Content uploaded by Ryan J Andres
Author content
All content in this area was uploaded by Ryan J Andres on Mar 21, 2016
Content may be subject to copyright.
crop science, vol. 56, m ay–june 2016 www.crops.org 1
ReseaRch
C is the world’s most important source of natural ber
as well as one of its leading oilseed crops. Cotton belongs
to the genus Gossypium, which is comprised of both tetraploid
and diploid species. Currently, about 44 species of diploid cotton
(2n = 2x = 26) are known to exist across the eight genomes
(A–G and K), which are spread throughout the arid and semiarid
regions of the tropics (Hutchinson et al., 1947; Saunders, 1961;
Wendel et al., 2009; Wendel and Grover, 2015). Spinnable cotton
bers evolved only in the A-genome diploids, and both remaining
A-genome species, G. arboreum and G. herbaceum L., were domes-
ticated independently (Wendel et al., 2009). Of these A-genome
species, G. arboreum is a commercially important diploid cotton
mostly grown in Asia. About 1 to 2 million years ago (MYA), an
A-genome diploid hybridized with a D-genome diploid to form
an allopolyploid species (2n = 4x = 52, AADD) (Wendel et al.,
2009). The diploid D-genome donor was most closely related to
G. raimondii, while the two remaining A-genome diploids appear
equally closely related to the A-genome donor (Wendel et al.,
2009). The polyploid then spread throughout the tropics of the
New World, spawning at least six dierent species, two of which
(G. hirsutum and G. barbadense L.) were independently domesti-
cated (Wendel et al., 2009).
The genus Gossypium shows a wide geographical distribution
across the globe with a multitude of growth, developmental, and
morphological attributes (Hutchinson et al., 1947; Wendel and
Major Leaf Shape Genes, Laciniate in Diploid
Cotton and Okra in Polyploid Upland Cotton,
Map to an Orthologous Genomic Region
Baljinder Kaur, Ryan Andres, and Vasu Kuraparthy*
ABSTRACT
Gossypium arboreum L, which produces spin-
nable cotton bers, is an A-genome diploid
progenitor species of tetraploid cotton. With
its diploid genome, publicly available genome
sequence, adapted growth, and developmental
and agronomic attributes, G. arboreum could
make an ideal cot ton spe cies to study the genetic
basis of biological traits that are controlled by
orthologous loci in diploid and polyploid spe-
cies. Leaf shape is an important agronomic trait
in cotton. Normal, subokra, okra, and laciniate
are the predominant leaf shapes in cotton culti-
vars. Laciniate in diploids is phenotypically simi-
lar to okra leaf shape in tetraploid. In the pres-
ent study, a population of 135 F2 plants derived
from accessions NC 501 and NC 505 was used
for genetic and molecular mapping of lacini-
ate leaf shape in diploid cotton (G. arboreum).
An inheritance study showed that laciniate leaf
shape was controlled by a single incompletely
dominant gene (LL–A2). Molecular genetic map-
ping using simple-sequence repeat (SSR) mark-
ers placed the leaf shape locus L- A 2 on chromo-
some 2. Targeted mapping using putative genes
from the delineated region established that
laciniate leaf shape in G. arboreum and okra leaf
shape in Gossypium hirsutum L. were controlled
by genes at orthologous loci. Collinearity was
well conserved between the diploid A- (G. arbo-
reum) and D- (G. raimondii Ulbr.) genomes in the
targeted genomic region narrowing the candi-
date region for the leaf shape locus (L- A 2) to nine
putative genes. Establishing the orthologous
genomic region for the L loci could help use the
diploid cotton resources toward map-based
cloning of leaf shape genes in Gossypium.
B. Kaur, R. Andres, and V. Kuraparthy, Crop Science Dep., North Caro-
lina State Univ., Raleigh, NC 27695, USA. Accepted 8 Jan. 2016. Received
14 Oct. 2015. *Corresponding author (vasu_kuraparthy@ncsu.edu).
Abbreviations: HDC, high-density consensus; LOD, logarithm of
odds; PCR, polymerase chain reaction; SSR, simple-sequence repeat;
STS, sequence-tagged site.
Published in Crop Sci. 56:1–11 (2016).
doi: 10.2135/cropsci2015.10.0627
© Crop Science Societ y of America | 5585 Guilford Rd., Madison, WI 53711 USA
All rights reserved.
Published March 11, 2016
2 www.c rops .org crop scie nce, v ol. 56, may–june 2016
Cronn, 2003; Wendel et al., 2010). Remarkable phenotypic
diversity exists for leaf shape in cotton, ranging widely
from overtly simple to deeply lobed leaves across both the
diploids and polyploids (Hammond, 1941; Hutchinson et
al., 1947; Saunders, 1961). Although the role of leaf shape
diversity in the evolution and adaptation of the Gossyp-
ium genus is not yet clearly established, leaf shape plays an
important role in cotton production. Leaf shape aects the
plant and crop canopy architecture and can inuence yield,
biotic stress tolerance, earliness, input use eciency, and
other production characteristics in cotton.
The predominant leaf shapes in cultivated upland
cotton are normal, subokra, and okra. All of these leaf
shapes, along with super okra, form an allelic series and
map to a single locus (L; (renamed here as L-D1) in the
D-subgenome of upland cotton ( Jones, 1982). Among
these leaf shapes, okra leaf shape (Fig. 1) is of particular
interest in cotton production (Green, 1953; reviewed in
Andres et al., 2014; Jiang et al., 2000). It is also used to
study the biological basis of leaf shape variation in plants
(Dolan and Poethig, 1991). The okra leaf shape gene (LO)
of the D-subgenome was shown to be incompletely domi-
nant to normal leaf shape and the L-D1 locus was mapped
using SSR markers on chromosome 15 of upland cotton
(Andres et al., 2014; Jiang et al., 2000).
Genes at the homoeologous locus in the A-genome
were also found to control leaf shapes such as laciniate
in tetraploid cotton (Endrizzi and Stein, 1975). However,
to date, mapping information for the allelic series of leaf
shape at the A-subgenome locus is not available, and the
orthologous relationship with the major leaf shape alleles
of the D-subgenome has not been established.
Cultivated diploid cottons (G. arboreum and G. herba-
ceum) show wide variation for leaf shape and size (Hutchin-
son, 1934). In a series of crosses among and between the
diploid A-genome species, Hutchinson (1934) demonstrated
that there existed ve leaf shapes in the Asiatic cottons G.
arboreum and G. herbaceum, all of which are allelomorphic:
laciniate (LL), arboreum (L), recessive broad (l), mutant
broad (LB), and mutant intermediate (LI). However, linkage
and chromosome map location of these genes are not avail-
able to date, and their orthologous relationship with major
leaf shape genes in tetraploid cotton are not known.
Gossypium arboreum, which is a domesticated, diploid
(2n = 2x = 26, A2A2) cotton, oers a unique opportu-
nity to study the molecular genetic basis of agronomically
important biological traits. Because of its diploid nature,
trait mapping is less cumbersome in G. arboreum than
mapping in the tetraploid with its duplicated genome and
corresponding genetic redundancy (Li et al., 2014). Gos-
sypium arboreum shows similar growth, developmental, and
agronomic attributes to tetraploid cotton (Hutchinson et
al., 1947), making it a valid comparative model. Further-
more, the availability of a draft genome sequence and the
presence of higher allelic diversity (Ma et al., 2008; Li et
al., 2014; Lu et al., 2015) could make G. arboreum an ideal
cotton species to study the genetic basis of biological traits,
specically the traits that are controlled by orthologous
loci in diploid and polyploid species.
The objectives of the current study were to (i) study
the inheritance of the laciniate leaf shape trait in diploid
cotton, (ii) genetically map the laciniate leaf shape gene
(LL–A2) in diploid cotton, and (iii) study the orthologous
relationship between the major leaf shape genes in diploid
and tetraploid cottons.
MATERIALS AND METHODS
Plant Material
Accession NC 505 (PI 615700, A2-0191), named as Chinese
Narrow Leaf, is a laciniate leaf shape line of G. arboreum from
China, whereas NC 501 (PI 167905, A15.02SD) with recessive
broad leaf shape (Fig. 1; Supplemental Fig. S1) was originally
collected by J. R. Harlan from Turkey in 1948. Seeds of both
lines were procured from the USDA Cotton Germplasm Col-
lection in College Station, TX. Line NC 505 was crossed with
NC 501 during fall 2013. A single F1 plant was selfed in the
greenhouse to obtain F2 seed in spring 2014. During fall 2014,
135 F2 plants along with the two parental lines were planted
in the greenhouse to record the phenotype for leaf shape and
Fig. 1. Leaf shape phenotypes of diploid cotton (Gossypium arbo-
reum) and tetraploid upland cotton (G. hirsutum). Top: (left) normal
broad-shaped leaf of the accession NC 501, (right) laciniate leaf of
the breeding line NC 505. Bottom: (left) normal leaf of accession
NC11-2100, (right) okra leaf of the breeding line NC05AZ21.
crop science, vol. 56, may–jun e 2016 www.crops.org 3
For nding closely linked markers and genomic targeting of
LL–A2 gene, current and previously mapped markers (SSR and
sequence-tagged site [STS]) were BLAST searched against the
publically available G. arboreum genome sequence (G. arboreum
A-genome BGI-CGP v2.0 [annotation v1.0]). Sequence-tagged
site markers were designed from putative genes from the ten-
tatively identied genomic region in G. arboreum. All the
primers were designed using Primer3 software (http://bioto-
ols.umassmed.edu/bioapps/primer3_www.cgi) and synthesized
by Integrated DNA Technologies (Coralville, IA). A M13 tail
sequence 5- CACGACGT TGTA A A ACGAC-3 was added to
the 5 end of all the forward primers to resolve the PCR prod-
ucts with capillary-based gel electrophoresis (Schuelke, 2000).
Mapped SSR and STS markers closely linked to the leaf
shape trait in the current linkage map and from Andres et al.
(2014) were then used to establish the orthologous relationship
between the L-D1 locus of upland cotton and L-A2 locus of dip-
loid cotton. The candidate genomic region for leaf shape genes
was established in the sequenced diploid A- and D-genomes and
the A-subgenome of upland cotton using markers mapping to
the orthologous gene sequences. To do this, marker sequences
were BLAST searched against sequenced G. hirsutum accession
TM1 (Zhang et al., 2015), G. raimondii (DOE Joint Genome
Institute: Cotton D V2.0), and G. arboreum cultivar Shixiya 1
(Li et al., 2014) genomes. Further, putative genes placed in the
genomic region of leaf shape in the A-genome by the BGI-
CGP sequence annotation were studied for collinearity to the
JGI G. raimondii genome ( JGI assembly v2.0 [annotation v2.1];
Paterson et al., 2012) and BLAST searched to nd the correct
orthologous genes. Usage of the JGI G. raimondii genome at
Phytozome allowed gene function to be predicted through the
protein homologs and gene ancestry tools of Phytozome. Of
the three sequenced genomes used for BLAST analysis G. hir-
sutum accession TM1 is a normal leaf shaped upland cotton, G.
arboreum cultivar Shixiya1 was a broad-leaf cultivar (Chen et
al., 2015) and G. raimondii shows simple leaf shape (Hammond,
1941; Hutchinson et al., 1947; Saunders, 1961).
Based on the collinearity map using sequenced A- and
D-genome maps as bridging species, STS markers were then
designed to nd newer and closely linked markers to the LL–
A2 gene and to further establish the orthologous relationship
between A- and D-genome leaf shape loci. Polymorphic STS
markers were synthesized, amplied, and integrated into the
genetic map as described above. A comparative map between
chromosome 2 of the G. raimondii draft genome, chromosome
2 of the G. arboreum draft sequence, chromosome 15 of the G.
hirsutum LSMapPop map (Andres et al., 2014), and molecular
genetic map of the LL–A2 gene of the current study was con-
structed using the Strudel software ( JHI Plant Bioinformatics)
and redrawn to the scale in Microsoft PowerPoint.
RESULTS
Inheritance of Leaf Shape Trait
in Gossypium arboreum
The F1 hybrid between NC 505 NC 501 showed inter-
mediate phenotype compared with the parents (Fig. 1;
Supplemental Fig. S1). It indicated that the leaf shape trait
was incompletely dominant in the heterozygous condition.
collect leaf samples for molecular genetic mapping. All plants
were grown in 25.4-cm pots (Hummert International, Inc.) in
the greenhouse under short day photoper iod conditions with
supplemental lighting. Day temperature was set at 88C and
night temperature was set at 70C.
Phenotyping
Phenotypic data was recorded by visual observations 2 mo after
planting. The F2 plants were scored either as normal type or
okra type because the dierence between the heterozygotes and
okra types was occasionally ambiguous.
Molecular Genetic Mapping
Leaf tissue samples from parental lines and F2 plants were col-
lected and ground in liquid nitrogen. DNA was extracted using
a modied miniprep extraction protocol reported by Li et al.
(2001). Quantity and quality of the DNA was estimated by
using a NanoDrop 2000 UV-Vis spectrophotometer (Thermo
Fisher Scientic). Samples were diluted to a concentration of
20 ng mL−1 to carr y out the polymerase chain reaction (PCR)
amplication reactions. A nal volume of 6 mL per reaction was
used for PCR. Each reaction included 20 ng of genomic DNA,
1 reaction buer with 7.5 mM MgCl2, 0.24 m M dNTPs, 0.5
units of Taq DNA polymerase, 0.48 µM forward primer, 3.6
µM reverse primer, and 3.6 µM M13 primer labeled with either
HEX (hexachlorouorescein) or 6-FAM (6-carboxy uorescein)
uorescent tags. A touchdown program was used to amplify all
the primers starting with 5 min denaturation at 95C, then 15
cycles of 94C for 45 s, and 65 to 51C (1 cycle each degree) for
45 s, and 72C for 1 min followed by 25 cycles of 94C for 45
s, 50C for 45 s, 72C for 1 min, and a nal extension at 72C
for 10 min. To study the size polymorphism, the PCR products
from all primer pairs were run on a 3% GenePure high-res-
olution agarose gel (ISC BioExpress) and on an ABI 3730
capillary-based electrophoresis sequencer (Applied Biosystems).
For all capillary-based gel electrophoresis GeneScan–500 LIZ
(Applied Biosystems) was used as size standard. GeneMarker
V2.6.0 (SoftGenetics, 2013) software was used to visualize and
analyze the data obtained from the ABI 3730 sequencer.
Genetic Map Construction
A Chi-square test was conducted to check the goodness of t
for both the phenotypic and genotypic (marker) data. JoinMap
4.1 (Van Ooijen, 2006) was used to develop the linkage map
for the leaf shape trait. A logarithm of odds (LOD) score of 10.0
was used to develop the linkage map.
Marker and Comparative Genomic Analysis
Studies conducted in tetraploid upland cotton (G. hirsutum)
mapped the leaf shape locus (L-D2) on chromosome 15 in G.
hirsutum and its homoeologous chromosome 2 in G. raimondii
(Andres et al., 2014). Chromosome 15 of D-subgenome was
homoeologous to chromosome 1 of A-subgenome in upland
cotton (Blenda et al., 2012). For mapping the leaf shape gene in
G. arboreum, 33 SSR markers from chromosome 1 of the high-
density consensus (HDC) map (Blenda et al., 2012) and 23 SSR
markers from the Cotton Marker Database (http://www.cot-
tonmarker.org/) were selected.
4 www.c rops .org crop scie nce, v ol. 56, may–june 2016
While normal leaf shape phenotype was unambiguously
determined in the F2 population, the distinction between
okra type and heterozygotes was not pronounced for a
few F2 individuals. Therefore, the F2 plants were scored
as either okra type or normal leaf shape. Phenotypic data
showed that the F2 population segregated in a ratio of 3:1
of okra to normal type (2 = 0.062, p-value = 0.8033)
conrming monogenic control of the laciniate leaf shape
trait in G. arboreum.
Simple-Sequence Repeat Marker Analysis
and Genetic Mapping of LL–A2 Gene
Chromosome 15, which carries the okra leaf shape gene
of D-subgenome (Andres et al., 2014), was reported to
be homoeologous to chromosome 1 of tetraploid upland
cotton and chromosome 2 of diploid D-genome cotton
(Blenda et al., 2012; Li et al., 2014). Fifty-six SSR mark-
ers genetically and physically mapped on chromosome
1 in tetraploid cotton were used to assess the polymor-
phism between the parental accessions NC 505 and NC
501. Out of the 46 SSR markers that amplied, 18 (39.1%)
were polymorphic between the two parents. All the
SSR markers were codominant except for one (Table 1;
Supplemental Table S1). Polymorphic markers were used
for genotyping the F2 population and mapping of the
leaf shape gene. Genetic analysis using JoinMap with a
LOD score of 10.0 mapped the leaf shape locus (L-A2) on
chromosome 2. Six SSR markers showed linkage to the
L-A2 locus mapping at var ying genetic distances (Fig. 2).
Linked markers of the L-A2 locus showed a genetic map
length of 40.5 cM (Fig. 2).
Comparative Analysis of the LL–A2 Genetic
Map with Gossypium hirsutum High-Density
Consensus Map
Two linked SSR markers (DPL526 [MON_DPL0526] and
BNL1693 [CLU16]) anking the leaf shape locus L-A2 in the
current map (Fig. 2) were mapped toward the telomeric end
of chromosome 1 of the upland cotton HDC map (Blenda
et al., 2012; Fig. 2). DPL526 maps to 20 cM and BNL1693/
CLU16 to 32 cM on the 152-cM-long chromosome 1 in
the HDC map. This supported the tentative localization
of the L-A2 locus to the distal region of chromosome 2 in
Asiatic cotton. The SSR marker HAU2936, which maps
between DPL526 and the L-A2 locus (Fig. 2) was not pres-
ent on chromosome 1 of the HDC map but did map near
the telomere on the homeologous chromosome 15. The
remaining three linked SSRs in Fig. 2 all mapped proxi-
mally to BNL1693 (CLU16) on chromosome 1 in the HDC
map (Fig. 2). The order of the genetically mapped SSRs
in the present study (Fig. 2) was identical to their order on
chromosome 1 of the HDC map with minor dierences in
the marker distances (Fig. 2). This suggested that no major
rearrangements existed between G. arboreum chromosome 2
and G. hirsutum chromosome 1 in the genomic region of the
L-A2 locus. Although most of the SSRs linked to the okra
leaf shape locus (L-D1) in G. hirsutum (Andres et al., 2014)
showed amplication in the G. arboreum parents, none were
polymorphic (Supplemental Table S1).
Comparative Mapping of the LL–A2 Genetic
Map to the Diploid A- and D-Genomes
The six polymorphic SSRs were BLAST searched against
both the G. arboreum and the G. raimondii genomes to
establish the orthologous genomic region in both species.
Only two of the SSRs (DPL526 and HAU2936) identied
an orthologous genomic region on chromosome 2 of both
G. arboreum and G. raimondii (Table 2). Three of the mark-
ers (BNL1693, NAU2095, and BNL2921) showed high
sequence similarity to appropriate physical regions only on
chromosome 2 of G. raimondii, while CIR18 did not have
a high-scoring match on chromosome 2 in either species
(Table 2). Because none of the four proximal SSR mark-
ers could be found in the G. arboreum physical sequence of
chromosome 2, a physical candidate region could not be
constructed in the G. arboreum genome. Thus, a genome
annotation to identify candidate genes could not be com-
pleted using SSRs alone.
Table 1. Polymorphic simple-sequence repeat (SSR) and sequence-tagged site (STS) markers used for molecular mapping of
LL-A2 gene in diploid cotton Gossypium arboreum.
Marker
name
Marker
type Forward sequence 5-3Reverse sequence 5-3
Allele size
NC 501 NC 505
——————— b p ———————
MON_DPL0526 SSR GTTCTTGGTCATGCTGGTAAGA A A TAG C CATATC CA CCT TA GCA GAT T 176 17 3
HAU2936 SSR TGCGGGGACCAG AA AGAG AGT TTTGTCCTGGCCACCCAAGG 289 283
LS-GA-24 STS GCAACCCATTTTCATTCCAC TCCCTCTCATCCTCTGCA AT 221. 2 215.2, 225.3
LS-GA-23 STS GTGGCACTTCACCCATTTTT ATTCCATCA AACACGGCAAT 225.9 223.6
LS-GA-13 STS AAGGATGGTACCGGGGTAAG TGTGGCCATCTGCTAAATCA 227.3 226 .1
15- LSF M-7 STS TCATATAG ATATC GT T T T T GAC T TC C T TCCATGATTCCCA A AGACAAG ~480 Absent
BNL1693 SSR CCCTTGGGAATAGCAGGTG CATGTGTCTCCGTGTGTGTGTG 249 251
NAU2095 SSR GGGACACAA ACA AA ACACAC GGAACTTGAGAACTTGAAGG 194 200
BNL 2921 SSR CGAGAGATT TTAAAGG GA AACA GGGAGTGGTCTGATGGA AA A 19 3 243
CI R18 SSR T CA AC TATC AG TC CA AT AA AGAGACCCACA AG 195.5, 2 08 208
crop science, vol. 56, may–jun e 2016 www.crops.org 5
Sequence-Tagged Site Marker Development
and Genomic Targeting of LL–A2 Gene
To nd markers that mapped proximally to the L-A2 locus
in the genetic map, but could also be found in the G.
arboreum physical map, 93 STS markers were designed o
of 22 putative genes located proximally to HAU2936 in
the G. arboreum physical sequence. Of these 93 STS mark-
ers, 86 amplied in both parental lines and 7 (8.1%) were
polymorphic (Supplemental Table S1). With the excep-
tion of 15-LSFM-7, all the polymorphic STS markers
were codominant (Supplemental Table S1). Four of the
polymorphic STS markers (15-LSFM-7 and LS-GA-13,
23, and 24) were run on the F2 population and all four
markers showed tight linkage to the L-A2 gene (Fig. 2,
3). Two anking STS markers (LS-GA-13 and 15-LSFM-
7) showed especially close linkage with the L-A2 locus
mapping the LL–A2 gene within a 1.3-cM region on
chromosome 2 of G. arboreum. With respect to the L-A2
locus, LS-GA-13 mapped 0.9 cM distally and 15-LSFM-7
mapped 0.4 cM proximally on chromosome 2 of G. arbo-
reum. The order of the mapped STS markers in the genetic
map was consistent with their physical locations in the G.
arboreum genome sequence (Fig. 3; Table 2).
Fig. 2. Linkage map of LL–A2 gene on chromosome 2 of Gos-
sypium arboreum and its comparative map analysis with high-
density consensus map of homoelogous chromosome 1 of the
A-subgenome of upland cotton. In both maps, genetic distance in
centimorgans is on the left with marker names on the right, while
the top of the maps are oriented toward the telomere.
Table 2. Simple-sequence repeat (SSR) and sequence-tagged site (STS) markers used for orthologous mapping and genomic targeting of leaf shape gene (LL-A2) in cot-
ton. The SSR markers showing <90% similarity were considered to have no homologues in the physical maps of A-subgenome of Gossypium hirsutum (Zhang et al., 2015),
diploid A-genome (Li et al., 2014), and diploid D-genome (DOE Joint Genome Institute: Cotton D V2.0).
Marker
name
Marker
type
Physical position in
Chr.A01 NBI G. hirsutum
Physical position in
Chr.02 BGI.v2 G. arboreum
Physical position in
Chr.02 JGI G. raimondii
Gene no. in
G. arboreum
Gene no. in
G. raimondii
Similarity
between A and D
gene sequences
———————————————————————————— b p ———————————————————————————— %
DPL526 SSR 98,93 6,617–98,936,725 6 9,515 ,2 84– 69,515,7 09 61,837,914–61,838,339 n/a n/a n/a
HAU 2936 SSR 98 ,513, 87 7– 98, 512, 979 69,124,435– 69,124,705 61,40 0,237– 61,400,507 n/a n/a n /a
LS-GA-24 STS 9 8 , 0 6 7, 2 74 – 9 8 , 0 6 7, 3 72 68,723,133 – 68,723,3 35 61,010,898–61,011,100 Cotton_A_00484 Gorai.002G246000 79.9
LS-GA-23 STS 9 8 , 0 2 7, 4 3 8 – 9 8 , 0 2 7, 6 4 3 68,665,556 – 68,665,761 60,989,413–60,98 9,618 Cotton_A_00488 Gorai.002G245700 89.6
LS-GA-13† STS 9 7,8 74 ,74 7– 9 7,8 74 ,9 6 1 68,525,6 94–68,525,908 60,892,051–60,892,259 Cotton_A_00499 Gorai.002G244900 89.6
15- LS FM-7† STS 9 7, 7 71 ,15 6 – 9 7, 7 71 , 6 0 7 6 8 ,413 ,5 93 – 6 8 ,414, 0 44 60,79 6,820– 6 0,797,271 Cotton_A_00509 Gorai.002G243800 96.6
BNL1693 SSR 96,240,192–96,240,148 Chromosome 13 and 11 59,691,657–59,691,887 n/a n/a n /a
NAU2095 SSR 95,794,416– 95,794,196 Chromosome 13 59,354,418–59,354,593 n/a n/a n/a
BNL 2921 SSR 40,133,071–40,133,251 Chromosome 7 and 8 27,353,767–27,353,935 n/a n/a n/a
CI R18 SSR Chromosome A13 Chromosome 7 and 13 Chromosome 13 n/a n/a n/a
† Markers are flank ing marke rs to the laciniate leaf shape ge ne in both ge netic and physica l maps.
6 www.c rops .org crop scie nce, v ol. 56, may–june 2016
Annotation and Orthologous Mapping of
the Leaf Shape Gene (LL–A2) Candidate
Region using Gossypium arboreum,
Gossypium raimondii, and Gossypium
hirsutum Genomes
BLAST analysis using genetically mapped marker
sequences against sequenced diploid A-and D-genomes
and A-subgenome of G. hirsutum showed that the order
of the genetically mapped markers was similar to their
physical order in each of the sequence based maps,
suggesting that collinearity is well conserved among tet-
raploid A-subgenome and diploid A- and D-genomes in
the targeted candidate L-A2 leaf shape region. The collin-
earity of the putative orthologous gene sequences among
the sequenced diploid A- and D-genomes and A-subge-
nome of G. hirsutum and their collinear map location with
respect to the genetically linked markers to the leaf shape
genes LL–A2 and LO–D1 (Fig. 3; Table 2, 3) suggest that
laciniate leaf shape locus in diploid cotton is orthologous
to okra leaf shape locus of upland cotton.
The laciniate leaf shape candidate region in the diploid
A-genome spans a physical distance of ~108 kb from STS
marker LS-GA-13 (located within gene Cotton_A_00499)
to STS marker 15-LSFM-7 (located within gene
Cotton_A_00509). Nine genes (Cotton_A_00500 through
Cotton_A_00508) are predicted to lie between these two
markers (Table 3). Comparative genomic analysis using
these putative gene sequences showed that the gene order
and sequence similarity are highly conserved in the region
of the L-A2 locus among G. arboreum and G. raimondii and
A-subgenome of G. hirsutum (Table 2, 3). The one di erence
between the diploid genomes is that the ortholog of G.
arboreum gene Cotton_A_00502 is broken into two separate
genes (Gorai.002G244500 and Gorai.002G244600) in
G. raimondii (Table 3). Nevertheless, all three genes are
putatively characterized as exostosins and the sequence
similarity of the region between the two species is high at
88.8% (Table 3). Therefore, it is likely that this dierence
is the result of a dierence in annotation between the two
genomes rather than a large insertion or deletion. Almost
all of the genes share sequence similarity above 87%
between the two species (Table 3). The exceptions are
Cotton_A_00503 and Cotton_A_00506, both of which
are serine–threonine protein kinases (Table 3). However,
both Cotton_A_00503 and Cotton_A_00506 are predicted
to be considerably shorter than their D-genome orthologs.
Extension of the Cotton_ A_00503 and Cotton_A_00506
Fig. 3. Molecular mapping of the LL–A2 gene in Gossypium arboreum and its genomic location in relation to the sequence-based physical
maps of G. raimondii (2n = 2x = 26, DD) and G. arboreum (2n = 2x = 26, AA) genomes and to the tetraploid genetic map of Andres et al. (2014).
crop science, vol. 56, may–jun e 2016 www.crops.org 7
sequences to match that of their D-genome orthologs
resulted in signicantly improved percentage similarities.
O the nine putative genes annotated in the diploid
A-genome, genes Cotton_A_00505 and Cotton_A_00507
showed high homology to genes implicated in leaf shape
determination in other studies (Saddic et al., 2006; Andres
et al., 2014; Vlad et al., 2014; Sicard et al., 2014) (Table 3).
These two candidate leaf shape genes were 48.9 kb apart and
were separated by single gene Cotton_ A_00506 encoding a
serine–threonine protein kinase (Table 3). Further, genes
Cotton_ A_00505 and Cotton_A_00507 showed high
DNA sequence similarity (94.3 and 87.4%, respectively)
with their orthologs in the G. raimondii genome (Table 3).
In upland cotton, the orthologs of these two genes in the
D-subgenome were also identied as possible leaf shape
candidates by Andres et al. (2014). A marker developed
from one of the candidate genes cosegregated with the
okra leaf shape locus L-D1 (Andres et al., 2014). However,
preliminary attempts at developing markers based on its
diploid A-genome homeolog in the current study were not
Table 3. Annotation and comparative genomic analysis of putative gene sequences identified in the genomic region of ortholo-
gous leaf shape locus (L) using sequence based physical maps of diploid progenitor cotton species Gossypium arboreum (BGI-
CGP assembly v2.0 [annotation v1.0]; Li et al., 2014) and G. raimondii (JGI assembly v2.0 [annotation v2.1]; Paterson et al., 2012).
Physical (sequence-based) map coordinates
Similarity
between A and D
sequences Putative function
A-genome (G. arboreum)D-genome (G. raimondii)
Cotton_A_00
Physical position
chromosome 2 Gorai.002G
Physical position
chromosome 2
bp bp %
Gh565 6 8 ,74 5, 8 87– 6 8 ,74 6 ,114 Gh565 61, 03 1,9 75 – 61, 03 2,19 9 87. 5 SSR marker
482 6 8, 741, 40 9 – 6 8 ,74 5, 8 08 246200 61,02 7,658 – 61,032,210 82.6 Pectate Lyase
483 68,728,321–68,730,997 24610 0 61,015,799–61,018,395 8 7.7 Carbonic anhydrase
484 68,721,745– 6 8,724,118 246000 61,0 0 9, 256 –61, 011,9 27 79.9 Carbonic anhydrase
485 68,708,231–68,710,581 246000 61, 00 9 ,2 56 – 61, 011, 927 78.9 Carbonic anhydrase
486 68,701,883–68,703,112 2459 00 6 1,0 0 4,7 59 – 6 1,0 0 6,4 20 9 7. 2 Aquaporin transporter
487 68,669,424– 68,670,317 24580 0 60,992,779–60,993,648 95.2 Unknown
488 68,641,68–68,666,437 245700 60,987,773–60,99 0,298 89.6 Ankyrin repeat
489 68,657,366– 68,6 62,624 24560 0 60,979,680–60,986,732 97. 5 Kinase
490 68,615,778–68,623,408 245 50 0 60,950,898 –6 0,959,619 95.5 PH-D finger
491 68,591,320–68,599,029 24540 0 60,934,232–60,942,403 92.4 PH-D finger
492 68,572,895–68,576,363 245300 60,918,534– 60,922,736 95.3 Aspartyl protease
493 68,56 9,096– 68,572,179 245200 6 0,915,361–6 0,918,548 91.3 Isomerase
494 68,563,933–68,565,933 245100 60,909,449–60,912,281 96.4 Hydrolase
495 68,556,349– 68,563,129 245000 60,902,520– 60,910,968 8 1.2 Ubiquitin transferase
496 68,552,461–6 8,5 52,923 n/a 60,899,991–60,90 0,421 81.5 Unknown
497 68,550,904 –68,552,090 n /a No Match on Chr02 n/a Unknown
498 68,536,247–68,539,114 n/a 60,889,803– 60,892,670 87. 8 Epimerase
499† 68,523,600 –6 8,526,467 24 49 0 0 60,889,583–60,893,220 89.6 Epimerase
500 68, 517,140–68 ,519, 951 244 80 0 60,882,963 –60,886,378 91 Synthetase
501 68,513,273–68,515,930 24470 0 60,878,611–60,881,268 98 Pentratricopeptide repeat
502 68,493,999–68,502,968 24450 0 60,863,949–60,866,523 25.8 Exostosin
503 68,487,614–68,491,649 244400 60,856,168–60,863,202 60.3 Ser–Thr protein kinase
504 68,483,664 –6 8,485,162 24430 0 60,852,082–60,854,077 96.6 Ribosomal protein L24e
505 68,480,260 –68,481,502 24420 0 60,848,207–60,849,955 94.3 HD-Zip transcription factor
506 68,442,940– 68,444,771 24410 0 60,818,675–60,821,101 69.2 Ser–Thr protein kinase
507 68,432,568–68,433,521 244000 60,816,695 –60,817,56 5 8 7. 4 HD-Zip transcription factor
508 68 ,419, 35 2– 6 8, 420,719 2439 00 6 0,802,515–60,804,472 93.6 Hypoxia response
509† 68,410,422–68,415,291 2438 00 6 0,793,342–60,798,916 96.6 Pyruvate kinase
510 68,40 0,894 –68,405,445 243700 60,784,25 8–60,789, 34 9 9 3.1 Pyruvate kinase
511 68,389,173–68,391,283 243600 60,775,425– 60,777,957 98.6 Ser–Thr protein kinase
512 68,377,517–68,379,223 24 35 00 60,763, 514– 60,767,171 9 7. 8 Unknown
513 68,351,915–68,356,673 243400 60,739,751– 60,745,4 94 88.6 Glycosyl transferase
514 68,346,525–68,348,978 24 330 0 60,734,924– 60,737,571 9 7. 8 Pentratricopeptide repeat
515 68,345,531–68,346,214 2432 00 6 0,733, 89 2– 6 0,73 4,9 23 98.8 Redoxin
516 68,327,789–68,331,092 243100 6 0,725,5 36 – 6 0,729,527 94.5 Enolase
517 68,281,323–68,284,053 243000 60,721,26 5–60,724,3 66 79.8 Unknown
518 68,274,602–68,280,195 24290 0 60,700,190–60,718,341 23.3 Ergosterol biosynthesis
NAU2343 68,283,937–68,284,089 NAU2343 60,699,180–60,69 9,332 89.2 SSR marker
† Putative genes used for devel oping ge netically mapp ed flanking STS markers LS-G A-13 and 15-LSFM-7.
8 www.c rops .org crop scie nce, v ol. 56, may–june 2016
successful as the primer pairs did not show polymorphism
between the parents (Supplemental Table S1).
The genomic region delineated by anking STS
markers (LS-GA-13 and 15-LSFM-7) in sequenced
A-genome was ~108 kb. This is equivalent to 95 kb in the
homoeologous diploid D-genome sequence and 103.14 kb
in the A-subgenome of G. hirsutum (Table 2, 3; Fig. 3).
Based on the anking STS markers, the physical sequence
size to genetic distance ratio in the genomic region of LL–
A2 was 86.3 kb cM−1.
Extension of Comparative Genomic Analysis
of Orthologous Leaf Shape Region in Diploid
A- and D-Genome Physical Maps
The annotation in G. arboreum was extended beyond the
L-A2 locus to range from the SSR markers Gh565 and
NAU2343. These SSRs dened the L-D1 candidate region
established in Andres et al. (2014) (Table 3). Neither Gh565
nor NAU2343 were polymorphic on the G. arboreum map-
ping population parents (Supplemental Table S1), but both
had high scoring matches in the G. arboreum physical
sequence. This expanded candidate region contained genes
Cotton_A_00482 through Cotton_A_00518 (Table 3).
The order of the genes remains highly conserved
between the two diploid species and all gene sequences
have percentage similarities of at least 80% with the two
exceptions noted in the previous section (Table 3). The
annotation of the G. arboreum sequence contains three
genes (Cotton_ A_00496–00498) that appear to lack a
clear homeolog in G. raimondii. Genes Cotton_A_00496
and Cotton_A_00498 show high sequence similarity to
unannotated regions of G. raimondii chromosome 2 located
between the closest anking annotated genes in G. raimondii.
This indicates that these two genes are possibly the result of
dierences in the annotation of the two genomes rather than
a sizeable insertion or deletion. However, Cotton_A_00497
does not have a high-scoring match on G. raimondii
chromosome 2 and therefore may be a gene unique to the
diploid cotton A-genome. Gossypium arboreum also appears
to carry an extra carbonic anhydrase in the candidate region
as both Cotton_A_00484 and Cotton_A_00485 are most
similar to Gorai.002G246000. Nevertheless, the continued
strong collinearity between the anking genes of the L-A2
and L-D1 loci strengthens the notion that the two regions
may be conditioned by the orthologous genes.
The L-D1 locus maps to a position at ~60.8 Mb
on the ~62.8-Mb chromosome 2 of G. raimondii,
establishing that the gene is physically located close to
the telomere. In G. arboreum the L-A2 locus maps to
~68.4 Mb on chromosome 2. However, the ~100 Mb G.
arboreum chromosomes 2 is much longer than ~62.8 Mb
G. raimondii chromosome 2. Therefore, the L-A2 locus
does not appear to be telomeric in the A-genome diploid
cotton compared with its D-genome ortholog.
DISCUSSION
In a series of crosses among and between the dip-
loid A-genome species G. arboreum and G. herbaceum,
Hutchinson (1934) demonstrated ve leaf shapes in Asiatic
or diploid cotton, all of which are allelomorphic: lacini-
ate (LL), arboreum (L), recessive broad (l), mutant broad
(LB), and mutant intermediate (LI). Only laciniate, which
is phenotypically similar to okra, was transferred to G.
hirsutum by a Dr. C. Rhyne ca. 1960 (Endrizzi and Stein,
1975; Jones, 1982). The laciniate locus was placed on
chromosome 1 of the tetraploid A-subgenome in cytoge-
netic work using monosomes (White and Endrizzi, 1965).
Since okra and laciniate alleles have similar eects on leaf
shape, they were considered to be genes at duplicate loci
in the two genomes (White and Endrizzi, 1965). This
served as the basis for establishing that chromosome 1 and
chromosome 15 are homeologous chromosomes in tetra-
ploid cotton (White and Endrizzi, 1965). There exists no
mention of any of the other three-leaf shape alleles being
transferred to G. hirsutum from Asiatic cotton. However,
since they are considered an allelic series in Asiatic cotton,
it is assumed that they could also make up an allelic series
at chromosome 1, the A-genome homoeologue, in G. hir-
sutum ( Jones, 1982; Meredith, 1984). Some G. hirsutum
lines still exist today that purportedly carry the lacini-
ate allele. However, the term laciniate is occasionally used
interchangeably with okra and super okra in the litera-
ture. Therefore, which of these lines, if any, truly carry
the laciniate allele is unknown. In the current study, two
of the leaf shapes described by Hutchinson (1934) were
investigated for their genetics and relationship with the
okra leaf shape of the tetraploid upland cotton. Genetic
analysis showed that laciniate leaf shape gene (LL–A2) of
G. arboreum showed incomplete dominance similar to the
okra leaf shape of tetraploid upland cotton (Andres et al.,
2014). Molecular genetic mapping of the laciniate leaf
shape in diploid cotton indicated that LL–A2 gene mapped
on chromosome 2 of diploid cotton, and L-A2 locus was
orthologous to the okra leaf shape locus (L-D1) of upland
cotton (Fig. 3; Table 2).
Fine mapping and chromosome walking toward the
target gene is dicult in polyploids because of the genetic
redundancy of the homeoalleles in duplicated genomes.
In many cases, the progenitor or related diploid species
with smaller genomes have been sequenced ahead of
their agriculturally important polyploids. Such wild and
progenitor diploid species of polyploid crop plants were
successfully used for ne mapping and map-based cloning
of biological traits. Examples include the vernalization
genes VRN1 and VRN2 in wheat using Triticum monococcum
L. (Yan et al., 2003, 2004), disease resistance genes Lr10,
Pm3b, and Sr35 of wheat using T. monococcum (Feuillet et
al., 2003; Yahiaoui et al., 2004; Saintenac et al., 2013),
Lr21 disease resistance gene of wheat using Aegilops tauschii
crop science, vol. 56, may–jun e 2016 www.crops.org 9
Coss. (Huang et al., 2003), and the late blight resistance
gene RB in potato using the wild diploid potato species
Solanum bulbocastanum Dunal (Song et al., 2003). Although
the diploid D-genome of species G. raimondii is sequenced,
genetic an alysis is not a fea sible option in this spec ies because
these accessions are photoperiod-sensitive perennial wild
species with narrow morphological diversity within a
species. Wide phenotypic diversity exists among the
14 extant wild diploid D-genome species for leaf shape
(Hutchinson 1934; Fryxell, 1979; Ulloa, 2014). However,
developing interspecic crosses and segregating mapping
populations in the diploid D-genome species could be a
cumbersome process as a result of gametic and sporophytic
incompatibility, sterility in hybrids, hybrid breakdown,
segregation distortion, etc. (Saunders, 1961; He and Liang,
1989). In addition, genetic analysis in these wild diploid
D-genome species also would not be feasible because most
of the accessions are photoperiod-sensitive perennials that
show poor seed germination. Thus, genomic targeting
and orthologous mapping of LL–A2 gene in the adapted,
photoperiod-insensitive, and cultivated diploid A-genome
species oers an ideal opportunity to ne map and clone
the orthologous leaf shape locus (L) in cotton.
Orthologous relationships between the laciniate
gene LL–A2 of diploid cotton and the okra leaf shape
gene (LO–D1) of upland cotton was established by using
a combination of genetic mapping and comparative
mapping using sequenced A- and D-genomes (Fig. 3;
Table 2). This showed the utility of the physical maps,
especially the sequence-based maps of the progenitor
diploid species G. arboreum and G. raimondii in improving
the genetic mapping eciency in polyploid cotton. An
ideal orthologous map would involve cross-validating
the orthologous anking markers in the two segregating
mapping populations in diploid and polyploid cottons.
However, microsatellite markers are mostly genome
specic and, as such, cannot be used for orthologous
mapping between homoeologous chromosomes (Roder
et al., 1998). Therefore, the sequenced diploid A- and
D-genomes were used as bridging species in the shuttle
mapping while establishing orthologous relationship
between laciniate of diploid and okra of tetraploid cottons.
The okra leaf shape locus was previously targeted to
a genomic region of 337 kb containing 34 putative genes
in the diploid D-genome of cotton (Andres et al., 2014).
In the present study, by establishing the orthologous
relationship between okra leaf shape (LO-D1) of upland
cotton and laciniate leaf shape (LL–A2) of G. arboreum,
the candidate genomic region of the leaf shape locus was
narrowed to a ~108-kb region in the diploid A-genome,
which contains nine putative gene sequences (Fig. 3; Table
3). Its orthologous collinear region is equivalent to 95 kb
in the diploid D-genome and 103.14 kb in the G. hirsutum
A-subgenome sequences (Table 2, 3; Fig. 3). The physical
to genetic map ratio within the delineated A-genome
region was estimated to be 86.3 kb cM−1. This ratio is less
than the genomic average of 276 to 352 kb cM−1 estimated
previously in G. hirsutum (Xu et al., 2008). A smaller
physical size to genetic distance ratio further validates the
previous observation by Andres et al. (2014) that leaf shape
locus (L-D1) is localized toward the distal region of the
chromosome, which was characterized by its propensity
for high recombination and gene density (Rong et al.,
2004; Li et al., 2014; Wang et al., 2013; Werner et al.,
1992; Gill et al., 1996). Thus, eciency of genetic analyses
of agronomic traits in upland cotton can be improved by
using the diploid species mapping and genomic resources.
Since the major leaf shape loci (L) have been mapped
to homeologous regions in both cotton genomes using
orthologous gene sequences in the sequenced A- and
D-genomes (Fig. 3; Table 2), it is likely that the same gene
is responsible for leaf shape in both genomes. However,
most genes in the region of interest appear to have a gene of
similar function either in tandem or close proximity (Table
3), and it remains possible that leaf shape in the two genomes
may be inuenced by either or both of these two related
genes. Of particular interest from these nine genes are two
putative genes (Cotton_ A _00505 and Cotton_A_00507)
that code for HD-Zip transcription factor proteins. These
HD-Zip transcription factors were previously implicated in
leaf morphological dierences (Saddic et al., 2006; Vlad et
al., 2014; Sicard et al., 2014). An STS m arker developed based
on one of these genes showed cosegregation with leaf shape
phenotype in 236 F2 plants in a previous study by Andres
et al. (2014). Markers from its A-genome homoeologue
Cotton_A_00507 did not show polymorphism between
the parents used in the current study. Expanded eorts are
currently underway to ne map the leaf shape locus L-A2
and test the role of the two candidate genes in leaf shape
variation at L locus of cotton.
Supplemental Information Available
Supplemental Fig. S1. Leaf shape phenotypes of the G.
arboreum parental accessions and their F1 hybrid at ~65 d
after germination.
Acknowledgments
We thank North Carolina Cotton Growers Association Inc.
and Cotton Incorporated for funding this research through an
assistantship to Ms. Baljinder Kaur. We also thank Mr. Lin-
glong Zhu for his technical help and assistance at various stages
of this research. We appreciate the excellent technical help by
Jared Smith and Sharon Williamson with the sequencer-based
genotyping work. We appreciate Drs. Richard Percy and James
Frelichowski of the USDA–National Cotton Germplasm Col-
lection for supplying the G. arboreum parental accessions.
10 www.c rops .org crop scie nce, v ol. 56, may–june 2016
References
Andres, R.J., D.T. Bowman, B. Kaur, and V. Kurapar thy. 2014.
Mapping and genomic targeting of the major leaf shape gene (L)
in upland cotton (Gossypium hirsutum L.). Theor. Appl. Genet.
127:167–177. doi:10.1007/s00122-013-2208-4
Blenda, A., D. D. Fang, J.F. Rami, O. Garsmeur, F. Luo, and J.M.
Lacape. 2012. A high-density consensus genetic map of tetra-
ploid cotton that integrates multiple component maps through
molecular marker redundancy check. PLoS ONE 7:e45739.
doi:10.1371/j ou r n a l.pone.0045739
Chen, Y., Y. Wang, T. Zhao, J. Yang, S. Feng, W. Nazeer, T. Zhang,
and B. Zhou. 2015. A new synthetic amphiploid (AADDAA)
between Gossypium hirsutum and G. arboreum lays the foundation
for transferring resistances to ver ticillium and drought. PLoS
ONE 10:e0128981. doi:10.1371/j ou r n a l.pone.012 8981
Dolan, L., and R.S. Poethig. 1991. Genetic analysis of leaf develop-
ment in cotton. Development 1:39–46.
Endrizzi, J.E., and R. Stein. 1975. Association of t wo marker loci
with chromosome 1 in cotton. J. Hered. 66:75–78.
Feuillet, C., S. Travella, N. Stein, L. Albar, A. Nublat, and B. Keller.
2003. Map-based isolation of the leaf r ust disease resistance gene
Lr10 from the hexaploid wheat (Triticum aestivum L .) g enome.
Proc. Natl. Acad. Sci. USA 100:15253–15258. doi:10.1073/
pn a s.2435133100
Fry xell, P.A. 1979. The natural history of the cotton tribe. Texas
A&M Univ. Press, College Station, TX.
Gill, K.S., B.S. Gill, T.R. Endo, and T. Taylor. 1996. Identication
and high-densit y mapping of gene-rich regions in chromosome
group 1 of wheat. Genetics 144:1883–1891.
Green, J.M. 1953. Sub-okra, a new leaf shape in upland cotton. J.
Hered. 44:229–232.
Hammond, D. 1941. The expression of genes for leaf shape in Gossy-
pium hirsutum L. and Gossypium arboreum L. II. The expression of
genes for leaf shape in Gossypium hirsutum L. A m. J. Bot. 28:138–
150. doi:10.2307/2436937
He, J.X., and Z.L. Liang. 1989. Embryological studies on interspe-
cic cross bet ween Gossypium arboream L. and G. davidsonii Kel-
log. Acta Genet. Sin. (2006) 16:256–262.
Huang, L., S.A. Brooks, W. Li, J .P. Fellers, H.N. Trick, and B.S.
Gill. 2003. Map-based cloning of leaf rust resistance gene L r21
from the large and polyploid genome of bread wheat. Genetics
164:655– 664.
Hutchinson, J.B. 1934. The genetics of cotton. J. Genet. 28:437–
513. doi:10.1007/BF 029 81765
Hutchinson, J.B., R.A. Silow, and S.G. Stephens. 1947. The evolu-
tion of Gossypium. Oxford Univ. Press, London.
Jiang, C.X., R.J. Wrig ht, S.S. Woo, T. A . DelMonte, and A.H. Pat-
erson. 2000. QTL analysis of leaf morphology in tetraploid Gos-
sypium (cotton). Theor. Appl. Genet. 100:409–418. doi:10.100 7/
s001220050054
Jones, J.E. 1982 . The present state of the art and science of cotton
breeding for leaf-morphological types. Proc. Belt wide Cotton
Production Res. Conf. The National Cotton Council of Amer-
ica. Memphis, TN. p. 93–99.
Li, F., G. Fan, K. Wang, F. Sun, Y. Yu a n , G. Song, Q. Li, Z. Ma,
C. Lu, C. Zou, W. Chen, X. Liang, H. Shang, W. Liu, C. Shi,
G. Xiao, C. Gou, W. Ye , X. Xu, X. Zhang, H. Wei, Z. Li,
G. Zhang, J. Wa n g, K. Liu, R.J. Kohel, R.G. Percy, J.Z. Yu ,
Y.X . Zhu, J. Wa n g, and S. Yu. 2014. Genome sequence of the
cultivated cotton Gossypium arboreum. Nat. Genet. 46:567–572.
doi:10.1038/n g.2 987
Li, H., J. Luo, J.K. Hemphill, and J .T. Wan g. 2001. A rapid and high
yielding DNA miniprep for cotton (Gossypium spp.). Plant Mol.
Biol. Rep. 19:183. doi:10.10 07/BF0 2772162
Lu, C., C. Zou, Y. Zhang, D. Yu , H. Cheng, P. Jiang, W. Yan g,
Q. Wang, X. Feng, M.A. Prosper, X. Guo, and G. Song. 2015.
Development of chromosome-specic markers with high poly-
morphism for allotetraploid cotton based on genome-wide
characterization of simple sequence repeats in diploid cottons
(Gossypium arboreum L. and Gossypium raimondii U lbr ich). BMC
Genomics 16:55. doi:10.1186/s12 864-015 -1265 -2
Ma, X.X., B.L. Zhou, Y.H. Lu, W.Z . Guo, and T.Z. Zhang. 2008.
Simple sequence repeat genetic linkage maps of A-genome dip-
loid cotton (Gossypium arboreum). J. Integr. Plant Biol. 50:491–
502. do i:10.1111/j.1744 -7 909.2008.0 0 6 36.x
Meredith, W.R . 1984. Inuence of leaf morphology on lint yield of
cotton-enhancement by the sub okra trait. Crop Sci. 24:855–857.
doi:10.2135/cropsci1984.0011183X002400050007x
Paterson A.H., J.F. Wendel, H. Gundlach, H. Guo, J. Jenkins, D.
Jin, D. Llewellyn, K.C. Showmaker, S. Shu, J. Udall, M. Yoo, R.
Byers, W. Chen, A. Doron-Faigenboim, M.V. Duke, L. Gong,
J. Grimwood, C. Grover, K. Grupp, G. Hu, T. Lee, J. Li, L. Lin,
T. Liu, B.S. Marler, J.T. Page, A.W. Roberts, E. Romanel, W.S.
Sanders, E. Szadkowski, X. Tan, H. Tang, C. Xu, J. Wang, Z.
Wang, D. Zhang, L. Zhang, H. Ashra, F. Bedon, J. E. Bowers,
C.L. Brubaker, P.W. Chee, S. Das, A.R. Gingle, C.H. Haigler,
D. Harker, L.V. Homann, R. Hovav, D.C. Jones, C. Lemke,
S. Mansoor, M. Rahman, L.N. Rainville, A. Rambani, U.K.
Reddy, J. Rong, Y. Saranga, B.E. Scheer, J.A. Scheer, D.M.
Stelly, B.A. Triplett, A.V. Deynze, M.F.S. Vaslin, V.N. Wagh-
mare, S.A. Walford, R.J. Wright, E.A. Zaki, T. Zhang, E.S.
Dennis, K.F.X. Mayer, D.G. Peterson, D.S. Rokhsar, X. Wang
and J. Schmutz. 2012. Repeated polyploidization of Gossypium
genomes and the evolution of spinnable cotton bres. Nature
492:423–427. do i:10.1038/nat u re11798
Roder, M.S., V. Korzun, K. Wendeh a ke, J. Plaschke, M.H. Tixier,
P. Leroy, and M .W. Ganal. 199 8. A microsatel lite map of wheat.
Genetics 149:2007–2023.
Rong, J., C. Abbey, J.E. Bowers, C.L. Brubaker, C. Chang, P.W.
Chee, T.A. Delmonte, X. Ding, J.J. Garza, B.S. Marler, C. Park,
G.J. Pierce, K.M. Rainey, V.K . Rastogi, S.R. Schulze, N.L.
Trolinder, J.F. We ndel , T.A. Wilkins, T.D. Williams-Coplin,
R.A. Wing, R.J. Wr ight, X. Zhao, L. Zhu, and A.H. Paterson.
2004. A 3347-locus genetic recombination map of sequence-
tagged sites reveals features of genome organization, transmis-
sion and evolution of cotton (Gossypium). Genetics 16 6:389–417.
doi:10.1534/genetics.16 6.1.389
Saddic, L.A., B. Huver mann, S. Bezhani, Y. Su, C.M. Winter, C.S.
Kwon, R.P. Collum, and D. Wa g ner. 2006. The LEAFY tar-
get LMI1 is a meristem identity regulator and acts together w ith
LEAFY to regulate expression of CAULIFLOWER. Develop-
ment 133:1673–1682. doi :10.1242/dev.02 331
Saintenac, C., W. Zhang, A. Salcedo, M.N. Rouse, H.N. Trick, E.
Akhunov, and J. Dubcovsky. 2013. Identication of wheat gene
Sr35 that confers resistance to Ug99 stem rust race group. Science
341:783–786. doi:10.1126/scien ce.12 390 22
Saunders, J.H. 1961. The wild species of Gossypium. Oxford Univ.
Press, London.
Schuelke, M. 2000. An economic method for the uorescent
labeling of PCR fragments. Nat. Biotechnol. 18:233–234.
doi:10.1038/ 72708
Sicard, A., A. Thamm, C. Marona, Y.W. Lee, V. Wah l , J.R. Stinch-
combe, S.L. Wright, C. Kappel, and M. Lenhard. 2014 . Repeated
crop science, vol. 56, may–jun e 2016 www.crops.org 11
evolutionary changes of leaf morpholog y caused by mutations
to a homeobox gene. Curr. Biol. 24:1880 –188 6. doi:10.1016/j.
cub.2014.06.061
SoftGenetics. 2013. GeneMarker genotyping software. Release
2.6.0. SoftGenetics LLC, State College, PA
Song, J., J.M. Bradeen, S.K. Naess, J.A. Raasch, S.M. Wielgus, G.T.
Haberlach, J. Liu, H. Kuang, S. Austin-Phillips, C.R. Buell, J . P.
Helgeson, and J. Jiang. 2003. Gene RB cloned from Solanum
bulbocastanum confers broad spectrum resistance to potato late
blight. Proc. Natl. Acad. Sci. USA 10 0:9128–9133. doi:10.10 73/
pn a s .15335 0110 0
Ulloa, M. 2014. The diploid D genome cottons (Gossypium spp.) of
the New World. In: I. Abdurakhmonov, editor, World cotton
germplasm resources. .InTech, Rijeka, Croatia.
Van Ooijen, J.W. 2006. JoinMap 4, software for the calculation of
genetic linkage maps in exper imental populations. Kyazma B. V.,
Wageningen, Netherlands.
Vlad, D., D. Kierzkowski, M.I. Rast, F. Vuolo, R. Dello Ioio, C.
Galinha, X. Gan, M. Hajheidari, A. Hay, R.S. Smith, P. Huijser,
C.D. Bailey, and M. Tsia ntis. 2014. Leaf shape evolution through
duplication, reg ulatory diversication, and loss of a homeobox
gene. Science 343:780–783. doi:10.112 6/science.124 838 4
Wang , Z., D. Zhang, X. Wang , X. Ta n , H. Guo, and A. H. Pater-
son. 2013. A whole genome DNA marker map for cotton based
on the D-genome sequence of Gossypium raimondii L. G3: Genes,
Genomes, Genet. 3:1759–1767.
Wen del, J.F., C. Brubaker, I. Alvarez, R. Cronn, and J.M. Stewart.
2009. Evolution and natural history of the cotton genus. In: A.H.
Patterson, editor, Genetics and genomics of cotton. Springer,
New York . p. 3–22.
Wen del, J.F., C. Brubaker, and T. Seelanan. 2 010. The origin and
evolution of Gossypium. In: J.M. Stewart, D.M. Oosterhuis,
J.J. Heitholt, and J.R. Mauney, editors, Physiology of cotton.,
Springer, Netherlands. p. 1–18.
Wen del, J .F., and R.C. Cronn. 2003. Polyploidy and the evolu-
tionary histor y of cotton. Adv. A g r on. 78:139–186. doi:10.1016/
S0065-2113(02)78004- 8
Wen del, J.F., and C.E. Grover. 2015. Taxonomy and evolution of
the cotton genus. In: D. Fang and R. Percy, editors, Cotton,
Agronomy Monograph 57. ASA, CSSA, and SSSA, Madison,
WI. doi:10.2134/agronmonogr57.2013.0020
Werner, J.E., T.R. Endo, and B.S. Gil l. 1992. Toward a cytogeneti-
cally based physica l map of the wheat genome. Proc. Natl. Acad.
Sc i . US A 89:11307–11311. doi:10.1073/pnas.89. 23.113 07
White, T.G., and J.E. Endrizzi. 1965. Tests for the association of
marker loci with chromosomes in Gossypium hirsutum by the use
of aneuploids. Genetics 51:605–612.
Xu, Z., R.J. Kohel, G. Song, J. Cho, J. Yu, S. Yu , J. Tomkins,
and J.Z. Yu. 2008. An integrated genetic and physical map of
homoeologous chromosomes 12 and 26 in upland cotton (G. hir-
sutum L.). BMC Genom ics 9:108. doi:10.1186/1471-216 4 -9-108
Yah i a ou i , N., P. Srichumpa, R. Dudler, and B. Keller. 2004.
Genome analysis at dierent ploidy levels allows cloning of the
powdery m ildew resistance gene Pm3b from hexaploid wheat.
Plant J. 37:528–538. doi:10.104 6/j.1365-313X.2 0 03.01977.x
Yan, L., A. Loukoianov, A. Blechl, G. Tran qui l l i , W. Ramakrishna,
P. San-Miguel, J.L. Bennetzen, V. Echenique, and J. Dubcovsky.
2004. The wheat VR N2 gene is a owering repressor down-
regulated by vernalization. Science 303:1640 –164 4. doi:10.1126/
science.1094305
Yan, L., A. Loukoianov, G. Tra nqu i l l i, M. Helguera, T. Fahima,
and J. Dubcovsky. 2003. Positional clon ing of wheat vernaliza-
tion gene V R N1. Proc. Natl. Acad. Sci. USA 10 0:6263–6268.
doi:10.1073/pnas.0 937399100
Zhang, T., Y. Hu, W. Jiang, L. Fang, X. Guan, J. Chen, et al. 2015.
Sequencing of allotetraploid cotton (Gossypium hirsutum L. acc.
TM-1) provides a resource for ber improvement. Nat. Biotech-
nol. 33:531–537. doi:10.1038/nbt .32 07
