Index | Search | Home | Table of Contents

link to pdf file

Tomasi, P., D. Dierig, and G. Dahlquist. 2002. An ovule culture technique for producing interspecific Lesquerella hybrids. p. 208–212. In: J. Janick and A. Whipkey (eds.), Trends in new crops and new uses. ASHS Press, Alexandria, VA.

An Ovule Culture Technique for Producing Interspecific Lesquerella Hybrids

Pernell Tomasi, David Dierig, and Gail Dahlquist


Lesquerella [Lesquerella fendleri (Gray) Wats., Brassicaceae], is a potential new oilseed crop, producing an unsaturated, hydroxy fatty acid (lesquerolic acid, C20:1-OH) with properties similar to imported castor oil (Dierig et al. 1993). The cultivated L. fendleri has demonstrated a narrow range of genetic variation determining hydroxy fatty acid seed oil content (Dierig et al. 2001). Germplasm evaluations completed at the USDA, Agricultural Research Service, US Water Conservation Laboratory revealed several other Lesquerella species having significantly elevated hydroxy fatty acid contents or different types of hydroxy fatty acids other than that found primarily in L. fendleri (Table 1). These wild species may be suitable donors for possible novel gene introgression by interspecific hybridization.

Table 1. Chromosome and oil characteristics of selected Lesquerella species evaluated at the USWCL, Phoenix, Arizona.

Lesquerella species n = x Content (%)
Oil content Densipolic acid
Lesquerolic acid
Auricolic acid
L. auriculata 8 33.0 1.60 8.1 17.4
L. fendleri 6 23.6 0.89 50.2 trace
L. lindheimeri 6 21.6 1.36 81.7 0.35
L. lyrata 8 25.8 26.9 trace trace
L. pallida 6 NAz trace 81.4 3.68

zNA=data not available

Several barriers to interspecific hybridization exist. The first, some type of late-acting sporophytic self-incompatibility, prevents fertilization and silique development following hand pollinations. Simple bud pollinations performed several days before floral anthesis circumvents this minor problem and allows for embryo fertilization. Secondly, an undetermined type of post-zygotic rejection further prevents the embryo from developing into an ovule and subsequent mature seed (Table 2), even with normal silique formation (parthenocarpy). Kumar et al. (1988) have demonstrated in Brassica fruticulosa × B. campestris interspecific hybrids that the utilization of ovule culture can usually overcome this ovular inhibition, thus producing sufficient F1 hybrid plants for further breeding purposes.

Table 2. Number of putative hybrid plants surviving greenhouse germination from reciprocal interspecific crosses between L. fendleri and three other Lesquerella species completed in Spring 1999.

Hybrid cross Line No. buds
No. seeds
No. seeds/
No. plants Fertility
L. auriculata L. fendleri A F 80 2 0.025 0 NAz
L. fendleri L. auriculata F A 208 6 0.029 0 NA
L. pallida L. fendleri P F 246 79 0.321 1 sterile
L. fendleri L. pallida F P 1069 3 0.003 1 fertile
L. lindheimeri L. fendleri L F 969 13 0.013 0 NA
L. fendleri L. lindheimeri F L 2280 11 0.005 1 fertile

zNA=data not available

Hybrids resulting from interspecific crosses usually exhibit various forms of partial to full male sterility due to incomplete homology between the parental species utilized (Kalloo and Chowdhury 1992). Sterility is not a severe limitation, as long as some type of fertility can be restored by inducing amphidiploidy with colchicine or by further producing fertile backcross generations from the sterile hybrid intermediaries. Colchicine treatments have systematically been utilized to induce amphidiploidy and restore fertility between different species of Brassica (Kumar and Shivana 1991), wide intergeneric hybrids involving Diplotaxis erucoides (Vylas et al. 1995) and with clover (Trifolium sp.) (Anderson et al. 1990).

The objectives of this study were twofold. The first was to develop an ovule culture protocol specific to Lesquerella for the regeneration of immature ovules from interspecific hybrid crosses and, if necessary, from subsequent backcross generations. The second objective was to restore fertility in the male sterile hybrid intermediaries by inducing amphidiploidy with in vitro colchicine treatments.


Two species, L. lindheimeri and L. pallida, were hybridized with L. fendleri because of the elevated concentration of lesquerolic acid (C20:1-OH) present in their seed oil (Table 1). Two other more distantly related species, L. auriculata and L. lyrata, were also hybridized with L. fendleri because of the predominately different hydroxy fatty acids present in their seed oil (C20:1-OH and C18:2-OH, respectively). Hybrid crosses are referred to with line names A × F (L. auriculata × L. fendleri), L × F (L. lindheimeri × L. fendleri), Ly × F (L. lyrata × L. fendleri), P × F (L. pallida × L. fendleri) and reciprocal crosses have respectively reversed line names.

All flowers past anthesis (if any) were removed from three to five inflorescences of each maternal plant used in this study. Bud pollinations were accomplished by removing approximately one-eight to one-fourth of the top of the calyx without damaging the emerging stigma. Pollen was immediately applied to the exposed stigma from a dehisced anther of the selected donor parent. Swollen siliques were excised 7 to 13 days after pollination (DAP) and aseptically dissected in the laboratory. Developing ovules (both healthy and shriveled) were placed into 60 × 15 mm plates containing 5 mL of Murashige and Skoog (MS) media supplemented with 0.5 gL-1 casein hydrolysate and 1.0 mgL-1 gibberellic acid, at 25°C under continuous illumination. After 4 weeks, germinated ovules were transferred to 100 × 25 mm plates containing MS media supplemented with 1.0 mgL-1 kinetin and 4.25 mgL-1 silver nitrate (AgNO3) for shoot initiation. The meristems of recently initiated shoot explants were then completely submerged in a MS shoot media containing 0.1% colchicine for 48 hr to overcome male sterility by inducing amphidiploidy. Surviving explants were rooted in 100 × 25 mm plates containing MS media supplemented with 0.1 mgL-1 auxin (NAA) and transferred to Jiffy peat pellets in the greenhouse. This process was repeated on the F1 interspecific hybrids, utilizing primarily a L. fendleri, high oil germplasm line, WCL-LY2 (Dierig et al. 2001), as the paternal donor, to produce several BC1 and one BC2 backcross generation.


Putative F1 Hybrids

Two of the three seeds harvested in spring 1999 from the first set of interspecific crosses (Table 2) were later germinated, grown, and confirmed morphologically as true F1 hybrids (data not shown). The third was determined to be a maternal self produced most likely from contamination because of an ineffective emasculation procedure. One of the confirmed hybrid plants was sterile (P × F03) and the other partially fertile (F × L11). Five shoot explants of P × F03 were in vitro colchicine treated and fertility was restored in a majority of the clones produced. These fertile clones were subsequently selfed, but not enough mature seed was produced for fatty acid analysis. The F × L11 F1 hybrid was partially fertile, thus it was selfed and the harvested seed analyzed for fatty acid content. The lesquerolic acid % was not significantly different than that of the selfed maternal parent.

A sufficient number of putative F1 hybrids were produced in fall 1999 and spring 2000 from ovule cultures between L. fendleri and the four other wild species utilized (Table 3). No reciprocal cross of Ly × F was attempted. The F × L survival percentage is low because many of the plants were eliminated due to labor constraints. Chromosomal mispairing may account for the low survival percentage in both the A × F and Ly × F crosses. Endogenous contamination and transplant shock effected most of the cross groups similarly. All of the crosses in Table 3 were treated with colchicine (except F × A). None of the A × F or F × L crosses have responded favorably to date, thus all the plants in both cross groups remain sterile. The L × F and F × P crosses have responded with some limited fertility restoration depending on family. We expect the P × F cross to respond favorably to colchicine, considering that the closely related P × F03 F1 hybrid had earlier in spring 1999. The only two surviving Ly × F F1 hybrid plants have not flowered to date, so their fertility status remains undetermined. We suspect they may respond similarly (by remaining sterile) as the A × F F1 hybrid did because L. lyrata has the same chromosomal number as L. auriculata and is also more closely related.

Table 3. Number of putative hybrids established in the greenhouse from in vitro ovule cultures of seven different F1 interspecific Lesquerella crosses completed in Fall 1999 and Spring 2000.

Hybrid cross Line No.
No. ovules
No. putative
hybrids established
L. auriculata
L. fendleri
A F 100 458 46 10.0 3 plants from 2 families 6.5
L. fendleri
L. auriculata
F A 91 684 1 0.15 0 plants from 1 family 0
L. lindheimeri
L. fendleri
L F 134 514 113 21.9 20 plants from 2 families 17.7
L. fendleri
L. lindheimeri
F L 399 2646 593 22.4 1 plant from 1 family 0.2
L. lyrata
L. fendleri
Ly F 14 66 8 12.1 2 plants from 1 family 25.0
L. pallida
L. fendleri
P F 36 230 139 60.4 26 plants from 2 families 18.7
L. fendleri
L. pallida
F P 155 403 60 14.9 8 plants from 2 families 13.3

BC1 and BC2 Backcross Generations

First and second backcross generations were also successfully produced from ovule culture in fall 2000 and spring 2001 (Table 4). Ovule germination rates ranged from 0% to 40%, depending on the species involved in the cross and generational stage. None of the (A × F) × F BC1 ovules that were cultured germinated. This was not entirely unexpected, as most of the dissected ovules were shriveled when they were removed from the deformed siliques. We suspect that the majority of embryos within the ovules were never fertilized due to a possible lack of genetic homology and related sterility. We have completed another colchicine treatment on the three A × F F1 hybrid plants in anticipation of inducing amphidiploidy, possibly restoring fertility and performing more BC1 backcrosses in the future breeding season.

Table 4. Number of seeds harvested and putative hybrids established in the laboratory from subsequent in vitro ovule cultures of seven different interspecific Lesquerella BC1 or BC2 backcrosses completed between fall 2000 and spring 2001.

Hybrid backcross Line Gen. No. buds
No. seeds
germ. %
No. hybrid
(L. auriculata L. fendleri) L. fendleri (A F) F BC1 246 0 471 0 0 0 from 7 families
(L. lindheimeri L. fendleri) L. fendleri (L F) F BC1 2598 4 407 46 11.3 20 from 9 families
[(L. lindheimeri L. fendleri) L. fendleri] L. fendleri [(L F) F] F BC2 4137 170 1635 134 8.2 91 from 19 families
(L. pallida L. fendleri) L. fendleri (P F) F BC1 809 2 411 80 20.4 77 from 6 families
(L. pallida L. fendleri) L. pallida (P F) P BC1 29 0 20 8 40.0 8 from 1 family
(L. fendleri L. pallida) L. fendleri (F P) F BC1 86 0 142 19 13.4 17 from 2 families
(L. fendleri L. pallida) L. pallida (F P) P BC1 41 0 42 8 19.0 5 from 1 family

All 20 plants from the 9 families of the (L × F) × F BC1 cross were derived from noncolchicine treated, sterile maternal parents. This may be why the ovule germination percentage (11.3%) is lower than all of the (P × F) × (F or P) BC1 crosses that were derived from colchicine treated, fertile maternal parents. Included in Table 4 is a column of the number of seeds harvested. Comparing the (L × F) × F BC1 and [(L × F) × F] × F BC2 hybrid cross groups, one notices that the number of seeds harvested increased. This may be due to a genome stabilizing effect that occurs with each ensuing backcross generation. A sample of seeds from the [(L × F) × F] × F BC2 group was analyzed for fatty acid concentration. The lesquerolic acid content of 35% was low compared to 55% to 60% in other samples and was likely due to the seed not developing fully within the siliques. Further BC3 and/or BC4 generations will need to be produced so more fatty acid content analyses can be completed.


A successful ovule culture protocol may allow breeders to introgress increased variation of fatty acid profile into L. fendleri from related wild species through hybridization and backcrossing. Distantly related species with different chromosomal numbers than L. fendleri, such as: L. auriculata and L. lyrata, may need diligent and repeated colchicine treatments on broader sample sizes (more families) to successfully attain fertile F1 hybrid intermediaries. Further research is needed on the morphological and cytological properties of these hybrids. These techniques have the potential to greatly improve the genetic variability of the cultivated species of Lesquerella and through further selection improve the commercialization efforts of this new crop.