Using Raman spectroscopy to discriminate viability states of Bacillus cereus exposed to saline or disinfectant stress

Using Raman spectroscopy to discriminate viability states of Bacillus cereus exposed to saline or disinfectant stress | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Using Raman spectroscopy to discriminate viability states of Bacillus cereus exposed to saline or disinfectant stress Jonathan Dikec, Eglantine Chalivat, Darius L. Abedini, Sabine Debuiche, and 4 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9112872/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Bacillus cereus is a foodborne pathogen able to enter in a viable but non-culturable (VBNC) state, in which cells remain metabolically active but escape detection by conventional culture-based methods used in the food industry. Raman microspectroscopy has emerged as a promising tool for VBNC detection due to its high sensitivity and single-cell resolution. This study evaluated the ability of Raman spectroscopy to discriminate VBNC B. cereus cells from other viability states. Specifically, we investigated (i) whether stressful conditions representative of food-processing environments can induce the VBNC state in B. cereus , and (ii) whether Raman spectral profiles allow differentiation among viability states. Three environmental B. cereus strains were exposed to saline solution and two commonly used food-industry disinfectants for 20 min or 24 h to induce stress. Results demonstrated that such conditions can induce a VBNC state in B. cereus . One strain was further labeled with deuterium, and Raman spectra were collected. Analyses focused on the C–D band and the fingerprint regions. The C–D region enabled discrimination between unstressed and stressed cells, while clustering analysis of the fingerprint region successfully separated unstressed, stressed/injured, VBNC, and dead cells. Mean spectra of each cluster revealed that VBNC cells exhibited marked changes in bands associated with DNA, proteins, cell wall components, and lipid membranes. Overall, this study demonstrates that Raman microspectroscopy, particularly fingerprint region analysis, provides a rapid and non-destructive approach to reliably distinguish VBNC B. cereus cells from other viability states, highlighting its potential for detecting dormant bacteria in food environments. Applied & Industrial Microbiology Food Science & Technology Spectroscopy Food security Raman spectroscopy Bacillus cereus viable but non-culturable (VBNC) viability state Figures Figure 1 Figure 2 Figure 3 1. Introduction Bacillus cereus is a spore forming foodborne pathogen widespread in the environment. This bacteria has been responsible for several outbreaks around the world(Choi & Kim, 2020 ). In France, B. cereus was ranked in 2014 as the second cause of foodborne outbreaks, leading to 18 hospitalizations(Glasset et al., 2016 ). Moreover, according to the European Food Safety Authority (EFSA), this species has been evaluated as an increasing cause of foodborne outbreaks in Europe(Authority (EFSA) & European Centre for Disease Prevention and Control (ECDC), 2024). In France, over the past two decades, the proportion of foodborne outbreaks attributed to B. cereus has increased significantly (from 2% in 1996 to 23% in 2013 of all reported foodborne outbreaks in the country). This increase is reflected in the number of human cases, which grew from 703 in 2009 to 2,246 in 2013(Authority & European Centre for Disease Prevention and Control, 2012, 2013, 2014, 2015). In parallel, some studies also pointed out that the presence and distribution of B. cereus in different food products around the world could potentially be impacted by climate change(Carlin et al., 2010 ; Feliciano et al., 2020 ). One explanation of the prevalence of this bacterium is its ability to sporulate. Bacterial spore is a dormant form of bacteria that can resist to extreme conditions and can sometimes pass-through decontaminations processes if they are not properly performed. Resistances to high temperature, pressure and other decontamination processes has been extensively studied over the past years and is tightly controlled in food processing(Setlow, 2006 ; Soni et al., 2016 ). However, in the environment it exists another dormant form called “viable but not culturable” (VBNC) state. In this state, bacteria can no longer grow, divide, or produce toxins, but they retain reduced metabolic activity and maintain membrane integrity. Like spores, VBNC cells can be reactivated when placed under favorable conditions, resuming growth and toxin production(Balagurusamy et al., 2024 ). However, they are less resistant to extreme environmental conditions than bacterial spores. Unlike spores, VBNC cells cannot be detected using the standard protocols commonly used in the food industry, which rely on bacterial growth on culture media. The VBNC state has been studied far less than the bacterial spores, particularly in B. cereus . Only a limited number of studies have demonstrated the ability of Bacillus spp. to enter a VBNC state. In particular, VBNC B. cereus has recently been reported in food matrices such as meats and milk(El-Aziz et al., 2018 ; Morawska & Kuipers, 2022 ; Rowan, 2004 ; Tabassum et al., 2020 ; Xiong et al., 2024 ). However, the origin of these VBNC cells in food products remains unclear: they may either result from environmental contamination or be induced by stresses encountered within the food industry, such as cleaning and disinfection procedures. Indeed, the VBNC state can be induced by many factors, including cold temperatures or disinfectants, with variable exposure times and concentrations(Balagurusamy et al., 2024 ). Therefore, it is crucial to determine whether the VBNC state of this pathogen can be triggered by stresses typically encountered in the food industry, as this would have important implications for both food safety and public health. One approach has been developed and applied to detect VBNC Bacillus cereus . Indeed, the propidium monoazide quantitative polymerase chain reaction (PMA-qPCR) combined with cultural methods has already been employed to assess the presence of VBNC state in meat and milk samples(Cattani et al., 2016 ; El-Aziz et al., 2018 ). Nonetheless, these methods remain subject to major limitations. Biomolecular approaches are primarily constrained by their detection threshold, typically ranging from 10³ to 10⁴ cells/mL. Recently, Raman microspectroscopy has been used as a new approach to detect the VBNC state of different foodborne pathogens. Indeed, using this technique allow to work at the scale of the single-cell and potentially avoid cultural phase which is a real limitation to study VBNC cells. Several studies have used the microspectroscopy Raman coupled with deuterium isotopic probing (Raman-DIP), allowing detection at single cell level, the viable cells and cells with less active metabolism identified as VBNC cells(Qi et al., 2022 ; Zhang et al., 2018 ). With this method, the viable and culturable cells with higher metabolic activity uptake the deuterium, resulting in a Carbone-Deuterium (C-D) band visible on the Raman spectra around 2040–2300 cm − 1 (Trigueros et al., 2023 ). For VBNC cells, this band shows a reduced intensity. Raman vibrational spectroscopy also has the advantage to identify quickly bacteria using the bands signatures in the “fingerprint” region between 700 cm − 1 and 1800 cm − 1 (Rebuffel et al., 2019 ; Strola et al., 2014 ). Because the cellular compounds of VBNC cells are different from vegetative cells, this region should vary and may be used to distinguish the VBNC from viable culturable and dead state. However, no study has attempted to determine whether this spectral region alone is sufficient to distinguish VBNC cells from other states of viability in B. cereus . The use of this spectral region could have the advantage of avoiding the time-consuming labeling of bacteria with deuterium, in addition to providing an informative spectral signature that would allow for further research beyond simple VBNC detection. Based on this observation and regarding the state of knowledge on the VBNC state of B. cereus , the main objective of this study was to evaluate how Raman spectroscopy can allow to discriminate VBNC B. cereus cells from other viability states. From this general aim, two specific objectives were defined: (i) to determine whether B. cereus can be induced into a VBNC state under stresses commonly encountered in the food industry, and (ii) to discriminate VBNC cells from other viability states based on their Raman spectra. To achieve these goals, the viability of three environmental strains of B. cereus were assessed through a combination of molecular approaches (qPCR and PMA-qPCR) and colony-forming unit (CFU) counting after exposure to stress. Three types of stress conditions were applied, each at short (20 min) and long (24 h) durations: saline solution (NaCl) and two widely used disinfectants in the food industry, with triamine or hydrogen peroxide as the main active compounds. One environmental strain was then selected for Raman spectroscopic analysis. The C-D bands region and the fingerprint region have been both used to observe the performance of Raman spectroscopy to discriminate VBNC cells from viable culturable and dead cells. 2. Materials & methods 2.1 Bacterial strains & culture conditions Three environmental strains of B. cereus (AF16-61-CCGS1, AF16-71-CCPE1, AF16-86-CCCH2) isolated from cooked shrimps, were used in this study and are referred in the manuscript as GS1, PE1 and CH2 respectively. All bacterial strains were stored at -80°C in a tryptone soy broth (TSB,Oxoid, Basingstoke, United Kingdom) supplemented with 20%(v/v) of glycerol. These strains were plated on Trypticase Soy Agar with 0.6% Yeast Extract (TSAYE, Oxoid, Basingstoke, United Kingdom) and incubated for 24 h at 30°C. Several colonies were then picked with a loop and cultured in TSB for 24 h at 30°C under agitation (400 rpm). The bacterial suspension was adjusted to reach a final concentration of 1⋅10 7 CFU/mL. A volume of 40 µL of each adjusted suspension was added to 3960 µL stress solution (10 − 2 dilution) for 20 min or 24h at 30°C under agitation (400 rpm). The stress solution consisted of a 1:1 mixture of TSB and either one of the following: the disinfectant Topax (P3-Topax 990, Ecolab, Arcueil, France) or Topactive (P3-Topactive DES, Ecolab, Arcueil, France) or NaCl at various concentrations. As both Topax and Topactive, contain a mix of active compounds, they are referred to by their main active ingredient: Topax as TA (TriAmine) and Topactive as HP (Hydrogen Peroxide). Control experiments were carried out using sterile water. After incubation, the samples were centrifuged at 5000 g for 10 min. The supernatants were removed and the pellet resuspended in 4 mL of NaCl physiological solution. The suspensions of cells were homogenized by vortexing. These prepared samples were then used for viable culturable bacterial enumeration and quantification of total and viable bacteria. In order to verify the absence of bacterial spores in the samples, part of these suspensions was heated to 80°C for 20 minutes, then cultured on TSAYe medium for 24 hours at 30°C, in parallel with microscopic observation. If no colony developed after this heat treatment and no spores were visible under the microscope, the samples were considered spore-free. 2.2 Viable culturable bacterial enumeration Final suspensions were diluted in NaCl physiological solution until 10 − 2 for the stressed samples and until 10 − 4 for control samples to have colonies that are sufficiently separated and be able to count them. A volume of 50 µL of from each dilution was plated using a spiral plater (Easyspiral, Interscience, Saint Nom la Breteche, France) on TSAYE (Oxoid, Basingstoke, United Kingdom), then incubated for 24 h at 30°C. Viable culturable bacteria were enumerated using a colony counter (Scan500, Interscience). All cultures and enumerations were performed in triplicate. 2.3 Quantification of total, viable bacteria Total DNA and DNA from viable cells were quantified by qPCR and PMA-qPCR, respectively. From the final suspension, 495 µL was mixed with 5 µL of nuclease-free water (Qiagen, Hilden, Germany) for total DNA extraction for standard qPCR. For PMA-qPCR, another 495 µL aliquot was treated with 5 µL of PMA (Propidium monoazide, Biotium, Fremont, USA) to a final concentration of 50 µM. PMA-treated samples were incubated for 5 min at room temperature in the dark, followed by light exposure at 100% for 10 min in an Eppendorf tube using a PhAST Blue lamp (GenIUL, Terrassa, Spain). Cells were then centrifuged at 5,000 g for 10 min at room temperature. The pellet was resuspended in 180 µL of a lysis buffer composed of Tris(hydroxymethyl)aminomethane hydrochloride at 20 mM (TrisHCl, Sigma Aldrich), Ethylenediaminetetraacetic acid (EDTA, Sigma Aldrich) at 2 mM; Triton X-100 (Sigma Aldrich) at 1.2% and lysozyme (Roche, Meylan, France) at 20 mg/mL. DNA extraction for both qPCR and PMA-qPCR, was performed with the DNeasy® Blood & Tissue kits for the Purification of Total DNA from Animal Tissues (Qiagen, Hilden, Germany). Elution was carried out in 100 µL of AE buffer. Extracted DNA was stored at -20°C for up to 5 days. Quantification of total and viable bacteria by qPCR and PMA-qPCR was performed as described in Brauge et al. ( 2018 )(Brauge, Faille, Sadovskaya, et al., 2018 ) using the strain B. cereus CUETEM98/4 as a positive control and B. spizizenii ATCC 6633 as a negative control. Standard curves were obtained using DNA covering the range from 10 1 to 10 9 genome equivalents (GE) per mL. The GE was calculated on the basis of a standard B. cereus ATCC 14579 genome of 5.92⋅10 6 bp and following the protocol described by Brauge et al. in 2025(Brauge et al., 2025 ). The quantification limit was around 4.12log(GE/mL). The bclA gene was used as the target for qPCR amplification. Amplification were carried out in a final volume of 25 µL, containing 2.5 µL of extracted DNA, 12.5 µL of SYBR Premix Ex Taq (2X SYBR qPCR Premix Ex Taq, Takara), 1.5 µM of forward primer C1 (5′-CAT CCG GAC TAG GAC TTC CA-3′) (Eurobio, Les Ulis, France), and 1.5 µM of reverse primer C2 (5′-TTG CCG CAG TAT ATA CGA TAA CA-3′) (Eurobio, Les Ulis, France), as described by Brauge et al. in 2017(Brauge, Faille, Inglebert, et al., 2018 ). Amplification and data acquisition were performed in a LightCycler 480 System (Roche). The amplification protocol was as follows: an initial denaturation at 90°C for 30s, followed by 45 cycles of: 10 seconds at 95°C, 10 s at 55°C, and 10 s at 72°C. Cycle threshold (Cq) values were calculated automatically by the LightCycler 480 software using the second derivative method. Cq value below the quantification threshold were set to zero. All DNA extractions and qPCR assays were performed in triplicate. 2.4 Raman microspectroscopic acquisition conditions From treated and untreated bacterial suspensions, 4 mL have been centrifuged at 5000 g for 10 min. The pellets were resuspended in 1 mL of deuterated nutrient media composed of 25% (v/v) 4x concentrated TSB and 75% (v/v) of D 2 O at 99.9% atom (Sigma Aldrich, Saint Quentin Fallavier, France). The suspensions were incubated for 3 h at 30°C under agitation (400rpm). Following incubation, samples were washed twice by centrifugation at 5000 g for 10 min, and the pellets were resuspended in 1 mL of phosphate-buffered saline (PBS, Merck, Germany). When bacterial concentration was low, the resuspension volume was reduced to concentrate the bacteria. For Raman acquisition, 4 µL of suspension was deposited on a quartz slide and air-dry 15 min at room temperature. Raman spectra were acquired using an XploRA PLUS microspectrometer (HORIBA, Palaiseau, France) and LabSpec 6.3 software (HORIBA, Palaiseau, France). An 8 mW laser beam at l = 532 nm was focused through a 100x (0.8 NA) microscope objective (Olympus LMPLFLN). From each experiment around 35 single-celled spectra have been acquired in each tested condition. All experiments have been performed in triplicate with three independent culture to acquire a total number of around 100 single-celled spectra for each condition. 2.5 Spectra preprocessing, data & statistical analysis All spectral preprocessing and analysis have been performed using Python (version 3.11). Spectra have been cropped to only the Raman shift between 700 and 3050 cm − 1 or between 700 and 1800 cm − 1 . Then spectra have been preprocessed using RamanSpy library following a despiking using a Whitaker-Hayes algorithm with standard parameters, a Gaussian smoothing, with standard parameters expected for the parameters σ = 1, a baseline correction with an asymmetric least squares algorithm (AsLS) with standard parameters excepted for the lam parameters set at 10 4 and the p parameters set at 0.005. Finally, spectra have been normalized by dividing the intensity of each Raman shift by the calculated ‘ L2 ’ norm. From that, median spectrum has been plotted for each condition. To identify groups of similar Raman spectra, clustering was performed by combining Principal Component Analysis (PCA) on the whole spectra dataset (cropped from 700 to 1800 cm − 1 ) with HDBSCAN ( Hierarchical Density-Based Spatial Clustering of Applications with Noise ) algorithm on the PCA values. PCA was first applied to reduce the dimension of the Raman spectra and project them onto the two principal components explaining most of the variance. HDBSCAN, from hdbscan library (v0.8.1.), was then used to identify dense groups of similar spectra, using default parameters except for two parameters were the values have been set based on the clustering stability observed with the HDBSCAN condensed tree. The min_sample value corresponds to a measure of the degree of conservatism required for clustering. We therefore set this value to 13 so that areas with the highest point density are considered clusters, without considering all points outside these areas as noise, as clusters can sometimes have a more widespread structure on the PCA. The results obtained were very similar for min_samples set between 12 and 20. The min_cluster_size value corresponds to the smallest group size considered to be a cluster. We therefore set this value to 5 in order to be able to observe small potential clusters that could represent a subpopulation within the sample. As the min_sample parameters, the stability of the clustering has been tested across different parameters values and the results obtained were very similar for min_cluster_size set between 2 and 13. For each defined cluster, the corresponding spectra and their PCA values were selected. Ellipses of each cluster defined by HDBSCAN were calculated with the covariance of the selected PCA values. They were then plotted in the reduced space, to represent the spatial dispersion of the cluster. This approach was used to perform a non-subjective and unsupervised clustering. Finally, for each cluster, the spectra of each stress conditions representing at least 25% of the total spectra number inside the cluster were extracted. These extracted spectra allow to compare the specific bands linked to various cellular components between the different clusters. For the comparison of the different means bands intensity, a Welch test has been performed with p-value < 0.05. 3. Results & discussion 3.1. Induction of the VBNC state in Bacillus cereus under different stress conditions 3.1.1. Short exposure The first objective was to evaluate the VBNC state induction by saline and chemical stresses with two different exposure times for the three environmental strains of B. cereus . Three strains of B. cereus have been exposed to disinfectant or saline solution at different concentrations, then the different viability states have been observed using qPCR (total population), PMA-qPCR (viable population) and CFU (viable culturable population) methods (Fig. 1 A). The number of VBNC cells corresponds to the difference between the viable population and the culturable state. For the short exposure, all strains maintained high culturable levels under control conditions, around 5.0–6.0 log(CFU/mL). Exposure to 15% NaCl induced no difference with the control for all three strains, suggesting that saline stress alone caused moderate physiological adaptation without significant lethality (Fig. 1 A). The Bacillus genus is indeed well-known for its innate abilities to survive in saline stressful conditions, for instance by producing osmoprotectant and antioxidant compounds(Valencia-Marin et al., 2024 ). In contrast, exposure to disinfectant solutions caused more substantial decreases in viability. For a short exposure to TA at 0.025% a decrease in cultivability of around 3 log(CFU/mL) was observed for the GS1 strain compared to the control. This decrease in cultivability underlines the induction of a VBNC population with a gap of around 2 log between viable and culturable populations. For the two other strains, when bacteria are exposed to 0.025% TA for 20 min, the gap between culturable and viable population did not increase compared to the control. Nevertheless, the strain CH2 was hugely impacted by this stress with a reduction of the viable population of around 2 log (GE/ml) compared to the total population, traducing lethal impact of TA for this strain (Fig. 1 A). For the short exposure to HP at 0.1%, the strain GS1 exhibited a marked difference between viable and culturable populations, with around 2 log differences versus only 1 log difference for the control population. In contrast, for the two other strains, the gap between culturable and viable populations were consistent with the control condition, with a difference of 1.5 to 2 log (Fig. 1 A). These results reflect the heterogeneity response from different strains to the same stress. Indeed, previous researches have already demonstrated that some differences can be observed in terms of stress response like exposure to salt, temperature and oxidative stress between B. cereus strains ATCC 10987 and ATCC 14579(den Besten et al., 2006 , p. 14579; Mols et al., 2007 ). Moreover, these results are consistent with the literature where previous work demonstrated that bacteria from the Bacillus genus can enter in the VBNC state when exposed to different stresses like antibiotics, antibacterial nanoparticles or pulsed electric fields(Morawska & Kuipers, 2022 ; Rowan, 2004 ; Xiong et al., 2024 ). Here, for the first time, the results demonstrated that VBNC state in B. cereus can be induced by various types of conventional disinfectants. On the contrary, a short exposure to 15% NaCl solutions no VBNC induction is observed. These results are particularly relevant since the B. cereus strains used here were recently isolated from a food industry environment. Indeed, recent studies have also reported the presence of VBNC B. cereus in different types of meats and in milk samples, detected using PMA-qPCR(Cattani et al., 2016 ; El-Aziz et al., 2018 ). Nevertheless, in those investigations, the time and conditions leading to VBNC induction remained unknown, as the studies focused on the detection and quantification of the pathogen. It is worth noting that the duration and mechanisms of VBNC induction can vary widely among studies, from a few minutes to several months, depending on the nature of the chemical or physical stress applied. Here, we provided the first evidence that B. cereus strains can rapidly (20 min) enter the VBNC state following exposure to conventional disinfectants commonly used in industrial environments. Nevertheless, short-term exposure (20 min) to saline stress (15% NaCl) did not induce a detectable VBNC state under the tested conditions. 3.1.2. Long exposure time Based on these first results, a much longer exposure time has been tested to determine whether the induction of the VBNC state and the stress responses of the different strains could vary over time. Under prolonged stress exposure, bacterial growth was clearly inhibited, leading to a strong reduction in total cell counts compared with the control condition for all stress treatments (Fig. 1 B). For the control condition, the total population was around 9 to 10 log(GE/mL) for the three strains, whereas under stress conditions it ranged from 5 to 7 log(GE/mL). For long exposure to the NaCl solution, results differed from those obtained under short exposure, with induction of VBNC populations showing differences of around 5 and 2.5 log between viable and culturable populations for strains GS1 and PE1, respectively. Both strains also exhibited a total population reduced by 4 log(GE/mL) compared to their respective control. Following exposure to 0.025% TA for 24 h, cells of strain GS1 appeared to be affected by the treatment, with a 3 log(GE/mL) difference in total population compared with the control. Nevertheless, no VBNC induction and mortality were detected because a difference under 1 log is observed between total and viable population and between viable and viable culturable population. For the other two strains, VBNC induction is observed under the same long-exposure condition, with 0.025% TA, with a mortality induction for the CH2 strain (Fig. 1 B). After a high oxidative stress, i.e. 24 h exposure to 0.5% HP, the GS1 strain showed a 3 log (GE/mL) in total population, with complete loss of cultivability and a viable population around 5 log(GE/mL), revealing a full VBNC induction. The other two strains also exhibited a 4 log reduction in total population and a large proportion of VBNC cells, with approximately 3.5 log difference between viable and culturable counts (Fig. 1 B). It should be noted that for the short exposure condition, concentration of disinfectant based on HP have been divided by 5 for short exposure compared to long exposure, shifting from 0.5% to 0.1% (v/v). Indeed, the threshold of PMA-qPCR was not sensitive enough to allow quantification of the viable population when cells were exposed to 0.5% HP for 20 min and cells exposed to 0.1% HP for 24 h were only lightly affected with no VBNC induction. Regarding these results, it appeared that the exposure time played a significant role on VBNC induction, particularly marked by the NaCl 15% exposure where no VBNC population has been observed for short exposure, contrary to the long exposure. Previous investigations have indicated that the transition of bacterial cells into the VBNC state is a progressive process that may extend over several days(Chen et al., 2018 ; Wong & Wang, 2004 ). For instance, in the studies cited, all viable Escherichia coli populations entered the VBNC state within a period ranging from a few hours to five days following exposure to chloramine, depending on the concentration applied. Likewise, Vibrio parahaemolyticus was reported to enter the VBNC state over periods varying from several days to two months, according to the specific treatment conditions imposed on the cells. Therefore, in the case of exposure to 15% NaCl, time exposure needs to be longer than 20 min to induce a quantifiable cell population in VBNC state in the strains AF16-61ccGS1 and AF16-71ccPE1. Moreover, despite different mechanisms of action, all tested stresses appeared to induce the VBNC state in the strains of B. cereus under certain exposure time. 3.1.3 Selection of the strain for spectroscopic studies From all these results, the GS1strain has an interesting profile because it exhibited various susceptibilities to stress. First, for short exposure to NaCl 15%, the three strains seemed unaffected because the bacterial population still has the same viability as the control. For the short exposure to the TA and HP, VBNC state was induced but a viable culturable population remained. In long exposure to NaCl and HP, the totality of the viable culturable population shifted into the VBNC state. For these two conditions, bacteria have been plate on TSAYe media at 30°C to observed if resuscitation process can be observed. After 72h of incubation resuscitation happen with visible colonies on the plates for these two conditions. This last result support the evidence that the gap between VC population and viable population are due to the presence of VBNC cells. Finally, long exposure to TA seemed to lead to a light induction into VBNC state with a large proportion of viable culturable cells. However, the total population is still drastically diminished compared to the control. In summary, strain GS1 exhibited a different mix of viability population depending on the treatment. This is why, for the next phase of this study, the strain GS1, showing multiple viability profiles, has been selected. 3.2. VBNC state investigated through Raman spectroscopy The next objective was to assess if Raman spectra can be used to discriminate B. cereus VBNC of the selected strain AF16-61ccGS1 from the other viability states, using the information contained in C-D bands region and in the fingerprint region. 3.2.1. Metabolic activity measurement through Raman DIP for VBNC discrimination Strain GS1 has been exposed to the different stress solutions described previously and subsequently incubated in deuterated TSB (75% v/v D 2 O) at 30°C for 3 h prior to Raman spectral acquisition. Figure 2 represents the median Raman spectra obtained from around 100 single-cell spectra acquired over three independent cultures. Between 800 and 1800 cm − 1 , the spectra are dominated by vibrational bands corresponding mainly to amide and DNA regions. A broad C-D Raman band (2040 to 2300 cm − 1 ), indicative of deuterium incorporation and thus metabolic activity, was observed in the control condition for both short and long incubations. Similarly, a C–D band comparable to the control was also detected following short exposure to 15% NaCl (v/v). This result reinforces the molecular biology data presented in Fig. 1 A, showing that cells exposed briefly to 15% NaCl (v/v) maintained high viability. The presence of a clear C–D signal in these conditions reflects significant and measurable metabolic activities in both unstressed control and cells briefly exposed to NaCl. In contrast, no C–D bands were detected under the other stress conditions. Instead, broad, unspecific bands appeared in the same spectral region (2040–2300 cm⁻¹) across all stressed samples, suggesting that these features are not biological but rather artifacts arising from spectral noise. Because no biomolecular Raman signatures are expected in this region, these artifacts are likely due to the low signal-to-noise ratio of spectra obtained from stressed cells. Indeed, when bacteria are treated to be induced in VBNC state, intensity of the Raman signal decreased a lot compared to control condition and bacteria shortly exposed to NaCl ( Supplementary data Fig. 1 ). The low signal-to-noise ratio (SNR) measured for VBNC bacteria can be attributed to reduced biomasses, observed experimentally through smaller sizes in microscopy ( Supplementary data Fig. 2 ). Similar observations were made for the short TA exposure condition. In this condition, the cells are not supposed to be mostly in the VBNC state but still stressed. Due to their smaller cell biomass, exposure to the laser leads to a reduced Raman signal acquisition compared to larger control cells. This signal reduction makes spectra preprocessing more susceptible to noise, particularly in spectral regions without Raman signal, where the signal-to-noise ratio is less pronounced. In these areas, the baseline correction algorithm (asymmetric least squares) tends to use noise spikes as a relevant band, resulting in the appearance of artifacts. If the bacteria were more metabolically active, like unstressed cells or those briefly exposed to a 15% NaCl concentration, distinct C-D bands should have appeared and thus overwhelmed the background noise. This would have resulted in the absence of artifacts due to the algorithm. This absence of C-D bands for the stressed conditions is consistent with the previous results (Fig. 1 ), showing that in these conditions, viability was impacted by disinfectant or saline stress. In these latter conditions, bacteria were mostly in dead or VBNC state. Some research reported a measurable band in the C-D regions after cell induction to the VBNC state with chlorine stress or UV exposure(Qi et al., 2022 ; Zhang et al., 2018 ). However, these studies are different from this current one, because the authors have not worked on B. cereus and the induction was not done with the same disinfectants or saline exposure. To our knowledge, these artifacts masking the C-D band have never been reported in the literature. Nevertheless, despite these artifacts in this current study C-D bands allow in the end to distinguish the unstressed cells (Control and 20 min saline exposure) from stressed cells. This result showed the limit of using C-D bands when bacteria are considerably stressed. Nevertheless, in the fingerprint region (700–1800 cm − 1 ) where SNR is good, contrary to the C-D band, some differences can be observed between controls conditions and stress conditions. Therefore, this study proposes to investigate deeper this spectra region to see if it possible to discriminate the different viability states of a B. cereus population based on the observable differences. 3.2.2. Viability state clusterization based on fingerprint region Given that the fingerprint region (700–1800 cm⁻¹) provided consistent and interpretable signals free from the artifacts observed in the silent region, subsequent analyses were realized, focusing on this spectral window. This region encompasses vibrational modes of key biomolecules such as proteins, lipids, carbohydrates, and nucleic acids, and thus offers a robust basis for the comparison of cellular physiological states. Recent studies have also analyzed this region to investigate the VBNC induction of Lacticaseibacillus paracasei under cold temperature. Some notable differences have been observed in different parts of the spectra compared with the control condition, sufficient to discriminate between control and VBNC cells (Bao et al., 2023 ). Here, the objective was to evaluate whether the VBNC cells of B. cereus were also sufficiently different to discriminate VBNC from control cells, but also to discriminate VBNC from dead cells. Unsupervised classification by a hierarchical density-based spatial clustering of application with noise (HDBSCAN) algorithm was applied to the two first principal component (PC) of the principal component analysis (PCA) was therefore applied to identify potential spectral clusters represented by different ellipses and corresponding to different viability states (Fig. 3 ). Only the first two PCs were used for clustering because they capture the most variance. Indeed, PC1 represented 41.02% and 38.56% of the variance for short and long exposure condition respectively, while and PC2 represented 10.46% and 12.80% of the variance for short and long exposure condition respectively. Each PC3 to PC5 contributes less than 7% of the variance, and their spectral characteristics were less comprehensive and redundant than those of PC1 and PC2. Moreover, performing PCA in three dimensions did not substantially improve cluster separation. Therefore, using the first two PCs provided a sufficient representation of the data for HDBSCAN clustering (Fig. 3 ). Finally, clustering stability was assessed using the HDBSCAN condensed tree (Fig. 3 in the supplementary data). For the short exposure condition, the branch depth of each cluster in the condensed tree supported the evidence that all detected clusters were stable. For the long exposure condition, only three of the four detected clusters were considered, since one cluster, represented by green ellipses in Fig. 3 , had a shallow branch, indicating low stability, and was therefore excluded from further analysis and considered as an artifact. The result obtained for short exposure showed that spectra belonging to unstressed cells (from control conditions, plotted in yellow) and those belonging to bacteria exposed to 15% NaCl (v/v, plotted in grey), are clustered together. This is in accordance with our previous results, where no difference has been observed between control conditions and short exposure to 15% NaCl in terms of population viability (Fig. 1 ) and in their median spectra (Fig. 2 ). On Fig. 3 A, it can also be observed that bacteria exposed to HP or TA are clustered in two different clusters (plotted in blue and green), separately from the control/15% NaCl cluster. This latter result tends to show a heterogeneity in the VBNC spectra related to their way of induction. Previous studies have also observed heterogeneity in VBNC cells, even for cells induced with the same stress. For example, Bao et al. 2023 (Bao et al., 2023 ) highlighted through PCA a remarkable heterogeneity of micro-Raman spectra for L. paracasei VBNC cells. Nevertheless, their PCA does not contain a reference condition of dead cells unlike in the results presented here in Fig. 3 A&B. Thus, it is possible that a part of the heterogeneity they observe may be due to the presence of spectra belonging to dead bacteria which is not the case in the results presented Fig. 3 A&B. Here, all spectra obtained from bacteria induced in VBNC state by exposure to HP (blue cross) or TA (green triangle) are different from the dead bacteria exposed to very high concentrations of disinfectant (red cross and triangle). For long exposure to stress, it can be observed that spectra belonging to unstressed control cells were clustered separately from all the other stress conditions (Fig. 3 B). Contrary to short exposure, spectra from cells exposed to 15% NaCl for a long period are this time different from control, and some spectra corresponding to saline stress (NaCl, yellow square) tend to be clustered with spectra corresponding to bacteria exposed HP (blue cross) and a small fraction of the bacteria exposed to TA (green triangle). Molecular biology proved that the viable cells exposed to HP and NaCl were identified to be all in VBNC state, in contrast with TA exposure (green plots) where the great majority of the cell population tends to still be viable and culturable, forming a cluster separated from the other conditions (Fig. 1 & Fig. 3 B). However, this small TA cluster is likely an artifact of the clustering process, as it does not appear to be stable in the HDBSCAN condensed tree and will not be considered for the next of this study (supplementary data Fig. 3 ).In the qPCR results showed in Fig. 1 B and Fig. 3 B for cells exposed for a long period to TA, no VBNC induction was considered because there is a difference of less than one log between viable and viable culturable population. However, it can be observed a large standard deviation for the viable culturable population. In consequence, it is possible that a fraction of the viable population can shift into VBNC state but still no be detected through the use of qPCR/CFU method. This difference of observation between results obtained in molecular biology coupled with culturable methods and between micro-Raman results underline the advantage to use micro-Raman to detect VBNC population. All these results tend to show that because there are clustered together, VBNC cells spectra induced by different ways during long exposure tends to share more similarity than tothose induced by short exposure,. Thus, a longer induction period to the VBNC state, seems to homogenize the VBNC population. On the other hand, a large heterogeneity of data was observed among the spectra of bacteria exposed to saline stress over 24 h, where some of these spectra were closer to the control or dead conditions (Fig. 3 B). This latter result might indicate that even if a part of VBNC spectra tend to be similar, there may remain some subpopulation inside VBNC. Other research also revealed that some heterogeneity can be observed in the VBNC population, even in their morphology. Coutard et al. 2007 (Coutard et al., 2007 ) have observed under scanning electron microscope that V. parahaemolyticus shifted from rod shape to coccoid form when they enter the VBNC state, with a morphological heterogeneity with tiny and larger coccoid forms. A few years later, it has been observed that larger coccoid forms of V. parahaemolyticus showed a better fitness for revival and that this form was present in seafood samples(Wagley et al., 2021 ). In this present study, brightfield microscopy observation before Raman acquisition (supplementary Fig. 2), outlined a morphological shift, from the rod-shaped cells of control conditions to the coccus-shaped VBNC cells. However, no morphological subpopulation has been observed. Untreated cells or cells shortly exposed to NaCl were rod-shaped but cells treated with HP, TA (short and long exposure), or exposed to NaCl for long exposure showed tiny round forms. Other studies have observed on Listeria monocytogenes and Escherichia coli that when these cells are in the VBNC state, they shifted to an homogenous round and smaller shape without morphological visible subpopulations(Carvalho et al., 2024 ; Se et al., 2021 ). For E. coli , Se and his colleagues(Se et al., 2021 ) suggested that this morphotype change was related to regulation of protein expression occurring in the VBNC state. In B. cereus , the VBNC cells generated through different stress conditions shared similar morphological characteristics. The limited variability observed in PCA analysis reflects subtle differences rather than distinct subpopulations. After 24 h of induction, VBNC cells from all treatments converged into a comparable spectral cluster, suggesting a common adaptive physiological state regardless of the initial stress. 3.2.3. Assessment of molecular bands associated with bacterial viability states To gain deeper insight into the molecular differences associated with the various bacterial viability states, the maximal intensities of Raman bands were compared for each VBNC cluster identified with PCA and HDBSCAN to control and dead cluster, under both short and long exposure times (Table 1 ). This analysis aimed to identify which biochemical components were most affected during the transition to the VBNC state. Refer to the supplementary data Fig. 4 to see the means spectra of each cluster identify Fig. 3 . A. Nucleic acid–related bands Previous studies have reported that numerous genes and proteins are differentially regulated in VBNC cells, reflecting major shifts in cellular metabolism and molecular composition(Casasola-Rodríguez et al., 2018 ; Dong et al., 2020 ; Pazos-Rojas et al., 2019 ). First regarding on DNA/RNA bands, with bands at 780, 1099,1477, 1574, 1609 cm − 1 , clear differences in band intensity were observed for bacteria belonging to the VBNC cluster, both after short and long stress exposures, compared with control and dead cells (Table 1 ). As a general observation, both for short and long exposure to stress, bacteria belonging to the VBNC cluster showed a majority of band intensities significantly lower than control bacteria, particularly for the three bands at 780, 1477 and 1574 cm − 1 , corresponding to components only related to DNA/RNA. Surprisingly, for long exposure to stress compared to short exposure, the band intensity at 780 cm − 1 was significantly higher for bacteria belonging to VBNC compared to control cells at the opposite is observed for band at 1609 cm − 1 . This changes in bands intensities between short and long exposure suggested substantial molecular rearrangements under prolonged stress. The previously cited study from Bao and his colleagues(Bao et al., 2023 ) induced L. paracasei in VBNC state through incubation at 4°C on a De Man, Rogosa et Sharpe (MRS) medium for 120, 180 and 220 days. They have observed that Raman signal intensity changed for VBNC cells depending on the inducing time. They concluded that the entry in VBNC is a gradual process that can take several weeks for L. paracasei . This could explain what was observed here with an increase in band intensity at 780 cm − 1 for bacteria exposed longer to stress. For the comparison with the dead cluster, the VBNC cluster shows more contrasting results, with half of the bands having lower intensities and the other half having higher intensities. It is known that in the VBNC state, bacteria are still metabolically active and some genes can be up or down regulated but with a global metabolic activity lower compared to viable culturable cells(Casasola-Rodríguez et al., 2018 ; Dong et al., 2020 ; Pazos-Rojas et al., 2019 ). Therefore, all these observed results can be explained by this typical lower metabolic activity than control bacteria. This also explains why some bands intensities were higher than the signal collected for dead bacteria, where no metabolic activity remained. Several past studies had shown that genes can be up and down regulated in VBNC bacteria(Casasola-Rodríguez et al., 2018 ; Dong et al., 2020 ; Pazos-Rojas et al., 2019 ). To our knowledge, no study has investigated the process of gene regulation at different times of induction. In regards to the results in the Table 1 and those obtained by Bao and his colleagues(Bao et al., 2023 ) where a lot of variation through time can be observed on spectra belonging to VBNC bacteria, it leads to the hypothesis that during an early phase of induction, some genes transcription activities may be more intense than in bacteria in dormancy state for 24 h. Subsequent further investigations are needed to conclude on this point. B. Lipid-related bands After examining nucleic acid–related bands, attention was next directed towards spectral regions associated with other major biomolecules. Since cell envelope integrity is known to play a key role in bacterial adaptation to stress, lipid- and membrane-related bands were analyzed to assess whether structural modifications accompany the VBNC transition (Table 1 ). The results are similar for bands at 871, 1025, 1053 and 1076 cm − 1 for short and long exposure with an increase in bands intensities for spectra acquired on bacteria in VBNC state compared both to spectra belonging to control and dead cluster. In the opposite, the bands at 1312 and 1450 cm − 1 exhibited a lower intensity in spectra belonging to VBNC cluster compared to control and dead state, both for short and long exposure. All these results traduced major changes in lipid membrane that seemed deeply affected, which is not surprising in regard to the VBNC morphology seen in supplementary Fig. 2, that changed drastically compared to the control. Some other publications observed that the cell membrane of bacteria in VBNC state was altered by changes in fatty acids composition in V. vulnificus , V. parahaemolyticus and Pseudomonas putida (Day & Oliver, 2004 ; Pazos-Rojas et al., 2019 ; Yoon & Lee, 2022 ). In these publications, the VBNC state has been induced by starvation at cold temperatures and by desiccation, respectively for Vibrio and Pseudomonas . Moreover, the Raman spectra obtained by Bao and his colleagues(Bao et al., 2023 ) on L. paracasei and the results presented in the Table 1 are consistent with this line of evidence, showing that the lipid membrane of bacteria appears to be deeply affected by the entry into the VBNC state, with some bands increasing or decreasing signal intensity. C. Protein-related bands Beyond lipid alterations, modifications in protein composition and cell wall structure can also provide valuable insights into bacterial adaptation to stress. To explore these aspects, we examined Raman bands corresponding to amide and amino acid vibrations, which reflect changes in proteins and peptidoglycan organization (Table 1 ). The bands only related to amide and amino acids located at 1002, 1243, 1332, and 1660 cm⁻¹ showed significantly reduced intensity for spectra belonging to VBNC cluster compared to control and dead cluster. As observed for DNA/RNA, it is reasonable to think that this may result from the reduced metabolic activity of these dormant bacteria(Balagurusamy et al., 2024 ; Se et al., 2021 ). Moreover, previous studies also demonstrated that the cell wall constituted by peptidoglycan with amino acids and amide bonds was strongly affected during the entry into the VBNC state(del Mar Lleò et al., 2000 ; Signoretto et al., 2001 , 2002 ). From these studies, it can be observed that VBNC cells of E. coli and Enterococcus faecalis reported changes in peptidoglycan chemistry with increased crosslink and O-acetylation. Interestingly, the band at 1206 cm − 1 related to tyrosine(Cui et al., 2022 ) exhibited an intensity significantly higher in VBNC bacteria compared to the control bacteria for both short and long exposure. This amino acid is known to be affected by response to oxidative stress in bacteria such as B. subtilis , for which phosphorylation tends to be attenuated compared to eukaryote cells(Shi et al., 2024 ). Moreover, a recent study on B. cereus showed that L-tyrosine was involved in biofilm formation(Huijboom et al., 2023 , p. 14579). A hypothesis that can explain why the intensity of this tyrosine band increased in VBNC bacteria is that when bacteria are exposed to saline (NaCl) or oxidative stress (TA, HP) it can lead to the beginning of the metabolic process that could lead to biofilm formation, before bacteria finally enter into the VBNC state or die. D. Carbohydrates-related bands In addition to proteins and peptidoglycan, carbohydrates also play a central role in bacterial structure and stress response. To complete this molecular overview, Raman bands associated with carbohydrate vibrations were analyzed to assess potential modifications in polysaccharide composition during the VBNC transition (Table 1 ). In both short and long exposure conditions, the intensities of the carbohydrate-associated bands at 871, 1053, and 1076 cm⁻¹ were higher in spectra from bacteria belonging to the VBNC clusters compared to control and dead cells. This increase suggests a reorganization of polysaccharide structures or an enhanced synthesis of carbohydrate-rich compounds during the transition to the VBNC state. Carbohydrates are present in cells in multiple locations. In bacteria such as B. cereus , carbohydrates are distributed throughout the cell, forming essential components of the cell wall and serving as important metabolic energy sources inside the cytoplasm. As discussed earlier, previous studies have observed that the cell wall was impacted by the entrance of the cell in VBNC state(del Mar Lleò et al., 2000 ; Signoretto et al., 2001 , 2002 ). Therefore, it is not surprising to see changes in bands related to carbohydrates, particularly for B. cereus which is a Gram-positive bacterium with a thick layer of peptidoglycan. Moreover, many Bacillus species respond to environmental stress by modulating polysaccharide metabolism, especially increasing exopolysaccharide production. This adaptive mechanism enhances environmental persistence by improving biofilm formation, cell aggregation, and protection against desiccation and oxidative stress and can be link to the increase in the tyrosine band intensity at 1206 cm − 1 observed previously(Vardharajula & Z, 2014 ; Yin et al., 2024 ). Finally, results presented in Table 1 indicated that VBNC from B. cereus induced by exposure to disinfectant or saline solution seems to be strongly impacted in their cell wall composition, lipid membrane, proteins and DNA/RNA. 4. Concluding remarks This study provides a comprehensive demonstration of how B. cereus responds to environmental stresses typically encountered in food processing environments, combining molecular biology and Raman spectroscopy to characterize its physiological states. In the first part, molecular analyses confirmed that exposure to conventional disinfectants and saline conditions can induce a VBNC state in environmental B. cereus strains. It is remarkable to observe that the VBNC state, beyond spore formation, represents an additional persistence strategy for this species, which should be considered in food safety assessments. Building on these molecular data, Raman microspectroscopy was then applied to evaluate whether bacterial viability states could be determined based on their spectral fingerprints. The results clearly showed that Raman spectroscopy is capable of distinguishing viable culturable, VBNC, and dead B. cereus cells, even without relying on the C–D bands commonly used for metabolic activity assessment. Moreover, the fingerprint region (700–1800 cm⁻¹) provided robust molecular information that enabled the identification of distinct cluster corresponding to each physiological state (VC, VBNC and dead) using HDBSCAN coupled to PCA. Beyond simple discrimination, spectral analysis revealed consistent molecular trends associated with the VBNC state. Specifically, variations in nucleic acid-related bands indicated changes in DNA conformation or stability, while lipid-associated bands reflected alterations in membrane composition. Differences observed in protein and peptidoglycan related regions, notably the increased intensity of the tyrosine band at 1206 cm⁻¹, suggested adaptive responses linked to oxidative stress and possibly early mechanisms of biofilm formation. Altogether, these biochemical signatures represent potential Raman markers of VBNC induction in B. cereus . These findings demonstrate that Raman spectroscopy can serve as a rapid, non-destructive, and complementary approach to molecular biology for evaluating bacterial viability under industrially relevant stress conditions. By simultaneously providing structural and metabolic information, this method offers new opportunities for monitoring bacterial persistence in food environments and for identifying molecular indicators associated with dormant but potentially resuscitable cells. Declarations Acknowledgements: This work was supported by a the STIMulE facility from Hauts-de-France region. Declaration of competing interests For the purpose of Open Access, the authors have applied a CC BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission. References Authority EFS, European Centre for Disease Prevention and Control (2012) The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2010. EFSA J 10(3):2597. ttps://doi.org/10.2903/j.efsa.2012.2597 Authority EFS, European Centre for Disease Prevention and Control (2013) The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2011. EFSA J 11(4):3129. ttps://doi.org/10.2903/j.efsa.2013.3129 Authority EFS, European Centre for Disease Prevention and Control (2014) The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2012. EFSA J 12(2):3547. ttps://doi.org/10.2903/j.efsa.2014.3547 Authority EFS, European Centre for Disease Prevention and Control (2015) The European Union summary report on trends and sources of zoonoses, zoonotic agents and food-borne outbreaks in 2013. EFSA J 13(1):3991. ttps://doi.org/10.2903/j.efsa.2015.3991 Authority (EFSA), European Centre for Disease Prevention and Control (ECDC) (2024) The European Union One Health 2023 Zoonoses report. EFSA J 22(12):e9106. ttps://doi.org/10.2903/j.efsa.2024.9106 Azan A, Untereiner V, Gobinet C, Sockalingum GD, Breton M, Piot O, Mir LM (2017) Demonstration of the Protein Involvement in Cell Electropermeabilization using Confocal Raman Microspectroscopy. Sci Rep 7:40448. ttps://doi.org/10.1038/srep40448 Balagurusamy R, Gopi L, Kumar DSS, Viswanathan K, Meganathan V, Sathiyamurthy K, Athmanathan B (2024) Significance of Viable But Non-culturable (VBNC) State in Vibrios and Other Pathogenic Bacteria: Induction, Detection and the Role of Resuscitation Promoting Factors (Rpf). Curr Microbiol 81(12):417. ttps://doi.org/10.1007/s00284-024-03947-8 Bao Q, Bo X, Chen L, Ren Y, Wang H, Kwok L-Y, Liu W (2023) Comparative Analysis Using Raman Spectroscopy of the Cellular Constituents of Lacticaseibacillus paracasei Zhang in a Normal and Viable but Nonculturable State. Microorganisms 11:1–14 Brauge T, Bellay M, Midelet G, Soumet C (2025) Viability Detection of Foodborne Bacterial Pathogens in Food Environment by PMA-qPCR and by Microscopy Observation. In A. Bridier (Ed.), Foodborne Bacterial Pathogens: Methods and Protocols (pp. 33–46). Springer US. ttps://doi.org/10.1007/978-1-0716-4100-2_3 Brauge T, Faille C, Inglebert G, Dubois T, Morieux P, Slomianny C, Midelet-Bourdin G (2018) Comparative evaluation of DNA extraction methods for amplification by qPCR of superficial vs intracellular DNA from Bacillus spores. Int J Food Microbiol 266:289–294. ttps://doi.org/10.1016/j.ijfoodmicro.2017.12.012 Brauge T, Faille C, Sadovskaya I, Charbit A, Benezech T, Shen Y, Loessner MJ, Bautista JR, Midelet-Bourdin G (2018) The absence of N-acetylglucosamine in wall teichoic acids of Listeria monocytogenes modifies biofilm architecture and tolerance to rinsing and cleaning procedures. PLoS ONE 13(1):e0190879. ttps://doi.org/10.1371/journal.pone.0190879 Carlin F, Brillard J, Broussolle V, Clavel T, Duport C, Jobin M, Guinebretière MH, Auger S, Sorokine A, Nguyen-Thé C (2010) Adaptation of Bacillus cereus, an ubiquitous worldwide-distributed foodborne pathogen, to a changing environment. Food Res Int 43(7):1885–1894. ttps://doi.org/10.1016/j.foodres.2009.10.024 Carvalho F, Carreaux A, Sartori-Rupp A, Tachon S, Gazi AD, Courtin P, Nicolas P, Dubois-Brissonnet F, Barbotin A, Desgranges E, Bertrand M, Gloux K, Schouler C, Carballido-López R, Chapot-Chartier M-P, Milohanic E, Bierne H, Pagliuso A (2024) Aquatic environment drives the emergence of cell wall-deficient dormant forms in Listeria. Nat Commun 15(1):8499. ttps://doi.org/10.1038/s41467-024-52633-7 Casasola-Rodríguez B, Ruiz-Palacios GM, Pilar R-C, Losano L, Ignacio M-R, de Orta MT (2018) Detection of VBNC Vibrio cholerae by RT-Real Time PCR based on differential gene expression analysis. FEMS Microbiol Lett 365(15):fny156. ttps://doi.org/10.1093/femsle/fny156 Cattani F, Barth VC, Nasário JSR, Ferreira CAS, Oliveira SD (2016) Detection and quantification of viable Bacillus cereus group species in milk by propidium monoazide quantitative real-time PCR. J Dairy Sci 99(4):2617–2624. ttps://doi.org/10.3168/jds.2015-10019 Chen S, Li X, Wang Y, Zeng J, Ye C, Li X, Guo L, Zhang S, Yu X (2018) Induction of Escherichia coli into a VBNC state through chlorination/ chloramination and differences in characteristics of the bacterium between states. Water Res 142:279–288 Choi W, Kim S-S (2020) Outbreaks, Germination, and Inactivation of Bacillus cereus in Food Products: A Review. J Food Prot 83(9):1480–1487. ttps://doi.org/10.4315/0362-028X.JFP-19-429 Coutard F, Crassous P, Droguet M, Gobin E, Colwell RR, Pommepuy M, Hervio-Heath D (2007) Recovery in culture of viable but nonculturable Vibrio parahaemolyticus: Regrowth or resuscitation? ISME J 1:111–120 Cui D, Kong L, Wang Y, Zhu Y, Zhang C (2022) In situ identification of environmental microorganisms with Raman spectroscopy. Environ Sci Ecotechnology 11:100187. ttps://doi.org/10.1016/j.ese.2022.100187 Day AP, Oliver JD (2004) Changes in membrane fatty acid composition during entry of Vibrio vulnificus into the viable but nonculturable state. J Microbiol 42(2):69–73 De Gelder J, De Gussem K, Vandenabeele P, Moens L (2007) Reference database of Raman spectra of biological molecules. J Raman Spectrosc 38(9):1133–1147. ttps://doi.org/10.1002/jrs.1734 del Mar Lleò M, Pierobon S, Tafi MC, Signoretto C, Canepari P (2000) mRNA Detection by Reverse Transcription-PCR for Monitoring Viability over Time in an Enterococcus faecalis Viable but Nonculturable Population Maintained in a Laboratory Microcosm. Appl Environ Microbiol 66(10):4564–4567. ttps://doi.org/10.1128/aem.66.10.4564-4567.2000 den Besten HMW, Mataragas M, Moezelaar R, Abee T, Zwietering MH (2006) Quantification of the Effects of Salt Stress and Physiological State on Thermotolerance of Bacillus cereus ATCC 10987 and ATCC 14579. Appl Environ Microbiol 72(9):5884–5894. ttps://doi.org/10.1128/AEM.00780-06 Dong K, Pan H, Yang D, Rao L, Zhao L, Wang Y, Liao X (2020) Induction, detection, formation, and resuscitation of viable but non-culturable state microorganisms. Compr Rev Food Sci Food Saf 19(1):149–183. ttps://doi.org/10.1111/1541-4337.12513 El-Aziz NKA, Tartor YH, El-Aziz Gharib AA, Ammar AM (2018) Propidium Monoazide Quantitative Real-Time Polymerase Chain Reaction for Enumeration of Some Viable but Nonculturable Foodborne Bacteria in Meat and Meat Products. Foodborne Pathog Dis 15(4):226–234. ttps://doi.org/10.1089/fpd.2017.2356 Feliciano RJ, Boué G, Membré JM (2020) Overview of the potential impacts of climate change on the microbial safety of the dairy industry. Foods 9(12). ttps://doi.org/10.3390/foods9121794 Glasset B, Herbin S, Guillier L, Cadel-Six S, Vignaud M-L, Grout J, Pairaud S, Michel V, Hennekinne J-A, Ramarao N, Brisabois A (2016) Bacillus cereus-induced food-borne outbreaks in France, 2007 to 2014: Epidemiology and genetic characterisation. Eurosurveillance 21(48):30413. ttps://doi.org/10.2807/1560-7917.ES.2016.21.48.30413 Huijboom L, Tempelaars M, Fan M, Zhu Y, Boeren S, van der Linden E, Abee T (2023) L-tyrosine modulates biofilm formation of Bacillus cereus ATCC 14579. Res Microbiol 174(6):104072. ttps://doi.org/10.1016/j.resmic.2023.104072 McEwen GD, Wu Y, Zhou A (2010) Probing Nanostructures of Bacterial Extracellular Polymeric Substances (EPS) versus Culture Time by Raman Microspectroscopy and Atomic Force Microscopy. Biopolymers 93(2):171–177. ttps://doi.org/10.1002/bip.21315 Mols M, de Been M, Zwietering MH, Moezelaar R, Abee T (2007) Metabolic capacity of Bacillus cereus strains ATCC 14579 and ATCC 10987 interlinked with comparative genomics. Environ Microbiol 9(12):2933–2944. ttps://doi.org/10.1111/j.1462-2920.2007.01404.x Morawska LP, Kuipers OP (2022) Transcriptome analysis and prediction of the metabolic state of stress-induced viable but non-culturable Bacillus subtilis cells. Sci Rep 12(1):18015. ttps://doi.org/10.1038/s41598-022-21102-w Pazos-Rojas LA, Muñoz-Arenas LC, Rodríguez-Andrade O, López-Cruz LE, López-Ortega O, Lopes-Olivares F, Luna-Suarez S, Baez A, Morales-García YE, Quintero-Hernández V, Villalobos-López MA, Torre JD la, Muñoz-Rojas J (2019) Desiccation-induced viable but nonculturable state in Pseudomonas putida KT2440, a survival strategy. PLOS ONE , 14 (7), e0219554. ttps://doi.org/10.1371/journal.pone.0219554 Pistiki A, Monecke S, Shen H, Ryabchykov O, Bocklitz TW, Rösch P, Ehricht R, Popp J (2022) Comparison of Different Label-Free Raman Spectroscopy Approaches for the Discrimination of Clinical MRSA and MSSA Isolates. Microbiol Spectr 10(5):e00763–e00722. ttps://doi.org/10.1128/spectrum.00763-22 Qi Z, Huang Z, Liu C (2022) Metabolism differences of biofilm and planktonic Pseudomonas aeruginosa in viable but nonculturable state induced by chlorine stress. Sci Total Environ 821:153374. ttps://doi.org/10.1016/j.scitotenv.2022.153374 Rebuffel V, Baritaux J-C, Bergmann E, Marcoux P, Espagnon I, Morales S (2019) Results on identification of bacteria aging in complex environmental samples using Raman spectroscopy. Novel Biophotonics Techniques Appl V 11075:57–59. ttps://doi.org/10.1117/12.2527141 Rowan NJ (2004) Viable but non-culturable forms of food and waterborne bacteria: Quo Vadis? Trends Food Science Technology 15(9):462–467. ttps://doi.org/10.1016/j.tifs.2004.02.009 Se J, Fu Y, Xie Y, Xu F, Shen C, Nannipieri P (2021) Proteomic changes of viable but nonculturable (VBNC) Escherichia coli O157:H7 induced by low moisture in an artificial soil. Biol Fertil Soils 57(2):219–234. ttps://doi.org/10.1007/s00374-020-01520-6 Setlow P (2006) Spores of Bacillus subtilis: Their resistance to and killing by radiation, heat and chemicals. J Appl Microbiol 101(3):514–525. ttps://doi.org/10.1111/j.1365-2672.2005.02736.x Shi L, Derouiche A, Pandit S, Alazmi M, Ventroux M, Køhler JB, Noirot-Gros M-F, Gao X, Mijakovic I (2024) Connection between protein-tyrosine kinase inhibition and coping with oxidative stress in Bacillus subtilis. Proceedings of the National Academy of Sciences , 121 (25), e2321890121. ttps://doi.org/10.1073/pnas.2321890121 Signoretto C, Lleo MdelM, Canepari P (2001) Modification of the Peptidoglycan of Escherichia coli in the Viable But Nonculturable State. Curr Microbiol 44:125–131 Signoretto C, Lleò MM, Canepari P (2002) Modification of the peptidoglycan of Escherichia coli in the viable but nonculturable state. Curr Microbiol 44(2):125–131. ttps://doi.org/10.1007/s00284-001-0062-0 Soni A, Oey I, Silcock P, Bremer P (2016) Bacillus Spores in the Food Industry: A Review on Resistance and Response to Novel Inactivation Technologies. Compr Rev Food Sci Food Saf 15(6):1139–1148. ttps://doi.org/10.1111/1541-4337.12231 Strola SA, Marcoux PR, Schultz E, Perenon R, Simon A-C, Espagnon I, Allier CP, Dinten J-M (2014) Differentiating the growth phases of single bacteria using Raman spectroscopy. Biomedical Vib Spectrosc VI: Adv Res Ind 8939:16–24. ttps://doi.org/10.1117/12.2041446 Tabassum T, Tabassum T, Ma M, S., Noor R (2020) Transformation of Heat Stressed Non-culturable Bacillus cereus Cells by Extracellular Extracts from Pseudomonas aeruginosa. Asian J Appl Sci 13(4):152–156. ttps://doi.org/10.3923/ajaps.2020.152.156 Trigueros S, Brauge T, Dedole T, Debuiche S, Rebuffel V, Morales S, Marcoux PR, Midelet G (2023) Optimisation and impact ofdeuterium isotope probing on Listeria innocua. PLoS ONE 18:1–11 Valencia-Marin MF, Chávez-Avila S, Guzmán-Guzmán P, Orozco-Mosqueda M del, Santos-Villalobos C, Glick S, B. R., Santoyo G (2024) Survival strategies of Bacillus spp. in saline soils: Key factors to promote plant growth and health. Biotechnology Advances , 70 , 108303. ttps://doi.org/10.1016/j.biotechadv.2023.108303 Vardharajula S, Z AS (2014) Exopolysaccharide Production by Drought Tolerant Bacillus Spp. And Effect on Soil Aggregation. Under Drought Stress 4(1):51–57. ttps://doi.org/10.15414/jmbfs.2014.4.1.51-57 Wagley S, Morcrette H, Kovacs-Simon A, Yang ZR, Power A, Tennant RK, Love J, Murray N, Titball RW, Butler CS (2021) Bacterial dormancy: A subpopulation of viable but non-culturable cells demonstrates better fitness for revival. PLoS Pathog 17(1):e1009194. ttps://doi.org/10.1371/journal.ppat.1009194 Wang Y, Huang WE, Cui L, Wagner M (2016) Single cell stable isotope probing in microbiology using Raman microspectroscopy. Curr Opin Biotechnol 41:34–42. ttps://doi.org/10.1016/j.copbio.2016.04.018 Wong Hc, Wang P (2004) Induction of viable but nonculturable state in Vibrio parahaemolyticus and its susceptibility to environmental stresses. J Appl Microbiol 96(2):359–366. ttps://doi.org/10.1046/j.1365-2672.2004.02166.x Xiong Z, Zhao Q, Zhao M, Liu L, Zeng J, Zhang S, Deng S, Liu D, Zhang X, Xing B (2024) Unveiling the mechanisms of black phosphorus nanosheets-induced viable but non-culturable state in Bacillus tropicus (p. 2024.06.17.599389). bioRxiv. ttps://doi.org/10.1101/2024.06.17.599389 Yin Z, Yuan Y, Zhang R, Gan J, Yu L, Qiu X, Chen R, Wang Q (2024) Understanding Bacillus response to salt stress: Growth inhibition, enhanced EPS secretion, and molecular adaptation mechanisms. Process Biochem 146:412–422. ttps://doi.org/10.1016/j.procbio.2024.09.023 Yoon J-H, Lee S-Y (2022) Characteristics of viable-but-nonculturable Vibrio parahaemolyticus induced by nutrient-deficiency at cold temperature. Crit Rev Food Sci Nutr 60(8):1302–1320. ttps://doi.org/10.1080/10408398.2019.1570076 Zhang S, Guo L, Yang K, Zhang Y, Ye C, Chen S, Yu X, Huang WE, Cui L (2018) Induction of Escherichia coli into a VBNC state by continuous-flow UVC and subsequent changes in metabolic activity at the single-cell level. Front Microbiol 9(SEP):1–11. ttps://doi.org/10.3389/fmicb.2018.02243 Zheng X, Li R, Wang T, Li X, Han X, Dai Y, Liu J, Xu J (2025) Unraveling Antibacterial Mechanisms of Surfactants against Staphylococcus aureus via Single-Cell Raman Spectroscopy. Anal Chem 97(17):9202–9211. ttps://doi.org/10.1021/acs.analchem.4c06380 Tables Table 1 is available in the Supplementary Files section. Additional Declarations The authors declare no competing interests. Supplementary Files Supplementarymaterial.pdf Table11.docx Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9112872","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":605556221,"identity":"7ea35c30-9e18-42cd-9928-6bf4a58f2ac8","order_by":0,"name":"Jonathan Dikec","email":"","orcid":"https://orcid.org/0000-0002-1713-8175","institution":"ANSES, Laboratory for Food Safety, Bacteriology and Parasitology of Fishery and Aquaculture Products Unit, Boulogne sur Mer, France","correspondingAuthor":false,"prefix":"","firstName":"Jonathan","middleName":"","lastName":"Dikec","suffix":""},{"id":605563061,"identity":"7910f4d2-fb68-4ed5-a659-002c538d82c4","order_by":1,"name":"Eglantine Chalivat","email":"","orcid":"https://orcid.org/0009-0005-6736-3687","institution":"ANSES, Laboratory for Food Safety, Bacteriology and Parasitology of Fishery and Aquaculture Products Unit, Boulogne sur Mer, France","correspondingAuthor":false,"prefix":"","firstName":"Eglantine","middleName":"","lastName":"Chalivat","suffix":""},{"id":605563062,"identity":"8a8a18ac-efee-4c65-bb87-cf51cd1d7a05","order_by":2,"name":"Darius L. Abedini","email":"","orcid":"https://orcid.org/0009-0006-6792-9668","institution":"Univ. Grenoble Alpes, CEA, LETI, DTBS, L4IV, 38054 Grenoble, France","correspondingAuthor":false,"prefix":"","firstName":"Darius","middleName":"L.","lastName":"Abedini","suffix":""},{"id":605563063,"identity":"2c7c9ba4-bdd5-4085-9afb-f316e6c6fd0d","order_by":3,"name":"Sabine Debuiche","email":"","orcid":"","institution":"ANSES, Laboratory for Food Safety, Bacteriology and Parasitology of Fishery and Aquaculture Products Unit, Boulogne sur Mer, France","correspondingAuthor":false,"prefix":"","firstName":"Sabine","middleName":"","lastName":"Debuiche","suffix":""},{"id":605563064,"identity":"26700635-f300-454a-97aa-8e4766b7e20f","order_by":4,"name":"Thomas Dubois","email":"","orcid":"https://orcid.org/0000-0003-2712-5877","institution":"Univ. Lille, CNRS, INRAE, ENSCL, UMET, F-59650, Villeneuve d’Ascq, France","correspondingAuthor":false,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Dubois","suffix":""},{"id":605563065,"identity":"522d4c9e-2a04-4c7c-a311-d80dde08ca3a","order_by":5,"name":"Christine Faille","email":"","orcid":"https://orcid.org/0000-0002-2786-1412","institution":"Univ. Lille, CNRS, INRAE, ENSCL, UMET, F-59650, Villeneuve d’Ascq, France","correspondingAuthor":false,"prefix":"","firstName":"Christine","middleName":"","lastName":"Faille","suffix":""},{"id":605563066,"identity":"0a4e54fb-0704-432b-bb1f-21130b60d03e","order_by":6,"name":"Pierre R. Marcoux","email":"","orcid":"https://orcid.org/0000-0003-3855-9662","institution":"Univ. Grenoble Alpes, CEA, LETI, DTBS, L4IV, 38054 Grenoble, France","correspondingAuthor":false,"prefix":"","firstName":"Pierre","middleName":"R.","lastName":"Marcoux","suffix":""},{"id":605563067,"identity":"752632fa-5230-452e-b773-d6f8d0b278b1","order_by":7,"name":"Thomas Brauge","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA8UlEQVRIiWNgGAWjYDACCQiVAKEqGFBpbIAHVcsZBlSasBbGNiK02Es3H93w4w9Dnnn7GeMXH+fZ5cm3tz9gOLgHjy0yx9Ju9rYxFMucyTGznLktudjgzBkDhgPP8Dksx+wGbwND4gyGHDNj3m3MiRskchiYPxzAr+Xmnz9ALfxvzIz/zqlPnD//+QOGAwS03OZhA2qRyDF+zNhwOLHhBoMBfi030tJuy7ZJFEtIPCtj7Dl2PHHDmRyDA/i0sM9IPnbzzR+bPAn+5M0fftRUJ85vP/7wAT4tUACOHTYJGJewBihg/kCsylEwCkbBKBhZAABRvFh+VXaHcgAAAABJRU5ErkJggg==","orcid":"https://orcid.org/0000-0002-7649-3919","institution":"ANSES, Laboratory for Food Safety, Bacteriology and Parasitology of Fishery and Aquaculture Products Unit, Boulogne sur Mer, France","correspondingAuthor":true,"prefix":"","firstName":"Thomas","middleName":"","lastName":"Brauge","suffix":""}],"badges":[],"createdAt":"2026-03-13 09:24:39","currentVersionCode":1,"declarations":{"humanSubjects":false,"vertebrateSubjects":false,"conflictsOfInterestStatement":false,"humanSubjectEthicalGuidelines":false,"humanSubjectConsent":false,"humanSubjectClinicalTrial":false,"humanSubjectCaseReport":false,"vertebrateSubjectEthicalGuidelines":false},"doi":"10.21203/rs.3.rs-9112872/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9112872/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":104682890,"identity":"fdbd4bc4-5aa5-46a6-bdcf-8b924b6dfd50","added_by":"auto","created_at":"2026-03-16 03:16:33","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":480156,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eQuantification of viable culturable (Log(CFU/mL)), viable (Log(GE/mL)) and total (Log(GE/mL)) populations of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eB. cereus\u003c/strong\u003e\u003c/em\u003e treated with water (control), Topax and Topactive disinfectant containing mostly Triamine (TA) and Hydrogen Peroxyde (HP). After 24 h of growth, cells were exposed for 20 min (A) or 24 h (B) to the different stresses. The error bars represent the standard deviation (n = 3).\u003c/p\u003e","description":"","filename":"figure1.png","url":"https://assets-eu.researchsquare.com/files/rs-9112872/v1/37557e2106306baed7805ca2.png"},{"id":104682894,"identity":"770efa85-f5e2-47dd-8029-821503f778bc","added_by":"auto","created_at":"2026-03-16 03:16:33","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":638157,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eMedian spectra of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eB. cereus\u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eGS1\u003c/strong\u003e exposed for 20 min (A) or 24 h (B) to stress conditions. Grey spectrum represents the median spectrum of bacteria exposed to water (control condition), yellow represents the median spectrum of bacteria exposed to NaCl 15 %, green the median spectrum of bacteria exposed to TA 0.025 % and blue the median spectrum of bacteria exposed to HP 0.1 % for 20min or 0.5 % for 24 h.\u003c/p\u003e","description":"","filename":"figure2.png","url":"https://assets-eu.researchsquare.com/files/rs-9112872/v1/8c6368c39580a72215380a59.png"},{"id":104682893,"identity":"f06a54aa-da00-4910-a76b-4bd441408f56","added_by":"auto","created_at":"2026-03-16 03:16:33","extension":"png","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":725224,"visible":true,"origin":"","legend":"\u003cp\u003e\u003cstrong\u003eClustering analysis of the different viability states through PCA and HDBSCAN (ellipses) of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eB. cereus \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eGS1\u003c/strong\u003e exposed to short (20 min) stress (A) and long (24 h) stress (B). The percentage corresponding to each PC is its variance explainability.\u003c/p\u003e","description":"","filename":"figure3.png","url":"https://assets-eu.researchsquare.com/files/rs-9112872/v1/fe7002a4223c312ed25ac7c3.png"},{"id":104784732,"identity":"3e075c92-2c0a-4fe0-bc7c-a88b2a75f5d8","added_by":"auto","created_at":"2026-03-17 08:08:45","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":2872494,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9112872/v1/27a5c0c0-0806-4bb7-aee1-2ac0850edc3a.pdf"},{"id":104782129,"identity":"9b121652-e012-44a4-bb62-798f1d103498","added_by":"auto","created_at":"2026-03-17 07:56:52","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":1283209,"visible":true,"origin":"","legend":"","description":"","filename":"Supplementarymaterial.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9112872/v1/5d9198376f8f6528b2986bcb.pdf"},{"id":104682891,"identity":"3c8d6cc9-3ace-4b09-89ab-c84fc771ff82","added_by":"auto","created_at":"2026-03-16 03:16:33","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":37604,"visible":true,"origin":"","legend":"","description":"","filename":"Table11.docx","url":"https://assets-eu.researchsquare.com/files/rs-9112872/v1/0b47dfc7790ef53c48960bff.docx"}],"financialInterests":"The authors declare no competing interests.","formattedTitle":"\u003cp\u003e\u003cstrong\u003eUsing Raman spectroscopy to discriminate viability states of \u003c/strong\u003e\u003cem\u003e\u003cstrong\u003eBacillus cereus \u003c/strong\u003e\u003c/em\u003e\u003cstrong\u003eexposed to saline or disinfectant stress\u003c/strong\u003e\u003c/p\u003e","fulltext":[{"header":"1. Introduction","content":"\u003cp\u003e \u003cem\u003eBacillus cereus\u003c/em\u003e is a spore forming foodborne pathogen widespread in the environment. This bacteria has been responsible for several outbreaks around the world(Choi \u0026amp; Kim, \u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). In France, \u003cem\u003eB. cereus\u003c/em\u003e was ranked in 2014 as the second cause of foodborne outbreaks, leading to 18 hospitalizations(Glasset et al., \u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). Moreover, according to the European Food Safety Authority (EFSA), this species has been evaluated as an increasing cause of foodborne outbreaks in Europe(Authority (EFSA) \u0026amp; European Centre for Disease Prevention and Control (ECDC), 2024). In France, over the past two decades, the proportion of foodborne outbreaks attributed to \u003cem\u003eB. cereus\u003c/em\u003e has increased significantly (from 2% in 1996 to 23% in 2013 of all reported foodborne outbreaks in the country). This increase is reflected in the number of human cases, which grew from 703 in 2009 to 2,246 in 2013(Authority \u0026amp; European Centre for Disease Prevention and Control, 2012, 2013, 2014, 2015). In parallel, some studies also pointed out that the presence and distribution of \u003cem\u003eB. cereus\u003c/em\u003e in different food products around the world could potentially be impacted by climate change(Carlin et al., \u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e2010\u003c/span\u003e; Feliciano et al., \u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e2020\u003c/span\u003e). One explanation of the prevalence of this bacterium is its ability to sporulate. Bacterial spore is a dormant form of bacteria that can resist to extreme conditions and can sometimes pass-through decontaminations processes if they are not properly performed. Resistances to high temperature, pressure and other decontamination processes has been extensively studied over the past years and is tightly controlled in food processing(Setlow, \u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e2006\u003c/span\u003e; Soni et al., \u003cspan citationid=\"CR42\" class=\"CitationRef\"\u003e2016\u003c/span\u003e). However, in the environment it exists another dormant form called \u0026ldquo;viable but not culturable\u0026rdquo; (VBNC) state. In this state, bacteria can no longer grow, divide, or produce toxins, but they retain reduced metabolic activity and maintain membrane integrity. Like spores, VBNC cells can be reactivated when placed under favorable conditions, resuming growth and toxin production(Balagurusamy et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). However, they are less resistant to extreme environmental conditions than bacterial spores. Unlike spores, VBNC cells cannot be detected using the standard protocols commonly used in the food industry, which rely on bacterial growth on culture media. The VBNC state has been studied far less than the bacterial spores, particularly in \u003cem\u003eB. cereus\u003c/em\u003e. Only a limited number of studies have demonstrated the ability of \u003cem\u003eBacillus\u003c/em\u003e spp. to enter a VBNC state. In particular, VBNC \u003cem\u003eB. cereus\u003c/em\u003e has recently been reported in food matrices such as meats and milk(El-Aziz et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e; Morawska \u0026amp; Kuipers, \u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Rowan, \u003cspan citationid=\"CR36\" class=\"CitationRef\"\u003e2004\u003c/span\u003e; Tabassum et al., \u003cspan citationid=\"CR44\" class=\"CitationRef\"\u003e2020\u003c/span\u003e; Xiong et al., \u003cspan citationid=\"CR51\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). However, the origin of these VBNC cells in food products remains unclear: they may either result from environmental contamination or be induced by stresses encountered within the food industry, such as cleaning and disinfection procedures. Indeed, the VBNC state can be induced by many factors, including cold temperatures or disinfectants, with variable exposure times and concentrations(Balagurusamy et al., \u003cspan citationid=\"CR7\" class=\"CitationRef\"\u003e2024\u003c/span\u003e). Therefore, it is crucial to determine whether the VBNC state of this pathogen can be triggered by stresses typically encountered in the food industry, as this would have important implications for both food safety and public health.\u003c/p\u003e \u003cp\u003eOne approach has been developed and applied to detect VBNC \u003cem\u003eBacillus cereus\u003c/em\u003e. Indeed, the propidium monoazide quantitative polymerase chain reaction (PMA-qPCR) combined with cultural methods has already been employed to assess the presence of VBNC state in meat and milk samples(Cattani et al., \u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e2016\u003c/span\u003e; El-Aziz et al., \u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Nonetheless, these methods remain subject to major limitations. Biomolecular approaches are primarily constrained by their detection threshold, typically ranging from 10\u0026sup3; to 10⁴ cells/mL. Recently, Raman microspectroscopy has been used as a new approach to detect the VBNC state of different foodborne pathogens. Indeed, using this technique allow to work at the scale of the single-cell and potentially avoid cultural phase which is a real limitation to study VBNC cells. Several studies have used the microspectroscopy Raman coupled with deuterium isotopic probing (Raman-DIP), allowing detection at single cell level, the viable cells and cells with less active metabolism identified as VBNC cells(Qi et al., \u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhang et al., \u003cspan citationid=\"CR54\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). With this method, the viable and culturable cells with higher metabolic activity uptake the deuterium, resulting in a Carbone-Deuterium (C-D) band visible on the Raman spectra around 2040\u0026ndash;2300 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Trigueros et al., \u003cspan citationid=\"CR45\" class=\"CitationRef\"\u003e2023\u003c/span\u003e). For VBNC cells, this band shows a reduced intensity. Raman vibrational spectroscopy also has the advantage to identify quickly bacteria using the bands signatures in the \u0026ldquo;fingerprint\u0026rdquo; region between 700 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e and 1800 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e (Rebuffel et al., \u003cspan citationid=\"CR35\" class=\"CitationRef\"\u003e2019\u003c/span\u003e; Strola et al., \u003cspan citationid=\"CR43\" class=\"CitationRef\"\u003e2014\u003c/span\u003e). Because the cellular compounds of VBNC cells are different from vegetative cells, this region should vary and may be used to distinguish the VBNC from viable culturable and dead state. However, no study has attempted to determine whether this spectral region alone is sufficient to distinguish VBNC cells from other states of viability in \u003cem\u003eB. cereus\u003c/em\u003e. The use of this spectral region could have the advantage of avoiding the time-consuming labeling of bacteria with deuterium, in addition to providing an informative spectral signature that would allow for further research beyond simple VBNC detection.\u003c/p\u003e \u003cp\u003eBased on this observation and regarding the state of knowledge on the VBNC state of \u003cem\u003eB. cereus\u003c/em\u003e, the main objective of this study was to evaluate how Raman spectroscopy can allow to discriminate VBNC \u003cem\u003eB. cereus\u003c/em\u003e cells from other viability states. From this general aim, two specific objectives were defined: (i) to determine whether \u003cem\u003eB. cereus\u003c/em\u003e can be induced into a VBNC state under stresses commonly encountered in the food industry, and (ii) to discriminate VBNC cells from other viability states based on their Raman spectra. To achieve these goals, the viability of three environmental strains of \u003cem\u003eB. cereus\u003c/em\u003e were assessed through a combination of molecular approaches (qPCR and PMA-qPCR) and colony-forming unit (CFU) counting after exposure to stress. Three types of stress conditions were applied, each at short (20 min) and long (24 h) durations: saline solution (NaCl) and two widely used disinfectants in the food industry, with triamine or hydrogen peroxide as the main active compounds. One environmental strain was then selected for Raman spectroscopic analysis. The C-D bands region and the fingerprint region have been both used to observe the performance of Raman spectroscopy to discriminate VBNC cells from viable culturable and dead cells.\u003c/p\u003e"},{"header":"2. Materials \u0026 methods","content":"\u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003e2.1 Bacterial strains \u0026amp; culture conditions\u003c/h2\u003e \u003cp\u003eThree environmental strains of \u003cem\u003eB. cereus\u003c/em\u003e (AF16-61-CCGS1, AF16-71-CCPE1, AF16-86-CCCH2) isolated from cooked shrimps, were used in this study and are referred in the manuscript as GS1, PE1 and CH2 respectively. All bacterial strains were stored at -80\u0026deg;C in a tryptone soy broth (TSB,Oxoid, Basingstoke, United Kingdom) supplemented with 20%(v/v) of glycerol. These strains were plated on Trypticase Soy Agar with 0.6% Yeast Extract (TSAYE, Oxoid, Basingstoke, United Kingdom) and incubated for 24 h at 30\u0026deg;C. Several colonies were then picked with a loop and cultured in TSB for 24 h at 30\u0026deg;C under agitation (400 rpm). The bacterial suspension was adjusted to reach a final concentration of 1\u0026sdot;10\u003csup\u003e7\u003c/sup\u003e CFU/mL. A volume of 40 \u0026micro;L of each adjusted suspension was added to 3960 \u0026micro;L stress solution (10\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e dilution) for 20 min or 24h at 30\u0026deg;C under agitation (400 rpm). The stress solution consisted of a 1:1 mixture of TSB and either one of the following: the disinfectant Topax (P3-Topax 990, Ecolab, Arcueil, France) or Topactive (P3-Topactive DES, Ecolab, Arcueil, France) or NaCl at various concentrations. As both Topax and Topactive, contain a mix of active compounds, they are referred to by their main active ingredient: Topax as TA (TriAmine) and Topactive as HP (Hydrogen Peroxide). Control experiments were carried out using sterile water. After incubation, the samples were centrifuged at 5000 g for 10 min. The supernatants were removed and the pellet resuspended in 4 mL of NaCl physiological solution. The suspensions of cells were homogenized by vortexing. These prepared samples were then used for viable culturable bacterial enumeration and quantification of total and viable bacteria. In order to verify the absence of bacterial spores in the samples, part of these suspensions was heated to 80\u0026deg;C for 20 minutes, then cultured on TSAYe medium for 24 hours at 30\u0026deg;C, in parallel with microscopic observation. If no colony developed after this heat treatment and no spores were visible under the microscope, the samples were considered spore-free.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec4\" class=\"Section2\"\u003e \u003ch2\u003e2.2 Viable culturable bacterial enumeration\u003c/h2\u003e \u003cp\u003eFinal suspensions were diluted in NaCl physiological solution until 10\u003csup\u003e\u0026minus;\u0026thinsp;2\u003c/sup\u003e for the stressed samples and until 10\u003csup\u003e\u0026minus;\u0026thinsp;4\u003c/sup\u003e for control samples to have colonies that are sufficiently separated and be able to count them. A volume of 50 \u0026micro;L of from each dilution was plated using a spiral plater (Easyspiral, Interscience, Saint Nom la Breteche, France) on TSAYE (Oxoid, Basingstoke, United Kingdom), then incubated for 24 h at 30\u0026deg;C. Viable culturable bacteria were enumerated using a colony counter (Scan500, Interscience). All cultures and enumerations were performed in triplicate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec5\" class=\"Section2\"\u003e \u003ch2\u003e2.3 Quantification of total, viable bacteria\u003c/h2\u003e \u003cp\u003eTotal DNA and DNA from viable cells were quantified by qPCR and PMA-qPCR, respectively. From the final suspension, 495 \u0026micro;L was mixed with 5 \u0026micro;L of nuclease-free water (Qiagen, Hilden, Germany) for total DNA extraction for standard qPCR. For PMA-qPCR, another 495 \u0026micro;L aliquot was treated with 5 \u0026micro;L of PMA (Propidium monoazide, Biotium, Fremont, USA) to a final concentration of 50 \u0026micro;M. PMA-treated samples were incubated for 5 min at room temperature in the dark, followed by light exposure at 100% for 10 min in an Eppendorf tube using a PhAST Blue lamp (GenIUL, Terrassa, Spain). Cells were then centrifuged at 5,000 g for 10 min at room temperature. The pellet was resuspended in 180 \u0026micro;L of a lysis buffer composed of Tris(hydroxymethyl)aminomethane hydrochloride at 20 mM (TrisHCl, Sigma Aldrich), Ethylenediaminetetraacetic acid (EDTA, Sigma Aldrich) at 2 mM; Triton X-100 (Sigma Aldrich) at 1.2% and lysozyme (Roche, Meylan, France) at 20 mg/mL. DNA extraction for both qPCR and PMA-qPCR, was performed with the DNeasy\u0026reg; Blood \u0026amp; Tissue kits for the Purification of Total DNA from Animal Tissues (Qiagen, Hilden, Germany). Elution was carried out in 100 \u0026micro;L of AE buffer. Extracted DNA was stored at -20\u0026deg;C for up to 5 days.\u003c/p\u003e \u003cp\u003eQuantification of total and viable bacteria by qPCR and PMA-qPCR was performed as described in Brauge et al. (\u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2018\u003c/span\u003e)(Brauge, Faille, Sadovskaya, et al., \u003cspan citationid=\"CR11\" class=\"CitationRef\"\u003e2018\u003c/span\u003e) using the strain \u003cem\u003eB. cereus\u003c/em\u003e CUETEM98/4 as a positive control and \u003cem\u003eB. spizizenii\u003c/em\u003e ATCC 6633 as a negative control. Standard curves were obtained using DNA covering the range from 10\u003csup\u003e1\u003c/sup\u003e to 10\u003csup\u003e9\u003c/sup\u003e genome equivalents (GE) per mL. The GE was calculated on the basis of a standard \u003cem\u003eB. cereus\u003c/em\u003e ATCC 14579 genome of 5.92\u0026sdot;10\u003csup\u003e6\u003c/sup\u003e bp and following the protocol described by Brauge \u003cem\u003eet al.\u003c/em\u003e in 2025(Brauge et al., \u003cspan citationid=\"CR9\" class=\"CitationRef\"\u003e2025\u003c/span\u003e). The quantification limit was around 4.12log(GE/mL). The \u003cem\u003ebclA\u003c/em\u003e gene was used as the target for qPCR amplification. Amplification were carried out in a final volume of 25 \u0026micro;L, containing 2.5 \u0026micro;L of extracted DNA, 12.5 \u0026micro;L of SYBR Premix Ex Taq (2X SYBR qPCR Premix Ex Taq, Takara), 1.5 \u0026micro;M of forward primer C1 (5\u0026prime;-CAT CCG GAC TAG GAC TTC CA-3\u0026prime;) (Eurobio, Les Ulis, France), and 1.5 \u0026micro;M of reverse primer C2 (5\u0026prime;-TTG CCG CAG TAT ATA CGA TAA CA-3\u0026prime;) (Eurobio, Les Ulis, France), as described by Brauge et al. in 2017(Brauge, Faille, Inglebert, et al., \u003cspan citationid=\"CR10\" class=\"CitationRef\"\u003e2018\u003c/span\u003e). Amplification and data acquisition were performed in a LightCycler 480 System (Roche). The amplification protocol was as follows: an initial denaturation at 90\u0026deg;C for 30s, followed by 45 cycles of: 10 seconds at 95\u0026deg;C, 10 s at 55\u0026deg;C, and 10 s at 72\u0026deg;C. Cycle threshold (Cq) values were calculated automatically by the LightCycler 480 software using the second derivative method. Cq value below the quantification threshold were set to zero. All DNA extractions and qPCR assays were performed in triplicate.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003e2.4 Raman microspectroscopic acquisition conditions\u003c/h2\u003e \u003cp\u003eFrom treated and untreated bacterial suspensions, 4 mL have been centrifuged at 5000 g for 10 min. The pellets were resuspended in 1 mL of deuterated nutrient media composed of 25% (v/v) 4x concentrated TSB and 75% (v/v) of D\u003csub\u003e2\u003c/sub\u003eO at 99.9% atom (Sigma Aldrich, Saint Quentin Fallavier, France). The suspensions were incubated for 3 h at 30\u0026deg;C under agitation (400rpm). Following incubation, samples were washed twice by centrifugation at 5000 g for 10 min, and the pellets were resuspended in 1 mL of phosphate-buffered saline (PBS, Merck, Germany). When bacterial concentration was low, the resuspension volume was reduced to concentrate the bacteria. For Raman acquisition, 4 \u0026micro;L of suspension was deposited on a quartz slide and air-dry 15 min at room temperature. Raman spectra were acquired using an XploRA PLUS microspectrometer (HORIBA, Palaiseau, France) and LabSpec 6.3 software (HORIBA, Palaiseau, France). An 8 mW laser beam at l\u0026thinsp;=\u0026thinsp;532 nm was focused through a 100x (0.8 NA) microscope objective (Olympus LMPLFLN). From each experiment around 35 single-celled spectra have been acquired in each tested condition. All experiments have been performed in triplicate with three independent culture to acquire a total number of around 100 single-celled spectra for each condition.\u003c/p\u003e \u003c/div\u003e \u003cdiv id=\"Sec7\" class=\"Section2\"\u003e \u003ch2\u003e2.5 Spectra preprocessing, data \u0026amp; statistical analysis\u003c/h2\u003e \u003cp\u003eAll spectral preprocessing and analysis have been performed using Python (version 3.11). Spectra have been cropped to only the Raman shift between 700 and 3050 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e or between 700 and 1800 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. Then spectra have been preprocessed using RamanSpy library following a despiking using a Whitaker-Hayes algorithm with standard parameters, a Gaussian smoothing, with standard parameters expected for the parameters σ\u0026thinsp;=\u0026thinsp;1, a baseline correction with an asymmetric least squares algorithm (AsLS) with standard parameters excepted for the lam parameters set at 10\u003csup\u003e4\u003c/sup\u003e and the p parameters set at 0.005. Finally, spectra have been normalized by dividing the intensity of each Raman shift by the calculated \u0026lsquo;\u003cem\u003eL2\u003c/em\u003e\u0026rsquo; norm. From that, median spectrum has been plotted for each condition.\u003c/p\u003e \u003cp\u003eTo identify groups of similar Raman spectra, clustering was performed by combining Principal Component Analysis (PCA) on the whole spectra dataset (cropped from 700 to 1800 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) with HDBSCAN (\u003cem\u003eHierarchical Density-Based Spatial Clustering of Applications with Noise\u003c/em\u003e) algorithm on the PCA values. PCA was first applied to reduce the dimension of the Raman spectra and project them onto the two principal components explaining most of the variance. HDBSCAN, from hdbscan library (v0.8.1.), was then used to identify dense groups of similar spectra, using default parameters except for two parameters were the values have been set based on the clustering stability observed with the HDBSCAN condensed tree. The \u003cem\u003emin_sample value\u003c/em\u003e corresponds to a measure of the degree of conservatism required for clustering. We therefore set this value to 13 so that areas with the highest point density are considered clusters, without considering all points outside these areas as noise, as clusters can sometimes have a more widespread structure on the PCA. The results obtained were very similar for \u003cem\u003emin_samples\u003c/em\u003e set between 12 and 20. The \u003cem\u003emin_cluster_size\u003c/em\u003e value corresponds to the smallest group size considered to be a cluster. We therefore set this value to 5 in order to be able to observe small potential clusters that could represent a subpopulation within the sample. As the \u003cem\u003emin_sample\u003c/em\u003e parameters, the stability of the clustering has been tested across different parameters values and the results obtained were very similar for \u003cem\u003emin_cluster_size\u003c/em\u003e set between 2 and 13. For each defined cluster, the corresponding spectra and their PCA values were selected. Ellipses of each cluster defined by HDBSCAN were calculated with the covariance of the selected PCA values. They were then plotted in the reduced space, to represent the spatial dispersion of the cluster. This approach was used to perform a non-subjective and unsupervised clustering. Finally, for each cluster, the spectra of each stress conditions representing at least 25% of the total spectra number inside the cluster were extracted. These extracted spectra allow to compare the specific bands linked to various cellular components between the different clusters.\u003c/p\u003e \u003cp\u003eFor the comparison of the different means bands intensity, a Welch test has been performed with p-value\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"3. Results \u0026 discussion","content":"\u003cdiv id=\"Sec9\" class=\"Section2\"\u003e\n \u003ch2\u003e3.1. Induction of the VBNC state in Bacillus cereus under different stress conditions\u003c/h2\u003e\n \u003cdiv id=\"Sec10\" class=\"Section3\"\u003e\n \u003ch2\u003e3.1.1. Short exposure\u003c/h2\u003e\n \u003cp\u003eThe first objective was to evaluate the VBNC state induction by saline and chemical stresses with two different exposure times for the three environmental strains of \u003cem\u003eB. cereus\u003c/em\u003e. Three strains of \u003cem\u003eB. cereus\u003c/em\u003e have been exposed to disinfectant or saline solution at different concentrations, then the different viability states have been observed using qPCR (total population), PMA-qPCR (viable population) and CFU (viable culturable population) methods (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). The number of VBNC cells corresponds to the difference between the viable population and the culturable state. For the short exposure, all strains maintained high culturable levels under control conditions, around 5.0\u0026ndash;6.0 log(CFU/mL). Exposure to 15% NaCl induced no difference with the control for all three strains, suggesting that saline stress alone caused moderate physiological adaptation without significant lethality (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). The \u003cem\u003eBacillus\u003c/em\u003e genus is indeed well-known for its innate abilities to survive in saline stressful conditions, for instance by producing osmoprotectant and antioxidant compounds(Valencia-Marin et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). In contrast, exposure to disinfectant solutions caused more substantial decreases in viability. For a short exposure to TA at 0.025% a decrease in cultivability of around 3 log(CFU/mL) was observed for the GS1 strain compared to the control. This decrease in cultivability underlines the induction of a VBNC population with a gap of around 2 log between viable and culturable populations. For the two other strains, when bacteria are exposed to 0.025% TA for 20 min, the gap between culturable and viable population did not increase compared to the control. Nevertheless, the strain CH2 was hugely impacted by this stress with a reduction of the viable population of around 2 log (GE/ml) compared to the total population, traducing lethal impact of TA for this strain (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). For the short exposure to HP at 0.1%, the strain GS1 exhibited a marked difference between viable and culturable populations, with around 2 log differences versus only 1 log difference for the control population. In contrast, for the two other strains, the gap between culturable and viable populations were consistent with the control condition, with a difference of 1.5 to 2 log (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA). These results reflect the heterogeneity response from different strains to the same stress. Indeed, previous researches have already demonstrated that some differences can be observed in terms of stress response like exposure to salt, temperature and oxidative stress between \u003cem\u003eB. cereus\u003c/em\u003e strains ATCC 10987 and ATCC 14579(den Besten et al., \u003cspan class=\"CitationRef\"\u003e2006\u003c/span\u003e, p. 14579; Mols et al., \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e). Moreover, these results are consistent with the literature where previous work demonstrated that bacteria from the \u003cem\u003eBacillus\u003c/em\u003e genus can enter in the VBNC state when exposed to different stresses like antibiotics, antibacterial nanoparticles or pulsed electric fields(Morawska \u0026amp; Kuipers, \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e; Rowan, \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e; Xiong et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Here, for the first time, the results demonstrated that VBNC state in \u003cem\u003eB. cereus\u003c/em\u003e can be induced by various types of conventional disinfectants. On the contrary, a short exposure to 15% NaCl solutions no VBNC induction is observed. These results are particularly relevant since the \u003cem\u003eB. cereus\u003c/em\u003e strains used here were recently isolated from a food industry environment. Indeed, recent studies have also reported the presence of VBNC \u003cem\u003eB. cereus\u003c/em\u003e in different types of meats and in milk samples, detected using PMA-qPCR(Cattani et al., \u003cspan class=\"CitationRef\"\u003e2016\u003c/span\u003e; El-Aziz et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). Nevertheless, in those investigations, the time and conditions leading to VBNC induction remained unknown, as the studies focused on the detection and quantification of the pathogen. It is worth noting that the duration and mechanisms of VBNC induction can vary widely among studies, from a few minutes to several months, depending on the nature of the chemical or physical stress applied. Here, we provided the first evidence that \u003cem\u003eB. cereus\u003c/em\u003e strains can rapidly (20 min) enter the VBNC state following exposure to conventional disinfectants commonly used in industrial environments. Nevertheless, short-term exposure (20 min) to saline stress (15% NaCl) did not induce a detectable VBNC state under the tested conditions.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec11\" class=\"Section3\"\u003e\n \u003ch2\u003e3.1.2. Long exposure time\u003c/h2\u003e\n \u003cp\u003eBased on these first results, a much longer exposure time has been tested to determine whether the induction of the VBNC state and the stress responses of the different strains could vary over time. Under prolonged stress exposure, bacterial growth was clearly inhibited, leading to a strong reduction in total cell counts compared with the control condition for all stress treatments (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). For the control condition, the total population was around 9 to 10 log(GE/mL) for the three strains, whereas under stress conditions it ranged from 5 to 7 log(GE/mL). For long exposure to the NaCl solution, results differed from those obtained under short exposure, with induction of VBNC populations showing differences of around 5 and 2.5 log between viable and culturable populations for strains GS1 and PE1, respectively. Both strains also exhibited a total population reduced by 4 log(GE/mL) compared to their respective control. Following exposure to 0.025% TA for 24 h, cells of strain GS1 appeared to be affected by the treatment, with a 3 log(GE/mL) difference in total population compared with the control. Nevertheless, no VBNC induction and mortality were detected because a difference under 1 log is observed between total and viable population and between viable and viable culturable population. For the other two strains, VBNC induction is observed under the same long-exposure condition, with 0.025% TA, with a mortality induction for the CH2 strain (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). After a high oxidative stress, \u003cem\u003ei.e.\u003c/em\u003e 24 h exposure to 0.5% HP, the GS1 strain showed a 3 log (GE/mL) in total population, with complete loss of cultivability and a viable population around 5 log(GE/mL), revealing a full VBNC induction. The other two strains also exhibited a 4 log reduction in total population and a large proportion of VBNC cells, with approximately 3.5 log difference between viable and culturable counts (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB). It should be noted that for the short exposure condition, concentration of disinfectant based on HP have been divided by 5 for short exposure compared to long exposure, shifting from 0.5% to 0.1% (v/v). Indeed, the threshold of PMA-qPCR was not sensitive enough to allow quantification of the viable population when cells were exposed to 0.5% HP for 20 min and cells exposed to 0.1% HP for 24 h were only lightly affected with no VBNC induction. Regarding these results, it appeared that the exposure time played a significant role on VBNC induction, particularly marked by the NaCl 15% exposure where no VBNC population has been observed for short exposure, contrary to the long exposure. Previous investigations have indicated that the transition of bacterial cells into the VBNC state is a progressive process that may extend over several days(Chen et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e; Wong \u0026amp; Wang, \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e). For instance, in the studies cited, all viable \u003cem\u003eEscherichia coli\u003c/em\u003e populations entered the VBNC state within a period ranging from a few hours to five days following exposure to chloramine, depending on the concentration applied. Likewise, \u003cem\u003eVibrio parahaemolyticus\u003c/em\u003e was reported to enter the VBNC state over periods varying from several days to two months, according to the specific treatment conditions imposed on the cells. Therefore, in the case of exposure to 15% NaCl, time exposure needs to be longer than 20 min to induce a quantifiable cell population in VBNC state in the strains AF16-61ccGS1 and AF16-71ccPE1. Moreover, despite different mechanisms of action, all tested stresses appeared to induce the VBNC state in the strains of \u003cem\u003eB. cereus\u003c/em\u003e under certain exposure time.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec12\" class=\"Section3\"\u003e\n \u003ch2\u003e3.1.3 Selection of the strain for spectroscopic studies\u003c/h2\u003e\n \u003cp\u003eFrom all these results, the GS1strain has an interesting profile because it exhibited various susceptibilities to stress. First, for short exposure to NaCl 15%, the three strains seemed unaffected because the bacterial population still has the same viability as the control. For the short exposure to the TA and HP, VBNC state was induced but a viable culturable population remained. In long exposure to NaCl and HP, the totality of the viable culturable population shifted into the VBNC state. For these two conditions, bacteria have been plate on TSAYe media at 30\u0026deg;C to observed if resuscitation process can be observed. After 72h of incubation resuscitation happen with visible colonies on the plates for these two conditions. This last result support the evidence that the gap between VC population and viable population are due to the presence of VBNC cells. Finally, long exposure to TA seemed to lead to a light induction into VBNC state with a large proportion of viable culturable cells. However, the total population is still drastically diminished compared to the control. In summary, strain GS1 exhibited a different mix of viability population depending on the treatment. This is why, for the next phase of this study, the strain GS1, showing multiple viability profiles, has been selected.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e\n\u003cdiv id=\"Sec13\" class=\"Section2\"\u003e\n \u003ch2\u003e3.2. VBNC state investigated through Raman spectroscopy\u003c/h2\u003e\n \u003cp\u003eThe next objective was to assess if Raman spectra can be used to discriminate \u003cem\u003eB. cereus\u003c/em\u003e VBNC of the selected strain AF16-61ccGS1 from the other viability states, using the information contained in C-D bands region and in the fingerprint region.\u003c/p\u003e\n \u003cdiv id=\"Sec14\" class=\"Section3\"\u003e\n \u003ch2\u003e3.2.1. Metabolic activity measurement through Raman DIP for VBNC discrimination\u003c/h2\u003e\n \u003cp\u003eStrain GS1 has been exposed to the different stress solutions described previously and subsequently incubated in deuterated TSB (75% v/v D\u003csub\u003e2\u003c/sub\u003eO) at 30\u0026deg;C for 3 h prior to Raman spectral acquisition. Figure \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e represents the median Raman spectra obtained from around 100 single-cell spectra acquired over three independent cultures. Between 800 and 1800 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, the spectra are dominated by vibrational bands corresponding mainly to amide and DNA regions. A broad C-D Raman band (2040 to 2300 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e), indicative of deuterium incorporation and thus metabolic activity, was observed in the control condition for both short and long incubations. Similarly, a C\u0026ndash;D band comparable to the control was also detected following short exposure to 15% NaCl (v/v). This result reinforces the molecular biology data presented in Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eA, showing that cells exposed briefly to 15% NaCl (v/v) maintained high viability. The presence of a clear C\u0026ndash;D signal in these conditions reflects significant and measurable metabolic activities in both unstressed control and cells briefly exposed to NaCl. In contrast, no C\u0026ndash;D bands were detected under the other stress conditions. Instead, broad, unspecific bands appeared in the same spectral region (2040\u0026ndash;2300 cm⁻\u0026sup1;) across all stressed samples, suggesting that these features are not biological but rather artifacts arising from spectral noise. Because no biomolecular Raman signatures are expected in this region, these artifacts are likely due to the low signal-to-noise ratio of spectra obtained from stressed cells. Indeed, when bacteria are treated to be induced in VBNC state, intensity of the Raman signal decreased a lot compared to control condition and bacteria shortly exposed to NaCl (\u003cstrong\u003eSupplementary data\u003c/strong\u003e Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). The low signal-to-noise ratio (SNR) measured for VBNC bacteria can be attributed to reduced biomasses, observed experimentally through smaller sizes in microscopy (\u003cstrong\u003eSupplementary data\u003c/strong\u003e Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). Similar observations were made for the short TA exposure condition. In this condition, the cells are not supposed to be mostly in the VBNC state but still stressed. Due to their smaller cell biomass, exposure to the laser leads to a reduced Raman signal acquisition compared to larger control cells. This signal reduction makes spectra preprocessing more susceptible to noise, particularly in spectral regions without Raman signal, where the signal-to-noise ratio is less pronounced. In these areas, the baseline correction algorithm (asymmetric least squares) tends to use noise spikes as a relevant band, resulting in the appearance of artifacts. If the bacteria were more metabolically active, like unstressed cells or those briefly exposed to a 15% NaCl concentration, distinct C-D bands should have appeared and thus overwhelmed the background noise. This would have resulted in the absence of artifacts due to the algorithm. This absence of C-D bands for the stressed conditions is consistent with the previous results (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e), showing that in these conditions, viability was impacted by disinfectant or saline stress. In these latter conditions, bacteria were mostly in dead or VBNC state. Some research reported a measurable band in the C-D regions after cell induction to the VBNC state with chlorine stress or UV exposure(Qi et al., \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e; Zhang et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e). However, these studies are different from this current one, because the authors have not worked on \u003cem\u003eB. cereus\u003c/em\u003e and the induction was not done with the same disinfectants or saline exposure. To our knowledge, these artifacts masking the C-D band have never been reported in the literature. Nevertheless, despite these artifacts in this current study C-D bands allow in the end to distinguish the unstressed cells (Control and 20 min saline exposure) from stressed cells.\u003c/p\u003e\n \u003cp\u003eThis result showed the limit of using C-D bands when bacteria are considerably stressed. Nevertheless, in the fingerprint region (700\u0026ndash;1800 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e) where SNR is good, contrary to the C-D band, some differences can be observed between controls conditions and stress conditions. Therefore, this study proposes to investigate deeper this spectra region to see if it possible to discriminate the different viability states of a \u003cem\u003eB. cereus\u003c/em\u003e population based on the observable differences.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec15\" class=\"Section3\"\u003e\n \u003ch2\u003e3.2.2. Viability state clusterization based on fingerprint region\u003c/h2\u003e\n \u003cp\u003eGiven that the fingerprint region (700\u0026ndash;1800 cm⁻\u0026sup1;) provided consistent and interpretable signals free from the artifacts observed in the silent region, subsequent analyses were realized, focusing on this spectral window. This region encompasses vibrational modes of key biomolecules such as proteins, lipids, carbohydrates, and nucleic acids, and thus offers a robust basis for the comparison of cellular physiological states. Recent studies have also analyzed this region to investigate the VBNC induction of \u003cem\u003eLacticaseibacillus paracasei\u003c/em\u003e under cold temperature. Some notable differences have been observed in different parts of the spectra compared with the control condition, sufficient to discriminate between control and VBNC cells (Bao et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e). Here, the objective was to evaluate whether the VBNC cells of \u003cem\u003eB. cereus\u003c/em\u003e were also sufficiently different to discriminate VBNC from control cells, but also to discriminate VBNC from dead cells. Unsupervised classification by a hierarchical density-based spatial clustering of application with noise (HDBSCAN) algorithm was applied to the two first principal component (PC) of the principal component analysis (PCA) was therefore applied to identify potential spectral clusters represented by different ellipses and corresponding to different viability states (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Only the first two PCs were used for clustering because they capture the most variance. Indeed, PC1 represented 41.02% and 38.56% of the variance for short and long exposure condition respectively, while and PC2 represented 10.46% and 12.80% of the variance for short and long exposure condition respectively. Each PC3 to PC5 contributes less than 7% of the variance, and their spectral characteristics were less comprehensive and redundant than those of PC1 and PC2. Moreover, performing PCA in three dimensions did not substantially improve cluster separation. Therefore, using the first two PCs provided a sufficient representation of the data for HDBSCAN clustering (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e). Finally, clustering stability was assessed using the HDBSCAN condensed tree (Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e in the supplementary data). For the short exposure condition, the branch depth of each cluster in the condensed tree supported the evidence that all detected clusters were stable. For the long exposure condition, only three of the four detected clusters were considered, since one cluster, represented by green ellipses in Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e, had a shallow branch, indicating low stability, and was therefore excluded from further analysis and considered as an artifact.\u003c/p\u003e\n \u003cp\u003eThe result obtained for short exposure showed that spectra belonging to unstressed cells (from control conditions, plotted in yellow) and those belonging to bacteria exposed to 15% NaCl (v/v, plotted in grey), are clustered together. This is in accordance with our previous results, where no difference has been observed between control conditions and short exposure to 15% NaCl in terms of population viability (Fig. \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e) and in their median spectra (Fig. \u003cspan class=\"InternalRef\"\u003e2\u003c/span\u003e). On Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA, it can also be observed that bacteria exposed to HP or TA are clustered in two different clusters (plotted in blue and green), separately from the control/15% NaCl cluster. This latter result tends to show a heterogeneity in the VBNC spectra related to their way of induction. Previous studies have also observed heterogeneity in VBNC cells, even for cells induced with the same stress. For example, Bao et al. \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e(Bao et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) highlighted through PCA a remarkable heterogeneity of micro-Raman spectra for \u003cem\u003eL. paracasei\u003c/em\u003e VBNC cells. Nevertheless, their PCA does not contain a reference condition of dead cells unlike in the results presented here in Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA\u0026amp;B. Thus, it is possible that a part of the heterogeneity they observe may be due to the presence of spectra belonging to dead bacteria which is not the case in the results presented Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eA\u0026amp;B. Here, all spectra obtained from bacteria induced in VBNC state by exposure to HP (blue cross) or TA (green triangle) are different from the dead bacteria exposed to very high concentrations of disinfectant (red cross and triangle).\u003c/p\u003e\n \u003cp\u003eFor long exposure to stress, it can be observed that spectra belonging to unstressed control cells were clustered separately from all the other stress conditions (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). Contrary to short exposure, spectra from cells exposed to 15% NaCl for a long period are this time different from control, and some spectra corresponding to saline stress (NaCl, yellow square) tend to be clustered with spectra corresponding to bacteria exposed HP (blue cross) and a small fraction of the bacteria exposed to TA (green triangle). Molecular biology proved that the viable cells exposed to HP and NaCl were identified to be all in VBNC state, in contrast with TA exposure (green plots) where the great majority of the cell population tends to still be viable and culturable, forming a cluster separated from the other conditions (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e \u0026amp; Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). However, this small TA cluster is likely an artifact of the clustering process, as it does not appear to be stable in the HDBSCAN condensed tree and will not be considered for the next of this study (supplementary data Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e).In the qPCR results showed in Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003eB and Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB for cells exposed for a long period to TA, no VBNC induction was considered because there is a difference of less than one log between viable and viable culturable population. However, it can be observed a large standard deviation for the viable culturable population. In consequence, it is possible that a fraction of the viable population can shift into VBNC state but still no be detected through the use of qPCR/CFU method. This difference of observation between results obtained in molecular biology coupled with culturable methods and between micro-Raman results underline the advantage to use micro-Raman to detect VBNC population. All these results tend to show that because there are clustered together, VBNC cells spectra induced by different ways during long exposure tends to share more similarity than tothose induced by short exposure,. Thus, a longer induction period to the VBNC state, seems to homogenize the VBNC population. On the other hand, a large heterogeneity of data was observed among the spectra of bacteria exposed to saline stress over 24 h, where some of these spectra were closer to the control or dead conditions (Fig.\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003eB). This latter result might indicate that even if a part of VBNC spectra tend to be similar, there may remain some subpopulation inside VBNC. Other research also revealed that some heterogeneity can be observed in the VBNC population, even in their morphology. Coutard et al. \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e(Coutard et al., \u003cspan class=\"CitationRef\"\u003e2007\u003c/span\u003e) have observed under scanning electron microscope that \u003cem\u003eV. parahaemolyticus\u003c/em\u003e shifted from rod shape to coccoid form when they enter the VBNC state, with a morphological heterogeneity with tiny and larger coccoid forms. A few years later, it has been observed that larger coccoid forms of \u003cem\u003eV. parahaemolyticus\u003c/em\u003e showed a better fitness for revival and that this form was present in seafood samples(Wagley et al., \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). In this present study, brightfield microscopy observation before Raman acquisition (supplementary Fig.\u0026nbsp;2), outlined a morphological shift, from the rod-shaped cells of control conditions to the coccus-shaped VBNC cells. However, no morphological subpopulation has been observed. Untreated cells or cells shortly exposed to NaCl were rod-shaped but cells treated with HP, TA (short and long exposure), or exposed to NaCl for long exposure showed tiny round forms. Other studies have observed on \u003cem\u003eListeria monocytogenes\u003c/em\u003e and \u003cem\u003eEscherichia coli\u003c/em\u003e that when these cells are in the VBNC state, they shifted to an homogenous round and smaller shape without morphological visible subpopulations(Carvalho et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e; Se et al., \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). For \u003cem\u003eE. coli\u003c/em\u003e, Se and his colleagues(Se et al., \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e) suggested that this morphotype change was related to regulation of protein expression occurring in the VBNC state. In \u003cem\u003eB. cereus\u003c/em\u003e, the VBNC cells generated through different stress conditions shared similar morphological characteristics. The limited variability observed in PCA analysis reflects subtle differences rather than distinct subpopulations. After 24 h of induction, VBNC cells from all treatments converged into a comparable spectral cluster, suggesting a common adaptive physiological state regardless of the initial stress.\u003c/p\u003e\n \u003c/div\u003e\n \u003cdiv id=\"Sec16\" class=\"Section3\"\u003e\n \u003ch2\u003e3.2.3. Assessment of molecular bands associated with bacterial viability states\u003c/h2\u003e\n \u003cp\u003eTo gain deeper insight into the molecular differences associated with the various bacterial viability states, the maximal intensities of Raman bands were compared for each VBNC cluster identified with PCA and HDBSCAN to control and dead cluster, under both short and long exposure times (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). This analysis aimed to identify which biochemical components were most affected during the transition to the VBNC state. Refer to the \u003cstrong\u003esupplementary data Fig.\u0026nbsp;4\u003c/strong\u003e to see the means spectra of each cluster identify Fig. \u003cspan class=\"InternalRef\"\u003e3\u003c/span\u003e.\u003c/p\u003e\n \u003cp\u003e\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eA. Nucleic acid\u0026ndash;related bands\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003ePrevious studies have reported that numerous genes and proteins are differentially regulated in VBNC cells, reflecting major shifts in cellular metabolism and molecular composition(Casasola-Rodr\u0026iacute;guez et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e; Dong et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e; Pazos-Rojas et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). First regarding on DNA/RNA bands, with bands at 780, 1099,1477, 1574, 1609 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, clear differences in band intensity were observed for bacteria belonging to the VBNC cluster, both after short and long stress exposures, compared with control and dead cells (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). As a general observation, both for short and long exposure to stress, bacteria belonging to the VBNC cluster showed a majority of band intensities significantly lower than control bacteria, particularly for the three bands at 780, 1477 and 1574 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e, corresponding to components only related to DNA/RNA. Surprisingly, for long exposure to stress compared to short exposure, the band intensity at 780 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e was significantly higher for bacteria belonging to VBNC compared to control cells at the opposite is observed for band at 1609 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e. This changes in bands intensities between short and long exposure suggested substantial molecular rearrangements under prolonged stress. The previously cited study from Bao and his colleagues(Bao et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) induced \u003cem\u003eL. paracasei\u003c/em\u003e in VBNC state through incubation at 4\u0026deg;C on a De Man, Rogosa et Sharpe (MRS) medium for 120, 180 and 220 days. They have observed that Raman signal intensity changed for VBNC cells depending on the inducing time. They concluded that the entry in VBNC is a gradual process that can take several weeks for \u003cem\u003eL. paracasei\u003c/em\u003e. This could explain what was observed here with an increase in band intensity at 780 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for bacteria exposed longer to stress. For the comparison with the dead cluster, the VBNC cluster shows more contrasting results, with half of the bands having lower intensities and the other half having higher intensities. It is known that in the VBNC state, bacteria are still metabolically active and some genes can be up or down regulated but with a global metabolic activity lower compared to viable culturable cells(Casasola-Rodr\u0026iacute;guez et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e; Dong et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e; Pazos-Rojas et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). Therefore, all these observed results can be explained by this typical lower metabolic activity than control bacteria. This also explains why some bands intensities were higher than the signal collected for dead bacteria, where no metabolic activity remained.\u003c/p\u003e\n \u003cp\u003eSeveral past studies had shown that genes can be up and down regulated in VBNC bacteria(Casasola-Rodr\u0026iacute;guez et al., \u003cspan class=\"CitationRef\"\u003e2018\u003c/span\u003e; Dong et al., \u003cspan class=\"CitationRef\"\u003e2020\u003c/span\u003e; Pazos-Rojas et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e). To our knowledge, no study has investigated the process of gene regulation at different times of induction. In regards to the results in the Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e and those obtained by Bao and his colleagues(Bao et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) where a lot of variation through time can be observed on spectra belonging to VBNC bacteria, it leads to the hypothesis that during an early phase of induction, some genes transcription activities may be more intense than in bacteria in dormancy state for 24 h. Subsequent further investigations are needed to conclude on this point.\u003c/p\u003e\n \u003cp\u003e\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eB. Lipid-related bands\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003eAfter examining nucleic acid\u0026ndash;related bands, attention was next directed towards spectral regions associated with other major biomolecules. Since cell envelope integrity is known to play a key role in bacterial adaptation to stress, lipid- and membrane-related bands were analyzed to assess whether structural modifications accompany the VBNC transition (Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). The results are similar for bands at 871, 1025, 1053 and 1076 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e for short and long exposure with an increase in bands intensities for spectra acquired on bacteria in VBNC state compared both to spectra belonging to control and dead cluster. In the opposite, the bands at 1312 and 1450 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e exhibited a lower intensity in spectra belonging to VBNC cluster compared to control and dead state, both for short and long exposure. All these results traduced major changes in lipid membrane that seemed deeply affected, which is not surprising in regard to the VBNC morphology seen in supplementary Fig.\u0026nbsp;2, that changed drastically compared to the control. Some other publications observed that the cell membrane of bacteria in VBNC state was altered by changes in fatty acids composition in \u003cem\u003eV. vulnificus\u003c/em\u003e, \u003cem\u003eV. parahaemolyticus\u003c/em\u003e and \u003cem\u003ePseudomonas putida\u003c/em\u003e(Day \u0026amp; Oliver, \u003cspan class=\"CitationRef\"\u003e2004\u003c/span\u003e; Pazos-Rojas et al., \u003cspan class=\"CitationRef\"\u003e2019\u003c/span\u003e; Yoon \u0026amp; Lee, \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e). In these publications, the VBNC state has been induced by starvation at cold temperatures and by desiccation, respectively for \u003cem\u003eVibrio\u003c/em\u003e and \u003cem\u003ePseudomonas\u003c/em\u003e. Moreover, the Raman spectra obtained by Bao and his colleagues(Bao et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e) on \u003cem\u003eL. paracasei\u003c/em\u003e and the results presented in the Table \u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e are consistent with this line of evidence, showing that the lipid membrane of bacteria appears to be deeply affected by the entry into the VBNC state, with some bands increasing or decreasing signal intensity.\u003c/p\u003e\n \u003cp\u003e\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eC. Protein-related bands\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003eBeyond lipid alterations, modifications in protein composition and cell wall structure can also provide valuable insights into bacterial adaptation to stress. To explore these aspects, we examined Raman bands corresponding to amide and amino acid vibrations, which reflect changes in proteins and peptidoglycan organization (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). The bands only related to amide and amino acids located at 1002, 1243, 1332, and 1660 cm⁻\u0026sup1; showed significantly reduced intensity for spectra belonging to VBNC cluster compared to control and dead cluster. As observed for DNA/RNA, it is reasonable to think that this may result from the reduced metabolic activity of these dormant bacteria(Balagurusamy et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e; Se et al., \u003cspan class=\"CitationRef\"\u003e2021\u003c/span\u003e). Moreover, previous studies also demonstrated that the cell wall constituted by peptidoglycan with amino acids and amide bonds was strongly affected during the entry into the VBNC state(del Mar Lle\u0026ograve; et al., \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e; Signoretto et al., \u003cspan class=\"CitationRef\"\u003e2001\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e). From these studies, it can be observed that VBNC cells of \u003cem\u003eE. coli\u003c/em\u003e and \u003cem\u003eEnterococcus faecalis\u003c/em\u003e reported changes in peptidoglycan chemistry with increased crosslink and O-acetylation. Interestingly, the band at 1206 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e related to tyrosine(Cui et al., \u003cspan class=\"CitationRef\"\u003e2022\u003c/span\u003e) exhibited an intensity significantly higher in VBNC bacteria compared to the control bacteria for both short and long exposure. This amino acid is known to be affected by response to oxidative stress in bacteria such as \u003cem\u003eB. subtilis\u003c/em\u003e, for which phosphorylation tends to be attenuated compared to eukaryote cells(Shi et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Moreover, a recent study on \u003cem\u003eB. cereus\u003c/em\u003e showed that L-tyrosine was involved in biofilm formation(Huijboom et al., \u003cspan class=\"CitationRef\"\u003e2023\u003c/span\u003e, p. 14579). A hypothesis that can explain why the intensity of this tyrosine band increased in VBNC bacteria is that when bacteria are exposed to saline (NaCl) or oxidative stress (TA, HP) it can lead to the beginning of the metabolic process that could lead to biofilm formation, before bacteria finally enter into the VBNC state or die.\u003c/p\u003e\n \u003cp\u003e\u003cspan type=\"BoldUnderline\" class=\"BoldUnderline\" name=\"Emphasis\"\u003eD. Carbohydrates-related bands\u003c/span\u003e\u003c/p\u003e\n \u003cp\u003eIn addition to proteins and peptidoglycan, carbohydrates also play a central role in bacterial structure and stress response. To complete this molecular overview, Raman bands associated with carbohydrate vibrations were analyzed to assess potential modifications in polysaccharide composition during the VBNC transition (Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e). In both short and long exposure conditions, the intensities of the carbohydrate-associated bands at 871, 1053, and 1076 cm⁻\u0026sup1; were higher in spectra from bacteria belonging to the VBNC clusters compared to control and dead cells. This increase suggests a reorganization of polysaccharide structures or an enhanced synthesis of carbohydrate-rich compounds during the transition to the VBNC state. Carbohydrates are present in cells in multiple locations. In bacteria such as \u003cem\u003eB. cereus\u003c/em\u003e, carbohydrates are distributed throughout the cell, forming essential components of the cell wall and serving as important metabolic energy sources inside the cytoplasm. As discussed earlier, previous studies have observed that the cell wall was impacted by the entrance of the cell in VBNC state(del Mar Lle\u0026ograve; et al., \u003cspan class=\"CitationRef\"\u003e2000\u003c/span\u003e; Signoretto et al., \u003cspan class=\"CitationRef\"\u003e2001\u003c/span\u003e, \u003cspan class=\"CitationRef\"\u003e2002\u003c/span\u003e). Therefore, it is not surprising to see changes in bands related to carbohydrates, particularly for \u003cem\u003eB. cereus\u003c/em\u003e which is a Gram-positive bacterium with a thick layer of peptidoglycan. Moreover, many \u003cem\u003eBacillus\u003c/em\u003e species respond to environmental stress by modulating polysaccharide metabolism, especially increasing exopolysaccharide production. This adaptive mechanism enhances environmental persistence by improving biofilm formation, cell aggregation, and protection against desiccation and oxidative stress and can be link to the increase in the tyrosine band intensity at 1206 cm\u003csup\u003e\u0026minus;\u0026thinsp;1\u003c/sup\u003e observed previously(Vardharajula \u0026amp; Z, \u003cspan class=\"CitationRef\"\u003e2014\u003c/span\u003e; Yin et al., \u003cspan class=\"CitationRef\"\u003e2024\u003c/span\u003e). Finally, results presented in Table\u0026nbsp;\u003cspan class=\"InternalRef\"\u003e1\u003c/span\u003e indicated that VBNC from \u003cem\u003eB. cereus\u003c/em\u003e induced by exposure to disinfectant or saline solution seems to be strongly impacted in their cell wall composition, lipid membrane, proteins and DNA/RNA.\u003c/p\u003e\n \u003c/div\u003e\n\u003c/div\u003e"},{"header":"4. Concluding remarks","content":"\u003cp\u003eThis study provides a comprehensive demonstration of how \u003cem\u003eB. cereus\u003c/em\u003e responds to environmental stresses typically encountered in food processing environments, combining molecular biology and Raman spectroscopy to characterize its physiological states. In the first part, molecular analyses confirmed that exposure to conventional disinfectants and saline conditions can induce a VBNC state in environmental \u003cem\u003eB. cereus\u003c/em\u003e strains. It is remarkable to observe that the VBNC state, beyond spore formation, represents an additional persistence strategy for this species, which should be considered in food safety assessments.\u003c/p\u003e \u003cp\u003eBuilding on these molecular data, Raman microspectroscopy was then applied to evaluate whether bacterial viability states could be determined based on their spectral fingerprints. The results clearly showed that Raman spectroscopy is capable of distinguishing viable culturable, VBNC, and dead \u003cem\u003eB. cereus\u003c/em\u003e cells, even without relying on the C\u0026ndash;D bands commonly used for metabolic activity assessment. Moreover, the fingerprint region (700\u0026ndash;1800 cm⁻\u0026sup1;) provided robust molecular information that enabled the identification of distinct cluster corresponding to each physiological state (VC, VBNC and dead) using HDBSCAN coupled to PCA.\u003c/p\u003e \u003cp\u003eBeyond simple discrimination, spectral analysis revealed consistent molecular trends associated with the VBNC state. Specifically, variations in nucleic acid-related bands indicated changes in DNA conformation or stability, while lipid-associated bands reflected alterations in membrane composition. Differences observed in protein and peptidoglycan related regions, notably the increased intensity of the tyrosine band at 1206 cm⁻\u0026sup1;, suggested adaptive responses linked to oxidative stress and possibly early mechanisms of biofilm formation. Altogether, these biochemical signatures represent potential Raman markers of VBNC induction in \u003cem\u003eB. cereus\u003c/em\u003e.\u003c/p\u003e \u003cp\u003eThese findings demonstrate that Raman spectroscopy can serve as a rapid, non-destructive, and complementary approach to molecular biology for evaluating bacterial viability under industrially relevant stress conditions. By simultaneously providing structural and metabolic information, this method offers new opportunities for monitoring bacterial persistence in food environments and for identifying molecular indicators associated with dormant but potentially resuscitable cells.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAcknowledgements:\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported by a the STIMulE facility from Hauts-de-France region.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eDeclaration of competing interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eFor the purpose of Open Access, the authors have applied a CC BY public copyright licence to any Author Accepted Manuscript (AAM) version arising from this submission.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eAuthority EFS, European Centre for Disease Prevention and Control (2012) The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2010. EFSA J 10(3):2597. ttps://doi.org/10.2903/j.efsa.2012.2597\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuthority EFS, European Centre for Disease Prevention and Control (2013) The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2011. EFSA J 11(4):3129. ttps://doi.org/10.2903/j.efsa.2013.3129\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuthority EFS, European Centre for Disease Prevention and Control (2014) The European Union Summary Report on Trends and Sources of Zoonoses, Zoonotic Agents and Food-borne Outbreaks in 2012. EFSA J 12(2):3547. ttps://doi.org/10.2903/j.efsa.2014.3547\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuthority EFS, European Centre for Disease Prevention and Control (2015) The European Union summary report on trends and sources of zoonoses, zoonotic agents and food-borne outbreaks in 2013. EFSA J 13(1):3991. ttps://doi.org/10.2903/j.efsa.2015.3991\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAuthority (EFSA), European Centre for Disease Prevention and Control (ECDC) (2024) The European Union One Health 2023 Zoonoses report. EFSA J 22(12):e9106. ttps://doi.org/10.2903/j.efsa.2024.9106\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eAzan A, Untereiner V, Gobinet C, Sockalingum GD, Breton M, Piot O, Mir LM (2017) Demonstration of the Protein Involvement in Cell Electropermeabilization using Confocal Raman Microspectroscopy. Sci Rep 7:40448. ttps://doi.org/10.1038/srep40448\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBalagurusamy R, Gopi L, Kumar DSS, Viswanathan K, Meganathan V, Sathiyamurthy K, Athmanathan B (2024) Significance of Viable But Non-culturable (VBNC) State in Vibrios and Other Pathogenic Bacteria: Induction, Detection and the Role of Resuscitation Promoting Factors (Rpf). Curr Microbiol 81(12):417. ttps://doi.org/10.1007/s00284-024-03947-8\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBao Q, Bo X, Chen L, Ren Y, Wang H, Kwok L-Y, Liu W (2023) Comparative Analysis Using Raman Spectroscopy of the Cellular Constituents of Lacticaseibacillus paracasei Zhang in a Normal and Viable but Nonculturable State. Microorganisms 11:1\u0026ndash;14\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrauge T, Bellay M, Midelet G, Soumet C (2025) Viability Detection of Foodborne Bacterial Pathogens in Food Environment by PMA-qPCR and by Microscopy Observation. In A. Bridier (Ed.), \u003cem\u003eFoodborne Bacterial Pathogens: Methods and Protocols\u003c/em\u003e (pp. 33\u0026ndash;46). Springer US. ttps://doi.org/10.1007/978-1-0716-4100-2_3\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrauge T, Faille C, Inglebert G, Dubois T, Morieux P, Slomianny C, Midelet-Bourdin G (2018) Comparative evaluation of DNA extraction methods for amplification by qPCR of superficial vs intracellular DNA from \u003cem\u003eBacillus\u003c/em\u003e spores. Int J Food Microbiol 266:289\u0026ndash;294. ttps://doi.org/10.1016/j.ijfoodmicro.2017.12.012\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBrauge T, Faille C, Sadovskaya I, Charbit A, Benezech T, Shen Y, Loessner MJ, Bautista JR, Midelet-Bourdin G (2018) The absence of N-acetylglucosamine in wall teichoic acids of Listeria monocytogenes modifies biofilm architecture and tolerance to rinsing and cleaning procedures. PLoS ONE 13(1):e0190879. ttps://doi.org/10.1371/journal.pone.0190879\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarlin F, Brillard J, Broussolle V, Clavel T, Duport C, Jobin M, Guinebreti\u0026egrave;re MH, Auger S, Sorokine A, Nguyen-Th\u0026eacute; C (2010) Adaptation of Bacillus cereus, an ubiquitous worldwide-distributed foodborne pathogen, to a changing environment. Food Res Int 43(7):1885\u0026ndash;1894. ttps://doi.org/10.1016/j.foodres.2009.10.024\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCarvalho F, Carreaux A, Sartori-Rupp A, Tachon S, Gazi AD, Courtin P, Nicolas P, Dubois-Brissonnet F, Barbotin A, Desgranges E, Bertrand M, Gloux K, Schouler C, Carballido-L\u0026oacute;pez R, Chapot-Chartier M-P, Milohanic E, Bierne H, Pagliuso A (2024) Aquatic environment drives the emergence of cell wall-deficient dormant forms in Listeria. Nat Commun 15(1):8499. ttps://doi.org/10.1038/s41467-024-52633-7\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCasasola-Rodr\u0026iacute;guez B, Ruiz-Palacios GM, Pilar R-C, Losano L, Ignacio M-R, de Orta MT (2018) Detection of VBNC Vibrio cholerae by RT-Real Time PCR based on differential gene expression analysis. FEMS Microbiol Lett 365(15):fny156. ttps://doi.org/10.1093/femsle/fny156\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCattani F, Barth VC, Nas\u0026aacute;rio JSR, Ferreira CAS, Oliveira SD (2016) Detection and quantification of viable \u003cem\u003eBacillus cereus\u003c/em\u003e group species in milk by propidium monoazide quantitative real-time PCR. J Dairy Sci 99(4):2617\u0026ndash;2624. ttps://doi.org/10.3168/jds.2015-10019\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChen S, Li X, Wang Y, Zeng J, Ye C, Li X, Guo L, Zhang S, Yu X (2018) Induction of Escherichia coli into a VBNC state through chlorination/ chloramination and differences in characteristics of the bacterium between states. Water Res 142:279\u0026ndash;288\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eChoi W, Kim S-S (2020) Outbreaks, Germination, and Inactivation of \u003cem\u003eBacillus cereus\u003c/em\u003e in Food Products: A Review. J Food Prot 83(9):1480\u0026ndash;1487. ttps://doi.org/10.4315/0362-028X.JFP-19-429\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCoutard F, Crassous P, Droguet M, Gobin E, Colwell RR, Pommepuy M, Hervio-Heath D (2007) Recovery in culture of viable but nonculturable Vibrio parahaemolyticus: Regrowth or resuscitation? ISME J 1:111\u0026ndash;120\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCui D, Kong L, Wang Y, Zhu Y, Zhang C (2022) In situ identification of environmental microorganisms with Raman spectroscopy. Environ Sci Ecotechnology 11:100187. ttps://doi.org/10.1016/j.ese.2022.100187\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDay AP, Oliver JD (2004) Changes in membrane fatty acid composition during entry of Vibrio vulnificus into the viable but nonculturable state. J Microbiol 42(2):69\u0026ndash;73\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDe Gelder J, De Gussem K, Vandenabeele P, Moens L (2007) Reference database of Raman spectra of biological molecules. J Raman Spectrosc 38(9):1133\u0026ndash;1147. ttps://doi.org/10.1002/jrs.1734\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003edel Mar Lle\u0026ograve; M, Pierobon S, Tafi MC, Signoretto C, Canepari P (2000) mRNA Detection by Reverse Transcription-PCR for Monitoring Viability over Time in an Enterococcus faecalis Viable but Nonculturable Population Maintained in a Laboratory Microcosm. Appl Environ Microbiol 66(10):4564\u0026ndash;4567. ttps://doi.org/10.1128/aem.66.10.4564-4567.2000\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eden Besten HMW, Mataragas M, Moezelaar R, Abee T, Zwietering MH (2006) Quantification of the Effects of Salt Stress and Physiological State on Thermotolerance of Bacillus cereus ATCC 10987 and ATCC 14579. Appl Environ Microbiol 72(9):5884\u0026ndash;5894. ttps://doi.org/10.1128/AEM.00780-06\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDong K, Pan H, Yang D, Rao L, Zhao L, Wang Y, Liao X (2020) Induction, detection, formation, and resuscitation of viable but non-culturable state microorganisms. Compr Rev Food Sci Food Saf 19(1):149\u0026ndash;183. ttps://doi.org/10.1111/1541-4337.12513\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eEl-Aziz NKA, Tartor YH, El-Aziz Gharib AA, Ammar AM (2018) Propidium Monoazide Quantitative Real-Time Polymerase Chain Reaction for Enumeration of Some Viable but Nonculturable Foodborne Bacteria in Meat and Meat Products. Foodborne Pathog Dis 15(4):226\u0026ndash;234. ttps://doi.org/10.1089/fpd.2017.2356\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eFeliciano RJ, Bou\u0026eacute; G, Membr\u0026eacute; JM (2020) Overview of the potential impacts of climate change on the microbial safety of the dairy industry. Foods 9(12). ttps://doi.org/10.3390/foods9121794\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eGlasset B, Herbin S, Guillier L, Cadel-Six S, Vignaud M-L, Grout J, Pairaud S, Michel V, Hennekinne J-A, Ramarao N, Brisabois A (2016) Bacillus cereus-induced food-borne outbreaks in France, 2007 to 2014: Epidemiology and genetic characterisation. Eurosurveillance 21(48):30413. ttps://doi.org/10.2807/1560-7917.ES.2016.21.48.30413\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHuijboom L, Tempelaars M, Fan M, Zhu Y, Boeren S, van der Linden E, Abee T (2023) L-tyrosine modulates biofilm formation of \u003cem\u003eBacillus cereus\u003c/em\u003e ATCC 14579. Res Microbiol 174(6):104072. ttps://doi.org/10.1016/j.resmic.2023.104072\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMcEwen GD, Wu Y, Zhou A (2010) Probing Nanostructures of Bacterial Extracellular Polymeric Substances (EPS) versus Culture Time by Raman Microspectroscopy and Atomic Force Microscopy. Biopolymers 93(2):171\u0026ndash;177. ttps://doi.org/10.1002/bip.21315\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMols M, de Been M, Zwietering MH, Moezelaar R, Abee T (2007) Metabolic capacity of \u003cem\u003eBacillus cereus\u003c/em\u003e strains ATCC 14579 and ATCC 10987 interlinked with comparative genomics. Environ Microbiol 9(12):2933\u0026ndash;2944. ttps://doi.org/10.1111/j.1462-2920.2007.01404.x\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eMorawska LP, Kuipers OP (2022) Transcriptome analysis and prediction of the metabolic state of stress-induced viable but non-culturable Bacillus subtilis cells. Sci Rep 12(1):18015. ttps://doi.org/10.1038/s41598-022-21102-w\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePazos-Rojas LA, Mu\u0026ntilde;oz-Arenas LC, Rodr\u0026iacute;guez-Andrade O, L\u0026oacute;pez-Cruz LE, L\u0026oacute;pez-Ortega O, Lopes-Olivares F, Luna-Suarez S, Baez A, Morales-Garc\u0026iacute;a YE, Quintero-Hern\u0026aacute;ndez V, Villalobos-L\u0026oacute;pez MA, Torre JD la, Mu\u0026ntilde;oz-Rojas J (2019) Desiccation-induced viable but nonculturable state in Pseudomonas putida KT2440, a survival strategy. \u003cem\u003ePLOS ONE\u003c/em\u003e, \u003cem\u003e14\u003c/em\u003e(7), e0219554. ttps://doi.org/10.1371/journal.pone.0219554\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003ePistiki A, Monecke S, Shen H, Ryabchykov O, Bocklitz TW, R\u0026ouml;sch P, Ehricht R, Popp J (2022) Comparison of Different Label-Free Raman Spectroscopy Approaches for the Discrimination of Clinical MRSA and MSSA Isolates. Microbiol Spectr 10(5):e00763\u0026ndash;e00722. ttps://doi.org/10.1128/spectrum.00763-22\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQi Z, Huang Z, Liu C (2022) Metabolism differences of biofilm and planktonic Pseudomonas aeruginosa in viable but nonculturable state induced by chlorine stress. Sci Total Environ 821:153374. ttps://doi.org/10.1016/j.scitotenv.2022.153374\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRebuffel V, Baritaux J-C, Bergmann E, Marcoux P, Espagnon I, Morales S (2019) Results on identification of bacteria aging in complex environmental samples using Raman spectroscopy. Novel Biophotonics Techniques Appl V 11075:57\u0026ndash;59. ttps://doi.org/10.1117/12.2527141\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eRowan NJ (2004) Viable but non-culturable forms of food and waterborne bacteria: Quo Vadis? Trends Food Science Technology 15(9):462\u0026ndash;467. ttps://doi.org/10.1016/j.tifs.2004.02.009\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSe J, Fu Y, Xie Y, Xu F, Shen C, Nannipieri P (2021) Proteomic changes of viable but nonculturable (VBNC) Escherichia coli O157:H7 induced by low moisture in an artificial soil. Biol Fertil Soils 57(2):219\u0026ndash;234. ttps://doi.org/10.1007/s00374-020-01520-6\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSetlow P (2006) Spores of Bacillus subtilis: Their resistance to and killing by radiation, heat and chemicals. J Appl Microbiol 101(3):514\u0026ndash;525. ttps://doi.org/10.1111/j.1365-2672.2005.02736.x\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eShi L, Derouiche A, Pandit S, Alazmi M, Ventroux M, K\u0026oslash;hler JB, Noirot-Gros M-F, Gao X, Mijakovic I (2024) Connection between protein-tyrosine kinase inhibition and coping with oxidative stress in Bacillus subtilis. \u003cem\u003eProceedings of the National Academy of Sciences\u003c/em\u003e, \u003cem\u003e121\u003c/em\u003e(25), e2321890121. ttps://doi.org/10.1073/pnas.2321890121\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSignoretto C, Lleo MdelM, Canepari P (2001) Modification of the Peptidoglycan of Escherichia coli in the Viable But Nonculturable State. Curr Microbiol 44:125\u0026ndash;131\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSignoretto C, Lle\u0026ograve; MM, Canepari P (2002) Modification of the peptidoglycan of Escherichia coli in the viable but nonculturable state. Curr Microbiol 44(2):125\u0026ndash;131. ttps://doi.org/10.1007/s00284-001-0062-0\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoni A, Oey I, Silcock P, Bremer P (2016) Bacillus Spores in the Food Industry: A Review on Resistance and Response to Novel Inactivation Technologies. Compr Rev Food Sci Food Saf 15(6):1139\u0026ndash;1148. ttps://doi.org/10.1111/1541-4337.12231\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eStrola SA, Marcoux PR, Schultz E, Perenon R, Simon A-C, Espagnon I, Allier CP, Dinten J-M (2014) Differentiating the growth phases of single bacteria using Raman spectroscopy. Biomedical Vib Spectrosc VI: Adv Res Ind 8939:16\u0026ndash;24. ttps://doi.org/10.1117/12.2041446\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTabassum T, Tabassum T, Ma M, S., Noor R (2020) Transformation of Heat Stressed Non-culturable Bacillus cereus Cells by Extracellular Extracts from Pseudomonas aeruginosa. Asian J Appl Sci 13(4):152\u0026ndash;156. ttps://doi.org/10.3923/ajaps.2020.152.156\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTrigueros S, Brauge T, Dedole T, Debuiche S, Rebuffel V, Morales S, Marcoux PR, Midelet G (2023) Optimisation and impact ofdeuterium isotope probing on Listeria innocua. PLoS ONE 18:1\u0026ndash;11\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eValencia-Marin MF, Ch\u0026aacute;vez-Avila S, Guzm\u0026aacute;n-Guzm\u0026aacute;n P, Orozco-Mosqueda M del, Santos-Villalobos C, Glick S, B. R., Santoyo G (2024) Survival strategies of \u003cem\u003eBacillus\u003c/em\u003e spp. in saline soils: Key factors to promote plant growth and health. \u003cem\u003eBiotechnology Advances\u003c/em\u003e, \u003cem\u003e70\u003c/em\u003e, 108303. ttps://doi.org/10.1016/j.biotechadv.2023.108303\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVardharajula S, Z AS (2014) Exopolysaccharide Production by Drought Tolerant Bacillus Spp. And Effect on Soil Aggregation. Under Drought Stress 4(1):51\u0026ndash;57. ttps://doi.org/10.15414/jmbfs.2014.4.1.51-57\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWagley S, Morcrette H, Kovacs-Simon A, Yang ZR, Power A, Tennant RK, Love J, Murray N, Titball RW, Butler CS (2021) Bacterial dormancy: A subpopulation of viable but non-culturable cells demonstrates better fitness for revival. PLoS Pathog 17(1):e1009194. ttps://doi.org/10.1371/journal.ppat.1009194\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWang Y, Huang WE, Cui L, Wagner M (2016) Single cell stable isotope probing in microbiology using Raman microspectroscopy. Curr Opin Biotechnol 41:34\u0026ndash;42. ttps://doi.org/10.1016/j.copbio.2016.04.018\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eWong Hc, Wang P (2004) Induction of viable but nonculturable state in Vibrio parahaemolyticus and its susceptibility to environmental stresses. J Appl Microbiol 96(2):359\u0026ndash;366. ttps://doi.org/10.1046/j.1365-2672.2004.02166.x\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eXiong Z, Zhao Q, Zhao M, Liu L, Zeng J, Zhang S, Deng S, Liu D, Zhang X, Xing B (2024) \u003cem\u003eUnveiling the mechanisms of black phosphorus nanosheets-induced viable but non-culturable state in Bacillus tropicus\u003c/em\u003e (p. 2024.06.17.599389). bioRxiv. ttps://doi.org/10.1101/2024.06.17.599389\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYin Z, Yuan Y, Zhang R, Gan J, Yu L, Qiu X, Chen R, Wang Q (2024) Understanding \u003cem\u003eBacillus\u003c/em\u003e response to salt stress: Growth inhibition, enhanced EPS secretion, and molecular adaptation mechanisms. Process Biochem 146:412\u0026ndash;422. ttps://doi.org/10.1016/j.procbio.2024.09.023\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eYoon J-H, Lee S-Y (2022) Characteristics of viable-but-nonculturable Vibrio parahaemolyticus induced by nutrient-deficiency at cold temperature. Crit Rev Food Sci Nutr 60(8):1302\u0026ndash;1320. ttps://doi.org/10.1080/10408398.2019.1570076\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhang S, Guo L, Yang K, Zhang Y, Ye C, Chen S, Yu X, Huang WE, Cui L (2018) Induction of Escherichia coli into a VBNC state by continuous-flow UVC and subsequent changes in metabolic activity at the single-cell level. Front Microbiol 9(SEP):1\u0026ndash;11. ttps://doi.org/10.3389/fmicb.2018.02243\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZheng X, Li R, Wang T, Li X, Han X, Dai Y, Liu J, Xu J (2025) Unraveling Antibacterial Mechanisms of Surfactants against Staphylococcus aureus via Single-Cell Raman Spectroscopy. Anal Chem 97(17):9202\u0026ndash;9211. ttps://doi.org/10.1021/acs.analchem.4c06380\u003c/span\u003e\u003c/li\u003e\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1 is available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":true,"hideJournal":true,"highlight":"","institution":"Agence nationale de sécurité sanitaire de l'alimentation, de l'environnement et du travail","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":true,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"Food security, Raman spectroscopy, Bacillus cereus, viable but non-culturable (VBNC), viability state","lastPublishedDoi":"10.21203/rs.3.rs-9112872/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9112872/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e \u003cem\u003eBacillus cereus\u003c/em\u003e is a foodborne pathogen able to enter in a viable but non-culturable (VBNC) state, in which cells remain metabolically active but escape detection by conventional culture-based methods used in the food industry. Raman microspectroscopy has emerged as a promising tool for VBNC detection due to its high sensitivity and single-cell resolution. This study evaluated the ability of Raman spectroscopy to discriminate VBNC \u003cem\u003eB. cereus\u003c/em\u003e cells from other viability states. Specifically, we investigated (i) whether stressful conditions representative of food-processing environments can induce the VBNC state in \u003cem\u003eB. cereus\u003c/em\u003e, and (ii) whether Raman spectral profiles allow differentiation among viability states. Three environmental \u003cem\u003eB. cereus\u003c/em\u003e strains were exposed to saline solution and two commonly used food-industry disinfectants for 20 min or 24 h to induce stress. Results demonstrated that such conditions can induce a VBNC state in \u003cem\u003eB. cereus\u003c/em\u003e. One strain was further labeled with deuterium, and Raman spectra were collected. Analyses focused on the C\u0026ndash;D band and the fingerprint regions. The C\u0026ndash;D region enabled discrimination between unstressed and stressed cells, while clustering analysis of the fingerprint region successfully separated unstressed, stressed/injured, VBNC, and dead cells. Mean spectra of each cluster revealed that VBNC cells exhibited marked changes in bands associated with DNA, proteins, cell wall components, and lipid membranes. Overall, this study demonstrates that Raman microspectroscopy, particularly fingerprint region analysis, provides a rapid and non-destructive approach to reliably distinguish VBNC \u003cem\u003eB. cereus\u003c/em\u003e cells from other viability states, highlighting its potential for detecting dormant bacteria in food environments.\u003c/p\u003e","manuscriptTitle":"Using Raman spectroscopy to discriminate viability states of Bacillus cereus exposed to saline or disinfectant stress","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-03-16 03:16:28","doi":"10.21203/rs.3.rs-9112872/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"f6c5a33c-8149-490b-b33f-55c6f2d8b420","owner":[],"postedDate":"March 16th, 2026","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[{"id":64446138,"name":"Applied \u0026 Industrial Microbiology"},{"id":64446139,"name":"Food Science \u0026 Technology"},{"id":64446140,"name":"Spectroscopy"}],"tags":[],"updatedAt":"2026-03-16T03:16:28+00:00","versionOfRecord":[],"versionCreatedAt":"2026-03-16 03:16:28","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9112872","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9112872","identity":"rs-9112872","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}