The present invention relates to direct detection of resistance mechanisms by MALDI-TOF Mass Spectrometry to guide appropriate treatment and prevent potential dissemination events. Background of the Invention

Resistance to third generation cephalosporins (TGC) is one of the most relevant problems in current gram-negative antimicrobial therapy, leading to overuse of carbapenemes and emergence of different carbapenemases. Outstanding from others, the most relevant resistance mechanisms to TGC are different plasmid borne extended spectrum β-lactamases (ESBL) families and plasmid-determined AmpC-type β-lactamases. Among them, CTX-M-type and CMY-2 β-lactamases are the most frequent mechanisms worldwide, respectively (1 ).

Among carbapenemases, KPC is one of the most significant enzymes, and its presence in epidemic strains makes it one of the most relevant resistance markers.

Phenotypical methods as double disk synergytests using β-lactam and β-lactamase-inhibitor disks, and even the original double disk approximation tests (with the incorporation of different inhibitors) remain convenient, easy to perform, but not fast methods for detecting these β- lactamase in gram-negative bacilli.

Clinical microbiology laboratories are today expected to provide a proper characterization of involved mechanisms for an effective antimicrobial therapy (2). In this case, fast typing methods for TGC and carbapenemases resistance mechanisms are needed to guide appropriate treatment and prevent potential dissemination events. This is also the case for any resistance mechanism, and different genotypic systems have been developed for achieving this goal.

Incorporation of matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) has been one of the most significant breakthrough technology included in modern clinical microbiology, and is today routinely used in high complexity laboratories for rapid species identification (3, 4). Based on MALDI-TOF MS availability, evaluation of β-lactamases production (especially carbapenemases) is also possible, provided that different antibiotic standards are available. In this case, antibiotics are incubated with the bacteria and the hydrolysis is followed for at least one hour (5, 6). Following this methodology, enzyme detection is feasible but only reflects the presence of at least one enzyme with hydrolytic activity on the provided substrate. Direct detection of a specific enzyme by analysis of a protein profile obtained by MALDI-TOF MS has been the subject of research in different studies. Among them, Camaro and Hays were the first to detect a peak corresponding to the β-lactamase TEM-1 , by analyzing the spectrum of proteins starting from bacterial lysates of E. coli susceptible and resistant to ampicillin (7). Papagiannitsis et al, detected the CMY-2 enzyme in different enterobacteria after extraction of periplasmic proteins using a multiple steps procedure (8, 9). Detection of KPC has also been inferred after detection of an accompanying protein in KPC-producing microorganisms.

There are known some patents and patent applications in the field.

For example, WO2018074762-A1 discloses a method for rapidly discriminating Gram-positive and negative bacteria and Candida species and determining whether the bacteria are resistant like MRSA, VRE, and bacteria having genes corresponding to ESBLs (TEM-, CTX-M-, SHV- type), AmpC (ACT, CMY2, DHA, CMY-1 like/MOX, ACC-1 , FOX), and carbapenemases (OXA- 48 like, IMP, VIM, NDM, KPC, SPM).

US9874570-B2 discloses a method of detecting a TEM protein in a sample. The method comprises subjecting the sample to MS/MS spectrometry in MRM mode and detecting whether one or more TEM fragments selected from the group consisting of SEQ ID NOS: 166-230, 232- 257, 259-261 , 1923, 1927, and 1928 is present, wherein detection of any of the TEM fragments by the MRM mass spectrometry indicates the presence of TEM protein in the sample.

WO2018005370-A1 relates to a method for determining presence of none, one or more ambler class carbapenemases expressed by enteric bacteria, that involves (a) providing sample comprising enteric bacteria, (b) applying the enteric bacteria in the test sample to at least four test compositions for a duration of time, where each of several test compositions comprises growth medium and an antibiotic, and at least one test composition further comprises at least one carbapenemase inhibitor, and (c) determining the presence of none, one or more ambler class carbapenemases expressed by the enteric bacteria by detecting a presence or an inhibition of growth of the enteric bacteria in each test composition after specific time.

US9506932-B2 discloses a method of detecting a CTX-M protein in a sample from a microorganism. In this case, the sample is subjected to MRM mass spectrometry and is detected whether one or more CTX-M fragments selected from the group consisting of in SEQ ID NOS: 446-478, 480-495, 2009-2013, 2015, 2017, 2019, 2021 -2027, 2029, 2030, 2032, 2034-2039, 2042-2051 , 2054, 2055, 2057-2063, 2065-2067, 2069-2074, 2076-2078, and 2081 -2092 is present, wherein detection of any of the CTX-M fragments by the MRM mass spectrometry indicates the presence of CTX-M protein in the sample.

Other resistance mechanisms detection methods are also known by EP2972351 -A1 , US9834823-B2, WO2017187178-A1 and WO2016016580. Summary of the Invention

Present invention provides a short and fast methodology for the identification of resistance mechanisms, such as CMY-2, KPC-2 or ΟΤΧ-Μβ-lactamases (among others) in clinical cultures of £ coli, by MALDI-TOF MS. The method is based on total intact protein extraction analysis using an organic solvent, and detection of the mature protein according to its theoretical molecular weight.

Embodiments of the invention provide a method for direct detection of resistance mechanisms by MALDI-TOF Mass Spectrometry, comprising a) preparing different bacterial cultures, such as £ constrains, among others, using a double layer sinapinic acid technique for detection of intact protein, wherein each bacterial culture is contained in a MALDI target spot; b) acquiring a spectrum of each culture by setting the spectrum range of a mass spectrometer using a MALDI- TOF software to a range comprised between 17,000 and 50,000 Da; and c) comparing the different acquired spectra taking into account a theoretical weight of the protein to be detected. According to the proposed method, besides E. coli, the bacterial cultures may also include enterobacteriaceae including Klebsiella pneumoniae, Proteus mirabilis, Citrobacter freundii, Citrobacter braakii, Enterobacter cloacae, Enterobacter aerogenes, Serratia marscesens, Salmonella enterica and Shigella sonnei. Moreover, in some embodiments, Gram negative non- fermented glucose bacilli including Pseudomonas aeruginosa may also be prepared. Likewise, according to the proposed method, the protein to be detected may comprise CMY-2, CTX-M group 1 such as CTX-M-15, CTX-M group 2 such as CTX-M-2, CTX-M group 8 such as CTX-M-8, CTX-M group 9 such as CTX-M-9 and also CTX-M-14, TEM, KPC-2, PER-3 or APH(3"). Moreover, in other embodiments, detection of the following protein families is also evaluated: OXA, IMP, VIM, NDM, MCR-1 , AAC(6 ^' ), AAD, or other AmpC. In said step a), the different bacterial cultures may be extracted by using an organic solvent extraction method. Then, the bacterial cultures are analyzed with a protein detection process. Moreover, the bacterial cultures are fresh (no more than 24 h of incubation).

In an embodiment, the organic solvent extraction method comprises suspending a loopful of the bacterial cultures in 300μΙ distilled water; adding 900μΙ of absolute ethanol and vortexing for 30 seconds; centrifuging at 13,000 rpm for 2 minutes; discarding the supernatant, centrifuging again and removing the remaining absolute ethanol; resuspending the MALDI target spot in 100μΙ of extraction solvent including formic acid-isopropyl alcohol-water, 17:33:50 ratio by volume; vortexing for 30seconds; and centrifuging at 13,000 rpm for 2 minutes to obtain a clean supernatant extract.

In an embodiment, the double layer sinapinic acid techniques comprises firstly, forming a first layer by loading 0.7 μΙ of a saturated solution of sinapinic acid in absolute ethanol onto the MALDI target spot and drying at room temperature. Secondly, a second layer is formed by mixing the protein extracts 1 :1 with sinapinic acid 10 g/L solution 30:70 [v/v] acetonitrile: 0.1 % trifluoroacetic acid in water. Finally, 1 μΙ_ of the second layer is deposited onto the first layer and drying at room temperature.

In an embodiment, said range is comprised between 26,000 and 29,000 Da, the protein to be detected comprising CTX-M, KPC-2, TEM or APH(3"). In another embodiment, said range is comprised between 30,000 and 34,000 Da, the protein to be detected comprising PER-3. In yet another embodiment, said range is comprised between 37,000 and 40,000 Da, the protein to be detected comprising CMY-2.

In a particular embodiment, present invention provides a short and fast methodology for the identification of CMY-2 β-lactamases in clinical cultures of £ coli by MALDI-TOF MS. In particular, according to this embodiment, fifty two characterized £ coli strains were analyzed using the double layer sinapinic acid technique for detection of intact protein. Rapid and easy one-step extraction was applied in each culture. Comparison among mass spectral profile of different strains showed that two groups of cultures could be differentiated after peak analysis. A single distinctive peak with different intensities, at approximately m/z 39,800 Da was found in all CMY-2 producing strains (wild type and transconjugants) and consistently absent in the control groups (ESBL producers and susceptible strains). Statistical results showed 100 % values for sensitivity and specificity, indicating a perfect test and a high discriminative power. Therefore, present invention shows that MALDI-TOF MS has the potential to detect directly the most clinically relevant acquired AmpC β-lactamase, the CMY-2-enzyme, in £ coli with a less time- consuming process. Present invention results may constitute the basis for further developments to detect other β-lactamases, or even other resistance markers.

Brief Description of the Drawings

The previous and other advantages and features will be more fully understood from the following detailed description of embodiments, with reference to the attached figures, which must be considered in an illustrative and non-limiting manner, in which:

Fig. 1 shows peaks of MALDI-TOF mass spectra of the CMY-2 and non-CMY-2 producing £. coli. a) Peak detection of 39,805 Da in the £ coli transconjugant strain is indicated in green line; Absence of the ca. 39,805-m/z peaks in the recipient strain is indicated in red line. Peak corresponding to CMY-2 is indicated with arrows, b) Peaks of different intensities corresponding to mature CMY-2 protein are shown with blue lines. Spectra on non-CMY-2- producing cultures (negative controls) are shown with red lines, c) Induction assay with antibiotic. Comparison of spectra of three different CMY-2 positive £ coli grown in presence (black, blue and green lines) and in absence (gray, light blue and light green lines, respectively) of ampicillin (100 g/ml).

Fig. 2 shows the spectrum of recombinant strains expressing CTX-M- 9 (idem CTX-M-14), -15 and -2 proteins. The unprocessed E. coli XL1 Blue receptor strain transformed with the expression vector pK19 was used as negative control.

Fig. 3 shows the spectrum of clinical cultures for the detection of β-lactamase (a) CTX-M-2, (b) CTX-M-15 and (c) CTX-M-14. As a negative control, strains of the same species were used with another type of resistance. As a positive control for the detection and location of the peaks, strains transformed with the recombinant vector were used for each of the variants evaluated.

Fig. 4 shows the detection spectrum of KPC-2 from colony of the transforming strain and clinical cultures in enterobacteria. a) spectrum of strain E. coli XL1 Blue transformed with vector carrying carbapenemase KPC-2, negative control E. coli XL1 Blue, b) spectrum of clinical cultures for the detection of KPC-2.

Detailed Description of the Invention

The present invention relates to direct detection of resistance mechanisms by MALDI-TOF Mass Spectrometry, for instance discrimination of CMY-2-producing E. coli, CTX-M or KPC-2 producing Enterobacteriaceae, not limitative as the teaching of this invention could be applied to the detection of other resistance mechanisms.

In an embodiment, the invention obtains characteristic MALDI-TOF-MS spectra from presence and absence of CMY-2 producing strains. Thus, £ coli HBC1 a1 (transconjugant) and £ coli HB101 (recipient) were processed, and a characteristic peak at 39,805 Da (slightly differing from the expected value for the mature CMY-2 enzyme, 39,854 Da) was observed only in the transconjugant strain (Fig. 1 a) which may be used for the purposes of identification of this β- lactamase. Besides, when these strains were analyzed in the identification range (2,000 to 20,000 Da), no distinctive peak was identified.

Subsequently, all cultures (n = 52) included in the study were processed under the same conditions by MALDI-TOF MS. Comparison among mass spectral profile of different strains showed that two groups of cultures could be differentiated after peak analysis. A single distinctive peak with different intensities, at approximately m/z 39,800 Da was found in all CMY-2 producing strains (wild type and transconjugants) and consistently absent in the control groups (ESBL producers and susceptible strains) (Fig. 1 b). Analysis of replicate cultures of all cultures gave simple and consistent mass spectral profiles within this range (17,000 to 50,000 Da).

Analysis using the raw spectra and the "'Peak Statistic Calculation" generated a list of 15 main peaks, including the 39,800 Da-peak that corresponds to mature CMY-2 protein. Statistical results showed a p-value < 0.000001 (by both PAD and PWKW tests) which confirm a significant differences for the selected peak (17). Consistently, the area under the curve (AUC) of the ROC curve for this specific peak was 1 .0 (100 % values for sensitivity and specificity), indicating a perfect test and a high discriminative power.

In order to assess if peak production was depending upon ampicillin induction, about one half (17/29) CMY-2 producers were compared after growth in medium with or without ampicillin (100 mg/L). Results did not show a difference in intensity or position (m/z) of this distinctive peak (39,800 Da) in the spectra (Fig. 1 c).

Finally, a preliminary test on blood cultures was carried out in order to evaluate the proposed methodology directly in these samples. Three blood culture bottles (BACTEC® Plus Aerobic bottles) were inoculated with 2 ml of human blood mixed with a bacterial suspension (approximately 10 ⁴ CFU/ml). Each bottle was inoculated different cultures; £ co//9238-3013 and E. CO//47688 (both CMY-2 producers) and E. coli T1 (CTX-M-15 producer, as negative control). Blood cultures were incubated at 37 ° C for 18 hours. With the purpose of removing blood cells, 1 .4 ml of positive culture from each bottle was centrifuged at 200 x g for 5 min. The supernatant was then centrifuged at 14,170 x g for 1 min at room temperature (18). Further, each MALDI target spot was treated with the same protein extraction methodology described above. All samples were tested in duplicate.

In good agreement with previous results, a peak of approximately 39,800 Da was observed in the spectra of both CMY-2 producing £ coli, while in the CTX-M-15 producer was absent. Present invention demonstrates that MALDI-TOF MS has the potential to detect directly the most clinically relevant acquired AmpC β-lactamase, the CMY-2-enzyme, both on £ coli plates, and (even if preliminary) directly on positive blood cultures bottles.

MALDI-TOF MS applications (once the technology is in place) are fast and low-cost techniques. This new procedure (developed for detection of CMY-2 producing £ coli) employs a single solvent extraction (formic acid - isopropyl alcohol - water) step without any incubation period. As an estimation, the described protocol has a less than 1 -h turnaround time, which is faster than any phenotypic or genotypic procedure (considering disk diffusion method and PCR plus sequencing) currently used in the clinical laboratory for its detection. Moreover, this can be run while processing samples for microbial identification and thus several samples may be run almost simultaneously without a mandatory end point time. Although present invention results are preliminary, MALDI-TOF spectra contain information that allows distinguishing CMY-2 producers on a rapid and reliable way. The time required is substantially lower than that needed for hydrolysis products detection or after extraction of periplasmic proteins as well as less labour-consuming (8, 19). It does not require additional reagents to those available and used routinely for microorganism identification, except perhaps the use of sinapinic acid.

The results may constitute the basis for further research to detect other protein under similar conditions. Overall, if the method can be extended to other β-lactamases, or even other resistance markers, may prove a significant improvement in laboratory abilities for their detection with extensive clinical implications, and the inventors strongly believe that appropriate validations will definitely lead to establishing a MALDI-TOF supplementary database for future applications in diagnostic laboratories and reference centers.

Bacterial strains and β-lactamases characterization

Fifty two characterized E. coli strains deposited at the laboratory collection in Buenos Aires, Argentina were used (Table 1 ).

Table 1. Characterization of CMY-2-producing cultures by MALDI-TOF MS analysis

Even if the initial attempts were done on bona fide clean transconjugants and the recipient strains analysis (that allowed us for a proof concept), it finally incorporated 29 different CMY-2 producing clinical cultures of £ coli (10), 21 different blac -2 negative cultures, one transconjugant strain with a plasmid carrying blacm-2 and the corresponding recipient strain (£. coli HBC1 a1 , £ coli HB101 respectively). In the method £ co// ^' ATCC 25922 and £ co// ^' ATCC 35218 were also included also as negative controls.

All cultures were previously characterized both phenotypically (by disk diffusion method and synergy test) and genotypically (by PCR and sequencing) aimed at different β-lactamase gene families (CMY-, TEM-, CTX-M-). Assays were performed as already described (10, 1 1 , 12). There was no epidemiological link among these cultures (Table 1 ). As test material, cultures were obtained on Mueller Hilton (MH) (Britiania, Argentina) agar plates without antibiotics (for all susceptible and resistant cultures) or in MH agar plates supplemented with ampicillin (Sigma- Aldrich, USA) 100 mg/L (for all resistant cultures), incubated overnight at 37°C. Sample preparation for MALDI-TOF MS

The organic solvent extraction method is almost the same already in use in most clinical laboratories, aimed at total intact protein analysis after extraction. Briefly, a loopful of bacterial material (4-5 colonies) was suspended in 300 μΙ distilled water, mixed by vortexing for 30 s; after that 900 μΙ of room temperature absolute ethanol (Sigma-Aldrich, USA) was added. The suspension was vortexed vigorously (30 s) and centrifuged at 17,000 X g for 2 min at room temperature (13). After discarding the supernatant, the MALDI target spot was re-suspended in 100 μΙ of extraction solvent (formic acid - isopropyl alcohol - water, 17:33:50 ratio by volume) (Sigma-Aldrich, USA), vortexed for 30 s, and then centrifuged (17,000 X g, 2 min, room temperature) to obtain the clean supernatant extract (7). Target spot loading

The double layer sinapinic acid (SA) (Bruker Daltonics, Germany) technique was used for intact protein detection. To form the first layer, 0.7 μΙ of a saturated solution of SA in absolute ethanol was loaded onto the MALDI target spot and allowed to dry at room temperature. As a second layer, the protein extracts were mixed 1 :1 with SA 10 g/L solution (30:70 [v/v] acetonitrile: 0.1 % trifluoroacetic acid in water) (Sigma-Aldrich, USA). One μΙ_ of the sample/matrix mixture was then deposited onto the first layer. Samples were allowed to air dry before being loaded in the mass spectrometer (14, 15). Each sample was pipetted 3 times on a stainless steel MALDI target plate (in triplicate).

Spectra acquisition and statistical analysis The mass spectra were obtained using a Microflex LT mass spectrometer by flexControl 3.4 software (Bruker Daltonics, Germany). The parameters were set up as follows: positive reflector mode within the mass range of 17,000 Da to 50,000 Da. Spectrometer ion source 1 , 19.98 kV; ion source 2, 17.93 kV; lens 5.50 kV; pulsed ion extraction 260 ns; detection gain, 2877 V; sample rate and electronic setting 0.50 GS/s; High mass range. Laser frequency 60 Hz and laser range between 80-100%.

Spectra of each spot were captured twice in automatic mode. Each spectrum was obtained after 800 shots (16x50 laser shots for the acquisition control software (Bruker Daltonics, Germany). Spectra were analyzed using flexAnalysis 3.4 software (Bruker Daltonics). Before each run, the spectra were externally calibrated using the Standard II calibration mixture (Bruker Daltonics).

All spectra were assessed with the software ClinProTools 3.0 (Bruker Daltonics, Germany) (16).

In another embodiment, the invention obtains characteristic MALDI-TOF-MS spectra from presence and absence of KPC-2 carbapenemase. In this case, clinical cultures of K. pneumoniae, S. marcescens, E. coli and E. cloacae were prepared. Susceptibility tests were performed according to CLSI. Cultures were analyzed by PCR and grouped according to the presence of β-lactamases genes. Detection of KPC-2 β-lactamase was carried out by MALDI- TOF after protein extraction with organic solvents. MALDI-TOF spectra were recorded by MALDI Biotyper system and analyzed using the ClinProTool software. A KPC-2 producing E. coli XL1 blue transformant was used as a positive control. Those cultures negative for blaKPc (ESBL producers or susceptible strains) and E. coli XL1 blue were included as negative controls.

The results showed that 57clinical strains (41 K. pneumoniae, 4 S. marcescens, 5 E. coli and 7 E. cloacae) (TABLE 3) were categorized as KPC-2 producers. A distinct peak at m/z -28.477 Da, in agreement with the theoretical MW of the mature protein of KPC-2 was detected in the E. coli XL1 transformant when the instrument operated at a mass range of 17,000 to 50,000 Da (Fig. 4A). This distinctive peak was present in all KPC-2 producing clinical cultures spectra, and consistently absent in the negative control group (n = 76) (Fig. 4B)

In another embodiment, the invention obtains characteristic MALDI-TOF-MS spectra from presence and absence of CTX-M-2 and KPC-2 producing Serratia marcescens. In this case, twenty clinical cultures of Serratia marcescensand 2 controls were prepared (see table 2). Susceptibility tests were performed according to CLSI. Cultures were analyzed by PCR and grouped according to the presence of β-lactamasegenes.

Table 2.

After extractions of the proteins with organic solvents the detection of CTX-M-2 and KPC-2 β- lactamases was carried out by MALDI-TOF. MALDI-TOF spectra were recorded on a MALDI Biotyper system (Bruker) at mass ranges betweenl 7,000 to 50,000 Da. All spectra were analyzed using Flex Analysis 3.0 software and the ClinProTools 3.0, to generate statistical analysis between CTX-M-2 and KPC-2 producing S. marcescensto the other.

The results showed that a distinctive peak at m/z of 28,265 Da in the recombinant strain of CTX- M-2 and another at m/z 28,487 Da in the transconjugant strain of KPC-2 were observed. These peaks were not found in the recipient strain. Both characteristic peaks would correspond to the mass of mature proteins of CTX-M-2 and KPC-2 which have a theoretical weight of 28,287 and 28,477 Da, respectively.

The MALDI-TOF test was able to differentiate the presence of both β-lactamases in these cultures. Four cultures were categorized as CTX-M-2 producers and three as KPC-2 producers. To highlight, two strains that possessed both enzymes simultaneously, were successfully discriminated and detected by MALDI-TOF. Both characteristic peaks were absent in the 15 S.marcescens cultures lacking CTX-M-2and/or KPC-2.

In another embodiment, the invention obtains characteristics MALDI-TOF-MS spectra from presence and absence of CTX-M producing Enterobacteriaceae. The CTX-M enzymes are the most prevalent extended-spectrum β-lactamases, both in nosocomial and in community settings. In this case, 54 clinical cultures, positive for any CTX-M genes, of E. coli, Salmonella enterica, Klebsiella pneumoniae, Serratia marcescens and Proteus mirabilis were prepared (Table 4). Susceptibility tests were performed according to CLSI. Cultures were analyzed by PCR and sequencing, and grouped according to the CTX-M variant. Subsequently, CTX-M- detection was carried out by MALDI-TOF after proteins extraction with organic solvents. MALDI-TOF spectra were recorded by MALDI Biotyper system and analyzed using the ClinProTool software. E. coli transformants producing CTX-M-2, CTX-M-15 and CTX-M-9 enzymes were used as positive controls (Fig. 2) and isolates lacking any blac7x-/w- as negative controls. Negative cultures included, also, different species of Citrobacter. The results showed that by operating at a mass range of 17,000 to 50,000 Da, three peaks at approximately m/z 28,287, 28,093 and 27,950 Da were detected by MALDI-TOF and they were correlated with the mature protein of CTX-M-2, CTX-M-15 and CTX-M-14 (Fig. 3), respectively. After statistical analysis of peaks, two groups of isolates could be distinguished: the CTX-M- producing enterobacteria (clinical cultures and transformants) and non-CTX-M-producing enterobacteria (including susceptible strains and CMY-2, KPC-2 and TEM producers). Furthermore, the 54 CTX-M-producing cultures were correctly classified into CTX-M-1 , CTX-M-2 and CTX-M-9 groups.

Table 4. CTX-M Beta-lactamases strains and others

In yet another embodiment, and as an example of other proteins responsible for bacterial resistance, one laboratory strain producing the aminoglycoside modifying enzyme APH(3') was also tested, giving a peak at m/z 29,047 Da. While the foregoing written description of the invention enables one of ordinary skill to make and use what is considered presently to be the best mode thereof, those of ordinary skill will understand and appreciate the existence of variations, combinations, and equivalents of the specific embodiment, method, and examples herein. The invention should therefore not be limited by the above described embodiment, method, and examples. The scope of the present invention is defined by the claims that follow.

REFERENCES

1. Gutkind G, Di Conza J, Power P, Radice M. 2013. β-lactamase-mediated Resistance: A Biochemical, Epidemiological and Genetic Overview. Current Pharmaceutical Design 19: 164-208.

2. Morency-Potvin P, Schwartz DN, Weinstein RA. 2017. Antimicrobial Stewardship: How the Microbiology Laboratory Can Right the Ship. Clin Microbiol Rev 30:381-407.

3. Claydon MA, Davey SN, Edwards-Jones V, Gordon DB. 1996. The rapid identification of intact microorganisms using mass spectrometry. Nat Biotechnol 14:1584-1586.

4. Krishnamurthy T, Ross PL. 1996. Rapid identification of bacteria by direct matrix-assisted laser desorption/ionization mass spectrometric analysis of whole cells. Rapid Commun Mass Spectrom 10: 1992-1996.

5. Sparbier K, Schubert S, Weller U, Boogen C, Kostrzewa M. 2012. Matrix-assisted laser desorption ionization-time of flight mass spectrometry-based functional assay for rapid detection of resistance against beta-lactam antibiotics. J Clin Microbiol 50:927-937.

6. Hooff GP, van Kampen JJ, Meesters RJ, van Belkum A, Goessens WH, Luider TM. 2012.

Characterization of beta-lactamase enzyme activity in bacterial lysates using MALDI-mass spectrometry. J Proteome Res 1 1 :79-84.

7. Camara JE, Hays FA. 2007. Discrimination between wild-type and ampicillin-resistant Escherichia coli by matrix-assisted laser desorption/ionization time-of-flight mass spectrometry. Anal Bioanal Chem 389:1633-1638.

8. Papagiannitsis CC, Kotsakis SD, Tuma Z, Gniadkowski M, Miriagou V, Hrabak J. 2014.

Identification of CMY-2-type cephalosporinases in clinical isolates of Enterobacteriaceae by MALDI-TOF MS. Antimicrob Agents Chemother 58:2952-7.

9. Hart PJ, Wey E, McHugh TD, Balakrishnan I, Belgacem O. 2015. A method for the detection of antibiotic resistance markers in clinical strains of Escherichia coli using MALDI mass spectrometry. J Microbiol Methods 1 1 1 :1-8.

10. Cejas D, Fernandez Canigia L, Quinteros M, Giovanakis M, Vay C, Lascialandare S, Mutti D, Pagniez G, Almuzara M, Gutkind G, Radice M. 2012. Plasm id-Encoded AmpC (pAmpC) in Enterobacteriaceae: epidemiology of microorganisms and resistance markers. Rev Argent Microbiol 44: 182-186.

1 1. Rincon Cruz G, Radice M, Sennati S, Pallecchi L, Rossolini GM, Gutkind G, Di Conza JA. 2013.

Prevalence of plasmid-mediated quinolone resistance determinants among oxyiminocephalosporin-resistant Enterobacteriaceae in Argentina. Memorias do Instituto Oswaldo Cruz 108:924-927.

12. Saba Villarroel PM, Gutkind GO, Di Conza JA, Radice MA. 2017. First survey on antibiotic resistance markers in Enterobacteriaceae in Cochabamba, Bolivia. Revista Argentina de

Microbiologia 49:50-54.

13. Matsuda N, Matsuda M, Notake S, Yokokawa H, Kawamura Y, Hiramatsu K, Kikuchi K. 2012.

Evaluation of a simple protein extraction method for species identification of clinically relevant staphylococci by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J Clin Microbiol 50:3862-3866.

14. Dai Y, Whittal RM, Li L. 1996. Confocal Fluorescence Microscopic Imaging for Investigating the Analyte Distribution in MALDI Matrices. Anal Chem 68:2494-2500.

15. Keller BO, Li L. 2006. Three-Layer Matrix/Sample Preparation Method for MALDI MS Analysis of Low Nanomolar Protein Samples. Journal of the American Society for Mass Spectrometry 17:780- 785.

16. Ketterlinus R, Hsieh SY, Teng SH, Lee H, Pusch W. 2005. Fishing for biomarkers: analyzing mass spectrometry data with the new ClinProTools software. Biotechniques 38:37-40.

17. Stephens MA. 1974. EDF statistics for goodness of fit and some comparisons. Journal of the American Statistical Association 69:730-737.

18. Rodriguez-Sanchez B, Sanchez-Carrillo C, Ruiz A, Marin M, Cercenado E, Rodriguez-Creixems M, Bouza E. 2014. Direct identification of pathogens from positive blood cultures using matrix- assisted laser desorption-ionization time-of-flight mass spectrometry. Clinical Microbiology and Infection 20:0421-0427. Mirande C, Canard I, Buffet Croix Blanche S, Charrier JP, van Belkum A, Welker M, Chatellier S. 2015. Rapid detection of carbapenemase activity: benefits and weaknesses of MALDI-TOF MS. Eur J Clin Microbiol Infect Dis 34:2225-2234.