Reed (Phragmites communis) is a potential bioenergy plant. We report on its first Agrobacterium-mediated transformation using mature seed-derived calli. The Agrobacterium strains LBA4404, EHA105, and GV3101, each harboring the binary vector pIG121Hm, were used to optimize T-DNA delivery into the reed genome. Bacterial strain, cocultivation period and acetosyringone concentration significantly influenced the T-DNA transfer. About 48% transient expression and 3.5% stable transformation were achieved when calli were infected with strain EHA105 for 10 min under 800 mbar negative pressure and cocultivated for 3 days in 200 μm acetosyringone containing medium. Putative transformants were selected in 25 mg l−1 hygromycin B. PCR, and Southern blot analysis confirmed the presence of the transgenes and their stable integration. Independent transgenic lines contained one to three copies of the transgene. Transgene expression was validated by RT-PCR and GUS staining of stems and leaves.
The common reed (Phragmites communis Trinius) is a large perennial grass found in wetlands throughout temperate and tropical regions of the world (Clevering et al., 2001). Vigorous spreading by horizontal runners (over 5 m yr−1) makes reed an aggressive wetland invader in many parts of the world (Foggi et al., 2011). Reed produces fertile seeds predominantly by cross pollination. Self pollination or agamospermy also occurs in lesser extent, however, seed set was much lower (Ishii & Kadono, 2002; Gucker, 2008). Reed has considerable economic importance, such as a source for grazing by livestock, habitat for wildlife, pulp and paper production, removal of toxic metals including uranium, thorium, lead, and organic compounds and, more recently, as a source of biomass for energy production (Hijosa-Valsero et al., 2011; Li et al., 2011). Reeds are recognized as a potential high yielding biofuel crop with an estimated yield of up to 30 tons ha−1 lignocellulosic biomass annually (Allirand & Gosse, 1995). Lignocellulosic biomass offers the possibility of a renewable, geographically distributed, and relative greenhouse-gas-favorable source of ethanol and other liquid fuels. The ability of the candidate plants to grow on marginal agricultural land indicates that they can probably have a large impact on transportation needs without significantly compromising the land needed for food crop production (Himmel et al., 2007; Rubin, 2008). For instance, the common reed grows well in neutral pH or alkaline tropical and temperate water lands or wetlands unsuitable for most crops, thereby avoids competition with food crops for lands (Sathitsuksanoh et al., 2009). Reed grows well on marginal and nonagricultural lands with no or low fertilizer requirement and irrigation requirements. The biomass has carbohydrate content similar to those of agricultural residues, such as corn stover and wheat straw. Bioconversion of the reed biomass to ethanol has shown the potentiality of this species as feedstock for second generation bioethanol production (Scordia et al., 2011). Such multiple advantages have established a role for reed as a potential bioenergy plant.
Developing sustainable biofuel crops involves improving the productivity and biochemical composition of the biomass for efficient ethanol conversion (Himmel et al., 2007). In biomass crops, the secondary cell walls constitute the major fractions of lignin, hemi-cellulose, and cellulose. Among these, lignin, because of its linkage with phenolic acids (p-coumaric and ferulic acids), is not easily biodegradable and hinders the deconstruction of lignocellulosic biomass for the release of sugars for subsequent fermentation to form ethanol. In addition, lignin is indigestible and reduces fiber digestibility in ruminants. Moreover, the presence of lignin in cell walls also negatively affects paper pulping from reed. These properties have prompted research to reduce lignin content through genetic modification for increasing ethanol conversion and fiber digestibility. Thus, along with recognition of reed as a potential biomass/bioenergy crop, breeding programs should focus on its improvement for bioethanol production as well as drought tolerance for intensive commercial cultivation.
Conventional approaches of breeding such as exploitation of natural genetic variation (Casler, 1987) or divergent selection (Méchin et al., 2000) have many drawbacks in improving bioenergy-associated traits as these affect many desirable traits, such as concentrations of nonstructural carbohydrates, cell wall polysaccharide composition, fiber components, lignin-polysaccharide cross-linkages, and phenolic acids. These traits influence microbial and/or enzyme accessibility in industrial applications (Titgemeyer et al., 1996). Genetic engineering complements the conventional breeding and opens new avenues to the genetic modification of forage and bioenergy grasses. Transfer of gene(s) from heterologous species provide the means of selectively introducing/modifying new traits into crop plants and expanding the gene pool beyond what has been available to conventional breeding approaches. Setting up a high-frequency transformation system would greatly accelerate grass improvement via genetic engineering. Successful plant regeneration has been reported in reed from stem segments (Poonawala et al., 1999; Yang et al., 2003; Guo et al., 2004), immature inflorescences (Lauzer et al., 2000) and mature seeds (Straub et al., 1988; Wang et al., 2001; Cui et al., 2002; Kim et al., 2011). To date, no reports regarding genetic transformation of reed using either particle bombardment or an Agrobacterium-mediated approach have been published. We have established a reproducible and efficient callus-based transformation system mediated by A. tumefaciens. Genetic transformation could facilitate the genetic improvement of reed with respect to its use as an energy crop, feedstock, and forage material, and for the improvement of its disease and pest resistance, and tolerance to abiotic stresses.
Materials and methods
Plant materials and culture condition
Mature seeds of common reed were collected in early winter from local vegetation in Jinju, Republic of Korea (35°12′N 128°05′E), and kept at 4 °C. Callus induction and plant regeneration procedure was conducted following our earlier study (Kim et al., 2011). Seeds were submerged in 50% sulfuric acid for 30 min and manually dehusked. Sterilization was conducted by dipping seeds in 70% (v/v) ethanol for 2 min followed by immersion in 30% (v/v) Clorox for 30 min with vigorous agitation. They were then rinsed three to five times thoroughly with sterilized water and planted on callus induction media containing MS basal salts and vitamins fortified with 1.0 mg l−1 2,4-D, 30 g l−1 sucrose, and 2 g l−1 Gelrite, and incubated at 25 °C in the dark. About 25 seeds were planted in Ф87 mm Petri dish containing 25 ml of the medium. After 4–5 weeks of incubation, selected shiny and friable embryogenic calli (Fig. 2a) were maintained in a medium of same composition by subculturing every 2 weeks prior to transformation.
Plasmid and Agrobacterium strains
Three A. tumefaciens strains, LBA4404, EHA105, and GV3101, bearing the binary vector pIG121Hm (Hiei et al., 1994) were used for transformations. The T-DNA region of the binary vector contained a neomycin phosphotransferase II gene (nptII) under the control of the nopaline synthase (NOS) promoter, a hygromycin phosphotransferase gene (HPT) under the control of the CaMV 35S promoter, and a β-glucuronidase gene with a catalase intron (int-GUS) also driven by the CaMV 35S promoter (Fig. 1).
Inoculation and cocultivation
Agrobacterium cells were grown in YEP medium containing 50 mg l−1 kanamycin and 50 mg l−1 hygromycin B (DUCHEFA Biochemie, Haarlem, Netherlands) at 28 °C. Following overnight growth, the cells were harvested by centrifugation at 2500 g for 10 min and resuspended in liquid MS medium supplemented with 30 g l−1 sucrose with or without acetosyringone. The OD600 of the suspension was adjusted to approximately 0.8. Sixty pieces of embryogenic calli were immersed in 30 ml of the Agrobacterium suspension and kept in a vacuum chamber at 800 mbar for 10 min. The calli were then incubated at normal atmospheric pressure for another 30 min (Lee et al., 2004), blotted with filter paper to remove excess bacteria, and transferred to callus induction medium as above with different concentrations of acetosyringone. Cocultivation was performed in the dark at 25 °C for 1–7 days.
Selection and regeneration of transgenic plants
Following cocultivation, calli were washed with sterile distilled water containing 250 mg l−1 cefotaxime and cultured on resting medium (callus induction medium with 250 mg l−1 cefotaxime) for another 5 days at 25 °C in the dark to inhibit excess growth of Agrobacterium and allow the calli to recover from cocultivation shock. The calli were then transferred to selection medium (MS regeneration medium supplemented with 1.0 mg l−1 Dicamba, 30 g l−1 sucrose, and 25 mg l−1 hygromycin B). The cultures were incubated at 25 °C under a 16-h light photoperiod (100 μmol m−2 s−1) and subcultured in a medium of same composition every 2 weeks. After 4–5 weeks, hygromycin-resistant calli were selected and subsequently cultured in a medium of same composition for another 4–6 weeks. After regeneration, putative transgenic shoots were separated and transferred onto square culture vessel containing 50 ml rooting medium consisting of half-strength MS medium supplemented with 30 g l−1 sucrose, 2 g l−1 Gelrite, and 25 mg l−1 hygromycin B. After 3–4 weeks, hygromycin-resistant well-rooted plants were transferred to soil.
β-glucuronidase activity assay
To determine the frequency of gene transfer, transient GUS expression in the calli were tested after 3-day cocultivation. GUS staining of callus, leaves, and stem of the putative transgenic plants was detected according to our previous report (Lee et al., 2004). Leaf stem or callus tissues were incubated in 50 mm sodium phosphate buffer (pH 7.0) containing 2 mm 5-bromo-4-chloro-3-indolyl glucuronide (X-gluc) for 48 h at 37 °C. Tissues were then washed once with sterile distilled water and soaked in 70% ethanol for several days to remove chlorophyll when necessary. Transient GUS expression in post cocultivated calli was calculated by counting the GUS-positive calli appearing as blue zones (at least 1 mm in diameter) after staining.
Molecular analysis of transgenic plants
Hygromycin-resistant and GUS-positive putative transgenics were analyzed by PCR and Southern blot analysis. Genomic DNA was isolated from the leaves of putative transgenic and control plants as described previously (Murray & Thompson, 1980). PCR was performed using the following primers: 5′-TCTAGAATGGCGTCTGTTGCT-3′ and 5′-TTCTCAGCTATTTAGGAGC-TC-3′ for amplification of the 448 bp GUS fragment, and 5′-CCTGAACTCACGACG-3′ and 5′-AAGA-CCAAGGAGCATAT-3′ for amplification of the 804 bp HPT fragment. The PCR reaction contained 50 ng of genomic DNA as template, and amplification was performed by 1 min at 95 °C for preheating, 30 cycles of 30 s at 94 °C for denaturation, 1 min at 58 °C for annealing, and 1 min at 72 °C for elongation. A final extension was carried out by one cycle of 10 min at 72 °C. The amplified PCR products were analyzed by electrophoresis on a 1% (w/v) agarose gel.
For Southern blot analysis, genomic DNA (15 μg) from the leaves of PCR-positive plantlets was digested with HindIII. Digested DNA was separated by electrophoresis on a 0.8% agarose gel and blotted onto a nylon membrane (Hybond™-XL, Amersham Biosciences, Little Chalfont, U.K.) using the alkaline transfer method. Gene-specific probe for GUS was obtained by PCR amplification of a 0.4-kb fragment using pIG121Hm as the template with the previously described primers and conditions. Southern hybridization was performed according to our previous report (Lee et al., 2004).
Expression of the GUS transgene was also validated by Northern blot. Total RNA was isolated from the leaves of PCR-positive plants using the Plant RNeasy mini kit (Qiagen, CA, USA) following the manufacturer's instructions. To eliminate the residual genomic DNA, the RNA samples were treated with the RNase-free DNase I (Qiagen). Reverse transcription (RT)-PCR amplification was performed with the SuperScript First-Strand Synthesis System kit (Invitrogen, CA, USA) according to the manufacturer's instructions with oligo(dT)12–18 primers. The GUS transcript was identified by amplifying a 0.4 kb cDNA fragment. The primer set used was the same as that used for genomic PCR analysis described above. This limited cycle PCR consisted of 2 min at 94 °C; 20 cycles of 30 s at 94 °C, 30 s at 58 °C, and 1 min at 72 °C; and 1 cycle of 10 min at 72 °C.
Data recording and analysis
Effects of Agrobacterium strains, acetosyringone concentrations, and cocultivation period were optimized step-by-step. All data presented are the average of three repetitions. Analysis of variance (anova) followed by Duncan's multiple range test (DMRT) were carried out to test statistical difference presented in each table.
Results and Discussion
Agrobacterium-mediated transformation is a multifactor, complex interaction. Therefore, systematic optimization of such factors has proven to be of considerable importance for the establishment of successful transformation systems in monocot plants. A number of factors, such as plant genotype, explant types, strains of Agrobacterium, selection marker genes, selection agents, and various tissue culture conditions, are critical for grass transformation. Because there has been no report of transformation of reed using A. tumefaciens, it was considered important to investigate the effects of some common factors influencing T-DNA delivery. In the present study, various factors affecting gene delivery were determined primarily by assaying the activity of the GUS gene in calli. Then, the optimized conditions were used to develop stable transgenic plants. Leaves and stems of the putatively transformed plants were assessed by histochemical assays and molecular investigations for the confirmation of the stability of the transformation.
Callus induction and plant regeneration
Callus was induced on 1 mg l−1 2,4-D supplemented MS medium from mature caryopsis. After 6–8 weeks of culture, approximately 70% of calli were creamy in color, friable embryogenic in nature with numerous embroid-like structure on the surface (Fig. 2a). These calli were subcultured in a medium of similar composition and maintained for transformation experiments. Upon transferring to regeneration medium (MS supplemented with 1.0 mg l−1 of Dicamba), 78% of the explants developed into green shoots. Six callus lines tested in a replicate experiment, which were able to regenerate green shoots after 12 months in culture without producing any abnormal shoots (data not shown).
Factors influencing transformation efficiency in reed
Effect of Agrobacterium strains and vacuum infiltration
The bacterial strains play significant roles in transformation efficiency. To improve transformation rates, we tested three different A. tumefaciens strains for their virulence to seed-derived calli of reed. All three A. tumefaciens strains, LBA4404, EHA105, and GV3101, harbored the binary vector pIG121Hm. Our experiment showed that Agrobacterium strain EHA105 is the most virulent strain for reed calli compared with LBA4404 and GV3101, producing more hygromycin-resistant GUS-positive calli (Table 1). The frequency of transient GUS expression in the EHA105 strain-infected calli was approximately 2-fold higher than that of LBA4404 or GV3101. This difference might be due to the different virulence (vir) regions and different chromosomal backgrounds of the strains (Hellens et al., 2000). The EHA105 succinamopine strain contains a disarmed pEHA105 plasmid present in the C58 chromosomal background. The LBA4404 octopine strain contains a disarmed pAL4404 plasmid in the TiAch5 chromosomal background. Thus, the differences in the vir region and chromosomal backgrounds between A. tumefaciens EHA105 and LBA4404 may affect the range of plants susceptible to T-DNA transfer via their vir genes (Hood et al., 1993). Our results indicate that EHA105 (pIG121Hm) strains are more suitable for reed transformation.
Table 1. Effects of Agrobacterium tumefaciens strain on transient GUS expression in mature seed-derived calli. Data were recorded after 3 days of cocultivation. Values represent the mean of three independent experiments. Both the inoculation and cocultivation media were supplemented with 200 μm acetosyringone. The data represent the mean values and SE of three independent experiments. Different letters in the GUS expression frequency indicate statistically significant differences (P <0.05)
No. of inoculated calli
Number of GUS expressing calli
Percentage of GUS expressing calli
34.2 ± 1.9
56.3 ± 1.3a
22.6 ± 3.2
36.7 ± 2.4b
19.3 ± 0.38
31.7 ± 0.12c
In our experiment, prior to cocultivation, a negative pressure of 800 mbar created by a vacuum pump for 10 min in a vacuum chamber was used based on our earlier experiment with tall fescue (Lee et al., 2004). Vacuum creates a negative pressure environment that effectively increases Agrobacterium volatilization, a condition conducive to the transfer of a foreign gene into plant cells (Gu et al., 2008). However, incubation for longer times resulted in damage to the calli and decreased transformation rates (Lee et al., 2004).
Effect of hygromycin selection
The effects of various concentrations of hygromycin B were evaluated on calli and subsequent regeneration processes to determine the appropriate selection dose. Virtually, all calli lose embryogenic potential and fewer than 5% of the calli were able to survive in the presence of 25 mg l−1 hygromycin B. However, these calli did not show any sign of shoot regeneration after 1 month. In 6–8 weeks they died, therefore, discarded. No calli survived in the presence of 40 mg l−1 or higher concentrations of hygromycin B (data not shown). Therefore, 25 mg l−1 hygromycin B was used in subsequent transformation experiments.
Effect of acetosyringone on transformation efficiency
Among various factors, inclusion of acetosyringone is one of the key factors in grass transformation experiments. As the first report of reed transformation, we investigated the possible effects of acetosyringone on T-DNA delivery. The transfer of T-DNA is mediated by virulence genes, which form the vir region of the Ti plasmid (Klee et al., 1983). Transcription of the vir region is induced by various phenolic compounds, such as acetosyringone and alpha-hydroxy-acetosyringone, which may be released by wounding in dicot cells (Stachel et al., 1985). Because grasses are monocots, inclusion of these inducers is essential to support gene transfer during transformation (Stachel et al., 1985; Hiei et al., 1994). Using the infection period of 30 min and cocultivation period of 3 days, the effect of varying concentrations of acetosyringone (0–300 μm) in cocultivation medium was investigated. Sixty callus pieces were used for each treatment. The supplementation of acetosyringone to both inoculation and cocultivation media greatly increased transient GUS expression. The highest frequency (55%) of GUS-expressing calli was obtained when 200 μm acetosyringone was added to the medium (Table 2). Higher concentrations of acetosyringone resulted in the reduction in GUS expression, indicating a possible negative effect on T-DNA transfer. Approximately 15% of the calli showed GUS expression in the absence of acetosyringone, indicating that acetosyringone is not essential, but highly influential for Agrobacterium-mediated transformation of seed-derived embryogenic calli of reed. It has been shown that the growth of certain bacterial strains is inhibited by acetosyringone, and this growth inhibition was accompanied by the loss of virulence (Fortin et al., 1992). Some evidence also suggests that acetosyringone may suppress virulence in some strain/plant species interactions at a high concentration of acetosyringone (Godwin et al., 1991). By contrast, positive effects of acetosyringone on transformation efficiency have been demonstrated in other monocot species, such as rice (Zhao et al., 2011), maize (Ishida et al., 1996), barley (Trifonova et al., 2001), and wheat (He et al., 2010).
Table 2. Effects of acetosyringone on transient GUS expression in mature seed-derived calli. Calli were inoculated with EHA105/pIG121Hm, an infection time of 30 min and cocultivated for 3 days. Values represent the mean of three independent experiments. The data represent the mean values and SE of three independent experiments. Different letters in the GUS expression frequency indicate statistically significant differences (P <0.05)
Acetosyringone concentration (μm)
No. of inoculated calli
No. of GUS expressing calli
Percentage of GUS expressing calli
9.3 ± 0.88
15.0 ± 1.2d
16.8 ± 1.96
26.7 ± 2.7c
33.7 ± 3.37
55.0 ± 3.6a
21.5 ± 1.18
35.0 ± 1.7b
Effect of cocultivation period
In addition to the bacterial strain, reed transformation efficiency also differed with different cocultivation periods. The cocultivation period was further optimized for the EHA105 strain with 200 μm acetosyringone. Cocultivation periods of 1, 3, 5, and 7 days were tested (Table 3). GUS activity was observed in all cases. However, 3-day cocultivation resulted in a much higher numbers of GUS-positive calli (53.3%). Longer than 3 days of cocultivation resulted in decreased transient GUS expression. Cocultivation periods of 2–9 days have been shown as efficient for transformation of several grass species, including tall fescue (Lee et al., 2004), rice (Nishimura et al., 2007; Tyagi et al., 2007), barley (Wu et al., 1998), and zoysiagrass (Toyama et al., 2003; Li et al., 2010). Calli that were cocultivated for more than 3 days exhibited some GUS expression. However, they were negatively affected by over growth of Agrobacterium, and a tissue hypersensitivity response was observed. Thus, a 3-day cocultivation period is recommended for efficient transformation of reed.
Table 3. Effects of Cocultivation period on transient GUS expression in mature seed-derived calli. Calli were inoculated with EHA105 harboring pIG121Hm, an infection time of 30 min on cocultivation medium containing 200 μm of acetosyringone. Values represent the mean of three independent experiments. The data represent the mean values and SE of three independent experiments. Different letters in the GUS expression frequency indicate statistically significant differences (P <0.05)
No. of inoculated calli
No. of GUS expressing calli
Percentage of GUS expressing calli
9.1 ± 2.2
15.5 ± 4.2c
32.7 ± 3.6
53.3 ± 0.56a
24.7 ± 1.5
40.0 ± 2.50b
2.3 ± 0.26
3.30 ± 1.10d
Regeneration of stably transformed plants
Four- to five-week-old light yellowish, friable embryogenic calli (Fig. 2a) were used for transformation experiments. Using our optimized protocol, three independent stable transformation experiments, each including 82–100 embryogenic calli, were performed to obtain transgenic reed plants. The results are summarized in Table 4. Before selection, the cocultivated calli were transferred to resting medium as described in Materials and Methods to avoid the possible stress of Agrobacterium infection. After this recovery phase, the calli were subcultured in selection medium containing 25 mg l−1 hygromycin B in 2-week intervals. During the selection, a majority of the calli turned brown and died, whereas some resistant calli were observed after 3–4 weeks of selection. The hygromycin-resistant calli were then regenerated into plantlets. Of the 144 hygromycin-resistant calli, 9 produced multiple green shoots (Fig. 2c, Supplementary Fig. S1c; Table 4). All these transformed plants showed GUS expression in leaves and stems (Fig. 2e; Supplementary Fig. S1e). Upon transferring to rooting media containing hygromycin, the plantlets developed sufficient roots (Fig. 2d) and survived successfully after transplantation (Supplementary Fig. S1d).
Table 4. Stable transformation efficiency of reed by Agrobacterium tumefaciens. The calli were cocultivated with the strain EHA105/pIG121Hm supplemented with 200 μm of acetosyringone for 3 days. Leaves of regenerated plants were stained by GUS staining
No. of inoculated calli
No. of resistant calli to hygromycin B
No. of GUS positive plants
Transformation frequency (%)
48.0 ± 5.29
3.0 ± 0.58
3.47 ± 0.35
Molecular analysis of transgenic reed plants
Integration of the T-DNA into the putative transgenic plant genomes was confirmed by PCR and Southern blot analyses. PCR analyses were conducted for the HPT and GUS genes. All of the tested plants showed the expected band sizes of 0.8 and 0.45 kb, respectively (Figs. 3a and b). No bands were detected with either primer set in nontransgenic plants. To avoid any false positives, Southern blot analysis was carried out using the HPT gene as a probe. Genomic DNA from putative transgenic plants was digested with HindIII and hybridized with the HPT probe. All of the plants tested were confirmed to be transgenic (Fig. 3c). One to three copies of the transgene were estimated to integrate into the genome of the reed. No hybridized band was observed in the nontransformed control plants. The various hybridization patterns across the transgenic lines suggest that the plants were derived from independent transformation events and no escape was observed, indicating specificity of the selection pressure. Incorporation of a single or very few copy of transgenes is one of the advantages of Agrobacterium-mediated transformation. Similar results were also obtained in other grass species (Lee et al., 2004, 2008; Kim et al., 2010; Zhao et al., 2011). RT-PCR analysis also revealed high-level expression of GUS mRNA as the gene is driven by a constitutively expressing CaMV 35S promoter (Fig. 3d). On the contrary, no transcripts were detected in nontransgenic control plants.
In conclusion, we established the first stable Agrobacterium-mediated genetic transformation system for reed. Among the various conditions, the best transformation rate was achieved when the calli were cocultivated with A. tumefaciens strain EHA105 for 3 days in the presence of 200 μm acetosyringone using a vacuum infiltration method. Putative hygromycin-resistant calli were selected and regenerated, and several independent transgenic plants were recovered. PCR and Southern blot analyses confirmed that transgenes were integrated into the genome of the hygromycin-resistant plants and that the individual plants were from independent transformation events. Our work will be useful for developing transgenic reed with specific breeding objectives.
This work was supported by a grant from the KRIBB Research Initiative Program (KGM0531113), a grant from the Next-Generation BioGreen 21 Program (No. PJ008139), Rural Development Administration, and, in part, by the support of “Cooperative Research Program for Agricultural Science & Technology Development (Project No. PJ 00651401)”, RDA, Republic of Korea. YG Kim, SA Sharmin, and KH Kim are supported by scholarships, and I Alam is supported by a postdoctoral grant from the BK21 Program at Gyeongsang National University, Republic of Korea.