- Email: [email protected]
Variability in amylose content of Bangladeshi rice cultivars due to unique SNPs in Waxy allele
Accepted Manuscript Variability in amylose content of Bangladeshi rice cultivars due to unique SNPs in Waxy allele Saima Shahid, Rokeya Begum, Samsad Razzaque, Jesmin, Zeba I. Seraj PII:
S0733-5210(16)30128-X
DOI:
10.1016/j.jcs.2016.07.006
Reference:
YJCRS 2178
To appear in:
Journal of Cereal Science
Received Date: 19 August 2015 Revised Date:
2 July 2016
Accepted Date: 7 July 2016
Please cite this article as: Shahid, S., Begum, R., Razzaque, S., Jesmin, , Seraj, Z.I., Variability in amylose content of Bangladeshi rice cultivars due to unique SNPs in Waxy allele, Journal of Cereal Science (2016), doi: 10.1016/j.jcs.2016.07.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Variability in amylose content of Bangladeshi rice cultivars due to unique SNPs in Waxy allele
2
Saima Shahida,c,1, Rokeya Beguma,c, Samsad Razzaquea, Jesminb and Zeba I. Seraja*
3
a
4
Dhaka, Dhaka-1000, Bangladesh
5
b
6
Bangladesh
Plant Biotechnology Laboratory, Department of Biochemistry and Molecular Biology, University of
RI PT
Department of Genetic Engineering and Biotechnology, University of Dhaka, Dhaka-1000,
7 8
c
Contributed equally
SC
9
*Corresponding Author: Zeba I. Seraj, Plant Biotechnology Laboratory, Department of Biochemistry
11
and Molecular Biology, University of Dhaka, Dhaka-1000, Bangladesh. Tel: +880-2-8614708; fax:
12
+880-2-8615583/9127051. Email: [email protected].
M AN U
10
13 14
1
15
Life Sciences, Penn State University, University Park, PA 16802, USA
TE D EP AC C
16
Present address: Plant biology PhD program, Department of Biology, and the Huck Institutes of the
1
ACCEPTED MANUSCRIPT List of Abbreviations
18
AC = Amylose content
19
bp =Base pairs
20
cDNA = Complementary DNA
21
CTAB = Cetyltrimethylammonium bromide
22
DNA =Deoxyribonucleic acid
23
dNTP = Deoxynucleoside triphosphate
24
ESE = Exonic splicing enhancer
25
GBSS-I = Granule-bound starch synthase I
26
G3PDH = Glyceraldehyde-3-phosphate dehydrogenase
27
indel = Insertion/deletion
28
kb = Kilobase pairs
29
mM =Millimolar
30
mRNA = Messenger RNA
31
ng = Nanogram
32
pmol = Picomole
33
QTL = Quantitative trait loci
34
RNA = Ribonucleic acid
35
RT-PCR = Reverse transcription polymerase chain reaction
36
SF2/ASF = Splicing factor-2/ alternative splicing factor
37
SNP = Single nucleotide polymorphism
38
UTR =Untranslated region
39
Wx = Waxy
40
µl = Microlitre
41
µM = Micromolar
AC C
EP
TE D
M AN U
SC
RI PT
17
42 43 44 2
ACCEPTED MANUSCRIPT 45 46
Abstract
47 Waxy gene (Granule Bound Starch Synthase I) is responsible for amylose synthesis in the rice
49
endosperm. Several mutations in this gene have been shown to be responsible for variable amylose
50
content (AC) phenotypes. The G/T mutation in 5′ splice site of Waxy intron 1 has been traced as the
51
origin of the glutinous rice phenotype and differentiates low AC from non-glutinous intermediate/high
52
AC rice. Sequencing of Waxy promoter and 5′ noncoding regions from 22 rice cultivars showed that
53
the evolutionary pattern of all Bangladeshi non-glutinous and most glutinous rice accessions are in
54
line with the general patterns of South and Southeast Asia. However, three cultivars Khara Beruin,
55
Modhu Beruin White and Kathali Beruin Red with low to very low amylose lacked the G/T splice site
56
mutation. These were more closely related to non-glutinous cultivars based on their SNP patterns in
57
promoter and noncoding regions. Further sequencing revealed a unique C deletion at a pyrimidine
58
tract of intron 5 of these three cultivars that may cause slippage of intron splicing. Additional SNPs at
59
intron 9 and 10 were also identified among these cultivars. These Bangladeshi-genotype-specific
60
mutations could be the cause of waxy or low amylose phenotypes in these glutinous accessions.
SC
M AN U
TE D
Key words: Amylose content; Waxy gene; Glutinous and non-glutinous rice; SNPs
EP
62
AC C
61
RI PT
48
3
ACCEPTED MANUSCRIPT 1. Introduction
64
Starch in rice endosperm contains amylose and amylopectin polysaccharides. Amylose is principally a
65
linear molecule containing α(1→4) linked glucose units and makes up approximately 0-30% of total
66
starch. In contrast amylopectin is a branched molecule which contributes to approximately 70-100%
67
of total starch in rice endosperm (Martin and Smith, 1995). Higher amylose levels (20–30%) are
68
observed in many indica rice varieties in South Asia (Morishima et al., 1992). Lower amylose levels
69
(10–20%) are more common in japonica varieties that predominate in East Asia (Yamanaka et al.,
70
2004). Amylose content (AC) can be thus used to classify milled rice samples into different categories
71
such as glutinous/waxy (0–5% amylose), low AC (6–18%), intermediate AC (19–23%), or high AC
72
(>23%) types (Bergman et al., 2004). Low AC is usually associated with tender, cohesive, and glossy
73
cooked rice while high AC is associated with firm, fluffy, and separated grains in cooked rice (Juliano
74
et al., 1981). Therefore, AC is considered as one of the major characteristics for assessing rice
75
cooking and eating qualities. In addition, low AC rice varieties have long been used as a tool of
76
improving grain quality through conventional breeding as these represent an intermediate type
77
between glutinous and non-glutinous rice varieties (Dong, 2000; Sato et al., 2002).
SC
M AN U
TE D
78
RI PT
63
Genetic and molecular marker-based QTL analyses have revealed that the wide range of AC variation
80
in rice endosperm is mainly controlled by a major locus (Wx or Waxy gene) and multiple minor loci
81
(Fan et al., 2005; Inukai et al., 2000). The Wx gene in rice encodes the granule-bound starch synthase
82
I (GBSS-I) enzyme responsible for amylose biosynthesis (Okagaki and Wessler, 1988). The origin of
83
the glutinous phenotype such as those found in japonica rice has been traced to the Wxb allele, which
84
contains a G/T mutation in intron 1 splice donor site of the Wx gene (Olsen and Purugganan, 2002).
85
Presence of this splice donor site mutation leads to inefficient splicing of Wx pre-mRNA followed by
86
reduced expression of functional GBSS-I (Cai et al., 1998). In contrast, non-glutinous phenotype
87
observed in many indica rice varieties has been traditionally associated with Wxa allele, which lacks
88
the intron 1 G/T mutation. However many Northeast Asian rice varieties carrying this intron 1 G/T
89
mutation still show non-glutinous phenotype. This suggests that partial suppression of the intron 1
90
G/T mutation may have an important role in the development of non-glutinous rice of this region
AC C
EP
79
4
ACCEPTED MANUSCRIPT (Olsen and Purugganan, 2002). Amylose content of rice endosperm is therefore a major domestication
92
trait that has evolutionary importance. Among the minor alleles affecting amylose content, a
93
polymorphic CTn microsatellite in the 5′-untranslated region (UTR) of Wx exon 1 has been shown to
94
be correlated with various AC classes (Ayres et al., 1997). Other SNPs in coding regions of different
95
Wx alleles lead to decreased AC by dropping the binding efficiency of GBSS-I to starch granules,
96
changing amino acid at respective sites (Liu et al., 2009; Sato et al., 2002) and altering expression
97
pattern of GBSS-I in endosperm and anthers (Mikami et al., 1999). In addition to Wx, mutations in
98
genes such as Dull (Zeng et al., 2007) and shrunken (Asaoka et al., 1993) also indirectly affect AC in
99
rice.
SC
RI PT
91
100
Bangladesh is the world’s fourth largest rice producing country. It is considered as an enriched rice
102
germplasm reservoir with over 6500 wild, landraces and modern high yielding varieties (Elias et al.,
103
2011). Majority of the rice species in Bangladesh are indica varieties with high amylose content and
104
non-glutinous phenotype (Olsen and Purugganan, 2002). Several glutinous rice varieties, which are
105
locally known as ‘Beruin’ or sticky rice, are also cultivated in the Northeastern region of Bangladesh
106
(Dipti et al., 2003). These Beruin cultivars generally have low amylose content and are popular for
107
making traditional dishes on special occasions. However the molecular basis of low amylose content
108
in these local Beruin varieties is unknown. In this study we examined the nucleotide diversity at the
109
Wx locus of selected Bangladeshi glutinous Beruin and non-glutinous rice cultivars to identify the
110
underlying cause of AC variation in these genotypes. We also compared these Bangladeshi varieties
111
with known Wx haplotypes in an attempt to understand their evolutionary pattern across Asia.
113
TE D
EP
AC C
112
M AN U
101
2. Materials and Methods
114 115
2.1. Plant materials
116
Seeds of eleven Beruin cultivars, which are representative of the diverse traditional landraces unique
117
to the Northeastern region of Bangladesh (Sylhet), were collected from the local farmers. An equal
118
number of non-glutinous cultivars was then selected to allow identification of SNPs underlying the
5
ACCEPTED MANUSCRIPT glutinous phenotype in Bangladeshi cultivars (Fig. 1A and 1B). Nine of the selected non-glutinous
120
cultivars were from the same region as the Beruin cultivars. Two other popular aromatic cultivars,
121
Kalijira and Kataribhog, were collected from the Northwestern region of Bangladesh (Dinajpur) to
122
expand the genetic diversity of sampled non-glutinous cultivars. All of the selected cultivars are
123
traditional transplanted Aman or monsoon season varieties. The collected seeds were multiplied
124
during the planting season in the net house. Three plants of each cultivar were grown in a single pot to
125
collect immature seeds.
RI PT
119
126 2.2. Estimation of amylose content
128
The amylose content (AC) of all cultivars was determined according to previously published protocol
129
(Juliano et al., 1981). Nazirshail rice flour (AC 25%, Bangladesh Rice Research Institute) was used as
130
a control and potato starch flour (E. Merck, Germany) was used for generating the standard curve.
M AN U
131
SC
127
2.3. Genomic DNA Extraction
133
Genomic DNA was extracted from leaves of 3-week-old pooled plants (10-12 plants per cultivar)
134
using CTAB extraction method and in some cases by using the Qiagen DNeasy® plant mini kit. DNA
135
of all cultivars was quantified by spectrophotometry. Quality of extracted DNA was checked by 1%
136
agarose gel electrophoresis and compared with known concentration of Lambda DNA (Invitrogen,
137
USA).
EP AC C
138
TE D
132
139
2.4. PCR amplification and DNA sequencing
140
A 2.7 kb upstream regulatory region of the Wx gene including promoter, exon 1, intron 1 and part of
141
exon 2 was PCR amplified and sequenced in all 22 Bangladeshi rice cultivars. PCR amplification was
142
performed in two steps to amplify overlapping ~1.5 kb segments of the above upstream regulatory
143
region using primers described by Olsen and Purugganan (2002) (Supplementary Table 1). For each
144
primer set, 15 µl PCR reaction mixture containing 60 ng of DNA template, 1 × PCR buffer, 1.67 mM
145
MgCl2, 100 µM of each dNTP, 0.33 pmol Forward /Reverse primer, 1 unit Taq DNA polymerase and
146
0.1 unit Pfx® DNA polymerase (Invitrogen, USA) was used. Amplification conditions were 94 °C for
6
ACCEPTED MANUSCRIPT 147
5 min, followed by 35 cycles of 95 °C for 1 min, 65.2 °C for 1 min, 72 °C for 1 min and a final
148
extension of 72 °C for 8.30 min.
149 For further characterization of the Wx coding region in Shamudrophena, Kathali Beruin Red, Modhu
151
Beruin White, Khara Beruin and Mou Beruin, two different segments were amplified and sequenced
152
using four sets of primers (Supplementary Table 1). The first segment (~0.2 kb) included part of exon
153
2, which is known to contain a 23 bp duplication in some glutinous varieties. This segment was
154
amplified using primer pair sequences Glu-23F and Glu-23R as described by Wanchana et al. (2003).
155
The other ~2 kb segment including exons 3-10 was amplified using three overlapping primer pairs
156
(Supplementary
157
(http://frodo.wi.mit.edu/primer3/) based on sequence of Nipponbare Wx locus (OS06G0133000) in
158
Gramene database (www.gramene.org). PCR amplification conditions were the same as for the Wx
159
upstream regulatory region except the annealing temperature, which was optimized for each primer
160
pair (Supplementary Table 1).
1).
These
were
designed
using
primer3
program
TE D
161
primers
M AN U
Table
SC
RI PT
150
All PCR amplicons were purified using QiAquick gel extraction kit (Qiagen, Germany) and directly
163
sequenced in ABI 3730XL genetic analyzer using both forward-reverse and internal primers
164
(Supplementary Table 1). Each fragment was sequenced twice for SNP confirmation. All of the
165
sequences from this study have been deposited in GenBank (accession no. JF834042.1-JF834063.1
166
and KP675770-KP675774).
AC C
167
EP
162
168
2.5. Sequence analysis
169
Wx upstream regulatory region sequences from 22 Bangladeshi rice landraces were compared to 18
170
previously reported haplotypes across Asia (Genbank accessions AY136760–AY136784). All
171
multiple sequence alignments were generated using ClustalW (Thompson et al., 2002). SNP/indels
172
were identified using TASSEL (Bradbury et al., 2007). Neighbour-joining tree of aligned Wx
173
upstream regulatory region (2.7 kb) sequences was constructed using MEGA 6.06 software package
174
(Tamura et al., 2013) with Kimura two-parameter model and complete deletion of alignment gaps.
7
ACCEPTED MANUSCRIPT Bootstrap test with 1000 replicates was performed to ensure confidence in phylogeny. A median-
176
joining haplotype network was also generated based on the aligned sequences using SplitsTree 4.13.1
177
(Huson and Bryant, 2006). Motif change patterns due to SNPs in Wx promoter region and intron 1 of
178
Bangladeshi cultivars were predicted using the PLACE database (www.dna.affrc.go.jp/PLACE).
179
Amino acid changes due to SNP in sequenced Wx coding exons from selected Bangladeshi landraces
180
were predicted using the ExPASy translation tool (http://web.expasy.org/translate/). SplicePredictor
181
(http://bioservices.usd.edu/splicepredictor/) was used to predict splice-site modifying SNPs in intronic
182
regions of sequenced varieties. RegRNA (http://regrna.mbc.nctu.edu.tw/) was used to predict cis-
183
regulatory elements involved in mRNA splicing or transcriptional regulation.
184
SC
RI PT
175
2.6. Semi-quantitative RT-PCR and gene expression
186
Total RNAs from rice endosperm at 16-18 days after flowering were extracted using TRIZOL. First
187
strand cDNA were synthesized from total RNA following manufacturer’s protocol (Invitrogen, USA).
188
G3PDH (glyceraldehyde-3-phosphate dehydrogenase) gene-specific primers (Supplementary Table 1)
189
were used to optimize the cDNA concentration for estimating the Wx mRNA level. Nanodrop
190
spectrophotometer was used to measure and optimize the working concentration of the cDNA. Wx
191
mRNA level was assessed using cycle-dependent semi-quantitative RT-PCR with the WxRT1 primer
192
pair (Supplementary Table 1) for 30, 32 and 35 cycles. Expression levels of samples were estimated
193
by visual inspection after electrophoresis in a 0.9% agarose gel based on the band intensity of the
194
loading control G3PDH. Two sets of primer pairs - Wx-RT2 and Wx-RT3 (Supplementary Table 1)
195
were used to confirm the G/T SNP at the 5´-splice donor site of Wx mRNA of Bangladeshi cultivars.
196
Wx-RT2 primer pair was used to determine the splicing efficiency of the first intron in the selected
197
landraces. Nipponbare was used as a control as it contains G/T SNP at intron 1 splice donor site of Wx
198
gene (Liu et al., 2009).
AC C
EP
TE D
M AN U
185
199 200
3. Results
201 202
3.1. Amylose content of Bangladeshi cultivars
8
ACCEPTED MANUSCRIPT In order to identify the molecular basis of glutinous rice in Bangladesh, we first evaluated the amylose
204
content (AC) of 22 Beruin and non-Beruin traditional landraces that are locally considered as
205
glutinous and non-glutinous cultivars respectively. These selected cultivars were sampled from
206
intensive rice cultivation zones in Northeastern and Northwestern regions of Bangladesh that are
207
known to harbour many diverse landraces (Elias et al., 2011). Based on their AC, the selected
208
cultivars were assigned into three major categories - intermediate, low and glutinous/waxy. All non-
209
Beruin Bangladeshi cultivars in this study had intermediate AC ranging from 19.1% to 23.3%
210
(Fig.1A), confirming that these are indeed non-glutinous. In contrast, only seven out of eleven Beruin
211
varieties, traditionally associated as glutinous or sticky rice, had amylose content within the waxy
212
range (0-5.4%), as shown in Fig. 1B. Among the remaining four Beruin varieties, Kathali Beruin Red
213
and Kalo Beruin belonged to the low AC group (AC 13.7% and 8.2% respectively). The two other
214
cultivars Modhu Beruin Red and Push Beruin (AC 23.4% and 22.6% respectively) had intermediate
215
AC, similar to the Bangladeshi non-glutinous cultivars.
M AN U
SC
RI PT
203
216
3.2. Nucleotide variation at Wx upstream regulatory region of Bangladeshi cultivars
218
We compared the ~2.7 kb Wx upstream noncoding region sequences from selected Bangladeshi
219
cultivars to Olsen and Purugganan (2002) reported progenitor haplotypes F (non-glutinous) and G
220
(glutinous). These progenitor haplotypes differ only in the intron 1 G/T splice donor site mutation
221
(Olsen and Purugganan, 2002). Overall, Bangladeshi cultivars had 29 single nucleotide
222
polymorphisms (SNP) and 11 insertions/deletions (indel) in the sequenced upstream regulatory region
223
(Fig. 1C). Out of these, 15 SNP and 3 indels were found in the promoter regions and 13 SNP and 7
224
indels in intron 1. These SNP and indels in promoter and intron 1 were more frequent in non-
225
glutinous cultivars than the Beruin cultivars. In total, seven different microsatellite alleles: CT8, CT10,
226
CT11, CT12, CT16, CT17 and CT18, were identified in the Wx exon 1 (encoding 5′ untranslated region) of
227
Bangladeshi cultivars (Figure 1C). An additional SNP within Wx exon 2 that also codes for 5′ UTR
228
was observed only in one of the non-glutinous Bangladeshi cultivars.
229
All of the Bangladeshi non-glutinous cultivars in this study had G-SNP at the 5′ splice donor site of
AC C
EP
TE D
217
9
ACCEPTED MANUSCRIPT Wx intron 1 (Fig. 1C) similar to the known non-glutinous haplotype F. The CT10 or CT11 alleles within
231
Wx exon 1 were observed in six of these non-glutinous Bangladeshi cultivars. The CT17 allele was
232
present in cultivars with variable AC, ranging from intermediate AC in Bangladeshi non-glutinous
233
cultivars to low and waxy AC content in the Beruin cultivars (Fig. 1). The CT12/ CT18 and CT8/ CT16
234
alleles were specific to non-glutinous and glutinous Bangladeshi cultivars, respectively.
RI PT
230
235
G/T SNP at the 5' splice site of Wx intron 1 could explain the waxy or low AC phenotype (AC 4.4-
237
8.2%) of six out of eleven Bangladeshi Beruin cultivars. These cultivars also carried CT17 allele in
238
exon 1 similar to the known glutinous progenitor haplotype G (Fig. 1C). Among these six Beruin
239
cultivars, Pak Beruin, Kathali Beruin white and Kalo Beruin had an additional novel G/A SNP at
240
position 718 in the Wx promoter region (Fig. 1C). The remaining five Beruin cultivars lacked the G/T
241
SNP in intron 1 splice donor site. Among these, Push Beruin and Modhu Beruin Red also showed
242
intermediate amylose phenotype (AC 22.6% and AC 23.4% respectively) like non-glutinous
243
Bangladeshi cultivars in this study (Fig. 1A). These varieties also had other nucleotide variations in
244
Wx regulatory region that were similar to the Bangladeshi non-glutinous cultivars (Fig. 1C). This
245
confirmed that Push Beruin and Modhu Beruin Red might be locally misnamed as glutinous varieties
246
even though they have non-glutinous phenotype.
M AN U
TE D
EP
247
SC
236
The other three Beruin cultivars that lacked the G/T SNP in splice donor site (Khara Beruin, Modhu
249
Beruin White and Kathali Beruin Red) had waxy or low AC phenotype (Fig. 1B). Among these only
250
Khara Beruin showed several unique mutations in the Wx upstream noncoding region (T/A SNP at
251
216 position in promoter, G/T SNP at position 1934 in intron 1). This indicated that polymorphisms
252
other than G/T SNP in intron 1 splice donor site and CTn allele in exon 1 may play a major role in
253
regulating Wx gene expression in these three genotypes.
AC C
248
254 255
3.3. Evolutionary relatedness of Bangladeshi cultivars
256
An unrooted phylogenetic tree based on the aligned ~2.7 kb Wx upstream regulatory region sequences
257
of selected Bangladeshi cultivars and 18 other reported haplotypes (Olsen and Purugganan, 2002) was
10
ACCEPTED MANUSCRIPT constructed to explore the evolutionary relationships among these accessions. These 18 reported
259
haplotypes represent the landrace diversity across Asia. In the phylogenetic tree, six Bangladeshi
260
glutinous cultivars (Akia Beruin White, Akia Beruin Red, Mou Beruin, Kalo Beruin, Kathali Beruin
261
White and Pak Beruin) clustered with reported glutinous haplotypes G, I, K, L and Q at high bootstrap
262
(98%) value (Fig. 2). Pak Beruin, Kalo Beruin, Kathali Beruin White along with a non-glutinous
263
cultivar Shamudrophena also formed a sub clade depending on the G/A SNP in position 718 of the Wx
264
promoter region (Fig. 2). These cultivars may carry the same Wx allele or with an additional mutation
265
(718 G/A).
RI PT
258
SC
266
On the other hand, five glutinous Beruin cultivars that lacked the intron 1 G/T SNP were grouped
268
with reported non-glutinous haplotypes and Bangladeshi non-glutinous cultivars. Intermediate AC
269
cultivars Push Beruin and Modhu Beruin Red clustered with reported non-glutinous haplotypes B, A
270
and Bangladeshi non-glutinous cultivars (Fig. 2). Waxy cultivar Khara Beruin clustered with reported
271
non-glutinous haplotypes E and D at 64% bootstrap and with non-glutinous haplotype C at 93%
272
bootstrap (Fig. 2). Based on the aligned Wx upstream regulatory region sequences, waxy AC Modhu
273
Beruin White and low AC Kathali Beruin Red appeared to be very close to intragenic recombinant
274
haplotype R and non-glutinous haplotype S, respectively, even though neither shared similar amylose
275
phenotype with the corresponding haplotypes (Fig. 2). Bangladeshi non-glutinous cultivars were
276
clustered with reported non-glutinous haplotypes but were more close to the haplotype F, A and B
277
(Fig. 2). In addition, Chinigura, Kalijira and Kataribhog distinctly clustered with non-glutinous
278
progenitor haplotype F at 48% bootstrap (Fig. 2). It might be possible that the Bangladeshi non-
279
glutinous cultivars originally carried the non-glutinous progenitor Waxy allele and then evolved into
280
haplotypes A and B through subsequent mutations at promoter region and intron 1.
AC C
EP
TE D
M AN U
267
281 282
3.4. Nucleotide variations at coding region of Wx gene in selected Beruin cultivars
283
Since G/T SNP at intron 1 splice donor site was not enough to explain the waxy AC of Modhu Beruin
284
White and Khara Beruin and low AC of Kathali Beruin Red, we further sequenced two segments in
285
the coding region of Wx gene. Mou Beruin (AC 4.4%) and Shamudrophena (AC 23.4%) were
11
ACCEPTED MANUSCRIPT considered as references for Bangladeshi glutinous and non-glutinous cultivars respectively for this
287
analysis. The first sequenced segment in Wx gene included part of coding region of exon 2 (~196 bp)
288
which has been reported to contain a 23 bp frame-shift duplication in some glutinous varieties that
289
lead to non-functional Wx proteins (Mikami et al., 2008; Wanchana et al., 2003). Among the selected
290
cultivars, this 23 bp duplication was only identified in Mou Beruin, consistent with its glutinous
291
phenotype (Table 1). The second segment analyzed in Wx gene spanned ~1.9 kb genomic region
292
including 3' end of exons 2-10 and introns 2-10. Only the non-glutinous reference Shamudrophena
293
among the five cultivars carried A/C SNP in exon 6 (position 671 from start codon), which resulted in
294
nonsynonymous amino acid change from tyrosine to serine (Table 1). This SNP has been shown to be
295
associated with intermediate AC (21-24%) previously (Dobo et al., 2010). Two other coding region
296
SNPs (T/C in exon 9, C/T in exon 10) were identified in the three selected Beruin cultivars but not in
297
the non-glutinous or glutinous reference (Table 1). Exon 9 SNP (T/C, position 1109 from start codon)
298
was found in all three selected cultivars (Table 1). The exon 9 SNP results in a synonymous change in
299
amino acid and has been previously reported as a silent mutation (Larkin and Park, 2003). Exon 10
300
SNP (C/T) was found only in low AC Kathali Beruin Red (Table 1). Exon 10 SNP (C/T) caused a
301
nonsynonymous change from a nonpolar amino acid proline (CCT) to polar amino acid Serine (TCT),
302
and it has been reported that this SNP is linked to high amylose content (Larkin and Park, 2003).
303
Therefore these exon-specific mutations were not directly responsible for the variant AC phenotype
304
traits observed for the selected Beruin cultivars.
SC
M AN U
TE D
EP
AC C
305
RI PT
286
306
Most interestingly, Kathali Beruin Red, Modhu Beruin White and Mou Beruin had overall four SNPs
307
and two indels in Wx introns which were absent in the Bangladeshi non-glutinous and glutinous
308
references (Table 1). Among these variations, C deletion in intron 5, G/A (at positions 81 and 95 from
309
5' splice site of intron 10) and ATA deletion at position 104-106 in intron 10 were identified in all
310
these three cultivars. On the other hand, G/A in intron 10 (position 118) was found only in low AC
311
Kathali Beruin Red and A/G at position 29 from 5' splice site of intron 9 was only observed in the
312
waxy AC Khara Beruin and Modhu Beruin White. A search for putative cis-elements that might be
313
affected by these SNPs/indel revealed that the C indel was located within a poly-C tract in intron 5,
12
ACCEPTED MANUSCRIPT 314
and this indel resulted in absence of a putative INTRONLOWER motif in these three cultivars
315
(Supplementary Table 2). Also, the G/A indel in intron 10 (at position 95 from 5' splice site) resulted
316
in a missing putative ESE-SF2/ASF motif (Exonic Splicing Enhancer) in these three cultivars
317
(Supplementary Table 2).
RI PT
318 Further homology search using NCBI BLASTn revealed that the sequenced coding regions of Khara
320
Beruin and Modhu Beruin White shared ~99% identity with previously reported Wxop (opaque) allele
321
containing cultivars ARC10818, ACC35618 and ARC6622 (Genbank accessions AB281448.1,
322
AB281447.1, and AB281453.1) from Nepal and India (Mikami et al., 2008, 1999). These Wxop
323
accessions did not contain the G/T SNP at the 5´ splice donor site of intron 1 and also had
324
nonsynonymous A/G SNP in exon 4 at position 521 from start codon (Mikami et al., 2008, 1999).
325
However the latter exon 4 SNP was not present in the Khara Beruin and Modhu Berun White (Table
326
1). Comparison of the rice grains revealed that only Khara Beruin had a completely opaque or chalky
327
endosperm, while Modhu Beruin White had partially chalky endosperm (Fig. 3), which supports the
328
low to very low amylose content of these varieties.
M AN U
TE D
329
SC
319
3.5. Wx gene expression in selected Beruin cultivars without G/T SNP in splice donor site
331
Semi-quantitative RT-PCR was used to estimate total expression levels of Wx gene in Modhu Beruin
332
White (waxy AC), Kathali Beruin Red (low AC), Push Beruin and Modhu Beruin Red (intermediate
333
AC). Nipponbare was used as a control since it is known to show decreased splicing efficiency of Wx
334
transcript due to G/T SNP at intron 1 splice donor site (Liu et al., 2009). The results showed that Wx
335
transcript levels in Push Beruin, Modhu Beruin Red and Kathali Beruin Red were relatively much
336
higher compared to the control Nipponbare, (Fig. 4A). This suggests that the splicing efficiency of Wx
337
transcript is not affected in Push Beruin, Modhu Beruin Red as well as Kathali Beruin Red. This
338
observation is consistent with the absence of G/T SNP at intron 1 and the intermediate-to-low AC in
339
these cultivars. The expression level of Wx transcript in Modhu Beruin White (intron 1 splice site
340
mutation absent) was also higher than that of Nipponbare but lower than the aforementioned three
341
Beruin cultivars (Fig. 4A). The transcript level however correlates with the waxy AC of Modhu
AC C
EP
330
13
ACCEPTED MANUSCRIPT Beruin White.
343
RT-PCR with primers specific for unspliced variant of Wx transcript (Liu et al., 2009) confirmed the
344
absence of intron 1 containing 1.1 kb fragment in all of the selected Beruin cultivars (Fig. 4B). A
345
second set of RT-PCR primers previously described by Prathepha (2007) was used to re-confirm the
346
absence of G/T SNP at intron 1 splice donor site of Wx gene of these selected cultivars (Fig. 4C).
347
Nipponbare showed a predominant band at 120 bp and another band at 210 bp, while the selected
348
Beruin cultivars only showed the 210 bp band (Fig. 4C). This is consistent with previous observations
349
that cultivars carrying intron 1 G/T SNP showed a predominant band at 120 bp with an additional
350
band at 210 bp, indicating at least two alternative variants of Wx transcript, but cultivars without the
351
mutation showed only a predominant band at 210 bp (Prathepha, 2007).
SC
RI PT
342
353
M AN U
352 4. Discussion
354
4.1. Distribution of Bangladeshi rice cultivars in context with Asian rice cultivars
356
In this study, 22 Bangladeshi glutinous and non-glutinous cultivars were placed in specific positions
357
along with 18 previously reported Wx haplotypes (Olsen and Purugganan, 2002) based on
358
phylogenetic- and haplotype network analyses of Wx upstream regulatory region sequences (Fig. 2
359
and 5). The previously reported Wx haplotypes were identified using 105 accessions from different
360
countries of Asia (Olsen and Purugganan, 2002). Bangladeshi non-glutinous cultivars Kalijira,
361
Kataribhog and Chinigura have the same Wx allele as haplotype F with no additional mutation. As
362
such, these cultivars were placed with non-glutinous progenitor haplotype F. Previously, Olsen and
363
Purugganan (2002) also showed that 39% of accessions in Southeast Asia and 15% of the South Asian
364
accessions contain this F haplotype. Four non-glutinous cultivars (Lathial White, Lathial Red,
365
Girishail, and Raujan 2) and one intermediate amylose containing cultivar Modhu Beruin Red bear
366
closer identity to haplotype A and were placed with it (Fig. 5).
AC C
EP
TE D
355
367 368
Three other Bangladeshi non-glutinous cultivars - Balam White, Balam Red and Raujan, and one
369
intermediate amylose containing cultivar Push Beruin were placed in the same node as haplotype B
14
ACCEPTED MANUSCRIPT (Fig. 5). Non-glutinous haplotypes A and B were derived from haplotype F by subsequent mutations
371
at both promoter and intron1 region. Haplotype A is more frequent in Southeast Asia and haplotype B
372
is more frequent in South Asia (Olsen and Purugganan, 2002). Therefore, Bangladeshi non-glutinous
373
and Beruin rice cultivars with intermediate AC carry the general pattern prevalent in South and
374
Southeast Asia.
RI PT
370
375
Among the Bangladeshi glutinous cultivars, Mou Beruin, Akia Beruin White and Akia Beruin Red
377
have the same sequence identities as the glutinous progenitor haplotype G in Wx upstream region
378
(Fig. 5). Most of the glutinous rice accessions in Southeast, North and South Asia have been reported
379
to carry haplotype G. Bangladeshi glutinous cultivars Kalo Beruin, Pak Beruin and Kathali Beruin
380
White contained an additional G/A SNP at position 718 of Wx promoter along with the G/T SNP at
381
the 5′ splice donor site of intron 1 (Fig. 1). These cultivars clustered with the glutinous haplotype
382
group (L, K, O and I), which also have additional mutations located at promoter and intron 1 but at
383
different positions (Fig. 5). The 718 (G/A) SNP has not yet been reported and is therefore unique to
384
Bangladeshi glutinous cultivars as well as the non-glutinous Shamudrophena.
TE D
M AN U
SC
376
385
From the median-joining network analysis (Fig. 5), low AC Bangladeshi Kathali Beruin Red was
387
found to be close to the non-glutinous haplotype S. Two other waxy AC cultivars Khara Beruin and
388
Modhu Beruin White were placed with non-glutinous haplotype C and recombinant haplotype R in
389
the Wx haplotype network (Fig. 5). This suggests that some Bangladeshi glutinous Beruin cultivars
390
may have a different allele compared to glutinous cultivars of Asia responsible for waxy or low AC
391
phenotype.
AC C
392
EP
386
393
4.2. Bangladesh genotype-specific changes
394
Several studies have reported specific SNP or indels in Wx exons that decrease AC in rice seeds by
395
reducing the binding of GBSS-I to starch granules or creating stop codon that terminates the
396
translation of Wx transcript (Liu et al., 2009; Wanchana et al., 2003). However, waxy Khara Beruin,
397
Modhu Beruin White and Kathali Beruin did not have any of those SNPs. Pairwise sequence
15
ACCEPTED MANUSCRIPT comparisons of these cultivars with Olsen and Purugganan (2002) reported haplotypes showed that
399
some mutations at both promoter and intron 1 region were shared between Khara Beruin and
400
haplotype C (99.93% sequence identity), Modhu Beruin White and haplotype R (99.19% identity) as
401
well as Kathali Beruin Red and haplotype S (99.26% identity). Some of these mutations were found to
402
be located within putative cis-regulatory motifs in the Wx promoter site in Beruin cultivars
403
(Supplementary Table 2). Most interestingly, these three cultivars have a C deletion at a
404
polypyrimidine tract of intron 5 as well as two G/A SNPs and an ATA deletion at intron 10. This C
405
deletion in intron 5 consequentially causes loss of the putative INTRONLOWER motif in that region.
406
This may cause slippage of intron splicing and subsequent translation (Brown, 1986; Sridharan and
407
Singh, 2007). Also, the G/A indel in intron 10 (at position 95 from 5' splice site) which affected a
408
putative Exonic Splicing Enhancer motif might cause exon skipping in these three cultivars
409
(Supplementary Table 2). Therefore, these mutations might be another cause of the low AC in Khara
410
Beruin, Modhu Beruin White and Kathali Beruin Red despite lack of the G/T SNP at the intron 1
411
splice donor site, which is known as the main cause of waxy or low amylose phenotype.
M AN U
SC
RI PT
398
TE D
412
Waxy amylose containing Khara Beruin and Modhu Beruin White also contained one SNP at intron 9
414
(G/A) and showed 99% similarity with opaque or chalky endosperm containing rice cultivars found in
415
Nepal, India, Myanmar and Indonesia. On the other hand, low amylose containing Kathali Beruin Red
416
contained one G/A SNP in position 118 in intron 10 that might create an alternative splice acceptor
417
site in intron 10 (Supplementary Table 2). This SNP in intron 10 has not been reported previously.
418
These SNPs at introns 9 and 10 also might have an effect on splicing as well as gene expression and
419
varying amylose between waxy Khara Beruin, Modhu Beruin and Low AC Kathali Beruin Red.
420
However, the transcript levels of the Wx gene in these three Beruin cultivars were not low compared
421
to Nipponbare, which carries the G/T splice site SNP in intron 1. Whether the above SNPs in the
422
promoter and introns alone are responsible for regulating Wx transcript levels in these cultivars or
423
other players are involved requires further investigation.
AC C
EP
413
424 425
5. Conclusion
16
ACCEPTED MANUSCRIPT Bangladeshi non-glutinous and glutinous cultivars with intermediate AC carried the general pattern of
427
nucleotide variation in Wx upstream region as observed in South and Southeast Asian cultivars. Some
428
exceptional Wx alleles were also identified in Bangladeshi glutinous cultivars Khara Beruin, Modhu
429
Beruin White and Kathali Beruin Red. These cultivars lack the G/T splice site mutation in intron 1
430
region as non-glutinous cultivars, but yet have waxy and/or low amylose phenotype. However these
431
cultivars have some SNPs in promoter region that may alter putative cis-regulatory motifs involved in
432
Wx gene regulation. Additionally, a putative INTRONLOWER motif (consensus sequence for plant
433
introns) and an Exonic Splicing Enhancer motif were lost in these cultivars due to indels in Wx introns
434
5 and 10 respectively. Khara Beruin and Modhu Beruin White (waxy AC) also contained a SNP at
435
intron 9 and Kathali Beruin Red (low AC) contained two SNPs at introns 6 and 10. It can be
436
hypothesized that the intron- and promoter-specific mutations found in these three cultivars could be
437
the cause of the waxy or low amylose phenotype. This hypothesis however needs to be investigated
438
further through functional characterization of these motifs and confirmation of alternative Wx variants
439
due to misregulation of splicing.
TE D
440
M AN U
SC
RI PT
426
Acknowledgement
442
Funding for this research, including fellowships for RB and SS and consumables and equipment, was
443
received from the Bangladesh Chapter of USDA under the 416-(B) grant aid. BAS-USDA-PALS
444
project also provided a 6-month extension fellowship to RB. We would like to thank Dr. Abdul
445
Chaudhury and Munir Hasan for providing information on the Beruin cultivars of Bangladesh and
446
encouraging us to undertake the work, Md. Sazzadur Rahman (Senior Scientific Officer, Bangladesh
447
Rice Research Institute) for helping with seed collection and Md. Shamim Hossain for taking care of
448
the plants.
AC C
EP
441
449 450
References
451
Asaoka, M., Okuno, K., Yano, M., Fuwa, H., 1993. Effects of Shrunken and Other Mutations on the
452 453
Properties of Rice Endosperm Starch. Starch/Stärke 45, 383–387. Ayres, N.M., McClung, A.M., Larkin, P.D., Bligh, H.F.J., Jones, C.A., Park, W.D., 1997.
17
ACCEPTED MANUSCRIPT 454
Microsatellites and a single-nucleotide polymorphism differentiate apparentamylose classes
455
in an extended pedigree of US rice germ plasm. Theoretical and Applied Genetics 94, 773–
456
781.
458
Bergman, C., Bhattacharya, K., Ohtsubo, K., 2004. Rice end-use quality analysis. In: Champagne, E.T. (Ed.), Rice Chemistry and Technology. AACC, MN, USA, pp. 415–472
RI PT
457
459
Bradbury, P.J., Zhang, Z., Kroon, D.E., Casstevens, T.M., Ramdoss, Y., Buckler, E.S., 2007.
460
TASSEL: software for association mapping of complex traits in diverse samples.
461
Bioinformatics 23, 2633–2635.
463
Brown, J.W.S., 1986. A catalogue of splice junction and putative branch point sequences from plant introns. Nucleic Acids Research 14, 9549–9559.
SC
462
Cai, X.L., Wang, Z.Y., Xing, Y.Y., Zhang, J.L., Hong, M.M., 1998. Aberrant splicing of intron 1
465
leads to the heterogeneous 5′ UTR and decreased expression of waxy gene in rice cultivars of
466
intermediate amylose content. The Plant Journal 14, 459–465.
468
Dipti, S., Bari, M., Kabir, K., 2003. Grain quality characteristics of some Beruin rice varieties of Bangladesh. Pakistan Journal of Nutrition 2, 242-245.
TE D
467
M AN U
464
Dobo, M., Ayres, N., Walker, G., Park, W.D., 2010. Polymorphism in the GBSS gene affects amylose
470
content in US and European rice germplasm. Journal of Cereal Science 52, 450–456.
471
Dong, B., 2000. The potential commercial value of Yunnan “Zhefang rice.” Yunnan Agric. Sci.
473 474 475
Technol. 5, 13–15.
Elias, S.M., Hasan, A.K.M.M., Seraj, Z.I., 2011. Microsatellite marker diversity and sequence
AC C
472
EP
469
polymorphism in the red gene locus of indigenous rice populations of Bangladesh. Plant Systematics and Evolution 296, 157–165.
476
Fan, C.C., Yu, X.Q., Xing, Y.Z., Xu, C.G., Luo, L.J., Zhang, Q., 2005. The main effects, epistatic
477
effects and environmental interactions of QTLs on the cooking and eating quality of rice in a
478
doubled-haploid line population. Theoretical and Applied Genetics 110, 1445–1452.
479 480 481
Huson, D.H., Bryant, D., 2006. Application of Phylogenetic Networks in Evolutionary Studies. Molecular Biology and Evolution. 23, 254–267. Inukai, T., Sako, A., Hirano, H.Y., Sano, Y., 2000. Analysis of intragenic recombination at wx in rice:
18
ACCEPTED MANUSCRIPT 482
Correlation between the molecular and genetic maps within the locus. Genome 43, 589–596.
483
Juliano, B.O., Perez, C.M., Blakeney, A.B., Castillo, T., Kongseree, N., Laignelet, B., Lapis, E.T.,
484
Murty, V.V.S., Paule, C.M., Webb, B.D., 1981. International Cooperative Testing on the
485
Amylose Content of Milled Rice. Starch/Stärke 33, 157–162.
487
Larkin, P.D., Park, W.D., 2003. Association of waxy gene single nucleotide polymorphisms with
RI PT
486
starch characteristics in rice (Oryza sativa L.). Molecular Breeding 12, 335–339.
Liu, L., Ma, X., Liu, S., Zhu, C., Jiang, L., Wang, Y., Shen, Y., Ren, Y., Dong, H., Chen, L., others,
489
2009. Identification and characterization of a novel Waxy allele from a Yunnan rice landrace.
490
Plant Molecular Biology 71, 609–626.
SC
488
Martin, C., Smith, A.M., 1995. Starch biosynthesis. Plant Cell 7, 971–985.
492
Mikami, I., Aikawa, M., Hirano, H.Y., Sano, Y., 1999. Altered tissue-specific expression at the Wx
493
M AN U
491
gene of the opaque mutants in rice. Euphytica 105, 91–97.
Mikami, I., Uwatoko, N., Ikeda, Y., Yamaguchi, J., Hirano, H., Suzuki, Y., Sano, Y., 2008. Allelic
495
diversification at the wx locus in landraces of Asian rice. Theoretical and Applied Genetics
496
116, 979–989.
499 500 501 502 503 504
relatives. Oxford Surveys in Evolutionary Biology 8, 135–135. Okagaki, R.J., Wessler, S.R., 1988. Comparison of non-mutant and mutant waxy genes in rice and
EP
498
Morishima, H., Sano, Y., Oka, H.I., 1992. Evolutionary studies in cultivated rice and its wild
maize. Genetics 120, 1137–1143. Olsen, K.M., Purugganan, M.D., 2002. Molecular evidence on the origin and evolution of glutinous
AC C
497
TE D
494
rice. Genetics 162, 941–950.
Prathepha, P., 2007. Identification of variant transcripts of waxy gene in non-glutinous rice (O. sativa L.) with different amylose content. Pakistan Journal of Biological Sciences 10, 2500–2504.
505
Sato, H., Suzuki, Y., Sakai, M., Imbe, T., 2002. Molecular Characterization of Wx-mq, a Novel
506
Mutant Gene for Low-amylose Content in Endosperm of Rice (Oryza sativa L.). Breeding
507
Science 52, 131–135.
508 509
Sridharan, V., Singh, R., 2007. A conditional role of U2AF in splicing of introns with unconventional polypyrimidine tracts. Molecular and Cellular Biology 27, 7334–7344.
19
ACCEPTED MANUSCRIPT 510 511
Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution 30, 2725–2729. Thompson, J.D., Gibson, T.J., Higgins, D.G., 2002. Multiple Sequence Alignment Using ClustalW
513
and ClustalX. In: Baxevanis, A.D., Stein, L.D., Stormo, G.D., Yates, J.R. (Eds.), Current
514
Protocols in Bioinformatics. John Wiley & Sons, New York, pp. 2.3.1–2.3.22.
515 516
RI PT
512
Wanchana, S., Toojinda, T., Tragoonrung, S., Vanavichit, A., 2003. Duplicated coding sequence in the waxy allele of tropical glutinous rice (Oryza sativa L.). Plant Science 165, 1193–1199. Yamanaka, S., Nakamura, I., Watanabe, K.N., Sato, Y.I., 2004. Identification of SNPs in the waxy
518
gene among glutinous rice cultivars and their evolutionary significance during the
519
domestication process of rice. Theoretical and Applied Genetics 108, 1200–1204.
SC
517
Zeng, D., Yan, M., Wang, Y., Liu, X., Qian, Q., Li, J., 2007. Du1, encoding a novel Prp1 protein,
521
regulates starch biosynthesis through affecting the splicing of Wxb pre-mRNAs in rice (Oryza
522
sativa L.). Plant Molecular Biology 65, 501–509.
M AN U
520
AC C
EP
TE D
523
20
ACCEPTED MANUSCRIPT Figure Legends
2
Fig. 1: Amylose content and nucleotide variation in Wx upstream regulatory region of twenty-two
3
Bangladeshi rice cultivars. Comparison of amylose contents in (A) non-glutinous and (B) Beruin
4
(locally known as glutinous) cultivars revealed three distinct categories. Nucleotide polymorphisms in
5
~2.7 kb Wx upstream noncoding region of these varieties with respect to known non-glutinous and
6
glutinous progenitor haplotypes (Olsen and Purugganan, 2002) are shown in (C).
RI PT
1
7
Fig. 2: Phylogenetic tree based on Wx upstream regulatory region showing the relationship among
9
Bangladeshi cultivars and other reported haplotypes. Neighbour-joining method along with Kimura 2-
10
parameter distance were utilized for phylogenetic analysis. Bootstrap values (>40%) for 1000
11
replicates are shown at the nodes.
M AN U
12
SC
8
13
Fig. 3: Chalky endosperm of Khara Beruin, partially chalky in Modhu Beruin White and nearly
14
transparent endosperm of non-glutinous cultivar Kataribhog.
TE D
15
Fig. 4: Expression of Wx gene in Beruin cultivars lacking G/T SNP in intron 1 splice donor site. (A)
17
Semi-quantitative RT-PCR analysis of Wx gene expression in developing seeds of selected Beruin
18
cultivars. Nipponbare, which carries G/T SNP in intron 1, was used as a control. RT- PCR with two
19
different primer sets (B and C) re-confirmed that intron 1 is efficiently spliced in the selected Beruin
20
cultivars without the G/T splice site SNP.
AC C
21
EP
16
22
Fig. 5: Median-joining network for Bangladeshi cultivars and reported haplotypes based on Wx
23
upstream regulatory region. Here magenta and blue colors indicate Bangladeshi glutinous Beruin and
24
non-glutinous cultivars, respectively. Purple color indicates cultivars that were newly assigned as non-
25
glutinous (this study) despite being locally known as Beruin or sticky rice. Squares and diamonds
26
represent previously reported non-glutinous and glutinous haplotypes across Asia (Olsen and
27
Purugganan, 2002), respectively. Circles represent nodes in the network. Node size is proportional to
28
number of cultivars/haplotypes present in each node.
1
ACCEPTED MANUSCRIPT
Table 1. Nucleotide polymorphisms in Wx coding region of selected Bangladeshi cultivars.
RI PT
Region in Position Position Polymorph Reported Shamudro Kathali Modhu Khara Mou Wx locus from from 5' -ism type Wx allele -phena Beruin Beruin Beruin Beruin start splice Name (nonRed White (glutinous codon site of glutinous reference) intron reference) Exon 2 100 23 bp wx,a No No No No Yes duplication Exon 4 476 SNP (ns) Wxhp(A/G)b A A A A A Exon 4 497 SNP (ns) Wxmq(G/A) G G G G G
SC
c
SNP (ns) Wx (A/G)a A A SNP(ns) Wxmq(T/C)c T T SNP (ns) Wxin(A/C) d C A SNP (s) T C SNP (ns) C T 94 Indel C 29 SNP A A 81 SNP G A 95 SNP G A 104- 106 indel ATA 118 SNP G A Here ‘ns’ and ‘s’ refers to nonsynonymous and synonymous amino acid 521 574 671 1109 1266
M AN U
Exon 4 Exon 5 Exon 6 Exon 9 Exon 10 Intron 5 Intron 9 Intron 10
op
A A A T T T A A A C C T C C C C G G A A A G A A G ATA G G G changes, respectively.
TE D
Sites without polymorphism in any of the cultivars are shaded in grey. a(Mikami et al., 2008,
AC C
EP
1999) ,b(Liu et al., 2009), c(Sato et al., 2002), d(Dobo et al., 2010)
ACCEPTED MANUSCRIPT
Non-glutinous cultivars 23.1
23.3 23.1 22.5 23.1 23.3
21.48
25
21.48
23.4 22.6
20
15
15 10
5
5
0
4.5 4.5 5.4
5.2 4.9 4.4
0
Low AC (6-18%)
C 1
2
Promoter
5 6 7
8
9
23 56 57 67 216 255 256 307 333 350 387 398 475 534 677 718 1250 1283
Exon 1 5ʹ UTR
10
11 12 13
14
Intron 1
Exon 2 5ʹ UTR
A A A A A G G G G G G G
C C C C C T T T T T T T
T T T T T T T T T T T T
G G G G G G G
C C C C C C C
G G G G G G G G G G G G
T T T T T C C C C C C C
T T T T T G G G G G G G
T T T T T C C C C C C C
A A A A A G G G G G G G
C C C C C T T T T T T T
G G G G G A A A A A A A
T -
G G G G A G G G G G G G
A A A A A G G G G G G G
C 16-20 G G T A 5 A C A G C A C 18 G G T A 5 A C A G C A C 18 G G T A 5 A C A G C A C 17 G G T A 6 A C A G C A C 17 G G T A 5 A C A G C A T 12 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 10 G A T A 5 G C G G T C T 10 G A - - 5 A C G G T C T 10 G A - - 5 A C G G T C
T T T T T G G G G G G G
G G G G G G G G G G G G
C C C C C T T T T T T T
T T T T T C C C C C C C
C -
A A A A A A A A A A -
A A A A A A A A A G A A
T T T T T C C C C C C C
T T
G G G G G G G G G G A G
C C C C C C C C C A A A
T T T T T T T C T C C C
A A A A A A A G A G G G
C C C C C C C T C T T T
T T T T T T T A T T T T
G G
C C
G G G G G G G G G G G C
T T T T T T T T T T C C
T T T T T T T G T T G G
T T T T T T T C T T C C
A A A A A A A A A A G G
C C C C C C C T C C T T
G G G G G G G A G G A A
T -
G G G G A A G A A G G A
A A A A A A G G A G G G
C 16-18 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A T 16 G A T A 5 G C G - T C T 8 G A T A 6 G T G - T C C 17 T G T A 5 A C A G C A T 11 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 10 G A - - 5 A C A G T C
T T T T T T G G T G G G
G G G G G G G T G G G G
C C C C C C T T C T T T
T T T T T T C C T C C C
-
A A A A A A A A A A -
A A A A A A A A A A A A
T T T T T T C C T C C C
T
G G G G G G G G G G G G
TE D
T T T T T C C C C C C C
EP
Haplotype G Akia Beruin White Akia Beruin Red Mou Beruin Pak Beruin Kathali Beruin White Modhu Beruin White Khara Beruin Kalo Beruin Kathali Beruin Red Modhu Beruin Red Push Beruin
4
C C C C C A A A A C A A
AC C
Non-glutinous
Haplotype F Kataribhog Chinigura Kalijira Shamudrophena Lathial White Lathial Red Girishail Raujan 2 Balam Red Balam White Raujan
Beruin (local glutinous)
Type Cultivar name
3
4.2
Waxy AC (0-5%)
M AN U
~2.7kb
1505,(CT)n
Wx Exons
SC
Ak ia
Ka
Intermediate AC (19-23%)
8.2
RI PT
10
13.7
2733
19.1
20.5
1600, Splice site 1685 1686 1687 1747, (AATT)n 1819 1846 1863 1902 1914 1917 1927 1934 1951 1953 2016 2171 2285 2304 2584
20
20.7
Amylose Content (%) B Ak eru ia in Be W ru hite M in R ou e Ka d th P Be al ak r u M i B B in od e r er u uin uin Be W ru h Kh in W ite ar a hit Ka K Be e a r th a lo ui M li B Be n od e r u hu ru in Be in R ru e Pu in d sh Re Be d ru in
25
Beruin (local glutinous) cultivars
B
ta ri C bho hi ni g gu Sh am K ra a ud liji La rop ra th he ia na La l W th hit ia e lR G ed iri R sha au il Ba jan la 2 Ba m la Re m d W hi t R e au ja n
Amylose Content (%)
A
ACCEPTED MANUSCRIPT
64
Bangladeshi non-glutinous Beruin cultivars Bangladeshi non-glutinous cultivars Known non-glutinous haplotypes Non-glutinous progenitor haplotype
82
Modhu Beruin White Kathali Beruin Red
52 64
AC C
EP
0.001
66
TE D
50
Haplotype S Khara Beruin Haplotype C Haplotype E Haplotype D Push Beruin Balam Red Balam White Haplotype B Raujan Haplotype A Modhu Beruin Red Raujan 2 Girishail 74 Lathial Red Lathial White
Clade 1 CT16-20
Sub clade 1.2 G at Intron 1 splice site (except haplotype R)
M AN U
93
59
Sub clade 1.1 T at Intron 1 splice site (except Shamudrophena)
RI PT
Glutinous progenitor haplotype Bangladeshi glutinous Beruin cultivars
SC
Haplotype K Haplotype L Akia Beruin Red Mou Beruin Haplotype M 44 Haplotype N 64 Haplotype Q Akia Beruin White Haplotype I Haplotype G Shamudrophena Sub clade 1.1.1 Kalo Beruin 718 G/A at 98 54 Kathali Beruin White Wx promoter 46 Pak Beruin Kalijira Kataribhog Chinigura Haplotype F 48 Haplotype P Haplotype J Haplotype H 64 Haplotype R
Known Glutinous haplotypes
Sub clade 2.1 CT8 (except haplotype E) Sub clade 2.2 CT10
Sub clade 2.3 CT 11-12
Clade 2 CT8-12 G at Intron 1 splice site
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C EP
200 bp 120 bp 100 bp
C
210 bp
1.6 kbp 1.1 kbp 1 kbp
SC
Push Beruin
Kathali Beruin Red
Modhu Beruin Red
Modhu Beruin White
Nipponbare
Marker
RI PT
Kathali Beruin Red
Push Beruin
Modhu Beruin Red
M AN U
Kathali Beruin Red
TE D
Push Beruin
Modhu Beruin Red
B Modhu Beruin White
Nipponbare
Marker
A
Modhu Beruin White
Nipponbare
Marker
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
SC
Raujan 2
Push Beruin C
Khara Beruin
M AN U
Modhu Beruin White
TE D
B , Raujan
K
R
Kathali Beruin Red
EP
Balam Red
Q
AC C
E
RI PT
A , Girishail, Lathial Red, Balam White Lathial White, Modhu Beruin Red
D
S
Pak Beruin, Kalo Beruin, Kathali Beruin White
Mou Beruin, Akia Beruin Red, Akia Beruin White
Chinigura, Kataribhog
Shamudrophena P
G F
M
Kalijira
J
I N
L
H
ACCEPTED MANUSCRIPT
•
Sequence polymorphism in Wx locus of 22 Bangladeshi rice landraces was analyzed
•
Most of the Bangladeshi landraces showed SNP patterns reported in South & Southeast Asian varieties Three Beruin cultivars lacked the known glutinous alleles but had waxy or low amylose
•
Novel Wx promoter/intronic mutations in these 3 Beruin cultivars may explain their
RI PT
•
AC C
EP
TE D
M AN U
SC
glutinous phenotype
S0733-5210(16)30128-X
DOI:
10.1016/j.jcs.2016.07.006
Reference:
YJCRS 2178
To appear in:
Journal of Cereal Science
Received Date: 19 August 2015 Revised Date:
2 July 2016
Accepted Date: 7 July 2016
Please cite this article as: Shahid, S., Begum, R., Razzaque, S., Jesmin, , Seraj, Z.I., Variability in amylose content of Bangladeshi rice cultivars due to unique SNPs in Waxy allele, Journal of Cereal Science (2016), doi: 10.1016/j.jcs.2016.07.006. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Variability in amylose content of Bangladeshi rice cultivars due to unique SNPs in Waxy allele
2
Saima Shahida,c,1, Rokeya Beguma,c, Samsad Razzaquea, Jesminb and Zeba I. Seraja*
3
a
4
Dhaka, Dhaka-1000, Bangladesh
5
b
6
Bangladesh
Plant Biotechnology Laboratory, Department of Biochemistry and Molecular Biology, University of
RI PT
Department of Genetic Engineering and Biotechnology, University of Dhaka, Dhaka-1000,
7 8
c
Contributed equally
SC
9
*Corresponding Author: Zeba I. Seraj, Plant Biotechnology Laboratory, Department of Biochemistry
11
and Molecular Biology, University of Dhaka, Dhaka-1000, Bangladesh. Tel: +880-2-8614708; fax:
12
+880-2-8615583/9127051. Email: [email protected].
M AN U
10
13 14
1
15
Life Sciences, Penn State University, University Park, PA 16802, USA
TE D EP AC C
16
Present address: Plant biology PhD program, Department of Biology, and the Huck Institutes of the
1
ACCEPTED MANUSCRIPT List of Abbreviations
18
AC = Amylose content
19
bp =Base pairs
20
cDNA = Complementary DNA
21
CTAB = Cetyltrimethylammonium bromide
22
DNA =Deoxyribonucleic acid
23
dNTP = Deoxynucleoside triphosphate
24
ESE = Exonic splicing enhancer
25
GBSS-I = Granule-bound starch synthase I
26
G3PDH = Glyceraldehyde-3-phosphate dehydrogenase
27
indel = Insertion/deletion
28
kb = Kilobase pairs
29
mM =Millimolar
30
mRNA = Messenger RNA
31
ng = Nanogram
32
pmol = Picomole
33
QTL = Quantitative trait loci
34
RNA = Ribonucleic acid
35
RT-PCR = Reverse transcription polymerase chain reaction
36
SF2/ASF = Splicing factor-2/ alternative splicing factor
37
SNP = Single nucleotide polymorphism
38
UTR =Untranslated region
39
Wx = Waxy
40
µl = Microlitre
41
µM = Micromolar
AC C
EP
TE D
M AN U
SC
RI PT
17
42 43 44 2
ACCEPTED MANUSCRIPT 45 46
Abstract
47 Waxy gene (Granule Bound Starch Synthase I) is responsible for amylose synthesis in the rice
49
endosperm. Several mutations in this gene have been shown to be responsible for variable amylose
50
content (AC) phenotypes. The G/T mutation in 5′ splice site of Waxy intron 1 has been traced as the
51
origin of the glutinous rice phenotype and differentiates low AC from non-glutinous intermediate/high
52
AC rice. Sequencing of Waxy promoter and 5′ noncoding regions from 22 rice cultivars showed that
53
the evolutionary pattern of all Bangladeshi non-glutinous and most glutinous rice accessions are in
54
line with the general patterns of South and Southeast Asia. However, three cultivars Khara Beruin,
55
Modhu Beruin White and Kathali Beruin Red with low to very low amylose lacked the G/T splice site
56
mutation. These were more closely related to non-glutinous cultivars based on their SNP patterns in
57
promoter and noncoding regions. Further sequencing revealed a unique C deletion at a pyrimidine
58
tract of intron 5 of these three cultivars that may cause slippage of intron splicing. Additional SNPs at
59
intron 9 and 10 were also identified among these cultivars. These Bangladeshi-genotype-specific
60
mutations could be the cause of waxy or low amylose phenotypes in these glutinous accessions.
SC
M AN U
TE D
Key words: Amylose content; Waxy gene; Glutinous and non-glutinous rice; SNPs
EP
62
AC C
61
RI PT
48
3
ACCEPTED MANUSCRIPT 1. Introduction
64
Starch in rice endosperm contains amylose and amylopectin polysaccharides. Amylose is principally a
65
linear molecule containing α(1→4) linked glucose units and makes up approximately 0-30% of total
66
starch. In contrast amylopectin is a branched molecule which contributes to approximately 70-100%
67
of total starch in rice endosperm (Martin and Smith, 1995). Higher amylose levels (20–30%) are
68
observed in many indica rice varieties in South Asia (Morishima et al., 1992). Lower amylose levels
69
(10–20%) are more common in japonica varieties that predominate in East Asia (Yamanaka et al.,
70
2004). Amylose content (AC) can be thus used to classify milled rice samples into different categories
71
such as glutinous/waxy (0–5% amylose), low AC (6–18%), intermediate AC (19–23%), or high AC
72
(>23%) types (Bergman et al., 2004). Low AC is usually associated with tender, cohesive, and glossy
73
cooked rice while high AC is associated with firm, fluffy, and separated grains in cooked rice (Juliano
74
et al., 1981). Therefore, AC is considered as one of the major characteristics for assessing rice
75
cooking and eating qualities. In addition, low AC rice varieties have long been used as a tool of
76
improving grain quality through conventional breeding as these represent an intermediate type
77
between glutinous and non-glutinous rice varieties (Dong, 2000; Sato et al., 2002).
SC
M AN U
TE D
78
RI PT
63
Genetic and molecular marker-based QTL analyses have revealed that the wide range of AC variation
80
in rice endosperm is mainly controlled by a major locus (Wx or Waxy gene) and multiple minor loci
81
(Fan et al., 2005; Inukai et al., 2000). The Wx gene in rice encodes the granule-bound starch synthase
82
I (GBSS-I) enzyme responsible for amylose biosynthesis (Okagaki and Wessler, 1988). The origin of
83
the glutinous phenotype such as those found in japonica rice has been traced to the Wxb allele, which
84
contains a G/T mutation in intron 1 splice donor site of the Wx gene (Olsen and Purugganan, 2002).
85
Presence of this splice donor site mutation leads to inefficient splicing of Wx pre-mRNA followed by
86
reduced expression of functional GBSS-I (Cai et al., 1998). In contrast, non-glutinous phenotype
87
observed in many indica rice varieties has been traditionally associated with Wxa allele, which lacks
88
the intron 1 G/T mutation. However many Northeast Asian rice varieties carrying this intron 1 G/T
89
mutation still show non-glutinous phenotype. This suggests that partial suppression of the intron 1
90
G/T mutation may have an important role in the development of non-glutinous rice of this region
AC C
EP
79
4
ACCEPTED MANUSCRIPT (Olsen and Purugganan, 2002). Amylose content of rice endosperm is therefore a major domestication
92
trait that has evolutionary importance. Among the minor alleles affecting amylose content, a
93
polymorphic CTn microsatellite in the 5′-untranslated region (UTR) of Wx exon 1 has been shown to
94
be correlated with various AC classes (Ayres et al., 1997). Other SNPs in coding regions of different
95
Wx alleles lead to decreased AC by dropping the binding efficiency of GBSS-I to starch granules,
96
changing amino acid at respective sites (Liu et al., 2009; Sato et al., 2002) and altering expression
97
pattern of GBSS-I in endosperm and anthers (Mikami et al., 1999). In addition to Wx, mutations in
98
genes such as Dull (Zeng et al., 2007) and shrunken (Asaoka et al., 1993) also indirectly affect AC in
99
rice.
SC
RI PT
91
100
Bangladesh is the world’s fourth largest rice producing country. It is considered as an enriched rice
102
germplasm reservoir with over 6500 wild, landraces and modern high yielding varieties (Elias et al.,
103
2011). Majority of the rice species in Bangladesh are indica varieties with high amylose content and
104
non-glutinous phenotype (Olsen and Purugganan, 2002). Several glutinous rice varieties, which are
105
locally known as ‘Beruin’ or sticky rice, are also cultivated in the Northeastern region of Bangladesh
106
(Dipti et al., 2003). These Beruin cultivars generally have low amylose content and are popular for
107
making traditional dishes on special occasions. However the molecular basis of low amylose content
108
in these local Beruin varieties is unknown. In this study we examined the nucleotide diversity at the
109
Wx locus of selected Bangladeshi glutinous Beruin and non-glutinous rice cultivars to identify the
110
underlying cause of AC variation in these genotypes. We also compared these Bangladeshi varieties
111
with known Wx haplotypes in an attempt to understand their evolutionary pattern across Asia.
113
TE D
EP
AC C
112
M AN U
101
2. Materials and Methods
114 115
2.1. Plant materials
116
Seeds of eleven Beruin cultivars, which are representative of the diverse traditional landraces unique
117
to the Northeastern region of Bangladesh (Sylhet), were collected from the local farmers. An equal
118
number of non-glutinous cultivars was then selected to allow identification of SNPs underlying the
5
ACCEPTED MANUSCRIPT glutinous phenotype in Bangladeshi cultivars (Fig. 1A and 1B). Nine of the selected non-glutinous
120
cultivars were from the same region as the Beruin cultivars. Two other popular aromatic cultivars,
121
Kalijira and Kataribhog, were collected from the Northwestern region of Bangladesh (Dinajpur) to
122
expand the genetic diversity of sampled non-glutinous cultivars. All of the selected cultivars are
123
traditional transplanted Aman or monsoon season varieties. The collected seeds were multiplied
124
during the planting season in the net house. Three plants of each cultivar were grown in a single pot to
125
collect immature seeds.
RI PT
119
126 2.2. Estimation of amylose content
128
The amylose content (AC) of all cultivars was determined according to previously published protocol
129
(Juliano et al., 1981). Nazirshail rice flour (AC 25%, Bangladesh Rice Research Institute) was used as
130
a control and potato starch flour (E. Merck, Germany) was used for generating the standard curve.
M AN U
131
SC
127
2.3. Genomic DNA Extraction
133
Genomic DNA was extracted from leaves of 3-week-old pooled plants (10-12 plants per cultivar)
134
using CTAB extraction method and in some cases by using the Qiagen DNeasy® plant mini kit. DNA
135
of all cultivars was quantified by spectrophotometry. Quality of extracted DNA was checked by 1%
136
agarose gel electrophoresis and compared with known concentration of Lambda DNA (Invitrogen,
137
USA).
EP AC C
138
TE D
132
139
2.4. PCR amplification and DNA sequencing
140
A 2.7 kb upstream regulatory region of the Wx gene including promoter, exon 1, intron 1 and part of
141
exon 2 was PCR amplified and sequenced in all 22 Bangladeshi rice cultivars. PCR amplification was
142
performed in two steps to amplify overlapping ~1.5 kb segments of the above upstream regulatory
143
region using primers described by Olsen and Purugganan (2002) (Supplementary Table 1). For each
144
primer set, 15 µl PCR reaction mixture containing 60 ng of DNA template, 1 × PCR buffer, 1.67 mM
145
MgCl2, 100 µM of each dNTP, 0.33 pmol Forward /Reverse primer, 1 unit Taq DNA polymerase and
146
0.1 unit Pfx® DNA polymerase (Invitrogen, USA) was used. Amplification conditions were 94 °C for
6
ACCEPTED MANUSCRIPT 147
5 min, followed by 35 cycles of 95 °C for 1 min, 65.2 °C for 1 min, 72 °C for 1 min and a final
148
extension of 72 °C for 8.30 min.
149 For further characterization of the Wx coding region in Shamudrophena, Kathali Beruin Red, Modhu
151
Beruin White, Khara Beruin and Mou Beruin, two different segments were amplified and sequenced
152
using four sets of primers (Supplementary Table 1). The first segment (~0.2 kb) included part of exon
153
2, which is known to contain a 23 bp duplication in some glutinous varieties. This segment was
154
amplified using primer pair sequences Glu-23F and Glu-23R as described by Wanchana et al. (2003).
155
The other ~2 kb segment including exons 3-10 was amplified using three overlapping primer pairs
156
(Supplementary
157
(http://frodo.wi.mit.edu/primer3/) based on sequence of Nipponbare Wx locus (OS06G0133000) in
158
Gramene database (www.gramene.org). PCR amplification conditions were the same as for the Wx
159
upstream regulatory region except the annealing temperature, which was optimized for each primer
160
pair (Supplementary Table 1).
1).
These
were
designed
using
primer3
program
TE D
161
primers
M AN U
Table
SC
RI PT
150
All PCR amplicons were purified using QiAquick gel extraction kit (Qiagen, Germany) and directly
163
sequenced in ABI 3730XL genetic analyzer using both forward-reverse and internal primers
164
(Supplementary Table 1). Each fragment was sequenced twice for SNP confirmation. All of the
165
sequences from this study have been deposited in GenBank (accession no. JF834042.1-JF834063.1
166
and KP675770-KP675774).
AC C
167
EP
162
168
2.5. Sequence analysis
169
Wx upstream regulatory region sequences from 22 Bangladeshi rice landraces were compared to 18
170
previously reported haplotypes across Asia (Genbank accessions AY136760–AY136784). All
171
multiple sequence alignments were generated using ClustalW (Thompson et al., 2002). SNP/indels
172
were identified using TASSEL (Bradbury et al., 2007). Neighbour-joining tree of aligned Wx
173
upstream regulatory region (2.7 kb) sequences was constructed using MEGA 6.06 software package
174
(Tamura et al., 2013) with Kimura two-parameter model and complete deletion of alignment gaps.
7
ACCEPTED MANUSCRIPT Bootstrap test with 1000 replicates was performed to ensure confidence in phylogeny. A median-
176
joining haplotype network was also generated based on the aligned sequences using SplitsTree 4.13.1
177
(Huson and Bryant, 2006). Motif change patterns due to SNPs in Wx promoter region and intron 1 of
178
Bangladeshi cultivars were predicted using the PLACE database (www.dna.affrc.go.jp/PLACE).
179
Amino acid changes due to SNP in sequenced Wx coding exons from selected Bangladeshi landraces
180
were predicted using the ExPASy translation tool (http://web.expasy.org/translate/). SplicePredictor
181
(http://bioservices.usd.edu/splicepredictor/) was used to predict splice-site modifying SNPs in intronic
182
regions of sequenced varieties. RegRNA (http://regrna.mbc.nctu.edu.tw/) was used to predict cis-
183
regulatory elements involved in mRNA splicing or transcriptional regulation.
184
SC
RI PT
175
2.6. Semi-quantitative RT-PCR and gene expression
186
Total RNAs from rice endosperm at 16-18 days after flowering were extracted using TRIZOL. First
187
strand cDNA were synthesized from total RNA following manufacturer’s protocol (Invitrogen, USA).
188
G3PDH (glyceraldehyde-3-phosphate dehydrogenase) gene-specific primers (Supplementary Table 1)
189
were used to optimize the cDNA concentration for estimating the Wx mRNA level. Nanodrop
190
spectrophotometer was used to measure and optimize the working concentration of the cDNA. Wx
191
mRNA level was assessed using cycle-dependent semi-quantitative RT-PCR with the WxRT1 primer
192
pair (Supplementary Table 1) for 30, 32 and 35 cycles. Expression levels of samples were estimated
193
by visual inspection after electrophoresis in a 0.9% agarose gel based on the band intensity of the
194
loading control G3PDH. Two sets of primer pairs - Wx-RT2 and Wx-RT3 (Supplementary Table 1)
195
were used to confirm the G/T SNP at the 5´-splice donor site of Wx mRNA of Bangladeshi cultivars.
196
Wx-RT2 primer pair was used to determine the splicing efficiency of the first intron in the selected
197
landraces. Nipponbare was used as a control as it contains G/T SNP at intron 1 splice donor site of Wx
198
gene (Liu et al., 2009).
AC C
EP
TE D
M AN U
185
199 200
3. Results
201 202
3.1. Amylose content of Bangladeshi cultivars
8
ACCEPTED MANUSCRIPT In order to identify the molecular basis of glutinous rice in Bangladesh, we first evaluated the amylose
204
content (AC) of 22 Beruin and non-Beruin traditional landraces that are locally considered as
205
glutinous and non-glutinous cultivars respectively. These selected cultivars were sampled from
206
intensive rice cultivation zones in Northeastern and Northwestern regions of Bangladesh that are
207
known to harbour many diverse landraces (Elias et al., 2011). Based on their AC, the selected
208
cultivars were assigned into three major categories - intermediate, low and glutinous/waxy. All non-
209
Beruin Bangladeshi cultivars in this study had intermediate AC ranging from 19.1% to 23.3%
210
(Fig.1A), confirming that these are indeed non-glutinous. In contrast, only seven out of eleven Beruin
211
varieties, traditionally associated as glutinous or sticky rice, had amylose content within the waxy
212
range (0-5.4%), as shown in Fig. 1B. Among the remaining four Beruin varieties, Kathali Beruin Red
213
and Kalo Beruin belonged to the low AC group (AC 13.7% and 8.2% respectively). The two other
214
cultivars Modhu Beruin Red and Push Beruin (AC 23.4% and 22.6% respectively) had intermediate
215
AC, similar to the Bangladeshi non-glutinous cultivars.
M AN U
SC
RI PT
203
216
3.2. Nucleotide variation at Wx upstream regulatory region of Bangladeshi cultivars
218
We compared the ~2.7 kb Wx upstream noncoding region sequences from selected Bangladeshi
219
cultivars to Olsen and Purugganan (2002) reported progenitor haplotypes F (non-glutinous) and G
220
(glutinous). These progenitor haplotypes differ only in the intron 1 G/T splice donor site mutation
221
(Olsen and Purugganan, 2002). Overall, Bangladeshi cultivars had 29 single nucleotide
222
polymorphisms (SNP) and 11 insertions/deletions (indel) in the sequenced upstream regulatory region
223
(Fig. 1C). Out of these, 15 SNP and 3 indels were found in the promoter regions and 13 SNP and 7
224
indels in intron 1. These SNP and indels in promoter and intron 1 were more frequent in non-
225
glutinous cultivars than the Beruin cultivars. In total, seven different microsatellite alleles: CT8, CT10,
226
CT11, CT12, CT16, CT17 and CT18, were identified in the Wx exon 1 (encoding 5′ untranslated region) of
227
Bangladeshi cultivars (Figure 1C). An additional SNP within Wx exon 2 that also codes for 5′ UTR
228
was observed only in one of the non-glutinous Bangladeshi cultivars.
229
All of the Bangladeshi non-glutinous cultivars in this study had G-SNP at the 5′ splice donor site of
AC C
EP
TE D
217
9
ACCEPTED MANUSCRIPT Wx intron 1 (Fig. 1C) similar to the known non-glutinous haplotype F. The CT10 or CT11 alleles within
231
Wx exon 1 were observed in six of these non-glutinous Bangladeshi cultivars. The CT17 allele was
232
present in cultivars with variable AC, ranging from intermediate AC in Bangladeshi non-glutinous
233
cultivars to low and waxy AC content in the Beruin cultivars (Fig. 1). The CT12/ CT18 and CT8/ CT16
234
alleles were specific to non-glutinous and glutinous Bangladeshi cultivars, respectively.
RI PT
230
235
G/T SNP at the 5' splice site of Wx intron 1 could explain the waxy or low AC phenotype (AC 4.4-
237
8.2%) of six out of eleven Bangladeshi Beruin cultivars. These cultivars also carried CT17 allele in
238
exon 1 similar to the known glutinous progenitor haplotype G (Fig. 1C). Among these six Beruin
239
cultivars, Pak Beruin, Kathali Beruin white and Kalo Beruin had an additional novel G/A SNP at
240
position 718 in the Wx promoter region (Fig. 1C). The remaining five Beruin cultivars lacked the G/T
241
SNP in intron 1 splice donor site. Among these, Push Beruin and Modhu Beruin Red also showed
242
intermediate amylose phenotype (AC 22.6% and AC 23.4% respectively) like non-glutinous
243
Bangladeshi cultivars in this study (Fig. 1A). These varieties also had other nucleotide variations in
244
Wx regulatory region that were similar to the Bangladeshi non-glutinous cultivars (Fig. 1C). This
245
confirmed that Push Beruin and Modhu Beruin Red might be locally misnamed as glutinous varieties
246
even though they have non-glutinous phenotype.
M AN U
TE D
EP
247
SC
236
The other three Beruin cultivars that lacked the G/T SNP in splice donor site (Khara Beruin, Modhu
249
Beruin White and Kathali Beruin Red) had waxy or low AC phenotype (Fig. 1B). Among these only
250
Khara Beruin showed several unique mutations in the Wx upstream noncoding region (T/A SNP at
251
216 position in promoter, G/T SNP at position 1934 in intron 1). This indicated that polymorphisms
252
other than G/T SNP in intron 1 splice donor site and CTn allele in exon 1 may play a major role in
253
regulating Wx gene expression in these three genotypes.
AC C
248
254 255
3.3. Evolutionary relatedness of Bangladeshi cultivars
256
An unrooted phylogenetic tree based on the aligned ~2.7 kb Wx upstream regulatory region sequences
257
of selected Bangladeshi cultivars and 18 other reported haplotypes (Olsen and Purugganan, 2002) was
10
ACCEPTED MANUSCRIPT constructed to explore the evolutionary relationships among these accessions. These 18 reported
259
haplotypes represent the landrace diversity across Asia. In the phylogenetic tree, six Bangladeshi
260
glutinous cultivars (Akia Beruin White, Akia Beruin Red, Mou Beruin, Kalo Beruin, Kathali Beruin
261
White and Pak Beruin) clustered with reported glutinous haplotypes G, I, K, L and Q at high bootstrap
262
(98%) value (Fig. 2). Pak Beruin, Kalo Beruin, Kathali Beruin White along with a non-glutinous
263
cultivar Shamudrophena also formed a sub clade depending on the G/A SNP in position 718 of the Wx
264
promoter region (Fig. 2). These cultivars may carry the same Wx allele or with an additional mutation
265
(718 G/A).
RI PT
258
SC
266
On the other hand, five glutinous Beruin cultivars that lacked the intron 1 G/T SNP were grouped
268
with reported non-glutinous haplotypes and Bangladeshi non-glutinous cultivars. Intermediate AC
269
cultivars Push Beruin and Modhu Beruin Red clustered with reported non-glutinous haplotypes B, A
270
and Bangladeshi non-glutinous cultivars (Fig. 2). Waxy cultivar Khara Beruin clustered with reported
271
non-glutinous haplotypes E and D at 64% bootstrap and with non-glutinous haplotype C at 93%
272
bootstrap (Fig. 2). Based on the aligned Wx upstream regulatory region sequences, waxy AC Modhu
273
Beruin White and low AC Kathali Beruin Red appeared to be very close to intragenic recombinant
274
haplotype R and non-glutinous haplotype S, respectively, even though neither shared similar amylose
275
phenotype with the corresponding haplotypes (Fig. 2). Bangladeshi non-glutinous cultivars were
276
clustered with reported non-glutinous haplotypes but were more close to the haplotype F, A and B
277
(Fig. 2). In addition, Chinigura, Kalijira and Kataribhog distinctly clustered with non-glutinous
278
progenitor haplotype F at 48% bootstrap (Fig. 2). It might be possible that the Bangladeshi non-
279
glutinous cultivars originally carried the non-glutinous progenitor Waxy allele and then evolved into
280
haplotypes A and B through subsequent mutations at promoter region and intron 1.
AC C
EP
TE D
M AN U
267
281 282
3.4. Nucleotide variations at coding region of Wx gene in selected Beruin cultivars
283
Since G/T SNP at intron 1 splice donor site was not enough to explain the waxy AC of Modhu Beruin
284
White and Khara Beruin and low AC of Kathali Beruin Red, we further sequenced two segments in
285
the coding region of Wx gene. Mou Beruin (AC 4.4%) and Shamudrophena (AC 23.4%) were
11
ACCEPTED MANUSCRIPT considered as references for Bangladeshi glutinous and non-glutinous cultivars respectively for this
287
analysis. The first sequenced segment in Wx gene included part of coding region of exon 2 (~196 bp)
288
which has been reported to contain a 23 bp frame-shift duplication in some glutinous varieties that
289
lead to non-functional Wx proteins (Mikami et al., 2008; Wanchana et al., 2003). Among the selected
290
cultivars, this 23 bp duplication was only identified in Mou Beruin, consistent with its glutinous
291
phenotype (Table 1). The second segment analyzed in Wx gene spanned ~1.9 kb genomic region
292
including 3' end of exons 2-10 and introns 2-10. Only the non-glutinous reference Shamudrophena
293
among the five cultivars carried A/C SNP in exon 6 (position 671 from start codon), which resulted in
294
nonsynonymous amino acid change from tyrosine to serine (Table 1). This SNP has been shown to be
295
associated with intermediate AC (21-24%) previously (Dobo et al., 2010). Two other coding region
296
SNPs (T/C in exon 9, C/T in exon 10) were identified in the three selected Beruin cultivars but not in
297
the non-glutinous or glutinous reference (Table 1). Exon 9 SNP (T/C, position 1109 from start codon)
298
was found in all three selected cultivars (Table 1). The exon 9 SNP results in a synonymous change in
299
amino acid and has been previously reported as a silent mutation (Larkin and Park, 2003). Exon 10
300
SNP (C/T) was found only in low AC Kathali Beruin Red (Table 1). Exon 10 SNP (C/T) caused a
301
nonsynonymous change from a nonpolar amino acid proline (CCT) to polar amino acid Serine (TCT),
302
and it has been reported that this SNP is linked to high amylose content (Larkin and Park, 2003).
303
Therefore these exon-specific mutations were not directly responsible for the variant AC phenotype
304
traits observed for the selected Beruin cultivars.
SC
M AN U
TE D
EP
AC C
305
RI PT
286
306
Most interestingly, Kathali Beruin Red, Modhu Beruin White and Mou Beruin had overall four SNPs
307
and two indels in Wx introns which were absent in the Bangladeshi non-glutinous and glutinous
308
references (Table 1). Among these variations, C deletion in intron 5, G/A (at positions 81 and 95 from
309
5' splice site of intron 10) and ATA deletion at position 104-106 in intron 10 were identified in all
310
these three cultivars. On the other hand, G/A in intron 10 (position 118) was found only in low AC
311
Kathali Beruin Red and A/G at position 29 from 5' splice site of intron 9 was only observed in the
312
waxy AC Khara Beruin and Modhu Beruin White. A search for putative cis-elements that might be
313
affected by these SNPs/indel revealed that the C indel was located within a poly-C tract in intron 5,
12
ACCEPTED MANUSCRIPT 314
and this indel resulted in absence of a putative INTRONLOWER motif in these three cultivars
315
(Supplementary Table 2). Also, the G/A indel in intron 10 (at position 95 from 5' splice site) resulted
316
in a missing putative ESE-SF2/ASF motif (Exonic Splicing Enhancer) in these three cultivars
317
(Supplementary Table 2).
RI PT
318 Further homology search using NCBI BLASTn revealed that the sequenced coding regions of Khara
320
Beruin and Modhu Beruin White shared ~99% identity with previously reported Wxop (opaque) allele
321
containing cultivars ARC10818, ACC35618 and ARC6622 (Genbank accessions AB281448.1,
322
AB281447.1, and AB281453.1) from Nepal and India (Mikami et al., 2008, 1999). These Wxop
323
accessions did not contain the G/T SNP at the 5´ splice donor site of intron 1 and also had
324
nonsynonymous A/G SNP in exon 4 at position 521 from start codon (Mikami et al., 2008, 1999).
325
However the latter exon 4 SNP was not present in the Khara Beruin and Modhu Berun White (Table
326
1). Comparison of the rice grains revealed that only Khara Beruin had a completely opaque or chalky
327
endosperm, while Modhu Beruin White had partially chalky endosperm (Fig. 3), which supports the
328
low to very low amylose content of these varieties.
M AN U
TE D
329
SC
319
3.5. Wx gene expression in selected Beruin cultivars without G/T SNP in splice donor site
331
Semi-quantitative RT-PCR was used to estimate total expression levels of Wx gene in Modhu Beruin
332
White (waxy AC), Kathali Beruin Red (low AC), Push Beruin and Modhu Beruin Red (intermediate
333
AC). Nipponbare was used as a control since it is known to show decreased splicing efficiency of Wx
334
transcript due to G/T SNP at intron 1 splice donor site (Liu et al., 2009). The results showed that Wx
335
transcript levels in Push Beruin, Modhu Beruin Red and Kathali Beruin Red were relatively much
336
higher compared to the control Nipponbare, (Fig. 4A). This suggests that the splicing efficiency of Wx
337
transcript is not affected in Push Beruin, Modhu Beruin Red as well as Kathali Beruin Red. This
338
observation is consistent with the absence of G/T SNP at intron 1 and the intermediate-to-low AC in
339
these cultivars. The expression level of Wx transcript in Modhu Beruin White (intron 1 splice site
340
mutation absent) was also higher than that of Nipponbare but lower than the aforementioned three
341
Beruin cultivars (Fig. 4A). The transcript level however correlates with the waxy AC of Modhu
AC C
EP
330
13
ACCEPTED MANUSCRIPT Beruin White.
343
RT-PCR with primers specific for unspliced variant of Wx transcript (Liu et al., 2009) confirmed the
344
absence of intron 1 containing 1.1 kb fragment in all of the selected Beruin cultivars (Fig. 4B). A
345
second set of RT-PCR primers previously described by Prathepha (2007) was used to re-confirm the
346
absence of G/T SNP at intron 1 splice donor site of Wx gene of these selected cultivars (Fig. 4C).
347
Nipponbare showed a predominant band at 120 bp and another band at 210 bp, while the selected
348
Beruin cultivars only showed the 210 bp band (Fig. 4C). This is consistent with previous observations
349
that cultivars carrying intron 1 G/T SNP showed a predominant band at 120 bp with an additional
350
band at 210 bp, indicating at least two alternative variants of Wx transcript, but cultivars without the
351
mutation showed only a predominant band at 210 bp (Prathepha, 2007).
SC
RI PT
342
353
M AN U
352 4. Discussion
354
4.1. Distribution of Bangladeshi rice cultivars in context with Asian rice cultivars
356
In this study, 22 Bangladeshi glutinous and non-glutinous cultivars were placed in specific positions
357
along with 18 previously reported Wx haplotypes (Olsen and Purugganan, 2002) based on
358
phylogenetic- and haplotype network analyses of Wx upstream regulatory region sequences (Fig. 2
359
and 5). The previously reported Wx haplotypes were identified using 105 accessions from different
360
countries of Asia (Olsen and Purugganan, 2002). Bangladeshi non-glutinous cultivars Kalijira,
361
Kataribhog and Chinigura have the same Wx allele as haplotype F with no additional mutation. As
362
such, these cultivars were placed with non-glutinous progenitor haplotype F. Previously, Olsen and
363
Purugganan (2002) also showed that 39% of accessions in Southeast Asia and 15% of the South Asian
364
accessions contain this F haplotype. Four non-glutinous cultivars (Lathial White, Lathial Red,
365
Girishail, and Raujan 2) and one intermediate amylose containing cultivar Modhu Beruin Red bear
366
closer identity to haplotype A and were placed with it (Fig. 5).
AC C
EP
TE D
355
367 368
Three other Bangladeshi non-glutinous cultivars - Balam White, Balam Red and Raujan, and one
369
intermediate amylose containing cultivar Push Beruin were placed in the same node as haplotype B
14
ACCEPTED MANUSCRIPT (Fig. 5). Non-glutinous haplotypes A and B were derived from haplotype F by subsequent mutations
371
at both promoter and intron1 region. Haplotype A is more frequent in Southeast Asia and haplotype B
372
is more frequent in South Asia (Olsen and Purugganan, 2002). Therefore, Bangladeshi non-glutinous
373
and Beruin rice cultivars with intermediate AC carry the general pattern prevalent in South and
374
Southeast Asia.
RI PT
370
375
Among the Bangladeshi glutinous cultivars, Mou Beruin, Akia Beruin White and Akia Beruin Red
377
have the same sequence identities as the glutinous progenitor haplotype G in Wx upstream region
378
(Fig. 5). Most of the glutinous rice accessions in Southeast, North and South Asia have been reported
379
to carry haplotype G. Bangladeshi glutinous cultivars Kalo Beruin, Pak Beruin and Kathali Beruin
380
White contained an additional G/A SNP at position 718 of Wx promoter along with the G/T SNP at
381
the 5′ splice donor site of intron 1 (Fig. 1). These cultivars clustered with the glutinous haplotype
382
group (L, K, O and I), which also have additional mutations located at promoter and intron 1 but at
383
different positions (Fig. 5). The 718 (G/A) SNP has not yet been reported and is therefore unique to
384
Bangladeshi glutinous cultivars as well as the non-glutinous Shamudrophena.
TE D
M AN U
SC
376
385
From the median-joining network analysis (Fig. 5), low AC Bangladeshi Kathali Beruin Red was
387
found to be close to the non-glutinous haplotype S. Two other waxy AC cultivars Khara Beruin and
388
Modhu Beruin White were placed with non-glutinous haplotype C and recombinant haplotype R in
389
the Wx haplotype network (Fig. 5). This suggests that some Bangladeshi glutinous Beruin cultivars
390
may have a different allele compared to glutinous cultivars of Asia responsible for waxy or low AC
391
phenotype.
AC C
392
EP
386
393
4.2. Bangladesh genotype-specific changes
394
Several studies have reported specific SNP or indels in Wx exons that decrease AC in rice seeds by
395
reducing the binding of GBSS-I to starch granules or creating stop codon that terminates the
396
translation of Wx transcript (Liu et al., 2009; Wanchana et al., 2003). However, waxy Khara Beruin,
397
Modhu Beruin White and Kathali Beruin did not have any of those SNPs. Pairwise sequence
15
ACCEPTED MANUSCRIPT comparisons of these cultivars with Olsen and Purugganan (2002) reported haplotypes showed that
399
some mutations at both promoter and intron 1 region were shared between Khara Beruin and
400
haplotype C (99.93% sequence identity), Modhu Beruin White and haplotype R (99.19% identity) as
401
well as Kathali Beruin Red and haplotype S (99.26% identity). Some of these mutations were found to
402
be located within putative cis-regulatory motifs in the Wx promoter site in Beruin cultivars
403
(Supplementary Table 2). Most interestingly, these three cultivars have a C deletion at a
404
polypyrimidine tract of intron 5 as well as two G/A SNPs and an ATA deletion at intron 10. This C
405
deletion in intron 5 consequentially causes loss of the putative INTRONLOWER motif in that region.
406
This may cause slippage of intron splicing and subsequent translation (Brown, 1986; Sridharan and
407
Singh, 2007). Also, the G/A indel in intron 10 (at position 95 from 5' splice site) which affected a
408
putative Exonic Splicing Enhancer motif might cause exon skipping in these three cultivars
409
(Supplementary Table 2). Therefore, these mutations might be another cause of the low AC in Khara
410
Beruin, Modhu Beruin White and Kathali Beruin Red despite lack of the G/T SNP at the intron 1
411
splice donor site, which is known as the main cause of waxy or low amylose phenotype.
M AN U
SC
RI PT
398
TE D
412
Waxy amylose containing Khara Beruin and Modhu Beruin White also contained one SNP at intron 9
414
(G/A) and showed 99% similarity with opaque or chalky endosperm containing rice cultivars found in
415
Nepal, India, Myanmar and Indonesia. On the other hand, low amylose containing Kathali Beruin Red
416
contained one G/A SNP in position 118 in intron 10 that might create an alternative splice acceptor
417
site in intron 10 (Supplementary Table 2). This SNP in intron 10 has not been reported previously.
418
These SNPs at introns 9 and 10 also might have an effect on splicing as well as gene expression and
419
varying amylose between waxy Khara Beruin, Modhu Beruin and Low AC Kathali Beruin Red.
420
However, the transcript levels of the Wx gene in these three Beruin cultivars were not low compared
421
to Nipponbare, which carries the G/T splice site SNP in intron 1. Whether the above SNPs in the
422
promoter and introns alone are responsible for regulating Wx transcript levels in these cultivars or
423
other players are involved requires further investigation.
AC C
EP
413
424 425
5. Conclusion
16
ACCEPTED MANUSCRIPT Bangladeshi non-glutinous and glutinous cultivars with intermediate AC carried the general pattern of
427
nucleotide variation in Wx upstream region as observed in South and Southeast Asian cultivars. Some
428
exceptional Wx alleles were also identified in Bangladeshi glutinous cultivars Khara Beruin, Modhu
429
Beruin White and Kathali Beruin Red. These cultivars lack the G/T splice site mutation in intron 1
430
region as non-glutinous cultivars, but yet have waxy and/or low amylose phenotype. However these
431
cultivars have some SNPs in promoter region that may alter putative cis-regulatory motifs involved in
432
Wx gene regulation. Additionally, a putative INTRONLOWER motif (consensus sequence for plant
433
introns) and an Exonic Splicing Enhancer motif were lost in these cultivars due to indels in Wx introns
434
5 and 10 respectively. Khara Beruin and Modhu Beruin White (waxy AC) also contained a SNP at
435
intron 9 and Kathali Beruin Red (low AC) contained two SNPs at introns 6 and 10. It can be
436
hypothesized that the intron- and promoter-specific mutations found in these three cultivars could be
437
the cause of the waxy or low amylose phenotype. This hypothesis however needs to be investigated
438
further through functional characterization of these motifs and confirmation of alternative Wx variants
439
due to misregulation of splicing.
TE D
440
M AN U
SC
RI PT
426
Acknowledgement
442
Funding for this research, including fellowships for RB and SS and consumables and equipment, was
443
received from the Bangladesh Chapter of USDA under the 416-(B) grant aid. BAS-USDA-PALS
444
project also provided a 6-month extension fellowship to RB. We would like to thank Dr. Abdul
445
Chaudhury and Munir Hasan for providing information on the Beruin cultivars of Bangladesh and
446
encouraging us to undertake the work, Md. Sazzadur Rahman (Senior Scientific Officer, Bangladesh
447
Rice Research Institute) for helping with seed collection and Md. Shamim Hossain for taking care of
448
the plants.
AC C
EP
441
449 450
References
451
Asaoka, M., Okuno, K., Yano, M., Fuwa, H., 1993. Effects of Shrunken and Other Mutations on the
452 453
Properties of Rice Endosperm Starch. Starch/Stärke 45, 383–387. Ayres, N.M., McClung, A.M., Larkin, P.D., Bligh, H.F.J., Jones, C.A., Park, W.D., 1997.
17
ACCEPTED MANUSCRIPT 454
Microsatellites and a single-nucleotide polymorphism differentiate apparentamylose classes
455
in an extended pedigree of US rice germ plasm. Theoretical and Applied Genetics 94, 773–
456
781.
458
Bergman, C., Bhattacharya, K., Ohtsubo, K., 2004. Rice end-use quality analysis. In: Champagne, E.T. (Ed.), Rice Chemistry and Technology. AACC, MN, USA, pp. 415–472
RI PT
457
459
Bradbury, P.J., Zhang, Z., Kroon, D.E., Casstevens, T.M., Ramdoss, Y., Buckler, E.S., 2007.
460
TASSEL: software for association mapping of complex traits in diverse samples.
461
Bioinformatics 23, 2633–2635.
463
Brown, J.W.S., 1986. A catalogue of splice junction and putative branch point sequences from plant introns. Nucleic Acids Research 14, 9549–9559.
SC
462
Cai, X.L., Wang, Z.Y., Xing, Y.Y., Zhang, J.L., Hong, M.M., 1998. Aberrant splicing of intron 1
465
leads to the heterogeneous 5′ UTR and decreased expression of waxy gene in rice cultivars of
466
intermediate amylose content. The Plant Journal 14, 459–465.
468
Dipti, S., Bari, M., Kabir, K., 2003. Grain quality characteristics of some Beruin rice varieties of Bangladesh. Pakistan Journal of Nutrition 2, 242-245.
TE D
467
M AN U
464
Dobo, M., Ayres, N., Walker, G., Park, W.D., 2010. Polymorphism in the GBSS gene affects amylose
470
content in US and European rice germplasm. Journal of Cereal Science 52, 450–456.
471
Dong, B., 2000. The potential commercial value of Yunnan “Zhefang rice.” Yunnan Agric. Sci.
473 474 475
Technol. 5, 13–15.
Elias, S.M., Hasan, A.K.M.M., Seraj, Z.I., 2011. Microsatellite marker diversity and sequence
AC C
472
EP
469
polymorphism in the red gene locus of indigenous rice populations of Bangladesh. Plant Systematics and Evolution 296, 157–165.
476
Fan, C.C., Yu, X.Q., Xing, Y.Z., Xu, C.G., Luo, L.J., Zhang, Q., 2005. The main effects, epistatic
477
effects and environmental interactions of QTLs on the cooking and eating quality of rice in a
478
doubled-haploid line population. Theoretical and Applied Genetics 110, 1445–1452.
479 480 481
Huson, D.H., Bryant, D., 2006. Application of Phylogenetic Networks in Evolutionary Studies. Molecular Biology and Evolution. 23, 254–267. Inukai, T., Sako, A., Hirano, H.Y., Sano, Y., 2000. Analysis of intragenic recombination at wx in rice:
18
ACCEPTED MANUSCRIPT 482
Correlation between the molecular and genetic maps within the locus. Genome 43, 589–596.
483
Juliano, B.O., Perez, C.M., Blakeney, A.B., Castillo, T., Kongseree, N., Laignelet, B., Lapis, E.T.,
484
Murty, V.V.S., Paule, C.M., Webb, B.D., 1981. International Cooperative Testing on the
485
Amylose Content of Milled Rice. Starch/Stärke 33, 157–162.
487
Larkin, P.D., Park, W.D., 2003. Association of waxy gene single nucleotide polymorphisms with
RI PT
486
starch characteristics in rice (Oryza sativa L.). Molecular Breeding 12, 335–339.
Liu, L., Ma, X., Liu, S., Zhu, C., Jiang, L., Wang, Y., Shen, Y., Ren, Y., Dong, H., Chen, L., others,
489
2009. Identification and characterization of a novel Waxy allele from a Yunnan rice landrace.
490
Plant Molecular Biology 71, 609–626.
SC
488
Martin, C., Smith, A.M., 1995. Starch biosynthesis. Plant Cell 7, 971–985.
492
Mikami, I., Aikawa, M., Hirano, H.Y., Sano, Y., 1999. Altered tissue-specific expression at the Wx
493
M AN U
491
gene of the opaque mutants in rice. Euphytica 105, 91–97.
Mikami, I., Uwatoko, N., Ikeda, Y., Yamaguchi, J., Hirano, H., Suzuki, Y., Sano, Y., 2008. Allelic
495
diversification at the wx locus in landraces of Asian rice. Theoretical and Applied Genetics
496
116, 979–989.
499 500 501 502 503 504
relatives. Oxford Surveys in Evolutionary Biology 8, 135–135. Okagaki, R.J., Wessler, S.R., 1988. Comparison of non-mutant and mutant waxy genes in rice and
EP
498
Morishima, H., Sano, Y., Oka, H.I., 1992. Evolutionary studies in cultivated rice and its wild
maize. Genetics 120, 1137–1143. Olsen, K.M., Purugganan, M.D., 2002. Molecular evidence on the origin and evolution of glutinous
AC C
497
TE D
494
rice. Genetics 162, 941–950.
Prathepha, P., 2007. Identification of variant transcripts of waxy gene in non-glutinous rice (O. sativa L.) with different amylose content. Pakistan Journal of Biological Sciences 10, 2500–2504.
505
Sato, H., Suzuki, Y., Sakai, M., Imbe, T., 2002. Molecular Characterization of Wx-mq, a Novel
506
Mutant Gene for Low-amylose Content in Endosperm of Rice (Oryza sativa L.). Breeding
507
Science 52, 131–135.
508 509
Sridharan, V., Singh, R., 2007. A conditional role of U2AF in splicing of introns with unconventional polypyrimidine tracts. Molecular and Cellular Biology 27, 7334–7344.
19
ACCEPTED MANUSCRIPT 510 511
Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: molecular evolutionary genetics analysis version 6.0. Molecular Biology and Evolution 30, 2725–2729. Thompson, J.D., Gibson, T.J., Higgins, D.G., 2002. Multiple Sequence Alignment Using ClustalW
513
and ClustalX. In: Baxevanis, A.D., Stein, L.D., Stormo, G.D., Yates, J.R. (Eds.), Current
514
Protocols in Bioinformatics. John Wiley & Sons, New York, pp. 2.3.1–2.3.22.
515 516
RI PT
512
Wanchana, S., Toojinda, T., Tragoonrung, S., Vanavichit, A., 2003. Duplicated coding sequence in the waxy allele of tropical glutinous rice (Oryza sativa L.). Plant Science 165, 1193–1199. Yamanaka, S., Nakamura, I., Watanabe, K.N., Sato, Y.I., 2004. Identification of SNPs in the waxy
518
gene among glutinous rice cultivars and their evolutionary significance during the
519
domestication process of rice. Theoretical and Applied Genetics 108, 1200–1204.
SC
517
Zeng, D., Yan, M., Wang, Y., Liu, X., Qian, Q., Li, J., 2007. Du1, encoding a novel Prp1 protein,
521
regulates starch biosynthesis through affecting the splicing of Wxb pre-mRNAs in rice (Oryza
522
sativa L.). Plant Molecular Biology 65, 501–509.
M AN U
520
AC C
EP
TE D
523
20
ACCEPTED MANUSCRIPT Figure Legends
2
Fig. 1: Amylose content and nucleotide variation in Wx upstream regulatory region of twenty-two
3
Bangladeshi rice cultivars. Comparison of amylose contents in (A) non-glutinous and (B) Beruin
4
(locally known as glutinous) cultivars revealed three distinct categories. Nucleotide polymorphisms in
5
~2.7 kb Wx upstream noncoding region of these varieties with respect to known non-glutinous and
6
glutinous progenitor haplotypes (Olsen and Purugganan, 2002) are shown in (C).
RI PT
1
7
Fig. 2: Phylogenetic tree based on Wx upstream regulatory region showing the relationship among
9
Bangladeshi cultivars and other reported haplotypes. Neighbour-joining method along with Kimura 2-
10
parameter distance were utilized for phylogenetic analysis. Bootstrap values (>40%) for 1000
11
replicates are shown at the nodes.
M AN U
12
SC
8
13
Fig. 3: Chalky endosperm of Khara Beruin, partially chalky in Modhu Beruin White and nearly
14
transparent endosperm of non-glutinous cultivar Kataribhog.
TE D
15
Fig. 4: Expression of Wx gene in Beruin cultivars lacking G/T SNP in intron 1 splice donor site. (A)
17
Semi-quantitative RT-PCR analysis of Wx gene expression in developing seeds of selected Beruin
18
cultivars. Nipponbare, which carries G/T SNP in intron 1, was used as a control. RT- PCR with two
19
different primer sets (B and C) re-confirmed that intron 1 is efficiently spliced in the selected Beruin
20
cultivars without the G/T splice site SNP.
AC C
21
EP
16
22
Fig. 5: Median-joining network for Bangladeshi cultivars and reported haplotypes based on Wx
23
upstream regulatory region. Here magenta and blue colors indicate Bangladeshi glutinous Beruin and
24
non-glutinous cultivars, respectively. Purple color indicates cultivars that were newly assigned as non-
25
glutinous (this study) despite being locally known as Beruin or sticky rice. Squares and diamonds
26
represent previously reported non-glutinous and glutinous haplotypes across Asia (Olsen and
27
Purugganan, 2002), respectively. Circles represent nodes in the network. Node size is proportional to
28
number of cultivars/haplotypes present in each node.
1
ACCEPTED MANUSCRIPT
Table 1. Nucleotide polymorphisms in Wx coding region of selected Bangladeshi cultivars.
RI PT
Region in Position Position Polymorph Reported Shamudro Kathali Modhu Khara Mou Wx locus from from 5' -ism type Wx allele -phena Beruin Beruin Beruin Beruin start splice Name (nonRed White (glutinous codon site of glutinous reference) intron reference) Exon 2 100 23 bp wx,a No No No No Yes duplication Exon 4 476 SNP (ns) Wxhp(A/G)b A A A A A Exon 4 497 SNP (ns) Wxmq(G/A) G G G G G
SC
c
SNP (ns) Wx (A/G)a A A SNP(ns) Wxmq(T/C)c T T SNP (ns) Wxin(A/C) d C A SNP (s) T C SNP (ns) C T 94 Indel C 29 SNP A A 81 SNP G A 95 SNP G A 104- 106 indel ATA 118 SNP G A Here ‘ns’ and ‘s’ refers to nonsynonymous and synonymous amino acid 521 574 671 1109 1266
M AN U
Exon 4 Exon 5 Exon 6 Exon 9 Exon 10 Intron 5 Intron 9 Intron 10
op
A A A T T T A A A C C T C C C C G G A A A G A A G ATA G G G changes, respectively.
TE D
Sites without polymorphism in any of the cultivars are shaded in grey. a(Mikami et al., 2008,
AC C
EP
1999) ,b(Liu et al., 2009), c(Sato et al., 2002), d(Dobo et al., 2010)
ACCEPTED MANUSCRIPT
Non-glutinous cultivars 23.1
23.3 23.1 22.5 23.1 23.3
21.48
25
21.48
23.4 22.6
20
15
15 10
5
5
0
4.5 4.5 5.4
5.2 4.9 4.4
0
Low AC (6-18%)
C 1
2
Promoter
5 6 7
8
9
23 56 57 67 216 255 256 307 333 350 387 398 475 534 677 718 1250 1283
Exon 1 5ʹ UTR
10
11 12 13
14
Intron 1
Exon 2 5ʹ UTR
A A A A A G G G G G G G
C C C C C T T T T T T T
T T T T T T T T T T T T
G G G G G G G
C C C C C C C
G G G G G G G G G G G G
T T T T T C C C C C C C
T T T T T G G G G G G G
T T T T T C C C C C C C
A A A A A G G G G G G G
C C C C C T T T T T T T
G G G G G A A A A A A A
T -
G G G G A G G G G G G G
A A A A A G G G G G G G
C 16-20 G G T A 5 A C A G C A C 18 G G T A 5 A C A G C A C 18 G G T A 5 A C A G C A C 17 G G T A 6 A C A G C A C 17 G G T A 5 A C A G C A T 12 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 10 G A T A 5 G C G G T C T 10 G A - - 5 A C G G T C T 10 G A - - 5 A C G G T C
T T T T T G G G G G G G
G G G G G G G G G G G G
C C C C C T T T T T T T
T T T T T C C C C C C C
C -
A A A A A A A A A A -
A A A A A A A A A G A A
T T T T T C C C C C C C
T T
G G G G G G G G G G A G
C C C C C C C C C A A A
T T T T T T T C T C C C
A A A A A A A G A G G G
C C C C C C C T C T T T
T T T T T T T A T T T T
G G
C C
G G G G G G G G G G G C
T T T T T T T T T T C C
T T T T T T T G T T G G
T T T T T T T C T T C C
A A A A A A A A A A G G
C C C C C C C T C C T T
G G G G G G G A G G A A
T -
G G G G A A G A A G G A
A A A A A A G G A G G G
C 16-18 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A C 17 T G T A 5 A C A G C A T 16 G A T A 5 G C G - T C T 8 G A T A 6 G T G - T C C 17 T G T A 5 A C A G C A T 11 G A T A 6 G T G - T C T 11 G A T A 6 G T G - T C T 10 G A - - 5 A C A G T C
T T T T T T G G T G G G
G G G G G G G T G G G G
C C C C C C T T C T T T
T T T T T T C C T C C C
-
A A A A A A A A A A -
A A A A A A A A A A A A
T T T T T T C C T C C C
T
G G G G G G G G G G G G
TE D
T T T T T C C C C C C C
EP
Haplotype G Akia Beruin White Akia Beruin Red Mou Beruin Pak Beruin Kathali Beruin White Modhu Beruin White Khara Beruin Kalo Beruin Kathali Beruin Red Modhu Beruin Red Push Beruin
4
C C C C C A A A A C A A
AC C
Non-glutinous
Haplotype F Kataribhog Chinigura Kalijira Shamudrophena Lathial White Lathial Red Girishail Raujan 2 Balam Red Balam White Raujan
Beruin (local glutinous)
Type Cultivar name
3
4.2
Waxy AC (0-5%)
M AN U
~2.7kb
1505,(CT)n
Wx Exons
SC
Ak ia
Ka
Intermediate AC (19-23%)
8.2
RI PT
10
13.7
2733
19.1
20.5
1600, Splice site 1685 1686 1687 1747, (AATT)n 1819 1846 1863 1902 1914 1917 1927 1934 1951 1953 2016 2171 2285 2304 2584
20
20.7
Amylose Content (%) B Ak eru ia in Be W ru hite M in R ou e Ka d th P Be al ak r u M i B B in od e r er u uin uin Be W ru h Kh in W ite ar a hit Ka K Be e a r th a lo ui M li B Be n od e r u hu ru in Be in R ru e Pu in d sh Re Be d ru in
25
Beruin (local glutinous) cultivars
B
ta ri C bho hi ni g gu Sh am K ra a ud liji La rop ra th he ia na La l W th hit ia e lR G ed iri R sha au il Ba jan la 2 Ba m la Re m d W hi t R e au ja n
Amylose Content (%)
A
ACCEPTED MANUSCRIPT
64
Bangladeshi non-glutinous Beruin cultivars Bangladeshi non-glutinous cultivars Known non-glutinous haplotypes Non-glutinous progenitor haplotype
82
Modhu Beruin White Kathali Beruin Red
52 64
AC C
EP
0.001
66
TE D
50
Haplotype S Khara Beruin Haplotype C Haplotype E Haplotype D Push Beruin Balam Red Balam White Haplotype B Raujan Haplotype A Modhu Beruin Red Raujan 2 Girishail 74 Lathial Red Lathial White
Clade 1 CT16-20
Sub clade 1.2 G at Intron 1 splice site (except haplotype R)
M AN U
93
59
Sub clade 1.1 T at Intron 1 splice site (except Shamudrophena)
RI PT
Glutinous progenitor haplotype Bangladeshi glutinous Beruin cultivars
SC
Haplotype K Haplotype L Akia Beruin Red Mou Beruin Haplotype M 44 Haplotype N 64 Haplotype Q Akia Beruin White Haplotype I Haplotype G Shamudrophena Sub clade 1.1.1 Kalo Beruin 718 G/A at 98 54 Kathali Beruin White Wx promoter 46 Pak Beruin Kalijira Kataribhog Chinigura Haplotype F 48 Haplotype P Haplotype J Haplotype H 64 Haplotype R
Known Glutinous haplotypes
Sub clade 2.1 CT8 (except haplotype E) Sub clade 2.2 CT10
Sub clade 2.3 CT 11-12
Clade 2 CT8-12 G at Intron 1 splice site
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C EP
200 bp 120 bp 100 bp
C
210 bp
1.6 kbp 1.1 kbp 1 kbp
SC
Push Beruin
Kathali Beruin Red
Modhu Beruin Red
Modhu Beruin White
Nipponbare
Marker
RI PT
Kathali Beruin Red
Push Beruin
Modhu Beruin Red
M AN U
Kathali Beruin Red
TE D
Push Beruin
Modhu Beruin Red
B Modhu Beruin White
Nipponbare
Marker
A
Modhu Beruin White
Nipponbare
Marker
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
SC
Raujan 2
Push Beruin C
Khara Beruin
M AN U
Modhu Beruin White
TE D
B , Raujan
K
R
Kathali Beruin Red
EP
Balam Red
Q
AC C
E
RI PT
A , Girishail, Lathial Red, Balam White Lathial White, Modhu Beruin Red
D
S
Pak Beruin, Kalo Beruin, Kathali Beruin White
Mou Beruin, Akia Beruin Red, Akia Beruin White
Chinigura, Kataribhog
Shamudrophena P
G F
M
Kalijira
J
I N
L
H
ACCEPTED MANUSCRIPT
•
Sequence polymorphism in Wx locus of 22 Bangladeshi rice landraces was analyzed
•
Most of the Bangladeshi landraces showed SNP patterns reported in South & Southeast Asian varieties Three Beruin cultivars lacked the known glutinous alleles but had waxy or low amylose
•
Novel Wx promoter/intronic mutations in these 3 Beruin cultivars may explain their
RI PT
•
AC C
EP
TE D
M AN U
SC
glutinous phenotype