A new allele for aluminium tolerance gene in barley (Hordeum vulgare L.)

Genetic materials

CXHKSL is a six-rowed Chinese variety that is tolerant to acid soils. The Al-sensitive
variety, Gairdner, is an Australian malting barley. The Al tolerant variety, Dayton,
was used as a control when investigating tolerance mechanisms of CXHKSL. One DH population
consisting of 210 lines derived from a cross between CXHKSL and Gairdner was used
for QTL mapping study.

Al tolerance and root growth

The relative Al tolerance of the different varieties and selected double haploid lines
(DHLs) were evaluated with hydroponic culture methods. Sterilized seeds were germinated
in the dark for 2 days at 4 °C and then 2 days at 28 °C. Root length of the seedlings
was measured and they were placed in an aerated nutrient solution containing 500 ?M
KNO
₃
, 500 ?M CaCl
₂
, 500 ?M NH
₄
NO
₃
, 150 ?M MgSO
₄
, 10 ?M KH
₂
PO4, 2 ?M Fe:EDTA, 11 ?M H
₃
BO
₃
, 2 ?M MnCl
₂
, 0.35 ?M ZnCl
₂
and 0.2 ?M CuCl
₂
. For barley, Al tolerance was estimated by measuring net root length after 4 days
in 0, 1, and 4 ?M AlCl
₃
(pH?=?4.3), respectively. Relative root length (RRL) was estimated as: (net root growth
in Al treatment/net root growth in control solution)?×?100 % 8]. Meanwhile, 4-day old control and 4 ?M AlCl
₃
treated seedlings roots were stained with haematoxylin for 15 min and rinsed for 10 min
to compare the density of Al accumulation at root apices. Haematoxylin could form
a purple-red complex with Al and provides an indirect of non-complexed Al in root
apices, with the intensity of staining correlated with sensitivity of Al toxicity
25].

Al tolerance was also scored using acid soil collected from the Northern Tasmania
(pH?=?4.3). Three seeds of each DHL and parent varieties were sown in the acid soil
in each replicate. Two independent experiments including six replicates were conducted
in April and June 2013, respectively. Four replications were applied in each experiment.
Both root length and root morphology were used to assess Al tolerance. Root length
(mm) of each seedling was measured seven days after sowing. Meanwhile, root tips were
screened for the absence or presence of thickening caused by Al toxicity.

Assaying citrate efflux and malate efflux from root apices

Seedlings were grown for 4 days in the nutrient solution described above (without
added AlCl
₃
). To study if the expression of HvAACT1 need longer Al treatment duration, half of the plants of each genotype were subjected
to 0.2 mM CaCl
₂
solution containing 10 ?M AlCl
₃
(pH?=?4.3) for overnight pre-treatment. Ten root apices (3-5 mm) with 4 replicates
were excised from the same line and washed in 1 ml 0.2 mM CaCl
₂
solution (pH?=?4.3) for on a platform shaker (60 rpm). After 30 min washing, 1 ml
0.2 mM CaCl
₂
solution (pH?=?4.3) with 30 ?M AlCl
₃
was added and shaken for 2 h at 60 rpm. The solutions were centrifuged to dryness
on a rotary vacuum drier for citrate efflux detection. The enzyme assay used to determine
citrate concentration is described by Wang et al. 16]. The initial citrate content in each sample was calculated from a standard curve.
Malate concentration was measured with an enzyme assay as described previously 26].

Molecular marker analysis

Three primer pairs were used to investigate allelic variation in the 5’UTR of HvAACT1. These were to detect the presence or absence of a ~1 kb transposon-like insertion
previously described in Al-tolerant genotypes of barley. The first pair of primers
was from Fujii et al. 23] with forward sequence 5’-GGTCCAACACTCTACCCTCCTT and reverse 5’-GGTGCGAG -TTGCCCCTAGCTATTA.
The second pair of primers described by Bian et al. 18] was forward 5’-CTTCATTTCAACCAAGCACTCC and reverse 5’-GCTTTTGGTCGAACAAA- GTATCG. The
third pair of primers was designed to amplify a slightly larger fragment that included
the above two pairs of primers comprised, forward 5’-TGTCGATATGGTGCTCTT -CG and reverse
5’-AGCTCCATGACAATTCTGGG. PCR reactions were performed at 20 ?l-volume including 10 ?l
HotstarTaq
^TM
master mix (Qiagen), 2 ?l primer mix (1:1 mix of forward and reverse primers at 10
nM), 3 ?l DNA template, and 5 ?l H
₂
O. Cycling conditions were as follows: 1 cycle of 1 min at 95 °C, 35 cycles of 1 min
at 95 °C, 30 s at 60 °C, 40 s at 72 °C, and finally with an extension step of 1 min
at 72 °C. All PCR reactions were run at C1000
^TM
Thermal cycler (BIO-RAD). PCR products were separated at 1 % agarose and visualized
by staining with 1 % Red safe under Gel Doc
^TM
XR
⁺
imagining system (BIO-RAD).

Another HvAACT1-specific marker, HvMATE-21, was used to genotype the population as well as three closely HvAACT1-linked SSR markers: Bmac310, Bmag353 and HVM03 14], 16], 19]. HvMATE-21 was a PCR marker that detected the presence or absence of a 21-bp fragment in
the 3’ UTR of HvAACT1. PCR reactions were carried out in a total volume of 15 ?l containing 25?~?30 ng
genomic DNA, 0.5 M of forward and reverse primers, 7.5 ?l GoTaq® Hot Start Colorless
Master Mix, 2X (Promega). The amplification of SSRs were performed by: 1 cycle of
3 min at 94 °C, 35 cycles of 1 min at 94 °C, 1 min at the annealing temperature 55 °C
and 1 min at 72 °C, with a final extension step of 5 min at 72 °C. The PCR profiles
for HvMATE-21 were almost the same as that for SSR markers except the annealing temperature
was 60 °C. All PCR reactions were run on Mastercycler Gradient 5331 (Eppendorf AG,
Germany). The PCR products were separated on 5 % denatured polyacrylamide gels and
visualized by a rapid silver staining method 27].

Isolation and sequence analysis of coding region of HvACCT1 gene in CXHKSL and Dayton

The published complete coding DNA sequence (CDS) of HvAACT1 gene (Genebank: AB302223.1) was retrieved from the National Center for Biotechnology
Information (NCBI, http://www.ncbi.nlm.nih.gov/gene) and aligned with barley reference genome data using IPK blast server (http://webblast.ipk-gatersleben.de/barley/viroblast.php). Based on the best hit sequence, a total of 4 pairs of primers (Additional file
2: Table S2) were used to amplify the whole HvACCT1 open reading frames (ORFs). The amplified PCR products was purified and cloned with
pGEM®-T Vector System (Promega). The final CDS were constructed using sequencing results
from 12 independent clones (3 clones for each pair of primers). Sequence analysis
was completed with software DNAMAN (version 7.0; Lynnon Biosoft, USA). The sequence
data of CDS have been deposited to NCBI Genbank Database with accession number of
KU725980 for variety CXHKSL and KU725981 for Dayton.

HvAACT1 expression

RNA was isolated from root apices (also from plants used for citrate efflux measurement)
by RNeasy
^TM
plant kit (Qiagen) and purified by inclusion of RNase-free DNase (Qiagen). One microgram
total RNA was used to synthesize cDNA by reverse transcriptase system (Invitrogen).
1.0 ?l oligo primer was added into 11.5 ?l reaction mixture including 1 ?g RNA. The
mixture was incubated at 70 °C for 10 min, and transferred to ice immediately. Each
aliquot included 4 ?l buffer, 2 ?l 0.1 M DTT, 1 ?l dNTP mix, and 0.5 ?l superscript
III Reverse Transcriptase was added into the mixture, and incubated at 42 °C for 1 h.
RNA degradation step was performed by addition of 0.25 ?l RNase H (Thermo Scientific™)
and incubated at 37 °C f or 30 min.

Quantification real time polymerase chain reaction(RT-PCR) was run in a C1000TM Thermal
cycler (BIO-RAD) with 10 ?l reaction mixture containing 4.0 ?l of cDNA diluted to
1:40, 5 ?l of SYBR Green Jumpstart Taq Readymix (Sigma) and 1 ?l primer mix (1:1 mix
of forward and reverse primers at 10 nM). Three pairs of primers used to measure expression
of HvAACT1 (HvAACT1-forward 5’-AGCAGCCAAGACCTTGAGAA and reverse 5’-AGCAG GAATCCACAACCAAG; New-HvAACT1-1-forward ACGGGGCTCTACCTCTTT -GT and reverse 5’-GGCAATAGAAACACCAACAGC; New-HvAACT1-2-forward CTGTGTCACTC TGGCATCGT, and reverse 5’-AAGCTGCAGAACACGAGAGGT). The constitutively
expressed barley glyceraldehyde-3-phosphate dehydrogenase (HvGAPDH) gene and barley
homologous to eukaryotic translation elongation factor 1A (HveEF-1A) gene was used
as reference genes. The sequences of primers are as follows: HvGAPDH-forward: 5’-GTGAGG CTGGTGCTGATTACG and reverse 5’-TGGTGCAGCTAGCATTTGAGAC, HveEF-1A-forward 5’-TTTCACTCTTGGTGTGAAGCAGAT and reverse 5’-GACTTCCTTCACGAT-TTCATCGTAA. Cycling
conditions were 3 min at 95 °C, followed by 40 cycles at 95 °C for 10 s, 60 °C for
20 s, 68 °C for 10 s. At the end, a melting curve of the amplified fragments was produced
by increasing the temperature every 0.5 °C from 60 °C to 95 °C.

Data analysis and QTL mapping

All phenotypic data was analysed by SPSS software package (Version 20.0, IBM), including
all basic statistics calculation and Chi-Square Goodness of Fit Test. For genetic
linkage map analysis, the genetic distances between molecular markers were calculated
using software JoinMap 4.0 28]. The mean values of root lengths of DHLs were used to detect QTL affecting root length
under Al toxicity with software MapQTL6 29]. Interval mapping (IM) method was first used to identify the major QTL. By selecting
significantly linked markers as cofactors, multiple QTL mapping (MQM) mapping method
based on the multiple-QTL model was used. A set of 1000 permutations was performed
to identify the LOD threshold corresponding to a genome-wide false discovery rate
of 5 % (P??0.05) 29].