Investigation of Antimicrobial Activity of Grapefruit Seed Extract and Its Application to Air Filters with Comparison to Propolis and Shiitake

Antimicrobial air filters using a new natural product agent, grapefruit seed extract (GSE) have been investigated for application in air purifiers or heating, ventilation, and air-conditioning (HVAC) systems. The disk diffusion method was used to evaluate the antimicrobial activity of GSE, propolis, and shiitake against four bacteria: Staphylococcus aureus, Micrococcus luteus, Psedomonas aeruginosa, and Escherichia coli. The inactivation of S. aureus was then investigated on air filters treated with two natural products, GSE and propolis (selected for comparison to GSE) using two test methods based on different deposition weights. GSE displayed the highest antimicrobial activity against S. aureus, P. aeruginosa, and E. coli of the three natural products tested by the disk diffusion method. Similar inactivation rates of 58–67% for GSE and propolis air filters were observed at a relatively low deposition weight of 94–98 μg/cm using the aerosol deposition method. However, the inactivation rate was considerably superior on the GSE air filter (~98%) compared to the propolis air filter (~75%) at deposition weights of 5000–8000 μg/cm using the film attachment method. The inactivation rate for both GSE and propolis air filters could be expressed as an exponential function [in the form of 1 – exp(–ax)] of the deposition weight per unit area of the natural product. The superior inactivation performance of the GSE air filter using the film attachment method was probably due to the high wettability of GSE for bacteria cultures, with a contact angle less than 20°. Therefore, GSE air filters will be more effective and economical than propolis air filters due to their high microbial activity and relatively low price.


INTRODUCTION
Bioaerosols are defined as airborne particles of biological origin consisting of bacteria, fungi, fungal spores, viruses, yeast, and pollen and their fragments, which include various antigens (Kalogerakis et al., 2005).Exposure to these bioaerosols in indoor environments has become a serious concern due to the wide range of adverse health effects, including contagious infectious diseases (Dales et al., 1991), hypersensitivity pneumonitis (Siersted and Gravesen, 1993), and respiratory problems such as asthma and rhinitis (Beaumont, 1988).Due to their presence in nature, exposure to bioaerosols is almost inevitable in most enclosed environments (Jones and Harrison, 2004;Ren et al., 1999).
Various techniques have been used to mitigate the viability of biological contaminants in indoor environments, such as ultraviolet (UV) irradiation (Lin and Li, 2002), air electric ion emission (Huang et al., 2008), ozone (Huang et al., 2012), and photocatalytic oxidation (Chen et al., 2010), among others.One promising method is an antimicrobial filtration system that physically captures bioaerosols on air filters and then inactivates the airborne microorganisms on the surface of the filters by treatment with antimicrobial materials such as silver or titanium-based nanoparticles (Li et al., 2006;Jin et al., 2007,), iodine powders (Lee et al., 2008), and carbon nanotubes (Brady-Estévez et al., 2008;Jung et al., 2011a).
Natural products are promising antimicrobial agents for indoor air quality improvement because they are typically less toxic compared to other chemicals or nanomaterials (Carson et al., 2006;Pibiri et al., 2006;Hwang et al., 2012).For example, Arabidopsis, rice, corn, soy bean, and the model legume Medicago truncatula are rich sources of antimicrobial indole, terpenoid, benzoxazinone, and flavonoid/isoflavonoid natural products (Dixon, 2001).However, only a few trials have been conducted to apply these natural products as antimicrobial agents in air filtration systems.Huang et al. (2010) investigated Melaleuca alternifolia (tea tree oil) as a disinfecting media for the inactivation of common environmental fungal spores on a filter surface and found that 50% of Aspergillus niger and 40% of Rhizopus stolonifer spores could be inactivated over a period of 60 min.Jung et al. (2011b) investigated a natural plant extract from Sophora flavescens for application to antimicrobial air filters and reported that the prepared natural product on an antimicrobial filter was effective for inactivating Gram-positive bacteria such as Staphylococcus epidermidis and Bacillus subtilis.Lee et al. (2013) investigated the antimicrobial effects of a natural extract of Mukdenia rossii and unipolar ion emission on air filters for use against S. epidermidis and noted that the inactivation rate was about 70% using natural extract-treated filters and was enhanced by a further 20% due to unipolar ion emission.
Shiitake, one of the most popular edible mushrooms in the world, is claimed to have antitumor, antiviral, and antimicrobial properties, as well as hypocholesterolemic and hypoglycaemic actions (Hearst et al., 2009).Propolis, a hard and resinous natural product derived by bees from plant juices, exhibits antibacterial activity due to its high flavonoid content (Benhanifia et al., 2013;Burdock, 1998;Kujumgiev et al., 1999).Grapefruit seed extract (GSE), a commercial product derived from the seeds and pulp of grapefruit, also has powerful antimicrobial properties (Lim et al., 2010) mainly due to the polyphenolic compounds it contains (Saito et al., 1998;Aloui et al., 2014).However, to our knowledge, no trial using antimicrobial air filters prepared with shiitake, propolis, and GSE natural products has been previously reported.
In this study, for the first time, antimicrobial air filters using new natural product agents, grapefruit seed extract (GSE) and propolis have been investigated for application in air purifiers or HAVC systems.The antimicrobial activities of three natural products, shiitake, propolis, and GSE were compared using a disk diffusion method for four test bacteria.The application to antimicrobial air filters treated with natural products was then performed using two natural product agents: GSE and propolis.GSE was selected because it had the highest antimicrobial activity among the three natural products and propolis was selected for comparison to GSE.The inactivation of Staphylococcus aureus on the antimicrobial air filters treated with natural products was investigated at a variety of deposition weights using two test methods with different orders of deposition weight.

Materials
Grapefruit seed extract (GSE, DF-100) prepared by grinding the grapefruit seed and juiceless pulp, then mixing with glycerine, was purchased from the Food Additives Bank Co., Ltd.(Anseong, Korea).Naringin, one of the major antimicrobial contents of GSE, was 0.51%(w/w) in the as-purchased sample.Propolis was acquired from Seoul Propolis Co., Ltd.(Seoul, Korea).CAPE (caffeic acid phenethyl ester) is known to be one of the main components of propolis.18 ppm of CAPE was measured in the as-obtained propolis sample according to the manufacturer.Shiitake mushroom (L.edodes) extract was obtained from YD Global Life Science (Seoul, Korea).The mushroom extract was obtained from dried mushroom caps through ethanol extraction.The extraction process started with 100.12 g of dried mushroom caps in the market (Mungyeong, Korea) and the extract was 9.09 g.Lentinan is one of the most active 1,3/1,6-β-Dglucans and its amount in the dried Japanese L. edodes was reported as 10~15 mg/g (Minato et al., 1999).
Four bacteria, S. aureus, Micrococcus luteus, Psedomonas aeruginosa, and Escherichia coli were prepared for antimicrobial activity tests of natural products using the disk diffusion method.S. aureus, which causes a diverse array of life-threatening infections and is capable of adapting to different environmental conditions (Lowy, 2003), and M. luteus, which is a nonmotile, spherical, saprotrophic bacterium that colonizes the human mouth, mucosae, oropharynx, and upper respiratory tract (Mauclaire and Egli, 2010), were selected as Gram-positive bacteria.P. aeruginosa, which is a typical opportunistic pathogen that colonizes the lungs of cystic fibrosis patients and causes severe infections in immunocompromised hosts (Köller et al., 2000), and E. coli, which is the most commonly isolated microorganism from human primary urinary tract infections (Hull et al., 1981), were selected as Gram-negative bacteria.E. coli (ATCC 11775) and P. aeruginosa (ATCC 10145) were cultured in nutrient broth at 37°C and M. luteus (ATCC 4698) was cultured in nutrient agar at 30°C for 24 h.They were provided by the Korean Research Institute of Bioscience and Biotechnology.
S. aureus was selected as a test bacterium for evaluation of the natural product antimicrobial filters because of its high persistence despite the use of effective antimicrobials (Mylotte et al., 1987).S. aureus KCTC 1621 (=ATCC 25923) was purchased from the Korean Collection for Type Cultures (KCTC, Korea).S. aureus was cultured in nutrient broth (Difco TM Nutrient Broth 234000) or nutrient agar (Difco TM Nutrient Agar 213000) at 37°C in an incubator.For nebulization, the concentration of bacterial suspension was set to about 1 × 10 8 CFU/mL by optical density.The sterile water was prepared by autoclaving 20 mL of triple distilled water for 15 min at 120°C.For the film attachment method, a 1000-fold dilution was prepared to make the concentration 1 × 10 5 CFU/mL.

Antimicrobial Tests Disk Diffusion Method
The disk diffusion method was conducted according to the standard method reported by Bauer et al. (1966) and was used to compare the antibacterial activities of three natural product extracts.Natural product extracts were diluted with an ethanol-water solution containing 95% ethanol (Samchun Pure Chemical Ind.Co., Ltd., Pyeoungtaek, Korea) with a concentration of 25% (w/v).The four bacteria were swabbed uniformly across petri plates, which were autoclaved at 120°C for 15 min.Three filter paper disks impregnated with three natural product extracts and one filter paper disk filled with a 25% ethanol aqueous solution as a control paper were placed on the surface of the agar.The plates were incubated at 5°C for 1 h to permit good diffusion and then transferred to an incubator at 37°C for 24 h.Antimicrobial activity was recorded by measuring the width of the clear inhibition zone around the disk using calipers.

Aerosol Deposition Method on Antimicrobial Filters
The aerosol deposition method on antimicrobial filters has been performed in many other studies (Eninger et al., 2009;Jung et al., 2011b;Miaśkiewicz-Peska and Łebkowska, 2011).Two natural product extracts, GSE and propolis, were diluted with 95% ethanol (Samchun Pure Chemical Ind.Co., Ltd.) with a concentration of 0.6% (w/v).The diluted natural product extracts were sprayed with a singlejet collision nebulizer (BGI Inc., Waltham, MA, USA) supplied with an airflow of 1 L/min and then passed through a diffusion dryer containing activated carbon to remove ethanol from the sprayed droplets.The aerosol particle sizes for the GSE and propolis produced via nebulization were similar, with a mode diameter of 0.63-0.68μm when measured with an aerodynamic particle sizer (APS, model 3321; TSI, Shoreview, MN, USA).The dried natural product particles were deposited on test filters composed of polyethylene terephthalate (PET) with a diameter of 13 mm (fiber diameter of 1.5 μm, thickness of 0.45 mm) for 60-120 s. S. aureus droplets were generated by a single-jet collision nebulizer and passed through a diffusion dryer containing silica gel to remove water vapor.The dried bioaerosols were then deposited on the natural product antimicrobial filters for 5 min.The PET filter had a singlepass particle collection efficiency of 98-99% at a face velocity of 0.2 m/s for the S. aureus bioaerosols.The filters deposited by the bioaerosols were extracted into a 5 mL buffer fluid consisting of phosphate-buffered saline (PBS) with 0.05% (v/v) Tween 80.The extraction fluid was vortexed for 2 min, agitated in an ultrasonic bath for 10 min, and then vortexed once more for 2 min.The extraction fluid was diluted with PBS at ratios of 1, 10, and 100 times, placed on a nutrient agar and then incubated for 24 h at 37°C.The numbers of CFU formed on the nutrient agar were counted after incubation.The inactivation rate was defined as Inactivation rate (%) = (1 -CFU/CFU control ) × 100 (1) where CFU control is the number of colonies recovered from the control filter on which no antimicrobial treatment using natural products was performed.

Film Attachment Method on Antimicrobial Filters
The antimicrobial properties of the filters were also determined by the film attachment method (Kawakami et al., 2008;Shin et al., 2011), which is a Japanese industrial standard test (JIS Z 2801(JIS Z :2000)).Two natural products, GSE and propolis, were diluted with 95% ethanol with a concentration of 0.6% (w/v).Air filters with a size of 40 × 40 mm 2 were uniformly coated with the natural products with a spray gun at an airflow of 5 L/min.We used two kinds of air filters (#56 and #90), which are used as pre filters (non-woven PET, MERV 10 class, fiber diameter of 30 μm, thickness of 0.26 mm) and medium filters (meltblown PP, H13 class, fiber diameter of 30 μm, thickness of 0.55 mm; 3AC ltd Co. Seoul, Korea) in commercial indoor air purifiers, respectively.They had a single-pass particle collection efficiency of 56% and 90%, respectively at a face velocity of 0.2 m/s for the S. aureus bioaerosols.The pre filter (#56), which is probably little suitable for bioaerosol filtration owing to its low collection efficiency, was selected to compare the antimicrobial performances of the filters which have different particle collection efficiencies.
The test samples were placed in petri dishes and inoculated with 0.4 mL of a bacterial culture containing 1×10 5 CFU/mL.The inoculum was covered with a polyester film (X-131 transparent copier film; Folex Imaging, Solihull, West Midlands, UK), and the petri dishes were incubated at 37°C for 24 h in a humid chamber to prevent desiccation.After the incubation period, 20 mL of extraction solution [0.1% (v/v) Tween 20, 145 mM sodium chloride, 20.5 mM sodium phosphate; pH 7.4] was added to the petri dishes, which were then shaken for 2 min.Subsequently, serial dilutions of the extraction solution were spread on agar plates in triplicate and incubated at 37°C overnight.Colonies were counted visually, and the numbers of CFU per sample were determined.The activity value was calculated using Eq. ( 1) to compare the results between the two different methods rather than using the original JIS standard tool, in which it is obtained by subtracting the log value determined for the test sample from the log value determined for the control.

RESULTS
Fig. 1 shows the size of the clear zones produced by the three natural products against the four bacteria.The size of the clear zone on control filters was approximately10 mm.GSE had the highest antimicrobial activity for both Grampositive and Gram-negative bacteria, except M. luteus.Of the three natural products, GSE produced a clear zone with the largest size for S. aureus, P. aeruginosa, and E. coli, and had the highest antimicrobial activity against S. aureus.Shiitake and propolis displayed little antimicrobial activity against Gram-negative bacteria, whereas they displayed a similar activity to GSE against Gram-positive bacteria.
Fig. 2 shows the change in the size of the clear zone for the action of GSE against three bacteria at different GSE concentrations [from 50 ppm to 250,000 ppm (25%)].The size of the clear zone increased proportionally to the GSE concentration, and was larger in the order of P. aeruginosa < E. coli < S. aureus at GSE concentrations larger than 5000 ppm (0.5%).Antimicrobial activity against S. aureus increased continuously with GSE concentration, whereas activity against E coli and P. aeruginosa was almost saturated at a GSE concentration of 5000 ppm.
Fig. 3 shows the size of the clear zones observed in this study compared to previous studies.The sizes of the clear zones of propolis against E. coli and S. aureus in the study of Rahman et al. (2010) were compared to those of GSE in our study and are shown in Fig. 3(a).Similarly to the result in Fig. 2, the sizes of the clear zones resulting from the action of GSE were larger than those for propolis at a given concentration of natural product, even after considering the different sizes of clear zones on the control filter papers (10 mm in this study and 7-8 mm in Rahman et al., 2010).At concentrations larger than 1000 ppm, the antimicrobial activities of GSE against E-coli (Gram-negative bacteria) were superior to those of propolis, which is similar to the result shown in Fig. 1.The sizes of the clear zones produced by a honey (Baidhyanath honey) against E. coli and S. aureus in Chute et al. (2010) were also compared to those of GSE in our study, as shown in Fig. 3(b).The initial size of the clear zone on control filter papers in Chute et al. (2010) was also about 10 mm.GSE at a concentration of 25% had a higher antimicrobial activity against S. aureus compared to the honey at a concentration of 40%, while it had a similar activity against E. coli to the honey at a concentration of 40%.Fig. 4 shows SEM images of filter #56 and filter #90 with GSE and propolis deposited for film attachment tests.The liquid phases of GSE and propolis were coated on to the filters and then dried on the surfaces, producing different shapes compared to other studies in which natural products were deposited on the filters in the form of already dried isolated particles (Jung et al., 2011b;Lee et al., 2013).As seen from Fig. 4, no significant morphological difference existed between the filters coated by GSE and propolis.
Figs. 5(a) and 5(b) show the inactivation rates of air filters containing GSE and propolis deposits against S. aureus for different colony numbers in the control filters obtained by the aerosol deposition and film attachment methods, respectively.In the aerosol deposition method, the inactivation rates of air filters were similar for both GSE and propolis at deposition weights of 94-98 μg/cm 2 , and they decreased as the colony number increased on the control filter.However, in the film attachment method, the inactivation rate of the propolis air filters decreased from 78 to 67-69% as the colony number increased on the control filter from 5000 to 20,000-25,000 at a high deposition weight of 5000 μg/cm 2 , whereas the inactivation rate of the GSE air filters was maintained at more than 95% at a high deposition weight of 8000 μg/cm 2 , even when the colony number on the control filter was increased from 5000 to 20,000-25,000.The deposition quantities of GSE and propolis in this study were as similar or quite low for use in air filters as compared to previous researches (75-300 μg/cm 2 in Jung et al. (2011b), 0.9 g/cm 2 in Huang et al. (2010)).Little difference was observed in the inactivation rates between filter #56 and filter #90 coated with GSE and propolis, respectively.Fig. 6 shows the inactivation rate of air filters containing GSE and propolis deposits against S. aureus at different deposition weights per unit area for both the aerosol deposition and film attachment methods.Here, PET filters of 13 mm in diameter and melt-blown PP filters (#90) with a 40 × 40 mm 2 were used in the aerosol deposition method and film attachment method, respectively.The colony number in the control filters was fixed to be about 1 × 10 4 .The inactivation rate of the GSE and propolis air filters increased with the deposition weight per unit area of natural products and represented a typical exponential function of 1 -exp(-ax b ).Similar to the result based on the disk diffusion method, the inactivation rate of GSE air filters was higher than for the propolis air filters at the higher deposition weights per unit area.Fig. 7 shows the contact angles of a bacterial culture droplet on the surface of the control, propolis, and GSE air filters.The contact angles of the droplet on the control filter and propolis-deposited filter were 122° and 93°, respectively.However, the angle was smaller than 20° on the GSE air filters, on which the bacterial culture quickly soaked in as it contacted the filter without a film for attachment.This indicates that a bacteria culture can be easily contacted to GSE natural product, which probably leads to an enhancement of the antimicrobial activity of the GSE air filter because the wettability of the antimicrobial filter surfaces is known to be highly related to its antimicrobial activity in previous literatures (Wang et al., 2008;Necula et al., 2009;Compagnoni et al., 2012;Agrawal et al., 2013).Therefore, the high wettability of the bacteria culture, as well as the inherent superior antimicrobial activity of GSE, ensured that the GSE natural product was an excellent agent for application to antimicrobial air filters.Furthermore, GSE is preferable for air filters because the cost is one or two orders of magnitude lower (several US dollars per 100 mL) than propolis, and it can be deposited on air filters with higher deposition weights at a set price.

DISCUSSIONS
To investigate the possibility of the application of the developed natural products on air filters, pressure drop and particle collection efficiency of the natural product-treated air filters were compared to the control filters.Pressure drops across the PET control filter used in the aerosol deposition method were 56.1 ± 1.7 Pa at a face velocity of 0.2 m/s.Whereas, pressure drops across the prepared GSE and propolis air filters with a deposition weight of 94-98 μg/cm 2 were 56.1 ± 2.0 Pa and 60.7 ± 3.4 Pa, respectively.Therefore, the increase of the pressure drop was negligible for the GSE air filters and only 8.2% for the propolis air filters compared to the control filter.Particle collection efficiency for the S. aureus bioaerosols was 98.6% for the control filter and 98.2%, 96.5% for the prepared GSE and propolis air filter, respectively.There was little change in the particle collection efficiency of the prepared GSE and propolis air filters compared to the control filter.Furthermore, the effect of life-time and clogging for the natural product antimicrobial filter is very important for application to air filters in HVAC systems or air cleaners.According to the previous antimicrobial durability study using Sophora flavescens (Chong et al., 2013), the antimicrobial activity remained stable for 90 days and however, rapid degradation of the activity owing to the noticeable decrease of the quantities of major antimicrobial compounds.It provides useful information regarding the average life expectation of antimicrobial air filters using Sophora flavescens but does not be generally prescribed for all natural products.For the case of GSE and propolis, it is necessary to investigate the effect of life time or clogging on the antimicrobial performances of the natural product filters, which will be prepared as our next study.

CONCLUSIONS
The antimicrobial activity of three natural products, grapefruit seed extract (GSE), propolis, and shiitake against four bacteria, S. aureus, M. luteus, P. aeruginosa, and E. coli, were compared using a disk diffusion method.Inactivation tests against S. aureus for the antimicrobial air filters treated with two natural products, GSE and propolis, were then performed at various deposition weights using two methods (aerosol deposition and film attachment).Of the three natural products, GSE displayed the highest antimicrobial activity against three bacteria, S. aureus, P. aeruginosa, and E. coli, and among the four bacteria, it had the highest antimicrobial activity against S. aureus.In the aerosol deposition method, the inactivation rates against S. aureus for GSE and propolis air filters were similar at the same deposition weights (< 100 μg/m 2 ).However, in the film attachment method, the inactivation rate of the GSE air filter was higher than that of the propolis air filter at the same deposition weights (> 1000 μg/m 2 ) due to the higher wettability of GSE air filters for bacteria culture.Therefore, GSE air filters will be more effective and economical than propolis air filters because of their higher microbial activity and relatively lower price.

Fig. 1 .Fig. 2 .
Fig. 1.Comparison of the size of clear zones of the four bacteria for the control paper and three natural products.

Fig. 3 .
Fig. 3. Comparison of the size of clear zones of E. coli and S. aureus for the GSE natural product in this study to those of previous studies: (a) Rahman et al. (2010) and (b) Chute et al. (2010).

Fig. 5 .
Fig. 5. Inactivation rates of S. aureus for the antimicrobial filters using GSE and propolis in different test methods: (a) aerosol deposition; (b) film attachment.

Fig. 6 .Fig. 7 .
Fig.6.Redrawing of the inactivation rates for the antimicrobial filters using GSE and propolis according to deposition weight per unit area in the two test methods.