Evaluation of Parameters Affecting Agrobacterium-Mediated Transient Gene Expression in Industrial Hemp (Cannabis sativa L.)

Abstract

Industrial hemp Cannabis sativa L. is an economically important crop mostly grown for its fiber, oil, and seeds. Due to its increasing applications in the pharmaceutical industry and a lack of knowledge of gene functions in cannabinoid biosynthesis pathways, developing an efficient transformation platform for the genetic engineering of industrial hemp has become necessary to enable functional genomic and industrial application studies. A critical step in the development of Agrobacterium tumefaciens-mediated transformation in the hemp genus is the establishment of optimal conditions for T-DNA gene delivery into different explants from which whole plantlets can be regenerated. As a first step in the development of a successful Agrobacterium tumefaciens-mediated transformation method for hemp gene editing, the factors influencing the successful T-DNA integration and expression (as measured by transient β-glucuronidase (GUS) and Green Florescent Protein (GFP) expression) were investigated. In this study, the parameters for an agroinfiltration system in hemp, which applies to the stable transformation method, were optimized. In the present study, we tested different explants, such as 1- to 3-week-old leaves, cotyledons, hypocotyls, root segments, nodal parts, and 2- to 3-week-old leaf-derived calli. We observed that the 3-week-old leaves were the best explant for transient gene expression. Fully expanded 2- to 3-week-old leaf explants, in combination with 30 min of immersion time, 60 µM silver nitrate, 0.5 µM calcium chloride, 150 µM natural phenolic compound acetosyringone, and a bacterial density of OD_600nm = 0.4 resulted in the highest GUS and GFP expression. The improved method of genetic transformation established in the present study will be useful for the introduction of foreign genes of interest, using the latest technologies such as genome editing, and studying gene functions that regulate secondary metabolites in hemp.

Keywords: Agrobacterium tumefaciens; Cannabis; GFP; GUS; genetic transformation; hemp; transient.

Figure 1

Histochemical GUS assays were performed on different transformed explants of Cannabis sativa L. Explants were transformed with the binary vector pCAMBIA1304 via Agrobacterium tumefaciens (strain GV3101). After 3 days of co-cultivation, histochemical staining for the GUS activity was performed to measure transient gene expression. The dark-blue-stained tissue indicates the presence of GUS activity. The experiments were repeated three times. (A) Three-week-old fully expanded wounded leaves using screen; (B) high-magnification (10 × 20; scale bar = 1.0 mm) image showing dark blue staining in three-week-old leaf tissues; (C) β-glucuronidase (GUS) gene expression in a single cell and multiple cells in leaf tissues. The arrow indicates the expression of the GUS gene in single to multiple cells in leaf tissues. (magnification = 10 × 100; scale bar = 100 µm); (D) seven-day-old seedling cotyledon with transient β-glucuronidase expression, indicated by dark blue staining at various levels (magnification = 10 × 5); (E) Seven to fourteen-day-old seedling hypocotyl expressing the β-glucuronidase gene; (F) the GUS gene expression in fourteen -day-old seedling roots; (G) expression of the β-glucuronidase gene in fourteen -day-old seedling nodal part; (H) two-week-old leaf-derived embryogenic callus pieces expressing the β-glucuronidase gene; (I) three-week-old leaf-derived callus culture with evident GUS-carrying globular embryo development; (J) three-week-old leaf-derived globular-shaped embryo under high magnification showing β-glucuronidase gene expression.

Figure 2

GFP protein fluorescence in Cannabis sativa L. wounded fully expanded 3-week-old leaf explants and leaf-derived embryogenic calli co-cultivated for 3 days with Agrobacterium suspension at final OD_600nm of 0.6 and washed 5× times with Murashige and Skoog washing medium before capturing the images using an Olympus SZX12 Stereo Fluorescence Microscope (Olympus America Inc., Melville, NY, USA). White arrows indicate GFP gene expression. The images were obtained using the fluorescence microscope mounted with a long-pass GFP filter and a DP72 camera. The GFP filter had excitation wavelengths of 460 to 480 nm and emission wavelengths of 495 to 540 nm, and the GFPA filter for the separation of GFP and blue-excited fluorophores had excitation wavelengths of 460 to 490 nm and emission wavelengths of 510 to 550 nm; these were used to identify the leaf and embryogenic callus tissues expressing GFP. (A,B) show the corresponding fluorescence image of 3-week-old fully expanded wounded leaves (cluster of cells) exhibiting GFP fluorescence and chloroplast autofluorescence, captured using a stereo fluorescence microscope under a GFP filter. The leaf tissues displayed green fluorescence that was partially masked by red fluorescence from chlorophyll (scale bar = 500 µm). (C) GFP protein fluorescence within the clusters of multiple cells in the leaf tissue (scale bar = 300 µm, GFPA filter); (D) strong expression of the GFP gene in primary vascular structures 3 days after co-cultivation (arrow); (E) strong expression of the GFP gene in the multiple isolated cells; (F,G) three-week-old leaf-derived callus pieces grown in a non-selection medium expressing GFP (globular embryos) (scale bar = 500 µm); (H) transformed globular-shaped embryogenic calli grown in a non-selection medium showing GFP expression (scale bar = 500 µm); (I) transformed nodular calli grown in the selection medium (10 mg/L hygromycin) showing GFP expression on day 14. (D,E,I) Magnification = 10 × 40.

Figure 3

Histochemical GUS assays were performed on transformed Cannabis sativa L. leaf tissues at different stages of development. One- to three-week-old leaves were transformed with the binary vector pCAMBIA1304 via Agrobacterium tumefaciens (strain GV3101). After 3 days of co-cultivation, histochemical staining for the GUS activity was performed. The dark blue staining indicates transient GUS gene expression. The experiments were repeated three times. (A) Expression of the β-glucuronidase gene in 1-week-old leaves with a surface area of 30.20 mm². (B) Expression of the β-glucuronidase gene in 2-week-old leaves with a surface area of 55.85 mm². (C) Expression of the β-glucuronidase gene in 3-week-old leaves with a surface area of 99.04 mm². Pictures were taken using a Leica EMSPIRA/3 Stereo Zoom Microscope (Leica Microsystems Inc., Deerfield, IL, USA) at a magnification of 0.743× (scale bar = 1.248 mm). The calculated surface area of the 1- to 3-week-old explants is indicated on the left side of (A–C).

Figure 4

Effect of bacterial density at OD_600nm = 0.6 on transformation efficiency based on transient β-glucuronidase (GUS) expression. Infection frequency was calculated as the percentage of GUS-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GUS foci per explant is the average number of GUS-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 5

Effect of bacterial density at OD_600nm = 0.6 on transformation efficiency based on GFP expression. Infection frequency was calculated as the percentage of GFP-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GFP foci per explant is the average number of GFP-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 6

Effect of wounding on GUS gene expression. Histochemical GUS assays were performed on transformed explants of Cannabis sativa L. Explants were transformed with the binary vector pCAMBIA1304 (1) via Agrobacterium tumefaciens. Explants were co-cultivated for 3 days with Agrobacterium, followed by histochemical staining for GUS activity. The dark blue staining indicates GUS activity in the various explants. The experiments were repeated three times with similar results. Three-week-old fully expanded and physically wounded by (A) using a hypodermic 23-gauge needle; (B) sonicated for 30 s twice with a 15 s interval; (C) physically pressing with 60 μM screen; (D) dark blue GUS staining in 3-week-old leaf tissues.

Figure 7

Effect of different wounding methods on transformation efficiency based on transient GUS expression. Infection frequency was calculated as the percentage of GUS-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GUS foci per explant is the average number of GUS-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 8

Effect of different wounding methods on transformation efficiency based on GFP expression. Infection frequency was calculated as the percentage of GFP-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GFP foci per explant is the average number of GFP-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 9

Influence of acetosyringone concentrations (µM) on transient GUS expression. Infection frequency was calculated as the percentage of GUS-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GUS foci per explant is the average number of GUS-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 10

Influence of acetosyringone concentrations (µM) on transient GFP expression. Infection frequency was calculated as the percentage of GFP-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GFP foci per explant is the average number of GFP-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 11

Effect of immersion period on transformation efficiency based on transient β-glucuronidase (GUS) expression. Infection frequency was calculated as the percentage of GUS-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GUS foci per explant is the average number of GUS-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 12

Effect of immersion period on transformation efficiency based on Green Florescent Protein (GFP) expression. Infection frequency was calculated as the percentage of GFP-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GFP foci per explant is the average number of GFP-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 13

Effect of different calcium chloride concentrations on transformation efficiency based on transient β-glucuronidase (GUS) expression. Infection frequency was calculated as the percentage of GUS-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GUS foci per explant is the average number of GUS-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 14

Effect of different calcium chloride concentrations on transformation efficiency based on transient GFP expression. Infection frequency was calculated as the percentage of GFP-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GFP foci per explant is the average number of GFP-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 15

Effect of different silver nitrate concentrations on transformation efficiency based on transient β-glucuronidase (GUS) expression. Infection frequency was calculated as the percentage of GUS-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GUS foci per explant is the average number of GUS-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).

Figure 16

Effect of different silver nitrate concentrations on transformation efficiency based on Green Florescent Protein (GFP) expression. Infection frequency was calculated as the percentage of GFP-positive explants from 3-week-old leaves out of the total number of explants examined. The number of GFP foci per explant is the average number of GFP-positive foci in at least three independent explants. The data are represented as means ± standard error (SE). To optimize the individual parameter, each experiment contained 25 to 30 samples and each experiment had 3 replications (n = 75).