Introduction

Date palm (Phoenix dactylifera L.) is a woody monocot tree with high nutritional value and economic importance. This tree is originated from the Middle East and North Africa and selected cultivars have been propagated by offshoot through a long time. For several years, date-palm groves have been facing to a series of challenges such as insect and disease attacks, salinity and extreme heat. For example, a vascular fusariosis infection resulted from the fungus, Fusarium oxysporum fsp albedinis has destroyed a large number of palm trees in some North African countries and has continued to spread to the East palm groves [26]. Red palm weevil (Rhynchophorus ferrugineus) is also reported as one of the most serious pests of date palm worldwide. It is native to southern Asia, however, since early 1980s, it is highly spread throughout the Middle East, North Africa and Mediterranean region [17]. Therefore, to avoid of more destroying date palms; we must develop new cultivars through advanced breeding methods. Considering the long lifespan and dioecious nature of date palm, it is difficult to breed it though classical programmes. Some of the most important barriers towards genetic improvement of trees, such as their large size and long breeding protocols, can be avoided by the application of genetic transformation techniques. Plant biotechnology as an interdisciplinary science is able to provide new breeding goals in fruit trees such as resistance breeding against pathogens [36].

Applications of gene transformation are increasing rapidly effected by an extensive research effort and initial achievements have been mostly in the development of herbicide, insect and disease resistant crops.

Transformation protocols have been reported for a variety of tropical and subtropical fruit species based upon organogenic and embryogenic regeneration pathways in both dicots such as citruses [16, 24, 58], mango [10, 34, 38], papaya [9, 12, 68], avocado [3, 39], coffee [4, 53], rubber tree [66] and monocots such as banana [23, 27, 59–61, 63, 65, 67], pineapple [15, 19, 21, 42] and oil palm [6, 29, 30, 32, 48]. Until now, there is no protocol available on stable transformation of date palm. We have previously reported the effect of major parameters on efficient transient expression of uidA gene in callus tissue of date palm cv. Kabkab [45].

In this study, we report for the first time an optimized procedure on somatic embryo transformation of date palm using particle bombardment and subsequent regeneration of transgenic plantlets in which the integration and expression of transgenes are proven. In order to achieve this objective, the effects of several physical and biological parameters that are expected to have a strong impact on DNA delivery and to improve regeneration of stable transformants were assessed.

Materials and methods

Preparation of somatic embryos for bombardment

Plant materials consisted of offshoots of date palm (Phoenix dactylifera L. cv. Estamaran) obtained from date palm collection orchard, Date Palm and Tropical Fruit Research Institute, Ahwaz, Iran. Embryogenic calli were initiated from offshoot meristems using an induction medium containing MS salts and vitamins [46] supplemented with 100 mg/l of 2,4-d, 3 mg/l of 2ip, 40 mg/l of adenine, 30 g/l of sucrose and 3 g/l of activated charcoal incubated for 16–20 weeks in the dark. Somatic embryos were produced after transfer of the embryogenic calli to another media containing MS salts and vitamins supplemented with 10 mg/l of 2,4-d, 3 mg/l of 2ip, 40 mg/l of adenine and 20 mg/l glutamine for 30–45 days. After elimination of glutamine from this media, somatic embryos were germinated and produced shoots. The media used for root induction on the bases of shoots was MS basal salts supplemented with 1 mg/l NAA, 3 g/l activated charcoal and 30 g/l of sucrose. All samples were incubated at 26 °C and transferred to a fresh medium every 4 weeks.

Somatic embryos were collected from induction cultures and placed in the center of Petri dishes (15–20 embryos in a 2.5 cm diameter circle) containing the same media of embryo induction except for glutamine which was substituted by 0.4 M mannitol (as an osmoticum agent) for 24 h in the dark before bombardment. Partial desiccation was performed by removal of the plate covers in a laminar flow bench for 60 min before bombardment.

Plasmid DNA

The plasmids used in this experiment were consisted of pCAMBIA 3301(Cambia, Australia), harboring the uidA and bar gene cassettes; pBI221 (Clontech, USA), a high copy number plasmid harboring the uidA gene under control of CaMV 35 promoter; and pAct1-d [44], carrying the uidA gene with the 5′ region of the rice actin1 promoter.

Bombardment conditions

Major physical and biological parameters were examined for obtaining efficient bombardment conditions in somatic embryos. For optimization of the physical parameters, the following experiments were performed individually; acceleration pressure (1,100, 1,350 and 1,550 psi), distance from stopping screen to target tissue (6, 9 and 12 cm), vacuum pressure (24, 26 and 28 inHg), particle size (gold in diameters of 0.6, 1.0 and 1.6 µm) and bombardment numbers (single and double). Regarding biological and DNA parameters, four experiments were conducted including plasmid type (pCAMBIA 3301, pBI221 and pAct1-d), DNA concentration (2.5, 12.5 and 25 µg per bombardment), embryo size (5.0> and 5.0< mm) and osmoticum type (mannitol, sorbitol, sucrose and glucose). Other parameters such as mannitol concentration, particle type, distance between rupture disk and macrocarrier and particle-DNA coating agents were set up as previously reported for callus tissue bombardment [45]. DNA was precipitated onto microcarriers and delivered according to the procedure given for the PDS-1000/He system (BioRad, USA) with some modifications. While continuously vortexing, 50 µl of particle solution (prepared in 50 % glycerol), 10 µl of DNA, 25 µl of 0.1 M spermidine and 50 µl of 2.5 M CaCl2 were added and the final mixture was vortexed for 2 min. The microparticles were allowed to settle for 4 min and then pelleted by spinning for 3 s. After removal of the supernatant, the pellet was washed twice with 140 µl of 70 and 100 % ethanol, respectively. Then 82 µl of 100 % ethanol was added and the pellet was resuspended by vortexing. 12.5 µl of the DNA-coated microparticles suspension was loaded onto the center of a macrocarrier, desiccated and used for bombardment.

For stable transformation experiments, immature somatic embryos with ≤1 mm size were collected and arranged on the center of Petri dishes containing MS medium supplemented with 10 mg/l 2,4-d, 3.0 mg/l 2ip, 40 mg/l adenine, 3.0 g/l charcoal active, 30 g/l sucrose and 0.4 M mannitol, 24 h before bombardment. Microprojectile delivery was performed according to the best parameters obtained in the first experiment including, 1,350 psi acceleration pressure, 6 cm target distance, 28 inHg vacuum pressure and 0.6 µm gold particles coated with 2.5 µg pAct1-d plasmid DNA. The embryos remained for 4 days in this media then transferred to the same media except for mannitol and incubate at 27 °C and 16 h daylight condition. After 4–6 weeks, the embryos grew and developed leaves and roots.

GUS expression assay

Bombarded embryos were left on the same plates for 72–92 h before being checked for GUS expression by transferring them to the staining solution, followed by incubation for 24 h at 37 °C, as described previously [45]. Transient GUS activity was detected as the number of blue foci in each treatment under a Stemi 2000-C binocular microscope (Zeiss, Germany) and photographed.

After regeneration of plantlets from bombarded somatic embryos, cut leaf pieces from each transformant were stained as explained above for 4 days, then washed with 70 % EtoH and photographed.

DNA isolation and PCR analysis

Total genomic DNA was extracted for PCR analysis from green leaves of each regenerated plantlet according to the procedure described by Quenzar et al. [51]. In order to identify transformed plantlets, genomic DNA was amplified with GUS-specific primers. PCR analysis was performed in a 25 µl total reaction volume comprising 1 µl template DNA, 0.5 µl of each primer pair, 0.5 µl of dNTPs, 0.7 µl MgCl2, 2.5 µl 10 × buffer and 0.5 µl of Taq polymerase (All from Fermentas). Amplification was carried out in a Techne Flexigene Thermal Cycler (Cambridge, UK) with the following protocol: 1 cycle of denaturation at 95 °C for 4 min, followed by 30 cycles of (94 °C for 1 min, annealing at 55 °C for 1 min, extension at 72 °C for 1 min), and then final extension at 72 °C for 5 min. The primers used were GUS+2: 5′-GGT GGT CAG TCC CTT ATG TTA CG-3′ and GUS−4: 5′-CCG GCA TAG TTA AAG AAA TCA TG-3′. Amplified products were analyzed by electrophoresis in 0.8 % (w/v) agarose gels then stained with ethidium bromide and photographed. Transformed plantlets carrying the gus gene were screened by detection of the bands at the expected molecular size (520 bp).

Southern blot hybridization

Total genomic DNA was obtained from fresh leaves of transformed and nontransformed control plants. A 20 μg portion of DNA was digested with EcoRI or XbaI restriction enzymes at 37 °C and separated on a 0.8 % (w/v) agarose gel by electrophoresis and blotted to a positively charged nylon (Hybond N+, Amersham) and hybridized with probe. The probe for uidA was a 560 fragment prepared by PCR from pBI221 using GUS−4 and GUS+2 primers described above. Probe labeling and hybridization were performed according to the instructions for the DIG Probe DNA Labeling and Detection kit I (Roche Germany). Southern hybridization was performed according to method described by Kahrizi et al. [31]. Hybridizing bands corresponding to the uidA gene were detected using the DIG Detection Kit.

Data analysis

Data statistical analysis (with at least three replicates for each experiments of optimization) was carried out using one-way ANOVA (PROC-GLM program of SAS). Analysis of variance was conducted for each treatment. Means were separated at the probability level of 5 % with the Duncan’s multiple range test when a significant F ratio observed (P < 0.05). Significant differences were shown by different letters above the bars.

Results and discussion

Determination of the optimum bombardment condition

Acceleration pressure: Bombardment of somatic embryos with 1,350 psi acceleration pressure exhibited higher level of transient GUS expression with 77 ± 3 blue spots compared to 1,100 and 1,550 psi with 49 ± 4 and 68 ± 3 blue spots, respectively (Fig. 1a; Table 1). Higher levels of transient GUS expression has also been reported in banana immature embryos [11], Onobrychis viciifolia cotyledons [49] and peanut cotyledons [13] using a 1,300 or 1,350 psi acceleration pressures.

Fig. 1
figure 1

Effect of physical and biological parameters on transient expression of the gus gene in bombarded date palm somatic embryos. a helium pressure (psi); b distance from stopping screen to target tissue; c vacuum pressure; d bombardment number; e particle size; f Plasmid type; g DNA concentration; h Somatic embryo size and i Osmoticum type. Data represent the mean ± SD determined from at least three replicates and were analyzed using one-way ANOVA (PROC-GLM program of SAS). Different letters indicate significant differences at the level of P < 0.05 within a parameter

Full size image
Table 1 An overview of different bombardment conditions tested during each parameter experiment of optimization
Full size table

Distance of target tissue: The distance traveled by the microcarriers (6, 9 and 12 cm) form the microcarrier launch assembly to the target tissues significantly affected the rate of transient GUS expression in somatic embryos. A maximum of 152 ± 10 blue spots was observed when microcarriers traveled a distance of 6 cm, and the expression decreased to 91 ± 9 and 24 ± 8 blue spots at 9 and 12 cm travel distances, respectively (Fig. 1b; Table 1). Similar results were reported for banana [11, 59], wheat [50] and triticale [54]. Moreover, this distance has been used for transformation of oil palm immature embryo [40], peanut cotyledons [13] and pine cotyledons [43]. Ruma et al. [55] reported that 7.5 cm travel distance of microcarrier showed more GUS activity than 2.5, 5 and 10 cm in shoot tips, hypocotyls and cotyledons of tomato.

Vacuum pressure: A higher level of GUS activity assessed as counted blue spots (312 ± 12) was obtained when somatic embryos bombarded at 28 inHg vacuum pressure, compared to 24 and 26 inHg pressures with 87 ± 11 and 137 ± 12 blue spots, respectively (Fig. 1c; Table 1). Sreeramanan et al. [59] have indicated that 28 inHg resulted better transient expression of gfp and gus genes in banana single bud and corm slices than 26, 27 and 29 inHg pressures. Similar results were found for transient expression of gus gene in wheat inflorescence and scutellum by Rasco-Gaunt et al. [52]. It is also reported that bombarded peanut cotyledons with 28 inHg chamber vacuum pressure gives successfully transient transformation [13]. In Dendrobium protocorm-like-body (PLB), a vacuum pressure of 27 inHg was also reported to cause more effective expression [28]. Briza et al. [7] have obtained reasonable levels of transgene expression with vacuum pressure of 28 inHg in Norway spruce embryogenic tissue. 27 inHg was also used for coffee transformation [4].

Number of bombardments and particle size: In order to better coverage of explants surface with coated microprojectiles, it is possible to repeat bombardment at the same condition with changing the explants direction compared to previously bombardment state. In this study, when number of bombardment doubled, the transient expression of GUS was decreased (less than half) from 152 ± 10 blue spots in single bombardment to 56 ± 10 blue spots in double bombardments (Fig. 1d; Table 1). This may be due to the increased damage done to the target tissues cells particularly with higher helium pressures. Sreeramanan et al. [59] were found that single bombarded banana corm slices showed better results in transient expression of reporter genes gfp and gus than double and triplicate bombarded slices.

In our study, the bombardment of somatic embryos with 0.6 µm gold particle represented a significantly higher number of blue spots per shot (192 ± 8) than 1 and 1.6 µm with 45 ± 2 and 82 ± 4 blue spots, respectively (Fig. 1e; Table 1). Gold particles are often used by several investigators, since they have more uniform and round structure, non-toxic, inert and do not degrade DNA bonds [56]. According to the explant type and bombardment parameters, different microparticle sizes have been efficiently used. Kruse et al. [35] have reported that the 0.6 µm gold microparticles showed better results than 1.0 and 1.6 µm for transformation of Wolffia columbiana plant. Khalafalla et al. [33] also obtained an optimum transient expression conditions for sGFP in soybean with gold particle that were 0.6 µm in diameter. It is also reported that bombardment of rice with 0.6 µm gold microcarriers indicated a higher level of transient expression of sGFP compared to 1 µm particles [8]. Decreasing gold microparticle size from 1 µm resulted in significant increase in the rate of recovery of bialaphos-tolerant clones from maize type II callus [20].

Plasmid (promoter) type

Three constructs harboring the uidA gene under control of two promoters were studied. Statistical analysis of the GUS activities checked showed significant differences among the plasmids tested. The best results were obtained using plasmid pAct1-d, with the uidA gene under control of the Act1 promoter, a constitutive promoter coding actin in rice [30]. This promoter increased GUS activity about fivefold over the double CaMV 35S promoter in pCAMBIA3301 and more than 12-fold over a CaMV 35S promoter in pBI221 construct. There was an average of 88 ± 7, 18 ± 2 and 7 ± 0 blue spots scored for pAct1-d, pCAMBIA 3301 and pBI221, respectively (Fig. 1f; Table 1). We concluded that CaMV 35S promoter may not work well in date palm tissues. Similarly, when three different promoters CaMV 35S, maize ubi1 and rice Act1 evaluated in barley, it was appeared that CaMV 35S promoter had the lowest level of transient expression of GUS while the highest activity was observed for the rice Act1 promoter [2]. Gallo-Meagher and Irvine [22] have also observed an increased level of transient GUS activity in sugarcane when using the Act1 promoter in comparison with the CaMV 35S promoter. However, the Act1 promoter has indicated the same level of activity as the CaMV 35S promoter in some Brassica genotypes [64]. In another study, large differences were demonstrated between Pinus nigra cotyledonary explants bombarded with the constructs having Act1 and CaMV 35S promoters [41]. Basu et al. [5] have reported a low level of GUS activity using the constitutive CaMV 35S promoter in Agrostis palustris in comparison with the Ubiquitin promoters of rice and corn. Schenk et al. [57] have reported that bombardment of sorghum and maize leaves with various promoters indicates that the CaMV 35S promoter has the lowest level of reporter proteins expression in both species as compared to those of the Ubiquitin and Act1 promoters. When banana transformed with gus gene fused with different promoters including CaMV 35S, ubi1 and BBTV through particle bombardment, it was a significant level of transient expression of GUS observed with ubi1 promoter compared to two other promoters [14]. The considerable efficiency of the actin promoter in controlling uidA gene expression as compared to CaMV 35S, suggests that Act1 can be practiced to drive the expression of economically important genes during date palm genetic transformation projects.

DNA concentration: When different concentrations of DNA plasmid (2.5, 12.5 and 25 µg per a shot in final volume of 12 µl) were bombarded, there were no significant differences observed among the three concentrations, but 12.5 µg per shot with 66 ± 11 blue spots showed superior effects compared to 2.5 and 25 µg per shot with 50 ± 6 and 51 ± 8 blue foci, respectively (Fig. 1g; Table 1). Ruma et al. [55] found that coating of microcarriers with 18 µl plasmid DNA showed more GUS activity than 6, 12 and 24 µl in tomato. Furthermore, coating 50 µl of particle solution by 10 µl of DNA with 1 µg/µl concentration was successfully used for transformation of sweet potato by Lawton et al. [37].

Embryo size and Osmoticum type: In order to define the most suitable embryo size in date palm for transformation using the particle delivery system, somatic embryos were screened before bombardment in two groups, small (less than 5 mm long) and large (more than 5 mm long) and evaluated for their GUS activity. There was a significant difference observed between the two embryo groups. As shown in Fig. 1h, Table 1 and Fig 2, the small embryos indicated the maximum average number of blue spots per shot (143 ± 8) compared to large embryos (38 ± 1). Similar results were observed for bombarded banana explants by Sreeramanan et al. [59]. They found that bombardment of 3 mm banana buds showed higher transient expression of GUS rather than 5 and 10 mm ones. Whereas, in oil palm there was no significant difference observed among bombarded embryos in three sizes <3, 3–6 and >6 mm [1].

Fig. 2
figure 2

Tranient GUS expression in different embryo sizes of date palm. L large embryos (more than 5 mm), S small embryo (less than 5 mm). Scale bar represents 3 mm

Full size image

High osmotic preconditioning (plasmolysis) treatment of target tissues increase transient and stable transformation by minimizing cytoplasm leakage from target cells [52]. Effect of osmoticum (mannitol alone or with sorbitol) on increasing efficiency of transient and stable transformation has been investigated by several researchers in various species such as Barely [25], peanut [13], citrus [18], peach [47], coffee [4] and Impatiens balsamina [62]. In this study, 24 h prior to bombardment, the somatic embryos were transferred to the medium containing four types of osmoticum agent including sorbitol, mannitol, glucose and sucrose, each at 0.4 M. After bombardment, the embryos were left on these media for 4 days, followed by GUS histochemical staining. The results indicated significant differences among osmoticums. Mannitol with 318 ± 12 blue spots was shown to be more effective than sorbitol (70 ± 7), sucrose (2 ± 0) and glucose (2 ± 0) with respect to transient expression of the reporter gene (Fig. 1i; Table 1).

Regeneration and analysis of transgenic plants

Eight weeks after bombardment of somatic embryos, the plantlets with extended leaves and root system were scored and subcultured to a fresh media. The results obtained in this study showed that 93.3 % of bombarded somatic embryos were able to recover and grew normally to produce healthy plantlets (Fig. 3).

Fig. 3
figure 3

Plantlets of date palm cv. Estamaran regenerated after bombardment of somatic embryos with pAct1-d construct harboring uidA gene under control of Act1 promoter. a regenerated plantlets from recovered embryos 8 weeks after bombardment; b extended leaves and root systems in regenerated transformed plantlets

Full size image

PCR analysis was performed on all of the plantlets regenerated from embryos transformed with pAct1-d, confirming the stable integration of gus transgene into the genome. Using the primers GUS+2 and GUS−4, the expected 520 bp band was obtained from the gDNA of young leaves of transformants, whereas no amplification was detected in nontransformed tissues. Out of 55 PCR tested plants, only 10 samples were positive (Fig. 4) indicating a transformation frequency of 18 % in this protocol.

Fig. 4
figure 4

PCR analysis of regenerated plantlets of date palm bombarded with pAct1-D. Genomic DNA was amplified with GUS+2 and GUS−4 specific primers. M 1 kb DNA molecular marker (Fermentas), l negative control of PCR reaction (water), 2 positive control (vector plasmid), 3 negative control for template DNA (untransformed plants), 4–17 samples from plantlet regenerated after bombardment of somatic embryos

Full size image

When assayed for GUS expression, the uidA gene driven by rice Act1 promoter resulted in strong GUS expression observed as a blue uniform color in the transformed plants leaf pieces, whereas nontransformed plants did not present any blue color (Fig. 5). The frequency of transformants calculated through GUS histochemical assay was 16 %.

Fig. 5
figure 5

Histochemical GUS staining of transformed and nontransformed plantlets. The uidA gene driven by rice actin1 promoter resulted in strong GUS activity in leaves (transformed) whereas no GUS activity was detected in tissues of wild type plantlets (nontransformed)

Full size image

Four GUS assay positive plantlets which confirmed by PCR and a non-transformed plantlet were selected and analyzed by southern blotting. To determine the copy number of GUS inserts, the uidA probe was used to hybridize genomic DNA from leaves of date palm plantlets digested by XbaI or EcoRI enzymes. 1–2 hybridizing bands were recognized which means that the plantlets contained single or double GUS. This confirmed that integration was occurred in transformed lines recognized by PCR analysis. Three transformed lines exhibited two hybridizing bands and one transformed line exhibited a single GUS copy. No transgene insertion or copy number was detected in nontransformed control (Fig. 6). After digestion and hybridization, the DNA of the transformed plants presented hybridizing bands at 1.7, 2.2, 2.4, 2.5, 4.5 and 8.1 kb. The results confirmed integration of transgenes into the genome of all the GUS-positive lines.

Fig. 6
figure 6

Southern blot analysis of transformed and nontransformed date palm plantlets for identification of uidA gene presence. M 1 kb molecular marker (Fermentas), P pBI221, C DNA from nontransgenic date palm plant; 1–4, represent hybridization pattern after digestion of genomic DNA from particle bombarded transgenic lines

Full size image

In conclusion, this investigation demonstrates that successful delivery of transgenes under control of Act1 promoter in somatic embryos of date palm is achievable using the established optimized protocol for microprojectile bombardment. This protocol which is presented for the first time in date palm will facilitate development of superior cultivars through introduction of useful genes. Similar to other monocotyledonous species, the application of the biolistic technology for direct transfer of DNA into the cells has facilitated date palm transformation. Particle bombardment delivery could potentially be used to introduce economically important traits such as resistance to disease and pests and quality improvement in the date palm.