Abstract
Cannabis (Cannabis sativa) growers seek greater crop uniformity and yield; however, use of vegetatively propagated plants has several drawbacks. There is increased interest in F1 hybrid seed, especially for flower production. Plant growth and performance of inbred lines of the cultivars Wife (WF) and Wilhelmina (WH), produced from one to four rounds of selfing (S1 to S4), and three F1 hybrid lines (WF S1 × WH S1, WF S2 × WH S2, and WF S3 × WH S3) were evaluated. As selfing increased from S1 to S4, within line plant variation did not change substantially for most measures for WF, but variation progressively decreased for plant height, total shoot length and plant dry weight for WH. S4 plants of WH were uniformly diminutive and likely exhibited inbreeding depression. F1 hybrids between WF and WH were larger than WH plants for all measures because WH is photoperiod insensitive (autoflowering) and dwarf. Compared with WF self-lines, F1 WF × WH hybrid lines exhibited greater uniformity, plant dry weight, flower dry weight, and percent cannabidiol and tetrahydrocannabinol. F1 WF × WH hybrid plants were able to achieve similar height and shoot growth to WF self-lines, even though they had a dwarf parent in WH. It is likely that heterosis is responsible for the growth enhancements observed in the F1 WF × WH hybrids. These findings indicate that F1 hybridization can produce uniform, stable, and high-yielding cannabis plants from seed.
Cannabis (Cannabis sativa) flower growers use stem-cutting propagation or micropropagation to generate clones of desirable female cultivars for controlled environment production; however, both propagation methods have limitations (Lubell-Brand et al. 2021). Stem cutting propagation requires maintenance of mother plants, which occupy growing space (Lubell-Brand et al. 2021) and frequently succumb to hop latent viroid (Adkar-Purushothama et al. 2023). Production of cannabis plants from micropropagation is hindered by the plant’s recalcitrance in tissue culture (Borbas et al. 2023).
The use of F1 hybrid seed has increased productivity for a range of agronomic crops including corn (Zea mays) and tomato [Solanum lycopersicum (Cheema and Dhaliwal 2004; Wright 1980)]. F1 hybrid seed is produced by crossing two inbred parent genotypes developed from repeated selfing. The benefits of F1 hybrid seed may include enhanced plant uniformity, vigor, and yield. There is a deficiency of published studies on selfing and F1 hybrid seed breeding for cannabis (Carlson et al. 2021).
Cannabis is a naturally dioecious, photoperiod sensitive, and outcrossing species (Bocsa and Karus 1999). Female plants can be treated with silver thiosulfate (STS), or other compounds, to induce the development of male flowers, which permits selfing and the generation of feminized (all female) seed (Lubell and Brand 2018). This method of creating feminized seed is used by flower growers to preserve cultivar genetics for their own production scheduling as well as for sale to cannabis growers and consumers. The objective of this research was to evaluate the growth and performance of cannabis seed produced from one to four rounds of selfing and F1 hybrid crosses.
Two cannabis cultivars, Wife (WF) and Wilhelmina (WH), which differ in photoperiod sensitivity and plant habit, were included in this research. WF is photoperiod sensitive and plants require 12 to 14 h of dark photoperiod to flower (Zhang et al. 2021). Plants of WF are tall and upright with long internodes and sparse foliage density. WH is photoperiod insensitive (autoflowering), and plants flower 35 to 50 d after seed germination regardless of photoperiod (Kurtz et al. 2023). In cannabis, autoflowering is a homozygous recessive trait controlled by a single locus (Toth et al. 2022). Plants of WH possess short internodes, small leaves, and stem diameters, and they produce few lateral shoots, which results in a dwarf habit.
Materials and Methods
Plant material, self-crosses, and F1 hybrid crosses.
WF was obtained as dioecious seed from US Hempcare (Niantic, CT, USA). A female plant was identified for the first self-cross, which produced an S1 seed line of WF. WH was obtained as feminized seed from Atlas Seed (Sebastopol, CA, USA), and a plant was selected for the first self-cross to produce an S1 seed line of WH. A self-cross was conducted by treating two shoots originating from the basal portion of the female plant with foliar sprays of 3 mM STS according to Lubell and Brand (2018), except applications were made every 4 d, to induce male flower production. For each cultivar, four self-crosses were conducted in succession to produce S1, S2, S3, and S4 seed lines. To accomplish this, an S1 plant was selfed to produce S2 seed, an S2 plant was selfed to produce S3 seed, and an S3 plant was selfed to produce S4 seed. Every cross conducted for this research was accomplished in isolation using a growth chamber with set point of 24 °C and 12-h photoperiod control provided by light-emitting diode lamps at 250 mmol⋅m−2⋅s−1. Pollen was dispersed by natural air flow in the growth chamber and by hand-plucking male open flowers and bringing their anthers in direct contact with female flowers. Seeds were harvested when their color had changed from green to brown at ∼35 to 42 d after pollination. Seed was stored in plastic bags in a refrigerator at 4 °C until needed.
The following F1 hybrid crosses were conducted using a single plant from each parent line: WF S1 × WH S1, WF S2 × WH S2, and WF S3 × WH S3. For each F1 hybrid cross, only the WH plant was masculinized by treating the entire plant with STS as described. Each F1 hybrid cross was conducted in isolation using a growth chamber as described. Pollination, seed harvest, and seed storage were as described for self-crosses. Self and F1 hybrid crosses were completed during a 16-month time-period spanning 2022 to 2023.
Study 1.
A greenhouse study was conducted from Sep 2023 to Nov 2023 to compare plant growth and performance of WF self-lines and F1 WF × WH hybrid lines. Plants were started from seed in 50-plug trays filled with peat-based medium (Metro-Mix 830; Sun Gro Horticulture, Agawam, MA, USA) and placed in a greenhouse with set points of 21/17 °C day/night temperature thresholds and photoperiod control provided by supplemental lighting using 1000-W high-pressure sodium lamps (Phantom HPS 100 W; Hydrofarm, Petaluma, CA, USA) and blackout curtains. Before sowing, seeds were soaked in water for 24 h then set in plastic petri dishes lined with moistened filter paper for 48 h. Fourteen days after sowing, seedlings were selected at random and potted to 5.68-L containers filled with the same medium. Containers were top-dressed with 20 g of 15N–3.9P–10K controlled-release fertilizer (Osmocote Plus 5- to 6-month formulation; Everris NA, Dublin, OH, USA). Plants were irrigated as needed and fertigated once weekly with a 20N–4.3P–16.6K water-soluble fertilizer (Peters 20 to 10–20; Scotts, Marysville, OH, USA) providing 100 mg⋅L−1 N. Plants were grown vegetatively under an 18-h photoperiod for 24 d; the photoperiod was then reduced to 12 h for 34 d for flowering. The experimental unit (EU) was a single potted plant, and plants were arranged in a randomized complete block design (RCBD) with eight replications. Initiation of terminal flowering was recorded when a minimum of three pistils bearing stigmas were visible at the shoot tips, according to Spitzer-Rimon et al. (2019). Plant height, measured from base to tip of the main shoot, was recorded on day 52 of the study. Shoot length was measured on day 56 of the study and summed for each plant. At harvest, plants were cut at the stem base, allowed to dry at room temperature for 14 d, and weighed; flowers were then separated from shoots and weighed. Percent flower weight was calculated by dividing the flower dry weight by the total plant dry weight and multiplying by 100%. Cannabinoid content of dry flower was analyzed by high-performance liquid chromatography at the University of Connecticut Center for Environmental Sciences and Engineering in Storrs, CT, USA.
Study 2.
A second greenhouse study was conducted from Jan 2024 to Mar 2024 to compare plant growth and performance of WH self-lines and F1 WF × WH hybrid lines. Plants were started from seed in a greenhouse as described. Fourteen days after sowing, WH plants were potted to 2.78-L containers and F1 WF × WH hybrid plants to 5.68-L containers, filled with the same medium as described for study 1, and top-dressed with the same controlled-release fertilizer as described at 8 g or 20 g, respectively. Irrigation and fertigation were as described for study 1. Plants were grown under 18-h photoperiod for 31 d, and then under 12-h photoperiod for 38 d. The EU was a single potted plant, and plants were arranged in an RCBD with 10 replications for WH self-lines and eight replications for F1 WF × WH hybrid lines. Initiation of terminal flowering, plant height and shoot length were measured as described on day 50 of the study for WH plants and on day 63 of the study for F1 WF × WH hybrid plants. Plant harvest was conducted as described on day 51 of the study for WH plants and on day 69 of the study for F1 WF × WH hybrid plants. Plant dry weight, percent flower dry weight and cannabinoid content of dry flower were determined as described for study 1.
Statistical analysis.
Data were subjected to the non-parametric Friedman test (P ≤ 0.05) to compare between self and F1 hybrid lines using statistical software (SAS version 9.4; SAS Institute, Cary, NC, USA).
Results and Discussion
Plants from WH self-lines were smaller and flowered earlier than plants from F1 WF × WH hybrid lines, which was not unexpected, because WH is autoflowering (Figs. 1, 2, and 5). The difference between WH self-lines and F1 WF × WH hybrid lines was not as substantial for percent flower dry weight as it was for other measures because autoflowering plants are dwarf. Kurtz et al. (2024) showed that autoflowering cultivars produced greater percent flower dry weight than some photoperiod sensitive cultivars. Means for all traits did not change as selfing increased from S1 to S4 for WH (Fig. 1; Table 1), but within line variation progressively decreased for plant height, total shoot length and plant dry weight (Fig. 1). Variation in days to flower was less for WH S4 plants than for S3 plants and so on. Visually, WH S4 plants were uniformly diminutive in stature and likely suffered from inbreeding depression (Fig. 2).
Boxplots of plant height (A), total shoot length (B), time to terminal flowering (C), plant dry weight (D), and percent flower dry weight (E) for cannabis (Cannabis sativa) plants from four selfed lines (S1, S2, S3, S4) of the cultivar Wilhelmina (WH) and three F1 hybrid lines of the cultivar Wife (WF) × WH (WF S1 × WH S1, WF S2 × WH S2, WF S3 × WH S3). Letters depict nonparametric Friedman test groupings at α = 0.05 (n = 10 for WH self-lines and n = 8 for F1 hybrid lines).
Citation: HortScience 59, 12; 10.21273/HORTSCI18197-24
Cannabis (Cannabis sativa) plants from four self-lines (S1, S2, S3, S4) of the cultivar Wilhelmina.
Citation: HortScience 59, 12; 10.21273/HORTSCI18197-24
Percent cannabidiol (CBD) and percent tetrahydrocannabinol (THC) for cannabis (Cannabis sativa) plants from three self-lines (S1, S3, S4) of the cultivar Wife (WF) and three F1 hybrid lines of WF × Wilhelmina [WH (WF S1 × WH S1, WF S2 × WH S2, WF S3 × WH S3)] in study 1, and from four self-lines (S1, S2, S3, S4) of WH in study 2.
Plants within each WF self-line were highly variable for all measured parameters (Figs. 3 and 4). WF S2 seeds did not germinate well due to pest pressure on the mother plant and/or damage to the seed during harvest or storage, therefore this line was not included in the study. As the level of selfing for WF increased from S1 to S4, the degree of within line variation did not change substantially for most measured parameters (Fig. 3).
Boxplots of plant height (A), total shoot length (B), time to terminal flowering (C), plant dry weight (D), and percent flower dry weight (E) for cannabis (Cannabis sativa) plants from three selfed lines (S1, S3, S4) of the cultivar Wife (WF) and three F1 hybrid lines of WF × Wilhelmina [WH (WF S1 × WH S1, WF S2 × WH S2, WF S3 × WH S3)]. Letters depict nonparametric Friedman test groupings at α = 0.05 (n = 8).
Citation: HortScience 59, 12; 10.21273/HORTSCI18197-24
Cannabis (Cannabis sativa) plants from three self-lines (S1, S3, S4) of the cultivar Wife.
Citation: HortScience 59, 12; 10.21273/HORTSCI18197-24
Other indicators of inbreeding depression within the self-lines were reduced fertility or infertility, leaf malformations, and super dwarf or stunted plants. For WF self-lines, 38% of S1 plants, 25% of S3 plants, and 50% of S4 plants produced ≤1 g of dried flower only. For WH self-lines, 40% of S1 plants, 40% of S2 plants, and 10% of S3 plants exhibited yellow, malformed leaves, super dwarf habit, and were infertile (Fig. 2). Interestingly, no WH S4 plants displayed this phenotype. Plants that possessed this phenotype could be identified at the seedling stage by their first true leaves, which were yellow. Out of 47 WH S4 seeds that germinated, none of the seedlings had yellow leaves, whereas 22% of S1 seedlings (n = 50), 21% of S2 seedlings (n = 48), and 19% of S3 seedlings (n = 37) did. Despite the small sample sizes, these observations suggest that the deleterious phenotype of yellow, malformed leaves, super dwarf habit, and infertility may have been eliminated by the S4 generation; however, additional seedlings would need to be evaluated to confirm this.
Plant performance was more uniform for F1 WF × WH hybrid lines than for WF self-lines (Figs. 3–5). F1 WF × WH hybrid lines had greater plant dry weight, flower dry weight and percent cannabidiol (CBD) and tetrahydrocannabinol (THC) than WF self-lines (Fig. 3; Table 1). F1 WF × WH hybrid plants were able to achieve similar height and shoot growth to WF self-lines, even though they had a dwarf parent in WH. It is likely that heterosis is responsible for the growth enhancements observed in the F1 WF × WH hybrids. Plants from the F1 hybrid line WF S3 × WH S3 were more uniform for total shoot length, plant dry weight, and percent flower dry weight than plants from the WF S1 × WH S1 and WF S2 × WH S2 lines in study 1 (Fig. 3). In addition, plants from the WF S2 × WH S2 and WF S3 × WH S3 lines were more uniform for plant height than plants from the WF S1 × WH S1 line. Terminal flowering occurred earlier for F1 WF × WH hybrid lines than WF self-lines, due to their autoflowering parent. Kurtz et al. (2023) also showed that hybrids between photoperiod sensitive and autoflowering cultivars flowered significantly earlier than the photoperiod sensitive parent.
Cannabis (Cannabis sativa) plants from three F1 hybrid lines from crossing cultivars Wife × Wilhelmina using plants derived from selfing at the S1, S2, or S3 level.
Citation: HortScience 59, 12; 10.21273/HORTSCI18197-24
There is a need within the cannabis flower industry for uniform stable cultivars (Smart et al. 2023). Many cannabis cultivars, likely derived from open pollination or random hybridization of cultivars, have shown significant within cultivar phenotypic variation (Stack et al. 2021). Plants from S2 seed lines developed and tested by Carlson et al. (2021) demonstrated variation in days to flower, slower growth, and inbreeding depression. Carlson et al. (2021) suggested that at least one inbred parent is needed for uniform cultivar development. Cannabis seed production companies such as Oregon CBD (Independence, OR, USA) and New West Genetics (Windsor, CO, USA) currently offer F1 hybrid seed, but little information about the breeding methods and the degree of parent line inbreeding has been provided (New West Genetics 2021; Oregon CBD 2024). Oregon CBD reports creating F1 hybrid seed by crossing an S1 photoperiod sensitive genotype with an autoflowering genotype (Oregon CBD 2024; Stack et al. 2022). Growers have indicated that Oregon CBD seed has produced more uniform crops than seed cultivars obtained from other sources (Joe Ullman, Co-owner and Director of Sales, Atlas Seed, Sebastopol, CA, USA, personal communication). In the 2021 and 2022 flower variety trials at the University of Vermont (Burlington, VT, USA), Oregon CBD cultivars were some of the largest and highest flower-yielding plants (Darby et al. 2021, 2022).
In the United States, cannabis breeding for flower production has predominantly been conducted by industry and not by academic institutions due to legal restrictions. Until recently, numerous cannabis seed suppliers sold S1 seed and marketed it as uniform and like the original mother plant. Our study indicates that S1 cannabis seed can be highly variable, like seed from open pollination or random hybridization, and this may negatively affect important traits such as flowering timing and yield (Smart et al. 2023).
Industry interest in F1 hybrid cannabis seed is increasing (Stevens 2023). Our results demonstrate that F1 hybrid seed bred from two inbred parental lines can produce uniform, stable and high yielding plants for production purposes (Figs. 1, 3, and 5). The methods used herein to produce feminized F1 hybrid seed can be readily adopted by cannabis seed producers. Additional benefits of F1 hybrid seed over traditional propagation methods, such as stem cuttings and micropropagation, may include stronger plant roots and stems, on demand use of seeds, and less reliance on propagation facilities’ infrastructure and supplies (Stevens 2023). The degree of selfing required for producing parental lines may vary depending on the inherent heterozygosity of the starting cultivars. For self-compatible crops such as corn and tomato, selfing to the S5 or S7 of the parental lines is recommended to optimize hybrid vigor effects in the F1 hybrid seed (Cheema and Dhaliwal 2004; Wright 1980). The optimal degree of selfing for parental lines could be less for cannabis because it is a naturally outcrossing species (Owens and Miller 2009).
References Cited
Adkar-Purushothama CR, Sano T, Perreault JP. 2023. Hop Latent Viroid: A Hidden Threat to the Cannabis Industry. Viruses. 15(3):681. https://doi.org/10.3390/v15030681.
Bocsa I, Karus M. 1999. The cultivation of hemp. Hemptech, Sebastopol, CA, USA.
Borbas LN, Kurtz LE, Lubell-Brand JD. 2023. A comparison of two media formulations and two vented culture vessels for shoot multiplication and rooting of hemp shoot tip cultures. HortTechnology. 33(2):233–238. https://doi.org/10.21273/HORTTECH05179-22.
Carlson CH, Stack GM, Jiang Y, Taşkıran B, Cala AR, Toth JA, Philippe G, Rose JKC, Smart CD, Smart LB. 2021. Morphometric relationships and their contribution to biomass and cannabinoid yield in hybrids of hemp (Cannabis sativa). J Exp Bot. 72(22):7694–7709. https://doi.org/10.1093/jxb/erab346.
Cheema DS, Dhaliwal MS. 2004. Hybrid tomato breeding. J New Seeds. 6(2-3):1–14. https://doi.org/10.1300/J153v06n02_01.
Darby H, Bruce J, Sullivan L, Zeigler S. 2021. 2021 Hemp flower variety trial. https://www.uvm.edu/sites/default/files/Northwest-Crops-and-Soils-Program/2021%20Research%20Rpts/2021_UVM_Hemp_Flower_Variety_Trial_Final.pdf. [accessed 24 Jul 2024].
Darby H, Bruce J, Sullivan L, Zeigler S. 2022. 2022 Hemp flower variety trial. https://www.uvm.edu/sites/default/files/Northwest-Crops-and-Soils-Program/2022%20Research%20Reports/2022_Hemp_Flower_Variety_Trial.pdf. [accessed 24 Jul 2024].
Kurtz LE, Brand MH, Lubell-Brand J. 2024. Cannabis triploids exhibit reduced fertility and similar growth and flower production compared to diploids. J Am Soc Hort Sci. 149(2):75–85. https://doi.org/10.21273/JASHS05359-23.
Kurtz LE, Brand MH, Lubell-Brand J. 2023. Gene dosage at the autoflowering locus effects flowering timing and plant height in triploid cannabis. J Am Hort Sci. 148(2):83–88. https://doi.org/10.21273/JASHS05293-23.
Lubell JD, Brand MH. 2018. Foliar sprays of silver thiosulfate produce male flowers on female hemp plants. HortTechnology. 28(6):743–747. https://doi.org/10.21273/HORTTECH04188-18.
Lubell-Brand J, Kurtz LE, Brand MH. 2021. An in vitro – ex vitro micropropagation system for hemp. HortTechnology. 31(2):199–207. https://doi.org/10.21273/HORTTECH04779-20.
New West Genetics. 2021. New West Genetics. https://newwestgenetics.com. [accessed 24 Jul 2024].
Oregon CBD. 2024. Oregon CBD. https://oregoncbdseeds.com/. [accessed 24 Jul 2024].
Owens SJ, Miller R. 2009. Cross and self-fertilization of plants—Darwin’s experiments and what we know now. Bot J Linnean Soc. 161:357–395. https://doi.org/10.1111/j.1095-8339.2009.01010.x.
Smart LB, Toth JA, Stack GM, Monserrate LA, Smart CD. 2023. Breeding of hemp (Cannabis sativa), p 239−288. In: Goldman I (ed). Plant breeding reviews. John Wiley & Sons Inc, Hoboken, NJ, USA.
Spitzer-Rimon B, Duchin S, Bernstein N, Kamenetsky R. 2019. Architecture and florogenesis in female Cannabis sativa plants. Front Plant Sci. 10:350. https://doi.org/10.3389/fpls.2019.00350.
Stack GM, Toth JA, Carlson CH, Cala AR, Marrero-Gonzalez MI, Wilk RL, Genter DR, Crawford JL, Philippe G, Rose JK, Viands DR, Smart CD, Smart LB. 2021. Season long characterization of high cannabinoid hemp reveals variation in cannabinoid accumulation, flowering time, and disease resistance. Glob Change Biol Bioenergy. 13:546–561. https://doi.org/10.1111/gcbb.12793.
Stack G, Murdock M, Wares A, Snyder S, Toth J, Crawford J, Schessel M, Genna N, Phillipe G, Stansell Z, Rose J, Moore G, Smart L. 2022. Triploid pollen-challenge trials. https://hemp.cals.cornell.edu/resources/reports-factsheets/2022-cornell-hemp-field-day-handouts/triploid-pollen-challenge-trials. [accessed 24 Jul 2024].
Stevens B. 2023. Benefits of F1 cannabis hybrids from Royal Queen Seeds. Business of Cannabis. https://businessofcannabis.com/benefits-of-f1-cannabis-hybrids-from-royal-queen-seeds/. [accessed 24 Jul 2024].
Toth JA, Stack GM, Carlson CH, Smart LB. 2022. Identification and mapping of major-effect flowering time loci Autoflower1 and Early1 in Cannabis sativa L. Front Plant Sci. 13:991680. https://doi.org/10.3389/fpls.2022.991680.
Wright H. 1980. Commercial hybrid seed production, p 161− 176. In: Fehr WR, Hadley HH (eds). Hybridization of crop plants. American Society of Agronomy and Crop Science Society of America, Madison, WI, USA. https://doi.org/10.2135/1980.hybridizationofcrops.c8.
Zhang M, Anderson SL, Brym ZT, Pearson BJ. 2021. Photoperiodic flowering response of essential oil, grain, and fiber hemp (Cannabis sativa L.) cultivars. Front Plant Sci. 12:694153. https://doi.org/10.3389/fpls.2021.694153.