Evaluation of Microbial Safety of Dairies using Bacterial Proteomic Profiling via MALDI Approach

Summary

This protocol enables rapid, cost-effective evaluation of microbial composition in dairy products by combining culturomics with MALDI-TOF MS. It delivers taxonomy results comparable to 16S rRNA sequencing with a shorter turnaround time than traditional culture methods.

Abstract

We present a protocol using the matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS, hereinafter referred to as "MS") identification method for monitoring the presence of microorganisms in different dairy products (butter, mozzarella cheese, milk, cream) using a proteomic approach, modern culturomics, and various sample preparation procedures that could be incorporated to the food safety quality control laboratories as a fast, reliable, and robust tool for monitoring microbial safety. It includes a microbial isolation step, picking grown colonies, sample preparation procedures (intact cells, on-spot and in-tube extraction), setting MS analysis parameters, generating and collecting MS spectra (protein fingerprints), and finally getting taxonomy using the MS platform, followed by data interpretation and visualization. This method, validated by numerous studies, offers a more reliable and cost-effective identification of microorganisms from milk compared to commercial biochemical test systems. It stands out as a promising advancement in food microbiology, meeting the diverse requirements of laboratories through its simple protocols and significantly reduced analysis time. The presented protocol addresses the needs of the food industry by providing reliable microbial identification in a proper turnaround time (TAT) that is well-suited to manufacturing processes.

Introduction

The overall goal of this protocol is to provide a comprehensive methodology for monitoring the microbial safety of various dairy products, including butter, cheese, milk, and cream, by employing MS for bacterial proteomic profiling. This approach, which combines modern culturomics with optimized sample preparation procedures, is intended to be a fast, reliable, and robust tool for food safety and quality control laboratories. The rationale behind creating and applying this methodology comes from the limitations of standard microbial identification methods. For many years, veterinary and food diagnostic laboratories have relied on conventional biochemical tests that examine bacterial growth and metabolism on differential culture media followed by various enzymatic assays¹. While these methods, frequently based on such standard guidelines as those of Bergey's Manual of Determinative Bacteriology, are relatively inexpensive and can produce both quantitative and qualitative data, they are laborious and time-consuming, requiring extensive media preparation, sample dilution, plating, incubation, colony counting, isolation, and detailed characterization steps^2,3.

The introduction of commercial semi-automated and automated systems such as Analytical Profile Index (API), BBL Crystal, Vitek, and Biolog MicroPlates streamlined microbial identification using biochemical assays, thereby reducing costs and turnaround times. However, these systems have some drawbacks, such as poor reproducibility, a lack of entries in their respective databases, and difficulties in determining phenotypic variations among strains, particularly for non-fermentative bacteria or strains within a species with minor biochemical differences, which could lead to incorrect in vitro results⁴. Given this, MS has evolved as a potent alternative, currently common in clinical microbiology and gaining attention in veterinary diagnostic and milk quality laboratories. This approach has been complemented by its speed, accuracy, and cost-effectiveness in detecting several microorganisms¹. Numerous studies have demonstrated that MS offers more reliable identification of microorganisms from milk, delivering faster and cheaper results than commercial biochemical test systems^5,6,7. While biochemical testing may suffice for genus-level or group-level identification, species-level diagnosis often necessitates more advanced methods like MS or 16S rRNA sequencing⁸. Indeed, MS is considered a promising platform for flexible and dependable identification of food microbial isolates, meeting diverse requirements of food microbiology laboratories through its simple protocols and significantly reduced analysis time, thereby enhancing food safety^4,5. The operational costs are minimal due to low reagent consumption, and sample handling allows for greater automation and high throughput⁸. However, it should be noted that this method's utility is inherently dependent on microbial growth conditions selected (including culture media type) and a reference database species coverage, which are recognized as the key factors limiting its usefulness.

MS plays a significant role in food microbiology by identifying and differentiating foodborne pathogenic, lactic acid bacteria (LAB), and other fermentative bacteria⁹. Its utility extends to modern culturomics, enabling the identification of all microbial colonies grown on culture media and facilitating the acquisition of colony-forming unit (CFU) information⁷. This approach has already been successfully applied to discover such complex microbial communities as those presented in human intestines, involved in diabetic foot infections, or connected with the urinary tract, revitalizing culture-dependent methods in microbiological practices^10,11.

This protocol is presented to guide laboratories in effectively implementing MS for the routine evaluation of microbial safety in dairy products, addressing the industry's need for reliable microbial identification within a practical turnaround time (TAT) suitable for manufacturing processes. The protocol details crucial steps, including (1) microbial isolation from liquid (milk, cream, whey) and solid (butter, cheese) dairy products using a selection of appropriate culture media (de Man, Rogosa and Sharpe agar - MRS agar, M-17 agar, All-Purpose Tween 80 agar - APT agar, China Blue Lactose agar - CBL agar, Milk Plate Count Agar - MPCA, Tryptic Soy Agar - TSA) and tailored incubation conditions; (2) standardized sample preparation procedures for MS analysis, offering options such as direct colony transfer, on-target formic acid extraction, and in-tube ethanol/formic acid protein extraction to accommodate different bacterial types and laboratory workflows; and (3) systematic MS analysis, including instrument calibration, defining optimal spectra acquisition parameters, and guidelines for data interpretation, including score value assessment and consistency checks for confident taxonomic assignment. By providing a structured and validated approach, this protocol aims to enhance the accuracy and efficiency of microbial monitoring in the dairy industry, ultimately contributing to improved food quality and public health.

Protocol

The reagents and the equipment used in this study are listed in the Table of Materials.

1. Isolation of dairy-related bacteria

1. Obtaining bacterial cultures from liquid or solid dairy products
   1. Culture media selection
      1. For bacterial isolation from dairy samples, select the culture media according to the expected microbial groups, including general growth media, media for Gram-negative bacteria, lactic acid bacteria (LAB), spore-forming bacteria, as well as slow-growing or fastidious microorganisms.
         NOTE: To maximize recovery of both dominant and rare species, multiple media should be inoculated in parallel. A set of example media for microbiome analysis from dairy samples is presented in Table 1.
   2. Inoculation and plating
      1. Obtain a representative sample of the dairy product and mix thoroughly.
      2. Aseptically transfer 1 mL (liquid samples: milk, cream, buttermilk) or 1 g (solid samples: cheese, butter) of the sample to a 15 mL conical centrifuge tube containing 9 mL of sterile liquid medium (peptone water / 0.85% NaCl solution). For solid samples, warm the liquid medium to 40-42 °C before adding the sample.
      3. Homogenize the sample by vortexing vigorously for 30-60 s. For fatty or hard samples, continue homogenization until a uniform suspension is obtained.
      4. Prepare a series of ten-fold dilutions by taking 1 mL of suspension and transferring it to successive tubes containing 9 mL of sterile solution. Mix each dilution well.
         NOTE: Depending on the sample type and expected level of microbial contamination, prepare dilutions ranging from 10^-1 to 10^-6. For most samples, dilutions up to 10^-3 will be sufficient; however, when analysing fresh milk or blue cheese, it will be necessary to prepare further dilutions. To ensure accurate isolation and ease of colony selection, the optimal inoculation dilution should give from 30-300 colonies per plate. This range allows for clear differentiation of individual colonies and minimizes the risk of overlapping growth, facilitating the obtaining of uncontaminated and homogeneous bacterial cultures for later analysis. For some processed dairy products (e.g., buttermilk), an undiluted fresh sample (e.g., 0.1 mL) might be used for direct plating.
      5. Take 0.1 mL of the sample or its dilution and transfer onto the surface of previously prepared Petri dishes containing the appropriate agar medium.
      6. Evenly spread the inoculum over the agar surface using a sterile spreader. Allow the inoculum to absorb into the medium for 15 min, then invert the plates.
   3. Incubation
      1. Aerobic Incubation: for all investigated samples, inoculate APT, M-17, CBL, MPCA, and TSA agar plates, incubate under aerobic conditions at 37 °C and 30 °C for 24 h. The incubation temperatures were determined based on the optimal growth requirements of different microbial groups crucial for assessing both the hygienic quality (30 °C for general mesophiles) and public health safety (37 °C for human-associated pathogens and indicator organisms) of dairy products.
         NOTE: If intensive growth of spore-forming bacteria is expected, in the case of TSA, a shorter (~12 h) incubation time is preferred.
      2. Anaerobic Incubation: for all investigated samples, inoculate MRS Agar plates, incubate under anaerobic conditions at 37 °C and 30 °C for 24-72 h, checking the microbial growth every day. The incubation temperatures were determined as mentioned in step 3.1.
         NOTE: Anaerobic incubation conditions can be achieved by placing plates in airtight containers with anaerobic atmosphere-generating sachets, which eliminate oxygen and provide a suitable atmosphere for the growth of anaerobic bacteria.
   4. Obtaining pure cultures
      1. After incubation, inspect the plates and select grown single colonies based on their morphological characteristics (e.g., size, shape, color, texture).
      2. Purify the selected colonies by subculturing (streaking) onto the same agar medium using 1 µL microbial loop and the same incubation conditions.

Culture medium	Application	Culture condition
Tryptic Soy Agar (TSA)	general medium for the cultivation of a wide range of bacteria	30–37 °C, 24–48 h, aerobic
Blood agar (BLA; with 5% sheep blood)	enriched, differentiation of hemolysis, useful for Streptococcus, Listeria, Bacillus	30–37 °C, 24–48 h, aerobic
MRS agar (de Man, Rogosa and Sharpe agar)	selective for Lactobacillus and other LAB Add 1% Tween-80 to support the growth of L. casei, L. delbrueckii, L. helveticus or L. acidophilus.	30 °C, 48–72 h, anaerobic or microaerophilic
M17 agar	selective for Lactococcus, Streptococcus thermophilus	30 °C, 48–72 h, anaerobic or microaerophilic
APT agar (All-Purpose Tween 80 agar)	selective for LAB and some Bifidobacteria	30 °C, 48–72 h, anaerobic or microaerophilic
Scheadler agar	selective and enriching, isolation of anaerobes, including Bifidobacterium, Clostridium	30–37 °C, 48–72 h, anaerobic
CBL agar (China Blue Lactose agar)	differential medium, for coliform bacteria and other lactose-fermenting bacteria	37 °C, 24–48 h, aerobic
Milk Plate Count Agar (MPCA)	universal, for the total number of aerobic mesophilic bacteria	30–37 °C, 24–48 h, aerobic

Table 1: List of suggested media for microbiome analysis of dairy samples with specified culture conditions.

2. Bacterial samples preparation for MS analysis

1. Before starting, prepare the MALDI matrix solution
   NOTE: The most commonly used matrix compound is HCCA, α-Cyano-4-hydroxycinnamic acid. This compound is photosensitive, so prepare and store the matrix in dark containers or away from light.
   1. Prepare the solvent mixture using the following proportions: 50% acetonitrile (HPLC-grade), 47.5% water (HPLC-grade), and 2.5% trifluoroacetic acid (TFA).
   2. Dissolve the matrix in the solvent mixture to a final concentration of approximately 10 mg/mL (or until saturation - until undissolved crystals remain visible).
   3. If the solution is not clear (crystals remain), centrifuge or filter the solution (e.g., through a 0.22 µm syringe filter) to obtain a clear liquid.
      ​NOTE: Storage in the refrigerator is allowed for short periods, provided a low number of defrost cycles.
2. To identify bacteria using MS, use one of the 3 sample preparation procedures:
   1. Direct Colony Transfer - Use this simplest and fastest method when dealing with well-growing, non-lysis-resistant bacteria, such as most Gram-negative bacteria.
      1. Using a sterile loop or toothpick, transfer a small amount (~1 µL) of the grown colony to a spot on the MALDI plate.
      2. Smear the bacterial material directly onto a designated spot on the MALDI target plate to create a thin, even layer.
      3. Once dried, overlay the spot with 1 µL of HCCA matrix solution and allow to air dry at room temperature.
   2. On Target Formic Acid Extraction - Use for difficult-to-identify bacteria (e.g., Gram-positive, yeast) whose cell walls make it difficult to release proteins without additional processing.
      1. Pick up a small amount of bacterial biomass from a single bacterial colony using a disposable microbial loop or sterile toothpick.
      2. Smear the bacterial material directly onto a designated spot on the MALDI target plate to create a thin, even layer.
      3. Overlay the smeared bacterial material with 1 µL of 70% formic acid solution and air dry at room temperature.
      4. Once dried, overlay the spot with 1 µL of HCCA matrix solution and allow to air dry at room temperature.
   3. In-Tube Protein Extraction (Ethanol/Formic Acid Method) - Use this method for difficult-to-identify microorganisms, including some Gram-positive bacteria, yeasts, and mycobacteria.
      1. Using a sterile loop or swab, transfer 1-3 complete colonies from an agar culture plate to a 1.5 mL microcentrifuge tube containing 300 µL of sterile water.
      2. Add 900 µL of absolute ethanol to adjust the final ethanol concentration to approximately 75%. Mix or vortex for a few seconds.
      3. Centrifuge the sample for 2 min at 15,000 x g (room temperature).
      4. Carefully discard the supernatant without disturbing the cell pellet.
         NOTE: In the case of mucous strains, additional centrifugation at 4 °C is recommended.
      5. Leave the pellet tube open until all residual ethanol has evaporated (5-10 min, room temperature).
      6. Add 20-50 µL 70% formic acid (FA) to the dried cell pellet. Mix by pipetting or vortexing until the pellet is fully resuspended.
         NOTE: The volume should be proportional to the amount of biological material.
      7. Add an equal volume of acetonitrile (ACN) to the formic acid suspension. Mix carefully.
      8. Centrifuge at 15,000 x g for 2 min (room temperature).
      9. Transfer 1 µL of the supernatant onto a designated spot on a MALDI target plate.
      10. Allow the spot to air dry completely at room temperature.
      11. Overlay the dried sample spot with 1 µL of α-Cyano-4-hydroxycinnamic acid (HCCA) matrix solution.
      12. Allow the matrix to air dry completely at room temperature before MS analysis (Figure 1).
          NOTE: 70% Formic Acid: Corrosive; causes burns. Handle in a fume hood. Acetonitrile: Flammable and toxic. Keep away from flames and avoid inhalation/contact. Trifluoroacetic Acid (TFA): Strong, corrosive, and volatile; avoid skin/eye contact and inhalation. Always wear a lab coat, chemical-resistant gloves, and safety goggles. Handle all reagents in a fume hood to avoid inhalation. Avoid skin and eye contact; work away from open flames. Collect acidic and organic solvent waste in clearly labeled, compatible containers for proper disposal.

Figure 1: A decision tree that visualizes the process of selecting an appropriate sample preparation procedure for bacterial isolates identification via the MS technique. Please click here to view a larger version of this figure.

3. Spectra acquisition and microbial identification using the MS technique

1. Calibration
   NOTE: For accurate mass determination in MS, prior to analyzing unknown samples, calibrate the instrument using appropriate standards. Proper instrument calibration is crucial for reliable identification.
   1. Prepare calibrator spots on the MALDI target plate containing 1 µL of: (a) Bacterial Test Standard (BTS), containing an extract of Escherichia coli DH5 alpha, covering a mass range of 3.6-17 kDa (for MBT Biotyper); (b) Microbiology Calibrator containing E. coli ATCC 25922 protein extract, ribonuclease, and myoglobin, covering a mass range of 3.6-17 kDa (for EXS2600 system). Allow to air dry and overlay the dried calibrator spots with 1 µL of HCCA matrix solution, similar to sample preparation.
   2. Load the MALDI target plate containing the prepared samples and calibrant spots into the instrument. Run analysis for calibrant spots using calibration mode (automatic).
   3. Verify the calibration to ensure mass accuracy is within acceptable limits across (100 ppm) the specified mass range (2-20 kDa ).
2. Spectra acquisition
   NOTE: When the instrument is calibrated, mass spectra from each prepared sample spot are acquired.
   1. General acquisition parameters
      1. Operate the mass spectrometer in linear positive ion mode.
      2. Set the mass detection range from 2,000-20,000 m/z (2-20 kDa).
   2. Use the following instrument-specific settings
      1. Ion source voltages: ion source 1 (IS1) at 20 kV, ion source 2 (IS2) at 18 kV, lens voltage at 9 kV. Acceleration voltage can be up to 25 kV.
      2. Detector voltage: Set to approximately 2.65 kV.
      3. Pulsed ion extraction settings: 100 ns, delayed extraction time fixed at 230 ns.
      4. Laser power settings: adjust manually to obtain the best spectrum quality (in view of the number of signals and signal intensities), typically ranging from 40%-65% of the minimum required power.
      5. Number of shots per spectrum: acquire a sufficient number of laser shots from different regions of each spot to generate a sum spectrum with good signal-to-noise ratio (S/N)→ (a) average of 200 shots, taken in 50-shot steps from 4 different regions of each spot or (b) acquisition in 500 shot steps - 50 shots from ten different positions of the target spot using AutoXecute mode.
   3. Sample measurement strategy
      1. Measure multiple spots per isolate (2 spots per isolate at least).
      2. Acquire spectra in replicate for each spot to ensure consistency and accuracy of findings → measure in duplicate, aiming for a certain number of total spectra per sample (4 spectra per sample at least → technical replicates).
         NOTE: Subject biological samples (at least 3) to get biological replicates.

4. Microorganism identification and data analysis

NOTE: Organism identification occurs by comparing the obtained protein profile (specific protein "fingerprint" of the microorganism) with a reference database, and the resulting correlation of both signal position and intensity is used to generate a matching value, which describes the confidence level with which the studied organism was matched to its corresponding reference microorganism. Identify microorganisms by comparing processed spectra to reference databases.

1. Spectra processing
   1. Use appropriate software for spectra processing: (a) flexAnalysis for MALDI Biotyper systems or (b) EX-Accuspec for EXS2600 systems.
   2. Apply a Savitzky-Golay method as a smoothing algorithm (width 2 m/z, 10 cycles).
   3. Perform baseline correction to remove background noise using the TopHat algorithm (S/N threshold = 2, TopHat filter width 0.2 Da ).
   4. Use a centroid mode as a peak detection algorithm (S/N threshold = 2 ).
2. Database comparison and score interpretation
   1. Upload the processed mass spectra of unknown isolates to the MALDI platform of microbial identification. Subsequently, run the identification.
      NOTE: The following software was used in these studies: Biotyper version 4.1.100, Ex-Accuspec version V1. In the case of MALDI Biotyper software, main spectra (MSPs) creation from raw spectra is recommended to ensure the best matching reliability.
   2. Evaluate the identification results according to manufacturer standard thresholds.
      NOTE: For MALDI Biotyper and EXS2600, score values are:
      2.300-3.000: highly probable species identification;
      2.000-2.299: secure genus identification, probable species identification;
      1.700-1.999: probable genus identification;
      0.000-1.699: not reliable identification.
      During evaluation of the identification result derived from MALDI Biotyper platform support findings using Consistency Categories: A- Species Consistent → best match and all other matches with a similarly high score value (>1.999 ) point to the same species; B - Genus Consistent → best match and other matches with similar score values (1.700 - 1.999) point to the same genus, but may differ at the species level; C- No Consistency → requirements for species or genus level consistency have not been met.

Results

Culturomic approach
The culturomic approach is a strategy aimed at maximally diversifying the conditions of microorganism cultivation in order to increase the chances of isolating, among others, rare, difficult to cultivate, or previously undescribed species¹². This makes it possible to obtain a much broader picture of the biodiversity of microflora present in various samples, e.g., food samples. Increased diversity of microorganisms can be observed already at the stage of vis...

Discussion

The protocol dedicated to microbial identification of dairy products offers an important advancement over traditional methods by ensuring optimal consensus on speed, reliability, and cost-effectiveness. Nevertheless, some crucial steps within the protocol demand careful attention. Firstly, the isolation and culturing phase is foundational. The culture media and incubation conditions must match the expected microbial diversity and load. Obtaining pure cultures is paramount, often requiring subculturing to avoid mixed prof...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
70% formic acid	Chempur	115676307
Acetonitrile	Merck KGaA	1.00014.1011
APT agar	Merck KGaA	1.10453.0500
Bacterial Test Standard	Bruker Daltonik GmbH	8255343
CBL Agar (China Blue Lactose Agar)	Merck KGaA	1.02348.0500
Columbia Blood agar	Oxoid	CM0331B
M17 agar
Milk Plate Count Agar (MPCA)	Oxoid	cm0681
MRS agar (de Man, Rogosa and Sharpe Agar)	Merck KGaA	1.10660.0500
MSP 96 target polished steel BC	Bruker Daltonik GmbH	8280800
Peptone water	Sigma-Aldrich	70179
Scheadler agar	Sigma-Aldrich	91019
Sheep blood	Biomaxima	SL0160-100
trifluoroacetic acid	Sigma-Aldrich	T6508
Tryptic Soy Agar (TSA)	Sigma-Aldrich	22091
Water	Merck KGaA	1.15333.2500
α-Cyano-4-hydroxycinnamic acid	Bruker Daltonik GmbH	8201344