Wild relatives of Stevia rebaudiana

The genus Stevia includes approximately 230 species, among which Stevia rebaudiana is notable for accumulating sweet steviol glycosides (Borgo et al., Reference Borgo, Laurella, Martini, Catalán and Sülsen2021). Wild relatives, such as S. eupatoria, S. lemmonii and S. micrantha, occur from the southern United States to Argentina, occupying diverse ecological niches and exhibiting cytogenetic variability (diploid, triploid, tetraploid) that underlies a multibasic genetic structure (Yadav et al., Reference Yadav, Singh, Dhyani and Ahuja2011; Borgo et al., Reference Borgo, Laurella, Martini, Catalán and Sülsen2021). Although not all wild species produce high glycoside levels, they possess adaptive traits, morphological differences in leaf architecture and growth habit, plus mechanisms to tolerate drought, fluctuating light and temperature extremes, that can be introduced into cultivated lines via conventional or biotechnological breeding to enhance biomass and glycoside yield under suboptimal conditions (Brandle et al., Reference Brandle, Starratt and Gijzen1998; Borgo et al., Reference Borgo, Laurella, Martini, Catalán and Sülsen2021; Gantait et al., Reference Gantait, Das and Banerjee2018; Al-Taweel et al., Reference Al-Taweel, Azzam, Khaled and Abdel-Aziz2021).

Because landraces and wild types often display greater genetic diversity and a higher number of private alleles than bred cultivars, these relatives serve as reservoirs of alleles for disease resistance, yield stability and steviol glycoside biosynthesis (Cosson et al., Reference Cosson, Hastoy, Errazzu, Budeguer, Boutié, Rolin and Schurdi-Levraud2019; Borgo et al., Reference Borgo, Laurella, Martini, Catalán and Sülsen2021; Gantait et al., Reference Gantait, Das and Banerjee2018). Molecular and cytogenetic analyses confirm that introgressing wild alleles can broaden the genetic base of commercial stevia cultivars, helping to overcome narrow genetic bottlenecks and improve overall performance (Yadav et al., Reference Yadav, Singh, Dhyani and Ahuja2011; Gantait et al., Reference Gantait, Das and Banerjee2018).

Intraspecific variability and cultivar diversity within Stevia genus

Stevia rebaudiana is a perennial shrub 60–120 cm tall, with opposite oval leaves with serrated margins, leathery texture, green stems, white-flowered panicle inflorescences, taproots and a fragrant sweet herbal aroma (Fig. 1). This species exhibits extensive varietal diversity, comprising over 90 varieties adapted to various climatic requirements and production systems (Angelini et al., Reference Angelini, Martini, Passera, Tavarini, Merillon and Ramawat2018). Many studies have demonstrated that significant heritability exists for key yield components, such as leaf yield and glycoside concentrations, indicating that genetic improvement is feasible through recurrent selection and breeding (Yadav et al., Reference Yadav, Singh, Dhyani and Ahuja2011). Modern germplasm collections incorporate both landraces and improved clones that have been developed through selective breeding and vegetative propagation methods (Clemente et al., Reference Clemente, Angelini, Ascrizzi and Tavarini2021). Evaluations of genetic variability using molecular markers, such as Random Amplified Polymorphic DNA (RAPD), Expressed Sequence Tag – Simple Sequence Repeat (EST-SSR) and Single Nucleotide Polymorphism analyses, have confirmed the presence of valuable polymorphisms underlying differences in key traits such as glycoside composition and stress tolerance (Cosson et al., Reference Cosson, Hastoy, Errazzu, Budeguer, Boutié, Rolin and Schurdi-Levraud2019). Detailed biochemical and molecular profiling of these materials reveals a wide range of steviol glycoside concentrations and complementarity in metabolic pathways, making them desirable targets for further improvement (Hastoy, Reference Hastoy2018).

Figure 1. Morphological features and cultivation of Stevia rebaudiana. (A) Field cultivation of stevia plants in a commercial production area in the South of Paraguay. (B) Stevia leaf showing its characteristic serrated margin and elongated oval shape (scale bar = 5 cm). (C) Stevia seed (achene) with its pappus hairs aiding in wind dispersal (scale bar = 1 mm). (D) Flowering stevia plant grown under plastic mulch conditions, displaying inflorescences with small white flowers and opposite leaves (scale bar = 10 cm).

Moreover, in trials of four elite stevia populations in North Carolina, key growth traits proved to be highly heritable, suggesting they respond well to selection (Kozik et al., Reference Kozik, Yücesan, Saravitz and Wehner2020). For example, plant height exhibited broad‐sense heritability values of 0.68 in June and 0.60 in August, branch width showed heritabilities of 0.59 and 0.55, and leaf area came in at 0.54 and 0.52. Under a 20% selection intensity, breeders could expect annual gains of roughly 1.2 Mg ha⁻¹ in dry biomass and about 24 mm in height. Similarly, the major steviol glycosides displayed strong genetic control: rebaudioside A had H²g = 0.60, rebaudioside C was 0.58, rebaudioside D was 0.50, stevioside was 0.52 and total steviol glycosides (TSGs) reached 0.62. These values translate to potential per-cycle increases of about 26.4 mg g⁻¹ for rebaudioside A (14.5 % of TSG) and 20.2 mg g⁻¹ for rebaudioside C (20.2% of TSG). Given this level of genetic variation, adopting marker-assisted selection makes it possible to accelerate improvements in both yield and glycoside composition.

Sexual reproduction as a bottleneck in the commercial scale-up of Stevia rebaudiana

Stevia rebaudiana exhibits two modes of reproduction: sexual propagation via seeds and asexual propagation via stem cuttings or tissue culture, each contributing differently to genetic diversity and uniformity in commercial plantations (Ramakrishnan et al., Reference Ramakrishnan, Sudheer, Banadka, Bajrang, Al-Khayri, Nagella, Al-Khayri, Jain and Penna2025). However, seeds present several challenges that limit their utility. Stevia rebaudiana individual seeds develop within achene-type fruits that act as the primary propagule or dispersal unit. Dark-coloured achenes tend to be heavier due to successful fertilization and embryo formation, yet they show highly variable viability, with germination rates ranging from less than 1% to as high as 59–86%. This variability in seed germinability results from intrinsic issues with pollination and fertilization compatibility, often leading to malformed embryos or endosperm. Consequently, viable seeds exhibit a wide range of physiological vigour, affecting key germinability parameters such as germination rate, uniformity and speed within a single seed batch – ultimately impairing seedling stand uniformity and crop establishment (Joosen et al., Reference Joosen, Kodde, Willems, Ligterink, van der Plas and Hilhorst2010).

In contrast, light and tan achenes are typically lighter because they result from self-incompatibility processes that inhibit viable seed formation, producing empty achenes or seeds without embryos, which fail to germinate. Furthermore, a sporophytic self-incompatibility mechanism renders most self-pollinated seeds infertile, requiring cross-pollination to produce seeds. This results in genetically heterogeneous half-sib progenies and inconsistent steviol glycoside profiles (Angelini et al., Reference Angelini, Martini, Passera, Tavarini, Merillon and Ramawat2018).

Additional reproductive and environmental factors, including premature seed harvest, protandry, low pollen viability, nutrient deficiencies affecting pollen tube growth and the need for specific light and temperature conditions, further depress viable seed production (Al-Taweel et al., Reference Al-Taweel, Azzam, Khaled and Abdel-Aziz2021). Moreover, stevia seeds are extremely small (∼3 mm), contain minimal endosperm reserves and deteriorate rapidly unless stored at low temperatures. These characteristics hinder seed handling, storage and quality maintenance, making sexual propagation inefficient for large-scale multiplication due to poor, inconsistent germination rates and genetically mixed offspring (Sharma et al., Reference Sharma, Gupta, Kumari, Kothari, Jain and Kachhwaha2023).

Consequently, vegetative propagation, either via stem cuttings or in vitro micropropagation, has become the preferred method for ensuring genetic uniformity, achieving higher propagation success, and faster multiplication of selected clones with consistent phytochemical profiles (Abdullateef and Osman, Reference Abdullateef and Osman2011; Al-Taweel et al., Reference Al-Taweel, Azzam, Khaled and Abdel-Aziz2021; Khan et al., Reference Khan, Verma, Rahman and Parasharami2021).

Linking genetic resources with strategies to cope with biotic and abiotic stress in stevia cultivation

Although less documented than in other crops, several biotic stressors, such as foliar and soilborne pathogens, significantly reduce yield and quality of stevia. Notable examples include Septoria leaf spot caused by Septoria steviae (Sanabria-Velazquez et al., Reference Sanabria-Velazquez, Enciso-Maldonado, Thiessen and Shew2024), foliar spots due to Alternaria steviae and A. alternata infections (Maiti et al., Reference Maiti, Sen, Acharya and Acharya2007; Yan et al., Reference Yan, Liu, Wang, Zhu, Yang, Xiao and Jiang2018), and rots caused by Sclerotinia sclerotiorum and Rhizoctonia solani (Koehler et al., Reference Koehler, Larkin, Rogers, Carbone, Cubeta and Shew2019; Kessler and Koehler, Reference Kessler and Koehler2020). Likewise, Fusarium oxysporum causes vascular wilt (Díaz-Gutiérrez et al., Reference Díaz-Gutiérrez, Poschenrieder, Arroyave, Martos and Peláez2019), and other secondary pathogens such as oomycetes, bacteria, viruses, phytoplasmas and various fungi may also be present (Samad et al., Reference Samad, Dharni, Singh, Yadav, Khan and Shukla2011; Chatzivassiliou et al., Reference Chatzivassiliou, Giakountis, Testa, Kienle and Jungbluth2015; Koehler and Shew, Reference Koehler and Shew2017; Rogers and Koehler, Reference Rogers and Koehler2021; Sanabria-Velazquez et al., Reference Sanabria-Velazquez, Cubilla, Flores-Giubi, Barua, Romero-Rodríguez, Enciso-Maldonado, Thiessen and Shew2023). Currently, studies reporting the extent to which the above-mentioned pathogens affect stevioside content are scarce. However, infections caused by A. alternata have been observed to reduce the commercial value and quality of stevia leaves, resulting in a negative impact on stevioside concentration (Prakash et al., Reference Prakash, Egamberdieva and Arora2022), although no specific quantitative data have been published. Since leaf quality is essential for stevioside content, it is reasonable to consider that any disease compromising leaf integrity could decrease the concentration of these compounds (Arturo et al., Reference Arturo, Torres González, Peña and Díaz2009). Nonetheless, further studies are needed to quantify these effects and better understand the relationship between infections by these pathogens and stevioside content in stevia.

Additionally, abiotic stresses, including drought, salinity, waterlogging and chemical treatments, can significantly alter steviol glycoside accumulation and plant growth. For instance, salinity and drought stress have been found to enhance the concentrations of stevioside and rebaudioside A, although they simultaneously limit biomass accumulation (Debnath et al., Reference Debnath, Ashwath and Midmore2019). Moreover, Gupta et al. (Reference Gupta, Sharma and Saxena2016) highlighted that stevia plant treated with compounds of diverse nature such as Na₂CO₃, proline and polyethylene glycol, displayed specific metabolic changes, which in some cases enhanced TSG production by up to threefold compared to non-treated plants, despite negatively impacting shoot growth. The fact that the expression of several members of the SrMYB transcription factor family which is one of the largest regulators of gene expression and hence a variety of plant functions, is tissue-specific, stress responsive and highly associated with steviosides content, turn them into potential targets for engineering weather extreme resilient varieties and enhanced stevioside profile (Chen et al., Reference Chen, Lyu, Jiang, Liu, Liu, Qu, Hou, Xu, Feng and Wu2024). Therefore, it is necessary to elucidate the molecular basis that regulates stevia physiological responses to stress combination, presence of both biotic and abiotic stressors, to ensure yield stability under adverse climatic conditions.

The species’ ability to withstand or tolerate these biotic and abiotic stressors largely depends on the genetic variability present in its populations. Several studies suggest that, while no genotype shows complete resistance to major pathogens, there is significant variation in how different lines respond to infections like Septoria leaf spot. Hastoy et al. (Reference Hastoy, Le Bihan, Gaudin, Cosson, Rolin and Schurdi-Levraud2019) have evaluated the response to Septoria sp. of 10 genotypes from different origins and identified 2 of them, ‘Gawi’ and ‘Esplac1’, as moderately susceptible, with only 10–15% of symptomatic leaf area, whereas 3 highly susceptible genotypes, named ‘E8’, ‘C’ and ‘E161718’, which reached up to 40% of symptomatic leaf area, were also reported. In a leaf-disk assay, Le Bihan et al. (Reference Le Bihan, Gaudin, Rasouli, Boutet, Laurencon, Cosson, Hastoy, Rolin and Schurdi-Levraud2025) have reported minimal severity (3% symptomatic area) for ‘Cult102_SPA’ and ‘Cult76_GER’ genotypes confirming their lower susceptibility. Moreover, Yadav et al. (Reference Yadav, Singh, Dhyani and Ahuja2011) highlighted the line SF5-1 (No. 103) for its enhanced resistance against Septoria sp., and Huber and Wehner (Reference Huber and Wehner2021) have found significant variability in disease resistance, biomass yield and glycoside content across various seed-derived cultigens, which emphasizes the pivotal role of genetic studies in unravelling the molecular basis underlying resistance to pathogen attack in stevia genotypes, while highlighting the importance of conserving and utilizing plant genetic resources for breeding programmes aimed at developing genotypes with superior adaptive profiles (Ramakrishnan et al., Reference Ramakrishnan, Sudheer, Banadka, Bajrang, Al-Khayri, Nagella, Al-Khayri, Jain and Penna2025).

Regarding cold stress, a study conducted by Kozik et al. (Reference Kozik, Yücesan, Saravitz and Wehner2020) exposed 14 different stevia lines to cold temperatures for up to 10 days. They discovered that only one line, 7947-3, showed almost no damage after six days at −2°C. When the temperature was held at 0°C for 8 days, a small handful of lines (including 7947-3 and 7990-17) still looked healthy, while the others began to suffer. Even at a milder 2°C, it took 10 days before most plants showed signs of distress, yet lines 7947-3, 7918-1 and 7686-6 continued to hold up well. Based on these results, a simple way to screen for cold tolerance is to check leaf damage after 6 days at −2°C, 8 days at 0°C, or 10 days at 2°C, all under a light intensity of about 500 µmol m⁻² s⁻¹. Therefore, lines like 7947-3, which stay largely unscathed after these treatments, are especially valuable because they offer the potential to develop stevia varieties capable of thriving in temperate regions.

Improving steviol glycoside profile

A glycoside contains glucose and non-sugar parts derivate from the diterpenoid steviol, and in stevia the major ones are named stevioside and rebaudioside A (Reb A) (Humphrey et al., Reference Humphrey, Richman, Menassa and Brandle2006; Brandle and Telmer, Reference Brandle and Telmer2007; Zhou et al., Reference Zhou, Gong, Lv, Liu, Li, Du and Liu2021). These compounds share a common precursor (kaurenoid) with that of the plant hormone gibberellic acid and their production occurs mainly in leaves. The first step of steviol glycoside biosynthesis pathway and its related genes is very well described (Bondarev et al., Reference Bondarev, Sukhanova, Reshetnyak and Nosov2003). In brief, through the action of four consecutive enzymes, geranylgeranyl-diphosphate is converted into steviol, followed by its glycosylation, via specific glycosyltransferases to form different steviol glycosides (Sharma et al., Reference Sharma, Gupta, Kumari, Kothari, Jain and Kachhwaha2023). Several studies have reported attempts to improve steviol glycosides through genetic transformation and metabolic pathway engineering. For instance, plants overexpressing UDP-glycosyltransferase 76G1(SrUGT76G1), enzyme that catalyses stevioside to Reb A conversion, displayed a significant increase of Reb A: stevioside ratio, which also improved organoleptic properties (Richman et al., Reference Richman, Swanson, Humphrey, Chapman, McGarvey, Pocs and Brandle2005; Kim et al., Reference Kim, Zheng, Liao and Jang2019). Note that 68 putative UGTs have been identified in stevia, and particularly, the functionality of SrUGT76G1 was characterized through transient expression in Nicotiana benthamiana. This approach could be used to study the genetic control of steviol glycoside composition in this crop, which remains unknown (Petit et al., Reference Petit, Berger, Camborde, Vallejo, Daydé and Jacques2020). Concordantly, Zheng et al. (Reference Zheng, Zhuang, Mao and Jang2019) reported that transgenic plants overexpressing 1-deoxy-d-xylulose-5-phosphate synthase 1 (SrDXS1) and kaurenoic acid hydroxylase (SrKAH), both enzymes involved in steviol synthesis, showed enhanced content of steviol glycosides, up to 42–54% and 67–88%, respectively.

In this context, Bogado-Villalba et al. (Reference Bogado-Villalba, Nakayama Nakashima, Britos, Iehisa and Flores Giubi2021) assessed the genetic relationship among several Paraguayan Stevia rebaudiana lines and varieties, along with their steviol glycoside profile using SSR and ISSR markers. Genotyping revealed two main clusters, one of which was predominantly composed of Eriete and Katupyry varieties, as well as other lines characterized by high steviol glycoside content. Similar clustering pattern have been reported by Bhandawat et al. (Reference Bhandawat, Sharma, Nag, Singh, Ahuja and Sharma2014), Dyduch-Siemińska et al. (Reference Dyduch-Siemińska, Najda, Gawroński, Balant, Świca and Żaba2020) and Subositi et al. (Reference Subositi, Sri and Daryono2011), with Cosson et al. (Reference Cosson, Hastoy, Errazzu, Budeguer, Boutié, Rolin and Schurdi-Levraud2019) identifying three clusters across 145 global genotypes using EST-SSR markers. These studies collectively demonstrate how molecular markers can unravel genetic structure and diversity, facilitating the identification and incorporation of valuable traits into stevia breeding programmes (Barbet-Massin et al., Reference Barbet-Massin, Giuliano, Alletto, Daydé and Bergeret2016).

Photoperiod plasticity and polyploidy as breeding targets for enhancing biomass and environmental resilience

Stevia rebaudiana is an obligate short-day plant with a critical day length of 13 h (Ramesh et al., Reference Ramesh, Singh, Megeji and Sparks2006), making its phenology highly sensitive to day length. Interestingly, several studies have reported enhanced stevia production under long-day conditions, likely due to reduced glycoside synthesis at or near flowering. Under extended photoperiods, genes for steviol glycoside biosynthesis are upregulated in leaves, while gibberellin-related genes are downregulated; in stems, gibberellins increase, promoting elongation and biomass (Yoneda et al., Reference Yoneda, Nakashima, Miyasaka, Ohdoi and Shimizu2017; Gantait et al., Reference Gantait, Das and Banerjee2018; Rengasamy et al., Reference Rengasamy, Othman, Che and Harikrishna2022; de Andrade Mv et al., Reference de Andrade Mv, Lucho, Do Amaral, Braga, Ribeiro and de Castro Rd2023). These responses can inform crop strategies, including artificial lighting or breeding, to boost vegetative growth and glycoside content. Photoperiodic responses in Stevia rebaudiana indicate the need for distinct cropping protocols: short days favour flowering and seed production, while long days or extended lighting enhance vegetative growth and glycoside yield. Optimizing protocols for each purpose, vegetative production or seed harvesting, requires aligning cultivation with suitable latitudes or seasons to meet photoperiod needs and maximize crop performance.

Stevia rebaudiana is cultivated across Asia, South America and North America, where it encounters diverse environmental conditions, including varying day lengths. Despite this wide distribution, no stevia species have shown long-day photoperiodic behaviour. A comprehensive phenological and physiological assessment of accessions from different latitudes is essential to uncover naturally occurring allelic diversity and to optimize cultivation under contrasting photoperiods (de Oliveira Vm et al., Reference de Oliveira Vm, Forni-Martins, Magalhães and Alves2004; Ramesh et al., Reference Ramesh, Singh, Megeji and Sparks2006; González-Delgado et al., Reference González-Delgado, Martínez-Rivas and Jiménez-Gómez2025; Ramakrishnan et al., Reference Ramakrishnan, Sudheer, Banadka, Bajrang, Al-Khayri, Nagella, Al-Khayri, Jain and Penna2025).

Genome-wide association studies (GWAS) are powerful tools to explore natural genetic variation in wild stevia and identify loci linked to adaptive traits like photoperiodic responses. Applying GWAS to native accessions may reveal quantitative trait loci (QTLs) associated with day-length sensitivity, improving understanding of ecological adaptation and supporting breeding for environmental suitability. Additionally, identifying stevia orthologs of florigen and components of the Arabidopsis circadian clock could enable targeted genetic engineering. This foundational knowledge may facilitate manipulation of growth habit genes, supporting the development of cultivars with extended vegetative phases, indeterminate growth or reduced photoperiod sensitivity – traits successfully modified in crops like tomato and sorghum (Murphy et al., Reference Murphy, Klein, Morishige, Brady, Rooney, Miller, Dugas, Klein and Mullet2011, Reference Murphy, Morishige, Brady, Rooney, Yang, Klein and Mullet2014; Klein et al., Reference Klein, Miller, Dugas, Brown, Burrell and Klein2015; Vicente et al., Reference Vicente, Zsögön, de Sá Afl, Ribeiro and Peres2015).

Polyploidization, the presence of three or more chromosome sets, can be induced in stevia using colchicine and has been applied as a breeding tool. Autotetraploid plants often develop larger and thicker leaves than diploids, increasing total glycoside yield (Yadav et al., Reference Yadav, Singh, Yadav, Dhyani, Bhardwaj, Sharma and Singh2013; Xiang et al., Reference Xiang, Tang, Liu and Song2019; Joshi et al., Reference Joshi, Venkatesha, Gupta, Padalia and Kumar2025). Beyond enhancing plant and organ size, polyploidy plays a key role in stress responses (Bhosale et al., Reference Bhosale, Maere and De Veylder2019; Lang and Schnittger, Reference Lang and Schnittger2020). Stress conditions such as heat and drought have been shown to raise ploidy levels, indicating a general adaptation mechanism involving upregulation of metabolic and defense-related genes (Cookson et al., Reference Cookson, Radziejwoski and Granier2006; Monjardino et al., Reference Monjardino, Smith and Jones2006; Scholes and Paige, Reference Scholes and Paige2015; Tossi et al., Reference Tossi, Martínez Tosar, Laino, Iannicelli, Regalado, Escandón, Baroli, Causin and Pitta-Álvarez2022). In support, Markosyan et al. (Reference Markosyan, Ong and Jing2021) found that polyploidy modified resistance gene expression in stevia, while in Arabidopsis, increased ploidy correlated with changes in cell wall structure linked to enhanced pathogen resistance (Hamdoun et al., Reference Hamdoun, Zhang, Gill, Kumar, Churchman, Larkin, Kwon and Lu2016; Bhosale et al., Reference Bhosale, Boudolf, Cuevas, Lu, Eekhout, Hu, Van Isterdael, Lambert, Xu, Nowack and Smith2018).

The genus stevia exhibits considerable chromosomal variation, with basic numbers of x = 11, 12 or 17. Diploid (2n = 22), triploid (2n = 33) and tetraploid (2n = 44) species have been documented (Yadav et al., Reference Yadav, Singh, Dhyani and Ahuja2011). Polyploid populations often differ ecologically and morphologically from diploids (de Oliveira Vm et al., Reference de Oliveira Vm, Forni-Martins, Magalhães and Alves2004). In Mexican stevia species, agamospermous polyploids occupy broader geographic ranges than diploids, suggesting greater colonization potential and ecological adaptability (Watanabe et al., Reference Watanabe, Yahara, Soejima and Ito2001). Their dominance in marginal habitats highlights polyploidy’s role in adaptation and range expansion. This natural chromosomal variation offers valuable genetic resources for breeding, especially to enhance resilience to environmental stress. Polyploid individuals, particularly those reproducing agamospermously, tend to exhibit reproductive stability and greater abiotic stress tolerance. Selecting triploid and tetraploid genotypes from diverse habitats could allow breeders to capture beneficial ploidy-related traits, such as increased leaf biomass and stress tolerance, without requiring artificial chromosome doubling.

Genome editing as a tool to harness the natural variation of stevia wild relatives

Although breeding efforts face methodological and conceptual challenges, genotypes with more desirable glycoside profiles have been identified. Nonetheless, a systematic framework for generating pathogen-resistant and enhanced glycoside profile varieties with large-scale availability still does not exist. The genetic complexity underlying steviol glycoside synthesis, combined with fragmented information on key genes, metabolic pathways and factors of disease resistance, limits the rational design of superior varieties (Singh et al., Reference Singh, Singh, Singh, Parmar, Paul, Vashist, Swarnkar, Kumar, Singh, Singh and Kumar2017; Xu et al., Reference Xu, Yuan, Yu, Huang, Sun, Zhang, Liu, Tong, Zhang, Wang and Liu2021).

Biotechnological tools, including genetic editing, enable the molecular breeding of crops with specific properties. However, it requires an in-depth understanding of the genetics basis that controls development and physiology in the crop of interest (Younes et al., Reference Younes, Aquilina, Engel, Fowler, Frutos Fernandez, Fürst, Gürtler, Gundert‐Remy, Husøy and Mennes2019; Taak et al., Reference Taak, Tiwari and Koul2020; Biswas et al., Reference Biswas, Saha and Dey2021). As previously described, this has been addressed for steviol glycoside biosynthesis and certain abiotic stress responses (Pal et al., Reference Pal, Masand, Sharma, Seth, Singh, Singh, Kumar and Sharma2023). The combination of this fundamental biological knowledge and state-of-the-art gene editing techniques, such as CRISPR and TALEN, can be used to target a specific gene to improve traits in a precise manner. Another strategy that could be implemented in stevia breeding is called de novo domestication, which exploits the existence of wild relatives adapted to challenging environments and a wide range of photoperiod, traits with a diffuse polygenic basis, as a suitable raw material where monogenic domesticated-related and yield-determinant traits can be manipulated, instead of introducing alleles from wild relatives into cultivated crops. This approach is suitable for stevia breeding as plant transformation and delivery of CRISPR-based vectors have been successfully achieved in this crop and its genome is sequenced (Ghose et al., Reference Ghose, Abdullah, Md Hatta and Megat Wahab2022). Lastly, de novo domestication has been reported in tomato, maize and wheat (Zsögön et al., Reference Zsögön, Cermak, Voytas and Peres2017, Reference Zsögön, Čermák, Naves, Notini, Edel, Weinl, Freschi, Voytas, Kudla and Peres2018; Fernie and Yan, Reference Fernie and Yan2019).

Besides molecular tools, other approaches have been reported to be applied to enhance steviol glycoside levels. The variability observed in different Stevia genotypes in nature results from the accumulation of naturally occurring mutations, at a very low rate, throughout evolution. Hence, to produce novel genetic variation, mutations can be induced via physical and chemical mutagens (Raina et al., Reference Raina, Laskar, Khursheed, Amin, Tantray, Parveen and Khan2016). Kumar et al. (Reference Kumar, Singh, Rana, Kumar, Bhushan, Pathania, Kumar, Singh and Arya2024) applied gamma ray to the stevia variety ‘Madhuguna’ to induce mutations, followed by selection of mutant with an improved steviol profile. This technique was found to be effective, at mild doses (5 and 10 kR), in improving steviol glycoside content.

Conservation and utilization of Stevia genetic resources: Bridging legal framework and breeding innovation

Conserving and harnessing Stevia’s genetic diversity is essential for creating lines with enhanced agronomic performance. Exploring this diversity enables researchers to identify genes that enhance disease resistance and stress tolerance, thereby reducing reliance on agrochemicals and promoting more sustainable production systems. Such variability also underpins stevia’s ability to maintain stable yields amid extreme weather, as shifting weather patterns can influence pathogen prevalence and crop productivity.

Although market demand for stevia is growing, much of the available germplasm remains underused (Borgo et al., Reference Borgo, Laurella, Martini, Catalán and Sülsen2021); therefore, it is crucial to evaluate how current conservation measures and breeding programmes preserve and deploy genetic variation, not only to refine sweetness and yield but also to bolster overall crop health and long-term resilience (Ramakrishnan et al., Reference Ramakrishnan, Sudheer, Banadka, Bajrang, Al-Khayri, Nagella, Al-Khayri, Jain and Penna2025). It is crucial to continue characterizing existing germplasm banks. This ongoing effort allows us to accurately assess their current diversity and, critically, evaluate the need for an influx of new germplasm (Ribeiro et al., Reference Ribeiro, Diamantino, Domingues, Montanari, Alves and Gonçalves2021).

Conservation strategies have been predominantly limited to ex situ methods (seed banks, field collections) and in vitro propagation, lacking comprehensive plans that integrate wild diversity to meet producers/farmers and market needs. In situ conservation, which would safeguard the dynamic evolution of wild populations and their co-adaptation with local pathogens, has received scant attention (Salgotra and Chauhan, Reference Salgotra and Chauhan2023).

Global conservation of genetic resources for crops like potato and maize has benefited from robust institutional support. The International Potato Center (CIP) in Peru maintains a comprehensive in-trust collection of virus-free potato germplasm using advanced cryopreservation protocols, ensuring safe and equitable distribution of genetic material worldwide (Vollmer et al., Reference Vollmer, Villagaray, Cárdenas, Castro, Chávez, Anglin and Ellis2017). Similarly, Brazil has developed a coordinated national system for germplasm conservation through Empresa Brasileira de Pesquisa Agropecuária (EMBRAPA), combining in situ and ex situ strategies including cryobanks for both plant and animal genetic resources, backed by international agreements and national laws protecting biodiversity and traditional knowledge (Machado et al., Reference Machado, Oliveira, Paraventi, Cardoso, Martins and Ambrósio2016; Morrell and Mayer, Reference Morrell and Mayer2017).

Moreover, international agricultural research centres such as Centro Internacional de Agricultura Tropical (CIAT) in Colombia have demonstrated the enormous economic value of the unrestricted flow of genetic resources across countries. For example, Latin American nations have significantly benefited from bean germplasm shared through CIAT, with improved varieties boosting agricultural productivity throughout the region, even in countries that did not originate the genetic material (Johnson et al., Reference Johnson, Pachico and Voysest2005). This reinforces the argument that shared access to well-preserved and documented genetic resources can yield widespread and equitable benefits, especially when supported by transparent legal frameworks.

These initiatives required the support of public policies and legal frameworks that guarantee access, traceability and fair benefit-sharing. The case of Stevia rebaudiana, native to the Paraguay–Brazil border and traditionally used by the Guaraní people, illustrates the critical need for such mechanisms. Research has shown that major companies are using stevia-derived products as sugar alternatives in their products without consent from the Guaraní communities, raising serious concerns under the Convention on Biological Diversity and exposing legal gaps, particularly in Paraguay, which has not ratified the Nagoya Protocol. While Brazil has developed legal mechanisms for access and benefit-sharing, its laws exclude food products like stevia, further complicating Guaraní efforts to claim compensation or legal protection (Relly, Reference Relly2023).

These examples show how the success of germplasm conservation and equitable access depends on institutional infrastructure, legal harmonization and long-term political will. To prevent future biopiracy and secure fair benefits for Indigenous peoples, especially in the case of stevia, it is urgent to consolidate similar international and national frameworks, such as those provided by CIP, CIAT or EMBRAPA, tailored to minor or neglected crops of high commercial interest.

New approaches and future perspectives

Integrating diversity from both cultivated germplasm and wild relatives remains a critical challenge in exploiting Stevia rebaudiana’s full genetic potential. The species’ high heterozygosity and sporophytic self-incompatibility pose significant challenges to traditional breeding efforts, hindering the development of stable seed-based lines and requiring the use of precise molecular tools to assemble and fix desirable alleles (Yadav et al., Reference Yadav, Singh, Dhyani and Ahuja2011; Ramakrishnan et al., Reference Ramakrishnan, Sudheer, Banadka, Bajrang, Al-Khayri, Nagella, Al-Khayri, Jain and Penna2025). In this context, reproductive biology emerges as a fundamental yet often overlooked bottleneck that must be addressed through targeted research, especially given the increasing global demand for steviol glycosides and the need for scalable propagation systems.

The development of curated germplasm collections, with robust phenotypic and genotypic characterization, lays the groundwork for effective introgression strategies while minimizing linkage drag and preserving genetic integrity (Abdullateef and Osman, Reference Abdullateef and Osman2011; Hastoy, Reference Hastoy2018). Wild relatives of stevia, adapted to environments from the Paraguayan highlands to arid zones, harbour traits for abiotic stress tolerance that could be introgressed into elite lines (Gantait et al., Reference Gantait, Das and Banerjee2018). However, fully harnessing such diversity will require resolving the challenges imposed by the species’ reproductive system.

To advance beyond current strategies, we propose an integrative framework (Fig. 2). First, expanding the genetic base through targeted collections in centres of origin, coupled with genomic and metabolomic profiling, enables a comprehensive characterization of available germplasm (Vallejo and Warner, Reference Vallejo and Warner2021). These data streams (genetic, phenotypic, chemical, phytosanitary) should feed continuously into breeding pipelines to optimize agronomic traits and glycoside profiles. Second, combining ex situ and in situ conservation ensures that populations can evolve under natural selection and pathogen pressure, yielding more adaptable genetic sources while engaging local communities in areas of original diversity (Benelli et al., Reference Benelli, Carvalho, El Merzougui and Petruccelli2021).

Figure 2. Proposed pipeline to promote conservation of Stevia rebaudiana genetic resources and their integration into breeding programmes and basic research. (A) The Stevia genus comprises close to 230 species, and their distribution areas extend from the southern United States to the South American Andean region. (B) This tremendous diversity should be conserved in situ, allowing the dynamic evolution of wild population and co-adaptation to local pathogens, or ex situ, under controlled environment where samples are stored (seedbanks, germplasms, gene banks, etc.). (C) Screening, through physiological and biochemical assays, of wild relatives to find key traits related to abiotic and biotic stress tolerance and steviol profile, followed by identification and selection of candidate genes or loci (QTL) via molecular tools. Altogether, via biotechnological approaches genes governing different economically important traits would be unravelled. (D) Introgression breeding and use of DNA marker technology in back cross programmes. Moreover, gene editing in breeding allows for precise manipulation of target traits by directly altering specific genes once their molecular basis has been identified. (E) Generation of new lines showing extreme weather resilient, pathogen attack resistance and improved glycosides profile. (F) Risk assessment and approval processes for the cultivation, consumption and commercialization of new materials. Lastly, adoption of the latter by growers and consumer markets.

Third, omic-based approaches, genomics, transcriptomics, metabolomics and phenomics, should be embedded into marker-assisted selection to accelerate the identification of loci controlling key traits (van Der Hooft Jj et al., Reference van Der Hooft Jj, Mohimani, Bauermeister, Dorrestein, Duncan and Medema2020). These methods can also support the dissection of complex traits related to sexual reproduction and germination performance. Finally, genome-editing tools such as CRISPR/Cas represent a promising frontier not only for enhancing steviol glycoside biosynthesis and stress tolerance but also for resolving reproductive barriers by targeting genes controlling self-incompatibility, embryo development and seed viability (Ghose et al., Reference Ghose, Abdullah, Md Hatta and Megat Wahab2022). Established transformation protocols in stevia already demonstrate that generating transgenic or edited lines is technically feasible (Taak et al., Reference Taak, Tiwari and Koul2020).

While Stevia rebaudiana has long been considered a species resistant to full domestication due to its reproductive complexity, emerging molecular technologies open new avenues to overcome these barriers. Future research should explicitly target the genetic and physiological basis of sexual reproduction in stevia, which remains one of the least understood yet most critical aspects limiting its scalable propagation. Addressing this challenge could transform stevia from a semi-domesticated medicinal plant into a fully cultivated, seed-propagated crop.