Subscriber access provided by Queen Mary, University of London
Article
Origin-Dependent Variations in the Atmospheric Microbiome Community in Eastern Mediterranean Dust Storms Daniella Gat, Yinon Mazar, Eddie Cytryn, and Yinon Rudich Environ. Sci. Technol., Just Accepted Manuscript • Publication Date (Web): 19 Apr 2017 Downloaded from http://pubs.acs.org on April 20, 2017
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Environmental Science & Technology is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 34
Environmental Science & Technology
1
Origin-Dependent Variations in the Atmospheric
2
Microbiome Community in Eastern Mediterranean
3
Dust Storms
4
Daniella Gat,1 Yinon Mazar,1 Eddie Cytryn,2 Yinon Rudich1*
5
1
6 7 8
Department of Earth and Planetary Sciences, Weizmann Institute of Science,
Rehovot 7610001, Israel. 2
Institute of Soil, Water and Environmental Sciences, The Volcani Center,
Agriculture Research Organization, Rishon Lezion 7528809, Israel
1
ACS Paragon Plus Environment
Environmental Science & Technology
9 10
Abstract Microorganisms carried by dust storms are transported through the atmosphere and
11
may affect human health and the functionality of microbial communities in various
12
environments. Characterizing the dust-borne microbiome in dust storms of different
13
origins or that followed different trajectories, provides valuable data to improve our
14
understanding of global health and environmental impacts. We present a comparative
15
study on the diversity of dust-borne bacterial communities in dust storms from three
16
distinct origins—North Africa, Syria and Saudi Arabia—and compare them with local
17
bacterial communities sampled on clear days, all collected at a single location, in
18
Rehovot, Israel. Storms from different dust origins exhibited distinct bacterial
19
communities, with signature bacterial taxa. Dust storms were characterized by a lower
20
abundance of selected antibiotic resistance genes (ARGs) compared with ambient
21
dust, asserting that the origin of these genes is local, possibly anthropogenic. With the
22
progression of the storm, the storm-borne bacterial community showed increasing
23
resemblance to ambient dust, suggesting mixing with local dust. These results show,
24
for the first time, that dust storms from different sources display distinct bacterial
25
communities, suggesting possible diverse effects on the environment and public
26
health.
27
Introduction
28
Dust-storms carry biological particles, such as plant debris, pollen, fungi and
29
bacteria, along with soil and mineral particles, over vast distances1,2. These biological
30
airborne particles play important roles in the Earth system by coupling between the
31
biosphere and the atmosphere3-6. Airborne bacteria may affect human health by
32
spreading pathogens7, they express proteins that serve as ice nuclei, potentially
33
affecting cloud formation8, and could affect agriculture and ecosystems health by 2
ACS Paragon Plus Environment
Page 2 of 34
Page 3 of 34
Environmental Science & Technology
34
spreading pathogens and bacteria that impact biogeochemical cycles 7,9,10. This wide
35
array of potential implications of bacterial transport through the atmosphere, brought
36
about an increasing interest in the bacterial diversity in aerosols and dust (e.g. 2,11-
37
14). The study of microbial diversity in dust previously relied mainly on culture-based
38
methods15, which are usually time consuming, and capture only a fraction of the total
39
bacterial diversity. Recent developments of culture independent methods for
40
microbiological analysis open-up the possibility to extensively study and characterize
41
the atmospheric microbiome12,15,16. A limited number of studies, using next-
42
generation sequencing techniques for microbial community characterization have
43
shown that different dust origins display diverse microbial communities2,11,16 in which
44
up to 25% of the bacteria remain viable17,18. This variety in bacterial community
45
composition could translate into various health, biogeochemical, or environmental
46
effects1,9,19.Previous studies of the dust microbiome have monitored ambient changes
47
in the airborne bacterial community in urban locations in the US, using PhyloChip
48
detection7, and examined bacterial community differences between clear days and
49
dust-storms in the Eastern Mediterranean using next-generation sequencing11. Other
50
studies have mapped the differences between bacterial and fungal communities using
51
next-generation sequencing in settled dust in different geographic locations across the
52
US13, the seasonal variations in concentrations and composition of allergenic fungal
53
spores were determined using qPCR and next-generation sequencing20,21, and the
54
seasonal variability of airborne microbial communities in rural and urban locations in
55
the US was examined over a period of 14 months14. Most of these studies explored
56
ambient aerosols and settled dust, which often reflect their immediate surroundings11-
57
13
58
represent the source bacterial community22,23. The role of dust storms in spreading
; in contrast, bacteria in dust storms are carried over large distances and may
3
ACS Paragon Plus Environment
Environmental Science & Technology
59
nutrients has been previously demonstrated, along with ecological changes attributed
60
to dust deposition24-26. It was also shown that dust-borne bacteria can contribute to N2
61
fixation in marine environments9, and that dust from different sources can produce a
62
distinct effect on bacterial production and community composition in seawater25.
63
However, the possible effect of deposition of dust-borne microorganisms on
64
indigenous microbial communities, and the relationship between the sources of the
65
atmospheric microbiome and microbial community composition are yet under-
66
characterized, and represent an important gap in the scientific knowledge about Earth
67
microbiomes27,28.
68
Dust storms are known to pose health risks due to inhalation of particulate
69
matter29,30, yet their role in spreading antibiotic resistance genes (ARGs), originating
70
in soil bacteria, is still being unraveled11,31,32. The introduction of ARGs into
71
pathogenic bacterial strains, via horizontal gene transfer, may result in increasing the
72
abundance of antibiotic resistant pathogens33. Thus, soil dust could potentially
73
become a vector for spreading antibiotic resistance genes, which is an increasing
74
global health problem34. Conversely, recent evidence suggests that airborne ARGs
75
originate from anthropogenic sources, such as chicken coops and agricultural plots,
76
and are less abundant in storm dust from North Africa11. This, again, highlights the
77
importance of the origin of the dust in assessing its potential affects.
78
The Eastern Mediterranean is frequented by seasonal dust storms, usually lasting 1–
79
2 days35. Global drying and warming trends are predicted to increase the frequency
80
and intensity of dust storms from this region, and particulate matter measurements
81
from Israel show that dust events have become more extreme, with higher daily and
82
hourly levels, since 200936,37. Most dust storms in the Eastern Mediterranean originate
4
ACS Paragon Plus Environment
Page 4 of 34
Page 5 of 34
Environmental Science & Technology
83
in the Sahara Desert, while some originate in south-eastern deserts, such as in the
84
Arabian Peninsula (Arabian Desert)37.
85
On September 8, 2015, a dust storm that was unprecedented with respect to the
86
conditions that initiated it, its origin, magnitude and duration, reached Israel from
87
Northern Syria35,38,39. Another intense dust storm from Jordan and the Arabian
88
Peninsula affected the region on November 3, 2015. These two storms were sampled
89
at a single location at the Weizmann Institute of Science in Rehovot, Israel, and were
90
compared to dust collected, in the same location, and described by Mazar et al.11
91
during two major North African dust storms, on March 3, 2014, and on February 10–
92
11, 2015. All storm samples were also compared to pooled aerosol samples collected
93
on nine clear days (i.e., PM10 < 53 µg m-3).
94
Here we present a first comparative study of bacterial communities in dust storms
95
originating from three different locations, using next-generation sequencing. We also
96
examine the contribution of these storms to the local airborne reservoir of selected
97
ARGs and horizontal gene transfer related elements, to detect possible differences in
98
ARG abundance between different dust origins.
99
Experimental
100
Dust collection
101
Atmospheric particulate matter having an aerodynamic diameter smaller than 10 µm
102
(PM10) was collected, as described in Mazar et al.11, on quartz microfiber filters
103
(Whatman, 203 mm × 254 mm) using a high-volume air sampler (Ecotech, model:
104
HiVol 3000, PM10 inlet, flow rate: 67.8 m3·h-1) situated on the roof of a four-story
105
building in the Weizmann Institute of Science, Rehovot, Israel (31.9070 N, 34.8102
106
E). Despite the wide distribution of particle sizes carried by dust storms, we focused
5
ACS Paragon Plus Environment
Environmental Science & Technology
107
on PM10 since small size particles tend to travel greater distances40, thus they should
108
better represent the source of the dust.
109
Prior to sampling, all filters were baked at 450°C for five hours to clean them of all
110
organic matter. Post-sampling, all filters were kept at -20°C until DNA was extracted.
111
Sampling took place during major dust storms between March 2014 and November
112
2015 and during clear days exhibiting PM10 concentrations under 53 µg·m3 (exact
113
dates of clear days are reported in Mazar et al.11 and in Table S1). The threshold for
114
clear days was established to be lower than the annual mean of urban PM10
115
concentrations in Israel, i.e., 53 µg·m3, according to the World Health Organization
116
database41. The total estimated weight of sampled dust was approx. 0.2 g per filter.
117
Backward trajectory calculations and meteorological data
118
Identification of the origin for each storm event was achieved using backward
119
trajectory calculations on the Hybrid Single Particle Lagrangian Integrated Trajectory
120
Model (HYSPLIT)42, via the web-based interface (READY,
121
http://ready.arl.noaa.gov/HYSPLIT_traj.php). The back-trajectories for the September
122
2015 storm were more complex, as a dust plume that formed over northern Syria
123
propagated south-west and was then spread along the eastern Mediterranean coast due
124
to a wind reversal35,38. Data of atmospheric PM10 concentrations were obtained from
125
the Israeli Ministry of Environmental Protection database, Rehovot Air Monitoring
126
station (http://www.svivaaqm.net/StationReportFast.aspx?ST_ID=32).
127
DNA extraction, amplification, and sequencing
128
The DNA extraction procedure was described in Mazar et al.11. In short, dust was
129
scraped from the filters using a sterile surgical scalpel and transferred to PowerSoil kit
130
bead-tubes (MoBio). The extraction procedure followed the manufacturer’s
6
ACS Paragon Plus Environment
Page 6 of 34
Page 7 of 34
Environmental Science & Technology
131
recommended protocol with slight modifications. All DNA samples were kept at -
132
20°C. DNA sequencing was conducted at the DNA Sequencing Facility (DNAS) at
133
the University of Illinois at Chicago (UIC), using Illumina Miseq instrument. The
134
targeted sequence is a segment of the V4 region in the 16S rRNA gene, the primers
135
used were: CS1_515F
136
(ACACTGACGACATGGTTCTACAGTGCCAGCMGCCGCGGTAA) and
137
CS2_806R (TACGGTAGCAGAGACTTGGTCTGGACTACHVGGGTWTCTAAT)
138
43
139
-SRR5247695). Merging of paired-end reads as well as quality trimming were
140
conducted by the Research Resources Center of the UIC. The produced Fasta files
141
were analyzed with mothur v. 1.39.044 software, according to the pipeline described in
142
Kozich et al.45 and on the website: http://www.mothur.org/wiki/MiSeq_SOP, a
143
complete command log is described in the supporting information. Taxonomic
144
classification of the assembled sequences was conducted using Silva rRNA
145
database46, v.123.
. All sequences were submitted to the NCBI SRA database (accession: SRR5247688
146
Quantification of total bacteria and of selected genes was achieved by qPCR. The
147
applied primers are listed in Table 1. The qPCR results for the storms dated March 3,
148
2014 and February 10–11, 2015, and for clear days were obtained from Mazar et al.11.
149
For samples obtained during the September 2015 and the November 2015 storms, the
150
following 25 µl reaction mixture was prepared: 12.5 µl of SensiFAST™ SYBR mix
151
(Bioline®), 1 µl of each primer (0.4 mM), and 7.5 µl of template DNA. Amplification
152
took place in a StepOnePlus™ real-time qPCR (ThermoFisher Scientific®) as follows:
153
3 min at 95°C, followed by 35 cycles of 2 s at 95°C and 30 s at 60°C. The standard
154
curve for all qPCR reactions was plotted based on amplification of known
7
ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 34
155
concentrations of pNORM1 plasmid (see supporting information for description),
156
along with a non-template control.
157
Table 1: qPCR primers list.
Target Gene intI1
sulI
qnrS
16S
Primer
Sequence
Reference
intI1LC5 (F)
GATCGGTCGAATGCGTGT
47
intI1LC1 (R)
GCCTTGATGTTACCCGAGAG
sul1-FW
CGCACCGGAAACATCGCTGCAC
sul1-RV
TGAAGTTCCGCCGCAAGGCTCG
qnrSrtF11
GACGTGCTAACTTGCGTGAT
qnrSrtR11
TGGCATTGTTGGAAACTTG
331F
TCCTACGGGAGGCAGCAGT
518R
ATTACCGCGGCTGCTGG
48
49
50
158 159
Results and Discussion
160
Dust sources
161 162
Sample names, dates and origins are listed in Table 2. Backward trajectories of the dust storms analyzed in this study are presented in
8
ACS Paragon Plus Environment
Page 9 of 34
Environmental Science & Technology
163
Table 3. The March 2014 and February 2015 dust storms originated in the Saharan
164
region of North Africa. However, the paths of these storms differed somewhat, as the
165
March 2014 storm’s trajectory passed over the Mediterranean Sea, north of the Nile
166
delta, whereas the February 2015 trajectory did not. During the first two sampling
167
days (Arabia_d1 and Arabia_d2) of the November 2015 storm, the dust originated in
168
Saudi Arabia and Jordan, whereas following a change in wind direction, the last
169
sampling day (Arabia_d3) represented dust from the Red Sea, the Gulf of Eilat/Aqaba
170
close to the Saudi Arabian coast, and the Israeli Negev. According to satellite images,
171
as well as published data39, the September 2015 storm originated in north-east Syria,
172
around the Euphrates valley and the city of A-Raqqa. Back trajectories show the dust
173
progressed southwards towards the Golan heights and from there it was carried and
174
spread over Israel due to wind reversal.
175
Table 2: Sample names, sampling dates, dust concentration on each sampling day and dust source. PM10
176
values represent the average daily concentration of particulate matter measuring 10 µm or less in diameter
177
on each sampling date, as measured by the Rehovot Air-Monitoring station and available on:
178
http://www.svivaaqm.net/StationReportFast.aspx?ST_ID=32
Sample name
Date
PM10 (µg·m-3)
Source
NCBI accession no.
Sahara1
03-Mar-14
838
North-Africa
SRR5247690
Sahara2_d1
10-Feb-15
482
North-Africa
SRR5247689
Sahara2_d2
11-Feb-15
1643
North-Africa
SRR5247688
Syria_d1
08-Sep-15
844
Syria
SRR5247693
Syria_d2
09-Sep-15
1859
Syria
SRR5247692
Syria_d3
10-Sep-15
928
Syria
SRR5247691
Arabia_d1
03-Nov-15
422
Saudi Arabia
SRR5247696
Arabia_d2
04-Nov-15
810
Saudi Arabia
SRR5247695
Arabia_d3
05-Nov-15
453
Saudi Arabia
SRR5247694
179
9
ACS Paragon Plus Environment
Environmental Science & Technology
180 181
Page 10 of 34
Table 3: Back-trajectories of each storm sample as obtained from the Hybrid Single Particle Lagrangian Integrated Trajectory Model (HYSPLIT)42 (http://ready.arl.noaa.gov/HYSPLIT_traj.php).
Sahara1
Sahara2_d1
Sahara2_d2
Syria_d1, d2, d3
Arabia_d1
Arabia_d2
182
10
ACS Paragon Plus Environment
Arabia_d3
Page 11 of 34
Environmental Science & Technology
183
Bacterial community composition
184
The bacterial community composition of the collected dust is presented in Figure 1.
185
For comparison, the 16S rRNA gene amplicon sequences extracted from dust samples
186
during clear days and during North-African storms, obtained and presented by Mazar
187
et al.11, were re-analyzed in tandem with the September 2015 and November 2015
188
dust samples (clear days’ samples are detailed in Table S1). The most abundant phyla
189
in all sampled storm dust microbiomes were varying proportions of Actinobacteria,
190
Proteobacteria, Firmicutes, Chloroflexi, and Bacteroidetes.
191
North African dust storms were characterized by the low abundance of Chloroflexi
192
and relatively high abundance of Deinococcus-Thermus, compared to the other storm
193
origins. The differences observed between the different North-African dust samples,
194
may have arisen from different paths, as observed by their back-trajectories (Table 3).
195
Sample Sahara1 represents dust that was carried shortly over the Mediterranean Sea,
196
north of the Nile delta, whereas samples Sahara2_d1 and Sahara2_d2 represent a
197
solely terrestrial trajectory, via northern Egypt and the Sinai desert. Other parameters
198
that might have caused differences between these storms may be the time of their
199
occurrence, almost a year apart; and/or the different seasons, since one took place
200
during early spring (Sahara1 on March 2014) and the other during the winter
201
(Sahara2 on February 2015).
202
The dust storm from Syria was characterized by a high relative abundance of
203
Cyanobacteria and Chloroflexi and low relative abundance of Firmicutes, compared
204
with other dust origins. During the storm, an increase in the relative abundance of
205
Firmicutes was observed, from 2% on sample Sahara_d1 to 8% on sample
206
Sahara_d3.
11
ACS Paragon Plus Environment
Environmental Science & Technology
207
The Saudi Arabian storm differed from the other two dust storms in the relatively
208
low abundance of Actinobacteria and the high abundance of Proteobacteria. A notable
209
change in bacterial community composition was observed during this storm; while the
210
first two days (samples Arabia_d1 and Arabia_d2) exhibited a similarity in bacterial
211
community composition with 30% Actinobacteria, 29% Proteobacteria, 10%
212
Bacteroidetes and 9% Chloroflexi in both samples, on the third day (Arabia_d3) the
213
samples exhibited a relatively high abundance of Firmicutes, 29%, compared to