Plant Cell Rep

DOI 10.1007/s00299-015-1769-x

ORIGINAL PAPER

Inside out: high-efficiency plant regeneration and Agrobacterium-

mediated transformation of upland and lowland switchgrass
cultivars
Yan-Rong Liu Hui-Fang Cen Jian-Ping Yan

Yun-Wei Zhang Wan-Jun Zhang

Received: 19 January 2015 / Revised: 7 February 2015 / Accepted: 10 February 2015

Springer-Verlag Berlin Heidelberg 2015

Abstract and transgenic plants could be obtained in 2 months after

Key message Selection of pre-embryogenic callus from
a core structure from mature seed-derived callus is the
key for high-efficiency plant regeneration and trans-
formation of switchgrass different cultivars.
Abstract Switchgrass (Panicum virgatum L.) has been
identified as a dedicated biofuel crop. For its trait
improvement through biotechnological approaches, we
have developed a highly efficient plant regeneration and
genetic transformation protocol for both lowland and
upland cultivars. We identified and separated a pre-
embryogenic core structure from the seed-derived cal-
lus, which often leads to development of highly regen-
erative type II calluses. From the type II callus, plant
regeneration rate of lowland cultivars Alamo and Perfor-
mer reaches 95 %, and upland cultivars Blackwell and
Dacotah, 50 and 76 %, respectively. The type II callus was
also amenable for Agrobacterium-mediated transformation.
Transformation efficiency of 72.8 % was achieved for
lowland cultivar Alamo, and 8.0 % for upland cultivar
Dacotah. PCR, Southern blot and GUS staining assays
were performed to verify the transgenic events. High
regenerative callus lines could be established in 3 months,
Agrobacterium infection. To our knowledge, this is the first
report on successful plant regeneration and recovery of
transgenic plants from upland switchgrass cultivars by
Agrobacterium-mediated transformation. The method pre-
sented here could be helpful in breaking through the bot-
tleneck of regeneration and transformation of lowland and
upland switchgrass cultivars and probably other recalci-
trant grass crops.
Keywords Biofuel crop  Embryogenic callus 
Regeneration  Switchgrass  Transgene
Abbreviations
2,4-D 2,4-Dichlorophenoxyacetic acid
BAP 6-Benzylaminopurine
CH Casein hydrolysate
GA Gibberellin
Introduction

Switchgrass (Panicum virgatum L.) is a C4 warm-season

Communicated by K. Wang.
Y.-R. Liu  H.-F. Cen  J.-P. Yan  Y.-W. Zhang 
W.-J. Zhang (&)
Department of Grassland Science, China Agricultural
University, No. 2 Yuanmingyuan Xilu, Haidian District,
Beijing 100193, Peoples Republic of China
e-mail: wjzhang@cau.edu.cn
Y.-W. Zhang  W.-J. Zhang
National Energy R&D Center for Biomass (NECB), China
Agricultural University, No. 2 Yuanmingyuan Xilu, Haidian
District, Beijing 100193, Peoples Republic of China
perennial grass, native to North America with the char-
acteristics of high biomass and low maintenance cost, and
has been identified as an important biofuel crop by US
Department of Energy (Vogel 2004; McLaughlin et al.
2006; Bouton 2007). With the increasing interests in con-
version of cellulosic biomass into biofuel, trait improve-
ment to reduce the cost in making biofuels is an important
research direction (Fu et al. 2011; Xu et al. 2011). Because
of the nature of outcrossing and polyploidy, breeding of
switchgrass is a major challenge through traditional

123

methods (Vogel and Mitchell 2008; Xu et al. 2011).
Molecular breeding of switchgrass through plant tissue
culture and genetic transformation is relatively new but a
promising alternative. However, the achievements so far
are limited mainly due to the recalcitrance in plant regen-
eration and transformation of switchgrass (Gressel 2008;
Xi et al. 2009). It is particularly true for the upland
switchgrass cultivars. So far no genetic transformation of
upland cultivars has been reported. Even callus regenera-
tion is very difficult to achieve (Song et al. 2012; Nages-
wara-Rao et al. 2013). Thus, establishment of a highly
efficient and amenable plant regeneration and transforma-
tion system of switchgrass is still in urgent need.
Various explants such as inflorescence (Richards et al.
2001; Burris et al. 2009; Xi et al. 2009), seedling segment
(Song et al. 2012) and mature seed (Xu et al. 2011; Li and
Qu 2011) have been used for callus initiation in switch-
grass. Basal medium used in the reported work included
MS (Li and Qu 2011; Xu et al. 2011; Song et al. 2012), NB
(Ramamoorthy and Kumar 2012), and LP (Burris et al.
2009) media. Plant regeneration and/or transformation of
different types of calluses has been described (Burris et al.
2009; Li and Qu 2011; Xu et al. 2011; Song et al. 2012).
Up to date, a few laboratories have reported plant regen-
eration and Agrobacterium-mediated transformation of the
lowland switchgrass cultivars, such as Alamo and Perfor-
mer (Somleva et al. 2002; Li and Qu 2011). Certain se-
lected genetic background of Alamo plant lines was
reportedly able to regenerate well, but to identify such lines
is time consuming and difficult, and the seed source of the
highly regenerative lines is very limited (Somleva et al.
2008; Xu et al. 2011). Regeneration of plants from upland
switchgrass cultivars is very difficult and has not been
reported (Xi et al. 2009; Burris et al. 2009; Song et al.
2012). Thus, establishing a plant regeneration procedure
with less genotype dependence in switchgrass is highly in
demand.
Mature seeds were used widely as explants, since it is
year-round available and easy to access (Xi et al. 2009; Xu
et al. 2011). However, calluses induced from switchgrass
mature seeds were actually a mixture of different types (Li
and Qu 2011). Only the Type II embryogenic callus (EC) is
highly regenerative (Burris et al. 2009). This phenomenon
was also found on other monocots, such as maize (Arm-
strong and Green 1985; Vasil and Vasil 1994; Welter et al.
1995), rice (Rueb et al. 1994), sugarcane (Guiderdoni and
Demarly 1988), wheat (Parmar et al. 2012), and various
non-food grasses (Lu and Vasil 1981; Zhang et al. 2007; Li
and Qu 2011). In general, two types of calluses had been
classified and described depending on color, regeneration
ability and texture. Type II ECs have been defined as fri-
able, rapidly growing, and highly regenerative, but it was
formed typically at a lower frequency than non-
123
Plant Cell Rep
embryogenic callus. Type I callus was described as slowly
growing, highly organized, and was hardly regenerative
(Vasil and Vasil 1986, 1994). In switchgrass, ap-
proximately 1015 % calluses in cv. Alamo are identified
as Type II calluses (Burris et al. 2009; Li and Qu 2011).
In this correspondence, we report that many switchgrass
calluses have a shell-core structure with the core
being able to develop to type II callus and the shell being
spongy and non-embryogenic callus. The discovery makes
it much easier to develop and propagate the type II callus at
a relatively early stage of tissue culture. The procedure
presented here is less genotype dependent: regeneration of
the lowland cultivars (Alamo and Performer) type II cal-
luses recovered this way was as high as 95 %. The type II
calluses were even observed from upland cultivars,
Blackwell and Dacotah, with high regeneration rates, at 50
and 76 %, respectively. The type II callus was also highly
amenable to Agrobacterium-mediated transformation.
Transformation efficiencies of 72.8 % for lowland cultivar
Alamo and 8 % for upland cultivar Dacotah were achieved.
To our knowledge, this is the first report of successful
recovery of transgenic plants from an upland switchgrass
cultivar.
Materials and methods
Callus induction and culture
Mature seeds of two lowland switchgrass cultivars, Alamo
and Performer, and two upland cultivars, Blackwell and
Dacotah, were used as explants for callus induction. The
seeds were purchased from Ernst Conservation Seeds Co.
(Meadville, PA, USA). The seeds were rinsed in 70 % (v/
v) alcohol for 1 min, surface sterilized with 50 % (v/v)
Clorox for at least 2 h with gentle stirring, and then rinsed
thoroughly with sterilized water at least five times. Ster-
ilized seeds were placed on the callus induction medium,
and cultured at 25 C in the dark for callus initiation. To
evaluate the effects of various culture media on callus
induction of these cultivars, media MS5:1, MS7 and NB7
were formulated (Table 1). All the media were adjusted to
pH 5.8 and sterilized at 121 C for 20 min. A hundred
seeds were used as a replicate, with three replicates per
experiment.
Observation of the shell and core structure
and development of type II embryogenic callus
Six weeks after callus induction, calluses were examined
under a stereomicroscope. When the spongy-like outer
layer of a callus was sliced open with sharp-tip tweezers, a
piece of compact, firm, yellowish, core-like structure was
Plant Cell Rep

Table 1 Culture media used in the experiments

Medium Composition

MS5:1 MS salts and vitamins, 30 g l-1 maltose, 5.0 mg l-1 2,4-D, 1.0 mg l-1 6-BA, 4.0 g l-1 phytagel
MS7 MS basic salt and vitamins, 30 g l-1 Maltose, 7.0 mg l-1 2,4-D, 4.0 g l-1 phytagel
NB7 N6 basic salt, 1 9 vitamin B stockA, 30 g l-1 Maltose, 500 mg l-1CH, 500 mg l-1 L-proline, 7.0 mg l-1 2,4-D, 4.0 g l-1 phytagel
MP MS5:1 medium with 2.0 g l-1 L-proline
REG MS basic medium with 0.2 mg l-1 NAA, 0.5 mg l-1 GA, 1.0 mg l-1 6-BA, 30 g l-1 maltose, 4.0 g l-1 phytagel
SM1 MP medium with 200 mg l-1 timentin and 50 mg l-1 hyg B
SM2 MP medium with 200 mg l-1 timentin and 100 mg l-1 hyg B
1/2TM 1/2 MS based medium with 200 mg l-1 timentin and 50 mg l-1 hyg B
A
1 9 vitamin B stock was composed of 9.9 mg l-1 thiamine, 9.5 mg l-1 pyridoxine hydrochloride, 4.5 mg l-1 nicotinic acid (Zhang et al.
2013)

often seen in the center of the callus. After separation and
subculture onto MP medium (Table 1) for proliferation,
these pre-embryogenic core structures often grew into type
II callus lines which were fast-growing, friable, granular
and had yellowish color (Burris et al. 2009; Li and Qu
2011).
To evaluate the occurrence of the core structure and
their development into the type II callus from different
cultivars of switchgrass, 180 pieces of mature seed-derived
calluses were randomly chosen from each cultivar in a
triplicated experiment. They were sliced open, and the
number of calluses with such core-like structures was
recorded. After subculture of the cores, the number of
cores that developed into type II calluses was counted
(Table 3).
Regeneration of the type II calluses
Sixty of 3-month-old type II calluses (about 2 mm in dia-
meter) from each switchgrass cultivar were placed onto
REG medium (Table 1) for shoot induction in a culture
room with photoperiod of 16 h light/8 h dark, light inten-
sity of 100 lmol m-2 s-1, and temperature at 25 C. After
3 weeks, the regenerated shoots were transferred to a
hormone-free 1/2 MS medium for rooting. The experiment
was carried out in three separated replicates. The regen-
eration rate was defined as the number of calluses with
shoot formation divided by the number of calluses utilized
in the experiment.
Microscopy
Calluses of different switchgrass cultivars were observed
under a stereomicroscope (Nikon, C-DSS230). To make
cross sections, calluses were fixed in FAA solution (50 %
ethanol, 5.0 % glacial acetic acid, and 3.7 % for-
maldehyde) and subsequently dehydrated in a graded
ethanol series (2 9 50, 60, 70, 85, 95 %, 3 9 100 %), then
embedded in paraffin, and sectioned using a microtome.
The mounted specimens were viewed and digital images
were captured.
Agrobacterium strain and transformation
Agrobacterium tumefaciens strain EHA105 (Hood et al.
1993) harboring a binary vector pCAMBIA1305.1 (infor-
mation see: http://www.cambia.org) was used in the
experiment. The T-DNA region included a hygromycin
(hyg) B resistance gene, hpt, as a selectable marker and an
intron interrupted b-glucuronidase gene (gus), each driven
by a CaMV 35S promoter. The Agrobacterium strain was
grown in YEP medium (Sambrook and Russell 2001)
containing 50 mg l-1 rifampicin, 100 mg l-1 kanamycin
and cultured in an incubator shaker at 28 C and 220 rpm
until OD600 reading reached 0.81.0. The cells were col-
lected by centrifugation at 3000g for 10 min. The pelleted
cells were re-suspended in liquid MP medium to OD600
reading of *0.5 before infection.
The transformation procedure was carried out in
accordance to Zhang et al. (2013). A cold treatment was
given to calluses (4 months old) in a maltose solution on
ice for 20 min prior to Agrobacterium infection. Calluses
were then immersed in Agrobacterium suspension in a
50 ml sterile plastic tube, and vacuum treated (-0.08 MPa,
10 min) to facilitate infection followed by gentle agitating
for 20 min at 80 rpm at 28 C. The bacterium suspension
was discarded, and the calluses were blotted with sterile
filter papers for 30 min to remove excessive Agrobacter-
ium. The calluses were then placed on two layers of sterile
filter paper wetted with 1 ml sterile water in a petri dish for
co-cultivation, at 25 C in darkness for 3 days. After co-
cultivation, calluses (about 3 mm in diameter) were cul-
tured on MP medium containing 200 mg l-1 timentin for
7 days for callus recovery and suppression of Agro-
bacterium growth.

123

Selection and regeneration of putative transgenic plants
After recovery, calluses were transferred to the first selection
medium (SM1) containing 50 mg l-1 hyg B (Roche Diag-
nostics GmbH, Germany). Two weeks later, calluses with
vigorous growth were transferred to the second selection
medium (SM2) containing 100 mg l-1 hyg B. In another
2 weeks, resistant calluses were transferred onto REG medium
supplied with 200 mg l-1 timentin and 20 mg l-1 hyg B for
shoot formation. Regenerated shoots were rooted on 1/2 MS
medium containing 200 mg l-1 timentin and 50 mg l-1 hyg B.
Regenerated plantlets from one original callus cluster were
counted as one putative transgenic event. Rooted resistant
plants were then transplanted in 10-inch pots in the greenhouse
for further growth. Putative transgenic plants were analyzed by
PCR and GUS staining assay, and some were verified by
Southern analysis. The transformation efficiency was defined
as the number of PCR/GUS staining positive plantlets forma-
tion divided by the total number of infected calluses.
GUS staining assay
GUS expression was first tested 1 week after infection.
When shoots were formed, leaf fragments were stained in a
GUS staining buffer (Jefferson 1987). After staining, 70 %
ethanol was used to remove the chlorophyll.
PCR and Southern blot analysis
Genomic DNA was extracted from young leaves of the hyg
BR and non-transgenic plants by the CTAB method (Gao
et al. 1989). For polymerase chain reaction (PCR) analysis,
100 ng DNA was used as template in a 20 ll PCR reaction.
Primers used to detect the CaMV 35S promoter were
50 -GAAACCTCCTCGGATTCCATT-30 (forward) and
50 -TGCTTTGAAGACGTGGTTGGA-30 (reverse). Annealing
temperature was 55 C in a standard PCR reaction to get a
207 bp PCR product. For Southern blotting assays, 25 lg of
plant genomic DNA was digested with restriction enzyme
BamHI, using hpt PCR products as the hybridization probe.
Primers used for hpt gene probe amplification were: 50 -
TACTTCTACACAAGCATCGGTCCAG-30 (forward) and
50 -CTTGACATTGGGGAGTTTAGCGAGA-30 (reverse).
Probe labeling and hybridization followed the directions of
DIG High primer DNA labeling and detection starter kit II
(Roche Applied Science. Mannheim, Germany). The hy-
bridized membranes were exposed to Kodak X-OMAT BT
X-ray film for autoradiography.
Statistical analyses
The tissue culture experiments were analyzed by one-way
analysis of variance (ANOVA). The comparison of
123
Plant Cell Rep
treatments was separated by Duncans multiple range test
(p \ 0.05). The proc GLM for ANOVA of SAS 8.2 (SAS
Institute, Cary, NC, USA) was used for the analyses.
Results
Callus induction
To optimize tissue culture conditions of switchgrass, the
callus induction responses of different cultivars on differ-
ent media were compared. Two lowland cultivars (Alamo
and Performer) and two upland cultivars (Blackwell and
Dacotah) were investigated in the experiments. As showed
in Table 2, callus induction rates of the Alamo were around
2733 %, Performer was around 4158 %, Blackwell was
around 6578 %, and Dacotah was around 7281 %.
Upland cultivars Blackwell and Dacotah had overall higher
callus induction rate, most likely reflecting the seed vigor
of various seed sources. Overall, callus induction rates of a
particular switchgrass cultivar on different media were
similar, and NB7 was chosen for future experiments.
Identification of the shell-core structure
in switchgrass callus and development of the type II
callus
In our pilot experiments, a shell-core structure in many
calluses derived from switchgrass mature seed was ob-
served after the calluses were sliced open. The core cell
clusters were yellowish, compact, firm, and granular,
morphologically different from the shell (Fig. 1b, c).
Interestingly, a majority of the callus proliferated in sub-
culture from the separated cores gradually developed
into type II callus, which was fast growing, friable, and
highly regenerative. Based on this observation, a procedure
to recover highly regenerative type II callus was developed.
Under a stereo microscope, the outer layer of callus was
a white, soft, spongy shell, which was characterized as
Table 2 Callus induction rates of switchgrass cultivars on different
media
Cultivar Callus induction rates on different media/%
MS5:1 MS7 NB7
Alamo 27.39 1.6 a 33.12 2.1 a 33.32 1.3 a
Performer 53.33 8.8 a 41.11 6.9 a 58.78 1.5 a
Blackwell 65.55 7.6 ab 68.89 1.9 ab 77.78 1.7 a
Dacotah 75.37 1.2 b 81.11 3.8 a 71.67 1.8 b
Each value represents mean SE from three replicates with a total of
300 seeds. The value followed with different letters in the same row
indicated statistical differences between the callus induction rate
according to ANOVA analysis (p \ 0.05)
Plant Cell Rep
Fig. 1 Characterize the shell and core structure and type II
embryogenic callus of switchgrass. a Six-week-old callus of switch-
grass derived from mature seed; b, c typical front views of 6-week-old
callus when the outer layer spongy shell was sliced open; (C indicate
the pre-embryogenic core structure); d some calluses do not have a
fast-growing, loosely organized, and hardly regenerative
(Fig. 1a, b). When the callus was sliced open, quite often, a
piece of core-like, yellowish, compact callus was seen in
the middle of the callus, loosely attached to the shell,
and was easy to separate (Fig. 1b, c). However, we also
saw that some calluses did not have such core-like struc-
tures (Fig. 1d). Histology analysis of the 3-week-old callus,
as showed in Fig. 1e, revealed the development process
from the core structure to type II callus with multiple
meristematic cell clusters shown. Two-month-old calluses
proliferated from the core did have the characters of type
II callus, and somatic embryos were forming on the surface
of callus (Fig. 1f). No obvious differences were observed
under the microscope between the type II calluses of
lowland cultivars and upland cultivars.
To evaluate the frequencies of the core structure for-
mation and their development into type II callus, calluses
of switchgrass lowland (Alamo, Performer) and upland
(Blackwell and Dacotah) cultivars were randomly chosen
and the spongy shells were sliced open. The cores were
separated and subcultured on MP medium. More than half
of them would become fast-growing, friable type II callus
lines, which would in turn be transferred to the REG
medium to evaluate their regeneration ability. As showed
in Table 3, the tested four cultivars had significant
pre-embryogenic core structure (nc no-core); e section of a 3-week-
old callus derived from mature seeds, showing the core development
from a cluster of meristem cells; f cross section of a 2-month-old type
II callus of switchgrass cultivar Alamo; The bar equals to 2 mm in
picture ad, and to 1 mm in picture e and f
differences in the percentages of calluses that had a core,
even the calluses looked alike from outside (Fig. 2a). The
lowland cultivars Alamo and Performer had remarkably
higher percentages of callus with a core (67 and 84 %,
respectively), than the upland cultivars Blackwell (30 %)
and Dacotah (45 %). Three weeks after subculture,
approximately 5572 % of the separated cores grew
fast, developed granular embryonic structures on the sur-
face of the callus, and became type II callus (Figs. 1f, 2b).
The percentages of cores developing into type II calluses
between upland and lowland cultivars were somewhat
similar. However, the regeneration rates of upland type II
calluses were significantly lower than lowland cultivars,
but still above 50 % (Table 3). Other calluses grew slowly
and was similar to the type I callus described in prior
reports (Vasil and Vasil 1986; Burris et al. 2009).
In general, 34 months of subculture may be needed to
obtain enough material from a callus line that was gener-
ated from a single seed for transformation experiments
(Fig. 2c). Of the tested cultivars, we only obtained regen-
erated plants from the type II calluses; neither the spongy-
shell callus nor type I callus was regenerative (Fig. 2d).
Normally, small shoots could be observed a week after the
type II callus was cultured on the REG medium. High re-
generation rates ([95 %) were observed from type II callus
123
 Plant Cell Rep

Table 3 Occurrence of core structure, type II callus and the regeneration rates of different switchgrass cultivars initiated on medium NB7
Cultivars Percentage of callus with Percentage of cultured cores that develop Regeneration rate of type II
a core-structure (%) to type II callus (%) callus (%)

Alamo 67.63 5.6 bA 55.85 5.1 b 96.56 1.1 a

Performer 84.10 1.3 a 72.99 3.5 a 97.78 2.5 a
Blackwell 29.70 9.7 c 63.23 8.1 ab 50.21 4.2 c
Dacotah 44.92 2.1 c 64.22 4.6 ab 76.24 8.7 b
A
Each value represents mean SE from three replicated experiments with a total of 180 pieces of initiated calluses. The value followed with
different letters in the same column indicated statistical differences between cultivars according to ANOVA analysis (p \ 0.05)

Fig. 2 Development of type II callus is the key to obtain regenerated
plants of switchgrass. a Six-week-old callus induced from switchgrass
mature seeds; b embryogenic calluses from the core structure after
3-week culture on MP medium; c type II callus proliferated on MP
medium for 1 month; d only the type II callus can regenerate into
shoots on REG medium (2 weeks), whereas type I callus only had
hairy roots formed, and spongy-like callus (III) cannot regenerate; e
h regeneration of type II callus of Alamo (e), Performer (f), Blackwell
(g) and Dacotah (h); il The regenerated plants of Alamo (i),
Performer (j), Blackwell (k) and Dacotah (l) growing in pots

123
Plant Cell Rep
lines of lowland cultivars Alamo (Fig. 2e) and Performer
(Fig. 2f). Type II callus lines of upland cultivars Blackwell
(Fig. 2g) and Dacotah (Fig. 2h) were also able to regen-
erate at rate of 50 and 76 %, respectively. Regenerated
shoots were rooted on 1/2 MS medium for 3 weeks, and
then transplanted to soil in pots. Most of them survived and
resumed growth (Fig. 2il).
Agrobacterium-mediated transformation of type II
calluses
To test whether the type II calluses were amenable in
Agrobacterium-mediated transformation, calluses of
switchgrass cultivars Alamo and Dacotah were transformed
with Agrobacterium strain EHA105 that harbored a binary
vector pCAMBIA1305.1 (Table 4). The T-DNA region
was showed in Fig. 3a. After co-cultivation on filter paper
for 3 days (Fig. 3b), calluses were transferred to MP
medium containing timentin to suppress Agrobacterium
growth but free of selection reagent hyg B for recovery of
the infected calluses. One week after Agrobacterium
infection, transient expression of GUS was tested; all the
Agrobacterium-infected calluses tested showed GUS
staining signals, but not in the non-infected callus (Fig. 3c).
In the following selection, calluses came from the same
original callus were put together as one resistant callus line
(Fig. 3d). After two rounds of selection with 50 and
100 mg l-1 hyg B, respectively, resistant calluses were
transferred to regeneration medium containing 20 mg l-1
hyg B for shoots induction. In the control, calluses that
were not infected with Agrobacterium had no green shoot
regenerated on the selection REG medium (Fig. 3e). In the
contrary, green shoots were regenerated from the hyg
B-resistant calluses (Fig. 3f). The regenerated shoots were
rooted on half strength MS medium supplemented with
50 mg l-1 hyg B for further selection (Fig. 3g). Rooted
putative transgenic plants were transplanted into soil in
pots for further growth (Fig. 3h). To analyze the expression
of the stable transformed gus gene, leaf segments from
Table 4 Agrobacterium-mediated transformation of type II callus of switchgrass cultivars Alamo and Dacotah
Varieties Experiment No. of callus infected
Alamo 1 53
2 49
3 49
Total 151
Dacotah 1 49
2 58
3 43
Total 150
individual putative transgenic plants were tested by GUS
staining assay. After removing the chlorophyll, intense
blue color as the result of GUS activity was observed on
the leaf fragments of transgenic plants, but not on the wild
type plant (Fig. 3i).
In the experiments, calluses of switchgrass cultivars
Alamo and Dacotah were transformed. A total of 110 and
12 putative transgenic plants lines were recovered from the
two cultivars, respectively. The putative transgenic plants
were first examined by PCR tests on the CaMV 35S pro-
moter. As showed in Fig. 4a, the expected size of DNA
fragment (207 bp) was amplified, no escapes were found.
For further confirmation, PCR positive plants were ran-
domly selected for Southern blot analyses. As showed in
Fig. 4b, the transgene in the genome of PCR positive plants
were detectable. Representative hybridization results were
shown from five Alamo transgenic plants and three Da-
cotah transgenic plants tested. Based on the PCR tests, the
transformation efficiencies were 72.8 and 8.0 % for Alamo
and Dacotah, respectively.
Discussion
A plant callus is a cluster of unorganized cells usually
consisting of two kinds of cells: small, dense, actively
dividing meristematic cells and larger, vacuolar, quiescent
parenchymatous cells. Calluses having various morpho-
logical phenotypes, mainly due to explant sources and
culture medium components, are often observed (Bho-
jwani and Dantu 2013). Quite often, calluses with various
morphologies are interconvertible, meaning that one
type of callus could gradually grow out of another type of
callus in culture (Aitchison et al. 1977). In our experi-
ence, grass calluses can be divided into two main cate-
gories based on the arrangement of the two kinds of cells:
one with meristematic cells mostly on the surface of the
calluses and the parenchymatous cells at the inside (such
as in rice and tall fescue), and another having the
No. of PCR?/Gus? plants Transformation efficiency (%)
39 73.6
36 73.5
35 71.4
110 72.8
4 8.1
5 8.6
3 7.0
12 8.0
123

Fig. 3 Illustration of Agrobacterium-mediated transformation and
plant regeneration of type II callus (c.v Alamo). a Schematic map of
T-DNA region of plasmid pCAMBIA1305.1; LB, left border; 35S,
CaMV 35S promoter; hpt, hygromycin B resistance gene; gus,
b-glucuronidase gene; pA, CaMV 35S polyA terminator; N, nopaline
synthase gene terminator; RB, right border; Probing region used in
Southern blotting is also shown. b Co-cultivation of Agrobacterium-
infected switchgrass type II callus on filter paper; c transient
expression assay of gus gene in the control (left) and
meristematic cells as a core at inside with par-
enchymatous cells surrounding the core at the outer
layer (such as in some bermudagrass cultivars). It is likely
caused by the genetic background. In most cases, the two
types of cells are packed tightly and not easy to
separate. In this report, we observed a shell and core
structure in many switchgrass calluses derived from ma-
ture seeds, indicating that they belong to the second
category with an exception: the core, mainly consisting of
the meristematic cells, is loosely connected to the sur-
rounding parenchymatous cells in the shell, and easy to
recognize and separate. With this discovery, a new
method is developed to more efficiently obtain type II
calluses. With this method, we not only achieved high
callus regeneration and transformation efficiencies for the
lowland cultivars, but also, for the first time, recovered
regenerated transgenic plants from an upland cultivar,
which was reportedly highly recalcitrant in regeneration
and transformation (Burris et al. 2009; Song et al. 2012).
After the core structures were separated and placed in
subculture, many of them grew fast and developed into
123
Plant Cell Rep
Agrobacterium-infected (right) calluses after 4 days culture on MP
medium containing 200 mg l-1 timentin; d Hyg B-resistant calluses
in second selection with 100 mg l-1 hyg B; e, f plant regeneration of
non-infected switchgrass type II calluses (e) and the hyg B-resistant
calluses (f) on REG medium containing 20 mg l-1 hygromycin B for
2 weeks; g resistant shoots on rooting medium containing 50 mg l-1
hygromycin B; h regenerated plants were transplanted to flowerpots;
i staining assay of the expression of gus gene in the transgenic (right)
and wild type switchgrass plants (left)
type II embryogenic calluses. The yellowish, friable, and
fast-growing type II calluses were vital for switchgrass
plant regeneration and transformation (Li and Qu 2011;
Ramamoorthy and Kumar 2012), most likely because, in
type II callus, the meristematic cells are exposed on the
surface which could facilitate callus regeneration and
Agrobacterium infection. Good transformation efficiencies
were achieved in a couple of previous reports using iden-
tified type II calluses in lowland cultivars (Li and Qu 2011;
Ramamoorthy and Kumar 2012). Based on our observa-
tion, those calluses could be originated from the core
structures, which may eventually grow out of the shell.
Lowland cultivars generated higher percentage of calluses,
had a core structure, and were also found with higher
plant regeneration and transformation rates. On the con-
trary, data indicate that upland cultivars had a low per-
centage of calluses with the core structure (Table 3).
Since no type II calluses were reported previously from
upland cultivars, it is hypothesized that the upland cultivar
calluses may have tighter and/or stronger shell structure
that prevents the core cells from growing out of the
Plant Cell Rep
Fig. 4 PCR and Southern blot analysis confirm the hyg B-resistant
switchgrass plants containing the transgene DNA. a PCR assay of
transgenic plants with a pair of primers of CaMV 35 promoter (target
band 207 bp); b Southern blot analysis with a fragment of hpt gene as
the hybridization probe. The gDNA samples were digested with
restriction enzyme BamHI that has one cleavage site within the
T-DNA and is out of the transgene; M, 1 kb DNA ladder, -, non-
transgenic plant DNA as a negative control; lane 15, randomly
chosen transgenic plants of Alamo; lane 68, randomly chosen
transgenic plants of Dacotah; ?, 741 bp hpt gene PCR fragment
(indicate as asterisk) was used as a positive control
shell. This might explain why the plant regeneration and
transformation of upland switchgrass are still low. And
more histology work is needed to validate this hypothesis.
We have a novel discovery of the shell and core Gressel J (2008) Transgenics are imperative for biofuel crops. Plant
structure from mature seeds-derived calluses in switch-
grass. The unique protocol developed based on this ob-
servation makes recovery of the type II calluses, which are
highly regenerable and transformable, much easier and
more efficient. It is particularly true for the labs with less
experience on switchgrass tissue culture. In addition, the
transformation protocol we developed (Zhang et al. 2013)
could have also enhanced Agrobacterium-mediated trans-
formation of switchgrass. The approach we used enhanced
lowland switchgrass transformation, which makes it pos-
sible for upland cultivar transformation, and could find
applications in similar cases in other grass species.
Author contribution statement Liu YR finished most of
the experiments and collected data, and wrote the first draft
of manuscript. Cen HF gave assistance in performing
experiments and data collection. Yan JP finished the
experiment of microscopy. Zhang YW involved in the
research project by financial support. Zhang WJ designed
experiments and involved in paper revision and
submission.
Acknowledgements The work was supported by the National High
Technology Research and Development Program (863
2012AA101801) of China, and Chinese Universities Scientific Funds
(2012QJ120) to W.J.Z. We are grateful to Dr. Rongda Qu of NCSU
for his critical reading and advices on the paper.
Conflict of interest The authors declare that they have no conflict
of interest.
References
Aitchison PA, Macleod AJ, Yeoman MM (1977) Growth patterns in
tissue (callus) cultures. In: Street HE (ed) Plant tissue and cell
culture. Univ. of California Press, Berkeley and Los Angeles,
pp 267387
Armstrong CL, Green CE (1985) Establishment and maintenance of
friable, embryogenic maize callus and the involvement of L-
proline. Planta 164:207214
Bhojwani SS, Dantu PK (2013) Tissue and Cell Culture. In: Bhojwani
SS, Dantu PK (eds) Plant tissue culture: an introductory text.
Springer, India, pp 3950
Bouton JH (2007) Molecular breeding of switchgrass for use as a
biofuel crop. Curr Opin Genet Dev 17:553558
Burris JN, Mann DGJ, Joyce BL, Stewart CN Jr (2009) An improved
tissue culture system for embryogenic callus production and
plant regeneration in switchgrass (Panicum virgatum L.).
Bioenergy Res 2:267274
Carvalho CHS, Bohorova N, Bordallo PN, Abreu LL, Valicentle FH,
Bressan W, Paiva E (1997) Type II callus production and plant
regeneration in tropical maize genotypes. Plant Cell Rep 17:7376
Fu CX, Xiao XR, Xi YJ, Ge YX, Chen F, Bouton J, Dixon RA, Wang ZY
(2011) Down regulation of cinnamyl alcohol dehydrogenase
(CAD) leads to improved saccharification efficiency in switch-
grass. Bioenergy Res 4:153164. doi:10.1007/s12155-010-9109-z
Gao YF, Zhu Z, Xiao GF, Zhu Y, Wu Q, Li XH (1989) Isolation of
soybean kunitz trypsin inhibitor gene and its application in plant
insect-resistant genetic engineering. Acta Bot Sin 40:405411
Sci 174:246263
Guiderdoni E, Demarly Y (1988) Histology of somatic embryogenesis
in cultured leaf segments of sugarcane plantlets. Plant Cell Tiss
Org Cult 14:7188
Hood EE, Gelvin SB, Melchers LS, Hoekema A (1993) New
Agrobacterium helper plasmids for gene transfer to plants.
Transgenic Res 2:208218
Jefferson R (1987) Assaying chimeric genes in plants: the GUS gene
fusion system. Plant Mol Bio Rep 5:387405
Li RY, Qu RD (2011) High throughput Agrobacterium-mediated
switchgrass transformation. Biomass Bioenergy 35:10461054
Lu C, Vasil IK (1981) Somatic embryogenesis and plant regeneration
from leaf tissues of Panicum maximum Jacq. Theor Appl Genet
59:275280
McLaughlin S, Bouton J, Bransby D, Conger B, Ocumpaugh W,
Parrish D, Taliaferro C, Vogel K, Wullschleger S (1999)
Developing switchgrass as a bioenergy crop. Perspectives on
new crops and new uses. In: Janick J (ed) Perspectives on new
crops and new uses. ASHS Press, Alexandria, pp 282299
123

McLaughlin SB, Kiniry JR, Taliaferro CM, De La Torre Ugarte D,
Donald LS (2006) Projecting yield and utilization potential of
switchgrass as an energy crop. Adv Agron 90:267297
Nageswara-Rao M, Soneji JR, Kwit C, Stewart CN Jr (2013)
Advances in biotechnology and genomics of Switchgrass.
Biotechnol Biofuels 6:77
Parmar SS, Sainger M, Chaudhary D, Jaiwal PK (2012) Plant
regeneration from mature embryo of commercial Indian bread
wheat (Triticum aestivum L.) cultivars. Physiol and Mol Biol
Plants 18:177183
Ramamoorthy R, Kumar PP (2012) A simplified protocol for genetic
transformation of switchgrass (Panicum virgatum L.). Plant Cell
Rep 31:19231931
Richards HA, Rudas VA, Sun H, McDaniel JK, Tomaszewski Z,
Conger BV (2001) Construction of a GFP-BAR plasmid and its
use for switchgrass transformation. Plant Cell Rep 20:4854
Rueb S, Leneman M, Schilperoort RA, Hensgens LAM (1994)
Efficient plant regeneration through somatic embryogenesis from
callus induced on mature rice embryos (Oryza sativa L.). Plant
Cell Tiss Org Cult 36:259264
Sambrook J, Russell DW (2001) Molecular cloning. A laboratory
manual, 3rd edn. Cold Spring Harbor Laboratory Press, Cold
Spring Harbor
Somleva MN, Tomaszewski Z, Conger BV (2002) Agrobacterium-
mediated genetic transformation of switchgrass. Crop Sci
42:20802087
Somleva MN, Snell DK, Beaulieu JJ, Peoples OP, Garrison BR,
Patterson NA (2008) Production of polyhydroxybutyrate in
switchgrass, a value-added co-product in an important lignocel-
lulosic biomass crop. Plant Biotechnol J 6:663e78
123
Plant Cell Rep
Song GQ, Aaron W, Hancock JF (2012) Factors influencing
Agrobacterium-mediated transformation of switchgrass culti-
vars. Plant Cell Tiss Org Cult 108:445453
Vasil V, Vasil IK (1986) Plant regeneration from friable embryogenic
callus and cell suspension cultures of Zea mays L. Plant Physiol
124:399408
Vasil IK, Vasil V (1994) In vitro cultures of cereals and grasses.
Kluwer Academic Publishers, Dordrecht, pp 293312
Vogel KP (2004) Switchgrass. In: Moser LE, Burson BL, Sollen-
berger LE (eds) Warm-season (C4) grasses. American Society of
Agronomy, Washington, pp 561588
Vogel KP, Mitchell RB (2008) Heterosis in switchgrass: biomass
yield in swards. Crop Sci 48:21592164
Welter ME, Clayton DS, Miller MA, Petolino JE (1995) Morphotypes
of friable embryogenic maize callus. Plant Cell Rep 14:725729
Xi YJ, Fu CX, Ge YX, Nandakumar R, Hisano H, Bouton J, Wang
ZY (2009) Agrobacterium-mediated transformation of switch-
grass and inheritance of transgenes. Bioenerg Res 2:275283
Xu B, Huang LK, Shen ZX, Welbaum GE, Zhang XZ, Zhao BY
(2011) Selection and characterization of a new switchgrass
(Panicum virgatum L.) line with high somatic embryogenic
capacity for genetic transformation. Sci Hortic 129:854861
Zhang S, Hanna W, Ozias-Akins P (2007) Comparison of callus
induction and plant regeneration from different explants in
triploid and tetraploid turf-type bermudagrasses. Plant Cell Tiss
Org Cult 90:7178
Zhang WJ, Dewey RE, Boss W, Phillippy BQ, Qu RD (2013)
Enhanced Agrobacterium-mediated transformation efficiencies
in monocot cells is associated with attenuated defense responses.
Plant Mol Biol 81:273286