To test and optimize the agroinfiltration system, a wide range of hemp tissues were tested using GUS and GFP as reporter genes. Successful transient GUS expression was observed in all the tested tissue types: leaves, cotyledons, hypocotyls, root segments, nodal parts, and 2- to 3-week-old callus cultures (Figure 1A–J). Among all the tested tissue types, 3-week-old callus explants showed the highest transient GUS expression percentage (81.56%) with an average of 49.22 blue spots per explant (Table 1). At the same time, transient GFP fluorescence was detected in leaves, petioles, root segments, and callus explants (Table 2). In comparison, 3-week-old fully expanded wounded leaves explants had the highest percentage (59.44%) of GFP expression with an average of 166.44 GFP spots per explant compared to an average of 163.89 blue spots (Table 1 and Table 2). A cluster of multiple cells exhibiting GFP protein green fluorescence and chloroplast autofluorescence was captured using a Fluorescence Stereo Microscope under a GFP filter in the 3-week-old fully expanded leaves (Figure 2A–C). Strong expression of the GFP gene in primary mesophyll cells, including vascular structures, was observed 3 days after co-cultivation (Figure 2C–E). In the case of both GUS and GFP expression, 3-week-old leaf tissues with a surface area of 99.04 mm² (Table 1 and Table 2 and Figure 3C) had the greatest number of blue or green fluorescence spots per explant (211.00 and 201.67, respectively). On the other hand, 1-week-old leaves with a surface area of 30.82 mm² and 2-week-old leaves with a surface area of 55.85 mm² had a lower number of blue foci and green fluorescence spots, as observed using a Stereo Zoom Microscope (Table 1 and Table 2 and Figure 3A,B). The lowest GUS expression was observed in hypocotyl explants (40.78%) while the lowest GFP expression was observed in root tissues (41.22%) (Table 1 and Table 2 and Figure 1F).

The transient expression frequency for GUS and GFP was generally higher in 3-week-old fully expanded leaves compared to 1-week-old and 2-week-old leaves (Table 1 and Table 2 and Figure 3A–C). One caveat of quantifying transient expression efficiency by counting the average number of spots per explant is that pretreatment of the explant tissue could influence the number of spots (Figure 2A–C). On average, 201 GFP-expressing cell clusters and 211 blue spots were observed.

The bacterial density of the Agrobacterium culture during co-cultivation is an important factor that not only affects the transient transformation efficiency but also the regeneration potential of the explant. In this study, five different optical densities at OD_600nm (0.2, 0.4, 0.6, 0.8, and 1.0) were tested for GUS and GFP activity in the 3-week-old leaves. At OD_600nm = 0.4, the highest percentage of GUS- and GFP-expressing explants were observed (89.67% and 82.78%, respectively) and the highest number of GFP and blue spots (156.70 and 154.33, respectively) after 3 days of co-cultivation (Figure 4 and Figure 5). As the bacterial density increased in the co-cultivation media, the transformation efficiency decreased for both the GUS and GFP genes, with the lowest values observed at OD_600nm = 1.0 (Figure 4 and Figure 5). The appropriate bacterial density also depended on the type of explant used for the transformation; a higher density might cause an overgrowth of the bacteria, resulting in necrosis or ineffective recovery of the explants. At the same time, a lower optical density might result in insufficient bacterial cells to infect most of the cells, leading to a lower percentage of T-DNA transfer.

In Agrobacterium-mediated transformation, wounding of the explant is necessary for efficient transformation. Generally, wounds provide Agrobacterium access to the plant cells. In this study, three different types of wounding were carried out using a surgical needle, a stainless steel metal screen, and sonication. The highest GUS and GFP expression was observed in 3-week-old leaf explants when wounded with a metal screen compared to a needle and sonication (Figure 6C,D).

Based on the observation 68.11% of leaf explants wounded using a screen were expressing GUS while 74.11% of leaf explants expressed the GFP gene. The presence of phenolic compounds, such as acetosyringone, after wounding, helps to activate the vir-inducing components, increasing the transformation efficiency (Figure 7 and Figure 8). However, excess wounding might affect tissue recovery. This may cause poor regeneration of the explant, and decrease the efficiency of obtaining a stable transgenic plant. Previous studies have stated that different wounding methods, such as sonication, might be needed for different types of explants such as embryos, immature cotyledons, apical meristems, calli, and hypocotyls [31].

Acetosyringone has been shown to enhance the transient expression of GUS in different species due to the activation of the viral gene [32,33]. The current results convincingly illustrate that the addition of acetosyringone improved the transformation efficiency in leaf explants compared to the control. The results of one study also suggest that acetosyringone augmentation in the co-cultivation medium can considerably increase the transformation efficiency and is necessary for the successful transformation of chickpeas [34]. This study’s results agree with others who demonstrated the successful application of acetosyringone at a concentration of 100 μM in enhancing the transformation efficiency in cowpeas [35] and cucumbers [36].

As hemp is considered recalcitrant to transformation, we also tested the effect of acetosyringone (AS, phenolic inducer) on transient expression efficiency. Several studies have reported that phenolic inducers can enhance the ability of Agrobacterium to infect and transform host plants. In genetic transformation, acetosyringone is necessary for the activation of the Agrobacterium virulence machinery. It induces virulence genes and facilitates the transfer of T-DNA from a plasmid into the plant cell and its integration into the host genome. The effect of six different concentrations of acetosyringone (0, 50, 11, 150, 200, and 250 µM) was investigated for co-cultivating leaf explants for transformation by subjecting 3-week-old leaves to a co-cultivation lasting 3 days. It was observed that acetosyringone at 150 µM resulted in the highest percentage of explants expressing both GUS and GFP (78.3% and 71.60%, respectively), while the 200 µM concentration resulted in the highest number of GUS and GFP spots per explant (152.78 and 154.56, respectively). The transient expression efficiency decreased drastically for both GUS (38.10%) and GFP (37.30%) at 250 µM (Figure 9 and Figure 10).

The immersion period of the explants in the Agrobacterium suspension similarly affects the transformation efficiency. A prolonged immersion time might lead to tissue necrosis, browning, and bacterial overgrowth. At the same time, a reduced time may be insufficient for T-DNA transfer. In this study, six different immersion periods (0, 10, 20, 30, 40, 50, and 60 min) were tested and an immersion period of 30 min in the Agrobacterium suspension resulted in the highest percentage expression for GUS (79.0%) and GFP (60.70%) with an average of 119.2 GFP- and 132.50 GUS-expressing events (Figure 11 and Figure 12). The lowest efficiency of gene transfer was observed after a 10-min immersion period, possibly due to insufficient time for Agrobacterium to infect a large number of cells. The gradual reduction in transient expression efficiency over time can be attributed to the browning of the leaf and or tissue death. Similar to optical density, immersion time may is also dependent on the type of tissue or explant. Young and small tissues might require less immersion time than tissues such as nodal sections.

Calcium is a macronutrient that is often present in plant media and cell walls and acts as a signaling molecule for changes in cell composition. In this study, five different concentrations of calcium chloride (0.2, 0.5, 0.75, 1, and 1.5 µM) were tested and the highest percentages of GUS- and GFP-expressing explants were observed at a concentration of 0.5 µM. At 0.5 µM, 76.5% of explants were expressing GUS and GFP with approximately 133 spots per explant (Figure 13 and Figure 14). As the concentration of calcium chloride increased, the efficiency of the transient transformation decreased. It has been reported that exogenous calcium in the media can suppress the expression of virulence genes in bacteria [37]. The high amount of exogenous calcium might also be responsible for the fortification of damaged cell walls, thereby limiting Agrobacterium infection.

Similarly, silver nitrate has been shown to affect gene transfer efficiencies in Agrobacterium-mediated transformation. In this study, five different concentrations of silver nitrate were tested and a concentration of 60 µM resulted in the highest transformation efficiency. The highest expression of GUS was 79.44% of explants and for GFP, 79.75% of explants showed gene expression. As the concentration of silver nitrate was increased (120 µM), the gene transfer efficiency decreased for both GUS and GFP (43.9% and 43.25%, respectively) (Figure 15 and Figure 16). Silver nitrate is a known antioxidant and ethylene inhibitor that affects cell division. The phenomenon of the gradual increase in transformation efficiency at low concentrations and decrease at high concentrations has been reported in several plants like apple, banana, Phalaenopsis violacea, and Primula vulgaris [38,39,40,41]. A lower transformation efficiency at higher concentrations can be attributed to the antimicrobial properties of silver ions, as it has been previously used to prevent Agrobacterium overgrowth and to prevent the browning of explants [41].

The C. sativa ‘Joey’ variety is a dual-purpose monecious variety grown for its grain and fiber. It has gone through several breeding generations with some heterogeneity due to its monecious characteristics. The flowering period is typically 55–70 days with the CBD content ranging from 1 to 2%. This variety was developed in Canada and was purchased from Parkland Industrial Hemp Growers Coop, Ltd. (3-126 Main St. N., Dauphin, MB R7N 1C2, Canada). The hemp strain ‘Joey’ was grown following the approved guidelines for industrial hemp provided by the Pennsylvania Department of Agriculture Bureau of Plant Industry under the regulated permits IH-16-P-2017 and IH-17-P-2017. The cultivar ‘Joey’ is an established line that was produced by several rounds of breeding. There is some phenotypic heterogeneity expected in the seedlings derived from the clones of ‘Joey’ in their response to the various parameters that were evaluated in this study. The seeds were harvested from mature plants under controlled conditions in the greenhouse at the University of Pennsylvania campus facilities (Middletown, PA, USA) [2,7,25]. The research experiments were conducted at Fort Valley State University and were approved by the Georgia Department of Agriculture (GDA), and followed the State of Georgia by House Bill 213 Georgia Hemp Farming Act, SB 195 and SB 324, and the University System of Georgia’s institutional policies and procedures. The mature seeds were rinsed with water for 20 min and surface-sterilized in an aqueous 30% Clorox (Clorox pro, 8.25% sodium hypochlorite) solution containing 3 drops of Tween-20 for 30 min on a shaker, and then rinsed 3 times, 5 min each, with sterile water. The sterilized seeds were soaked in sterile 1% hydrogen peroxide overnight to soften the seed coat and increase the rate of seed germination.

After soaking, the outer dark brown hard seed coats and inner dark green membranous coats were removed under sterile conditions with the aid of pointed tweezers and a surgical knife. Care was taken not to cause any injury to the naked seeds (including cotyledons, radicles, and embryos). The sterilized seeds were then germinated on a half-strength Murashige and Skoog’s (MS) basal medium [48] with vitamins containing 3% (w/v) sucrose, 10 mg L⁻¹ thiamine, 100 mg L⁻¹ myoinositol (pH 5.8), and solidified with 0.3% (w/v) Gelrite^®. All the cultures were incubated at 26 ± 2 °C under a 16/8 h (light/dark) photoperiod with light provided by cool-white, fluorescent lamps at an intensity of 31–35 µmol m⁻² s⁻¹. The explants from one- to three-week-old seedlings were used for genetic transformation.

For the transformation experiment, six different types of explants from three-week-old seedlings were used, such as leaves, cotyledons, hypocotyls, root segments, nodal parts, and leaf-derived embryogenic calli. Leaf explants at three developmental stages (1-, 2-, and 3-week-old fully expanded leaf from seedling) were used to investigate the effect of the explant and developmental stages on the transient transformation efficiency. In a separate experiment, the fully expanded 3-week-old leaves were cut into sections (1 cm each) and used as explants for embryogenic callus induction on Gamborg B5 basal salt medium [49] supplemented with 3.0% maltose, 5.1 mM calcium chloride, 5.5 mM glutamine, 32.5 µM glutathione, 95.1 µM serine, 7.4 µM adenine, 4.5 µM 2,4-D, 0.9 µM kinetin, and 0.3% Gelrite [50]. The pH of the medium was adjusted to 5.80 with 1 N NaOH or 1.0 N HCl and then autoclaved at 100 kPa (121 °C) for 20 min. After 4 weeks of culturing in the dark at 26 ± 2 °C, the effect of each treatment on callus induction was surveyed. All cultures were incubated at 26 ± 2 °C under a 16/8 h (light/dark) photoperiod with light provided by cool-white, fluorescent lamps at an intensity of 31 to 35 mol m⁻² s⁻¹.

A. tumefaciens strain GV3101 transformed with plasmid pCambia 1304 expressing both β-glucuronidase (GUS) and Green Fluorescent Protein (GFP) genes were used for transformation. The GFP-GUS gene fusion was under the control of a Cauliflower mosaic virus (CaMV) 35S promoter and NOS terminator. The T-DNA region of the plasmid also contained the hygromycin B phosphotransferase gene driven by the CaMV 35S promoter. The pCambia 1304 plasmid has been routinely used to demonstrate or establish transient expression in several plant species such as Arabidopsis [51], several medicinal plants [52], rice [53], tobacco plant [54], and Dendrobium Sonia [55], and tobacco leaves [56]. The vector pCambia 1300 was used as a control. A single Agrobacterium colony derived from the stock culture described by Dan et al. [57] was cultured in liquid Luria Bertani (LB) [58] medium containing 50 mg L⁻¹ kanamycin and was grown on a shaker at 120 rpm at a temperature of 28 °C for 16 h. The bacterium was pelleted at 5000× g for 10 min, washed in antibiotic-free LB medium, re-pelleted, and re-suspended in fresh LB medium twice. The control consisted of Agrobacterium which was not subcultured and was used directly for transformation after centrifugation and resuspension in liquid LB medium. Once the preferred OD was achieved, 100 μM acetosyringone was added to the bacterial suspension culture to increase the virulence. Explants of different ages were transferred into the Agrobacterium suspension for transformation. The explants were immersed in the Agrobacterium cell suspension for 30 min at 26 °C and gently shaken on a rotary shaker at 50 rpm to ensure that the entire explants were fully submerged for bacterial adherence. After different immersion periods (0, 10, 20, 30, 40, or 60 min), the explants were subsequently washed with sterile distilled water containing 200 mg L⁻¹ cefotaxime to remove any excess bacteria. The infected explants were blot-dried using sterile Whatman No. 1 filter paper for 5 min and then the explants were transferred to the co-cultivation medium. The explants were then inoculated onto the pre-selection medium (MS containing 0.23 mg L⁻¹ BA, 50 mg L⁻¹ cefotaxime, and 100 mg L⁻¹ carbenicillin) and incubated at 26 ± 2 °C for 3 days in the dark for co-cultivation. After 3 days, the explants were washed 4 to 5 times with sterile distilled water, immersed in a 200 mg L⁻¹ carbenicillin solution for 5 to 10 min, and blotted dry using sterilized filter paper.

To assess the factors affecting transformation efficiency, different treatments were analyzed. All the experiments were conducted with at least 3 replicates with each containing 25 to 30 samples for optimizing the individual parameter. The parameters included explants derived from 3-week-old seedlings such as a leaf, cotyledon, hypocotyl, root, nodal part, and leaf-derived 2- and 3-week-old embryogenic callus, age of explant (1- to 3-week-old fully expanded leaf), immersion period (0, 10, 20, 30, 40, and 60 min), acetosyringone concentration (0, 50, 11, 150, 200, and 250 µM) during co-cultivation, bacterial density at 600 nm (0.2, 0.4, 0.6, 0.8 and 1.0), calcium chloride concentration (0, 0.2, 0.5, 0.75, 1.0 and 1.5 µM), and silver nitrate concentration (0, 30, 60, 90, and 120 µM).

To test the influence of different wounding treatments, similar fully expanded 3-week-old leaf explants were gently stabbed 4 to 5 times using a sterile hypodermic needle (23 gauge), and/or a 60 µm stainless steel screen was pressed over the surface of the leaf. For the sonication treatment, the explants were immersed in 30 mL of Agrobacterium suspension and sonicated for 30 s twice with a 15 s interval between each treatment [Fisherbrand Model 120 Sonic Dismenbrator]. After wounding, the explants were transferred into a fresh Agrobacterium suspension and incubated at 30 °C for a further 30 min for the infection time with occasional gentle agitation. After co-cultivation, the explants were washed 3 to 4 times with liquid MS medium and blotted dry on sterile filter paper. Then, the explants were placed on culture media for three days in the dark at 25 ± 2 °C.

As a control, the explants were directly placed on the co-cultivation medium without being immersed in the Agrobacterium suspension. At the end of the co-cultivation period, the histochemical GUS assay and fluorescence microscopy were performed. A total of 25 to 30 explants were used for each treatment, and the experiments were repeated 3 to 4 times. All the parameters were evaluated based on observations of the average percentage of both GUS- and GFP-expressing explants and the number of GUS- and GFP-expressing spots per explant.

The method for the histochemical GUS assay was adopted from [59] with some modifications. The explants were incubated overnight at 37 ± 2 °C in a buffer containing 0.5 mg mL⁻¹ 5-bromo-4-chloro-3-indolyl-β-D-glucuronic acid (X-Gluc), 0.1 M sodium phosphate buffer (pH 7.0), 0.1 M potassium ferricyanide, 0.1 M potassium ferrocyanide, and 0.1% (v/v) Triton X-100. The X-Glue solution was removed, and the explants were washed with double distilled water followed by a wash with 70% ethanol. To remove the chlorophyll and other pigments, a mixture of acetone and methanol (1:3) was added and explants were incubated at 4 °C for 1 h before being visually inspected. After the tissue became transparent, the explants were washed with distilled water twice and stored in 50% glycerol [60,61]. Green fluorescent protein expression in leaves, hypocotyl segments, calli, and developing embryos was visualized using an Olympus SZX12 Stereo Fluorescence Microscope equipped with an HBO 100 W mercury bulb light mounted with a long pass GFP filter and DP74 Cooled Color camera (Olympus America Inc., Melville, NY, USA). The GFP filter had excitation wavelengths of 460–480 nm and emission wavelengths of 495–540 nm, and the GFPA filter for the separation of GFP and blue-excited fluorophores had excitation wavelengths of 460–490 nm and emission wavelengths of 510–550 nm. Photographs were taken using an Olympus SZX12 Automatic Exposure Photomicrographic System.

To optimize the individual parameter, each experiment contained at least 25 to 30 samples, and each experiment was replicated three times (n = 75). The experimental data were statistically analyzed using analysis of variance (ANOVA). The treatment means were separated by using the Tukey–Kramer honestly significance difference (HSD) test at p ≤ 0.05 [62].