UV-Tolerant Culturable Bacteria in an Asian Dust Plume Transported over the East China Sea

Airborne bacteria are dispersed along with Asian dust and have various influences on atmospheric characteristics, downwind ecosystems, and public health. In this study, aircraft observations were conducted during Asian dust passage and non-dust conditions over the East China Sea in December 2010, and the composition of culturable bacteria within the dust was investigated using culture-based gene analysis. The 16S rRNA gene analysis revealed that the 34 cultured strains isolated from the transported dust could be grouped into Bacillales and Actinomycetales. Highly tolerant endospore-forming bacteria (25 isolates) were predominant in the composition. On the other hand, all non-endospore-forming bacteria (9 isolates) were gram-positive bacteria with high guanine and cytosine contents, which have substantial ultraviolet (UV) resistance. Thus, the isolated culturable bacteria were strains having some tolerance for exposure to UV radiation. Although the atmosphere is an extreme environment for bacterial survival, Asian dust can facilitate atmospheric dispersion of culturable bacteria that are resistant to the harsh ambient air environment.


INTRODUCTION
Atmospheric dust derived from desert areas is frequently transported great distances (e.g., Uno et al., 2009).Dust particles influence the global radiative balance (Satheesh and Moorthy, 2005) and hydrologic cycle by acting as ice nuclei and cloud condensation nuclei (Andreae and Rosenfeld, 2008).Mineral dust also has a beneficial effect on fertilization of downwind ecosystems (Bishop et al., 2002).However, some studies have shown that dust particles cause adverse respiratory effects in humans (Kwon et al., 2002;Ichinose et al., 2005).
Recent studies have suggested that arid and semiarid regions of the Asian continent are potential sources of airborne microbes associated with desert dust (Kellogg and Griffin, 2006;Kobayashi et al., 2007;Kakikawa et al., 2008;Maki et al., 2008;Kenzaka et al., 2010;Nishimura et al., 2010; † Now at Arid Land Research Center, Tottori University, Tottori 680-0001, Japan Yamada et al., 2010).Given the long-range diffusion of Asian dust in the Northern Hemisphere (Uno et al., 2009), microorganisms internally and/or externally mixed with dust are expected to be transported in the same manner.Some field experiments coupled with 16S and 18S rRNA gene analysis revealed that Asian dust can possibly bridge distant microbial ecosystems (Yukimura et al., 2009;Jeon et al., 2011;Yamaguchi et al., 2012;Jeon et al., 2013).It has been also reported that biological materials in dust, such as bacteria and organic substances, might enhance the efficiency of ice-cloud formation and precipitation (Pratt et al., 2009;Creamean et al., 2013).Biological substances in dust can have positive and negative impacts on humans, and these substances increase adverse respiratory effects in comparison with pure minerals (Ichinose et al., 2008;Liu et al., 2014).
To estimate the influences of airborne microbes transported in Asian dust on downwind ecosystems, we need to consider what kind of microorganism can be transported with maintenance of microbial survivability through long-range atmospheric dispersion (Kellogg and Griffin, 2006;Iwasaka et al., 2010).Studies have indicated that culturable halobacteria were sometimes dispersed with Asian dust (Echigo et al., 2005;Hua et al., 2007;Maki et al., 2010Maki et al., , 2011)).On the other hand, culture-independent methods revealed that various types of bacteria were present in desert soils and were transported in Asian dust (Yamaguchi et al., 2012;Maki et al., 2014).Therefore, both tolerant and non-tolerant bacteria exist in deserts and dusty air, either viable or dead.Atmospheric stressors such as ultraviolet (UV) radiation, low temperature and desiccation are harmful to airborne microbes, which can be critical to their viability in atmospheric transport (Lighthart, 1997;Tong and Lighthart, 1997;Smith et al., 2011).However, there is limited information on the survival of airborne bacteria transported with Asian dust.
Here we present qualitative information of culturable bacteria associated with an Asian dust plume transported over the East China Sea.Aircraft observations were conducted over the sea offshore of western Japan during Asian dust passage and non-dust conditions in December 2010, and isolated bacteria were analyzed using culture-based gene analysis.The purpose of this study is to indicate the composition of culturable bacteria transported in an Asian dust plume over the East China Sea, and to estimate from the results bacterial tolerance associated with their survival during atmospheric transport.

Preparation
Air samples were collected using sterile gelatin filters (Sartorius AG, Germany).Sampling with these filters has high collection efficiency across a wide range of flow rates, for example, > 90% for Bacillus endospores (Parks et al., 1996;Burton et al., 2005).All sampling apparatus such as membrane filters, filter holders, tubes and tube clumps were first autoclaved at 121°C for 15 min.The sterile gelatin filters were cut to fit on 47-mm filter holders (Millipore, Billerica, MA) using sterile instruments.Gelatin filters were then placed on sterile 0.22-µm pore size membrane filters (Millipore) in the filter holders.After that, sampling units were assembled and the units were stored in sterile plastic bags at 4°C until sample collection.The above procedures were performed in a laminar flow hood sterilized with 70% ethanol and UV (254-nm wavelength) radiation.
Before sample collection, R2A agar media (BD Difco, Detroit, MI) and tryptic soy agar media (TSA; BD Difco) were prepared with 100 µg/mL cycloheximide added to inhibit fungal growth.To rule out artificial contamination, samples of all prepared agar media were incubated at room temperature for more than 10 days before culture experiments were initiated.

Airborne Observation and Sample Collection
Airborne observations were made over the East China Sea (Fig. 1) near Fukue Island, on 11, 12 and 14 December 2010.The base was at Fukue Airport, Nagasaki, western Japan (128.83°E, 32.67°N; black asterisk in Fig. 1).A Beechcraft King Air 200T was chartered from Diamond Air Service Inc. (Aichi, Japan) for observation flights.An inlet was installed on the upper portion of the airframe.Outside air from the inlet was introduced into a glass manifold and then supplied to air samplers and analyzers.During each observation flight, airspeed was maintained at approximately 80 m/s.We assumed a particle density of 1.0 g/cm 3 for bacteria and 2.3 g/cm 3 for mineral dust particles.Given these two airborne particle densities, estimated 50% cutoff aerodynamic diameters of the inlet at the aforementioned flight speed were approximately 3.6 µm and 2.4 µm, respectively.
During the observation flights, sampling units and air samplers (Kobayashi et al., 2007) were connected to the manifold using sterile silicon tubes with a closed tube clamp.Three trip blanks (gelatin filters in sterile 50 mL centrifugal tubes) were prepared for every flight.When the aircraft reached each sampling altitude, interiors of the inlet, pipe and manifold were primed with outside air for 5 minutes.A tube clamp was then opened and outside airborne particles were collected on the filters through the manifold, at a rate of 16 L/min for 30 minutes.The tube clamp was then closed until the next sample collection.After observations, tube ends of the sampling units were wrapped with Parafilm, and the units were placed into sterile plastic bags.The above procedures were done using sterile disposable gloves.Before observations began, an air sample from inside the aircraft was collected on 10 December following the same procedures, to evaluate the potential for contamination from inside the aircraft.All collected samples and trip blanks were stored at 4°C until further treatment.
Particle mass and number concentrations were also monitored during each flight.Particle mass concentrations were measured in diameter ranges 0.5-1.0µm, 0.5-2.5 µm, 0.5-7.0µm and 0.5-10 µm, using a particle mass monitor (GT-321; Shibata, Japan).Particle number concentrations with equivalent optical diameter D p were measured in D p > 2.5 µm, using an optical particle counter (GT-521; Shibata).Ambient temperature and other flight observation data (e.g., atmospheric pressure and flight speed) were monitored onboard the aircraft.The flight system is also described in other works (Hatakeyama et al., 2011;Fujiwara et al., 2014;Hatakeyama et al., 2014).

Cultivation
The gelatin filters were removed from the filter holders in a sterilized laminar flow hood at the Prefectural University of Kumamoto.Each gelatin filter was dissolved in sterile Milli-Q water at 37°C, and 0.2 mL sample aliquots were streaked onto TSA and R2A agar media and incubated at 25°C for approximately 1 month (e.g., van der Linde et al., 1999).Single isolated bacterial colonies were identified and transferred to fresh media three times to obtain pure cultures.

Analysis
DNA extraction and polymerase chain reaction (PCR) were carried out for bacterial isolates.Each isolated strain was incubated with R2A broth (Nihon Pharmaceutical Co., Ltd., Japan) or tryptic soy broth (BD Difco).A total 1 mL of broth culture was centrifuged at 6000 rpm for 5 min.Supernatant was removed and bacterial cells were mixed with 180 µL of lysis buffer, as recommended for DNA extraction from gram-positive bacteria in the DNeasy Blood and Tissue Kit (Qiagen Inc., Valencia, CA).Subsequent extraction procedures were followed according to the kit protocol, and 200 µL of DNA elute was obtained.Extracts were concentrated to 100 µL in TE buffer (pH.8.0) by ethanol precipitation, and stored at −20°C until subsequent PCR amplification.

RESULTS
On 11 December 2010, Asian dust was observed in western Japan.The airborne observation on that day was conducted during Asian dust passage over the East China Sea, at altitudes 500, 1000, 2000 and 3000 m.Table 1 shows particle concentrations and flight observation data for each level flight on 11, 12 and 14 December.Measurements by the particle mass monitor for each flight indicated that concentrations in diameter range 0.5-10 µm were almost equal to those of 0.5-7.0µm (data not shown).In this study, measured concentrations within 0.5-7.0µm were used as particle mass concentrations and shown in Table 1.Particle mass concentrations in level flights ranged between 1.3 and 62 µg/m 3 , and number concentrations of coarse particles (D p > 2.5 µm) were between 5.1 × 10 3 and 2.2 × 10 5 particles/m 3 .Particle mass and number concentrations in the dust (11 December) were approximately 7-30 times higher than those in non-dusty air (12 and 14 December).
Two samples collected at 500 and 1000 m altitudes on 11 December, a sample at 500 m on 12 December, a sample inside the aircraft (Control in Table 1), and trip blanks were used for cultivation.In total, 38 culturable bacterial strains were isolated from the four samples (Fig. 2).From Sample A, 15 strains were isolated on R2A agar and 10 strains were isolated on TSA.From Sample B, five bacterial strains were isolated on R2A agar and four strains were isolated on TSA.No bacteria were isolated from Sample C. Four strains were isolated on TSA from the control, but there was no bacterial growth in trip blanks.
The result of gene and phylogenetic analysis is shown in Figs.3(a) and 3(b).Based on 16S rRNA gene sequence similarity, 38 bacterial isolates were classified into species.For each isolate, the obtained sequences showed more than 99% sequence identity.All strains isolated from Samples A and B were gram-positive and grouped into two orders, Bacillales and Actinomycetales.The 25 bacterial isolates obtained from Sample A were assigned to the genera Bacillus (13 isolates), Streptomyces (6 isolates), Paenibacillus (2 isolates), Terribacillus (2 isolates), Gordonia (1 isolate), and Shimazuella (1 isolate).The nine strains obtained from Sample B were in the genera Bacillus (5 isolates), Terribacillus (2 isolates), Saccharopolyspora (1 isolate), and Streptomyces (1 isolate).Four bacterial strains obtained from the control were gram-positive and belonged to Actinomycetales (Fig. 3(a)).Isolated strains were assigned to the genera Streptomyces (3 isolates) and Micrococcus (1 isolate).Although Streptomyces species were also isolated from Samples A and B, these were not identical at the 16S rRNA gene level to Streptomyces species isolated from the control (Fig. 3(a)).

DISCUSSION
Culturable bacterial strains isolated from the dust predominantly belonged to the order Bacillales (Fig. 3(b)).Bacteria such as Bacillus, Terribacillus, Paenibacillus and Shimazuella species can form highly tolerant endospores (Nicholson et al., 2000;Park et al., 2007).In the case of genus Bacillus, spore coats, specific proteins and DNA repair machinery in the endospores are important factors to prevent damage from solar UV radiation, temperature, desiccation, and chemical substances (Nicholson et al., 2000).It is obvious that the endospore state is advantageous to maintain bacterial survivability during atmospheric transport.In past studies, endospore-forming Bacillus species were frequently isolated from long-range transported dust using the culturedependent and independent methods (Hua et al., 2007;Maki et al., 2010;Jeon et al., 2011;Maki et al., 2011;Tanaka et al., 2011;Yamaguchi et al., 2012).The result of the present study is thus consistent with past investigations.
We also isolated culturable non-endospore-forming bacteria belonging to the order Actinomycetales (Streptomyces, Saccharopolyspora and Gordonia species) from the dust (Fig. 3(a)).Characteristics of these species are high guanine and cytosine (GC) contents in their DNA (Singer and Ames, 1970;Arenskӧtter et al., 2004;Ventura et al., 2007;Qin et al., 2010).It has been reported that bacterial tolerance for solar UV radiation increases with greater G + C content (Singer and Ames, 1970), i.e., high GC gram-positive bacteria have a certain level of solar UV resistance.Griffin et al. (2006) also mentioned that high GC contents might enhance bacterial survival in atmospheric transport with African dust.
Overall, all isolated culturable bacteria in the present study were strains with some tolerance for UV radiation exposure.UV radiation can be separated into three wavelengths, UV-A (400-320 nm), UV-B (320-280 nm) and UV-C (< 280 nm).Solar UV-A and UV-B radiation is able to penetrate the ozone layer.UV-B radiation is absorbed by DNA in cells and inhibits DNA replication and RNA transcription (Mitchell and Karentz, 1993).In addition, UV-A and UV-B radiation cause the formation of reactive oxygen substances such as •OH, H 2 O 2 , 1 O 2 and O 2 -in the cells, which trigger disruption of cell membranes, inactivation of several enzymes, and DNA damage (Pattison and Davies, 2006).These damages induced by both direct and indirect reactions with solar UV radiation lead to substantial cell death (Demidova and Hamblin, 2004;Lukšienė, 2005).
Culturable UV-tolerant bacterial strains were also isolated in the previous studies carried on near grounds in the arrival areas of Asian, African and Australian dust (Griffin et al., 2006;Hua et al., 2007;Lim et al., 2011).However, these studies could not omit the background or bacteria from local sources, meaning that previous results could contain not only the compositions of bacteria transported in dust but also some few bacterial compositions aerosolized from near sampling sites.Sample collections over the surfaces by using aircrafts and/or air balloons can decrease inputs of microbes from various sources with respect to atmospheric samples obtained during dust passage, although those sample collections have other problems.One of the problems on Fig. 3(a).Phylogenetic tree from bacterial isolates cultured on TSA and R2A agar media.The phylogenetic tree was constructed using the neighbor-joining method.Reference sequences (> 1400 bp) are shown in regular font with GenBank accession numbers.Sequences of isolated bacteria are shown in bold, together with sample and culture media information.Scale bar indicates 0.1 changes per nucleotide.Sequence assembly had > 99% sequence identity.Phylogenetic relationships were bootstrapped 1,000 times, and bootstrap values > 50% are shown.

Fig. 3(b). Phylogenetic tree (continued).
aircraft observations was inevitable particle loss in inlets and pipes (e.g., Pui et al., 1987).Considering the cutoff diameter of the inlet and measurements of particle mass concentrations, the inevitable particle loss in our aircraft observations was occurred in the ranges of airborne particles larger than 7 µm.As a result, most airborne particles > 7 µm and some airborne particles < 7 µm in the air were not collected for analysis in the present study.Recent works show that mineral particles transported in long-range transported dust were dominated in the range of particle diameter between 2 to 5 µm in remote areas (Zhang et al., 2003;Perry et al., 2004).Other studies reported that almost bacteria in soils, water, atmosphere and long-range transported dust were present in the size range smaller than 1 µm (Bae et al., 1972;Watson et al., 1977;Hara and Zhang, 2012).Therefore, our aircraft observations were adequate to obtain airborne bacteria associated with or without mineral dust, even though the data included some underestimations.Another potential problem with aircraft observations could be contamination from inside the airplane.In the present study, culturable bacteria isolated from the dust were not identical at 16S rRNA gene level with culturable bacteria isolated from air inside the aircraft.Additionally, no bacterial growth was found in the trip blanks, indicating that artificial contamination during sample preparation, collection and cultivation processes was minimal for at least the culturable bacterial composition.Thus probably up to the 34 types of cultured UV-tolerant bacteria traveled in the Asian dust plume over the East China Sea.
Numerous case studies have indicated that solar UV radiation can negatively influence microbial survival in various environments (e.g., Llabrés and Agusti, 2006).The potential for bacterial cell membrane damage by the atmospheric stressors through long-range atmospheric transport has been also reported (Hara and Zhang, 2012).Even though the present study based on only one dust event over the East China Sea, the result indicates typical bacterial strains survived in long-range atmospheric transport without viable bacterial strains of local sources and artificial contamination.However, a lot of case studies in various dust events are needed to elucidate survival mechanisms of airborne bacteria transported in Asian dust.Furthermore, some laboratory experiments are also necessary to assess bacterial UV-tolerance in the air.Nevertheless, bacterial UV-tolerance might be related to bacterial survival in atmospheric transport from arid and semiarid regions of the Asian continent.

CONCLUSIONS
We conducted aircraft observations in western Japan during Asian-dust passage and non-dust conditions over the East China Sea.Atmospheric samples were cultured for approximately 1 month, and 38 cultured bacterial isolates were studied through 16S rRNA gene analysis.The results show that culturable bacterial strains isolated from Asian dust were different than strains isolated from air inside the aircraft, and were dominated by highly tolerant endosporeforming bacteria.Non-endospore-forming bacteria isolated from the dust were high G + C gram-positive bacteria, which have UV resistance.As a result, all cultured bacteria were strains with substaintial tolerance for exposure to UV radiation.Solar UV radiation likely affects the viability of airborne bacteria, and bacterial UV-tolerance can be a key factor for their survival in long-range atmospheric transport with Asian dust.

Fig. 1 .
Fig. 1.Flight and dust information.Flights on 11 and 12 December 2010 shown as red and blue lines, respectively.Flight on 14 December was similar to that on 12 December.Areas affected by Asian dust on 11 December marked by black dots (13 of 61 JMA monitoring sites; http://www.jma.go.jp/jma/index.html).Black asterisk indicates Fukue airport.