Surface sanitation against foodborne pathogens in domestic environments: efficacy of a handheld ultraviolet C light emitting diode device

Abstract

Aim

Ensuring food safety is fundamental to protect public health, especially in domestic environments where food is handled daily. This two-phase study aimed to investigate the efficacy of an ultraviolet-C light emitting diode (UV-C LED) handheld lamp through in vitro disinfection tests and disinfection tests on artificially contaminated surfaces.

Methods and results

The UV-C LED-based lamp efficacy was assessed at different initial microbial contamination titers and several UV doses, and both American Type Culture Collection (ATCC), foodborne, and clinical strains were considered. The UV-C LED lamp demonstrated high efficacy (log₁₀ reduction >1 log) against the standard bacteria strains tested using a UV dose of 21.77 mJ cm⁻². The greatest efficacy was achieved against E. coli (k = 0.0232) followed then by Bacillus subtilis (k = 0.0225) against which a titer of <1 CFU ml⁻¹ was achieved with a UV dose of 15.55 and 21.77 mJ cm⁻², respectively. Cladosporium spp. (k = 0.001) showed higher resistance against UV treatment, where a 50.00 ± 14.14 inactivation rate % (%IR) was achieved by applying the highest UV dose (31.1 mJ cm⁻²). Compared with B. subtilis, isolated L. monocytogenes 490 showed similar susceptibility (k = 0.0236), unlike isolated Listeria monocytogenes 1484 (k = 0.0146), isolated Salmonella Infantis 43 072/20 (k = 0.0126), and isolated S. Infantis 29 673/20 (k = 0.0124), which showed greater UV resistance. Considering the results obtained on the surfaces (stainless steel, polypropylene, and glass), the type of surface material influenced the susceptibility of isolated bacterial strains. However, the presence of organic matter (5% fetal bovine serum) on the treatment surface did not significantly affect device decontamination efficiency by applying a UV dose of 15.55, 21.77, and 31.1 mJ cm⁻². On both stainless steel and glass, a titer <1 CFU 169 cm⁻² was achieved against all the isolated bacterial strains applying a UV dose of 31.1 mJ cm⁻², except with L. monocytogenes 1484. Lastly, a titer <1 CFU 169 cm⁻² was never achieved on polypropylene contaminated with L. monocytogenes 1484.

Conclusion

The use of a UV-C LED handheld lamp (peak wavelength 265 nm) could be an efficient disinfection method to be applied in domestic or small-scale food-processing environments, to reduce the cross contamination of food.

Introduction

Despite numerous government and industrial interventions implemented in recent years, the number of reported cases of foodborne diseases is constantly increasing. Every year, an estimated 600 million—almost 1 in 10 people in the world—fall ill after eating contaminated food, resulting in a global annual burden of 33 million disability-adjusted life years and 420 000 deaths (WHO 2024).

Environments with a high risk of food contamination are mainly represented by facilities where huge quantities of food intended for large-scale distribution are processed. Nonetheless, domestic environments and small workplaces also fall into this category of at-risk environments (Byrd-Bredbenner et al. 2013, Silva et al. 2017).

An important point to be highlighted is the crucial role that can be represented by contaminated food preparation surfaces, for the disinfection of which chemical products are often used. However, despite their high effectiveness, disinfectants have a negative impact both on the environment and on our health. In particular, the production and sedimentation on surfaces of possible residual biocides can promote the emergence and spread of antibiotic resistance in bacteria present in those environments, including pathogens (Li et al. 2023). An innovative approach includes the design of surface materials that are “self-decontaminating” from microbes, but their application is still limited, and their effectiveness and potential risks are unknown (Filippidou et al. 2015, Querido et al. 2019, Maillard and Pascoe 2024).

A recent strategy adopted for the disinfection of surfaces in indoor environments (belonging to different at-risk environments) is ultraviolet-C light emitting diode (UV-C LED) technology (Beck et al. 2017, Santos and de Castro 2021, Palma et al. 2024). This strategy offers a rapid and non-destructive disinfection method, ensuring the microbiological safety of food materials without leaving any residue (Wang et al. 2023). To date, optimal disinfection results using UV-C LED have been obtained in surface disinfection, reaching inactivation rates up to 99.9% both against bacteria and viruses in indoor environments (Calle et al. 2021, Mariita et al. 2022). Additionally, UV-C LED technology is a chemical-free and heat-free alternative, making it environmentally friendly and cost-effective, also for domestic use. Furthermore, it overcomes the limitations of traditional mercury UV lamps, such as long warm-up times, short lifetime, and the risk of mercury contamination (Palma et al. 2022). However, it should be underlined that, particularly in water, some kind of microorganisms can repair UV-induced genetic damage, such as Mycobacterium avium or some Gram-positive bacteria known to be resistant to UV rays (Schiavano et al. 2018). In many cases, this genetic repair is due to the presence of some photolyase enzymes, such as DNA photolyase, that can bind the covalent bonds of thymine dimers (CPDs), formed during the UV exposure, to return the CPD to its original state using the visible light energy (Jones and Baxter 2017). To avoid photorepair, UV doses high enough to inactivate microorganisms should hence be used. Furthermore, the disinfection efficacy through UV-C irradiation can be impaired by the following various factors that should be considered and optimized in order to standardize the method and to reach the optimum efficiency; namely, the distance from the light source, environmental parameters such as temperature and humidity, surface material, and the presence of organic matters (Bolashikov and Melikov 2009, Querido et al. 2019, Kim and Kang 2020, Zhang et al. 2020, Fuchs et al. 2022).

The use and development of new UV-C-based technologies could represent an effective solution to guarantee the hygiene of the environments in which food is handled (Lim and Harrison 2016, Gabriel et al. 2018, Song et al. 2023, Park et al. 2024). However, almost all of the UV-C LED devices considered before are not always applicable to domestic environments. For this reason, in our study, a new UV-C LED handheld lamp, designed for decontamination of surfaces in small-scale food processing environments, was considered.

The purpose of this study was: (i) to validate the efficacy of UV-C LED hand lamp in in vitro decontamination tests using as test microorganism the Gram-negative Escherichia coli ATCC 25922, the Gram-positive bacteria Bacillus subtilis ATCC 6633 (vegetative form) and the filamentous fungus Cladosporium spp.; (ii) to evaluate the susceptibility of strains of two foodborne bacterial species isolated from food and clinical samples (Listeria monocytogenes and Salmonella Infantis) and to examine the efficacy of UV-C LED lamp, for decontamination of food handling-related surfaces, such as stainless steel, polypropylene (PP), and glass surfaces; (iii) to calculate the susceptibility constants (k) of the tested microorganisms. Indeed, to our knowledge, standards about the needed UV doses for microbial inactivation in in vitro settings and on different kinds of surfaces are not available yet.

Materials and methods

UV-C LED-based hand lamp, irradiation conditions, and UV Dose calculation

The device used was a UV-C LED-based hand lamp (59S^® UVC LED HANDHELD STERILIZER) composed of two parallel lines of 20 LEDs, tuned into a wavelength of 260–280 nm (peak wavelength 265 nm). As reported by the manufacturer, the average irradiance at a vertical distance of 1 cm below the top UV-C LED is 622 µW cm⁻². The UV-C LED lamp does not produce ozone and was equipped with a safety system that turned off the LEDs when the lamp rotated upward. During all the experiments, the irradiation treatment was performed with the UV-C LED lamp by scanning over the contaminated Petri dishes or surfaces 0, 3, 5, 7, and 10 times (corresponding to 0, 15, 25, 35, and 50 s of UV-C exposure, respectively), at a distance of 1 cm using a support. In order to standardize the treatment, the irradiation scan speed was kept manually around 0.026 m s⁻¹. The different test experimental conditions and the corresponding UV-C exposure times are reported below.

To calculate the UV Doses delivered during the experimental conditions, the following formula (Formula 1) was applied:

in which D (J m⁻²) was the UV Dose, E_t (seconds) was the irradiation exposure time of the samples, and I_r (W m⁻²) was the radiation emitted by the lamp.

Bacterial and fungi strains

The efficacy of the UV-C-based hand lamp described above was initially assessed on the following test microorganisms: E. coli ATCC 25922, B. subtilis ATCC 6633 (vegetative form), and Cladosporium spp. For these types of microorganisms (Gram-negative and Gram-positive bacteria, and fungi), characteristics and their different UV susceptibility constants are known, enabling their classification from the most to the least sensitive to UV-C irradiation (Beauchamp and Lacroix 2012, ISO 2019).

In addition, two strains of L. monocytogenes (L. monocytogenes 490 and L. monocytogenes 1484) collected within the framework of the annual official control plan activity (European_Commision 2005, 2007), carried out in the Marche region for routine and surveillance system, and two strains of S. enterica subsp. enterica serovar Infantis (S. Infantis 43072/20 and S. Infantis 29763/20) were used. Listeria monocytogenes 490 and S. Infantis 43072/20 were isolated from clinical samples, while L. monocytogenes 1484 and S. Infantis 29 763/20 were isolated from food matrices (ready-to-eat food and chicken meat, respectively). (Schiavano et al. 2022, Amagliani et al. 2021, Russo et al. 2024).

Test microorganism suspensions preparation

Escherichia coli ATCC 25922, S. Infantis 43072/20, and S. Infantis 29763/20 were grown on tryptic soy agar (TSA) plates, incubated at 37°C for 24 h; B. subtilis ATCC 6633 was grown on brain heart infusion (BHI) agar incubated at 37°C for 24 h. Overnight cultures of E. coli ATCC 25922, S. Infantis 43 072/20, S. Infantis 29763/20, and B. subtilis ATCC 6633, grown in tryptic soy broth (TSB) at 37°C for 18 h, were washed twice and resuspended in saline solution (NaCl 0.9% w/v); then, the optical densities at 600 nm (OD₆₀₀) were adjusted to values corresponding to a final concentration of 10⁸ CFU ml⁻¹ (for E. coli ATCC 25922, S. Infantis 43072/20, and S. Infantis 29763/20) and 10⁷ CFU ml^-1 (for B. subtilis ATCC 6633). During the preparation of B. subtilis ATCC 6633 suspension, a NaCl 0.9% solution with 0.05% TWEEN 80 was used to avoid the formation of bacterial clusters.

Cladosporium spp. was grown in agar slant tubes containing malt extract agar (MEA) with 0.05 g l⁻¹ chloramphenicol, at 25°C for at least 72 h. The fungal suspension was prepared by pouring NaCl 0.9% w/v solution over Cladosporium spp. slants, and gently swirling to detach fungal conidia. The fungal suspension was then diluted to an optical density at 530 nm (OD₅₃₀), corresponding to a final concentration of 10⁶ CFU ml⁻¹.

Listeria monocytogenes 490 and L. monocytogenes 1484 were grown on TSA supplemented with 6 g l⁻¹ Yeast Extract (TSAYE). Overnight cultures of L. monocytogenes 490 and L. monocytogenes 1484 were grown in TSB with Yeast Extract (6 g l⁻¹) (TSBYE) at 37°C for 18 h, then washed twice and resuspended in NaCl 0.9% w/v; the OD₆₀₀ was adjusted to values corresponding to a concentration of 10⁸ CFU ml⁻¹.

All microorganism suspension concentrations were experimentally confirmed by plate culture. All culture materials were purchased from Thermo Fisher Scientific (Waltham, MA, USA).

Disinfection tests in in vitro setting

The first part of this study consisted of irradiating microorganisms spread onto solid media (TSA for E. coli ATCC 25922 and S. Infantis; TSAYE for L. monocytogenes; BHI for B. subtilis ATCC 6633; and MEA for Cladosporium spp.) in Petri dishes (90 mm diameter). One hundred microliter of bacterial and fungal suspensions, in the range 10^2–10⁵ CFU ml⁻¹, were dispersed onto agar, then allowed to completely dry at 35°C up to 10 min and irradiated with the previously mentioned UV-C LED lamp, according to experimental conditions reported in the Section “UV-C LED-based hand lamp, irradiation conditions, and UV Dose calculation”. The tests were performed with three independent experiments, and each experiment was carried out in triplicate. After incubation at 37 °C for 24 h for bacterial species and at 25°C for 72 h (Cladosporium spp), the CFU ml⁻¹ were calculated. For safety reasons, the operators have worn protective glasses, laboratory coat, and protective gloves during the tests.

Disinfection tests on artificially contaminated surfaces

During the second part of this study, the disinfection efficiency of the UV-C LED hand lamp was assessed on surfaces of different materials commonly used in food processing environments for food handling. Three different surfaces were treated with UV-C LED irradiation: stainless steel, PP, and glass. Each surface was previously cleaned and sterilized in autoclave or in dry stove, to remove any biological contaminants. The contamination phase was then carried out under a laminal flow hood. Each surface (13 by 13 cm, 169 cm²) was contaminated by homogenously spreading, using a loop, 200 μl of bacterial suspensions (10³ CFU ml⁻¹) of each L. monocytogenes, and S. Infantis strain, individually. To simulate the presence of surface organic matter, 5% fetal bovine serum (FBS) (Sigma–Aldrich, St. Louis, MO, USA) was added to the bacterial suspension (ASTM 2017). Microbial concentration on the surfaces was evaluated by using Replicate Organism Detection and Counting (RODAC) plates (TSA for S. Infantis strains and TSAYE for L. monocytogenes strains), through the standardized method employing the Contact Plate Weight Applicator (VWR International Srl, Milan, Italy) and 10 s as sampling time (ISO 2018). The contaminated surfaces were divided into 4 sections: 2 control sections (in which the sampling was performed before the UV-C treatment) and 2 treated sections (in which the sampling was performed after the UV-C treatment). The experimental conditions for the irradiation times were the same as used for the in vitro tests and are reported in the Section “UV-C LED-based hand lamp, irradiation conditions, and UV Dose calculation”. The experiment was replicated three times. All contact plates (each of which had a surface area of 24 cm²) were incubated at 37 °C for 24 h and then the 169 cm² were calculated.

Moreover, the results were also converted to log₁₀ CFU 169 cm⁻² to calculate the log-reduction induced by all the UV-C irradiation times considered in the different surface coupons tested. As previously mentioned, the same operative safety measures were taken also at this stage of the study.

Inactivation rate calculation

To calculate the percentage of inactivation (%IR) of microorganism growth determined by UV-C LED irradiation, the following formula (Formula 2) was applied:

in which A was the microbial concentration in the contaminated not UV-C treated control, and B was the microbial concentration obtained after the UV-C disinfection at the different UV doses.

Determination of the D₉₀ values and UV inactivation constant (k)

Based on the %IRs obtained, the UV doses causing a 90% reduction of bacterial number (D₉₀) were calculated with GraphPad Prism 8.0 (San Diego, CA, USA), by a non-linear regression curve model. From the D₉₀ values, the UV inactivation constant (k) of each microorganism was calculated using the following formula (Formula 3) (Kesavan and Sagripanti 2013):

Statistical analysis

Statistical analyses were performed with GraphPad Prism 8.0 (San Diego, CA, USA), using a two-way Analysis of Variance (ANOVA) test, followed by Tukey’s multiple comparison test. A difference was identified as significant at a P value of less than 0.05 (*), 0.01 (**), and 0.001(***).

Results

UV doses used during the experiments

The UV Doses (D) used to inactivate all test microorganisms, calculated with Formula 1, were reported in Table 1.

Microbial inactivation in in vitro tests

The in vitro tests assessed the disinfection efficacy of the UV-C LED hand-lamp using three standard reference microorganisms (E. coli ATCC 25922, B. subtilis ATCC 6633, and Cladosporium spp.), with known UV susceptibility (ISO 2019), and two different strains for each foodborne bacterial species (L. monocytogenes and S. Infantis).

Effects of UV-C LED-based hand lamp on standard reference microorganisms

The colony plate count confirmed the expected starting concentration of each microbial suspension and allowed us to calculate the microbial titers in the control plates.

Data about initial microbial concentrations (Log CFU ml⁻¹), and about the %IRs and Log reductions (CFU ml⁻¹) obtained after the UV-C disinfection treatment, performed by irradiating bacterial cultures of the reference strains at various starting concentrations with four different UV Doses (9.33, 15.55, 21.77, and 31.1 mJ cm⁻²), are shown in Fig. 1.

As shown in Fig. 1, the %IRs obtained irradiating the bacterial strains were related, as expected, to the initial titer, and the effect was particularly evident at the lowest UV doses tested. This result was confirmed by the experiments with the fungal strain, in which its known higher resistance to UV rays led to less linear data. Interestingly, at the starting concentration of 10³ CFU ml⁻¹, a high microbial inactivation was already obtained with the lowest UV-dose (9.33 mJ cm⁻²), both with E. coli ATCC 25922 and B. subtilis ATCC 6633, (90.14 ± 4.30% and 88.65 ± 6.13%, respectively), while with Cladosporium spp. only a 21.41 ± 6.71%IR was achieved at the highest UV dose (31.1 mJ cm⁻²). However, a higher %IR (50 ± 14.14%) was obtained maintaining the UV dose at the maximal level (31.1 mJ cm⁻²), but with a starting fungal titer of 10² CFU ml⁻¹. Conversely, at the same UV dose, a complete (<1 CFU ml⁻¹, E. coli ATCC 25922, all starting titers and B. subtilis, 10³–10⁴ CFU ml⁻¹) or nearly complete (99.82 ± 0.06%IR, B. subtilis ATCC 6633, 10⁵ CFU ml⁻¹) inactivation was observed in the two bacterial species.

Evaluating the results obtained with the same UV doses, the initial bacteria titer significantly influenced the microbial reduction of E. coli ATCC 25922 only with a UV dose of 9.33 mJ cm⁻², while considering B. subtilis ATCC 6633 also with a UV dose of 15.55 mJ cm⁻². At higher UV doses, the initial titer didn’t affect the UV-C LED hand-lamp efficacy. Indeed, in these conditions, the lamp can inactivate the entire microbial population even with the highest microbial titer tested (10⁵ CFU ml⁻¹); even more so it is confirmed in case of lower initial concentrations spread on the agar plates, reaching a plateau starting from a UV dose of 21.77 mJ cm⁻².

Effects of UV-C LED-based hand lamp on bacteria isolated from clinical samples and food matrices

In a subsequent phase, the UV-C susceptibility of 2 strains of L. monocytogenes and two strains of S. Infantis, was evaluated. The %IRs, Log reductions (CFU ml⁻¹), and initial microbial concentrations (log CFU ml⁻¹) obtained by in vitro experiments (Fig. 2) confirmed the UV-C dose-dependent inactivating effect with all the tested strains.

Listeria monocytogenes 490 showed a higher susceptibility to UV-C irradiation with respect to the other strains, which anyway showed values always >85% applying a UV dose of 9.33 mJ cm⁻². In this case, the initial bacterial titer did not significantly affect lamp efficacy; only at the lowest UV dose (9.33 mJ cm⁻²) and only in L. monocytogenes 490 and S. Infantis 29 763/20 the %IR was related to the initial titer, with a higher effect at the lower initial microbial concentration.

Microbial inactivation in surface disinfection tests

In the second part of the study, the disinfection process with the same UV-C device was carried out by simulating a real application context on food handling surfaces made of three different materials in the presence of organic matter, artificially contaminated with the L. monocytogenes and S. Infantis strains. The microbial concentrations were evaluated through surface sampling (see Section “Disinfection tests on artificially contaminated surfaces”) before and after surface irradiation. Before the treatment, the microbial load sampled on the PP, stainless steel, and glass surfaces was experimentally measured by colony plate counting.

Data about the %IR for each strain of L. monocytogenes and S. Infantis on the different surfaces are reported in Fig. 3.

Analyzing the results obtained with the lowest UV dose (9.33 mJ cm⁻²), an %IR greater than 80% was achieved in all experimental conditions, except with the PP surface artificially contaminated with L. monocytogenes 1484 (58.86 ± 2.42%). This strain was the least susceptible one in the surface experiments, displaying a %IR slightly lower than the other strains, in all the experimental conditions tested, even in the irradiation tests with the highest UV dose.

In all the experiments conducted with an irradiation UV dose greater or equal to 15.55 mJ cm⁻², the stainless steel was the surface which led to the highest %IR in all the strains tested (%IR > 97%). Then, increasing the UV dose up to 21.77 or 31.1 mJ cm⁻², the %IR improved until no presence of any colony (<1 CFU 169 cm⁻²) after treatment. Even in glass surfaces the %IR was very high, with a range of 82.18–96.07% with the lowest UV dose, and no presence of any colony (<1 CFU 169 cm⁻²) with the highest one, except with L. monocytogenes 1484 (98.68 ± 1.32%). Considering the PP, a complete disinfection of the surface was achieved only by applying the highest UV dose (31.1 mJ cm⁻²), except with L. monocytogenes 1484 against which an IR% of 97.58 ± 1.45% was achieved. When applying a UV dose of 9.33, 15.55, or 21.77 mJ cm⁻², the statistical analysis showed significant differences between the %IR obtained on the different surfaces, whereas when using a UV dose of 31.1 mJ cm⁻² no statistically significant differences were found. Lastly, considering S. Infantis 29763/20, at the UV dose of 15.55 mJ cm⁻², significant differences emerged between PP and the other two surfaces, due to the higher residual viability found on PP. Comparing the results obtained in the presence of 5% FBS and in the absence of FBS (Supplementary material S1), it was found that the presence of organic soil load did not significantly influence the efficacy of the lamp (P > 0.05), except at the lowest UV dose (9.33 mJ cm⁻²) for L. monocytogenes 1484 and S. Infantis 29763/20 on stainless steel (P < 0.05). Data expressed as log reduction (CFU 169 cm⁻²) are reported in Fig. 4.

The log CFU 169 cm⁻² in the controls ranged from 3.30 to 3.60 log CFU 169 cm⁻²; in three out of the four strains tested (L. monocytogenes 490, S. Infantis 43 072/20, and S. Infantis 29 763/20), after the irradiation with the highest UV dose (31.1 mJ cm⁻²), the obtained Log reductions in glass were equal to the average initial population densities (log CFU 169 cm⁻²) recovered by the same CTR surface, not irradiated with UV. Contrarily, on glass and PP contaminated with L. monocytogenes 1484 a log reduction CFU 169 cm⁻² of 2.49 ± 0.92 and 1.71 ± 0.30 was achieved, respectively. Interestingly, applying a UV dose of 21.77 mJ cm⁻², a titer of <1 CFU 169 cm⁻² was achieved with all bacterial strains on stainless steel coupons. In general, we can assume that the irradiation treatments led to a UV dose-dependent log reduction in all the surfaces tested, even if in a non-linear manner.

D₉₀

Based on the inactivation results obtained, the D₉₀ of each microorganism used in the study was calculated considering both the initial microbial titers and the kind of surfaces used. The curves from which the D₉₀ values were extrapolated are shown in the Supplementary material S2. Subsequently, the UV Inactivation Constant (k) for each microorganism was calculated from the D₉₀ values (Formula 3). All the data are reported in Tables 2, 3, and 4.

As reported in Table 2, considering all initial microbial titers (10⁵, 10⁴, and 10³ CFU ml⁻¹), the UV dose needed to inactivate 90% of the initial microbial titer of E. coli ATCC 25922 was similar to B. subtilis ATCC 6633 (P > 0.05). Conversely, much higher UV doses were needed to inactivate 90% of Cladosporium spp., especially considering an initial mycotic titer of 10⁴ CFU ml⁻¹. Different D₉₀ values were also found between the two L. monocytogenes strains (Table 3), confirming that L. monocytogenes 1484 had greater UV resistance than L. monocytogenes 490. This difference between D₉₀ values was not identified between the two S. Infantis strains, confirming their similar susceptibility to UV treatment. A difference in D₉₀ values was also found considering different test surfaces at the same initial bacterial titer (10³ CFU ml⁻¹). The PP surface showed a greater residual bacterial viability after the UV-C treatment and, consequently, higher D₉₀ and lower k values with respect to glass and stainless steel surfaces. Considering the D₉₀ of L. monocytogenes 1484 obtained on PP, the D₉₀ value was always higher compared to the L. monocytogenes 1484 D₉₀ values obtained in vitro on TSAYE medium, therefore indicating an increase in microorganism UV resistance if dispersed on PP surface. This difference was also found in the D₉₀ values of L. monocytogenes 490. Listeria monocytogenes 1484 was more UV resistant than L. monocytogenes 490 on all surfaces; differences were also found between the D₉₀ values of L. monocytogenes 1484 and L. monocytogenes 490, both on PP, stainless steel, and glass. However, the two S. Infantis strains showed similar D₉₀ and susceptibility considering all three surfaces. Finally, considering all D₉₀ results obtained on surfaces, L. monocytogenes 1484 showed a higher D₉₀ value than L. monocytogenes 490, S. Infantis 43072/20, and S. Infantis 29763/20 on all types of surfaces.

Discussion

Based on UV-C LED technology, we designed this two-stage study to evaluate the disinfection efficacy of a UV-C LED hand lamp, as well as assess the susceptibility of microorganisms of relevance for food safety both on the plate and on surface material often in contact with food. Initially, test microorganisms whose susceptibility to UV is known were used to carry out the efficacy tests. Subsequently, the same in vitro tests were carried out with four field microorganisms, whose susceptibility was then compared with the susceptibility of the same field microorganisms but spread on three types of food surfaces of interest, going on to assess whether the nature of the surface material could in any way influence UV-C efficacy.

Considering the factors that can affect the efficacy of the irradiation during the disinfection step, it must be underlined that the exposure time, the distance (de Groot et al. 2019) and the initial microbial concentration (Jaiaue et al. 2022, Mathur 2022) can considerably modify the rate of microbial lethality. For this reason, in this study, the efficacy of a UV-C LED-based hand lamp was assessed at different starting microbial levels and with various UV exposure times. The corresponding UV doses used at each treatment time were also calculated. As expected, the initial microbial titer affected the %IR obtained and this effect was evident with the lowest UV dose used (9.33 mJ cm⁻²), in all the bacterial strains tested. Indeed, from the 15.55 mJ cm⁻² UV dose, the high efficacy of the UV-C LED hand-lamp led to a plateau of the %IR. Conversely, in the case of the fungal strain, for which a very high UV resistance is known (Mathur 2022), the lamp was not effective enough to lead to a high %IR, highlighting strong and statistically significant variability at all the UV doses tested, in a titer-dependent manner.

In an application context, besides the initial bacterial titer and the timing of UV-C exposure, the presence of surface organic matter (Kaur et al. 2024), and the kind of material and surface properties, especially the hydrophobicity, surface roughness, and porosity (Kim and Kang 2020), can also influence the germicidal action of the irradiation. For this reason, coupons of different materials have been artificially contaminated with a bacterial suspension to which 5% FBS was added to simulate the presence of organic matter as well as to assess the efficacy of the UV-C LED hand-lamp. The surfaces considered were stainless steel, glass, and PP, which are slightly different in several characteristics (Kim and Kang 2020, Calle et al. 2021), and which are daily used in domestic environments as food handling and food contact surfaces.

Considering the presence of organic matter on the surface, comparing the surface bacterial recovery obtained in the presence or absence of 5% FBS (Supplementary material S1), it was higher in the presence of organic soil load. These results are in line with those reported by Deckers and colleagues (Deckers et al. 2010), who demonstrated that the presence of tensioactive molecules of the organic matter added to the inoculum solutions plays a role in bacteria adherence during the RODAC sampling. On the other hand, comparing the %IRs obtained in the presence of 5% FBS and in the absence of FBS (Supplementary material S1), no significant differences were found when applying a UV dose of 15.55, 21.77, and 31.1 mJ cm⁻². Similar results were obtained by Moore et al. (2012), who reported that the presence of organic matter did not affect the decontamination efficacy of the UV-based device. Contrarly, Zwicker et al. (2022) observed how the germicidal efficacy of UV can be affected by the presence of soil load. However, this discrepancy can be explained by the different wavelengths used. In fact, in the same study using a wavelength of 222 nm, the authors obtained lower log reductions in the presence of soil load, whereas, at higher wavelengths, the differences were not observed using organic matter such as albumin, artificial sweat, and artificial wound exudate. As demonstrated by the same authors, this is due to the increased radiation depth penetration at a higher wavelength (from 254 nm), which becomes comparable to the one observed with NaCl solution (Zwicker et al. 2022).

Considering the different effects of the UV-C irradiation according to the surfaces, all the strains showed the highest susceptibility when spread in stainless steel and irradiated with a UV dose greater than or equal to 15.55 mJ cm⁻². Indeed, in stainless steel contamination tests, a UV dose of 15.55 mJ cm⁻² was sufficient to enable a %IR higher than 97% in all the strains (Fig. 2); furthermore, considering all bacteria tested strains, no colony (<1 CFU 169 cm⁻²) was obtained from stainless steel surface when it was treated with a 21.77 mJ cm⁻² UV dose, or higher (Fig. 2). Moreover, the UV-C LED hand lamp showed very good efficacy on the glass surface, leading to %IR comparable to those obtained in stainless steel in almost all the experimental conditions. Conversely, PP was the surface in which the strains proved to be more resistant, regardless of the bacterial strain tested or the UV dose impacted on the surface. Similar results have been obtained by Calle et al. (2021), in which, after contaminating stainless steel and high-density polyethylene surfaces with a strain of S. enterica, and irradiating the surfaces with UV-C light, the log reduction resulted higher in the stainless steel coupon than in polyethylene one. Particularly, the authors obtained a log reduction ranging from 1.97 to 3.48 log CFU cm⁻² in stainless steel, with UV doses of 12 and 24 mJ cm⁻² (calculated from the data reported in the manuscript. Irradiance: 2 and 4 W m⁻²; irradiation time: 60 s). These results are almost in accordance with the ones achieved in this work, with both strains of S. Infantis. The high efficiency of UV-C light in disinfecting artificially contaminated surfaces has also been assessed by other authors, in which a reduction of >5 logs was achieved with E. coli, Staphylococcus aureus, and Candida albicans (Duering et al. 2023). However, in this case, only aluminum was used as a test coupon, so data cannot be contrasted with those obtained by our study. Moreover, it is important to highlight that it is difficult to compare the available data about UV irradiation used in surface disinfection because of the wide differences in several factors of the protocols applied, as the UV doses, the starting titers used to contaminate the coupons, the irradiation times and microorganisms tested.

As suggested by other authors (Kim and Kang 2020), we can assume that the lower log reductions obtained in PP surfaces or similar materials may be due to the porosity of the coupons, which can create deep crevices. In these conditions, bacterial cells can settle, clump, and be hidden from the UV light rays (Calle et al. 2021).

Based on the data obtained, the disinfection process was found to be more efficient on the stainless steel surface. In general, this type of surface is recommended in those environments where food is handled. In fact, its low roughness reduces the adhesion of bacteria on it and makes the disinfection process more efficient (Boulané‐Petermann 1996, Di Cerbo et al. 2021). These characteristics, added to the high strength and malleability of the material, make stainless steel one of the most optimal surface materials for food processing. Compared with stainless steel, PP could be subjected to surface damage from sharp instruments, thus creating grooves within which food residues and microorganisms could harbor themselves, reducing the effectiveness of disinfection treatments. Considering glass, its low impact resistance could make it unsuitable and dangerous in environments where food is handled.

Moreover, even within the same species, genetic differences can play a significant role in influencing the inactivation of microorganisms by UV-C exposure (Soro et al. 2023). Genetic variants within a species may result in the presence of specific mutations that influence resistance ability or susceptibility to the damaging effects of UV-C radiation or could affect the production of enzymes involved in DNA repair mechanisms or oxidative stress management; these characteristics should make some strains more resistant against UV-C irradiation. Even though in this study the genetic bases of possible differences between the two strains of L. monocytogenes have not been already evaluated, we could hypothesize that their dissimilar behavior against UV-C irradiation, especially when spread on food-related surfaces’ coupons, could be linked to some genetic factors, which make L. monocytogenes 1484 more resistant. Given these results, further studies could address genetic variability potentially related to find out some possible genetic explanations.

It must be considered that a limitation of this study is that the disinfection efficacy of the lamp could be different depending on the operator. This is because variations in the timing of UV exposure, scan speeds different from those tested in the study, or a greater distance between lamp and surface could result in lower disinfection efficacy. To avoid this, we recommend following the procedure used in the present study.

Despite their high efficacy in microbial inactivation, it must be emphasized that overexposure to UV-C radiation is dangerous and harmful to the human skin, causing burns, severe forms of photokeratitis, and inflammation of the cornea. Due to this, many devices based on UV-C LED technology are equipped with safety mechanisms that prevent accidental operator exposure, such as motion sensors that turn off the LEDs when they sense a moving individual or when the device is rotated (like the device tested in this study). In addition to this, individual safety equipment (e.g. protective glasses, gloves, and UV protection face shield) should be used.

Collectively, our research has highlighted the effectiveness of a portable UV-C lamp as a disinfection tool in household settings, for food-handling surfaces. Its high effectiveness in reducing microbial load is a promising solution for improving hygiene practices and food safety in this context, reducing the spread of foodborne pathogens and the manifestation of related diseases. In addition, new data about the UV-C sensitivity of food and clinical L. monocytogenes and S. Infantis strains have been highlighted. To our knowledge, this is the first study that can compare the %IRs, D₉₀, and the k values of different microbial strains (collection and field strains), both in vitro and on food-handling-related surfaces in the presence of organic matter. In our opinion, this data can be useful as a starting point for further studies aiming to standardize the use of UV-C sources for disinfection in food-related environments.

Conclusions

The application of UV-C LED is emerging as a good alternative for surface disinfection applications and could be an easy-to-use disinfection method to be applied in domestic or small-scale food-processing environments, to reduce the spread of foodborne infections. This study demonstrated the effectiveness of UV-C LED handheld lamp (peak wavelength 265 nm) at reducing food pathogens’ microbial load on common food contact surfaces such as stainless steel, glass, and polyethylene, being a really promising approach in the prevention of foodborne diseases. It must be considered that in this study enteric viruses were not taken into account. Equally important to consider in the food and surface contamination field are the enteric viruses, which can give rise to foodborne infections, especially if contaminated water is used for washing/preparing food and cleaning surfaces (Kotwal and Cannon 2014). Their survival in the environment for long periods of time and their high tolerance to changing environmental conditions make them a public health concern (Lau et al. 2020). Future studies on the evaluation of the efficacy of these UV-C LED devices in inactivating this class of virus on different surfaces should be considered.

Acknowledgments

The authors would like to sincerely appreciate the Centro di Riferimento Regionale Patogeni Enterici dell’IZS Umbria e Marche “Togo Rosati” sede di Tolentino for providing the strains of Salmonella enterica subsp. enterica serovar Infantis (S. Infantis 43072/20 and S. Infantis 29763/20).

Author contributions

Francesco Palma (Conceptualization, Formal analysis, Methodology, Writing – review & editing), Giulia Baldelli (Conceptualization, Formal analysis, Methodology, Writing – original draft), Giulia Amagliani (Supervision, Writing – original draft), Asja Conti (Investigation, Methodology), Mauro De Santi (Formal analysis, Supervision, Writing – original draft), Giorgio Brandi (Supervision, Writing – original draft), and Giuditta Fiorella Schiavano (Conceptualization, Formal analysis, Project administration, Supervision, Writing – review & editing)

Conflict of interest

None declared.

Funding

None declared.

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

References

Author notes