Ultrasound-assisted blue light killing Vibrio parahaemolyticus to improve salmon preservation

Highlights

• •The bactericidal effect was positively correlated with the blue light dose.
• •Blue light damaged the cell membrane and caused a reactive oxygen species burst.
• •RNA-seq analysis was used to characterize the changes of metabolic pathways.
• •Ultrasonic field for 15 min assisted by blue light sterilization is more effective.
• •Blue light or ultrasonic field treatment maintained the quality of salmon.

Abstract

Vibrio parahaemolyticus is a typical marine bacterium, which often contaminates seafood and poses a health risk to consumers. Some non-thermal sterilization technologies, such as ultrasonic field (UF) and blue light (BL) irradiation, have been widely used in clinical practice due to their efficiency, safety, and avoidance of drug resistance, but their application in food preservation has not been extensively studied. This study aims to investigate the effect of BL on V. parahaemolyticus in culture media and in ready-to-eat fresh salmon, and to evaluate the killing effectiveness of the UF combined with BL treatment on V. parahaemolyticus. The results showed that BL irradiation at 216 J/cm² was effective in causing cell death (close to 100%), cell shrinkage and reactive oxygen species (ROS) burst in V. parahaemolyticus. Application of imidazole (IMZ), an inhibitor of ROS generation, attenuated the cell death induced by BL, indicating that ROS were involved in the bactericidal effects of BL on V. parahaemolyticus. Furthermore, UF for 15 min enhanced the bactericidal effect of BL at 216 J/cm² on V. parahaemolyticus, with the bactericidal rate of 98.81%. In addition, BL sterilization did not affect the color and quality of salmon, and the additive UF treatment for 15 min did not significant impact on the color of salmon. These results suggest that BL or UF combined with BL treatment has potential for salmon preservation, however, it is crucial to strictly control the intensity of BL and the duration of UF treatment to prevent reducing the freshness and brightness of salmon.

1. Introduction

Vibrio parahaemolyticus, one of the primary pathogens, is a gram-negative bacterium widely distributed in estuarine and marine environments [1]. Consumption of contaminated, raw or undercooked seafood may lead to acute gastroenteritis and even death [2]. Antibiotics have been traditionally used to treat infection with this bacterium. However, V. parahaemolyticus has high levels of resistance to certain antibiotics, posing a serious health hazard [3].

Non-thermal sterilization, rather than antibiotics, is commonly used to eliminate pathogenic microorganisms in food. Fleet [4] attempted to decontaminate mussels with a combination of chlorine, ultraviolet light and ozonated iodine, but had little effect on V. parahaemolyticus. Croci, Suffredini, Cozzi and Toti [5] used ozone-oxidized water purify mussels contaminated with Escherichia coli, V. cholera, and V. parahaemolyticus, resulting in only a decrease by 1 log in the count of V. cholerae and V. parahaemolyticus.

The bactericidal effect of electrolytic oxidation of water on V. parahaemolyticus in oysters was also limited, as treatment for more than 8 h resulted in oyster mortality [6]. Treatment with 300 MPa ultrahigh-pressure (UHP) for 180 s significantly eliminated V. parahaemolyticus (decreased by 5 log), but damaged the appearance of food materials and was costly [7], [8], [9]. Low doses of radiation (<3 kGy) effectively killed V. parahaemolyticus in oysters without affecting the organoleptic quality of the oyster [10]. However, we must take into account the current low consumer acceptance of irradiated foods and the inconvenience caused by the management of radioactive substances. Therefore, proper approaches to effectively kill pathogens and assure the quality of aquatic foods are needed to reduce the risk of foodborne illness caused by V. parahaemolyticus.

Both Gram-positive and Gram-negative bacteria can perceive blue light at 405–470 nm to respond physiologically [11]. Blue light has antibacterial or bactericidal abilities with little to no harmful effects on mammalian cells [12]. In an in vitro study, blue light exposure killed 15.7% of the tested Propionibacterium acnes immediately and 24.4% after 60 min [13]. Another research showed blue light irradiation successfully increased the cure rate (by 91%) in patients infected with antibiotic-resistant H. pylori [14]. It should be noted that blue light shows different bactericidal effects at different wavelengths. Blue light at 405 nm killed 99.9% of Helicobacter pylori, and effectively inhibited the growth of Porphyromonas gingivalis [15], [16]. However, blue light at wavelength above 430 nm was unable to inhibit P. gingivalis. In addition, Chen, Huang, Liu, Liu, Zhao and Wang [17] found that blue light bound V. parahaemolyticus curcumin and its sequestered biofilm through mediated photodynamic inactivation (PDI) and prolonged the storage quality of cooked oysters. Blue light showed a wide range of antibacterial activity, but whether it can be used alone to control V. parahaemolyticus in aquatic products and the associated mechanisms of action need further investigation. In addition, the application of ultrasonic sterilization is also a promising method that can inactivate microorganisms without harmful effects on the nutrition, quality and organoleptics of food products, and is often applied to the sterilization of fruit and vegetable juices [18], [19]. Therefore, in the present work ultrasound sterilization has been investigated as a complementary method to blue light and, to our knowledge, is the first validation of a sterilization strategy for V. parahaemolyticus on salmon.

In this study, the antibacterial activity of blue light irradiation against V. parahaemolyticus was evaluated and insights into the bactericidal mechanism were provided. In addition, the ability of blue light radiation as well as ultrasonic field treatment on ready-to-eat fresh salmon contaminated with V. parahaemolyticus were assessed, including its bactericidal activity and potential impact on salmon quality.

2. Materials and methods

2.1. Bacterial strain and culture condition

Vibrio parahaemolyticus RIMD 2210633 was obtained from Chinese Centre for Disease Control and Prevention (Beijing, China). This strain from frozen stocks at −80 °C grew at 37 °C for 16 h on Voges Proskauer (VP) agar (1% (w/v) tryptone, 3.5% (w/v) NaCl, 1% (w/v) fish peptone, 0.2% (w/v) yeast extract, and 15% (w/v) agar, adjust pH value of 8, Sinopharm Chemical Reagent, Beijing, China), and stored at 4 °C.

Propagation of this strain was performed by picking a single colony from VP agar into VP liquid medium (1% (w/v) tryptone, 3.5% (w/v) NaCl, 1% (w/v) fish peptone, and 0.2% (w/v) yeast extract), and grew 8–24 h at 37 °C. Then, the strain was transferred to VP liquid medium at the ratio of 1:200 for 2–3 h to complete the preparation of bacterial suspension. When the optical density at 600 nm (OD600) reached to 0.8–1.0, the suspension was diluted by VP liquid medium to OD600 of 0.1.

2.2. Blu-ray emitting diodes (LEDs) system

The luminance (Lux) was measured with a digital light meter (Lutron-LX-101A, Lutron electronic enterprise Co., Ltd, Taiwan, China) at the blue light wavelength of 460–475 nm. The irradiance (W/cm²) of light was calculated using the photometric conversion, defined in Eq. (1).

$$P=L\times {\left[\mathit{Km}\times V\left(\lambda \right)\right]}^{-1}$$ (1)

where P is the irradiance (W/cm²), L is the luminance (lux), Km is the maximum value of spectral luminous efficiency, and V (λ) is the photopic spectral function at a wavelength of 470 nm.

Under 10 W, 20 W, and 30 W blue light irradiance for 120 min, the energy per unit area (fluence) applied in each experiment were 72 J/cm², 144 J/cm², and 216 J/cm², respectively, which was calculated by Eq. (2) [20].

$$E=P\times t$$ (2)

where E is the absolute dose (energy density) (J/cm²), P is the irradiance (power density) (W/cm²), and t is the exposure time (in seconds).

2.3. Ultrasonic field (UF) treatment on salmon

The method of UF treatment was modified by Li, Ming, Liu, Xu, Xu, Hu, Mo and Zhou [21]. Salmon samples with the size of 2 cm × 2 cm × 0.5 cm (approximately 5–8 g of each piece) were placed in a 50 mL tube, and submerged by 10 mM phosphate buffered saline (PBS) (Sigma-Aldrich®, Darmstadt, Germany) about 3 mL higher than salmon. After sealing, samples were put in the center of an ultrasonic chamber (Bionoon-950 ultrasonic Crusher, Bonuo Biotechnology Co., Ltd., Shanghai, China), and were finally treated for total processing time of 15 min or 30 min at room temperature with the conditions of ultrasonic power at 25 KHz and 300 W.

2.4. Blue light inhibited bacterial growth in vitro

2.4.1. Blue light treatment

The growth of V. parahaemolyticus was checked by measuring the value of OD600. The inhibitory activity of blue light on V. parahaemolyticus was determined by the agar drop diffusion method. Briefly, fresh salmon fillets were cut into 2 cm × 2 cm × 0.5 cm pieces using a sterilized knife. Samples were divided into 6–7 groups, and each group included 13–15 pieces of salmon. Bacteria suspension of 200 µL (preciously prepared, OD600 of 0.1) was centrifuged at 8000 rpm for 3 min, and the pellets were washed with 10 mM (PBS). After the second centrifugation, cells were diluted in a tenfold series from 10⁰ to 10^-7 cells/mL with PBS, and inoculated on VP agar in triplicate. Before cultivation, all plates were placed on ice and irradiated with blue light at 460 nm, and the arrangement is shown in Table 1 (Fig. S1). The sensitivity of bacteria to blue light was determined by counting the cell population on plates after being cultivated at 37 °C for 12 h.

Table 1.

Preparation of different samples and irradiated conditions by blue light.

Samples’ name	Irradiations conditions	Irradiation time
CK	No blue light,put on ice	30 min	60 min	120 min
BL10	Blue light at 10 W,put on ice	30 min	60 min	120 min
BL20	Blue light at 20 W,put on ice	30 min	60 min	120 min
BL30	Blue light at 30 W,put on ice	30 min	60 min	120 min

2.4.2. Blue light bacterial inhibition

To further identify the death of V. parahaemolyticus, 200 µL bacterial suspension (OD600 of 0.1) was inoculated into VP agar, and irradiated with different intensities of bule light at 72 J/cm², 144 J/cm² or 216 J/cm², respectively. Once plates were washed with 10 mM PBS and centrifuged at 8000 rpm for 3 min, cells were collected, and resuspended with 1 mL PBS, as well as mixed by 1 μg propidium iodide (PI) solution (Yeasen, Shanghai, China). After incubation at 37 °C for 10 min in darkness, PI penetrated damaged cell membrane. PI-stained cells were photographed under a fluorescence microscope (Axio Vert.A1, Carl Zeiss AG, Oberkochen, Germany), and evaluated by a flow cytometer (Beckman CytoFLEX FCM, Beckman Coulter, Inc., State of California, USA) with a blue argon laser at 488 nm. Cell suspension without blue light treatment was used as the control.

The morphology of V. parahaemolyticus was observed by SEM according to Endo, Garcia Cortez, Ueda-Nakamura, Nakamura and Dias Filho [22]. After bacterial suspension was irradiated with blue light and centrifuged (referred to 2.4), cells were collected and washed twice with PBS, following by fixing with 2.5% glutaraldehyde (Sigma-Aldrich®, Darmstadt, Germany) at 4 °C for 2 h. Then, cells were washed with PBS and dehydrated sequentially in graded ethanol with the concentration of 35%, 50%, 70%, 80%, 90%, and 100% (v/v). Finally, dehydrated samples were coated with gold under vacuum and examined in a Quanta 200 scanning electron microscope (FEI Company, Hillsboro, USA) at 20 kV acceleration voltage and 12000 × magnification. Cells without blue light treatment were controls.

2.5. Reactive oxygen species (ROS) assay

2.5.1. Detection of ROS

The endogenous ROS generation in V. parahaemolyticus were measured using fluorescent dyes 2′,7′-dichlorodihydrofluorescein diacetate (DCFH-DA, Molecular Probes, Eugene, OR, USA) through flow cytometry CyFlowspace (Partec, Norderstedt, Germany), modified by Yang, Zhao, Zhu, Tian and Wang [23]. The dyes were added into previously prepared bacteria suspension (OD600 of 0.1), with a final concentration of 1 μM. After incubation at 37 °C and 150 rpm for 30 min in darkness, cells were washed with PBS, and centrifuged to remove extracellular dyes. Then, cells with fluorescent dyes were irradiated with different intensities of bule light at 72 J/cm², 144 J/cm² or 216 J/cm². The production of ROS was evaluated by fluorescence microscope and flow cytometer [24]. Cells without blue light treatment were used as controls. The effect of inhibition of ROS production on V. parahaemolyticus survival was explored by using imidazole (IMZ 0.6–2.4 mM, Sigma-Aldrich®, Darmstadt, Germany).

2.5.2. Detection of DNA damage

The TUNEL probe (Roche, Basel, Switzerland) was loaded with the prepared bacterial suspension (OD600 of 0.1) according to the instructions of the reactive oxygen kit (Yeasen Biotechnology Co., Ltd., Shanghai, China). Briefly, 1 mL of prepared V. parahaemolyticus suspension was added with paraformaldehyde (PA) (Macklin, Shanghai, China) to fix cells, and placed in a shaker at 200 rpm for 30 min in order to prevent the cells from aggregating into clumps. After elution with PBS, 10 µL of TUNEL probe was added, and incubated for 1 h at 37 °C protected from light, then eluted and resuspended in PBS. Finally, DNA damage was detected by the flow cytometry.

2.6. RNA-seq for transcriptome analysis

The aim of RNA-seq for transcription analysis was to reveal the changes in the metabolic pathways of V. parahaemolyticus under blue light irradiation. Specifically, the prepared bacterial suspension (OD = 0.1) of 200 µL was inoculated onto VP agar, and irradiated with 30 W blue light for 30 min (survival rates approximately 20%) and without blue light as the control group, which were recorded as T1 and CK, respectively. Triplicates were performed for each treatment. After the irradiation, cells were washed off with prepared PBS buffer, and centrifuged at 6000 rpm for 3 min. Then, the desired organisms were collected and transferred to sterilized lyophilized tubes. After collection, the organisms were immediately stored in liquid nitrogen for freezing (greater than2h) and then placed on dry ice and sent to Gene Denovo (Guangdong, China) for transcriptome sequencing analysis. The bioinformatics analysis of RNA-Seq data was referring to Li, Wang, Zhou, Fan, and Sui [25].

2.7. Physical and chemical indicators of salmon

2.7.1. Shear force analysis

The shear force was determined using a TA.XT.plus texture analyzer with the WBS probe (Stable Micro Systems, Surrey, UK). The prepared sample (2 cm × 2 cm × 0.3 cm) of voluntary muscle was vertical to the cutter at a distance of 1 cm from the shear head, so that the blade could penetrate the entire sample. The speed pre-test, test or post-test was 1 mm/s. The shear force (N) was measured in triplicate.

2.7.2. Measurement of 2-Thiobarbituric acid (TBA)

After mixing in a mini food processor (Sunbeam, New York, USA), 100 g of meat was divided into 15 portions. TBA-reactive substances (hexanal and valeraldehyde) in samples were measured immediately after mincing. The colorimetric absorbance at 532 nm obtained from a spectrophotometer (PerkinElmer, Waltham, USA) was converted to milligrams of malondialdehyde per kilogram of meat to represent the TBA content.

2.7.3. Color measurement

The appearance of salmon was measured with a Minolta CR-200 colorimeter (Konica Minolta Holding Inc., Tokyo, Japan). Color was described using the CIELAB scale: lightness (L*), green to red hue (a*), and blue to yellow hue (b*). Each sample was measured five times at different surface points.

2.7.4. Sensory analysis

The sensory analysis was performed by a panel of assessors consisted of six experts in sensory analysis who were trained according to the requirements of the Chinese standard GB/T 22210–2008. Sensory attributes were indicated by appearance (glossiness), aroma (freshness), flavor (taste) and texture (elasticity) according to Pedrós-Garrido, Clemente, Calanche, Condón-Abanto, Beltrán, Lyng, Brunton, Bolton and Whyte [26].

2.8. Statistical analysis

Statistically significant differences (one-way ANOVA) were analyzed by IBM SPSS Statistics version 23.0 (IBM, NY, USA). All graphs were generated using GraphPad Prism® version 9 (GraphPad Software, San Diego, CA, USA), or were drawn using Microsoft® PowerPoint 16.24 (Microsoft, Redmond, USA). Significant differences between samples were identified at the confidence level of 5% (P < 0.05).

3. Results and discussion

Based on the fact that ultrasonic field (UF) treatment and blue light (BL) irradiation have a bactericidal effect, this study investigated the effective bactericidal intensity and mechanism of blue light against V. parahaemolyticus in salmon, as well as explored the bactericidal effect in combination with ultrasonic field and the effect on salmon organoleptic quality. The exploration pathway is shown in Fig. 1.

Fig. 1.

Investigation pathways of ultrasonic field and blue light treatment on salmon.

3.1. The bactericidal activity of blue light irradiation on V. Parahaemolyticus

The kinetic inactivation of bacteria in visible light depends mainly on the absorption spectrum of the endogenous photosensitizer [27]. Specifically, different bacteria produce various porphyrins, and the difference between the peak absorption wavelengths of these porphyrins is used to determine the wavelengths required for their optimal photostimulation, which in turn determines the wavelength and intensity of the selected visible light for sterilization [20]. The wavelength range required for the PDI of different bacteria was confirmed to be between 400 and 500 nm, while 520 nm did not have a significant killing effect on V. parahaemolyticus [20], [28]. Based on the results from a previous experiment where we tested the bactericidal effect of three wavelengths of blue light at 380 nm, 420 nm and 460 nm, we discovered that 460 nm was the optical choice not only due to its effective of killing V. parahaemolyticus, but also considering the security risk of short-wavelength blue light (data not shown). In the present study, the mortality rate of V. parahaemolyticus was approximately 100% at 460 nm blue light irradiation (Fig. 2(a1)). This is consistent with the result that blue light at 460 nm at 25 °C was able to cause PDI of V. parahaemolyticus with a lethality rate of 99.9% [28].

Fig. 2.

The killing effect of Blu-ray on V. parahaemolyticus. A) Cells survival of V. parahaemolyticus after irradiation of blue light under different power for different irradiation time (survival rate (cells without dilution, a1) and gel image of total colonies at different dilution concentrations (a2) were under the blue light for 120 min). B) Fluorescence signal of PI-stained V. parahaemolyticus by flow cytometry and microscopy after irradiation of blue light under different energy density. Cells recorded by flow cytometry (b1) and with brightfield (b2, upper row) and fluorescent (b2, lower row). C) The morphology of V. parahaemolyticus observed by Scanning electron microscopy after blue light irradiation. CK - samples without blue light treatment; BL10, BL20, BL30 - samples irradiated with blue light at the power of 10 W, 20 W or 30 W, respectively; Ed72, Ed144, Ed216 - samples irradiated by blue light at the energy density of 72 J/cm², 144 J/cm² or 216 J/cm², respectively; Letters (a, b, c, etc.) in the figure mean the statistic significant difference (Duncan, p < 0.05).

The bactericidal effect of PDI on V. parahaemolyticus on seafood was much weaker than that on planktonic bacteria in solution. This phenomenon is highly attributed to the protective function of the food mechanism, which provides gaps and fatty layers for microorganisms to hide on the surface [29]. Therefore, the PDI of V. parahaemolyticus on salmon shows a dose-dependent pattern of irradiation. In order to achieve the desired bactericidal effect, sufficient visible light dose to enhance the bactericidal efficiency in addition to the effective bactericidal wavelength is necessary. To explore the effect of BL intensity on V. parahaemolyticus cells, undiluted cells on plates were exposed to blue light at three different emission powers (10, 20 and 30 W) for 30, 60 and 120 min, respectively. Results showed that the survival of V. parahaemolyticus had a significant decrease at 120 min irradiation of BL, even at a lower emission power of 10 W, the bactericidal effect was over 90% (Fig. 2(a1)). Also, in the gel imaging results of colonies, V. parahaemolyticus cells were barely detectable at 4-fold dilution after 20 W, 120 min BL irradiation (Fig. 2(a2)). Obviously, the bactericidal effect of blue light on V. parahaemolyticus was positively correlated with the irradiation dose, which is consistent with the results of other researches [17], [28], [30]. Similar findings were made for a gradual decrease in bacterial population with increasing dose of blue light irradiation (at 405 nm), although applied to Bacillus cereus, Clostridium difficile, and Staphylococcus aureus [31]. Therefore, we chose BL irradiation for 120 min for further treatment to assess the damage to cell membranes. The absolute dose was in 72, 144 or 216 J/cm² under blue light irradiation at 10, 20 or 30 W, respectively, which was given by the irradiance (W/cm²) times the exposure time (in seconds), would be used to express the intensity of BL.

To further confirm the mortality of V. parahaemolyticus after BL irradiation, a PI fluorescent probe was applied to determine the population of dead cells. PI can penetrate damaged cell membranes and combine with nucleic acid within cells, and then show a distinct red light under blue light, so it can be used to label dead cells [32]. The results of flow cytometry showed that under high intensity of BL, the fluorescence peak deviated to the right, clearly moving towards longer wavelengths (Fig. 2(b1)). As the wattage of the BL irradiation increased, the fluorescence intensity of PI was gradually increased as observed by fluorescence microscopy (Fig. 2(b2)). In addition, blue light has the potential to induce structural changes in the cells, either internally or externally, as this treatment leads to cell death. Therefore, we used SEM to characterize the changes in cell morphology of V. parahaemolyticus. It has been suggested that V. parahaemolyticus under PDI treatment exhibits a large number of ruptured cells [30]. In this study, SEM analysis results showed that BL irradiation caused cell surface collapse and contraction, and with the increase of intensity, most of the cells were broken and adhered (Fig. 2C), but whether the cells had ruptured needs to be further confirmed. The above results confirmed that BL irradiation damaged the cell membrane of V. parahaemolyticus and eventually led to cell death.

3.2. Blue light irradiation induces ROS burst in V. Parahaemolyticus

Although blue light-induced bacterial cell death is associated with endogenous photosensitizers, it is the burst of cytotoxic ROS that plays a decisive role [15]. An excess of ROS generated can lead to cell membrane damage and cell death under aerobic conditions by oxidizing functional proteins, DNA and polyunsaturated fatty acids in the cell membrane [33], [34], [35]. Considering that ROS may be responsible for blue light-induced mortality in V. parahaemolyticus, we used the redox indicator probe DCFH-DA to detect ROS levels in cells. The florescence peaks collected by flow cytometry from BL treated cells were shifted towards longer wavelengths compared to CK samples, with the most significant shift to the right at an intensity of 216 J/cm² (Fig. 3(a1)). Correspondingly, blue light radiation triggered ROS production in V. parahaemolyticus in a dosage-dependent pattern. showing the strongest green fluorescence at 216 J/cm² (Fig. 3(a2)). Therefore, BL irradiation was able to induce ROS production in V. parahaemolyticus.

Fig. 3.

Changes in intracellular ROS of V. parahaemolyticus and the effect of inhibition of ROS production on the survival of V. parahaemolyticus after bule light irradiation with different energy density. A) Fluorescence signal of ROS in V. parahaemolyticus recorded by flow cytometry (a1) and microscopy (a2, with brightfield - upper row; fluorescent - lower row) after irradiation of blue light under different energy density. B) The effect of imidazole (IMZ) on the survival of V. parahaemolyticus after blue light treatment with different energy density. C) Fluorescence signal of DNA damage in V. parahaemolyticus detected by flow cytometry (c1) and microscopy (c2, with brightfield - upper row; fluorescent - lower row) after irradiation of blue light at the maximum energy density. CK - samples without blue light treatment; Ed72, Ed144, Ed216 - samples irradiated by blue light at energy density of 72 J/cm², 144 J/cm² or 216 J/cm², respectively; Letters (a, b, c, d) in the figure mean the statistic significant difference (Duncan, p < 0.05).

To investigate whether ROS was associated with BL-induced cell death in V. parahaemolyticus, the reactive oxygen species scavenger IMZ was added to the bacterial solution to explore the cell survival status. IMZ is a specific inhibitor of NADPH (nicotinamide adenine dinucleotide phosphate) oxidase, could applied to prevent ROS generation [36]. The results showed a significant increase in the survival of V. parahaemolyticus under BL irradiation after the addition of IMZ, saving V. parahaemolyticus from damage induced by blue light irradiation. (Fig. 3B). At the same time, an antagonistic effect between the increased concentration of IMZ and the bactericidal effect of blue light was verified, confirming a positive role for ROS in mediating blue light bactericidal activity (Fig. 3A,3B). Therefore, the production of reactive oxygen species within V. parahaemolyticus cells following blue light irradiation was confirmed to promote cell death.

Due to the burst of ROS, bacteria generate oxidative stress and then cause damage to biomolecules, including DNA damage, which is considered to be lethal to living cells [37]. To investigate whether BL irradiation causes cellular DNA damage leading to cell death, we used TdT-mediated dUTP biotin nick end labelling to detect apoptosis. The results showed that the fluorescence peak of BL treated cells was no significant deviation compared to CK (Fig. 3(c1)) and the cells showed no fluorescence (Fig. 3(c2)). Consequently, we inferred that BL did not cause significant damage to DNA of V. parahaemolyticus. Similarly, Kim, Mikš-Krajnik, Kumar and Yuk [38] treated other Gram-negative bacteria with blue light at 405 nm for 7.5 h which inactivated 1.0, 2.0 and 0.8 lg CFU/mL Escherichia coli O157:H7, Salmonella typhimurium and Shigella sonnei, respectively, and no DNA degradation was detected.

3.3. Transcriptome analysis

To understand the changes in the metabolic pathways of V. parahaemolyticus under BL stress, we performed RNA-seq analysis of blue-irradiated cells (Fig. 4). The resulting deferentially expressed genes (DEGs) were statistically analyzed, and a total of 1765 DEGs were found in the CK-vs-T1 group, of which 1179 were up-regulated and 585 were down-regulated. The DEGs were plotted as a volcano plot (Fig. 4A). Base on the fact that both red (up-regulated) and blue (down-regulated) points in the plot indicated differential gene expression, it shows that BL irradiation reduced significant differential gene expression in V. parahaemolyticus. According to the above results heat map was used to present the results of the relational, intergenic hierarchical clustering of the samples. As shown in Fig. 4B, for genes that were functionally similar or involved in the regulation of the same metabolic pathway, differential gene expression showed significant differences between the CK and BL treated groups (the description of genes is shown in Table S1). The most significantly enriched GO terms among the DEGs were found to include 9 terms for biological processes, four terms for cellular components and four terms for molecular functions (Fig. 4C). GO functionally enriched DEGs demonstrated that BL treatment significantly interfered with the metabolic process of V. parahaemolyticus. Several studies verified that cell membrane damage caused by blue light irradiation played a key role in sterilization [38], [39], [40]. Our study identified an enrichment of cell membrane-associated genes in GO analysis (Fig. 4C), which may have stimulated bacterial repair of membrane damage induced by blue light radiation.

Fig. 4.

Statistical analysis and Gene Ontology (GO) enrichment of differentially expressed genes (DEGs) between groups of Vibrio parahaemolyticus treated without or with blue light. A) Volcano map of DEGs in Vibrio parahaemolyticus (up - up-regulation; ns - no signification; down - down-regulation). B) Heat map of DEGs in Vibrio parahaemolyticus. C) Gene Ontology (GO) enrichment analysis of significant DEGs in Vibrio parahaemolyticus. D) Pathway enrichment analyzed with the Kyoto Encyclopedia of Genes and Genomes (KEGG) database of DEGs in Vibrio parahaemolyticus. Log2(FC) means log2(Fold Change), which takes the logarithm of the difference multiplier with a base of 2 (genes with P value < 0.01, FDR greater than 0.05 and log2(FC) greater than 1 or lower than −1 were assigned as significant DEGs); CK1, CK2, CK3 - the group ofVibrio parahaemolyticus treated without blue light; T1, T2, T3 - the group ofVibrio parahaemolyticus treated with blue light.

In addition, as mentioned above, ROS in PDI can react with various amino acid residues in proteins, especially sulphur-containing amino acids, cysteine and methionine, which are particularly sensitive [41]. In this study, analysis of the RNA-seq data using KEGG database revealed BL treatment significantly affected the metabolic pathways of V. parahaemolyticus, particularly the glycolytic pathway and biosynthesis of amino acids, mainly including valine, leucine and isoleucine, as well as methionine and cysteine (Fig. 4D). This may be due to that as V. parahaemolyticus senses BL irradiation, it activated the two-component system to control pathways such as microbial metabolism and glycolysis in different environments to adapt to BL stress. Through the screening of key genes, it is hypothesized that a large number of genes may have the roles of inducing gene activators under the BL stress and may undergo oxidative stress, leading to the destruction of membrane tissue structure and consequently bacterial death. All above confirm that blue light-induced cell death is associated with a burst of ROS, mainly due to the effects on cell metabolism and disruption of cell membranes. Similar results were also confirmed that PDI treatment significantly inactivated V. parahaemolyticus mainly by inducing protein degradation and cell structure rupture [42].

3.4. The effect of blue light irradiation on salmon quality

Since salmon is susceptible to microbial contamination during processing and storage, ready-to-eat fresh salmon were chosen to test the potential of BL to kill V. parahaemolyticus (Fig. 5,6). We evaluated the effect of blue light radiation on salmon by inoculating fresh salmon with or without V. parahaemolyticus. When salmon samples irradiated with BL intensities at 216 J/cm², the population of V. parahaemolyticus cells reduced by 1.63 lg CFU/mL (Fig. 5A) and 2.3 lg CFU/mL (Fig. 6A), reaching a bactericidal effect almost 100%. Based on the texture characteristic play a crucial role in the quality of food during storage, shear force representing texture indicator was measured to investigate the effect of BL on salmon tenderness. The shear force of salmon was significantly reduced only by exposure to high intensity BL at 216 J/cm², while low intensity BL did not change the tenderness of salmon (Fig. 5B,6B).

Fig. 5.

Effect of blue light irradiation on quality parameters of fresh salmon infected by V. parahaemolyticus. A) Total cell population. B) Shear force of salmon. C) TBA content. D) Color index. CK - salmon no infected with bacteria and untreated with blue light; Ed0 - salmon infected with bacteria but without irradiated by blue light; Ed72, Ed144, Ed216 - salmon infected with bacteria and irradiated by blue light at energy density of 72 J/cm², 144 J/cm² or 216 J/cm², respectively. Letters (a, b, c, etc.) in the figure mean the statistic significant difference (Duncan, p < 0.05).

Fig. 6.

Effect of Blue light irradiation on quality parameters of ready-to-eat fresh salmon without contaminated with V. parahaemolyticus at different energy density A) Total cell population. B) Shear force of salmon. C) TBA content. D) Color index. E) Sensory analysis of appearance, aroma, flavor and texture. CK - ready-to-eat fresh salmon untreated with blue light; Ed72, Ed144, Ed216 - ready-to-eat fresh salmon irradiated by blue light at energy density of 72 J/cm², 144 J/cm² or 216 J/cm², respectively; Letters (a, b, c, etc.) in the figure mean the statistic significant difference (Duncan, p < 0.05); (*) show the significant difference of each sensory indicators compared with CK (LSD, p < 0.05).

Light projection may lead to lipid oxidation, discoloration, nutrient loss and off-flavors in food products [43]. When applying blue light sterilization to food products, whether photosensitive oxidation occurs should be considered. Photosensitive oxidation is the reaction of the more electrophilic 1O₂ in the raw material with a high density of electron double bonds to form hydroperoxides in the trans configuration [44], leading to fat oxidation. Short wavelength light (455 nm or lower) significantly affected the rate of oil oxidation [45]. In this study, simulated blue light irradiation of ready-to-eat fresh salmon under refrigerated conditions (4 ± 1 °C) did not increase TBA and thus did not cause lipid peroxidation (Fig. 5C,6C). This may be attributed to the fact that salmon flesh contains many carotenoids and β-carotene, which quench singlet oxygen [46] or photosensitizers [47], thereby inhibiting photosensitive lipid oxidation. It is also possible that PDI generates ROS by consuming large amounts of oxygen [48], which reduces the O₂ moles on the surface of salmon to prevent lipid oxidation. This result is consistent with that of Chen, Huang, Liu, Liu, Zhao and Wang [17], where PDI effectively inhibited lipid oxidation in oysters.

Color is another key determinant of the visual quality and market value of seafood [49]. The change of color in food are associated with various factors such as lipid oxidation and enzymatic reactions [50], [51]. In our study, when exogenous V. parahaemolyticus was inoculated on salmon, BL radiation at high intensities (144 and 216 J/cm²) did not give a significant difference in the color of salmon (Fig. 5D). This is consistent with the results of Josewin, Ghate, Kim and Yuk [52], blue light at 460 nm inactivated L. monocytogenes in smoked salmon without causing color changes despite being irradiated at high doses (2400 J/cm²). Similarly, irradiation of fresh salmon with blue light at 405 nm for an extended period of time (8 h) did not result in color changes [53]. In contrast, when blue light irradiated to salmon without inoculating exogenous V. parahaemolyticus, the high intensity of BL could significantly alter the brightness (represented by L values) of the salmon, with a difference in L values of 1.42 (reduced) and 1.02 (increased) under the intensity of 144 and 216 J/cm², respectively, compared to CK. In addition, the high intensity BL (216 J/cm²) significantly reduced the redness (represented by a*) of salmon by approximately 1.10, a value that was also a significant increase in the redness of salmon at 72 and 144 J/cm² intensities of BL. At the same time, BL treatment enhanced the yellowness (represented by b*) of salmon (Fig. 6D).

As an evaluation index, emotional test is aim to indicate the acceptance of consumers to products [54]. To evaluate the sensory characteristics of the ready-to-eat salmon after BL irradiation, appearance, aroma, flavor and texture were measured. The BL treatment resulted in a significant enhancement of the aroma, reducing the fishy smell, and the high intensity of BL at 144 J/cm² decreased the elasticity of salmon characterized by the texture (Fig. 6E). Therefore, the proposed intensity of blue light (216 J/cm²) in this study has potential application in killing V. parahaemolyticus, improving the quality of salmon.

3.5. Bactericidal effect of ultrasound field and blue light irradiation on V. Parahaemolyticus

As ultrasound field (UF) has a certain bactericidal effect [55]. We combined UF and BL to evaluate the bactericidal effect of UF assisted sterilization. Salmon samples were treated with UF followed by BL irradiation, and subsequently we evaluated the survival rate of V. parahaemolyticus and the organoleptic properties of the salmon. The results showed that single UF treatment could significantly kill approximately 90% of V. parahaemolyticus. Furthermore, UF and BL treatment have a superposition effect on sterilization. After UF treatment for 15 or 30 min combined with BL radiation, the sterilization rates were 98.81% or 99.99%, respectively, which were higher than that of single BL treatment (97%). Especially at UF for 15 min (sample of UF15 + BL), the cell population decreased from 8 to 4 lg CFU/mL (Fig. 7A). The microbial inactivation by ultrasonic field is usually attributed to the aerobic effect, which leads to cell membrane rupture and damage to DNA, ultimately leading to the inactivation of different microorganisms [55], [56]. A study from Khandpur and Gogate [19] confirmed that the combination of ultrasonic sterilization and ultraviolet (UV) radiation was able to significantly reduce the population of microorganisms. However, in this study, the mechanism of ultrasound treatment on V. parahaemolyticus needs further investigation.

Fig. 7.

Effect of ultrasonic field (UF) and blue light (BL) irradiation treatment on the quality of ready-to-eat fresh salmon infected by V. parahaemolyticus. A) total cell population. B) Shear force of salmon. C) TBA content. D) Color index of salmon. CK - salmon without treated by blue light and ultrasonic field; BL - salmon treated with blue light at 216 J/cm²; UF15, UF30 - salmon treated by ultrasonic field for 15 min or 30 min but without blur light, respectively; UF15 + BL, UF30 + BL - salmon treated by ultrasonic field for 15 min or 30 min and treated with blue light at 216 J/cm², respectively; Letters (a, b, c, etc.) in the figure mean the statistic significant difference (Duncan, p < 0.05).

Different from the BL radiation only, the UF treatment for 30 min was able to significantly reduce the shear force of the salmon, as in samples of UF30 and UF30 + BL (Fig. 7B), while 15 min UF treatment had no effect. Similarly, the addition of UF procedure did not lead to fat oxidation in salmon (Fig. 7C). Noteworthily, the UF treatment for 30 min was able to elevate the L value by 11.83 in UF30 and 12.79 in UF30 + BL, respectively, compared to the CK sample. At the same time, prolonged UF treatment significantly reduced the redness (a*) and increased the yellowness (b*) of the salmon. However, UF treatment of 15 min had no significant effect on color. Therefore, based on the above consequences, we conclude that the combination of UF and BL killed V. parahaemolyticus more effectively, compared to use single UF or BL treatment, but the time of UF needs to be strictly controlled because a long-term UF treatment could reduce the freshness and brightness of salmon. This might be due to the fact that ultrasound treatment of salmon cause muscle fiber contraction or browning caused by the Maillard reaction [57].

4. Conclusion

Blue light is effective in killing V. parahaemolyticus in salmon by inducing ROS production of V. parahaemolyticus, leading to 96.8–99.9% of cell death at 216 J/cm² intensity without effect in salmon quality. In addition, the additive ultrasonic field treatment could enhance the bactericidal effect by blue light, killing 98.81–99.99% of V. parahaemolyticus cells. However, the duration of ultrasonic field needs to be controlled to prevent affecting the freshness and color of the salmon. This study recommends blue light or a combination with ultrasonic filed as effective methods for cold sterilization of food.

CRediT authorship contribution statement

Xiaolin Zhu: Conceptualization, Methodology, Investigation, Data curation, Writing – original draft. Han Yan: Methodology, Data curation, Formal analysis, Software. Zhenkun Cui: Funding acquisition, Conceptualization, Project administration, Writing – review & editing. Hongbo Li: Data curation, Conceptualization, Methodology. Wei Zhou: Investigation, Data curation. Zhenbin Liu: Investigation, Writing – review & editing. Hao Zhang: Investigation, Data curation. Tatiana Manoli: Visualization, Writing – review & editing. Haizhen Mo: Conceptualization, Methodology, Project administration, Writing – review & editing. Liangbin Hu: Conceptualization, Methodology, Investigation, Data curation, Funding acquisition, Writing – review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: