Leaf explants of a hybrid poplar Populus davidiana Dode × P. bollena Lauche were cultured in media with 0.1~0.5 mg/L BA and 0.01~0.1 mg/L NAA. There were no explants regenerated for 0.1 mg/L BA and 0.01–0.03 mg/L NAA (Table 1). Low regeneration frequencies were induced using BA (0.1 mg/L) and NAA (0.05, 0.08, or 0.1 mg/L) and BA (0.5 mg/L) with NAA (0.01, 0.03, 0.05, 0.08, or 0.1 mg/L). Ninety percent of the explants regeneration shoots when BA (0.3 mg/L) and NAA (0.08 mg/L) were used. Regenerated shoots from leaf explants were shown in Figure 1a. Confalonieri et al.[20] reported the positive effect of BA and NAA on shoot regeneration in Populus nigra. To obtain shoot regeneration, leaf discs were cultivated in Petri dishes on MS medium with 0.5 mg/L BA and 0.2 mg/L NAA.

The rooting experiments showed that new root formation and growth were promoted by NAA at any concentration, and root regeneration was 100%. However, NAA at 0.25 mg/L was the optimal rooting concentration because the growth status of roots was long and polyrhizal (Figure 1b). After the hardening process, 90% of the poplar plantlets survived acclimatization.

Previously, various media were used to different genotypes and explants for shoot regeneration and rooting. IBA was used to induce roots for leaf of aspen and zeatin was used for shoot regeneration [24]. Adventitious shoots regenerated in a woody plant medium (WPM) supplemented with TDZ (Thidiazuron) from aspen stem explants [16]. Ferreira et al.[14] reported that adventitious shoot regeneration of P. euphratica from organogenic nodules of leaf explants was achieved within a range of concentrations of a-naphtalenacetic acid and 6-benzylaminopurine, at a fixed 2:1 ratio. Shoots of Populus ciliata Wall were regenerated at high frequencies from explants grown on Murashige and Skoog (MS) medium supplemented with 0.5 mg/L kinetin and 0.2 mg/L indole-3-acetic acid (IAA). Regenerated shoots developed roots in MS medium supplemented with 0.1 mg/L IAA [7].

The effect of Agrobacterium concentration on transformation frequency was shown in Figure 2. At OD₆₀₀ = 0.6, a lower transformation frequency was observed. As OD₆₀₀ ranging from 0.8 to 1.0, a higher transformation frequency was observed with little contamination. At OD₆₀₀ = 1.2, the transformation frequency was also reduced because of a higher contamination frequency. At OD₆₀₀ ranging from 0.8 to 1.0, the suitable incubation time of the explants in the bacterial suspension ranged from 20 min to 30 min (Figure 2). Previous study by Confalonieri et al.[20] optimized the production of transgenic plantlets from different P. nigra clones, Agrobacterium concentration (7 × 10⁸; 1.2 × 10⁹ cells/mL) were used. The results show that the optimal procedure involved dipping of leaf discs into a bacterial suspension (7 × 10⁸ cells/mL) for 20 min. Previous research has shown that the transformation efficiency increases in the presence of AS in the co-cultivation medium [32,39]. However, in the current experiment, the presence of AS during co-cultivation did not increase further the transformation efficiency (data not shown).

In addition to Agrobacterium concentration, and infection time, the transformation frequency was also influenced by leaf age. We selected three different leaf age groups, 30-, 60-, and 90-day-old (Figure 3). The transformation frequency of the 30-day-old leaves was more than that of the 60-day-old and 90-day-old leaves (Figure 4). As the age of the sample leaf increased, the transformation frequency decreased (Figure 4d). The transformation frequency was 10% for the 90-day-old samples, while ~30% for the 30-day-old samples. Although several factors affecting transformation efficiency have been studied, there was no focus on explant age. A previous study by Smith and McCown [40] reported that juvenile sources of explant tissues have superior growth and regeneration potential compared with other sources. Civinova and Sladky [41] reported that juvenile explants have superior regeneration capacity and they are less lignified. Han et al.[22] reported that incompletely lignified explant sections of hybrid cottonwoods have also been efficiently transformed with Agrobacterium.

Different genotypes may respond differently to the optimal treatment conditions, and genetic transformation efficiency would be changed. According to published works, transformation frequency appeared to be different: 4%–6% for Populus trichocarpa, genotype Nisqually-1 [33], 16% for Populus tremuloides[27], and 20% for Populus nigra var. italica[26]. To improve its utility for functional genomics research, mutant library construction using the FOX hunting system was necessary for an efficient means for transformation and regeneration. Using hybrid poplar Populus davidiana Dode × P. bollena Lauche for transformation, the transformation frequency exceeded 30% by controlling the age of the leaf.

In order to understand the higher transformation ability in the younger leaf explants, the Agrobacterium invasion of various leaves were analyzed to confirm that leaf age was an important factor in Agrobacterium-mediated transformation. Figure 5a,b showed the magnified TEM image of the intercellular spaces with no Agrobacterium invasion. For 30-day-old leaf explants, some bacteria were scattered throughout the intercellular space (Figure 5c,d). For the 60-day-old leaf explants, the bacteria decreased slightly in number, and some of the intercellular space appeared empty (Figure 5e,f). For the 90-day-old leaf explants, almost all the intercellular space appeared clean and just a few bacteria were observed in the intercellular space (Figure 5g,h). These results demonstrated that there were Agrobacterium-mediated transgenic activities in the leaf explants, as well as the 30-day-old leaf explants contained more Agrobacterium than the 60-day-old and 90-day-old explants. The highest transformation efficiency was generated by the 30-day-old leaf explants (Figure 4d). The transformation frequencies were influenced by the amount of invasive bacteria.

Younger leaf explants may have a higher capability of shoot regeneration. However, in our TEM observations, the younger leaf explants had higher Agrobacterium infection. The invasion behavior of Agrobacterium in plant cells and their effects on genetic transformation is interesting. Agrobacterium invasion into plant cells can be affected by plant tissue age [42]. Younger poplar leaves were more susceptible to Agrobacterium-mediated transformation. As leaf age increased, the susceptibility to genetic transformation by Agrobacterium invasion decreased. A possible reason for the observed differential response of Agrobacterium invasion in different tissues was because of changes in the cell wall composition at different physiologic and developmental stages. As the plant cells mature, their cell wall composition changes so that Agrobacterium-binding sites were gradually lost [43].

We observed the genetic transformation further using a GFP label and Southern and Northern blot analysis. A plant transformation vector designed to express the MD–GFP fusion protein was used to detect the GFP signals in transgenic poplar. There were no GFP signals in the non-transgenic plants, shoots (Figure 6c,d), or roots (Figure 6g,h). However, the GFP signals from the shoots (Figure 6a,b), roots (Figure 6e,f), and leaf epidermis (Figure 6i,j) of the transgenic plants were detected. These results demonstrated that the pBI121-MD-GFP plasmid was introduced into hybrid poplar P. davidiana Dode × P. bollena Lauche.

Genomic DNA from seven of the different independent transgenic lines was subjected to Southern blot analysis. In line 1, non-specific DNA bands appeared as a result of the continuous distribution (line 1, Figure 7a), but all other lines contained the distinct bands. The banded MD–GFP gene was associated with high molecular weight invaded transgenic DNA(T-DNA), and produced the diverse copies as shown in line 2~8 (Figure 7a). The T-DNA appeared to be randomly integrated into the poplar genome.

For Northern blot analysis, all transgenic lines displayed a distinct band consistent with the predicted MD–GFP mRNA (lines 2~8, Figure 7b), whereas no hybridization signal was detected from the non-transgenic wild-type plant (line 1, Figure 7b). These results indicated that the MD–GFP gene was successfully integrated and expressed in the seven transgenic poplar lines 2~8 (Figure 7b).

Populus davidiana Dode × P. bollena Lauche leaf explants were obtained randomly from in vitro shoot cultures and then used for regeneration and transformation. Populus davidiana Dode × P. bollena Lauche was hybrid by Xinjiang poplar as the male parent and aspen as female parent. The plasmid pBI121-MD-GFP and DIG-labeled MD gene probe were kindly provided by Prof. Che (Northeast Agricultural University) and used as a marker for plant transformation in our laboratory. The plasmid pBI121-MD-GFP with the CaMV 35S promoter was transferred into Agrobacterium tumefaciens EHA105 using electroporation method (Bio-Rad, Micropluser, Hercules, CA, USA).

Murashige and Skoog (MS) medium [38] was supplemented with 6-benzylaminopurine (BA) at 0.1 mg/L to 0.5 mg/L and naphthaleneacetic acid (NAA) at 0.01 mg/L to 0.1 mg/L (Table 1). The leaves were cut into 1 cm × 1 cm blades and were cultured adaxial. The leaves were transferred to fresh regeneration medium at 10-day intervals. The regenerated shoots (3 cm to 5 cm in length) were excised and rooted on MS medium containing NAA at 0.1, 0.2, 0.25, 0.3, or 0.4 mg/L. All media included 3% (w/v) sucrose and 0.8% (w/v) agar. The pH of the media was adjusted to 5.8 after adding the plant growth regulation, and then autoclaved at 121 °C for 20 min.

After 1 to 2 months, the root formation was lignified, and the well-developed plantlets were removed from the media. The plantlets were gently washed under running tap water to remove the adhering media. For the hardening process, the plantlets were transferred into plastic cups containing a mixture of sterile compost. The regeneration experiments were designed with 10 Petri dishes for shoot regeneration and 10 triangular flasks for rooting per treatment, and each dish contained at least six leaf explants per treatment for shoot regeneration analysis, and each flask contained three shoots per treatment for rooting frequency analysis. Experiments were done in triplicate. The cultures were incubated at 25 °C ± 2 °C under a 16 h photoperiod (45 μmol m⁻² s⁻¹).