Abstract
Micropropagation is a valuable production tool for the cultivation of hemp (Cannabis sativa), and development of optimal protocols is ongoing. The goal of this study was to evaluate a novel growing medium combination, consisting of Driver and Kuniyuki Walnut (DKW) medium as the nutrient source and glucose as the carbon source, and to investigate the link between in vitro and in vivo (i.e., greenhouse) plant performance. Among 10 accessions intended to represent a range of heterozygosity levels and various essential oil chemotypes, the DKW–glucose growing medium generally produced the most vigorous plantlets by all parameters evaluated in vitro (height, biomass, canopy area, vegetative growth rate, and regeneration rate). Across four growing media treatments, all of which included meta-topolin as the sole plant growth regulator, poor to no rooting was observed in vitro. Hybrids were more vigorous than nonhybrid selections in vitro, but not in vivo. No correlation was observed between in vitro and in vivo vigor, indicating that, with these media, plant performance in vitro is not predictive of that in vivo.
Micropropagation is a useful tissue culture (i.e., in vitro) technique that can be used for mass multiplication, germplasm maintenance, and virus mitigation in hemp production. Because hemp is still an emerging crop, the optimization process for hemp (Cannabis sativa) micropropagation is ongoing, and the primary focus has been on the selection of best-performing plant growth regulators (PGRs) in the growing medium, as reviewed by Boonsnongcheep and Pongkitwitoon (2020) and Lata et al. (2017). Notably, Lata et al. (2016) used the novel cytokinin meta-topolin to induce both rooting and shooting, thus bypassing the need for an additional medium containing auxin.
In contrast to PGRs, basal nutrient media and carbon sources have been explored minimally in hemp micropropagation. Murashige and Skoog (MS) medium (Murashige and Skoog, 1962) is the most common nutrient medium used in hemp in vitro studies (Boonsnongcheep and Pongkitwitoon, 2020), although Daria has been used for indirect regeneration (Wielgus et al., 2008). Recently, Page et al. (2021) noted an improved micropropagation multiplication rate and vigor in four of five high-cannabinoid-type [i.e., cannabidiol (CBD) and Δ-9-tetrahydrocannabinol (THC)] genotypes cultured on Driver and Kuniyuki Walnut medium (DKW) (Driver and Kuniyuki, 1984) compared with MS, Gamborg, and Babi media. Among carbon sources, sucrose has been the most widely used in hemp micropropagation studies (Boonsnongcheep and Pongkitwitoon, 2020; Lata et al., 2017). Recently, glucose was explored as a carbon source for in vitro seed germination of industrial hemp and resulted in lower germination rates in general compared with sucrose (Hesami et al., 2021). Nevertheless, glucose was shown to be superior to sucrose for in vitro growth of turmeric (Salvi et al., 2002), Prunus (Harada and Murai, 1996), and cork oak (Romano et al., 1995). In hops (Humulus lupulus), the most closely related species to hemp of economic importance (they are both in the family Cannabaceae), glucose resulted in greater multiplication rates than maltose (Smýkalová et al., 2001).
In vitro media optimization can be complicated by interactions between medium and genotype, as was observed for callus induction in hemp (Slusarkiewicz-Jarzina et al., 2005). With the high degree of genetic diversity among hemp cultivars (Lynch et al., 2016), it would be preferable to have a generally successful protocol that could be used across a range of genotypes. Furthermore, consistent performance between in vitro and in vivo environments is desirable. For example, in systems that rely on both in vitro and in vivo production stages, it is important that vigor in vivo translates to vigor in vivo and vice versa. A body of research compares field performance of plants that originated from micropropagation vs. conventional propagation (e.g., El-Shiekh et al., 1996; Nehra et al., 1994; Vuylsteke and Ortiz, 1996). However, the link between in vivo and in vitro vigor has rarely been investigated. In one study across 15 coffee varieties, no correlation was found between in vitro callogenesis and vegetative vigor in the field (Santos et al., 2013).
This study aims to evaluate a combination of two nutrient media (MS and DKW) and two carbon sources (glucose and sucrose) in a step toward improving hemp micropropagation protocols. Furthermore, because heterosis has been documented in hemp field studies (Kurtz et al., 2020), we used a selection of genotypes expected to have different heterozygosity levels to explore the relationship between plant performance in vitro and in vivo. We hypothesized that DKW in combination with glucose would produce the most vigorous plantlets in vitro, and that vigor in vitro and in vivo would be correlated.
Materials and Methods
Establishment of plant stock.
Ten accessions used in Oregon CBD’s (Oregon CBD, Hemp Seed Research and Development, Independence, OR) essential oil-type hemp production were evaluated: three F1 hybrids, three first-generation selfed selections, three-second generation selfed selections, and one backcross. We hereafter refer to the three F1 hybrids as “hybrids” and to the remaining nonhybrid genotypes as “selections.” It should be noted that all accessions used do not share the same parents, and that the second-generation selfed plants are not the second generation of the first-generation selfed plants included. Further information including cultivar numbers, names, pedigrees, and classifications are shown in Supplemental Table S1. Eight seeds were sown per cultivar in plug trays, over which plastic domes were placed to limit water loss during germination. Trays were maintained in a greenhouse under 24 h of light, with temperatures at 24 to 27 °C during the day and 20 to 26 °C at night. Plants were hand-watered as needed to sustain moderate to high soil moisture. One vigorous plant per cultivar was selected as the source of cuttings (i.e., mother plant).
After 2 months of growth from seed, ≈20 cuttings were taken from each mother plant. Cuttings were taken from the tips of lateral shoots, lightly scraped with a sharp scalpel, dipped in powdered indole-3 butyric acid (Hormex, Maia Products Inc., Westlake Village, CA), and placed in rockwool cubes (A-OK Starter Plugs; Grodan, Milton, Ontario, Canada) that had been soaked overnight in water adjusted to a pH of 6.8 using CalMag OAC (TPS Nutrients, Bellevue, WA). The rockwool was set in trays without holes, and a plastic dome was fitted over the top of the cuttings to maintain humidity of nearly 100%. Domed trays were placed under shadecloth in the same greenhouse conditions described earlier. Vents on the sides and tops of the domes were opened to decrease humidity. Cuttings that developed roots in rockwool (≈3 weeks) were transplanted to 4-L pots with a compost-based soil blend (Rexius, Eugene, OR). After another ≈4 weeks, successful cuttings (8–10 plants per genotype; i.e., daughter plants) were transplanted to 38-L pots. For each genotype, five daughter plants were used for the in vivo trial and the remaining daughter plants were used as explant donors for the in vitro trial. Daughter plants were fertilized for optimal production and treated for pests as needed, as detailed later.
In vivo trial.
The 10 genotypes were evaluated in the greenhouse with five replications arranged in a completely randomized design. The in vivo trial began after the daughter plants were transplanted to 38-L pots (growth stages 4–9) (Mediavilla et al., 1998). Plants were maintained in vegetative growth in the same greenhouse conditions described earlier for a week before being transitioned to another greenhouse at 24 to 27 °C with a 12-hour photoperiod to promote flowering, and stayed in these conditions through harvest. Plants were fertilized weekly for the first 5 weeks with 0.24 L of a three-part top-dress mixture [consisting of equal parts worm castings (Rexius), compost (Rexius), and chicken pellets (Stutzman Nutri-Rich 4-3-2, Canby, OR)], and an additional 0.24 L of granulated chicken manure applied every other week. Plants were treated periodically with Pyganic 5.0 (McLaughlin Gormley King Company, Minneapolis, MN) to manage aphid populations. Growth stage was evaluated weekly throughout the experiment. When plants were mature (≈8.5 weeks after being transitioned into the flowering environment), height was measured and aboveground biomass was harvested, and a subsample (i.e., one flowering lateral branch per plant, cut from the node at the main stem) was used to estimate harvest index by separating flowers and stems manually. Whole plants and subsampled flowers and stems were air-dried for several weeks before weighing for the biomass measurement. Flower yield was estimated as the whole plant weight multiplied by the harvest index. Vegetative growth rate (VGR) was determined as the average number of leaf pairs added per week during the 4-week period from the beginning of data collection to the end of vegetative growth.
In vitro trial.
Three to five daughter plants per accession were used as explant donors for the in vitro trial. Daughter plants were maintained in vegetative growth in the same greenhouse conditions described above for the establishment of plant stock. Plants were fertilized regularly with 0.24 L per plant of the three-part fertilizer mix described earlier.
Explants, comprised of nodal segments of ≈0.5 cm in length containing axillary buds, were cut from the daughter plants. Immediately after cutting, explants were surface-sterilized for 1 min in 70% ethanol. Explants were then transferred to a solution of 0.1% Plant Preservative Mixture (Plant Cell Technology, PPM, Washington, DC) and soaked for a minimum of 30 min and a maximum of 24 h for rehydration. Before inoculation on tissue culture medium, explants were surface-disinfected in a solution of 2% sodium hypochlorite [Bi-Mart brand bleach (5% sodium hypochlorite), Eugene, OR] and 0.1% Tween (VWR Life Science, Radnor, PA) for 10 min, transferred into the laminar flow hood, and rinsed in sterile water until no suds remained.
Sterilized explants were inoculated on 50 mL of solid, sterile medium contained in 372-mL culture vessels (C2100; PhytoTech Laboratories, Lenexa, KS) with one explant per vessel. Four media were evaluated, comprising a combination of two carbon sources (glucose and sucrose) and two nutrient media [MS (M5531, PhytoTech Laboratories) and DKW (D2470, PhytoTech Laboratories)]. Both carbon sources were used at 3% w/v (glucose at 0.17 m and sucrose at 0.09 m), and nutrient media were used at full strength. Based on the findings of Lata et al. (2016), all media contained 2 µm of the hormone meta-topolin. In addition, all media contained 0.5 g/L activated charcoal, 0.8% (w/v) agar, and 1% (w/v) PPM to reduce contamination. A detailed media preparation protocol as adapted from Lata et al. (2016) is provided in Supplemental Table S2.
Explants were incubated in growth chambers at 25 ± 2 °C with a 16-h photoperiod and 21 to 72 µmol⋅m–2⋅s–1 (variation a result of differences within the chamber). A total of 40 treatments (4 media × 10 genotypes), replicated eight times, were arranged in a completely randomized block design, with each block occupying the same section of a growth chamber to account for possible variation in light intensity. Numbers of leaf pairs were counted weekly, and VGR was determined as the average number of leaf pairs added per week in the 4-week period corresponding to weeks 2 through 5. After 5 weeks in culture, nonregenerating explants were counted and the regeneration rate was determined as the percentage of cuttings that grew at least one leaf pair during their time in culture. Canopy photos were taken using a smartphone held level above each culture vessel, and canopy area (as a percentage of the total vessel area) was estimated using a macro in ImageJ software version 1.5 K (Rasband, 2018) adapted from Bobadilla (2019) (Supplemental Fig. S1). Height was determined by measuring the total length of the shoot from base to apical meristem, and fresh biomass was determined by weighing whole plantlets. Because no roots were observed at this time, plantlets were transferred to fresh growing medium identical to their original treatment and cultured for an additional 5 weeks, after which the presence or absence of roots was recorded. Plantlets with roots were rinsed carefully in clean water, transferred to sterilized coconut coir (Eco Earth, Zoo Med Laboratories, San Luis Obispo, CA) in a sterile greenhouse plug tray, then returned to the growth chamber (in the same conditions as described earlier) for an additional 10 days before being transferred to sterile 0.5-L pots with a mix of two parts Sun Gro Professional Growing Mix potting soil (Sun Gro Horticulture, Agawam, MA) to one part perlite. Pots were maintained under vented, domed trays and shadecloth in the same greenhouse conditions described earlier for the establishment of plant stock.
Statistical analysis.
Orthogonal contrasts were used to compare media treatments and genotype groups for in vitro height, biomass, canopy, and VGR; and to compare genotype groups for in vivo height, biomass, flower yield, and VGR. To establish correlations between vigor parameters measured in vivo and in vitro, vigor parameters were averaged across replications (in vivo) or across both replications and media treatments (in vitro), and the mean for each genotype was used to perform a Spearman’s rank correlation for all possible combinations of parameters. Two experimental units were excluded from the in vitro data set as a result of contamination, and one in vivo unit was excluded because of molding during the postharvest drying process. All statistical analyses were performed using R statistical software version 1.4.1 (R Core Team, 2020).
Results
In vitro trial.
Orthogonal contrasts revealed significant effects of medium composition on plantlet vigor. For in vitro height, biomass, and canopy area, the interaction term between carbon source and nutrient medium was significant, whereas VGR displayed only significant main effects for these terms (Table 1).
Significance of orthogonal contrasts performed on vigor parameters measured in 10 hemp genotypes grown in vitro using different growing media (n = 318).
Significance of orthogonal contrasts performed on vigor parameters measured in 10 hemp genotypes grown in vivo (n = 49).
Overall, the DKW–glucose medium produced the most vigorous plantlets by all parameters (Fig. 1). Upon further exploration of the interaction between carbon source and nutrient medium, there was no difference between carbon sources on MS medium, whereas there was a significant difference between carbon sources on DKW medium. Plantlets were 47% taller and had a 32% greater biomass and 50% higher canopy area when DKW was used in combination with glucose instead of sucrose (Fig. 1A–C). Furthermore, VGR was 20% greater in glucose compared with sucrose and, similarly, 20% greater in DKW compared with MS (Fig. 1D). Across all 40 treatments, regeneration rates averaged 74%. There were generally small differences between growing media, with the highest and lowest rates observed for DKW–glucose (79%) and DKW–sucrose (71%), respectively (Fig. 1E).
In vitro vigor parameters measured in 10 hemp genotypes using different growing media (n = 320). (A–C) Based on orthogonal contrasts, experimental unit raw data were grouped by the interaction between basal nutrient medium [Driver and Kuniyuki Walnut medium (DKW) or Murashige and Skoog (MS) medium] and carbon source (glucose or sucrose), with the exception of (D), for which the interaction was not significant and nutrient medium and carbon groups are shown separately. Error bars indicate sem of the raw data. (A–D) Means include plants that did not regenerate. (E) Regeneration rate by genotype group.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vitro vigor parameters measured in 10 hemp genotypes using different growing media (n = 320). (A–C) Based on orthogonal contrasts, experimental unit raw data were grouped by the interaction between basal nutrient medium [Driver and Kuniyuki Walnut medium (DKW) or Murashige and Skoog (MS) medium] and carbon source (glucose or sucrose), with the exception of (D), for which the interaction was not significant and nutrient medium and carbon groups are shown separately. Error bars indicate sem of the raw data. (A–D) Means include plants that did not regenerate. (E) Regeneration rate by genotype group.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vitro vigor parameters measured in 10 hemp genotypes using different growing media (n = 320). (A–C) Based on orthogonal contrasts, experimental unit raw data were grouped by the interaction between basal nutrient medium [Driver and Kuniyuki Walnut medium (DKW) or Murashige and Skoog (MS) medium] and carbon source (glucose or sucrose), with the exception of (D), for which the interaction was not significant and nutrient medium and carbon groups are shown separately. Error bars indicate sem of the raw data. (A–D) Means include plants that did not regenerate. (E) Regeneration rate by genotype group.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Significant differences were observed between hybrid and selection groups for all in vitro parameters (Table 1), and hybrids outperformed selections in all vigor parameters measured. Compared with selections, hybrids were 60% taller and had 2.0 times the biomass, 2.3 times the canopy area, and a 30% greater VGR (Fig. 2A–D). Hybrids also had a greater rate of regeneration (85%) compared with selections (69%) (Fig. 2E).
In vitro vigor parameters measured in 10 hemp genotypes using different growing media (n = 318). Based on orthogonal contrasts, experimental units are shown by genotype group (hybrid or selection). Error bars indicate sem of the raw data. (A–D) Means include plants that did not regenerate. (E) Regeneration rate by genotype group. VGR = vegetative growth rate. ns, *, **, ***Nonsignificant or significant at P < 0.05, 0.01, or 0.001, respectively.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vitro vigor parameters measured in 10 hemp genotypes using different growing media (n = 318). Based on orthogonal contrasts, experimental units are shown by genotype group (hybrid or selection). Error bars indicate sem of the raw data. (A–D) Means include plants that did not regenerate. (E) Regeneration rate by genotype group. VGR = vegetative growth rate. ns, *, **, ***Nonsignificant or significant at P < 0.05, 0.01, or 0.001, respectively.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vitro vigor parameters measured in 10 hemp genotypes using different growing media (n = 318). Based on orthogonal contrasts, experimental units are shown by genotype group (hybrid or selection). Error bars indicate sem of the raw data. (A–D) Means include plants that did not regenerate. (E) Regeneration rate by genotype group. VGR = vegetative growth rate. ns, *, **, ***Nonsignificant or significant at P < 0.05, 0.01, or 0.001, respectively.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Rooting in vitro was variable and generally unsuccessful. No genotype had a rooting rate greater than 27%, and all roots were less than 2 cm long (examples shown in Supplemental Fig. S2). Of the four media evaluated, plantlets cultured on DKW–glucose had the greatest rooting rate of 26%, compared with 18%, 10%, and 5% for DKW–sucrose, MS–glucose, and MS–sucrose, respectively (Fig. 3A). With respect to genotype group, hybrids had a greater rooting rate compared with selections (25% vs. 10%, Fig. 3B). Despite careful adherence to protocol, no plants that developed roots transferred successfully to soil.
Percentage of hemp plantlets (n = 318) with roots after 10 weeks in vitro, grouped by media treatment (A) and genotype (B), calculated using raw data. DKW = Driver and Kuniyuki Walnut medium; MS = Murashige and Skoog medium.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Percentage of hemp plantlets (n = 318) with roots after 10 weeks in vitro, grouped by media treatment (A) and genotype (B), calculated using raw data. DKW = Driver and Kuniyuki Walnut medium; MS = Murashige and Skoog medium.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Percentage of hemp plantlets (n = 318) with roots after 10 weeks in vitro, grouped by media treatment (A) and genotype (B), calculated using raw data. DKW = Driver and Kuniyuki Walnut medium; MS = Murashige and Skoog medium.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vivo trial and correlations between in vivo and in vitro.
In contrast to in vitro results, orthogonal contrasts showed no significant differences between hybrids and selections for the in vivo metrics of height, biomass, flower yield, and VGR (Fig. 4, Supplemental Fig. S3).
In vivo (i.e., greenhouse) vigor parameters measured in 10 hemp genotypes (n = 49). Based on orthogonal contrasts, genotypes were grouped as hybrid or selection. Error bars indicate sem of the raw data. VGR = vegetative growth rate; ns = not significant.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vivo (i.e., greenhouse) vigor parameters measured in 10 hemp genotypes (n = 49). Based on orthogonal contrasts, genotypes were grouped as hybrid or selection. Error bars indicate sem of the raw data. VGR = vegetative growth rate; ns = not significant.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
In vivo (i.e., greenhouse) vigor parameters measured in 10 hemp genotypes (n = 49). Based on orthogonal contrasts, genotypes were grouped as hybrid or selection. Error bars indicate sem of the raw data. VGR = vegetative growth rate; ns = not significant.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Spearman’s rank correlation showed no correlations between in vitro and in vivo parameters. Furthermore, although there was a significant, strong positive correlation among all in vitro parameters (Table 2), there was only a significant positive correlation between height and flower yield among in vivo parameters.
Discussion
The high performance of DKW–glucose medium across these cultivars indicates a step toward refining hemp micropropagation protocols for improved vigor outcomes. This finding builds on that of Page et al. (2021), who observed greater multiplication rates and greater canopy areas of drug-type C. sativa when cultured on DKW compared with MS, using sucrose as the carbon source. Page et al. (2021) also noted that plants cultured on MS in the longer term developed nutrient deficiency and hyperhydricity. In contrast, Hesami et al. (2021) found MS–sucrose medium to produce greater rates of in vitro seed germination when compared with DKW–glucose medium, which indicates that nutrient medium and carbon sources best for one in vitro technique may not be the best for all other techniques—even within the same species. One important consideration is that, compared with the sucrose medium, the glucose medium has about twice the molarity at the same mass concentration (% w/v), thus, lower water potential. This could cause seeds to take up sucrose solutions more readily, whereas ease of water uptake may not be so important in developed stem tissues, possibly explaining the discrepancy between the results of Hesami et al. (2021) and those of our study. The widespread use of MS as the default nutrient medium for hemp tissue culture could be attributed to its proliferation in general tissue culture use. For example, Phillips and Garda (2019) found that 82% of tissue culture citations in a review used MS nutrient medium or some variation thereof, whereas just 6% of reviewed studies used DKW. However, results from our study and others show the need for further exploration of nutrient media, because the default does not always produce the best results.
It is difficult to single out the reason for hemp generally performing better under DKW compared with MS, given that there are several differences in the composition of these media. That said, the amount of sulfur in DKW is 8.5 times greater than in MS (Supplemental Table S3). Across 13 CBD hemp cultivars, Kalinowski et al. (2020) observed an average leaf sulfur concentration of 0.31%, which is greater than the average of 0.23% measured across 536 nonwoody plant species (Fratte et al., 2021). This relatively high leaf concentration of (and possibly requirement for) sulfur in hemp could explain the greater vigor observed with DKW compared with MS in our study. In addition, the lower than average phosphate levels in MS medium (1.25 mm, compared with 1.95 mm in DKW) have been reported to be inadequate for several plant species in both suspension and static cultures (George et al., 2007a).
Sucrose is the most widely used carbon source for in vitro growth across most plant species, and hemp is no exception. In a review of hemp micropropagation protocols, sucrose was used in 100% of the studies (Boonsnongcheep and Pongkitwitoon, 2020). However, glucose has been shown to be superior to sucrose in promoting growth and organogenesis in some species, including bitter almond (Yaseen et al., 2012) and some alder species (George et al., 2007b). Across the 10 hemp genotypes evaluated, glucose generally resulted in more vigorous plantlets. More important, there was an interaction between nutrient medium and carbon source for most vigor parameters. George et al. (2007b) noted that maintaining an adequate sugar-to-nitrate ratio in the medium is important because nitrogen uptake in vitro can be hindered by low levels of sugar in the medium. In our study, molar ratios of sugar to nitrate were greatest for DKW–glucose (5.75 vs. 2.22 for MS–sucrose, 4.23 for MS–glucose, and 3.03 for DKW–sucrose), and this could help explain why this medium resulted in the most vigorous plantlets across all vigor parameters measured (Fig. 1).
The poor rooting observed in the in vitro trial disagrees with results of Lata et al. (2016), in which high rooting rates were observed in medium containing the same major components as the MS–sucrose treatment in our study, which included meta-topolin as the only PGR. Although their study included one high-THC variety, all cultivars evaluated here are intended for the high-CBD and other non-THC cannabinoids market. The different genotypes used could explain these contrasting results, because there is a high degree of variation across hemp end-use types (Grassa et al., 2021; Johnson and Wallace, 2021; Lynch et al., 2016; Schwabe et al., 2019), and genotype-by-environment interactions are likely in effect, as documented in field studies (Petit et al., 2020). The use of 10 accessions in our study, representing a range of heterozygosity levels and various chemotypes, provides a fairly diverse sample (Supplemental Table S1). Therefore, at least for non-THC hemp, further research of PGRs is needed if a one-step protocol (using only one growing medium) is to be used. To our knowledge, no other studies have replicated the use of meta-topolin as the sole PGR in growing medium as described by Lata et al. (2016).
The clear pattern of hybrids significantly outperforming nonhybrid selections in all parameters measured in vitro is consistent with observed patterns of heterosis in conventionally grown hemp (Kurtz et al., 2020). It should be noted that the 10 accessions evaluated do not come from the same parents, and that the classification of cultivars as hybrids and nonhybrid selections is rendered on the assumption that the mother plants, randomly selected for each cultivar, were representative of the cultivar’s heterozygosity level. Furthermore, although the selections represent a group of accessions with an expected lesser degree of heterozygosity than the hybrids, it is difficult to define rigorously the degree of homozygosity in this germplasm. Cannabis is much less tolerant of inbreeding than many familiar horticultural crops. Therefore, comparisons between the hybrid and selection groups represent comparisons between germplasm groups expected to show more and less heterosis, respectively. If expectations are not met, it is possible that the differences observed between hybrid and selection groups could be a result of differences in genetic background rather than heterosis. However, there is some interrelation between cultivars (Supplemental Table S1), and the grouping structure used in all analyses, as well as the number of accessions included in the study, lend legitimacy to the comparison between hybrids and nonhybrid selections. Still, additional research using genetic measures of heterozygosity are required to extrapolate these results.
Heterosis patterns observed in vitro did not manifest in vivo (Fig. 4). Furthermore, the lack of correlation between in vitro and in vivo vigor metrics indicates that, at least by these measures and for these cultivars, in vitro and in vivo vigor are not aligned (Table 2). Multiple factors in gene expression could explain these discrepancies between in vitro and in vivo performance. For example, carbohydrate availability can regulate many genes and is connected to physiological and morphological outcomes such as photosynthesis, respiration, growth, hormone response, plant morphology, and carbon use. The greater abundance of carbohydrates in culture medium compared with in soil can result in the upregulation of carbon storage, carbon use, and growth genes (Koch, 1996). The greater expression of these genes in vitro could explain the manifestation of phenotypic differences in vitro that are not observed in vivo. In addition, the expression of other regulatory genes that affect in vitro growth specifically but not in vivo growth, such as those that regulate shoot regeneration and wound-induced differentiation in trees, could be involved (Nagle et al., 2018). This lack of correlation between in vitro and in vivo performance may have unwanted consequences for systems that rely on both in vitro and in vivo production. For example, these results suggest that breeding programs should not rely on in vitro selection.
Conclusion
Among four growing medium treatments and 10 accessions, the most vigorous plantlets were observed in the medium consisting of DKW as the nutrient source and glucose as the carbon source. Despite generally high shoot regeneration rates observed across all treatments, almost no rooting was observed. This suggests that, at least for a range of non-THC hemp accessions, a PGR other than meta-topolin is needed to promote rooting. F1 hybrid accessions outperformed selections in all vigor parameters in vitro but not in vivo (i.e., greenhouse). The lack of correlation observed between in vitro and in vivo vigor metrics indicates that in vitro vigor is not predictive of in vivo performance, and this may have unwanted consequences for breeding and production systems relying on both in vitro and in vivo growth.
Literature Cited
Bobadilla, L.K 2019 Frequency, distribution, ploidy diversity and control options of herbicide resistant italian ryegrass (Lolium perenne L. spp. multiflorum (Lam.) Husnot) populations in western Oregon Oregon State Univ., Corvallis MS thesis
Boonsnongcheep, P. & Pongkitwitoon, B. 2020 Factors affecting micropropagation of Cannabis sativa L: A review Pharm. Sci. Asia 47 21 29 https://doi.org/10.29090/psa.2020.01.019.0030
Driver, J. & Kuniyuki, A. 1984 In vitro propagation of Paradox Walnut root stock HortScience 18 506 509
El-Shiekh, A., Wildung, D.K., Luby, J.J., Sargent, K.L. & Read, P.E. 1996 Long-term effects of propagation by tissue culture or softwood single-node cuttings on growth habit, yield, and berry weight of “northblue” blueberry J. Amer. Soc. Hort. Sci. 121 339 342 https://doi.org/10.21273/JASHS.121.2.339
Fratte, M.D., Pierce, S., Zanzottera, M. & Cerabolini, B.E.L. 2021 The association of leaf sulfur content with the leaf economics spectrum and plant adaptive strategies Funct. Plant Biol. 48 9 924 993 https://doi.org/10.1071/FP20396
George, E.F., Hall, M.A. & Klerk, G.J.D. 2007a The components of plant tissue culture media I: Macro- and micro-nutrients 65 113 George, E.F., Hall, M.A. & Klerk, G.-J.D. Plant propagation by tissue culture. Springer Dordrecht, the Netherlands
George, E.F., Hall, M.A. & Klerk, G.J.D. 2007b The components of plant tissue culture media II: Organic additions, osmotic and pH effects George, E.F., Hall, M.A. & Klerk, G.-J.D. Plant propagation by tissue culture. Springer Dordrecht, the Netherlands
Grassa, C.J., Weiblen, G.D., Wenger, J.P., Dabney, C., Poplawski, S.G., Motley, S.T., Michael, T.P. & Schwartz, C.J. 2021 A new Cannabis genome assembly associates elevated cannabidiol (CBD) with hemp introgressed into marijuana New Phytol. 230 1665 1679 https://doi.org/10.1111/nph.17243
Harada, H. & Murai, Y. 1996 Micropropagation of Prunus mume Plant Cell Tissue Org. 46 265 267 https://doi.org/10.1007/BF02307104
Hesami, M., Pepe, M., Monthony, A.S., Baiton, A. & Phineas Jones, A.M. 2021 Modeling and optimizing in vitro seed germination of industrial hemp (Cannabis sativa L.) Ind. Crops Prod. 170 113753 https://doi.org/10.1016/j.indcrop.2021.113753
Johnson, M.S. & Wallace, J.G. 2021 Genomic and chemical diversity of commercially available high-CBD industrial hemp accessions Front. Genet. 12 682475 https://doi.org/10.3389/fgene.2021.682475
Kalinowski, J., Edmisten, K., Davis, J., McGinnis, M., Hicks, K., Cockson, P., Veazie, P. & Whipker, B.E. 2020 Augmenting nutrient acquisition ranges of greenhouse grown CBD (cannabidiol) hemp (Cannabis sativa) cultivars Horticulturae 6 98 https://doi.org/10.3390/horticulturae6040098
Koch, K.E 1996 Carbohydrate-modulated gene expression in plants Annu. Rev. Plant Physiol. 47 509 540 https://doi.org/10.1146/annurev.arplant. 47.1.509
Kurtz, L.E., Mahoney, J.D., Brand, M.H. & Lubell-Brand, J.D. 2020 Comparing genotypic and phenotypic variation of selfed and outcrossed progeny of hemp HortScience 55 1206 1209 https://doi.org/10.21273/HORTSCI15061-20
Lata, H., Chandra, S., Khan, I. & ElSohly, M. 2017 Micropropagation of Cannabis sativa L.: An update 285 297 Chandra, S., Lata, H. & ElSohly, M.A. Cannabis sativa L.: Botany and biotechnology. Springer Cham, Germany https://doi.org/10.1007/978-3-319-54564-6_13
Lata, H., Chandra, S., Techen, N., Khan, I.A. & ElSohly, M.A. 2016 In vitro mass propagation of Cannabis sativa L.: A protocol refinement using novel aromatic cytokinin meta-topolin and the assessment of eco-physiological, biochemical and genetic fidelity of micropropagated plants J. Appl. Res. Med. Aromat. Plants 3 18 26 https://doi.org/10.1016/j.jarmap.2015.12.001
Lynch, R.C., Vergara, D., Tittes, S., White, K., Schwartz, C.J., Gibbs, M.J., Ruthenburg, T.C., deCesare, K., Land, D.P. & Kane, N.C. 2016 Genomic and chemical diversity in cannabis Crit. Rev. Plant Sci. 35 349 363 https://doi.org/10.1080/07352689.2016.1265363
Mediavilla, V., Jonquera, M., Schmid-Slembrouck, I. & Soldati, A. 1998 Decimal code for growth stages of hemp (Cannabis sativa L.) J. Intl. Hemp Assoc. 5 2 65
Murashige, T. & Skoog, F. 1962 A revised medium for rapid growth and bioassays with tobacco tissue cultures Physiol. Plantarium 15 473 497 https://doi.org/10.1111/j.1399-3054.1962.tb08052.x
Nagle, M., Déjardin, A., Pilate, G. & Strauss, S.H. 2018 Opportunities for innovation in genetic transformation of forest trees Front. Plant Sci. 9 1443 https://doi.org/10.3389/fpls.2018.01443
Nehra, N.S., Kartha, K.K., Stushnoff, C. & Giles, K.L. 1994 Effect of in vitro propagation methods on field performance of two strawberry cultivars Euphytica 76 107 115 https://doi.org/10.1007/BF00024027
Page, S.R.G., Monthony, A.S. & Jones, A.M.P. 2021 DKW basal salts improve micropropagation and callogenesis compared with MS basal salts in multiple commercial cultivars of Cannabis sativa Botany 99 269 279 https://doi.org/10.1139/cjb-2020-0179
Petit, J., Salentijn, E.M.J., Paulo, M.-J., ThouMinot, C., van Dinter, B.J., Magagnini, G., Gusovius, H.-J., Tang, K., Amaducci, S., Wang, S., Uhrlaub, B., Müssig, J. & Trindade, L.M. 2020 Genetic variability of morphological, flowering, and biomass quality traits in hemp (Cannabis sativa L.) Front Plant Sci. 11 102 https://doi.org/10.3389/fpls.2020.00102
Phillips, G.C. & Garda, M. 2019 Plant tissue culture media and practices: An overview In Vitro Cell. Dev. Biol. Plant 55 242 257 https://doi.org/10.1007/s11627-019-09983-5
Rasband, W.S 2018 ImageJ. U.S. National Institutes of Health Bethesda MD https://imagej.nih.gov/ij/
R Core Team 2020 R: A language and environment for statistical computing R Foundation for Statistical Computing Vienna Austria https://www.R-project.org/
Romano, A., Noronha, C. & Martins-Loucao, M.A. 1995 Role of carbohydrates in micropropagation of cork oak Plant Cell. Tissue Org. 40 2 159 167 https://doi.org/10.1007/BF00037670
Salvi, N.D., George, L. & Eapen, S. 2002 Micropropagation and field evaluation of micropropagated plants of turmeric Plant Cell. Tissue Org. 68 2 143 151 https://doi.org/10.1023/A:1013889119887
Santos, M.R.A., Ferreira, M.G.R., Oliveira, C.L.L.G., Ramalho, A.R. & Espindula, M. 2013 Vegetative vigor conilon coffee and its potential for in vitro callus induction Coffee Sci. 8 432 438
Schwabe, A.L., Hansen, C.J., Hyslop, R.M. & McGlaughlin, M.E. 2019 Research grade marijuana supplied by the National Institute on Drug Abuse is genetically divergent from commercially available Cannabis bioRxiv 592725 https://doi.org/10.1101/592725
Slusarkiewicz-Jarzina, A.S., Ponitka, A. & Kaczmarek, Y. 2005 Influence of cultivar, explant source and plant growth regulator on callus induction and plant regeneration of Cannabis sativa L Acta Biol. Cracov. Bot. 47 145 151
Smýkalová, I., Ortová, M., Lipavská, H. & Patzak, J. 2001 Efficient in vitro micropropagation and regeneration of Humulus lupulus on low sugar, starch-gelrite media Biol. Plant. 44 7 12 https://doi.org/10.1023/A:1017901817063
Vuylsteke, D.R. & Ortiz, R. 1996 Field performance of conventional vs. in vitro propagules of plantain (Musa spp., AAB Group) HortScience 31 862 865 https://doi.org/10.21273/HORTSCI.31.5.862
Wielgus, K., Luwanska, A., Lassocinski, W. & Kaczmarek, Z. 2008 Estimation of Cannabis sativa L. tissue culture conditions essential for callus induction and plant regeneration J. Nat. Fibers 5 199 207 https://doi.org/10.1080/15440470801976045
Yaseen, M., Ahmad, T., Sablok, G., Standardi, A. & Hafiz, I.A. 2012 Review: Role of carbon sources for in vitro plant growth and development Mol. Biol. Rep. 40 2837 2849 https://doi.org/10.1007/s11033-012-2299-z
Macro written for ImageJ to determine percent of culture vessel area covered by hemp vegetation based on color thresholding. Provided by Lucas Bobadilla; adapted from Bobadilla (2019).
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Macro written for ImageJ to determine percent of culture vessel area covered by hemp vegetation based on color thresholding. Provided by Lucas Bobadilla; adapted from Bobadilla (2019).
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Macro written for ImageJ to determine percent of culture vessel area covered by hemp vegetation based on color thresholding. Provided by Lucas Bobadilla; adapted from Bobadilla (2019).
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Hemp plantlets cultured from micropropagation cuttings. (A) One replicate of genotype 1 grown in Driver and Kuniyuki Walnut medium with glucose. (B) One replicate of genotype 1 grown in Murashige and Skoog medium with sucrose. (C, D) Examples of plantlets that developed small roots at the base.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Hemp plantlets cultured from micropropagation cuttings. (A) One replicate of genotype 1 grown in Driver and Kuniyuki Walnut medium with glucose. (B) One replicate of genotype 1 grown in Murashige and Skoog medium with sucrose. (C, D) Examples of plantlets that developed small roots at the base.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Hemp plantlets cultured from micropropagation cuttings. (A) One replicate of genotype 1 grown in Driver and Kuniyuki Walnut medium with glucose. (B) One replicate of genotype 1 grown in Murashige and Skoog medium with sucrose. (C, D) Examples of plantlets that developed small roots at the base.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Greenhouse trial hemp plants at (A) 4 weeks, (B) 6 weeks, (C) 8 weeks, and (D) 10 weeks after initial propagation.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Greenhouse trial hemp plants at (A) 4 weeks, (B) 6 weeks, (C) 8 weeks, and (D) 10 weeks after initial propagation.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Greenhouse trial hemp plants at (A) 4 weeks, (B) 6 weeks, (C) 8 weeks, and (D) 10 weeks after initial propagation.
Citation: HortScience 57, 9; 10.21273/HORTSCI16651-22
Pedigree information for 10 essential oil–type hemp cultivars provided by Oregon CBD for in vitro (micropropagation) and in vivo culture.
Medium components and preparation of hemp micropropagation medium, adapted from Lata et al. (2016).
Product component comparison of Murashige and Skoog (MS), and Driver and Kuniyuki Walnut (DKW) plant nutrient media.
