Impact of factors affecting the efficacy of intense pulsed light for reducing Bacillus subtilis spores

Abstract

This study investigated how the following four intense pulsed light (IPL) treatment factors affect the inactivation of Bacillus subtilis (KCCM 11,315) spores: distance between the sample and IPL lamp (8, 13, and 18 cm), pulse width (0.5, 1.3, and 2.1 ms), charging voltage (1000, 1200, and 1400 V), and processing time (10, 20, and 30 s). The results showed that all four factors considerably influenced the spore inactivation rate in different ways. Excluding processing time, which does not affect the pulse itself, the effect was largest for pulse width, followed by distance, and charging voltage. The optimal treatment condition that maximized the inactivation rate was a distance of 8 cm, a pulse width of 2.1 ms, a charging voltage of 1000 V, and a processing time of 30 s, which together produced a 6 log reduction. It revealed that individual factors need to be investigated together for achieving the optimal condition of IPL.

Introduction

Intense pulsed light (IPL) is a nonthermal processing technique using polychromatic light for surface sterilization (Chen et al., 2019). IPL includes broad-spectrum radiation from the UV region to the near-infrared region, and the main wavelength range influencing the inactivation of microorganisms is UV-C. Chen et al. (2015) reported that compared with processing methods such as heat treatment and chemical sterilization, IPL is a rapid nonthermal treatment method that also has the advantage of leaving no residue (Bhavya and Umesh Hebbar, 2017; Mandal et al., 2020). Applying IPL treatment to foods has minimal effects on food quality and is effective in reducing spoilage and pathogenic microorganisms.

Bacillus subtilis is a spoilage and spore-forming bacteria that can cause problems if it survives at a high level after food decontamination. Spores of B. subtilis show higher resistance to sterilization processes such as heat and chemicals than plant cells, as well as ultraviolet light and radiation (Atrih and Foster, 2002; Carlin, 2011). There are several factors that affect the higher resistance of the spore. First, there are the chemical and physical substances that affect the sensitivity to stress like dipicolinic acid, α/β-type small acid-soluble proteins, and spore coat (Clair et al., 2020; Jo et al., 2019; Setlow, 2006). In particular, when light irradiation is performed when the microorganisms are in a spore state, photoproduct like pyrimidine dimers are better recovered due to their DNA repair mechanism (Setlow, 2006). Research on the mechanism of inactivation of spores is still actively underway and an in-depth study of the inactivation of B. subtilis is needed for food safety.

Various factors affect the efficacy of IPL treatment. Farrell et al. (2010) experimented with the main factors influencing IPL treatment using pathogenic bacteria isolated from patients. This study considered electrical parameters (lamp voltage and number of pulses), biological factors (population size, age, species, and pigments), and the distance of samples from the light source. Artíguez and Martínez de Marañón (2015) suggested the inherent resistance of spores and the dependence of light transmittance on cell density were major factors influencing the inactivation effects of IPL treatment. The smoothness and hydrophobicity of the sample surface affects inactivation levels due to microbial or light shading effects and cell clustering (Proulx et al., 2015). Sharma and Demirci (2003) investigated the effects of the distance between the source of pulsed UV light and samples, the sample thickness, and the treatment time on the decontamination of alfalfa seeds inoculated with Escherichia coli O157: H7. Krishnamurthy et al. (2008) used milk and milk-foam samples to test the effects of the volume/weight of the sample, the treatment time, and the distance between the sample and the lamp on the efficacy of pulsed UV-light treatment. Hwang et al. (2015) investigated the effects of optical factors such as permeability, extinction coefficient, and lightness on the inactivation effect of IPL by inoculating various liquid foods with Pseudomonas aeruginosa.

It has been found experimentally that many factors can individually affect IPL treatment. However, it is unknown how these factors influence the inactivation of spores when multiple factors are adjusted simultaneously. In this study, we investigated the effects of the distance between the sample and the IPL lamp and the lamp charging voltage, pulse width, and processing time on spore inactivation, with the aim of determining the optimal inactivation condition.

Materials and methods

Spore preparation

Bacillus subtilis KCCM 11,315 was obtained from the Korean Culture Center of Microorganisms for use as the experimental strain. This strain is a nonpathogenic spoilage and spore-forming bacteria that has a low usage risk in the laboratory environment. B. subtilis was cultured in 3 mL of Bacto™ Brain Heart Infusion (Difco™, BD, Sparks, MD, USA) broth at 30 °C for 24 h (Jo et al., 2019). Twenty microliters of the culture medium was inoculated on nutrient agar (Difco™, BD) with added sporulation salt (100 mg/L CaCl₂ and 40 mg/L MnSO₄) at 30 °C. After 7 days of cultivation the surface of the medium was scraped using a cell scraper to acquire formed spores, and 5 mL of sterilized 0.85% NaCl solution was added. A 5- mL aliquot of the scraped solution was collected and then centrifuged (406G, Gyrozen, Incheon, Korea) at 4000 rpm for 10 min. The supernatant was discarded and the pellet was washed, and then the centrifuge and washing procedure was repeated three times. After adding another 5 mL of sterilized 0.85% NaCl solution, heat treatment was applied at 80 °C for 10 min to kill the vegetative cells. Following the heat treatment, centrifugation and washing were carried out using the same method. The obtained spores were stored at –70 °C before the experiment.

The inactivation experiment using IPL was conducted after confirming that more than 99% of the B. subtilis spores had formed using the Schaeffer-Fulton Spore Stain Kit (Sigma-Aldrich, St. Louis, MO, USA). The staining method was as follows:

1. The suspension of the strain was smeared, dried, and fixed on a glass slide with mild heat.

2. The malachite green solution was dropped onto the glass slide, which was then heated with steam for 5 min. The reagent was not allowed to dry. During the heating processing with steam, the malachite green solution combined with endogenous spores to stain them.

3. The malachite green solution was rinsed. Since the vegetative cells did not exhibit strong binding with the malachite green solution, the cells were discolored during the washing process.

4. The safranin solution was added for 30 s.

5. The sample was washed with water, the moisture was removed, and then the glass slides were covered with cover glasses to enable observations under a microscope (CX21LED, Olympus Corporation, Tokyo, Japan). The spores were stained in green and vegetative cells were stained red, enabling the endospores to be distinguished from vegetative cells.

IPL treatments

IPL device

The self-designed laboratory-scale IPL treatment device consisted of a power supply, a cooling system, and a treatment chamber containing a xenon lamp which emits IPL with the 200–1100 nm range of wavelength, fans for ventilation, and a shelf whose distance from the lamp could be adjusted (Jo et al., 2019). The power supply was adjustable for processing time, pulse width (a measure of the time between the beginning and end of the pulse, typically based on the full width half maximum of the pulse shape), frequency, and charging voltage. Cooling water at 25 °C flowed around the lamp to prevent it from either overheating or cooling rapidly during the experiments.

The treatment chamber was 46 cm wide, 45 cm long, and 69 cm high, and had a front door for sample loading. The lamp was a xenon flash lamp (type NL 9553, XAP series, Heraeus Noblelight, Cambridge, UK) with a length of 27.1 cm and a diameter of 1.1 cm. The xenon lamp was located at the top of the treatment device, and the fans were located above both sides. The height of the shelf could be adjusted from 8 to 53 cm in 5-cm intervals (i.e., 10 levels). The chamber was constructed from acryl and was colored black in order to minimize the light loss.

Treatment conditions

Five milliliters of 0.85% sterile sodium chloride solution was added to the spore pellet to obtain 10⁷–10⁸ CFU/ mL. After being resuspended, the spore suspensions were serially diluted using 0.85% sterile sodium chloride solution and 0.1- mL of these solutions were spread each onto plate count agar (PCA; Difco™, BD, Sparks, MD, USA). After spreading the diluted samples on PCA plates, the untreated control samples were incubated for 48 h at 30 °C and the experimental samples were exposed to IPL under the conditions of 8, 13, and 18 cm of vertical distance between the plate and the lamp, 1000, 1200, and 1400 V of voltage, 0.5, 1.3, and 2.1 ms of pulse width, 2 Hz of frequency, and 10, 20, and 30 s of processing time. Spore inoculated plates were IPL treated within 2 min to limit the extent of germination of the spores caused by nutrients in PCA (Esbelin et al., 2016; Jo et al., 2019; Levy et al., 2012). The IPL-treated PCA plates were incubated for 48 h at 30 °C before counting the numbers of colonies formed. All experimental procedures mentioned in this session were repeated three times.

Measurement of total energy fluences

The total energy fluences (J/cm²) is mathematically calculated as the energy of each pulse (J/cm² pulse) $\times$ processing time (s) $\times$ pulse frequency (Hz) (Mandal et al., 2020). In this study, total energy fluences were measured by a spectroradiometer (ILT950, International Light Technologies, Peabody, MA, USA) covering a broad range of wavelengths from 250 to 1050 nm. The energy intensity was measured three times by placing the detector head of the spectroradiometer at the same position as where the sample was treated with IPL, and the average value was used.

Statistical analysis

The results were analyzed using MINITAB software (version 16, Minitab, State College, PA, USA) to find the optimal condition for B. subtilis spore inactivation. A 95% confidence level was used for the significance cutoff. The dependent variable was the log₁₀ reduction, and the independent variables were charging voltage, pulse width, distance, and processing time.

Fitting of the microbial inactivation model

A double Weibull model as proposed by Coroller et al. (2006) was used to fit the microbial log₁₀-reduction curves:

$$\frac{N}{N_0}=\frac{1}{1+{10}^{\alpha }}\left[{10}^{-{\left(\frac{F}{{\delta }_1}\right)}^P+\alpha }+{10}^{-{\left(\frac{F}{{\delta }_2}\right)}^P}\right]$$ 1

where F is the fluence (J/cm²), N is the CFU after IPL treatment, N₀ is the initial number of microorganisms, p is the shape parameter to determine curve concavity or convexity (p = 1, log-linear; p > 1, convex; p < 1, concave), α is the ratio between subpopulation 1 (f) and subpopulation 2 (1–f), and δ₁ and δ₂ are the fluences (J/cm²) for the first decimal reductions of subpopulations 1 and 2, respectively. Fluence was used instead of processing time as generally used by other studies. The nonlinear regression analysis tool GInaFiT (a freeware add-in for Microsoft Excel) was used to model the nonlinear log₁₀-reduction curves.

In addition, the response-surface methodology was used to perform regression analysis with OriginPro (version 9.0, OriginLabs Corporation, Northampton, MA, USA).

Results and discussion

Spore inactivation effect according to the total energy fluences

Before measuring the effects of IPL on Bacillus subtilis spore, the energy fluences of one pulse formed by control factors (distance, voltage, pulse width) was measured (Table 1). The energy fluence tended to increase with increasing pulse width and driving voltage, and decreasing distance. This trend is similar with the studies of Kramer et al. (2017) which found that the measured fluence increased when the charging voltage increased and the sample was closer to the lamp. Yi et al. (2017) also reported that the energy fluence increased with increasing the pulse width.

Table 1.

Fluences and 4D values of IPL according to treatment conditions

	TreatmentNo	Distance(cm)	Voltage (V)	Pulse width (ms)	Fluence (mJ/cm2 pulse)	4D (J/cm2)
Figure 2	1	8	1000	0.5	16.13	0.43
	2	13			12.20	1.32
	3	18			8.78	2.81
	4	8	1200		18.84	0.59
	5	13			14.14	0.87
	6	18			11.22	2.06
	7	8	1400		20.44	0.44
	8	13			16.04	0.75
	9	18			13.13	1.04
Figure 3	10	18	1000	0.5	8.78	2.81
	11		1200		11.22	2.06
	12		1400		13.13	1.04
	13		1000	1.3	14.06	1.21
	14		1200		15.99	0.92
	15		1400		17.90	0.39
	16		1000	2.1	16.69	0.90
	17		1200		18.78	0.34
	18		1400		20.26	0.44

The variations in spore inactivation values according to total fluences of IPL is shown in Fig. 1. The inactivation level increased with the total fluences, and 8 log reductions were achieved after 6 J/cm² of total fluence. Aguirre et al. (2015) treated B. cereus spores on agar medium using pulsed light, and they obtained 6 log reductions at total fluences of 2.1 J/cm². Levy et al. (2012) obtained 5 log reductions of B. subtilis spores at total fluences of 1.25 J/cm², respectively. When comparing the results obtained by other studies, this present study obtained similar results, in which the spores were reduced by 5, 6, and 7 log with 1.13, 2.93, and 4.39 J/cm², respectively. When vegetative microorganisms are exposed to IPL, it is generally recognized that photophysical effects such as membrane damage and protein elution, and photochemical effects such as dimers formation, occur in damaged cells (Bhavya and Umesh Hebbar, 2017; Krishnamurthy et al., 2010; Pollock et al., 2017). These mechanisms could also be applied to spores, but spores show lower susceptibility than vegetative microorganisms due to the formation of the spores coat and inherent DNA repair mechanisms (Anderson et al., 2000; Nicholson et al., 2000; Setlow, 2006). Levy et al. (2012) specifically stated that the energy required for the inactivation of spores is 18 times more than that of vegetative cells. Comparing with other non-thermal processes such as pulsed electric field (PEF) and high pressure processing (HPP), however, IPL effectively inactivates spores at a lower level of energy expenditure (Mandal et al., 2020; Rahman, 2007).

Fig. 1.

Inactivation of B. subtilis spores as a function of total fluences. Data are mean and SE values

Inactivation curves fitted with the double Weibull model

The inactivation curves of spore after IPL treatment do not show linear relationship. Coroller et al. (2006) tried to describe the inactivation curve more accurately using several Weibull models, and many studies use double Weill model to explain microbial inactivation mechanisms or kinetics after IPL treatment (Couvert et al., 2005; Crook et al., 2015; Ferrario et al., 2013; Levy et al., 2012). When the microorganisms exposed to the IPL, they had resistance first until the IPL exceeds a particular intensity threshold (shoulder), and over this threshold, they showed a rapid reduction. After that, a tail occurred. Tailing presents due to various reasons such as lack of homogeneous population or repair of DNA damage (Hwang et al., 2019). In this study, the double Weibull model was used to investigate the energy fluence and inactivation rates and to obtain the 4D values, corresponds to the fluences required for a 4 log reduction, of each different treatment conditions composed of various factors.

For the same pulse width, as shown in Table 1, a shorter distance and higher voltage required smaller fluences to achieve 4 log reduction. Moreover, for the same distance, the 4D value decreased as the voltage and pulse width increased. In this table, the fluences corresponding to various conditions that are numbered as 1 and 8 showed similar value (16.13 and 16.04 mJ/cm² per pulse, respectively). However, their 4D value was quite different, 0.43 and 0.75 J/cm², respectively. Converted to the number of pulses, they are represented as 26.66 and 46.76, respectively, and their difference is almost 1.75 times. In addition, in case of treatment number 4 and 17, their fluence were similar as 18.84 and 18.78 mJ/cm² per pulse, respectively. However, their 4D value were 0.59 and 0.34 J/cm², respectively and their difference in converted number of pulses (31.32 and 18.10, respectively) is also almost 1.73 times. It means that since even pulses with a similar degree of fluence are determined by a combination of several factors, the total number of pulses applied to achieve a specific level of inactivation will depend on the effect of each factor on the actual inactivation.

To confirm this hypothesis, inactivation curves of B. subtilis spore were derived using various conditions of IPL (Figs. 2 and 3). Overall, they showed that the rate of inactivation of spores increased with increasing voltage, pulse width and processing time (number of pulses), and decreasing distance. When the charging voltage and pulse width were fixed, the inactivation rate increased as the distance decreased (Fig. 2(A)–(C)). The inactivation rate was lowest for a distance of 18 cm (the longest distance between the lamp and the sample) and highest for the shortest distance of 8 cm. At distances of 18 cm it took more than 6 min to reach a 7 log reduction, while this was achieved within 4 min for a distance of 8 cm.

Fig. 2.

Inactivation of spores treated with IPL at different charging voltages: (A) 1000 V, (B) 1200 V, and (C) 1400 V and distance: (D) 8 cm, (E) 13 cm, and (F) 18 cm, with the same pulse width of 0.5 ms. Data are mean and SE values

Fig. 3.

Inactivation of spores treated with IPL at different charging voltages: (A) 1000 V, (B) 1200 V, and (C) 1400 V and pulse width: (D) 0.5 ms, (E) 1.3 ms, and (F) 2.1 ms, with the same distance of 18 cm. Data are mean and SE values

When the distance and charging voltage were fixed, the mortality rate increased with the pulse width (Fig. 3(A)–(C)). The inactivation rate was highest for a pulse width of 2.1 ms, reaching a 7 log reduction within 4 min under a treatment condition of 18 cm, 1400 V, and 2.1 ms. For a fixed distance and pulse width, the inactivation effect increased with the charging voltage (Figs. 2(D)–(F)) and 3(D)–(F)). The inactivation rate was highest at the highest charging voltage of 1400 V.

The inactivation curves were fitted to the double Weibull model, and as the voltage, pulse width, and processing time increased, and the distance decreased, the shoulder and tail of the curve were shortened or hard to figure out. As mentioned in Sect. 3.1, the intensity of IPL also can be determined by combined effects of charging voltage, pulse width, distance, and processing time. Through this, it could be estimated that these factors affect the energy fluence of the IPL and when the spore is exposed to strong conditions, it is more difficult to adapt to or tolerate the IPL. Kramer et al. (2017) also found that the inactivation rate increased with decreasing distance and increasing lamp voltage. In experiments with Enterobacter sakazakii, the inactivation rate tended to increase with the voltage and treatment time (Choi et al., 2009). In addition, Artíguez and Martínez de Marañón (2015) found that a higher energy fluence of IPL produced a greater inactivation effect.

These previous studies have focused on microbial inactivation according to changes in treatment conditions to maximize the log reduction. However, since IPL is a light sterilization technique, and the intensity of the light is complexly determined by several factors. Also, even with the same intensity, there may be a large difference in the microbial inactivation effect depending on the influence of each factor. For example, as shown in Fig. 2(A), 3 J/cm² of total fluences can be made by setting distance and processing time as followed three types; (1) 8 cm – 3.10 min, (2) 13 cm–4.10 min, and (3) 18 cm–5.69 min (the voltage and the pulse width were fixed as 1000 V and 0.5 ms, respectively). These three conditions have same total fluences but their spore inactivation level were about 6.0, 5.5, and 4.0, respectively. It means that narrowing the distance has more impact in inactivation than increasing the number of pulses if the same total fluences has to be used. In order to identify to what extent the factors influence log reduction, the main effect plot was identified using MINITAB. The magnitude of the effect of each factor on the inactivation level can be inferred by the slope obtained by the main effect plot. In case of this study, processing time showed the most influence on the inactivation level, followed by the pulse width, distance, and voltage. If the processing time, which does not affect the pulse itself, is excluded here, it can be concluded that the pulse width of one pulse has the most impact on the inactivation level. As mentioned above, previous studies figured out various factors of IPL which affects the microbial inactivation. In contrast, this present study has compared how much various treatment factors affect the log reduction of spore. To increase the inactivation rate while reducing energy usage, it is effective to control pulse width, distance, and voltage in order.

Response surface model of spore inactivation

Response surface plots of spore inactivation according to the treatment conditions are shown in Fig. 4. As mentioned before, when the stronger energy is delivered to the spore, the shoulder and tail were shortened or hard to figure out in their inactivation curve. As shown in Fig. 4(C), processing time is a powerful condition. Thus, spores could be inactivated effectively. However, it is difficult to conduct a regression analysis clearly that explains the effects of each factor on the reduction of B. subtilis spores. Excluding this data, regression equations describing Fig. 4(A) and (B) using the pulse log reduction (L, log), width (P, ms) and voltage (V, V) were yielded in the following order:

$$L=8.40048-0.05071P-0.01228V+0.21547P^2+4.5112910^6{V}^2-0.00137PV$$ 2

$$L=1.09835-0.42285P-0.00491V+0.21152P^2+1.2741710^6{V}^2-0.00118PV$$ 3

Fig. 4.

Response surface plots of the log reduction as a function of pulse width and charging voltage with fixed distance (18 cm) and processing time: (A) 10 s, (B) 120 s, and (C) 360 s

The adjusted R² value for these equation were 0.89 and 0.93, respectively. It was identified from the regression model that both individual factors and the interaction effects were significant (p < 0.05) for estimating spore inactivation.

In combination with the analysis results in Sect. 3.2, optimal conditions to achieve 6 log reduction of B.subtilis spore using four factors, distance, pulse width, voltage, and processing time were analyzed using MINITAB, and the results were respectively 8 cm, 2.1 ms, 1000 V, and 30 s. Within the set experimental conditions, this result could be obtained because the main effect of charging voltage was the lowest among the four factors and the optimal condition was determined when all four factors were considered. In conclusion, considering the processing factors affecting IPL treatment can help to maximize the reduction effect. When applying IPL in the food industry, consideration of treatment factors and the establishment of the optimal sterilization condition would allow the design of sterilization systems that meet both quality and safety requirements.

In conclusion, IPL treatment is an effective method for surface sterilization aimed at reducing or even eliminating the spores of putrefactive bacteria. This study investigated the inactivation effects on B. subtilis spores according to various IPL treatment factors. Distance, pulse width, charging voltage, and processing time all crucially affected the inactivation of spores. Excluding processing time, which did not affect the pulse itself, the pulse width had the greatest effect on spore inactivation, followed by distance, and charging voltage. Response surface analysis was used to determine the conditions under which the inactivation rate was maximized when all factors were considered. The inactivation rate increased with a shorter distance, a higher pulse width and charging voltage, and a longer processing time. This study has confirmed that the levels of resistant spores can be markedly reduced by using IPL, which is a nonthermal sterilization technology. The optimal treatment effect will be obtained by considering the influence of the inactivation factors. Sterilization is essential to ensure the quality control of food products, and IPL is demonstrably a useful method for spore inactivation.

Declarations

Conflict of interest

There is no conflict of interest between the authors of this manuscript.