Effect of Treatment with a Natural Extract of Mukdenia Rossii ( Oliv ) Koidz and Unipolar Ion Emission on the Antibacterial Performance of Air Filters

Airborne microorganisms are deposited on air filters when these are used to reduce aerosol concentrations in indoor environments. These microorganisms cause various undesirable effects, such as production of unpleasant odors and resuspension of bacterial and fungal spore bioaerosols into the indoor air. To resolve these issues, we investigated the antimicrobial effects of a natural extract of Mukdenia rossii (Oliv) Koidz and unipolar ion emission against bioaerosols deposited on air filters. The experimental results showed that the natural extract had an antimicrobial effect on the deposited bioaerosols, and that unipolar ion emission enhanced the antimicrobial performance of the natural extract.


INTRODUCTION
Indoor environmental conditions affect human health directly because people spend most of their life time indoors, such as in apartments, shopping malls, and public buildings.Therefore, concerns over indoor air quality, and the demand for artificial air conditioning and filtration of indoor air are rapidly increasing since people have become conscious of environmental diseases (Douwes et al., 2003).Air filtration by capture of aerosol particles is the representative method for improving indoor air quality (Lee, 2011).When aerosol particles are captured by air filters, airborne biological particles, such as bacteria and fungi, are simultaneously deposited onto the surface of the filters.These biological particles maintain their viability and grow on the filter under humid conditions.Furthermore, they multiply and spread debris in the air with producing rank odors.Therefore, development of an air filtration mechanism that can prevent the growth of biological particles on filters is necessary.Air filters treated with anti-microbial components such as iodine (Eninger et al., 2008;Lee et al., 2008b) and membranebreaking enzymes have been developed and tested.Nevertheless, many non-biological aerosol particles are present in the air, and when non-biological dust is deposited onto the surface of a filter, the attached anti-microbial component on the filter fibers are instantly covered by dusts and their effects are nullified.Therefore, typically, antimicrobial filters have short, effective anti-microbial activity owing to the accumulation of non-biological dust.A variety of methods, such as the use of airborne silver nanoparticles (Lee et al., 2008a(Lee et al., , 2010) ) and artificial ion emission (Lee et al., 2004(Lee et al., , 2005;;Huang et al., 2008;Park et al., 2009), have been suggested for resolving the problem of microbial contamination of air filters.
In this study, we provide a new approach involving the treatment of air filters with a leaf extract of Mukdenia rossii (Oliv) Koidz and an emission of positive or negative unipolar ions.

Experimental Setup
Fig. 1(a) shows a schematic diagram of the experimental setup for testing air filters.This system comprised an aerosolizer, an ion emitter, and a test filter.
In the aerosolizer, filtered air passed through the nebulizer (Microbiological Research Establishment 6 jet, Collision nebulizer, BGI Inc., MA, USA) at a rate of 5 liter per minute (LPM) and the filtered air aerosolized particles of a suspension of the nebulizer into the airflow.The particles were carried to the test filter for deposition on an air filter.
In the ion emitter, we used a carbon fiber ionizer (Sejin Electronics Inc., Seoul, Korea), which generated positive or negative ions.This ionizer was combined with airflow systems to generate unipolar ionic flows in the test filter.
In the test filter, aerosol particles or unipolar ions were propelled for deposition on the air filter.We used a filter with medium-sized pores (Cleanguard respirator, Yuhan-Kimberly, Korea) as a test filter, because the respirator with this filter was one of the most popular respirators during outbreaks of air infection such as the influenza A H1N1 2009 outbreak.Fig. 1(b) shows a respirator and the nonwoven fabric filter inside the respirator.We used the non-electrect filter with 99.15% collection efficiency for 0.3 μm aerosol particles.The filter consists of fibers that have a diameter of 4 μm, and the thickness of the tested filter was approximately 100 μm.
The complete experimental procedure was as follows.
(1) We tested untreated (raw) filters by using bacterial bioaerosols.We enumerated the number of bacterial bioaerosols that were deposited and maintained the culturability on the untreated filters with medium-sized pores.
(2) We deposited the antimicrobial natural extract on the test filters by the aerosolization of its particles.Next, we tested treated filters by using bacterial bioaerosols and evaluated their antimicrobial performance.(3) Unipolar ionic flows (of positive or negative ions) were introduced into the filters treated with the extract.Unipolar ions were emitted onto the filters after the deposition of bacterial bioaerosols on the extracttreated filters.We enumerated the number of bacterial bioaerosols deposited on these filters and evaluated the filter's antimicrobial performance.(4) To observe the effect of the sequence of unipolar ion emission and bioaerosol deposition, we switched the order in which the ion emission applied and bioaerosol deposited.We emitted unipolar ions onto the extracttreated filters, and then introduced bioaerosols on the filter systems.Thus, in this case, the bioaerosols were simultaneously exposed to the natural extract and to unipolar ions.Under all experimental conditions, more than three replications were conducted to confirm the results.

Tested Bacterial Bioaerosols, Determination of Bioaerosol Concentration, and Natural Extract
We used Staphylococcus epidermidis (KCTC1917) as a test microorganism in the experiments.S. epidermidis is a gram-positive organism with an aerosol particle size of 800 nm, which have been used in bioaerosol studies (Lee et al., 2008a(Lee et al., , 2010;;Hwang et al., 2010).The concentration of bacterial particles in the suspension in the nebulizer was 6× 10 7 cfu/mL.The solvent for the bacterial particles was filtered sterile water (Aquamax, YoungLin instrument, Seoul, Korea) mixed with phosphate-buffered saline (PBS) buffer (Biosesang Inc., Seoul, Korea; NaCl 2.7 M, KCl 54 mM, Na 2 HPO 4 86 mM, KH 2 PO 4 28 mM) in the ratio 20:1.After aerosolization of the bacterial particles, we measured the volume of the particles passing through the filter by using a particle size distribution analyzer (PSD 3603, TSI Inc., MN, USA) to ensure that bacterial bioaerosols with a constant volume weight of 3.68 × 10 -2 mm 3 were deposited on the test filter.
Fig. 1(c) shows the process of determining the concentration of bacteria on the air filters.The test filters were immersed into filtered sterile water mixed with PBS buffer.The suspension contacting the filters was treated by sonicwave stimulus for 3 minutes and was vortexed for 1 minute to seperate the bacterial particles from the filter.The suspension was then washed twice by centrifugation at 15,000 rpm for 3 minutes to prevent any extra antimicrobial activity of the debris of the natural extract.We inoculated 0.1 mL of the washed suspensions containing the separated bacterial particles onto nutrient agar (beef extract 0.3%, peptone 0.5%, agar 1.5%, Difco, Becton Dickinson, NJ, USA) and incubated the plate for 24 hours.We enumerated the number of colonies on the agar plate and used this value to determine the bacterial bioaerosol concentration.
The leaf extract of Mukdenia rossii (Oliv) Koidz (major ingredients: flavonoid and polyphenol) was used in the experiment.This leaf was collected from the Dae-kwanryung mountain in the Gang-won province of the Republic of Korea, and the extraction was carried out at the Korea Institute of Science and Technology (KIST), Gangneung.We investigated this natural extract, and not chemically synthesized particles or metals, because of its potential environment-friendly side effects.Moreover, the leaf extract showed antimicrobial effects in the initial test stage of the study; therefore, we chose this material for our experiments.The extracted material was diluted by mixing 50 mg of the extract in 1.2 × 10 -3 m 3 of filtered sterile water.The extract particles were deposited onto filters at an airflow rate of 5 LPM for 10 min by aerosolization using a nebulizer and a diffusion dryer, as shown in Fig. 1(a).

Unipolar Ion Emission
Unipolar ions were emitted onto the filter at an airflow rate of 5 LPM by using a carbon fiber ionizer.The concentration of the ions was measured using an electrometer (TSI 3068 Electrometer, TSI Inc., MN, USA).The concentrations of positive and negative unipolar ions were 350,000 e + /cm 3 and 760,000 e -/cm 3 , respectively, when the measuring point was at a distance of 5 cm from the ionizer.We used these 2 experimental conditions because each represented the maximum concentration that the ion emitter could produce in our experimental system.

Antimicrobial Efficiency of Air Filters Treated with the Natural Extracts
Fig. 2 shows the concentration of culturable bioaerosols on the filters.In the control experiments, bacterial bioaerosols were deposited on the control filter for 10 minutes.The control filter was neither treated with the natural extract nor with the unipolar ions.The culturable bioaerosol concentration on the control filter was 7.6 × 10 5 ± 2.0 × 10 5 cfu/m 3 (first column of Fig. 2).The same experimental conditions were applied when testing the extract-treated filters.Fig. 3 shows an image of the filter treated with the natural extract; it shows the position of the submicron-sized natural extract particles on the filter fibers.The concentration of culturable bioaerosols on the extract-treated filter was 2.2 × 10 5 ± 1.6 × 10 5 cfu/m 3 , indicating that the number of culturable bioaerosols on the filters had decreased (second column data of Fig. 2).This reduction in the number of culturable bioaerosols was statistically significant (t-test; p-value = 0.003).Fig. 4(a) shows that the inactivation rate due to the natural extract-treated filters without ion emission was 71%.The inactivation rate is defined as: Inactivation rate = (concentration on control filterconcentration on tested filter)/concentration on control filter (1) To the best of our knowledge, this is the first study to report the antimicrobial activity of the natural extract of Mukdenia rossii (Oliv) Koidz on bioaerosols on air filters.

Emission of Positive and Negative Ions on Air Filters Treated with the Natural Extract
The effects of unipolar ion emissions on the bioaerosols on the extract-treated filters are shown in the third and fourth columns of the graph of Fig. 2. The culturable bioaerosol concentrations after emission of positive of positive and negative ions were 1.1 × 10 5 ± 9.7 × 10 4 cfu/m 3 (350,000 e + /cm 3 at 5 cm from the ionizer) and 8.8 × 10 4 ± 2.4 × 10 4 cfu/m 3 (760,000 e -/cm 3 at 5 cm from the ionizer), respectively.
On application of the natural extract and emission of positive ions, the culturable bioaerosol concentrations on the extract-treated filters were 86% lesser than those on the untreated filters, and this difference in the bioaerosol concentrations was statistically significant (t-test; p-value: 0.03 < 0.05).In other words, the average inactivation rate achieved on emission of positive unipolar ions on the extracttreated filters was 86% as shown in the second data point of Fig. 4(a), and the emission of positive unipolar ions further inactivated 50% of the bioaerosols that had survived against the action of the natural extract, thus, enhancing the antimicrobial performance of the extract-treated air filter by an average 20%.However, the differences between the inactivation rates achieved on treatment with the natural extract alone and on combined treatment with the natural extract and positive ions were not statistically significant (t-test p-values : 0.269 > 0.05).The emission of negative ions caused an 89% decrease in the concentration of culturable bioaerosols on the extracttreated air filter.This difference in the concentration of culturable bioaerosols between the treated filter with negative ions and untreated filters was statistically significant (t-test; p-value: 0.01 < 0.05).The average inactivation rate achieved after the application of the natural extract and emission of negative ions was 89% as shown in the third point of Fig. 4(a).However, the differences in the inactivation rates achieved on treatment with the natural extract alone and on combined treatment with the natural extract and negative ions were not statistically significant (t-test p-values: 0.16 > 0.05).Emission of negative unipolar ions further inactivated 60% of the bioaerosols on the extract-treated filters, thus enhancing the antimicrobial performance of the extract-treated air filter by 25%, on average, with statistically insignificant difference.

Effect of the Sequence of Unipolar Ion Emission and Bioaerosol Deposition
In addition to the above experiments, we emitted unipolar ions on the extract-treated filter before the deposition of bioaerosols.Thus, in this case, the bioaerosols were simultaneously exposed to the natural extract and unipolar ions.The culturable bioaerosol concentrations after emission of positive and negative ions were 7.5 × 10 4 ± 6.3 × 10 4 cfu/m 3 and 1.1 × 10 5 ± 1.5 × 10 5 cfu/m 3 , respectively.These data indicate that the inactivation rates achieved because of the application of the natural extract and emission of positive unipolar ions and application of the natural extract and emission of negative unipolar ions before bioaerosol deposition were 90 ± 8% and 85 ± 20%, respectively, as shown in the second and fourth points in Fig. 4(b).Therefore, Fig. 4(b) shows the experimental results of variation in the time point of ion emissions and polarity.The t-test p-value for the effect of positive unipolar ion emissions before and after bioaerosol depositions on the activation rate was 0.28 (> 0.05), and the corresponding value for the effect of negative unipolar ion emissions was 0.41 (> 0.05); therefore, the difference in effect of the sequence of unipolar ion emission and bioaerosol deposition was not statistically significant.Thus, the effect of unipolar ion emissions on the antimicrobial efficiency of extract-treated filters did not depend on the sequence of deposition of bioaerosols and emission of unipolar ions.
In these experiments, the natural extract of Mukdenia rossii (Oliv) Koidz was found to have antimicrobial activity; however, the detailed mechanism underlying these results is another point of research which will be conducted by biological approaches.Air ions have long been considered as antimicrobial stimuli; however, the detailed mechanisms underlying the antimicrobial activity of air ions are under investigation (Lee, 2011).Fletcher et al. (2007) suggested that air ions exert bactericidal effects via the electro-poration of bacterial cells.The position and accumulation of air ions on the surfaces of airborne microorganisms may disrupt the electric fields of adjacent cell walls, which in turn may disrupt the transport of electrons and protons inside the cells (Lee, 2011).The mechanisms underlying the hybrid effects of the natural extract and ion emissions need to be investigated.

CONCLUSION
In this study, we found that the natural extract of Mukdenia rossii (Oliv) Koidz showed antimicrobial activity when applied to air filters.Unipolar ion emissions could enhance the antimicrobial performance of the air filters by 20-25% on an average.The sequence of bioaerosol deposition and unipolar ion emission did not significantly affect the inactivation rate.In the future, it may be necessary to investigate the effects of various environmental conditions on the antimicrobial performance of filters treated with this extract and on the concomitant effects of the natural extract and ion emission.

Fig. 1 .
Fig. 1.Schematic diagram of the experimental setup.(a) Overall experimental setup, (b) images of tested respirators, and (c) determination of bacterial concentrations on the air filters images of tested respirators.

Fig. 2 .
Fig. 2. Culturable bioaerosol concentrations on filters.[first column], control condition (without natural extract and ions); [second column], filters treated with the natural extract; [third column], filters treated with the natural extract and positive unipolar ion emissions; and [fourth column], filters treated with the natural extract and negative unipolar ion emission.

Fig. 4 .
Fig. 4. (a) Inactivation rate achieved for filters treated with the natural extract and unipolar ion emissions.(b) The effect of the sequence of unipolar ion emission and bioaerosol deposition on the inactivation rate.