Accepted Manuscript

Title: Effects of GhWUS from upland cotton (Gossypium

hirsutum L.) on somatic embryogenesis and shoot regeneration

Authors: Yanqing Xiao, Yanli Chen, Yanpeng Ding, Jie Wu,

Peng Wang, Ya Yu, Xi Wei, Ye Wang, Chaojun Zhang,
Fuguang Li, Xiaoyang Ge

PII: S0168-9452(17)31129-9
DOI: https://doi.org/10.1016/j.plantsci.2018.02.018
Reference: PSL 9762

To appear in: Plant Science

Received date: 29-11-2017

Revised date: 15-1-2018
Accepted date: 18-2-2018

Please cite this article as: Yanqing Xiao, Yanli Chen, Yanpeng Ding, Jie Wu, Peng
Wang, Ya Yu, Xi Wei, Ye Wang, Chaojun Zhang, Fuguang Li, Xiaoyang Ge, Effects of
GhWUS from upland cotton (Gossypium hirsutum L.) on somatic embryogenesis and
shoot regeneration, Plant Science https://doi.org/10.1016/j.plantsci.2018.02.018

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.
Effects of GhWUS from upland cotton (Gossypium hirsutum L.) on somatic
embryogenesis and shoot regeneration

Yanqing Xiaoa, Yanli Chena, Yanpeng Dingb, Jie Wua, Peng Wanga, Ya Yua, Xi Weic,
Ye Wanga, Chaojun Zhanga, Fuguang Lia,* , Xiaoyang Gea,

a
State Key Laboratory of Cotton Biology, Institute of Cotton Research of CAAS, Anyang,
455000 Henan, China

T
b
Hebei Agricultural University, Baoding, 071001 Hebei, China
c
Shenyang Agricultural University, Shenyang 110866, China

R IP

Corresponding Author. E-mail addresses: gexiaoyang613@163.com (Xiaoyang Ge),
aylifug@163.com (Fuguang Li)

SC
Highlights

U
Overexpression of GhWUS in Arabidopsis induces somatic embryo and shoot
N
formation from seedling root tips.
A
 GhWUS overexpression in Arabidopsis promotes phytohormone-independent
shoot and auxin-dependent embyos formation from excised root.
M

 Expression of chimeric GhWUS repressor inhibits cotton embryogenic callus

ED

formation.
E PT
CC
A

 Abbreviations: AtWUS, Arabidopsis thaliana WUSCHEL; WT, wild-type Arabidopsis thaliana plant; Col-0,
Columbia-0; IBA, 3-indole butyric acid; IAA , 3-indole acetic acid; RT-PCR, reverse transcription-PCR;
Abstract
The WUSCHEL (WUS) gene encodes a plant-specific homeodomain-containing
transcriptional regulator, which plays important roles during embryogenesis, as well
as in the formation of shoot and flower meristems. Here, we isolated two homologues
of Arabidopsis thaliana WUS (AtWUS), GhWUS1a_At and GhWUS1b_At, from
upland cotton (Gossypium hirsutum). Domain analysis suggested that the two putative
GhWUS proteins contained a highly conserved DNA-binding HOX domain and a

T
WUS-box. Expression profile analysis showed that GhWUSs were predominantly

IP
expressed during the embryoid stage. Ectopic expression of GhWUSs in Arabidopsis

R
could induce somatic embryo and shoot formation from seedling root tips.

SC
Furthermore, in the absence of exogenous hormone, overexpression of GhWUSs in
Arabidopsis could promote shoot regeneration from excised roots, and in the presence

U
of exogenous auxin, excised roots expressing GhWUS could be induced to produce
N
somatic embryo. In addition, expression of the chimeric GhWUS repressor in cotton
A
callus inhibited embryogenic callus formation. Our results show that GhWUS is an
important regulator of somatic embryogenesis and shoot regeneration.
M

Keywords: Gossypium hirsutum; Somatic embryogenesis; Shoot regeneration;

ED

GhWUS

1. Introduction
PT

The WUSCHEL (WUS) protein is a bifunctional homeodomain transcription factor,

which plays a vital role during somatic embryogenesis and shoot regeneration,
E

possibly by regulating the expression of specific target genes. In addition to highly

CC

conserved homeobox domain, WUS contains three functional domains at the

conserved C-terminal region; these include an acidic domain, a WUS-box
A

(TLPLFPMH), and an EAR-like motif, among which the WUS-box domain, a highly
conserved functional domain, has a significant role in the dual biological functions of
WUS and is remarkably essential for the maintenance of stem cell identity [1]. WUS
is involved in the maintenance of the stem cell pool possibly by specifying the stem
cell fate. Previous research has shown that WUS is essential for stem cell formation
and maintenance of the shoot apical meristems (SAM) [2]. In the SAM, the
expression of WUS is limited to a few cells of the organizing centre (OC), which is
regulated by the CLAVATA (CLV) signalling pathway [3,4]. The WUS protein can
move from the cells of the OC into the adjacent cells located in the central zone (CZ),
where it directly activates the transcription of a negative regulator, CLAVATA3 (CLV3),
by binding to its regulatory regions [5]. CLV3 then migrates into the extracellular
space to bind to CLAVATA1 (CLV1), which has been proven to inhibit the

T
transcription of WUS as CLV1 is activated with related receptor kinases; this is

IP
considered to be the WUS–CLV feedback system necessary for the maintenance of

R
the stem cell pool and SAM [5,6]. Both somatic embryogenesis and shoot

SC
regeneration require the establishment of SAM; therefore, WUS is essential for
somatic embryogenesis and shoot regeneration.

U
Plant somatic embryogenesis is a developmental process similar to that of zygotic
N
embryos. Developmentally and morphologically, both somatic embryogenesis and
A
zygotic embryo development share the same stages from globular-shaped embryo to
torpedo-shaped embryo and to cotyledonal embryo in dicots [7]. Somatic
M

embryogenesis may involve three developmental phases, namely the embryonic

ED

induction phase, embryonic phase, and development phase [8], each of which is a
very complicated process in the interaction and regulation of various genes and
signalling pathways. Genes, such as LEAFY COTYLEDON (LEC) [9], FUSCA3
PT

(FUS3) [10], ABAINSENSITIVE 3 (ABI3) [11], SOMATIC EMBRYOGENESIS

RECEPTOR KINASE 1 (SERK1) [12], BABY BOOM (BBM) [13],
E

AGAMOUS-LIKE15 (AGL15) [14], and WUS [15], involved in zygotic embryo

CC

development, also play a vital role in somatic embryogenesis. During somatic

embryogenesis in many plant species, embryogenic calli are first induced from
A

explants on a medium containing high levels of auxin and are then transferred into
auxin-free medium to initiate the development process, similar to that of zygotic
embryos [16]. The auxin-free medium could lead to the re-establishment of auxin
gradients in the embryogenic callus, which induces the local expression of WUS gene,
in the area of low auxin levels, marking the location of future shoot meristem
formation [16,17]. In addition, WUS can also directly promote cell fate transition from
somatic cells to embryogenic cells and maintain the embryonic cell identity in
Arabidopsis [15].
De novo shoot organogenesis refers to the process in which plant explants form
adventitious shoots under proper culture conditions, which is influenced by exogenous
phytohormones. In general, de novo shoot organogenesis consists of three consecutive
phases [18,19]—in the first phase, the cells dedifferentiate to gain competence for

T
shoot organogenesis, in the second phase, cell determination for specific organ

IP
formation occurs in response to exogenous hormone, and in the third phase, organ

R
morphogenesis proceeds without exogenous phytohormone induction. It is reported

SC
that a high cytokinin/auxin ratio induces shoot organogenesis, whereas a low ratio
induces root development, and a mass of proliferating cells (callus) could be induced

U
from explants when both auxin and cytokinin were at high levels [19,20]. In a
N
previous study, plant explants were observed to form adventitious shoot in vitro under
A
the effect of exogenous cytokinin and auxin [21]. Generally, it takes two steps to
regenerate shoots in Arabidopsis [20,22]; in the first step, explants are induced to
M

form a callus on auxin-rich callus induction medium (CIM), and in the second step,
ED

shoot formation occurs from the callus on cytokinin-rich shoot induction medium
(SIM). The stage of preincubation on CIM, in which the explants acquire the ability to
form shoot, was reported to be necessary for shoot regeneration [23,24]. The
PT

expression of the NAC transcription factors, CUP-SHAPED COTYLEDON 1 (CUC1)

and CUC2, was induced in CIM, which mediated the initiation of shoot meristem and
E

promoted adventitious shoot formation through the activation of STM expression

CC

[16,25]. Several studies have shown that the overexpression of either CUC1 or CUC2
could promote shoot formation and the cuc1 cuc2 double mutant is defective in shoot
A

regeneration [25,26]. After calli are transferred to the SIM, a gradient of auxin and
cytokinin is induced. In the region with high cytokinin levels, the up-regulated
expression of the WUS gene was reported to be mediated by cytokinin response
regulatory factors type-B ARABIDOPSIS RESPONSE REGULATORs (ARRs)
[22,27], which promoted cell fate transition from callus cells to OC, and was required
for specification of the shoot stem cell niche as well as for the establishment of the
shoot meristem. At the same time, the expression of CUC2 was restricted to the region
of low cytokinin levels, in which the CUC2-expression cells further developed into
shoot promeristem [16,22]. Thereafter, the local upregulation of key shoot regulatory
genes, such as PIN-FORMED1 (PIN1), SHOOT-MERISTEMLESS (STM), and CLV3,
promoted the formation of a developed shoot meristem [22].
In general, WUS can promote both shoot regeneration and somatic embryogenesis

T
through its participation in the maintenance of SAM or in the transition from the

IP
vegetative to embryonic stage. In this study, two homologous WUS genes,

R
GhWUS1a_At and GhWUS1b_At, were cloned from upland cotton and transformed

SC
into Arabidopsis. Our results show that GhWUSs are key regulators during somatic
embryogenesis and shoot regeneration.
2. Materials and methods
U
N
2.1. Plant materials and growth conditions
A
In this study, we used wild type Arabidopsis thaliana Columbia ecotype (Col-0)
plants and the pER8::eGFP-GhWUS1a_At and pER8::eGFP-GhWUS1b_At
M

transgenic plants. The plants were grown in a greenhouse at 22 °C under a 16 h light/8

ED

h dark photoperiod. The cotton materials used in the study were from Gossypium
hirsutum ‘CCRI24’ plant. For tissure culture of CCRI24, the hypocotyl segments were
culture on callus induction medium to induce calli and then the calli were transferred
PT

onto the embryogenic callus induction medium for inducing embryogenic callus
which were transferred to somatic embryo induction medium for somatic embryo
E

induction [28]. For expression analysis of GhWUSs, samples were collected from the
CC

non-embryogenic callus (NEC), embryogenic callus (EC), globular-stage embryos,

torpedo-shaped embryos, cotyledon-stage embryos, roots, stems, and leaves of
A

CCRI24.
2.2. Gene cloning, vector construction, and plant transformation
Total RNA was isolated from somatic embryoids of G. hirsutum ‘CCRI24’ plants
using a Plant RNAprep Pure Kit (Tiangen, Beijing, China). The cDNA was
synthesized using a PrimeScript™ RT Reagent Kit with gDNA Eraser (Perfect Real
Time; Takara, Dalian, China). The primers (Supplementary Table S1) for
GhWUS1a_At and GhWUS1b_At were designed using Premier Primer 5 software.
PrimeSTAR polymerase (Takara, Dalian, China) was used to amplify the full-length
coding sequence of GhWUS1a_At and GhWUS1b_At using the cDNA templates. For
the construction of two pER8::eGFP-GhWUS vectors, cDNA fragments encoding the
full length of GhWUS1a_At and GhWUS1b_At were inserted respectively into the
XVE-inducible expression vector pER8 with the hygromycin-selectable marker [29].

T
All transgenic Arabidopsis plants were obtained by Agrobacterium-mediated

IP
transformation of wild-type Col-0, as described previously [30]. The

R
pER8::eGFP-GhWUS transgenic plants were selected on hygromycin. For

SC
construction of two 35S::GhWUS-SRDX vectors, two GhWUS-SRDX fusion
fragments were inserted, respectively, into the pCAMBIA2300 vector with

U
kanamycin-selectable marker, which were transformed into hypocotyl explants of
N
‘CCRI24’ through the Agrobacterium-mediated transformation, as described
A
previously [28]. The empty vector of pCAMBIA2300 was used as a control (CK) in
the transformation experiments. The differentiation rate was calculated by measuring
M

the percentage of embryogenic callus derived from all callus.

ED

2.3. Domains analysis and sequence alignment

DNAMAN software (version 6.0) was used to perform multiple sequence
alignment. The conserved domains were analysed using the Simple Modular
PT

Architecture Research Tool (http://smart.embl-heidelberg.de/).

2.4. Arabidopsis thaliana tissue culture
E

Sterilized T3 homozygous seeds of the pER8::eGFP-GhWUS transgenic plants

CC

were cultured on solid Murashige and Skoog’s (MS) medium (1× MS salts, 2%
sucrose, 0.3% Gelrite gellan gum, pH 5.7) supplemented with 10 μM of the inducer,
A

17-β-estradiol, under a 16 h light/8 h dark photoperiod at 22 °C. After induction for 5,

7, and 9 days, samples were collected for qRT-PCR analysis and WT plants were used
as a control. The frequency of seedling phenotypes was calculated.
For shoot regeneration, the whole experiment was done as described previously
[20]. Sterilized seeds of WT and the pER8::eGFP-GhWUS transgenic plants were
germinated and grown on a non-inductive MS medium. After cultivation for 2 weeks,
the excised root segments (0.5–1 cm) were cut and incubated on callus induction
medium (CIM; 1× Gamborg’s B5 salts, 3% sucrose, 0.5 g/L MES, 0.05 mg/L kinetin,
0.5 mg/L 2,4-D, and 0.3% Gelrite gellan gum, pH 5.7) under continuous darkness at
22 °C. After cultivation for 7 days, the explants were transferred onto the SIM with 10
μM 17-β-estradiol (1× MS salts, 1% sucrose, 0.5 g/L MES, 2 mg/L zeatin, 1 mg/L
d-biotin, 0.4 mg/L IBA, 0.3% Gelrite gellan gum, pH 5.7) and incubated under

T
continuous light at 22 °C. Shoot regeneration frequency (%) was calculated at 0, 6, 8,

IP
10, 12, and 14 days after the explants were transferred onto SIM.

R
For root analysis, root explants were cut from seedlings of WT and the

SC
pER8::eGFP-GhWUS transgenic plants were germinated and grown on a
non-inductive MS medium for two weeks. Thereafter, the excised roots were

U
incubated on MS medium supplemented with 10 μM 17-β-estradiol and/or specified
N
phytohormone under continuous light at 22 °C.
A
2.5. Semi-quantitative and quantitative RT-PCR
The samples were collected as described above. The sequences of all the primers
M

are given in Supplementary Table S1. For semi-quantitative RT-PCR, 2× Taq PCR
ED

Mastermix (BIOTEKE, Beijing, China) was used to amplify the chimeric

eGFP-GhWUSs sequence and AtAPT1 (AT1G27450; internal control). The number of
PCR cycles was 20 for eGFP-GhWUSs and 25 for AtAPT1. For quantitative RT-PCR
PT

(qRT-PCR), endogenous G. hirsutum histone3 (GhHis3, Gh_D03G0370) and A.

thaliana actin2 (AtACT2, AT3G18780) genes were used as internal standards for G.
E

Hirsutum and A. thaliana, respectively. The quantitative RT-PCR was performed as

CC

described previously [31].

3. Results
A

3.1. Cloning and sequence analysis of GhWUS homologues

We used AtWUS amino acid sequence as a query sequence to search the G.
hirsutum protein database employing blastp program (https://www.cottongen.org/),
and obtained four homologous GhWUS amino acid sequences, namely GhWUS1a_At,
GhWUS1b_At, GhWUS1a_Dt, and GhWUS1b_Dt, which were the same as obtained
by Yang et al [32]. Protein sequence alignment showed that GhWUS1a_At shared
95.71% identity with GhWUS1a_Dt and GhWUS1b_At shared 99.31% identity with
GhWUS1b_Dt (Supplementary Fig. S1), which suggested that these two sets of genes
might be functionally redundant. Therefore, we cloned GhWUS1a_At and
GhWUS1b_At, from At genomes of the tetraploid G. hirsutum for our study. The
deduced protein sequences of GhWUS1a_At and GhWUS1b_At shared only 36.22%
and 40% identity with AtWUS, respectively. However, we found that the HOX

T
domains of those two sequences were highly similar with that of AtWUS (Fig. 1). The

IP
WUS-box (TLPLFPMH), which is the functional domain, was also the same as that of

R
AtWUS (Fig. 1). Therefore, we deduced that GhWUSs might serve a function similar

SC
to that of AtWUS.
3.2. Expression analysis of GhWUS homologues in G. hirsutum

U
We examined GhWUSs expression patterns by RT-PCR analysis in the samples of
N
non-embryogenic callus (NEC), embryogenic callus (EC), globular-stage embryos,
A
torpedo-shaped embryos, cotyledon-stage embryos, roots, stems, and leaves harvested
from upland cotton. As shown in Fig. 2, the expression levels of GhWUSs were low in
M

roots, stems, and leaves. During the various stages of somatic embryogenesis, higher
ED

expression levels of GhWUSs were found in the callus and embryoid stages especially
in the stage of torpedo-shaped embryos, but the GhWUSs had low abundant
expression in the EC, similar to that in the roots, stems, and leaves. These results
PT

showed that GhWUSs might play a crucial part in the transition from callus to EC and
embryoid development.
E

3.3. Effects of GhWUS overexpression on root development in Arabidopsis

CC

thaliana
To confirm the functions of the two GhWUS genes, we constructed two inducible
A

expression vectors, pER8::eGFP-GhWUS1a_At and pER8::eGFP-GhWUS1b_At, and

transformed them into wild-type Arabidopsis plants. Considering the differences in
the expression levels, we chose two pER8-GhWUS1a_At transgenic lines, A12-20 and
A12-27, with significantly different expression levels, as well as two
pER8-GhWUS1b_At transgenic lines, A10-1 and A10-4, for the subsequent study. The
transgenic plants were germinated and grown on MS medium without 17-β-estradiol
for 8 days, and were then transferred to MS medium with or without 10 μM
17-β-estradiol for 24 h to detect the expression. As shown in Fig. 3A, the results of
qRT-PCR and semi-quantitative RT-PCR confirmed that the expression was induced
by 17-β-estradiol in the pER8-GhWUSs transgenic plants. Line A12-27 and Line
A10-1 were strong expressing lines compared to Line A12-20 and Line A10-4. These
transgenic lines were sown on solid MS medium supplemented with 10 μM inducer,

T
17-β-estradiol. Eight days after the induction, the root tip of some seedlings

IP
developed into different tissues, which either substituted for the root tip or were kept

R
back while the root tip continued to grow (Fig. 3B, C, D, and E). These tissues were

SC
divided into four phenotypes, namely leaf-like organs (Fig. 3B), embryo-like
structures (Fig. 3C), callus-like structures (Fig. 3D), and normal root organs (Fig. 3E).

U
We found that the transgenic Arabidopsis lines with different expression levels
N
showed different frequencies of the four phenotypes. We counted the frequency of
A
four phenotypes of Line A12-20 (n = 336), Line A12-27 (n = 318), Line A10-1 (n =
307) and Line A10-4 (n = 331) (Fig. 3F) and found that the strong
M

GhWUS–expressing lines (Fig. 3A), Line A12-27 and Line A10-1, had higher
ED

frequency of leaf-like and embryo-like phenotypes compared to that of the relatively

weak GhWUS–expressing lines, Line A12-20 and Line A10-4 (Fig. 3A). The Leaf-like
phenotypes accounted for 35.85% of the Line A12-27 plants and 13.35% of the Line
PT

A10-1 plants. The primary root of Line A12-27 (27.67%) and Line A10-1 (43%)
plants had developed into embryo-like structures. In contrast, only 4.17% of Line
E

A12-20 and 3.93% of Line A10-4 exhibited the embryo-like phenotype, and 1.49% of
CC

Line A12-20 and 8.76% of Line A10-4 had leaf-like phenotypes. Furthermore,
qRT-PCR analysis suggested that the expression of the LEC1, FUS3, and ABI3, which
A

are key regulators in somatic embryogenesis, were up-regulated significantly in the

pER8-GhWUSs transgenic plants (Fig. 3G). The expression of CLV3, which is a shoot
meristem induction marker gene, was also enhanced in the transgenic plants (Fig. 3H).
Moreover, we found the expression of LEC1, FUS3, ABI3, and CLV3 in the strong
GhWUS-expressing lines, Line A10-1 and Line A12-27, to be higher than that in the
relatively weak GhWUS–expressing lines, Line A10-4 and A12-20 (Fig. 3G and H).
All these results showed that ectopic expression of GhWUSs can promote somatic
embryogenesis and shoot regeneration in Arabidopsis, and that the expression of
genes involved in somatic embryogenesis and shoot regeneration is significantly
induced in the GhWUS overexpression transgenic plants.
3.4. Ectopic expression of GhWUSs promotes de novo shoot organogenesis from
Arabidopsis root explants

T
The strong GhWUS1a_At–expressing line (Line A12-27) and the strong

IP
GhWUS1b_At–expressing line (Line A10-1) were selected to further understand the

R
role of GhWUSs in shoot regeneration. We tested the capacity of root explants from

SC
the pER8-GhWUSs transgenic plants to regenerate adventitious shoots. The shoot
regeneration experiment was done as described previously [20]. The root explants

U
were induced to form calli on CIM and were then transferred onto SIM supplemented
N
with 10 μM 17-β-estradiol to form shoots. As shown in Fig. 4, both Line A12-27 and
A
Line A10-1 had higher shoot regeneration frequency compared to that of wild-type
Columbia-0 (Col-0). At 6 days on inductive SIM (SIM6), the pER8::eGFP-GhWUS
M

root explants began to form shoots, whereas the Col-0 root explants formed shoots at
ED

SIM8. In addition, almost all the pER8::eGFP-GhWUS root explants formed shoots
and could produce more shoots at SIM12 than did the explants from wild-type. These
results indicate that overexpressing GhWUS in Arabidopsis could promote de novo
PT

shoot organogenesis.
3.5. GhWUS overexpression induces shoot formation from the excised roots of
E

Arabidopsis independent of the exogenous hormones

CC

Previous studies had shown that shoot regeneration required the stage of
preincubation on CIM which was necessary for WUS expression in respond to SIM
A

induction [23,24]. Moreover, activation of WUS was stimulated by SIM-direct in

drm1 drm2 cmt3 (ddc) triple mutant which exhibited increased capacity to regenerate
shoots even when roots were incubated directly on SIM, but required preincubation
on CIM in WT [33]. Here, We tested whether GhWUS overexpression promotes shoot
regeneration without the stage of preincubation on CIM. The root explants were
directly incubated on SIM supplemented with 10 μM 17-β-estradiol. After incubation
for 23 days, a mass of pER8::eGFP-GhWUSs root explants formed shoots whereas
there were fewer shoots formed from the wild-type Col-0 root explants
(Supplementary Fig. S2), which revealed that preincubation on CIM is not required
for shoot formation from the GhWUS overexpression root explants. A high level of
cytokinin and a low level of auxin in SIM plays a very important role in shoot
formation, and previous research has also shown that shoot meristems initiated in

T
regions of high cytokinin and low auxin [22]. Next, we tested the effects of these

IP
hormones on the GhWUS overexpressing root explants. The excised roots of the

R
2-week-old seedlings of WT, Line A12-27, and Line A10-1 were incubated on

SC
inductive MS medium supplemented with or without hormones (auxin or cytokinin).
As shown in Fig. 5B and C, the GhWUS overexpressing root explants could produce

U
shoots at a low frequency when put on the inductive MS medium containing auxin (2
N
μM IAA). Accidentally, we found that fewer embryos were induced from Line
A
A12-27 excised root on inductive MS medium containing auxin (Fig. 5C, D and E);
however, this phenomenon was not observed in Line A10-1 for the present. In
M

addition, on the inductive MS medium containing cytokinin (2 mg/L zeatin) (Fig. 5G

ED

and H), it was expected that these root explants would produce a mass of shoots.
Surprisingly, we also found the pER8::eGFP-GhWUS roots were induced to produce
shoots on the inductive MS medium without exogenous hormone (Fig. 5J and K). In
PT

contrast, the wild-type Col-0 root explants did not form shoots on all the above MS
media (Fig. 5A, F and I). These results indicated that ectopic expression of GhWUS in
E

root promoted adventitious shoot formation in Arabidopsis independent of the

CC

exogenous hormone. To further clarify the effects of GhWUS on shoot regeneration,

we detected the expression levels of GhWUS, CLV3, and CUC2 at different inducing
A

period after the excised roots were incubated on inductive MS medium without
exogenous hormone. The GhWUS transcript levels in the transgenic lines were
significantly increased after the induction, and these levels were maintained at least
for 6 days (Fig. 5L). The expression levels of CLV3 and CUC2, which are the shoot
regeneration-related genes, were obviously enhanced with the increase in GhWUS
transcription (Fig. 5M and N). These results showed that ectopic GhWUS possibly
promoted the CLV3 and CUC2 transcription in roots to induce shoots formation from
them.
3.6. Expression of the chimeric GhWUS repressor inhibits cotton embryogenic
callus formation
The overexpression of AtWUS can promote vegetative-to-embryonic transition in
Arabidopsis [15] and the formation of EC in G. hirsutum CRI12 which is a

T
recalcitrant cotton cultivar for plant regeneration via somatic embryogenesis [28].

IP
Thus, we analysed the effects of GhWUS on EC formation in G. hirsutum CRI24, a

R
cotton cultivar easy for somatic embryogenesis. A transcription factor fused to the

SC
SRDX repression domain can act as a dominant repressor which specifically suppress
the expression of target genes, even in the presence of redundant transcription factors

U
[34]. To analyse the role of GhWUS in cotton somatic embryogenesis, we constructed
N
two 35S::GhWUSs-SRDX vectors and an empty vector (CK) to transform the
A
hypocotyl segments of CRI24. The hypocotyl segments were cultured on callus
induction medium containing kanamycin and cefotaxime for 2 months, and then the
M

calli derived from these hypocotyl segments were transferred onto the EC induction
ED

medium (EIM) containing kanamycin and cefotaxime for inducing EC. After
incubation for 90 days on EIM, most of calli transfromed with 35S::GhWUSs-SRDX
vectors did not differentiate and kept on the status of calli even cultured for another 2
PT

months (Fig. 6A) while most CK calli differentiated into EC (Fig. 6B). We calculated
the EC differentiation rate (the embryogenic callus percentage of all callus) at 30, 60,
E

and 90 days after incubation on EIM. Both 35S::GhWUS1a_At-SRDX and

CC

35S::GhWUS1b_At-SRDX transformants presented a very low EC differentiation rate

compared to that of the empty vector transformants (Fig. 6C), indicating constitutive
A

expression of GhWUSs-SRDX signifcantly decrease the formation of EC. These

results indicated that the inhibition on target genes of GhWUS suppressed the
transition from callus to EC.
4. Discussion
WUSCHEL protein is an evolutionarily conserved plant-specific homeobox
transcription factor [35]. The expression of AtWUS is detected during both
embryogenic and postembryonic (shoot and floral meristems) development stages [3].
The main function of AtWUS is to specify stem cell fate and to keep stem cells in an
undifferentiated state in meristems, which is required for embryogenesis as well as for
shoot and floral meristem integrity [2,3]. The function of WUS has been well studied
in Arabidopsis. However, research on the homologues of AtWUS in cotton is missing.
In this study, two homologues of AtWUS, GhWUS1a_At and GhWUS1b_At, were

T
isolated from upland cotton for the first time. Sequence analysis suggested that the

IP
two putative GhWUS protein sequences contained a highly conserved HOX domain

R
and a WUS-box (Fig. 1), which indicated that they belong to the WUS

SC
homeobox-containing (WOX) protein family. In addition, the functional domain of
GhWUS, WUS-box (TLPLFPMH), is the same as that of AtWUS, which supports the

U
notion that GhWUSs may exhibit a function similar to that of AtWUS.
N
WUS is deemed to play an important role during embryogenesis. In Arabidopsis,
A
WUS is required for somatic embryo formation [17]. Overexpression of AtWUS has
been found to promote the transition from vegetative stage to embryonic stage in
M

Arabidopsis [15], as well as the formation of the embryogenic callus in cotton [28]. In
ED

the present study, the expression profile analysis showed that GhWUSs were
predominantly expressed during the embryoid stage, especially at the stage of
torpedo-shape embryos (Fig. 2). The expression of the chimeric GhWUS repressor in
PT

cotton could significantly suppress the transition from callus to embryogenic callus
(Fig. 6). Therefore, GhWUSs indeed have a crucial role during cotton somatic
E

embryogenesis. To analyse the conserved functions of GhWUSs during somatic

CC

embryogenesis, the pER8::eGFP-GhWUSs vectors were constructed and transformed

into the wild-type Arabidopsis plants. Ectopic expression of GhWUS in Arabidopsis
A

could induce somatic embryogenesis in roots (Fig. 3C), which was similar to the
result of a previous study wherein it was observed that the overexpression of AtWUS
in roots promoted somatic embryogenesis [15]. The root tips of AtWUS
overexpressing Arabidopsis plants developed into embryo-like structures at a very
low frequency without auxin, but the AtWUS overexpression in the presence of auxin
could induce somatic embryogenesis in virtually all the roots [36]. Unexpectedly, the
overexpression of GhWUSs in Arabidopsis roots in the absence of auxin could induce
somatic embryogenesis at a higher frequency (Fig. 3F). The roots excised from Line
A12-27 were induced to form embryos on inductive MS medium containing auxin
(Fig. 5C, D and E). Furthermore, the embryo marker genes, LEC1, FUS3, and ABI3,
were remarkably increased in the GhWUS-overexpressing roots compared to that in
the wild-type Col-0 roots (Fig. 3G). These results proved that GhWUSs have

T
important function in promoting somatic embryogenesis and may have more

IP
pronounced effects on somatic embryogenesis comparing to AtWUS.

R
Many studies have indicated that WUS has a very important role in shoot

SC
regeneration. The wus mutant was defective in shoot meristem development [2,3] and
failed to regenerate shoots from hypocotyl explants during de novo shoot regeneration

U
[37]. The overexpression of AtWUS in the arr1 12 double mutant could restore the
N
shoot regeneration defects [27]. In our research, the overexpression of GhWUSs in
A
Arabidopsis promoted de novo shoot organogenesis from callus on SIM (Fig. 4), and
could also produce shoots from the root explants without the stage of preincubation
M

on CIM (Supplementary Fig. S2). Besides, ectopic expression of GhWUSs in the

ED

Arabidopsis root could also induce shoot formation (Fig. 3B), which was similar to
observation made in Arabidopsis [36]. A recent article indicated that when the roots
excided from the 35S::WUS-GR plants were cultured on DEX and auxin, no shoots or
PT

embryos were formed [38]. However, we found that the excised

pER8::eGFP-GhWUS root explants were not only induced to produce shoots on
E

inductive MS medium in the presence of cytokinin or auxin (Fig. 5B, C, G, and H),
CC

but could also directly regenerate shoots on hormone-free MS medium supplemented

with inducer (Fig. 5J and K), indicating that the activation of GhWUS in root is
A

sufficient to induce shoot meristems. In addition, there is frequency difference

between the excised pER8::eGFP-GhWUS roots on the inductive medium with and
without hormones from our work, more shoots were produced from excised roots
cultute on cytokinin. This is probably because the up-regulated expression of
endogenous WUS, mediated by cytokinin response regulatory factors type-B ARRs
with high cytokinin levels [22,27], promoted more shoot formation. The expression of
CLV3 and CUC2, which are the shoot regeneration related genes, was obviously
enhanced with the increase of GhWUS transcription in the excised GhWUS-expression
roots (Fig. 5M and N). These results showed that ectopic GhWUS possibly activated
CLV3 and CUC2 in excised roots to promote shoot formation, and GhWUS might
exhibit different function in promoting Arabidopsis shoot regeneration comparing to
AtWUS.

T
In conclusion, we isolated two GhWUS genes from upland cotton to assess their

IP
function in shoot formation and somatic embryogenesis. We found that

R
overexpression of GhWUS in Arabidopsis roots could induce shoot formation and

SC
somatic embryogenesis. In excised roots, GhWUS overexpression promoted CLV3 and
CUC2 transcription to induce shoot formation independent of the exogenous hormone.

U
Besides, expression of the chimeric GhWUS repressor significantly blocked the
N
formation of cotton embryogenic callus, suggesting that GhWUS was a key regulator
A
during cotton somatic embryogenesis. Identification of GhWUS provides a candidate
gene for improving shoot formation and somatic embryogenesis in crop species.
M

However, further studies are required to decipher the downstream genes of GhWUS
ED

and to determine whether GhWUS promotes de novo shoot regeneration by directly

activating CLV3 and CUC2.
PT

Conflicts of interest
The authors declare that they have no conflict of interest.
E
CC

Acknowledgments
This research was funded by the National Key R&D Program of China
A

(2016YFD0100505).
References
[1] M. Ikeda, N. Mitsuda, M. Ohme-Takagi, Arabidopsis WUSCHEL is a bifunctional transcription
factor that acts as a repressor in stem cell regulation and as an activator in floral patterning, Plant Cell
21 (2009) 3493-3505.
[2] T. Laux, K.F. Mayer, J. Berger, G. Jurgens, The WUSCHEL gene is required for shoot and floral
meristem integrity in Arabidopsis, Development 122 (1996) 87-96.
[3] K.F. Mayer, H. Schoof, A. Haecker, M. Lenhard, G. Jurgens, T. Laux, Role of WUSCHEL in
regulating stem cell fate in the Arabidopsis shoot meristem, Cell 95 (1998) 805-815.
[4] U. Brand, J.C. Fletcher, M. Hobe, E.M. Meyerowitz, R. Simon, Dependence of stem cell fate in

T
Arabidopsis on a feedback loop regulated by CLV3 activity, Science 289 (2000) 617-619.
[5] R.K. Yadav, M. Perales, J. Gruel, T. Girke, H. Jonsson, G.V. Reddy, WUSCHEL protein movement

IP
mediates stem cell homeostasis in the Arabidopsis shoot apex, Genes Dev. 25 (2011) 2025-2030.
[6] A. Kinoshita, S. Betsuyaku, Y. Osakabe, S. Mizuno, S. Nagawa, Y. Stahl, R. Simon, K.

R
Yamaguchi-Shinozaki, H. Fukuda, S. Sawa, RPK2 is an essential receptor-like kinase that transmits the
CLV3 signal in Arabidopsis, Development 137 (2010) 3911-3920.

SC
[7] J.L. Zimmerman, Somatic embryogenesis: A Model for early development in higher plants, Plant
Cell 5 (1993) 1411-1423.
[8] M. Elhiti, C. Stasolla, A.M. Wang, Molecular regulation of plant somatic embryogenesis, In Vitro
Cell. Dev. Biol.-Plant 49 (2013) 631-642.
U
N
[9] M.D. Gaj, S. Zhang, J.J. Harada, P.G. Lemaux, Leafy cotyledon genes are essential for induction of
somatic embryogenesis of Arabidopsis, Planta 222 (2005) 977-988.
A
[10] H. Luerssen, V. Kirik, P. Herrmann, S. Misera, FUSCA3 encodes a protein with a conserved
VP1/ABI3-like B3 domain which is of functional importance for the regulation of seed maturation in
M

Arabidopsis thaliana, Plant J. 15 (1998) 755-764.

[11] H. Shiota, R. Satoh, K. Watabe, H. Harada, H. Kamada, C-ABI3, the carrot homologue of the
Arabidopsis ABI3, is expressed during both zygotic and somatic embryogenesis and functions in the
ED

regulation of embryo-specific ABA-inducible genes, Plant Cell Physiol. 39 (1998) 1184-1193.

[12] E.D. Schmidt, F. Guzzo, M.A. Toonen, S.C. deVries, A leucine-rich repeat containing receptor-like
kinase marks somatic plant cells competent to form embryos, Development 124 (1997) 2049-2062.
PT

[13] K. Boutilier, R. Offringa, V.K. Sharma, H. Kieft, T. Ouellet, L. Zhang, J. Hattori, C.M. Liu, A.A.
van Lammeren, B.L. Miki, J.B. Custers, M.M. van Lookeren Campagne, Ectopic expression of BABY
BOOM triggers a conversion from vegetative to embryonic growth, Plant Cell 14 (2002) 1737-1749.
E

[14] E.W. Harding, W. Tang, K.W. Nichols, D.E. Fernandez, S.E. Perry, Expression and maintenance of
CC

embryogenic potential is enhanced through constitutive expression of AGAMOUS-Like 15, Plant

Physiol. 133 (2003) 653-663.
[15] J.R. Zuo, Q.W. Niu, G. Frugis, N.H. Chua, The WUSCHEL gene promotes
A

vegetative-to-embryonic transition in Arabidopsis, Plant J. 30 (2002) 349-359.

[16] M. Ikeuchi, Y. Ogawa, A. Iwase, K. Sugimoto, Plant regeneration: cellular origins and molecular
mechanisms, Development 143 (2016) 1442-1451.
[17] Y.H. Su, X.Y. Zhao, Y.B. Liu, C.L. Zhang, S.D. O’Neill, X.S. Zhang, Auxin-induced WUS
expression is essential for embryonic stem cell renewal during somatic embryogenesis in Arabidopsis,
Plant J. 59 (2009) 448-460.
[18] M. Sugiyama, Organogenesis in vitro, Curr. Opin. Plant Biol. 2 (1999) 61-64.
[19] J. Duclercq, B. Sangwan-Norreel, M. Catterou, R.S. Sangwan, De novo shoot organogenesis: from
art to science, Trends Plant Sci. 16 (2011) 597-606.
[20] K. Sugimoto, E.M. Meyerowitz, Regeneration in Arabidopsis tissue culture, Methods Mol. Biol.
959 (2013) 265-275.
[21] F. Skoog, C.O. Miller, Chemical regulation of growth and organ formation in plant tissues cultured
in vitro, Symp. Soc. Exp. Biol. 11 (1957) 118-130.
[22] S.P. Gordon, M.G. Heisler, G.V. Reddy, C. Ohno, P. Das, E.M. Meyerowitz, Pattern formation
during de novo assembly of the Arabidopsis shoot meristem, Development 134 (2007) 3539-3548.
[23] P. Che, S. Lall, S.H. Howell, Developmental steps in acquiring competence for shoot development
in Arabidopsis tissue culture, Planta 226 (2007) 1183-1194.
[24] D. Valvekens, M.V. Montagu, M.V. Lijsebettens, Agrobacterium tumefaciens-mediated

T
transformation of Arabidopsis thaliana root explants by using kanamycin selection, Proc. Natl. Acad.

IP
Sci. 85 (1988) 5536-5540.
[25] Y. Daimon, K. Takabe, M. Tasaka, The CUP-SHAPED COTYLEDON genes promote adventitious

R
shoot formation on calli, Plant Cell Physiol. 44 (2003) 113-121.
[26] A. Kareem, K. Durgaprasad, K. Sugimoto, Y. Du, A.J. Pulianmackal, Z.B. Trivedi, P.V. Abhayadev,

SC
V. Pinon, E.M. Meyerowitz, B. Scheres, K. Prasad, PLETHORA genes control regeneration by a
two-step mechanism, Curr. Biol. 25 (2015) 1017-1030.
[27] W.J. Meng, Z.J. Cheng, Y.L. Sang, M.M. Zhang, X.F. Rong, Z.W. Wang, Y.Y. Tang, X.S. Zhang,

U
Type-B ARABIDOPSIS RESPONSE REGULATORs specify the shoot stem cell niche by dual
regulation of WUSCHEL, Plant Cell 29 (2017) 1357-1372.
N
[28] W. Zheng, X. Zhang, Z. Yang, J. Wu, F. Li, L. Duan, C. Liu, L. Lu, C. Zhang, F. Li, AtWuschel
promotes formation of the embryogenic callus in Gossypium hirsutum, PLoS One 9 (2014).
A
[29] J.R. Zuo, Q.W. Niu, N.H. Chua, An estrogen receptor-based transactivator XVE mediates highly
M

inducible gene expression in transgenic plant, Plant J. 24 (2000) 265-273.

[30] S.J. Clough, A.F. Bent, Floral dip: a simplified method for Agrobacterium-mediated
transformation of Arabidopsis thaliana, Plant J. 16 (1998) 735-743.
ED

[31] Q. Gong, Z. Yang, X. Wang, H.I. Butt, E. Chen, S. He, C. Zhang, X. Zhang, F. Li, Salicylic
acid-related cotton (Gossypium arboreum) ribosomal protein GaRPL18 contributes to resistance to
Verticillium dahliae, BMC Plant Biol. 17 (2017) 59.
PT

[32] Z.E. Yang, Q. Gong, W. Qin, Z. Yang, Y. Cheng, L. Lu, X. Ge, C. Zhang, Z. Wu, F. Li,
Genome-wide analysis of WOX genes in upland cotton and their expression pattern under different
stresses, BMC Plant Biol. 17 (2017) 113.
E

[33] O. Shemer, U. Landau, H. Candela, A. Zemach, L.E. Williams, Competency for shoot regeneration
from Arabidopsis root explants is regulated by DNA methylation, Plant Sci. 238 (2015) 251-261.
CC

[34] K. Hiratsu, K. Matsui, T. Koyama, M. Ohme-Takagi, Dominant repression of target genes by

chimeric repressors that include the EAR motif, a repression domain, in Arabidopsis, Plant J. 34 (2003)
733-739.
A

[35] E. van der Graaff, T. Laux, S.A. Rensing, The WUS homeobox-containing (WOX) protein family,
Genome Biol. 10 (2009) 248.
[36] J.L. Gallois, F.R. Nora, Y. Mizukami, R. Sablowski, WUSCHEL induces shoot stem cell activity
and developmental plasticity in the root meristem, Genes Dev. 18 (2004) 375-380.
[37] T.Q. Zhang, H. Lian, C.M. Zhou, L. Xu, Y. Jiao, J.W. Wang, A two-step model for de novo
activation of WUSCHEL during plant shoot regeneration, Plant Cell 29 (2017) 1073-1087.
[38] B. Negin, O. Shemer, Y. Sorek, L. Eshed Williams, Shoot stem cell specification in roots by the
WUSCHEL transcription factor, PLoS One 12 (2017) e0176093.

T
R IP
SC
U
N
A
M
ED
E PT
CC
A
Figure captions
Fig. 1. Sequence alignment and domain analysis of the WUS in Arabidopsis and
upland cotton. HOX domain is marked in red border and WUS-box is indicated in
blue border.
Fig. 2. Expression profile of GhWUS. Relative expression levels of GhWUS1b (A)
and GhWUS1a (B) in different tissues of Gossypium hirsutum were determined by
qRT-PCR using GhHis3 as the internal control gene. The callus expression level was

T
regarded as “1”, error bars indicate the standard deviation from three independent

IP
experiments.

R
Fig. 3. Ectopic expression of GhWUS in Arabidopsis thaliana. (A) Quantitative

SC
RT-PCR and semi-quantitative RT-PCR analysis of GhWUS expression in different
transgenic lines with and without inducer. (B-E) Phenotypes of pER8::eGFP-GhWUS

U
seedlings grown for 8 days on the Murashige and Skoog’s medium supplemented with
N
10 μM 17-β-estradiol, leaf-like (B), embryo-like (C), callus-like (D), and normal (E).
A
Bars = 1 mm. (F) The frequency of occurrence of four phenotypes of different
transgenic lines. (G-H) qRT-PCR analysis of LEC1, FUS3, ABI3, and CLV3
M

expression in different transgenic lines and wild-type (Col) plants. AtACT2 was used
ED

as the internal reference gene for qRT-PCR analysis and AtAPT was used for
semi-quantitative RT-PCR analysis. The expression level in wild-type (Col) plants
was set as 1, error bars indicate the standard deviation from three independent
PT

experiments.
Fig. 4. Overexpression of GhWUSs in Arabidopsis roots promotes de novo shoot
E

organogenesis. (A) The process of shoot regeneration and (B) shoot regeneration
CC

frequencies in Line A12-27, Line A10-1, and WT root explants transferred onto SIM
at indicated times. Bars = 2 mm. Error bars indicate the standard deviation from three
A

biological replicates.
Fig. 5. Effects of GhWUS overexpression on shoot regeneration from the excised
roots of Arabidopsis. Root explants derived from Col-0, Line A10-1, and Line
A12-27 seedlings were cultured on the inductive Murashige and Skoog’s (MS)
medium supplemented with auxin (A, B, and C) or cytokinin (F, G, and H), and in the
absence of hormone (I, J, and K). D shows the enlarged view of the area marked in
red border in C. E shows the enlarged view of the area marked in D. (L) qRT-PCR
analysis of the expression of GhWUSs in the transgenic root explants cultured on the
inductive MS medium for 0, 1, 3, and 6 days. (M-N) qRT-PCR analysis of the
expression of CLV3 and CUC2 in wild-type, Line A10-1, and Line A12-27 root
explants cultured on the inductive MS medium for 0, 1, 3, and 6 days using AtACT2
as the internal control gene. The expression level in wild-type (Col) plants was set as

T
1. Bars, 1 cm (A-C, F-H); 1 mm (D, I-K); 0.5 mm (E). Error bars indicate the standard

IP
deviation from three independent experiments.

R
Fig. 6. Expression of the chimeric GhWUS repressor inhibits cotton EC

SC
formation. (A) Calli transformed with 35S::GhWUSs-SRDX vector cultured for 3
months in EIM. (B) EC transformed with empty vector cultured for 2 months in EIM.

U
(C) Overexpression of GhWUSs-SRDX signifcantly decrease EC differentiation rate
N
(the embryogenic callus percentage of all callus) which was calculated at 30, 60, and
A
90 days after incubation on EC induction medium. Bars = 5 mm. Error bars indicate
the standard deviation from three independent experiments.
M
ED
E PT
CC
A
A Figures:

CC
EPT
ED
M
A
N
U
SC
RIP
T
A
CC
EPT
ED
M
A
N
U
SC
RIP
T
A
CC
EPT
ED
M
A
N
U
SC
RIP
T