Skip to main content
NCBI home page
Springer logoLink to Springer
. 2026 May 13;116(3):51. doi: 10.1007/s11103-026-01717-7

Development of efficient Agrobacterium-mediated transformation of a second-generation biofuel crop, Brassica carinata

Reshma Roy 1, Shilpika Kassety 2, Nelson Gororo 3, Rick Bennett 3, Anthony A Millar 1,✉, Srinivas Belide 2,✉
PMCID: PMC13167906  PMID: 42120638

Abstract

Brassica carinata (B. carinata) is a promising non-food biofuel crop with the potential for low carbon intensity biofuel production under suitable management conditions. However, B. carinata requires extensive genetic improvement of many agronomic traits for it to be widely deployed into agriculture systems. Its rapid improvement via gene editing or genetic modification requires robust, genotype-flexible transformation protocols, as current methods are limited and have only been applied to a narrow germplasm base, hindering broader genetic enhancement efforts. In this paper, cotyledonary petioles from five diverse B. carinata genotypes were evaluated for shoot regeneration under different cytokinin treatments. Among the genotypes tested, regeneration efficiency varied substantially across the five B. carinata genotypes. Transient expression of the GUS reporter gene in cotyledonary petioles confirmed high transformation proficiency, with expression frequencies varying from 96.4 to 100% among four genotypes. Consistently, high stable transformation efficiency was achieved for these four genotypes of at least 5%, with the highest efficiency being 12%. Interestingly, transformation and regeneration efficiencies did not correlate, indicating the latter does not serve as a reliable indicator of transformation efficiency when subjected to selection pressures. Transgene integration was validated by PCR and copy number analysis was performed using digital droplet PCR. Furthermore, T1 seedlings derived from selected transgenic events demonstrated resistance to glufosinate ammonium whereas wild-type seedlings exhibited mortality. Overall, this study presents a significantly improved and reproducible Agrobacterium-mediated transformation method for B.carinata, advancing the potential for rapid genetic improvement and trait development in this promising biofuel crop.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11103-026-01717-7.

Keywords: Non-food biofuel crop, Molecular farming, Herbicide resistance

Key message

We established an efficient and reproducible B. carinata regeneration and transformation system that works reliably across multiple genotypes.

Supplementary Information

The online version contains supplementary material available at 10.1007/s11103-026-01717-7.

Introduction

Brassica carinata (B. carinata), commonly known as Ethiopian or Abyssinian mustard, is a species of the Brassicaceae. It is one of the oldest cultivate leafy vegetables and oilseed crops originating from East Africa, particularly from Ethiopia (Maxted et al. 1997; Vaughan and Geissler 2009). Recently, B. carinata has emerged as a promising oilseed crop with considerable potential to support sustainable agriculture and low-emission feedstocks for renewable fuels and bioproducts (Seepaul et al. 2021). Unlike many conventional oilseed crops, B. carinata demonstrates strong abiotic stress tolerance, attributed to its robust taproot system, high water-use efficiency, and effective nutrient uptake mechanisms (Zoong Lwe et al. 2021; Limbalkar et al. 2021). This supports the maintenance of stable biomass and reproductive production in situations where water availability is low, positioning it as a prime choice for cultivation in resource-restrained agricultural environments (Gesch et al. 2015; Seepaul et al. 2019).

Complementing its agronomic resilience, B. carinata is instrumental in fostering environmental sustainability. As a winter cover crop, B. carinata could be integrated into double-cropping practices where it could potentially support soil health without compromising food yield (Bennett 2022). Moreover, its extensive root architecture fosters carbon sequestration in soils, while its application as a feedstock for sustainable aviation fuel (SAF) provides significant reductions in greenhouse gas emissions within the aviation industry (Karami et al. 2024). The oil extracted from B. carinata seeds, characterized by high erucic and linolenic acid content, is well-suited for conversion into drop-in fuels, including SAF and renewable diesel, which are compatible with existing fuel infrastructure and engines (Seepaul et al. 2021). In addition, the protein-rich meal left after oil extraction serves as a high-value livestock feed, while other industrial applications such as bioplastics, biolubricants, and biopesticides, expand its bioeconomic potential (Taylor et al. 2010).

As the interest in B. carinata is increasing, facilitating its extensive applicability through modern biotechnological methodologies becomes imperative. Agrobacterium-mediated transformation will help in contributing to establish a robust framework for functional genomics and targeted trait improvement in B. carinata. However, there have only been limited reports on the genetic transformation of this crop on a narrow range of genotypes (Babic et al. 1998; Chaudhary et al. 1998).

Based on the limited published reports, the type of explants and the selection agent used in the transformation has profound impact on the transformation efficiency (Babic et al. 1998; Chaudhary et al. 1998). Chaudhary et al. (1998) reported an efficiency of 22% for transformation in B. carinata Braun (C90 − 1089) by using cotyledonary petioles as explants with kanamycin as the selection agent. Similarly, Babic et al. (1998) reported a 30–50% transformation efficiency when kanamycin was used as the selection agent. However, the efficiency dropped dramatically to just 1–2% when L-phosphinothricin (L-PPT) was used as plant selectable markers (Babic et al. 1998) demonstrating the profound impact of selection agent choice on transformation outcomes. Both cotyledonary petioles and hypocotyls were used as explants; however, cotyledonary petioles responded best, while the transformation efficiency for hypocotyls was not quantified but was noted to be lower (Babic et al. 1998). In both these studies, the specific genotypes that were used was C90 − 1089 and C90 − 1163, both developed by Canadian researchers. Early regeneration studies in B. carinata demonstrated successful shoot regeneration from hypocotyl explants (Yang et al. 1991), and complete plant regeneration from hypocotyl-derived protoplasts (Narasimhulu et al. 1992). These studies demonstrated the in vitro regeneration and Agrobacterium-mediated transformation of B. carinata, with Babic et al. (1998) establishing a stable transformation protocol confirmed by molecular analyses and T1 inheritance. Subsequently, Agrobacterium-mediated transformation was applied to trait modification, including hairpin RNA-mediated silencing of the endogenous FAD2 gene combined with heterologous Crambe abyssinica FAE expression, which significantly increased erucic acid content in B. carinata seeds (Mietkiewska et al. 2008). More recently, simplified in planta approaches such as the floral dip method have been demonstrated in B. carinata cv. Pusa Gaurav, achieving an approximate 1.5% transformation efficiency, however, there have been no follow up reports (Verma et al. 2008). A comprehensive in vitro regeneration screen of 51 B. carinata accessions revealed wide variation in shoot regeneration per explant among genotypes (Gil-Humanes et al. 2011). Collectively, these studies confirm that while B. carinata is amenable to Agrobacterium-mediated transformation, success remains highly genotype-dependent and constrained by the choice of explant, selection system, and transformation approach, limiting the broader application of existing protocols to diverse germplasm. Transformation protocols that demonstrate efficacy for a specific genotype may prove ineffective for other genotypes, thereby reducing the versatility and wider applicability of the methodology. For example, the protocol for one maize inbred line cannot be applied to another (Brettschneider et al. 1997). Follow-up studies demonstrating reproducibility, stable inheritance, or integration into modern genome-editing pipelines remain absent.

To address this critical bottleneck, we evaluated the regeneration potential of five B. carinata genotypes selected to represent a broad genetic background and phenotypic diversity, using cotyledonary petioles as explants to assess the effects of plant growth regulators (PGRs) at varying concentrations across different phases of organogenesis. Furthermore, we present the successful implementation of genetic transformation across four distinct genotypes utilizing Agrobacterium-mediated techniques, employing L-PPT as a selection marker. Our reproducible transformation method, effective across four genotypes using L-PPT as a selection marker is expected to facilitate genetic improvement and support the future development of B. carinata varieties through gene editing.

Materials and methods

Plant material

Seeds of B. carinata, DHN1 and DHN2 were obtained from Nufarm, Australia. Genotypes ATC 94023 (hereafter referred to as AGG1) and ATC 94024 (AGG2) were obtained from Australian Grains GeneBank, Horsham, Australia. Genotype 52648 (DW3) was provided by Dr Harsh Raman, NSW Department of Primary Industries (DPI), Wagga Wagga, Australia. B. carinata seeds were sterilised using chlorine gas. Briefly, 100 mL of commercial bleach (White King premium bleach, sodium hypochlorite 42 g/L; sodium hydroxide 9 g/L; chlorine 4% (w/v)) was mixed with 3 mL concentrated HCL in a desiccator jar and sterilised for 3.5 h. Sterilized seeds were aseptically transferred to 30 mL plastic containers and stored at room temperature for further use.

Seed germination and explant preparation

Approximately 15 sterilised seeds were inoculated onto ½ strength MS media (Murashige and Skoog 1962) with 0.8% agar (pH 5.8) and incubated in low cool white, fluorescent light at 25–30 µmol m−2 s−1 with a 16/8 h (light/ dark cycle) at 23–24 °C for five days. Cotyledonary petioles (CPs) measuring up to 5 mm in length were aseptically isolated and used for regeneration and transformation with Agrobacterium.

Initial screening of regeneration potential and optimisation

To assess the regeneration potential of the five B. carinata genotypes, CPs excised from five-day old seedlings were cultured on callus induction media (CIM) supplemented with 3% sucrose containing either 6-benzylaminopurine (BAP) (at either 0.5 mg/L or 1.0 mg/L) or thidiazuron (TDZ) (at either 0.5 mg/L or 1.0 mg/L) with three different antioxidants, L-cysteine (Sigma-Aldrich, Cat#C7352), ascorbic acid (Sigma-Aldrich, Cat#A7506) and glutathione Sigma-Aldrich, Cat#G6013) for two weeks after which the cotyledon portion of the CP was excised and removed while the petiole portion was reinoculated onto CIM and cultured for another two weeks (Fig. 1; Cycle 3). At the end of the CIM cycle, callusing petioles were transferred to shoot induction media (SIM) and cultured for another two weeks. Regenerating shoot buds and a small amount of callus were transferred to shoot outgrowth media (SOG) which is designed to promote elongation and further shoot development. The compositions of all culture media (CIM, SIM, and SOG), including growth regulators and antioxidant additives are detailed in Table 1.

Fig. 1.

Fig. 1

Open in a new tab

Schematic representation of the different steps of the B. carinata transformation protocol (Created in BioRender. Kaines, S. (2025) https://BioRender.com/undefined )

Table 1.

Composition of media used in different steps of carinata transformation from cotyledonary petioles

Component
(mg/L)		 Cycle 1 Cycle 2 Cycle 3 Cycle 4 Cycle 5 Cycle 6 Cycle 7
Co-cultivation Pre-Selection Callus induction (CIM) Shoot initiation (SIM) Shoot outgrowth (SOG) Shoot elongation Root induction
Thidiazuron (TDZ) b 1.0 1.0 1.0 – – – –
1-Naphthaleneacetic acid (NAA) 0.05 0.05 0.05 – – – –
Kinetin (Kn) – – – 1.0 – – –
Indole-3-butyric Acid (IBA) – – – – – – 1.0
Iota-carrageenan a – 150 150 – – –
CaCl2·2 H2O a – – 10 10 10 10 –
Gibberellic acid (GA3) – – – – 0.5 0.5 –
L-cysteine (LC) 50 50 – – – – –
Ascorbic acid (AA) 15 15 – – – – –
Acetosyringone (AS) 16 – – – – – –
Glutathione (Reduced) – 19 19 19 19 19 19
PPT – – 5.0 5.0 5.0 5.0 5.0
Cefotaxime – 250 250 250 250 250 250
Timentin – 50 50 50 50 50 50
Silver Nitrate (AgNo3) – 3.0 3.0 1.5 1.5 1.5 1.5
Sucrose (g/L) 30 30 30 30 20 20 20
Agar Agar (g/L) 9.0 9.0 9.0 9.0 9.0 9.0 9.0
Open in a new tab

a added to AGG1 genotype in hyperhydration control experiment

b TDZ (1.0 mg L−1) was constant in transformation media but varied in regeneration treatments

Agrobacterium co-cultivation with CP explants

The binary vector p1305.1 (Supplementary Fig. 1S A) that was derived from the pCAMBIA series was used in this study for GUS (β-glucuronidase) assays to test explant suitability for transformation. It contains the catalase intron in the GUSPlus-His6 reporter gene (β-glucuronidase) driven by the CaMV35S promoter and contains the NOS terminator (https://www.snapgene.com/plasmids/plant_vectors/pCAMBIA1305.1). To generate stable transgenic B. carinata, the binary vector pBSV-004 was used (Supplementary Fig. 1S B; Petrie et al. 2012) and contains the phosphinothricin acetyltransferase (PAT) gene, which confers resistance to glufosinate ammonium (Basta resistance). Binary vectors (p1305.1 and pBSV-004) were transformed into Agrobacterium (AGL1) (Nagel et al. 1990). Single colonies of Agrobacterium grown on LB agar plates were inoculated into 10 mL of LB liquid medium containing kanamycin (50 mg/L) and rifampicin (25 mg/L) and cultured overnight at 28ᵒC with shaking at 150 rpm. Agrobacterium cells were processed as described by Belide et al. (2011) and were used to infect freshly isolated CPs as described by Belide et al. (2013). Briefly, overnight grown bacterial cells were pelleted, washed and re-suspended in 10 mL of MS media (adjusted to OD 600 nm = 0.45) with 2% glucose and grown with 100 µM acetosyringone (AS) for four hrs at 28ᵒC. For optimum vir gene induction, spermidine was added to a final concentration of 1 mM and grown for additional two hours as described by Belide et al. (2011 and 2013). Freshly isolated CPs (approximately 30) were infected with 12.5 mL of this Agrobacterium culture for 5–6 min. Infected explants were blotted on sterilised filter paper to remove the excess bacteria and then cultured in co-cultivation media (Table 1; Cycle 1) in the dark at 23–24ᵒC for two days (Fig. 1; Cycle 1).

Histochemical GUS assay

Histochemical GUS assays were performed on CPs as previously described (Jefferson, 1987). Agrobacterium-infected CPs were submerged in freshly prepared X-Gluc staining solution (2 mM X-Gluc dissolved in Dimethylformamide 0.1% [v/v] Triton X-100, and 0.1 M NaPO₄ buffer, pH 7.0) and vacuum-infiltrated for approximately 60 s. Samples were then incubated overnight at 37 °C in the dark. Following staining, the X-Gluc solution was removed, and tissues were destained in 80% (v/v) ethanol at 24 °C for ~ 8 h to remove chlorophyll. Blue-stained CPs, indicative of GUS expression, were examined and imaged using a Leica stereoscopic zoom microscope (M60). CPs that stained at the bottom of the petiole region was scored as positively transformed.

Selection and regeneration of putative transgenic shoots from CPs

Two days after co-cultivation with Agrobacterium, CPs were transferred to pre-selection media (Cycle 2; Table 1) and cultured for 3–4 days in the light (16 h, 45–50 µmol m−2 s−1) at 23–24 °C. After 3–4 days of culture the infected CPs were then transferred to CIM (Cycle 3; Table 1) and cultured for two weeks in cool light white, fluorescent light (16 h, 45–50 µmol m−2 s−1) at 23–24 °C. After two weeks of culture on CIM, the expanded cotyledon portion was removed, and callusing petioles were transferred to fresh CIM (Fig. 1, Cycle 3). After two weeks, callusing petioles were transferred to SIM (Cycle 4; Table 1), and then subsequently to SOG (Cycle 5; Table 1) and cultured for another two weeks. For the AGG1 genotype, Iota-carrageenan and CaCl2·2 H2O were added to assess the impact on hyperhydration. These additives were not components of the standard SOG but were applied as supplementary treatments. Shoot initials (approximately 1 cm) from each CP were isolated and cultured on shoot elongation media (Cycle 6; Table 1) for another four weeks (sub-culture after two weeks) as described above. All shoots from an individual CP are considered as clones and were maintained accordingly. Green and healthy shoots (2–3 cm) were transferred to root induction media (Cycle 7; Table 1) and cultured for another four weeks or until profuse roots were observed from the cut end of the shoot. Shootlings with well-established roots were transferred to the glasshouse and plants were grown to maturity. A schematic representation of different steps of this method is illustrated in Fig. 1.

DNA extraction and transgene copy number analysis by digital droplet PCR (ddPCR)

Genomic DNA from putative transgenic shoots was isolated as described by Mieog et al. (2013) and the transgene (PAT) copy number was determined as described Petrie et al. (2020). The High-mobility group (HMG) gene, a single copy in the genome of Brassica species (Weng et al. 2004), was used as a reference. Digital droplet PCR (ddPCR) data from putative transgenic shoots were analysed using QuantaSoft™ software (Bio-Rad) to determine the copy number of the PAT transgene.

Glufosinate ammonium foliar application and herbicide resistance conformation

To test for herbicide resistance, T1 seeds from three different B. carinata genotypes with varying transgene copy numbers (which may influence PAT expression and resistance level), along with non-transformed seeds, were sown in white trays containing soil and grown under controlled conditions at 24 °C with a 16/8 h light/dark photoperiod for approximately three weeks. A commercial glufosinate ammonium herbicide (SureFire, PCT holdings Pty Ltd) was diluted to 0.05% and sprayed uniformly until the leaf surfaces were visibly wet to ensure complete foliar coverage of both non-transformed parental lines and T1 transgenic seedlings. The herbicide application was repeated 2–3 weeks after the initial spray.

Statistical analysis

Data from regeneration and transformation experiments, including means and standard deviations, were analyzed using SigmaPlot (Version 12). Differences in regeneration and transformation frequencies between genotypes and treatments were considered statistically significant at P < 0.05.

Results and discussion

Transient GUS expression confirms transformability of cotyledonary petioles

To evaluate the transformability of CPs across genotypes, we conducted a transient GUS expression assay in four B. carinata genotypes. CPs were selected as explants for this study due to their proven efficiency in Agrobacterium-mediated transformation across Brassica species, including canola, and B. carinata, where they consistently support high regeneration and transformation frequencies (Cardoza and Stewart 2003; Bhalla and Singh 2008; Chaudhary et al. 1998). High expression frequencies (96.4%–100%) for all four genotypes tested were observed (Table 2). The remarkable consistency of transient GUS expression denotes successful Agrobacterium-mediated infection and the transfer of T-DNA across the four genotypes, highlighting CPs are suitable explant for transformation. The construct (p1305.1) used in this study contains a catalase intron within the GUS coding sequence, which prevents reporter gene expression in Agrobacterium and restricts GUS activity to eukaryotic plant cells. This design ensures that histochemical staining reflects plant-derived expression rather than bacterial background, thereby validating the specificity of the assay and justifying the omission of an antibiotic wash step prior to staining.

Table 2.

Transient GUS gene expression in CPs of different B. carinata genotypes

Genotype Total number of explants tested Frequency of GUS expression
AGG1 19 100
AGG2 15 100
DHN1 22 100
DW3 28 96.4
Open in a new tab

Optimization of cytokinin concentrations for callus-derived shoot regeneration

CPs from five-day old seedlings (five genotypes) grown in low cool white, fluorescent light (25–30 µmol m−2 s−1) were isolated and inoculated onto CIM supplemented with either BAP or TDZ, at two different concentrations (four treatments in total). In earlier regeneration studies of B. carinata, BAP was the main cytokinin employed, either alone (Gil-Humanes et al. 2011) or in combination with auxins such as picloram (Narasimhulu et al. 1992). BAP and TDZ were selected due to their proven efficacy in inducing shoot regeneration in Brassica species (Ravanfar et al. 2017; Yu et al. 2010). Around 15 days after inoculation, dark green callus formation was noticed around the CPs. The cotyledon portion (upper portion) of the explant which became brown was excised and the callusing petiole portion of the explant was reinoculated onto CIM and grown for another two weeks. After a total of one month on CIM, most of the explants from all the genotypes had formed green organogenic callus. Regeneration efficiency, calculated as (number of shoots regenerated / number of explants inoculated) × 100, showed that BAP appears to slightly outperform TDZ, reaching closer to 100% callus induction frequency in three of the genotypes (Table 3).

Table 3.

Effect of different concentration of cytokinins on regeneration efficiency using CPs. The numbers of inoculated explants represent pooled data from multiple independent Agrobacterium-mediated transformation experiments

Genotype Code of the treatment Cytokinin (mg/L) Number of explants inoculated Number of explants responded at C3 Callus induction frequency at C3 Average number of shoots / CP at C5 stage a
AGG1 Tr-1 0.5 TDZ 46 44 95.6 8.6 ± 5.3
Tr-2 1.0 TDZ 47 46 97.8 6.7 ± 2.0
Tr-3 0.5 BAP 46 46 100 5.7 ± 3.1
Tr-4 1.0 BAP 45 45 100 6.5 ± 3.4
AGG2 Tr-1 0.5 TDZ 36 36 100 16.9 ± 7.4
Tr-2 1.0 TDZ 36 36 100 19.6 ± 5.3
Tr-3 0.5 BAP 36 33 91.6 12.0 ± 5.8
Tr-4 1.0 BAP 36 36 100 12.9 ± 5.3
DHN1 Tr-1 0.5 TDZ 22 21 95.4 12.5 ± 6.7
Tr-2 1.0 TDZ 22 21 95.4 10.1 ± 4.7
Tr-3 0.5 BAP 22 22 100 8.7 ± 4.1
Tr-4 1.0 BAP 23 23 100 8.7 ± 4.1
DW3 Tr-1 0.5 TDZ 39 39 100 4.2 ± 4.4
Tr-2 1.0 TDZ 39 39 100 3.0 ± 3.0
Tr-3 0.5 BAP 34 34 100 2.1 ± 2.1
Tr-4 1.0 BAP 35 34 97.1 4.2 ± 1.7
DHN2 Tr-1 0.5 TDZ 20 16 80.0 3.4 ± 1.4
Tr-2 1.0 TDZ 20 16 80.0 4.4 ± 2.0
Tr-3 0.5 BAP 20 19 95.0 4.5 ± 2.8
Tr-4 1.0 BAP 56 34 64.2 3.4 ± 2.3
Open in a new tab

a Data represent the mean ± standard deviation (SD) from three biological replicates, with each replicate consisting at least six explants cultured under identical conditions per treatment

Among the tested genotypes, AGG2 demonstrated the highest regeneration potential, with the TDZ (Tr-2: 1.0 mg/L) treatment yielding an average of 19.6 ± 5.3 shoots per explant, with all explants producing shoots (100% regeneration rate). The BAP treatments also worked well on AGG2, producing around 12.0 shoots per CP. Similarly, DHN1 had a strong regeneration in all treatments, with the TDZ (0.5 mg/L) treatment being best, producing 12.5 ± 6.7 shoots per explant. On the other hand, DW3 and DHN2 had much lower regeneration potential across all treatments, with an average of less than five shoots per CP. Although DW3 has a 100% rate of callusing under most treatments, it only produced an average of 2.1–4.2 shoots per explant, suggesting that although it is capable of initiating organogenesis, the subsequent formation of shoots is limited. AGG1 demonstrated a moderate performance and achieved almost complete callusing rates (95.6–100%) for all treatments but produced fewer shoots per explant than AGG2 and DHN1.

The Tukey HSD post-hoc test was performed to identify which cytokinin treatments produced statistically significant shoot regeneration responses among the tested genotypes. The results revealed that AGG2 exhibited the most significant response, with TDZ treatments (Tr-1 and Tr-2) notably outperforming BAP treatments. Significant differences were observed between Tr-1 versus Tr-3 (p = 0.03), Tr-2 versus Tr-3 (p < 0.001), and Tr-2 versus Tr-4 (p = 0.01). AGG1 also showed modest but statistically significant improvement with TDZ (Tr-1) compared to BAP treatments. Among the tested genotypes, DW3 had poor overall regeneration; however, Tr-3 (0.5 mg/L BAP) was significantly inferior compared to Tr-1 (p = 0.03). In contrast, DHN1 and DHN2 showed no statistically significant differences between treatments.

Our results show that B. carinata’s regeneration potential depends heavily on the genotype, which is consistent with previous research (Gil-Humanes et al. 2011). The present study not only substantiated the applicability of CPs but also yielded quantitative data regarding the shoot regeneration frequency across five previously uncharacterized genotypes with respect to their in vitro regeneration response. The maximum efficiency of callus induction and shoot formation per explant of AGG2 and DHN1 is much higher than that of existing B. carinata reports, indicating the potential superiority of these genotypes and/or optimal media conditions in this study. Narasimhulu et al. (1992) achieved 22–74% regeneration from hypocotyl protoplasts, with a maximum of ~ 6 shoots per callus. Similarly, Gil-Humanes et al. (2011) evaluated 51 genotypes using cotyledonary explants and found an average of 2.9 shoots per explant, with only a few genotypes exceeding 6 shoots. By contrast, our study consistently achieved near-complete callus response and markedly higher shoot numbers. The inclusion of antioxidants, the precise timing of subculturing explants onto fresh medium and the adaptation of type of cytokinin may have contributed to the improved organogenic responses observed.

The generation of stable transgenic B. carinata following regeneration optimization

Next, we attempted stable Agrobacterium-mediated transformations on four B. carinata genotypes using the process shown in Fig. 2. Excised CPs from five-day-old seedlings (Fig. 2a) were co-cultivated with Agrobacterium harbouring the binary vector pBSV-004 (Petrie et al. 2012), which contains the PAT gene conferring Basta (glufosinate ammonium) resistance. For regeneration, the treatment chosen was Tr-2, as this was the best or second-best treatment for most of the genotypes, and was added to the co-cultivation, pre-selection and callus induction medias (Cycle 1, 2 & 3; Table 1). Co-cultivated CPs remain green after 2 days of growth on co-cultivation media (Fig. 2b), whereas CPs from non-transformed wild type were bleached and shrivelled after 2 weeks on CIM (Fig. 2c), confirming the appropriate concentration of selection was being used. The green CPs from co-cultivation media is transferred to pre-selection media and is incubated for 2–3 days. Towards the end of 2 weeks in CIM, green callus infused with shoots appeared (Fig. 2d), which further developed into shoots after 2 weeks on SOG media (Fig. 2e). After rooting, plants were transferred to soil (Fig. 2f and g), and regenerated plants were assayed for the presence of the transgene via PCR to test for the presence of the PAT gene. Based on this, the transformation efficiency for each genotype was calculated based on the number of transgene positive regenerated shoots (Table 4). All PCR-positive shoots successfully developed roots on rooting induction medium, and no losses were observed during the rooting stage.

Fig. 2.

Fig. 2

Open in a new tab

Selection and regeneration of transgenic B. carinata plants. a 5-day old, etiolated seedlings were the source of CPs; b Transformed CPs on co-cultivation media, c Negative control: non-transformed wild type (AGG2) CPs on CIM with 5 mg/L PPT; d Transformed CPs producing green shoots on SIM with 5 mg/L PPT; e Multiple shoots induced from transformed CPs on SOG media with 5 mg/L PPT; f &g Mature transgenic plants (AGG2) (6 and 10 weeks respectively)

Table 4.

Transformation efficiency of different genotypes of B. carinata

Genotype Total number of explants transformed Number of PCR positives shoots obtained Transformation efficiency (%) b
AGG1 95 1 1
659 a 33 5
AGG2 80 4 5
DHN1 130 15 12
DW3 35 3 9
Open in a new tab

a with the addition of Carrageenan and CaCl2

b Transformation efficiency values are rounded to the nearest whole number

It was found that the transformation efficiency varied greatly from genotype to genotype. DHN1 shows the highest transformation efficiency (12%). DW3 also performed moderately well (9%) despite its lower regeneration capacity (Table 3), whereas AGG2 had the best regeneration capacity (Table 3), however its transformation efficiency was only 5%. On the other hand, AGG1 showed a very low transformation efficiency (1%), and this was possibly due to the calli displaying the effects of hyperhydration. Previously, the use of Iota-carrageenan was successful in the control of hyperhydration of calli in safflower and eucalyptus transformation protocols, and in both instances was able to increase the transformation efficiency (Belide et al. 2011; Whitehouse et al. 2002). Therefore, the transformation was repeated with Iota-carrageenan and CaCl2 added to the media (Cycle 3; Table 1) for the AGG1 genotype. This resulted in reducing the hyperhydration of calli without any detrimental effect on regeneration and led to a substantially improved transformation efficiency of 5%. Therefore, a transformation efficiency of at least 5% was achieved for each of the four B. carinata genotypes. Interestingly, while AGG2 and DHN1 exhibited the strongest regeneration capacity, transformation efficiency did not strictly follow the same trend. For instance, DHN1 and DW3, despite moderate to lower regeneration responses, displayed the highest transformation efficiencies (12% and 9%, respectively). By contrast, AGG2, which had the highest regeneration ability, achieved only a 5% transformation efficiency. The results of this study indicate that the regenerative capacity does not serve as a reliable indicator of successful transformation when subjected to selection pressures. While genotypes that show the potential to develop multiple shoots often facilitate Agrobacterium-mediated transformation, the relationship between regeneration efficiency and transformation rate still varies (Park et al. 2012).

Relationship between regeneration and transformation efficiency

Regeneration efficiency was uniformly high across all genotypes (95.4–100%) for Tr-2 (1 mg/L TDZ), indicating minimal genotypic variation in callus survival and shoot induction frequency. In contrast, the average number of shoots per callus piece at the Cycle 5 varied markedly among genotypes, ranging from 3.0 ± 3.0 in DW3 to 19.6 ± 5.3 in AGG2. Despite this variation, shoot regeneration capacity showed no positive correlation with transformation efficiency. Accordingly, AGG2, which exhibited the highest shoot proliferation, displayed a low transformation efficiency (5%), whereas DHN1 and DW3 achieved higher transformation efficiencies (12% and 9%, respectively) despite moderate to low shoot numbers. These results indicate that regeneration traits alone are poor predictors of transformation success, supporting the conclusion that genotype-specific factors governing T-DNA delivery, integration, or response to selection primarily determine transformation efficiency.

In B. carinata, very few studies have explicitly examined the relationship between regeneration and transformation efficiency. Babic et al. (1998) reported that high transformation frequencies in B. carinata were associated with efficient regeneration from transformed callus rather than from non-transformed cells. Importantly, their study also showed that most shoot buds developing directly from cotyledonary petioles within the first week of culture were not transformed, highlighting that early and prolific regeneration does not necessarily correspond to successful T-DNA integration.

Similar observations have been reported in other Brassica species, including canola (Brassica napus), where regeneration efficiency is best viewed as a permissive prerequisite rather than a reliable predictor of stable transformation efficiency. Widely used Agrobacterium-mediated transformation systems based on cotyledonary petiole or hypocotyl explants routinely achieve high regeneration frequencies once culture conditions are optimised; however, transformation efficiency continues to vary substantially among genotypes and is largely influenced by downstream processes such as Agrobacterium susceptibility, T-DNA delivery, stable integration, and survival under selection (Moloney et al. 1989; Bhalla and Singh 2008; Bates et al. 2017).

Accordingly, even when regeneration performance is consistently high across genotypes, a clear positive correlation between regeneration metrics (e.g. percentage regeneration or shoots per explant) and transformation efficiency is often weak or absent. This is likely because the principal limiting steps occur after, or independently of, shoot induction, including genotype-specific defence responses and differential sensitivity to selection pressure (Bhalla and Singh 2007; Cao Chu et al. 2020). Overall, the Brassica transformation literature supports the interpretation that achieving a high regeneration threshold improves experimental throughput by increasing the number of shoots available for screening but does not necessarily increase the proportion of independently transformed events, which is primarily shaped by genotype × infection and selection biology (Bates et al. 2017; Cao Chu et al. 2020).

Transgene copy number and herbicide resistance validation

Following B. carinata transformation, digital droplet PCR (ddPCR) was used to quantify the transgene copy number of the PAT gene in 28 independent T₀ lines of the four different transformed genotypes (Fig. 3). The transgene copy number value (CNV) ranged from one to eight, meaning that most transgenic plants have multiple insertion events, which is typical of Agrobacterium-mediated transformation (Jupe et al. 2018).

Fig. 3.

Fig. 3

Open in a new tab

Copy number of the PAT gene in 28 independent T0 events of four different genotypes

In order to evaluate whether the PAT transgene conferred herbicide resistance, two applications of 0.05% glufosinate ammonium was sprayed onto T1 plants from three different genotypes (AGG2, DHN1, and DW3). Following the applications, the transgenic lines remained green and continued to grow, whereas the non-transformed controls displayed chlorosis and ultimately died (Fig. 4a–c). This demonstrates that the PAT transgene confers resistance to glufosinate ammonium in B. carinata across successive generations.

Fig. 4.

Fig. 4

Open in a new tab

Effect of foliar spray of 0.05% glufosinate ammonium on transgenic T1 seedlings. WT seedlings are on the left side of the panel and transgenic T1 seedlings are on the right side of the panel with three different genotypes a DHN1, CNV=1.97; b DW3, CNV =2.67, c AGG2, CNV=1.98)

Conclusions

This investigation represents a notable enhancement in the genetic engineering toolkit for B. carinata, introducing a reproducible regeneration and transformation methodology utilizing CPs as explants. Through the systematic optimization of hormonal combinations (specifically cytokinin) and the incorporation of agents to alleviate tissue hyperhydration, we achieved high frequencies of shoot regeneration across multiple genotypes. The transformation efficiencies documented, supported by molecular assays and functional herbicide resistance in T1 progeny, substantiate the efficacy of this protocol across a genetically diverse background genotypes.

To our knowledge, this is the first report to systematically evaluate regeneration efficiency, transformation competence, molecular validation, and trait expression across multiple B. carinata genotypes. Beyond mere proof-of-concept, the established platform exhibits substantial potential for expediting trait enhancement in B. carinata for a variety of applications. It can be directly applied to engineer traits associated with climate resilience (e.g., tolerance to drought, frost, and salinity), as well as herbicide resistance and enhanced oil content and composition.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary Material 1 (14.9MB, docx)

Acknowledgements

The authors are grateful to The Australian Research Council Training Centre for Future Crops Development (IC210100047) for providing support to this project along with a PhD stipend that was awarded to Reshma Roy.

Author contributions

Reshma Roy designed the study, performed all plant transformation experiments, collected and analysed the data, and drafted the manuscript. Shilpika Kassetty provided technical support with tissue culture maintenance. Nelson Gororo and Rick Bennett contributed domain knowledge and provided guidance on carinata agronomy and crop development. Anthony Millar provided critical scientific input, manuscript revision and editing. Srinivas Belide conceived the project, contributed to experimental design, supervised the transformation work, and contributed to manuscript drafting and editing. All authors read and approved the final manuscript .

Funding

Open access funding provided by CSIRO Library Services.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Anthony A. Millar, Email: tony.millar@anu.edu.au

Srinivas Belide, Email: Srinivas.belide@csiro.au.

References

  1. Babic V, Datla RS, Scoles GJ, Keller WA (1998) Development of an efficient Agrobacterium-mediated transformation system for Brassica carinata. Plant Cell Rep 17:183–188. 10.1007/s002990050375 [DOI] [PubMed] [Google Scholar]
  2. Bates R, Craze M, Wallington EJ (2017) Agrobacterium-mediated transformation of oilseed rape (Brassica napus). Curr Protoc Plant Biol 2(4):287–298. 10.1002/cppb.20060 [DOI] [PubMed] [Google Scholar]
  3. Belide S, Hac L, Singh SP, Green AG, Wood CC (2011) Agrobacterium-mediated transformation of safflower and the efficient recovery of transgenic plants via grafting. Plant Methods 7:12. 10.1186/1746-4811-7-12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Belide S, Zhou XR, Kennedy Y, Lester G, Shrestha P, Petrie JR, Sing SP (2013) Rapid expression and validation of seed-specific constructs in transgenic LEC2 induced somatic embryos of Brassica napus. Plant Cell Tiss Organ Cult 113:543–553. 10.1007/s11240-013-0295-1 [Google Scholar]
  5. Bennett R (2022) Carinata: An emerging biofuel feedstock platform. In: Proceedings of 2022 AOCS Annual Meeting & Expo
  6. Bhalla PL, Singh MB (2008) Agrobacterium-mediated transformation of Brassica napus and Brassica oleracea. Nat Protoc 3:181–189. 10.1038/nprot.2007.527 [DOI] [PubMed] [Google Scholar]
  7. Brettschneider R, Becker D, Lörz H (1997) Efficient transformation of scutellar tissue of immature maize embryos. Theor Appl Genet 94:737–748. 10.1007/s001220050473 [Google Scholar]
  8. Cao Chu U, Kumar S, Sigmund A, Johnson K, Li Y, Vongdeuane P, Jones TJ (2020) Genotype-independent transformation and genome editing of Brassica napus using a novel explant material. Front Plant Sci 8:11:579524. 10.3389/fpls.2020.579524 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cardoza V, Stewart CN (2003) Increased Agrobacterium-mediated transformation and rooting efficiencies in canola (Brassica napus L.) from hypocotyl segment explants. Plant Cell Rep 21:599–604. 10.1007/s00299-002-0560-y [DOI] [PubMed] [Google Scholar]
  10. Chaudhary S, Parmenter DL, Moloney MM (1998) Transgenic Brassica carinata as a vehicle for the production of recombinant proteins in seeds. Plant Cell Rep 17:195–200. 10.1007/s002990050377 [DOI] [PubMed] [Google Scholar]
  11. Gesch RW, Isbell TA, Oblath EA, Allen BL, Archer DW, Brown J, Hatfield JL, Jabro JD, Kiniry JR, Long DS, Vigil MF (2015) Comparison of several Brassica species in the north central U.S. for potential jet fuel feedstock. Ind Crops Prod 75:2–7. 10.1016/j.indcrop.2015.05.084 [Google Scholar]
  12. Gil-Humanes J, Martin A, Barro F (2011) Characterization of a collection of Brassica carinata genotypes for in vitro culture response and mode of shoot regeneration. Am J Plant Sci 2:27–34. 10.4236/ajps.2011.21003 [Google Scholar]
  13. Jefferson RA, Kavanagh TA, Bevan MW (1987) GUS fusions: β-glucuronidase as a sensitive and versatile gene fusion marker in higher plants. EMBO J 6:3901–3907 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Jupe F, Rivkin AC, Michael TP, Zander M, Motley ST, Sandoval JP, Slotkin RK, Chen H, Castanon R, Nery JR, Ecker JR (2018) The complex architecture and epigenomic impact of plant T-DNA insertions. 10.1101/282772. bioRxiv 282772 [DOI] [PMC free article] [PubMed]
  15. Karami O, Huang Y-K, Dwivedi P (2024) Does carinata, an oilseed crop for sustainable aviation fuel, improve the eco-efficiency of crop rotations in South Georgia, United States? J Clean Prod 453:141986. 10.1016/j.jclepro.2024.141986 [Google Scholar]
  16. Limbalkar OM, Singh R, Kumar P, Nanjundan J, Parihar CM, Vasisth P, Yadava DK, Chinnusamy V, Singh N (2021) Deployment of Brassica carinata A. Braun derived Brassica juncea (L.) Czern. Lines for improving heterosis and water use efficiency under water deficit stress conditions. Front Plant Sci 12:765645. https://www.ncbi.nlm.nih.gov/pubmed/34899786 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Maxted N, Ford-Lloyd B, Hawkes JG (1997) Plant genetic conservation: the in-situ approach. (eds). Chapman & Hall, London https://link.springer.com/10.1007/978-94-009-1437-7
  18. Mieog J, Howitt C, Ral J-P (2013) Fast-tracking development of homozygous transgenic cereal lines using a simple and highly flexible real-time PCR assay. BMC Plant Biol 13:71. 10.1186/1471-2229-13-71 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Mietkiewska E, Hoffman TL, Brost JM, Giblin EM, Barton DL, Francis T, Zhang Y, Taylor DC (2008) Hairpin-RNA mediated silencing of endogenous FAD2 gene combined with heterologous expression of Crambe abyssinica FAE gene causes an increase in the level of erucic acid in transgenic Brassica carinata seeds. Mol Breeding 22:619–627. 10.1007/s11032-008-9204-4 [Google Scholar]
  20. Moloney MM, Walker JM, Sharma KK (1989) High efficiency transformation of Brassica napus using Agrobacterium vectors. Plant Cell Rep 8(4):238–242. 10.1007/BF00778542 [DOI] [PubMed] [Google Scholar]
  21. Murashige T, Skoog F (1962) A Revised Medium for Rapid Growth and Bio Assays with Tobacco Tissue Cultures. Physiologia Plantarum 15:473–497. 10.1111/j.1399-3054.1962.tb08052.x
  22. Nagel R, Elliott A, Masel A, Birch RG, Manners JM (1990) Electroporation of binary Ti plasmid vector into Agrobacterium tumefaciens and Agrobacterium rhizogenes. FEMS Microbiol Lett 67:325–328. 10.1111/j.1574-6968.1990.tb04041.x [Google Scholar]
  23. Narasimhulu SB, Kirti PB, Prakash S, Chopra V (1992) Rapid and efficient plant regeneration from hypocotyl protoplasts of Brassica carinata. Plant Cell Rep 11:159–162. 10.1007/bf00232171 [DOI] [PubMed] [Google Scholar]
  24. Park J-I, Ahmed N, Kim H-R, Nou I-S (2012) Advances in in vitro culture of the Brassicaceae crop plants. J Plant Biotechnol 39:13–22. 10.5010/JPB.2012.39.1.013 [Google Scholar]
  25. Petrie JR, Shrestha P, Zhou XR, Mansour MP, Liu Q, Belide S, Nichols PD, Singh SP (2012) Metabolic engineering plant seeds with fish oil-like levels of DHA. PLoS ONE 7:e49165. 10.1371/journal.pone.0049165 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Petrie JR, Zhou X-R, Leonforte A, McAllister J, Shrestha P, Kennedy Y, Belide S, Buzza G, Gororo N, Gao W, Lester G, Mansour MP, Mulder RJ, Liu Q, Tian L, Silva C, Cogan NOI, Nichols PD, Green AG, de Feyter R, Devine MD, Singh SP (2020) Development of a Brassica napus (Canola) crop containing fish oil-like levels of DHA in the seed oil. Front Plant Sci 11:727. 10.3389/fpls.2020.00727 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Ravanfar SA, Orbovic V, Moradpour M, Abdul Aziz M, Karan R, Wallace S, Parajuli S (2017) Improvement of tissue culture, genetic transformation, and applications of biotechnology to Brassica. Biotechnol Genet Eng Rev 33:1–25. 10.1080/02648725.2017.1309821 [DOI] [PubMed] [Google Scholar]
  28. Seepaul R, Kumar S, Iboyi J, Bashyal M, Stansly T, Bennett R, Boote K, Mulvaney M, Small I, George S, Wright D (2021) Brassica carinata: biology and agronomy as a biofuel crop. GCB Bioenergy 13:582–599. 10.1111/gcbb.12804 [Google Scholar]
  29. Seepaul R, Marois J, Small IM, George S, Wright DL (2019) Carinata dry matter accumulation and nutrient uptake responses to nitrogen fertilization. Agron J 111:2038–2046. 10.2134/agronj2018.10.0678 [Google Scholar]
  30. Taylor D, Falk K, Palmer C, Hammerlindl J, Babic V, Mietkiewska E, Jadhav A, Marillia E-F, Francis T, Hoffman T, Giblin E, Katavic V, Keller W (2010) Brassica carinata - A new molecular farming platform for delivering bio-industrial oil feedstocks: case studies of genetic modifications to improve very long-chain fatty acid and oil content in seeds. Biofuels Bioprod Biorefin 4:538–561. 10.1002/bbb.231 [Google Scholar]
  31. Vaughan JG, Geissler C, Nicholson B (2009) The new Oxford book of food plants / J.G. Vaughan and C.A. Geissler; illustrations by B.E. Nicholson, Elisabeth Dowle & Elizabeth Rice, [2nd]. edn. Oxford University Press, Oxford
  32. Verma SS, Chinnusamy V, Bansa KC (2008) A Simplified Floral Dip Method for Transformation of Brassica napus and B. carinata. J Plant Biochem Biotechnol 17:197–200. 10.1007/BF03263286 [Google Scholar]
  33. Weng H, Pan A, Yang L, Zhang C, Liu Z, Zhang D (2004) Estimating number of transgene copies in transgenic rapeseed by real-time PCR assay withHMG I/Y as an endogenous reference gene. Plant Mol Biology Report 22:289–300. 10.1007/BF02773139 [Google Scholar]
  34. Whitehouse AB, Marks TR, Edwards GA (2002) Control of hyperhydricity in Eucalyptus axillary shoot cultures grown in liquid medium. Planr Cell Tissue Organ Cult 71:245–252. 10.1023/A:1020360120020 [Google Scholar]
  35. Yang M-Z, Jia S-R, Pua E-C (1991) High frequency of plant regeneration from hypocotyl explants of Brassica carinata A. Br. Planr Cell Tissue Organ Cult 24:79–82. 10.1007/BF00039734 [Google Scholar]
  36. Yu Y, Liu L, Zhao YQ, Zhao B, Guo YD (2010) A highly efficient in vitro plant regeneration and Agrobacterium-mediated transformation of Brassica oleracea var. botrytis. New Z J Crop Hortic Sci 38:235–245. 10.1080/01140671.2010.511126 [Google Scholar]
  37. Zoong Lwe Z, Sah S, Persaud L, Li J, Gao W, Raja Reddy K, Narayanan S (2021) Alterations in the leaf lipidome of Brassica carinata under high-temperature stress. BMC Plant Biol 21:404. 10.1186/s12870-021-03189-x [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary Material 1 (14.9MB, docx)

Data Availability Statement

No datasets were generated or analysed during the current study.

Articles from Plant Molecular Biology are provided here courtesy of Springer

RESOURCES