Page 1 of 29
Environmental Science & Technology
1
Transcriptome Analysis of Invasive Plants in Response to Mineral
2
Toxicity of Reclaimed Coal-mine Soil in the Appalachian Region
3
Thangasamy Saminathan1‡, Sridhar A. Malkaram1‡, Dharmesh Patel2, Kaitlyn Taylor1, Amir
4
Hass2, Padma Nimmakayala1, David Huber1, Umesh K. Reddy1*
5
1
Department of Biology, Gus R. Douglass Institute, West Virginia State University, Institute,
6 7 8
WV 25112-1000, USA. 2
Agricultural and Environmental Research Station, Gus R. Douglass Institute, West Virginia State University, Institute, WV 25112-1000, USA.
9 10
KEYWORDS: Coal mine; Soil; Toxicity; Transcriptome; RNA-seq; Mugwort; Goutweed;
11
Reclamation; Appalachian Mountains.
12 13 14
1
ACS Paragon Plus Environment
Environmental Science & Technology
15
ABSTRACT:
16
17
18
19
Efficient post-mining reclamation requires successful revegetation. By using RNA sequencing,
20
we evaluated the growth response of two invasive plants, goutweed (Aegopodium podagraria L.)
21
and mugwort (Artemisia vulgaris), grown in two Appalachian acid-mine soils (MS-I and -II, pH
22
~4.6). Although deficient in macronutrients, both soils contained high levels of plant-available
23
Al, Fe and Mn. Both plant types showed toxicity tolerance, but metal accumulation differed by
24
plant and site. With MS-I, Al accumulation was greater for mugwort than goutweed (385±47 vs
25
2151±251 µg g-1). Al concentration was similar between mine sites, but its accumulation in
26
mugwort was greater with MS-I than MS-II, with no difference in accumulation by site for
27
goutweed. An in situ approach revealed deregulation of multiple factors such as transporters,
28
transcription factors, and metal chelators for metal uptake or exclusion. The two plant systems
29
showed common gene expression patterns for different pathways. Both plant systems appeared
30
to have few common heavy-metal pathway regulators addressing mineral toxicity/deficiency in
31
both mine sites, which implies adaptability of invasive plants for efficient growth at mine sites
32
with toxic waste. Functional genomics can be used to screen for plant adaptability, especially for
33
reclamation and phytoremediation of contaminated soils and waters.
34
2
ACS Paragon Plus Environment
Page 2 of 29
Page 3 of 29
Environmental Science & Technology
35
INTRODUCTION:
36
Coal mining process in the Appalachian region
37
The Appalachian basin is a coal-rich region extending from Alabama to Maine in the United
38
States. Coal mining has been performed in West Virginia, in the Central Appalachian region, for
39
271 years. Mining activity results in a land cover change with huge environmental consequences.
40
In total, 5% (19,581 km2) of the land surface in southern West Virginia was converted to surface
41
mines in the 3 decades after 1976 1. Coal mining in Appalachia is expected to continue to be a
42
significant contributor to the region economy and US energy portfolio, with projected production
43
of 240 million tons per year by 2040 (reduced from 266 million tons in 2012) 2.
44
Mountain-top removal is used for near-surface coal extraction. Such a large-scale disturbance
45
and activity leads to coal slurry, or sludge, a mix of water, coal dust and clay containing toxic
46
heavy metals such as arsenic, mercury, lead and chromium and excess essential minerals. In
47
addition to an imbalance in toxic mineral and heavy metal content, soil pH decreases to < 5.5
48
because of weathering of rocks and oxidation of pyritic minerals into sulfuric acid during mining
49
3
50
forming ions (H, Al). Moreover, the removal of the native soil and its replacement with soil
51
replacement material, namely finely crushed rocks, creates additional challenges for plant
52
establishment and growth and overall success of revegetation and reclamation in the post-mining
53
area.
. The process involves removing the base-forming ions (Ca, Mg, K) and replacement with acid-
54
Heavy metal toxicity and plant signaling
55
Macronutrients such as N, P, K, Ca, Mg, and S, and micronutrients such as Fe, Cu, Mn, Zn,
56
Mo, Cl, and Ni are required for plant growth. Other minerals such as Na, Co, V, Se, Al and Si
57
are beneficial but not necessary to plants. Nutrient toxicity or deficiency occurs when nutrients
3
ACS Paragon Plus Environment
Environmental Science & Technology
Page 4 of 29
58
are present in excess or insufficient quantity 4. Basal heavy-metal toxicity occurs in plants by
59
uptake, sequestering and chelating the excess minerals that are present 5. The easiest strategy for
60
plants is to sequester or exclude minerals. However, this process causes homeostasis disruption
61
with symptoms such as impaired chlorophyll synthesis 6, diminished photosynthesis 7 or altered
62
root ultrastructure 8. Hyperaccumulator plants use different strategies for metal accumulation in
63
aboveground tissues.
64
Glutathione (GSH) plays a key role in metal scavenging because of the great affinity of metals
65
to bind its thiol (-SH) group and is a precursor of phytochelatins. Almost one third of
66
characterized proteins are metalloproteins, containing essential metals in their active center.
67
During metal homeostasis, metals enter the cell and are involved in two mechanisms: specific
68
chaperones deliver metals to the site of action 9 and chelators sequester excess free metal ions 10.
69
In addition to metal homeostasis, when plants are exposed to an abnormal concentration of
70
metals, they show oxidative stress, whereby the cellular redox balance between anti- and pro-
71
oxidants is further disturbed 11. To cope with oxidative damage, the level of pro-oxidants plays a
72
major role in cellular protection for Cd or Cu toxicity 12. During oxidative defense against metal
73
stress, the primary response of plants is to generate reactive oxygen species (ROS), namely
74
hydrogen peroxide (H2O2), superoxide (O2⋅–), and hydroxyl radical (HO⋅)
75
adversely affect plant growth at morphological, physiological and biochemical levels
76
sedentary, plants require both efficient perception and signaling systems for elevated ROS
77
production when experiencing metal stress. When plants sense Cd or Cu toxicity, NADPH
78
oxidase and lipoxygenase, respectively, are upregulated. Mitogen-activated protein kinases
79
(MAPKs), sensors of ROS, link the perception of metal stress to downstream phosphorylation of
4
ACS Paragon Plus Environment
13
. These species 14
. Being
Page 5 of 29
Environmental Science & Technology
80
transcription factors (TFs). The MAPK cascade plays key roles in signaling activated by
81
different metals 15-17.
82
Metal toxicity and gene regulation
83
The hyperaccumulator Arabidopsis halleri features the expression of > 30 candidate genes as 18
84
compared with the non-accumulator Arabidopsis thaliana on exposure to Zn
85
such as heavy-metal ATPases (HMA) transporters are required to absorb the metals
86
transport Zn2+, Cu+, Cu2+, Cd2+, Pb2+, Ni2+, and Co2+ across biological membranes to load them
87
into the xylem stream with the help of HMAs 20. Plant cells possess different transport systems
88
such as HMAs; natural resistance-associated macrophage proteins (Nramps); ZRT, IRT-like
89
proteins (ZIPs); ATP-binding cassette (ABC) transporters; antiporters of cations; and other metal
90
transporters. Arabidopsis iron-regulated transporter (IRT1), located in the root plasma
91
membrane, can transport a wide range of substrates including Fe
92
Zn tolerance via cross-homeostasis between Zn and Fe in Arabidopsis
93
cation/proton exchangers sequester metals into subcellular compartments such as the vacuole 24.
94
Malate transporter (AtALMT1), in a 14-member family, is critical for Al tolerance in Arabidopsis
95
25
96
Cu, Cd, Fe, Co, Zn, and Cr 26.
21, 22
. Metal-pumps 19
. They
. GSH plays a key role in 23
. ABC transporters or
. Multiple cellular responses and downstream signals exist for different heavy metals such as
97
TFs play key roles in metal toxicity. Basic helix-loop-helix TFs control the expression of key
98
genes such as IRT and ferric oxidase reductase (FRO) functioning in Fe acquisition in tomato 27.
99
Other TFs, bZIP, Myb and zinc-finger proteins are upregulated by Cd exposure
28
. More than
100
131 TFs were significantly upregulated in Thlaspi caerulescens grown under Zn-sufficient
101
conditions
102
element-binding factor and dehydration-responsive element-binding protein were up- or
29
. In addition to TFs, cis-element binding proteins such as ethylene-responsive
5
ACS Paragon Plus Environment
Environmental Science & Technology
103
downregulated with Cd and Cu treatment
104
metal hyperaccumulating abilities.
30, 31
. Therefore, TFs play pivotal roles in maintaining
105
Because of surface-mining operations used on a large scale in the Appalachian region for coal
106
excavation, successful post-mining revegetation processes are needed to restore ecosystem
107
services such as clean water and stable grounds. Use of genomic technologies such as RNA
108
transcriptome sequencing (RNA-seq) has great potential for studying and identifying
109
differentially regulated factors in plants, including hyperaccumulation of metals, when grown
110
under adverse environmental conditions.
111
In this study, we used two different invasive medicinal plant species, goutweed and mugwort,
112
to study the effect of two different mine soils on the transcriptome profiles. We examined 2 mine
113
sites in West Virginia, mine site I and II (MS-I and -II), that contained a range of metals at toxic
114
levels and analyzed the gene expression profile of the two plant species grown in soil from the
115
sites. We profiled the key TFs, metal transporters, and enzymes of biochemical pathways. Both
116
plants exhibited many common regulators across both mine soils. We discuss these results and
117
their implications for bioremediation and reclamation.
118
6
ACS Paragon Plus Environment
Page 6 of 29
Page 7 of 29
Environmental Science & Technology
119
EXPERIMENTAL METHODS:
120
Plant materials and soil collection
121
The invasive plants goutweed (Aegopodium podagraria L.; Ap) and mugwort (Artemisia
122
vulgaris; Av) were collected from local mountains in West Virginia. They were vegetatively
123
propagated from a single genotype by using stolon and rhizomes, respectively, to ensure
124
homogeneity. Young new shoots were chosen for pot experiments. We collected surface soil
125
samples (0-15 cm) from two different mining locations that were mined and reclaimed in 1980
126
(MS-I; 38°10’39.89” N; 81°31’46.29” W) and in 2005 (MS-II; 38°11’47.38” N; 81°38’35.14”
127
W).
128
Soil processing and experimental analysis
129
Mine soils were air-dried and sieved through a 2-mm screen before analysis. Soil samples were 32
130
analyzed for Mehlich-3 extractable elements as described
. Leaf and stem tissues were
131
microwave-digested as described 33. Soil extracts and digested plant materials were analyzed by
132
use of the Optima 8300 ICP-OES Spectrometer (Perkin Elmer, Waltham, MA, USA).
133
Pots were filled with the mine soils, and commercial Promix-filled pots were used as a control.
134
The vegetatively propagated plantlets were transplanted into pots, with three biological
135
replications. Plants were irrigated with no additional nutrients and grown for 40 days before
136
harvesting. Root and shoot tissues were collected from 40-day old plants for RNA extraction and
137
plant element analysis, respectively.
138
RNA extraction, Library preparation and RNA-seq
139
Roots of goutweed and mugwort were washed to remove dirt and stored at -70°C. Total RNA 34
140
was extracted as suggested
by using TRIzol reagent (Invitrogen, Carlsbad, CA, USA) and the
141
RNA MiniPrep kit (Zymo Research, Irvine, CA, USA). Later, on-column DNase digestion was
7
ACS Paragon Plus Environment
Environmental Science & Technology
Page 8 of 29
142
also performed. RNA integrity number (RIN) >8.0 was confirmed by use of the 2100
143
Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA). Beads with Oligo(dT) were used to
144
isolate poly(A) mRNA after total RNA was collected. Furthermore, mRNA was broken down
145
into short fragments by adding fragmentation buffer. Taking these short fragments as templates,
146
random hexamers were used to synthesize the first-strand cDNA. The second-strand cDNA was
147
synthesized by using GEX Second Strand buffer and DNA polymerase I (Illumina, San Diego,
148
CA, USA). The double-strand cDNA short fragments were then purified and resolved with EB
149
buffer for end repair and adding a poly(A). Next, the short fragments were connected with
150
sequencing adapters and purified by agarose gel electrophoresis. Suitable fragments were
151
selected as templates for the PCR amplification. Finally, the library from two biological
152
replications was used for 50-bp paired-end RNA-seq (Genomic Services Laboratory, Huntsville,
153
AL, USA) with the Illumina HiSeq 2500 system (Illumina, Inc., San Diego, CA, USA).
154
RNA-seq and data analysis
155
Poly(A) RNA from two biological replicates in each treatment was sequenced by the Ilumina
156
paired-end sequencing protocol. About 80 to 107 M total reads were generated for each
157
treatment.
158
(http://www.bioinformatics.babraham.ac.uk). The adapter sequence was removed from reads by
159
use of cutadapt v.1.2.1
160
reads were checked and trimmed for A/T homopolymer tails of >8 bases and filtered with >2 Ns
161
present or with quality score
