Introduction

One of the most consumed and commonly cultivated fruit crops worldwide is Citrus. In many nations, citriculture is an important industry. Oranges make up more than half of the world’s Citrus production and are one of the most widely sold and produced Citrus fruit varietals. The Citrus fruits grapefruits, tangerines, and lemons are also highly traded. In 2019, 143 thousand tons of Citrus were produced worldwide (FAO 2021). Maintaining economic competitiveness involves tactics that go beyond commercial concerns, and the genus Citrus has restrictions that are limiting production, such as Huanglongbing (HLB), which is gravely affecting citriculture globally. Effective disease control is an issue shared by all producing countries. Most of the time, the resources used to combat diseases are focused on chemicals, which can harm the environment and favor the formation of resistant microbes owing to continual application. In this circumstance, technology is useful for the creation of genetically resistant Citrus types that will allow for more sustainable citriculture with a lower environmental effect (Caserta et al. 2020).

Citrus is also one of the most difficult species to breed conventionally. It has challenges as a result of nucellar polyembryony, self-incompatibility, and high heterozygosity. On the other hand, according to literature, the Agrobacterium-mediated transformation of juvenile materials, such as zygotic embryos, hypocotyls, epicotyls, and cotyledons, has proven to be a successful way of genetically modifying Citrus (Fleming et al. 2000; Costa et al. 2002; Al-Bahrany 2002; De Oliveira et al. 2009). Recent breakthroughs in genome editing and gene stacking technologies are allowing improvements and trait modifications at unprecedented rates. However, rates of regeneration represent a bottleneck in genetic engineering of commercially relevant Citrus varieties. This presents a major challenge in generating modified crops for improvements testing. Citrus in vitro organogenesis responses are influenced by several internal and external factors. The differences in genotype and seasonal dependency make it challenging to standardize a protocol, as the explant varies from year to year, and the regeneration potential of the explant is correlated with the physiological conditions of the donor plants. Results are also influenced by levels of endogenous metabolites and plant growth regulators (Niedz and Evens 2011; Hu et al. 2017). Tissue culture conditions, in particular length of dark time and exposure to light intensity, seem to be critical, as they can alter the level of endogenous hormones such as cytokinins and auxins (Marutani-Hert et al. 2012; Germanà et al. 2020). Nutrient requirements also vary among plant species and micropropagation techniques.

Importantly, the impact of media for micropropagation has received significantly less focus in research. The choice of basal medium is often based on prior usage rather than optimal results (Greenway et al. 2012), even though the chemical composition of the culture media has been shown to have a significant impact on the success of in vitro tissue cultivation (Ružić et al. 2003). Higher plants cannot develop in vitro without the use of well-adjusted macronutrient salts, specifically N, P, K, Ca, Mg, and S (George et al. 2008; Lozzi et al. 2019). For several Citrus spp., in vitro culture growth has been demonstrated in MT (Murashige and Tucker 1969) and MS (Murashige and Skoog 1962), and more recently DKW (Driver and Kuniyuki 1984; de Oliveira et al. 2010). To enhance the likelihood of producing transgenic Citrus plants from transformed cells this study aimed to investigate the impact of varying salt concentrations in the media on shoot regeneration thereby improving the chances of a successful transformation (de Oliveira et al. 2010; Dominguez et al. 2022).

To further improve Citrus spp. transformation, we evaluated 21 published media (Table 1) formulations on 11 Citrus cultivars to identify optimal conditions for shoot regeneration. The results obtained from this study provide insights into the shoot regeneration performance of these cultivars, and suggestions are offered for development of improved Citrus regeneration techniques.

Table 1 The composition of the 21 media types used in the study are organized by PhytoTechnology Laboratories® code, name of the published media, species analyzed and reference
Full size table

Materials and methods

Plant material

The experiments were carried out at the Crop Improvement and Genetics Research Lab at the United States Department of Agriculture – Agricultural Research Service (USDA-ARS) in Albany, California, USA.

Seeds from 11 varieties of Citrus, including mandarins, sweet oranges, orange rootstock, and lemons, were grouped into the following categories:

  • Mandarins: W. Murcott (C. reticulata), Changsha Mandarin (C. reticulata Blanco), Kinnow (C. deliciosa x C. nobilis), Sunki Mandarin (C. sunki Hort. ex Tanaka), Fina Sodea Clementine (C. clementina);

  • Sweet Orange: Pineapple (C. sinensis L. Osbeck), Valencia Cutter (C. sinensis L. Osbeck);

  • Grapefruit: Grapefruit 343 (C. paradisi Macfadyen);

  • Orange Rootstock: US802 (C. grandis Osbeck × P. trifoliata), and.

  • Lemons: Lisbon Lemon ‘Frost’ 2 N (C. limon) and Lisbon Lemon 4 N (C. limon), a tetraploid developed by Dr. Roose at UC Riverside.

The seeds were obtained from Lyn Seeds Inc., Lindcove Citrus Research Station, and UC Riverside Citrus germplasm collection and stored dry at 4ºC.

For germination, the outer seed coats were removed manually, and seeds were treated with 70% alcohol for 1 min., followed by 1.8% sodium hypochlorite for 20 min., and then rinsed three times with sterile distilled water. Two or three seeds were cultured in assay tubes with MS (Murashige and Skoog 1962) supplemented with 0.1% of Plant Tissue Culture Contamination Control (PTC3) (PhytoTechnology Laboratories), 30 gL−1 sucrose, and 8 gL−1 agar, pH 5.8 ± 0.01. The seeds were stored in the dark for 15–30 days before being transferred to light conditions with a 16/8-hour (light/dark) photoperiod.

Shoot regeneration

Epicotyl segments approximately 1 cm long were harvested under aseptic conditions from two-month-old seedlings and gathered in MS liquid media before being deposited on fresh medium. The 21 media tested were obtained from PhytoTechnology Laboratories, and their ion composition and recipe are detailed in Supplementary Tables S1 and S2, respectively. All media were tested as they were described by the authors in Table 1; however they were normalized to 5 mgL−1 6-benzylaminopurine (BAP), 1 mgL−1 indole-3-acetic acid (IAA), 1mgL−1 phloroglucinol, 30 gL−1 sucrose, 2 gL−1 charcoal, and 8 mgL − 1 agar. The media were adjusted to pH 5.8 ± 0.01 and autoclaved at 120 °C for 15 min. Explants were initially incubated in 100 × 15 mm Petri dishes in the dark at 28 ± 2ºC for 30 days and then transferred to 100 × 25 mm Petri dishes for 16/8-hour (light/dark) photoperiod. 11 of the media were selected from the 21 media screening experiment. These were the best performing media and were used in the experiments described.

Statistical analysis

Treatments were tested using 2 petri dishes, each containing 9 explants for a total of 18 explants per treatment. Treatments were repeated once or twice over time, which were treated as blocks in the ANOVA. The cultures were sub-cultured every 15–20 days and the number of shoots per explant recorded each month. To determine if data transformation was required a Box-Cox analysis (Box and Cox 1964) was conducted to determine if data transformation was required, and if so, the data were transformed prior to ANOVA. Mean separation analyses used the Scott-Knott 120combination to provide a simple visual color intensity summary overview of the results. Nonmetric multidimensional Scaling (MDS) plots were constructed from using from a Bray-Curtis similarity matrix (Bray and Curtis 1957) to provide a visual illustration of the clustering relationships of the citrus types and media. Software used included DesignExpert (StatEase, MN), PRISM (GraphPad, Boston, MA), and R Studio Data Analysis (R Development Core Team 2008).

Results

The present study conducted statistical analyses for the number of shoots regenerated from 11 different cultivars after 3 months of culture, using 21 unique media (1st experiment, (Fig. 1). A 2nd experiment was conducted under duplicate conditions but only using the 11 best media from the 1st experiment, to confirm results. The cultivars evaluated in this study were: W Murcott (WMU), Sunki Mandarin (SKM), Fina Sodea (FS), Changsha Mandarin (CSM), Kinnow (KW), Valencia Cutter (VAL), Pineapple (PIN), Grapefruit 343 (GFR), Lisbon Lemon 4 N (L4), Lisbon Lemon Frost 2 N (LIS), and US802 (US8). The harvested juvenile tissue displayed a prompt response to the media. Within a few days, callus formation could be observed in the wounded region of nearly all explants. Upon the initial assessment, the cultivars W Murcott and Pineapple demonstrated the greatest number of shoots developed.

Fig. 1
figure 1

Number of shoots per explant on the first round and media interaction in Changsha Mandarin, Kinnow, Sunki Mandarin, Fina Sodea, Lisbon Lemon 4 N, Lisbon Lemon 2 N, W Murcott, Grapefruit, and Pineapple. Uppercase letters indicate difference between Genotype in the same Media. Lower case letters indicate statistical difference between Media in the same Genotype

Full size image

In terms of the number of shoots per explant, in general, of the 21 mediums tested E337, R7100, and W865 produced equivalent and statistically significant results for all cultivars examined. Figure 1 shows the effect of the 21 media on nine of the representative cultivars chosen for this study. It can be seen that certain media were more effective for shoot generation than others. For example, in the top right panel Changsha Mandarin displays a statistical preference for media B1471, R7100 and S813. These results were used to narrow the study to the most effective 10 media plus DKW for a confirmation assay (Fig. 1).

The second media study results are depicted in the heat map (Fig. 2). Results provide an overview of the most effective formulations for each cultivar, allowing identification of the optimal media for organogenesis in each Citrus genotype. Data obtained from the 11 media study, was further used for an ANOVA statistical examination of variations between genotypes, media and their interaction (ANOVA, Table 2, and statistical summary Table S3). Results suggest that Citrus genotype has the single greatest impact on rates of regeneration followed by media and interaction, respectively.

Table 2 ANOVA for the 11 media study examining variations across citrus genotype, medium, and their interaction. Citrus genotype appears to have the single greatest impact
Full size table
Fig. 2
figure 2

Heat Map of Citrus Genotype/Media combinations provides an overview of the number of shoots produced for each citrus/media combination. The greener the color, the more shoots produced by the combination. Changsha Mandarin (CSM), Kinnow (KW), Sunki Mandarin (SKM), Fina Sodea (FS), Lisbon Lemon 4 N (L4), Lisbon Lemon 2 N (LIS), W. Murcott (WMU), Grapefruit (GFR), Valencia (VAL), US802 (US8), and Pineapple (PIN)

Full size image

Based on the second experiment of 11 media selection experiments, the three best media formulations were identified for each cultivar. Table 3 lists the three recommended media for each cultivar based on the number of shoots and length of shoots (data not shown). The best common statistically equivalent media for mandarins were B1471, R7100, and S813, with an average of 1.58 shoots per explant. For sweet oranges and grapefruit, the most favorable group results were achieved with Q673, R7100, and W865, yielding 1.1 shoots per explant. Rootstock US802 had the highest success rates using media DKW, R7100, and M5525, with up to 0.46 shoots per explant. For lemons, the optimal media formulations were B1471, W865, and R7100, yielding 0.76 shoot per explant. Pineapple did not display a statistical preference for any one media type and shoot formation was seen for all tested, however media DKW, Q673 and R7100 appeared to give the greatest amount of regeneration.

Table 3 Recommended media for shoot regeneration on each cultivar using juvenile tissue of Changsha Mandarin (CSM), Kinnow (KW), Sunki Mandarin (SKM), Fina Sodea (FS), Lisbon Lemon 4N (L4), Lisbon Lemon 2N (LIS), W. Murcott (WMU), Grapefruit (GFR), Valencia (VAL), US802 (US8), and Pineapple (PIN). p-values were determined by nonparametric analyses using the Kruskal–Wallis test for one-way ANOVA
Full size table

Figure 3 depicts the clustering relationships for shoot regeneration performance between the data points using nonparametric views of the data Multidimensional Scaling (MDS) analysis. These results indicate that media B1471, R7100 and S813 contain a preferred mineral content over the other recipes examined. However, the data also indicates that the results were highly influenced by mandarins, oranges and grapefruit as they produced the greatest number of shoots from the media examined. Figure 3 further indicates that tissue culture conditions were not optimized sufficiently for the lemons or rootstock US802 to effectively regenerate shoots.

Fig. 3
figure 3

Clustering relationships between the data points based on the nonparametric views of the data (MDS) analysis from the Bray-Curtis coefficient. The distance between the points reflects the dissimilarity between the data points. A The data points are grouped according to performance, from the best shoot regenerating media to the lowest. B The data points are grouped according to the responsiveness of the genotypes, from the lowest shoot regenerating cultivar to the most responsive. Changsha Mandarin (CSM), Kinnow (KW), Sunki Mandarin (SKM), Fina Sodea (FS), Lisbon Lemon 4 N (L4), Lisbon Lemon 2 N (LIS), W. Murcott (WMU), Grapefruit (GFR), Valencia (VAL), US802 (US8), and Pineapple (PIN)

Full size image

The variability in media mineral content and differences in genotype significantly affected shoot regeneration, making it challenging to perform statistical analyses. Despite these challenges, observable differences in shoot regeneration can be attributed to variations in mineral source and balance across media formulations (Table 4). Specifically, media with reduced nitrogen levels, as well as higher levels of calcium and magnesium, promoted greater shoot regeneration. In addition, media formulations that include extra vitamins produced favorable outcomes. For instance, the M5525 (MT medium), which is a modified MS medium with elevated vitamin content, demonstrated a positive effect on shoot regeneration.

Table 4 Correlation of media nutrients significance on shoot regeneration per explant from Citrus Juvenile Tissue. P-values and correlation coefficient r-values are shown for all 11 citrus cultivars, and separated for Mandarins, Lemons, Oranges and Grapefruit (O&G), and US802
Full size table

Regarding ion composition (Table S1), media with lower nitrogen sources tended to yield the best results for shoot regeneration, especially in the mandarin, orange, and grapefruit genotypes tested (Table 4). Overall, the majority of Citrus varieties displayed improved shoot regeneration rates with media that contained 50% lower ammonium (5.00–10.00 mmolL − 1) than MS medium (20.61 mmolL − 1). Additionally, media formulations with slightly higher concentrations of calcium (4.00–9.00 mmolL − 1) and magnesium (3.00-7.51 mmolL − 1) than MS medium (3.00 mmolL − 1 calcium, 1.50 mmolL − 1 magnesium) also showed promising results for shoot regeneration. For US802, Ca2 + concentration showed a significant correlation with rate of regeneration. These findings suggest that optimizing the nitrogen source and balancing mineral concentrations, particularly calcium and magnesium, can lead to improved shoot regeneration rates in multiple Citrus genotypes.

Discussion

Factors that need to be considered for each Citrus regeneration include hormone levels, (Moreira-Dias et al. 2000; Molina et al. 2007; Rodríguez et al. 2008; da Silva et al. 2010), polarity of epicotyls (Hu et al. 2017), genotype (Jiménez et al. 2001), plant age (Long et al. 2022), and of course choice of basal media (Murashige and Skoog 1962).

The nutrient medium is an essential part of tissue culture methods, and several media formulations have been published (Table 1). Even though MS medium is the most commonly used for a range of species, many reports have described unique basal media formulations for specific species and techniques, including MT (Murashige and Tucker 1969) for Citrus, Y3 medium (Eeuwens 1976) for coconut and OM medium (Rugini 1984) for Olive.

Genotype plays a crucial role in determining the in vitro response of Citrus plants (Jimenez et al., 2001) 37 and comparing the results of different studies can be challenging due to variations in methodology and conditions. Nonetheless, several publications have demonstrated that Citrus genotypes show significant differences in their morphogenic response, even when exposed to the same treatment (Bordón et al. 2000; Rodríguez et al. 2008; Cardoso et al. 2016; Dominguez et al. 2022). Therefore, it is important to consider the genotype when designing in vitro experiments with Citrus.1.

The present study corroborated that the effectiveness of the media depended significantly on the specific Citrus cultivar. The genotype responsiveness rate is a determining factor regarding which cultivar and media should be chosen for genetic transformation. For example, Changsha Mandarin had no shoot regeneration from MT media, but demonstrated the most effective shoot regeneration results of all cultivars in medium S813 (Fig. 2). Pineapple (a sweet orange) was responsive in all media tested, confirming why this is the most commonly used cultivar for transformation (Marutani-Hert et al. 2012). This result highlights the need to critically examine the culture media if regeneration is underperforming.

Murashige and Tucker Medium (1969), initially described as favorable for in vitro growth of Citrus, has been widely used regardless of the cultivar (Ghorbel et al. 2000; Dutt et al. 2020). MT is a modification of Murashige and Skoog medium (Murashige and Skoog 1962)], containing increased levels of vitamins such as thiamine, pyridoxine, and nicotinic acid. MT medium with modifications (EME and H + H) have also produced successful results for Citrus organogenesis (Tavano et al. 2009). Initially the medium EME (MT with 500 mgL−1 of Malt Extract) and H + H medium (MT with half concentration of Nitrogen source) were described to initiate embryogenic suspension cultures from ovule-derived callus, providing an excellent source of totipotent protoplasts for use in fusion experiments (Grosser and Gmitter 1990). The vitamins act as enzymatic cofactors in universal pathways such as glycolysis and the TCA cycle, as well as primary and secondary metabolism in plants, and Glycine serves as an amino acid source.

The results of the present study indicate that high levels of nitrate or ammonium, such as those found in the MT (M5525) medium, had a negative impact on Citrus organo-genesis in most cultivars tested. Conversely, media formulations with lower nitrogen source concentrations, such as DKW and WPM, were more successful in promoting Citrus organogenesis. These findings suggest that examination of the nitrogen concentration and source in the culture media is critical for successful Citrus tissue culture and should be carefully considered when designing culture protocols for different cultivars (de Oliveira et al. 2016). Niedz and Evens (2008) discovered higher Citrus calli multiplication using MS medium with lower NH4+ concentrations of 2.50 to 3.75 mM resulted in the maximum rise in fresh weight. These ranges correspond to an NH4+:NO3− ratio of 1:3 at 37.5 mM total Nitrogen, whereas in conventional MS medium this ratio is 1:2 at 60 mM total Nitrogen (Niedz and Evens 2008).

Despite the preference of plants for nitrate over ammonium and the higher cost of nitrate uptake and assimilation, the effect of the NH4+:NO3− ratio on Citrus experiments does not appear to follow a clear pattern. As observed by this study and others, Citrus explant regeneration is dependent on genotype and use of both nitrate and ammonium in reduced concentrations. This suggests that the optimal NH4+:NO3− ratio for Citrus tissue culture may vary depending on the specific cultivar being used. However, we observed media R7100 (with a 1:4 ratio at 26.26mM total Nitrogen) and B1471 (with a 1:3 ratio at 39.38mM total of Nitrogen) both had increased rates of regeneration. Further research is needed to better understand the complex interactions between Citrus genotypes and nitrogen source preferences in tissue culture.

It has also been noted that higher levels of salts, such as Mg2+ and Ca2+, are reported as responsible for producing a greater number of shoots in Carrizo explants (Niedz et al. 2015). In sour orange mature tissue, the number of buds per regenerating explant increased when the DKW medium was used. The Ca2+ concentration in the DKW basal medium (9 mM) is three times higher than in WPM or MS (3 mM) (Tallón et al. 2013). DKW was reported to be the best media for inducing shoots per explant for Carrizo and Washington Navel Citrus (de Oliveira et al. 2016). While the results reported here suggest that while Mg2+ and Ca2+ were not significantly different for most genotypes they seem to help as found using the media R7100.

Conclusions

Developing universal transformation techniques is a crucial biotechnological objective. However, variations in cultivars of the same species can significantly impact their organogenesis response and the frequency of transformed explants after agrobacterium infection. Based on ANOVA results genotype is the single largest variable when it comes to media requirements for regeneration. Therefore, it is essential to understand the specific requirements of each cultivar to develop optimized media formulations for regeneration. According to Niedz and Evens (2008), creating superior media formulations involves techniques that (1) eliminate ion-related confounding effects, (2) treat all initial solution properties as dependent variables, and (3) differentiate between proportional and amount effects to determine the optimal ion concentration and media requirements. As this study demonstrates, each cultivar has its own unique media requirements that influence organogenesis and regeneration. Furthermore, the study reveals the negative effects of excessive nitrogen supply during the regeneration process. These results provide a starting point for developing a new media formulation for Citrus organogenesis. Finally, medium R7100 conforms with the observations of this study for optimal salt concentration, as it contains lower nitrogen and higher Mg2 + and Ca2 + content than MS. Although not the best for every cultivar, R7100 appeared overall to be the most effective for regeneration for the majority of Citrus genotypes tested.