL resistance with great concern. It is one of the most worrying concerns in infection research apart from the methicillin resistance of Staphylococcus aureus (MRSA).

[0014] As an adjuvant for combating beta-lactamases, beta-lactamase inhibitors are administered together with a beta-lactam antibiotic as a combination drug in order to inhibit or at least weaken the effect of the beta-lactamases. Beta-lactamase inhibitors currently have a beta- lactam ring, but usually have no appreciable antibacterial efficiency. There are licensed combination drugs with clavulanic acid (Figure 1A), sulbactam (Figure IB) and tazobactam (Figure 1C) as beta-lactamase inhibitors, which are used in the following fixed combinations: amoxicillin with clavulanic acid, ampicillin with sulbactam and piperacillin with tazobactam. Not all combinations have the optimum effect. These adjuvants should only be used after careful identification of the bacteria and determination of their resistance, because it is to be expected that resistances to combination drugs will also be formed very quickly.

Objective of the Invention

[0015] The objective of the invention is to provide a method with which the inhibitory effect of substances on beta-lactamases is determined quickly and with certainty, particularly the inhibitoiy effect of the inhibitor in a combination drug, and thus to provide a method for determining the beta-lactamase-based resistance of a bacterial sample to such a combination drug.

Summary of the Invention

[0016] The methods according to the invention are described in the Claims 1, 9 and 19. Preferred example embodiments are specified in the dependent Claims 2 to 8 and 10 to 18.

[0017] The invention relates to a method for determining the inhibitory effect of a test substance on beta-lactamases, wherein the test substance is brought together with a beta-lactamase and a substrate which has a beta-lactam ring, and afterwards a mass spectrum is acquired and evaluated as to whether the substrate has been enzymatically modified by the beta-lactamase. [0018] The substance under investigation (test substance) and the beta-lactamase (or beta- lactamase producing microbes) are brought together with the substrate, preferably in a liquid reaction mixture. After a certain reaction time, the mass spectrum is acquired from the liquid or a portion of it. If the enzymatic activity of the beta-lactamases is not inhibited by the test substance, the concentration of a substrate decreases and that of reaction products increases, so the mass signal of the substrate in the mass spectrum is small or completely absent, and one or more mass signals from reaction products of the substrate are present in the mass spectrum. The decrease in the substrate as a result of an enzymatic modification is preferably derived from the intensity ratio of the mass signal of the substrate to the mass signal of a reaction product or to the mass signal of a reference substance. However, the enzymatic activity of the beta-lactamase can also be derived from the mass signal of the substrate alone. [0019] In a preferred method, an additional mass spectrum is acquired from the mixture at the start time, in order to derive enzymatic modifications from the changed signal intensities of the corresponding mass signals in the two mass spectra. The beta-lactamases used in the method can be artificially synthesized or be produced by bacterial strains which exhibit a beta- lactamase resistance. Beta-lactamases usually cause a hydrolytic cleavage of the beta-lactam ring of the antibiotic, so a reaction product is heavier than the antibiotic by 18 atomic mass units.

[0020] The determination of the inhibitory effect of a test substance can also include a quantitative determination of the inhibitory effect, i.e. the strength of the inhibitory effect. The reaction speed with which a more or less inhibited beta-lactamase reacts with a substrate is a measure of the strength of the inhibitory effect of a test substance. If the concentrations of the beta-lactamases (or the quantity of biological material), the substrate and the test substance are known at the start of the method, or fixed by standardized operating instructions, the reaction speed can be determined from the concentrations of the substrate and its reaction products measured after a certain time (single-point measurement), where the concentrations are represented by the signal intensities of the relevant mass signals (or their ratios with respect to each other). However, the reaction speed can also be determined each with from the concentrations of the substrate or from the concentrations of reaction products. But it is also possible to acquire an additional mass spectrum (two-point measurement) at the start of the reaction process, or several additional mass spectra during the course of the reaction (multi-point measurement) in order to determine the reaction speed from the temporal changes of the mass signals in the mass spectra. As is usual in other mass spectrometric methods, at the end of the reaction process, the concentration of the substrate or the concentrations of reaction products can be derived from mass signals of calibration substances, which are added to the reaction mixture (preferably at the end of the reaction process) and cannot be enzymatically modified.

[0021] With the method according to the invention, the inhibitory effect of a test substance on several beta-lactamases, which can belong to different classes of beta-lactamase, can be determined simultaneously. Since the reaction products from different substrates can usually be distinguished clearly in a mass spectrum, it is also possible to use several substrates which are each specific for a certain beta-lactamase or for a certain class of beta-lactamases so that the inhibitory efficiency of the test substance on different beta-lactamases can be distinguished. [0022] The substrates used in the method can be effective beta-lactam antibiotics or derivatives of beta-lactam antibiotics. To increase the sensitivity of the method it is advantageous to use high substrate concentrations. However, this kills the bacteria, which are the preferred source of beta-lactamases, too quickly, even those with a beta-lactamase resistance. For this reason, substrates can be used which have a lower antibiotic efficiency than conventional beta- lactam antibiotics. The substrates then replicate the steric forms of different antibiotics in a preferred way.

[0023] The test substance is preferably added to a nutrient solution together with several substrates and resistant bacteria. The bacterial strains selected are particularly ones which have a marked beta-lactamase resistance to current beta-lactam antibiotics and combination drugs. After an incubation period, a mass spectrum is acquired from the supernatant of the nutrient solution or a portion of it.

[ 0024] In principle, the mass spectrum can be acquired with any type of mass spectrometer in the method according to the invention. It is, however, particularly advantageous to use a MALDI time-of-flight mass spectrometer with axial ion injection which is already used in routine operation for the mass spectrometric identification of microorganisms. Since the MALDI process generates a strong chemical background in the lower mass range of a few hundred atomic mass units, it is favorable to use substrates with masses in the mass ranges with low background. This is achieved, for example, by using substrates with molecular weights greater than 700 atomic mass units, preferably between 700 and 1200 atomic mass units. It is also advantageous to increase the proton affinity of the substrates in order to be easier ionized. A chemical derivatization of the substrate with regard to the molecular weight, the proton affinity and/or a covalently bound charge tag is preferably carried out at the end of the reaction process in order to leave the chemical properties of the substrate as similar as possible to those of an antibiotic during the reaction, and can also be applied to reaction products of the substrate. Furthermore, the substrates used can have an anchor group with which they can be extracted from the medium in which the reaction was prepared, and enriched. Examples of anchor groups are biotin and a 6-His tag (a chain of six histidine molecules), which are extracted with the aid of streptavidin or a chelate loaded with nickel ions respectively; the bonds here are reversible and the extracting binding partner can be bound to the walls of the reaction vessel or to magnetic beads. 100251 An application of the method according of the invention consists in screening a test substance library during a pharmacological development of drugs in order to find lead structures for beta-lactamase inhibitors which can be used in combination drugs together with an antibiotic. [0026] A further application consists in taking several test substances which have different inhibitory effects on different classes of beta-lactamases, and determining their inhibitory efficiency against a bacterial strain with a beta-lactamase resistance. The beta-lactamases of the bacterial strain are then characterized by the inhibitory efficiency or inefficiency of the substances, particularly with regard to the encoding site of the resistance gene. In the case of bacteria, the resistance gene can be encoded in a plasmid or in the bacterial chromosome. Since a resistance gene on a plasmid can be exchanged between bacterial strains of different species, or even of different genera, more easily than one on the bacterial chromosome— a fact which increases the speed with which the resistance gene spreads and thus the potential risk posed by the bacterial strain— the knowledge of the encoding site is of great importance for the epide- miological assessment of the bacterial sample. It is known, for example, that beta-lactamases of molecular group A (enzymes with a serine group) are mainly plasmid encoded, but are hardly inhibited, if at all, by clavulanic acid, while beta-lactamases of molecular group B (metalloen- zymes) are chromosomally encoded and are inhibited by clavulanic acid.

1 0271 The invention relates also to a method for determining whether a bacterial sample is beta-lactamase resistant to a combination drug, determining particularly the inhibitory efficiency of the inhibitor of a combination drug against beta-lactamases. The method according to the invention comprises: bringing together the bacterial sample and the combination drug, acquiring a mass spectrum, and determining whether a beta-lactam antibiotic of the combination drug has been enzymatically modified. A combination drug consists at least of one inhibitor and one beta-lactam antibiotic. The number of antibiotics or inhibitors in a combination drug can each be greater than one, however.

[0028] The term "bacterial sample" is to be understood in a general sense. It can be a sample which is known or suspected to comprise bacteria, or which is to be analyzed as to whether it comprises bacteria. A bacterial sample can, for example, be a portion of an already isolated bacterial culture, which is harvested from an agar plate, but can also be a bacteria pellet, which is obtained directly from a body fluid such as blood, urine or CSF (cerebrospinal fluid) or from a nutrient solution such as a blood culture or culture broth. Smears from mucous membranes, particularly a smear from the nasal or pharyngeal mucosa, and smears from wounds or stool samples, are also deemed to be bacterial samples, as are swab samples in hospitals or in food- processing companies, and food samples themselves. A bacterial sample can comprise bacteria of one or more strains. 10029) In a method according to the invention, the bacterial sample is incubated together with the combination drug in a nutrient solution, and afterwards the supernatant of the nutrient, solution or a portion of it is subjected to a mass spectrometric measurement. But it is also possible to apply the bacterial sample to a flat nutrient medium (e.g. an agar plate) which contains the combination drug. The mass spectrum is then acquired from the culture medium close to the bacterial sample.

[0030] It is especially preferred that, in an additional reaction mixture, the bacterial sample is brought together only with the beta-lactam antibiotic of the combination drug. The additional mass spectrum provides information on whether the bacteria of the sample produce beta- lactamases which enzymatically modify the beta-lactam antibiotic of the combination drug. The method according to the invention can furthermore be modified so that the bacterial sample is not brought together with the combination drug, but with the inhibitor of the combination drug and a substrate, e.g. a derivative of the beta-lactam antibiotic.

[0031] As has already been described above, an enzymatic modification of the beta-lactam antibiotic can be inferred from the absence or decrease of the mass signal of the beta-lactam antibiotic, or from the presence of one or more mass signals of reaction products, or from the intensity ratio of the mass signal of the beta-lactam antibiotic to the mass signal of a reaction product, or to the mass signal of a reference substance. The speed with which a beta-lactam antibiotic of a combination drug reacts is a quantitative measure for the strength of the beta- lactamase resistance of the bacteria against the combination drug, and can be determined by a single-point, two-point or multi-point measurement. As is usually the case with a functional test, the determination of the resistance can be carried out for different concentrations of the combination drug in order to determine the minimum inhibitory concentration.

[0032] Particularly when the mass spectra are acquired with a MALDI time-of-flight mass spectrometer, it is favorable that the reaction is followed by an additional chemical derealization which modifies the beta-lactam antibiotic and any reaction products which may be present in such a way that afterwards their molecular weight is greater than 700 atomic mass units, their proton affinity has increased and/or they have at least one covalently bound charge.

[0033] The method according to the invention can test rapidly and reliably whether a bacterial sample has a beta-lactamase resistance and whether a combination drug is effective against existing beta-lactamases. The test takes less than three hours and is thus very much faster than the routine method where the growth of the bacteria or the inhibition of growth is detected by optical means. In contrast to genetic methods, the method according to the invention can be used to functionally determine the efficiency of combination drugs against a bacterial sample with a beta-lactamase resistance.

[0034] In principle, the mass spectrum in the methods according to the invention can be acquired with any type of mass spectrometer, for example with a time-of-fiight mass spectrometer with orthogonal ion injection (OTOF), an ion cyclotron resonance mass spectrometer (ICR-MS), an electrostatic ion trap mass spectrometer or an RF ion trap mass spectrometer. A triple quadrupole filter mass spectrometer is particular suitable for a mass spectrometric determination of an enzymatic modification of the substrate because very high sensitivity can be achieved with it, and therefore very tiny initial quantities are sufficient to carry out the methods according to the invention. It is, however, desirable to use the same MALDI time-of-flight mass spectrometer with axial ion injection which is already used in routine operation for the taxonomic identification of the bacteria. Instead of an ionization with matrix-assisted laser desorption (MALDI ), other types of ionization, for example electrospray ionization (ESI) or ionization by matrix-free laser desorption (nanostructure initiator mass spectrometry (NIMS)), can be used for the method according to the invention. Those skilled in the art are familiar with these mass spectrometers and ionization methods, so we will forego a detailed explanation here.

Brief Description of the Illustrations

[0035] Figures 1A to 1 C show the molecular structural formula of three beta-lactamase inhibitors known from the prior art: clavulanic acid (Fig. 1A), sulbactam (Fig. IB) and tazobac- tam (Fig. 1 C).

[0036] Figure 2 is a schematic representation of the procedure for determining the inhibitory efficiency of test substances in order to identify lead structures in the search for pharmacologically active substances (high-throughput screening). [0037] Figure 3 is a schematic representation of the procedure for determining the susceptibility of a combination drug against a bacterial sample.

[0038] Figures 4A to 4D show four measured MALDI mass spectra of the ampicillin antibiotic in a mass range between 340 and 450 atomic mass units. Examples of Preferred Embodiments

[0039] Figures 1A to 1C show the molecular structural formula of three beta-lactamase inhibitors known from the prior art: clavulanic acid (Fig. 1 A), sulbactam (Fig. IB) and tazobac- tam (Fig. 1C). These inhibitors all have a beta-lactam ring, i.e. a cyclic amide with a heterocyclic structure of three carbon atoms and one nitrogen atom. [0040] Figure 2 is a schematic representation of the procedure for determining the inhibitory efficiency of test substances in order to identify lead structure elements in the search for pharmacologically active substances. Pharmaceutical companies keep large substance libraries which can easily contain a few hundred thousand different substances. With the aid of modern chemical methods, such as combinatorial synthesis or the highly parallel preparation of natural products, thousands of derivatives of existing original substances can be produced in a short time, or new original substances can be found. The method according to the invention makes it possible to test a large number of test substances in a short time for their inhibitory efficiency against beta-lactam ases (high-throughput screening).

[0041] In Step A, test substances are provided in a substance plate 100. The substance plate 100 is a microtiter plate with 24 spatially separate wells 101 to 124, which are arranged in rows and columns. The precise dimensions (length x width x height) of a microtiter plate according to the ANSI standard, on the recommendation of the Society of Biomolecular Screening, are 127.76 mm x 85.48 mm x 14.35 mm. There are a multitude of formats, e.g. with 24 wells (4x6, capacity: 0.5-3 ml per well), 96 wells (8 12, capacity: 0.3-2 ml per well) or 384 wells (16x24, capacity: 0.03-0.1 ml per well). Each well on the substance plate 100 contains only one single test substance in this embodiment. A pipetting robot 10 transfers the test substances from the substance plate 100 onto a reaction plate 200 in such a way that each well on the reaction plate

200 contains only one test substance. The reaction plate 200 is a microtiter plate with 24 wells

201 to 224. The pipettes of the pipetting robot 10 are replaced after each transfer in order to prevent contamination of the test substances in the wells 201 to 224. [0042] In Step B, a substrate mixture with three different substrates SI to S3, which are all beta-lactam antibiotics, is put into each well of the reaction plate 200 by means of a dispenser 20. The dispensing is a non-contact operation in order to prevent contamination between the reaction mixtures. [0043] In Step C, a few microliters of the supernatant from each well of the reaction plate 200 is transferred onto a first MALDI sample support and prepared there. The sample support 300 has the same format as the reaction plate 200 with respect to number and positions of the sample sites, but can also have more than the 24 sample sites 301 to 324 so that samples from other reaction plates can also be prepared on the MALDI sample support 300. After the sam- pies transferred from the reaction plate 200 have dried on the sample support 300, they are coated with about one microliter of matrix solution. The matrix used is a-cyano-4- hydroxycinnamic acid (HCCA) at a concentration of 10 milligrams per milliliter in a mixture of water, 50% acetonitrile and 2.5% trifmoroacetic acid. A pipetting robot (not shown) transfers the samples and the matrix solution onto the sample sites 301 to 324 of the MALDI sample support 300. After preparation, the sample support 300 is introduced into a MALDI time-of- flight mass spectrometer and mass spectra MSal to MSa24, of which only the mass spectra MSal and MSa24 are shown in Figure 3 for reasons of clarity, are acquired at each of the MALDI sample sites 301 to 324. Each of the MALDI mass spectra MSal to MSa24 comprises mass signals of the substrates S I to S3 and of the relevant test substance, for example the test substance Tl in the mass spectrum MSal . The mass spectra MSal to MSa24 acquired in Step C represent the signal strengths of the substrates SI to S3 at the start of the subsequent inhibition reactions and thus allow the determination of the inhibitory strength of the test substances under investigation Tl to T24 in a two-point measurement.

[ 00441 In Step D. a bacterial mixture is put into each well of the reaction plate 200 by means of a dispenser 30. The dispensing is a non-contact operation, as in Step C, in order to prevent contamination between the reaction mixtures. The bacterial mixture contains different bacterial strains for which a beta-lactamase resistance to the antibiotics used has already been proven. The reaction plate 300 is subsequently incubated for three hours at 37° Celsius under constant shaking and then centrifuged in order to precipitate the bacterial cells at the bottom of the wells.

[0045] In Step E, a few microliters of the supernatant are transferred from each well of the reaction plate 200 onto a second MALDI sample support and prepared there as described in Step C, The sample support 400 has the same format as the sample support 300. After the preparation, the sample support 400 is introduced into the MALDI time-of-flight mass spectrometer and one mass spectrum is acquired for each of the 24 MALDI sample sites 401 to 424 (MSbl to MSb24). The inhibitory efficiency of the test substances can be determined by com- paring the corresponding mass signals in the mass spectra before and after the inhibition reaction. The mass signals of the substrates SI to S3, for example, in the mass spectrum MSbl are unchanged in comparison to the mass spectrum MSal , i.e. no enzymatic modification of the substrates SI to S3 has taken place. The test substance Tl inhibits all the beta-lactamases produced by the bacterial mixture. In the mass spectrum MSb24, in contrast, the mass signals of the substrates SI to S3 are missing, and mass signals of corresponding reaction products Rl to R3 are present. Thus the test substance T24 does not inhibit the beta-lactamases of the bacterial mixture.

[0046] In order to search through large substance libraries, the method can be carried out in such a way that the inhibitory effect is at first determined in a large number of multiplex mix- tures, each with several test substances per well, and that subsequently those test substances which exhibited an inhibitory effect in one of the multiplex mixtures are tested individually.

[0047] Figure 3 is a schematic representation of a procedure for determining the susceptibility of a combination drug against a bacterial sample.

[0048] In Step A, small quantities of an isolated bacterial colony 3, which was cultured in the usual way on an agar plate, are each transferred into two Eppendorf tubes 1 and 2, which are then filled with a liquid nutrient medium. The Eppendorf tubes 1 and 2 are then vortexed for a short time to distribute the bacteria in the nutrient medium. To standardize the test, the number of bacterial cells per unit volume is adjusted to a specific value by the addition of further quantities of microbes, or by an incubated growth phase, or by dilution of the nutrient medium, until the optical density of the nutrient medium reaches a certain value.

[0049] In Step B, the combination drug under investigation 5 is added to the reaction mixture in the Eppendorf tube 1. Only the antibiotic 6 of the combination drug is added to the reaction mixture in the Eppendorf tube 2.

[0050] In Steps C and D, the two Eppendorf tubes 1 and 2 are incubated and shaked at 37°C for three hours and subsequently centrifuged for 2 minutes at 13,000 rpm in order to obtain a cell-free supernatant (7A, 8 A) and a bacterial pellet ( 7B. 8B) at the bottom of both Eppendorf tubes 1 and 2.

[0051] In Step E, one microliter from the supernatant 7A and one microliter from the supernatant 8A are transferred to two separate sample sites of a MALDI sample support. After the samples transferred from the supernatant have dried, they are coated with one microliter of a matrix solution. The matrix solution used is a-cyano-4-hydroxycinnamic acid (HCCA) at a concentration of 10 milligrams per milliliter in a mixture of water, 50% acetonitrile and 2.5% trifluoroacetic acid. After the matrix has crystallized out by vaporization of the solvent, the sample support is introduced into a MALDI time-of-flight mass spectrometer and MALDI mass spectra MSI and MS2 are acquired in the mass range from 100 to 1000 atomic mass units. Other matrices, solvents and types of MALDI preparation are possible.

[0052] The mass spectrum MS2 exhibits a mass signal of the antibiotic A and a reaction product R, i.e. that the antibiotic A has been enzymatically modified and the bacterial colony 3 comprises a beta-lactamase resistance. The mass spectrum MSI exhibits only the mass signal of the antibiotic A, but no mass signals of reaction products, i.e. the inhibitor of the combination drug inhibits the beta-lactamases formed by the bacterial colony, and the combination drug can be used for a therapy against this bacterial strain.

[0053] Figures 4A to 4D show four measured MALDI mass spectra with the ampicillin antibiotic in a mass range between 340 and 450 atomic mass units. Ampicillin has a molar mass of 349.41 g/mol and is a semi-synthetic, antibiotic active substance from the group of the beta-lactam antibiotics, which has been used since 1961. It is known as a broad-spectrum antibiotic due to its efficiency against Gram-positive bacteria and some Gram-negative bacterial species. Ampicillin belongs chemically to the aminopenicillins. The sample preparation (inoculation of a liquid nutrient medium, incubation and centrifugation) and the preparation of the MALDI samples correspond to the method in Figure 3.

[0054] The mass spectrum in Figure 4A shows an ampicillin spectrum after incubation with the DH5a bacteria strain of the species Escherichia coli. The mass spectrum exhibits the mass signals of the protonated ampicillin ([M+H]⁺, 350u) and of a sodium adduct ion of the ampicillin ([M+Na]⁺, 372u), but no mass signals of reaction products which could be attributed to an enzymatic modification, such as a hydrolytic cleaving of the beta-lactam ring. The bacterial strain DH5a does not exhibit any beta-lactamase resistance. [0055] The mass spectrum in Figure 4B depicts an ampicillin spectrum after incubation with an ESBL bacterial strain of the species Escherichia coli. The two mass signals of the ampicillin at 350u and 372u are absent, and hydrolyzed reaction products are present in the mass spectrum at 368u and 390u (18u heavier). The ESBL bacterial strain produces beta-lactamases which enzymatically modify the ampicillin and it is therefore resistant to ampicillin.

[0056] The mass spectra in Figures 4C and 4D shown ampicillin spectra after incubation with the ESBL bacterial strain from Figure 4B with the addition of clavulanic acid and tazobactam, respectively. In contrast to the mass spectrum of Figure 4B, the mass signals of the non- modified ampicillin at 350u and 372u are present in both cases, while the hydrolyzed reaction products at 368u and 390u are absent. Clavulanic acid and tazobactam both inhibit the beta- lactamases of the ESBL bacterial strain. In combination with clavulanic acid or tazobactam. ampicillin is effective against the ESBL bacterial strain.

[0057] To determine the resistance of combination drugs, some or all of the consumables necessary for this, such as MALDI sample supports, NIMS sample supports, Eppendorf tubes, a MALDI matrix, solvents, antibiotics, derivatives of antibiotics, substrates, inhibitors or combination drugs, can be supplied in sterile kits.

Claims

1. Method to determine the inhibitory effect of a test substance on beta-lactamases, comprising the steps:

(a) mixing the test substance with at least one beta-lactamase and at least one substrate which has a beta-lactam ring,

(b) acquiring a mass spectrum, and

(c) evaluating the mass spectrum to determine whether a substrate has been enzymatically modified in step (a).

2. Method according to Claim 1 , wherein the substrates exhibit little antibiotic effect com- pared to beta-lactam antibiotics.

3. Method according to Claim 1 or 2, wherein different types of substrate are used which replicate the steric forms of different antibiotics around the beta-lactam ring.

4. Method according to Claim 1, wherein at least one substrate is a beta-lactam antibiotic.

5. Method according to one of the Claims 1 to 4, wherein an enzymatic modification is

detected

by the presence of one or more mass signals of reaction products of one of the substrates or

by the intensity ratio of the mass signal of a substrate to one or more mass signals from its reaction products or to the mass signal of a reference substance.

6. Method according to one of the Claims 1 to 5, wherein an evaluation takes place in step (c) as to whether a hydrolytic cleaving of at least one substrate has taken place.

7. Method according to one of the Claims 1 to 6, wherein in step (a) the test substance, the at least one substrate and the bacteria are put into a nutrient solution, the bacteria exhibiting a beta-lactamase resistance.

8. Method according to one of the Claims 1 to 7, wherein for an enzymatic modification detected in Step (c) the reaction speed is determined and the strength of the inhibition is derived from the reaction speed.

9. Method for determining a beta-lactamase resistance of a bacterial sample to a combination drug which is comprised of at least one beta-lactam antibiotic and one inhibitor, with the steps: (a) mixing the bacterial sample and the combination drag,

(b) acquiring a mass spectrum, and

(c) evaluating the mass spectrum to determine whether a beta-lactam antibiotic of the combination drug has been enzymatically modified in step (a) .

10. Method according to Claim 9, wherein the bacterial sample is incubated together with the combination drug in a nutrient solution and the mass spectrum is acquired from the supernatant of the nutrient solution or a portion of it.

11. Method according to Claim 9 or 10, wherein an enzymatic modification is detected by the presence of one or more mass signals of reaction products of a beta-lactam antibiotic of the combination drug.

12. Method according to Claim 9 or 10, wherein an enzymatic modification is detected by the intensity ratio of the mass signal of a beta-lactam antibiotic of the combination drug to one or more mass signals of its reaction products or to the mass signal of a reference substance.

13. Method according to one of the Claims 9 to 12, wherein for an enzymatic modification detected in Step (c) the reaction speed is determined and the strength of the resistance is derived from the reaction speed.

14. Method according to Claim 13, wherein the reaction speed is determined from the individual mass spectrum and the time of its acquisition or from further mass spectra acquired at different times.

15. Method according to one of the Claims 9 to 13, wherein the Steps (a) to (c) are carried out for different concentrations of the combination drug.

16. Method according to one of the Claims 9 to 15, wherein between Steps (a) and (b) a chemical derivatization step is inserted in which the beta-lactam antibiotics of the combination drug and/or the enzymatically modified reaction products which may be present are modified in such a way that their masses are in each case larger than 700 atomic mass units and/or have at least one covalently bound charge.

17. Method according to one of the Claims 9 to 16, wherein in Step (b) the ionization is performed by matrix-assisted laser desorption.

18. Method according to Claim 9, wherein additionally a portion of the bacterial sample is mixed with only the antibiotic of the combination drug, then a mass spectrum is acquired and evaluated with respect to whether the antibiotic of the combination drug has been en- zymatically modified.

19. Method to determine the inhibitory effect of the inhibitors of a combination drug on a bacteria] sample which is resistant due to the production of beta-lactamases, with the steps:

(a) mixing the resistant bacterial sample with the inhibitors of the combination drug and derivatives of the beta-lactam antibiotics of the combination drug,

(b) acquiring a mass spectrum, and

(c) evaluating the mass spectrum to determine whether a derivative of a beta-lactam antibiotic of the combination drug has been enzymatically modified in Step (a).