First integrated in vitro regeneration protocol for the endangered aquatic plant Hydrocera triflora (Balsaminaceae) via optimized cytokinin–auxin balance

Abstract

Hydrocera triflora, a rare and endangered aquatic species of phylogenetic importance within the family Balsaminaceae, lacks any reported in vitro regeneration system. This study establishes the first complete in vitro propagation protocol for this taxon, integrating optimized sterilization, shoot induction, proliferation, and rooting. Stem explants (~ 3 cm) were disinfected using ethanol, 0.1% mercuric chloride with Tween, and 20% sodium hypochlorite, with the most effective method combining ethanol pre-wash and 0.1% mercuric chloride for 15 min, resulting in 35% contamination and 13% browning. Shoot induction reached 96% on medium supplemented with 1.0 mg L⁻¹ 6-benzylaminopurine and 0.2 mg L⁻¹ indole-3-butyric acid, producing 3–12 healthy shoots per explant with minimal callus. Moderate cytokinin concentrations optimized shoot proliferation, maximizing multiplication coefficient and vigor while avoiding excessive callusing. Rooting was highest on Murashige and Skoog medium with 0.2 mg L⁻¹ indole-3-butyric acid, where roots appeared by day 5, averaged 6.3 cm in length with 16.1 roots per plantlet, and achieved a 100% rooting rate. These results indicate that precise cytokinin–auxin balance governs organized morphogenesis in H. triflora, providing insights into the developmental plasticity and tissue-specific hormonal responses of this species. The established protocol offers a reliable platform not only for large-scale micropropagation and ex situ conservation but also for future biotechnological applications, including genetic improvement, secondary metabolite production, and translational studies in related recalcitrant aquatic species.

Introduction

Hydrocera triflora is an endangered aquatic perennial herb and the sole species of the monospecific genus Hydrocera within the family Balsaminaceae [1, 2]. In China, it was listed as regionally extinct (RE) in 2013 and reclassified as endangered in 2023 [3]. Balsaminaceae comprises only two genera, the species-rich Impatiens (~ 1,000 species) and Hydrocera, which are sister taxa based on morphological and molecular evidence and are classified as eudicots within the order Ericales [2, 4, 5]. Morphologically, H. triflora is an erect aquatic herb with narrow alternate leaves, zygomorphic flowers borne in short axillary pedicels, and berry-like indehiscent drupes; its submerged spongy stem facilitates oxygen uptake in wetland habitats [6–9].

At the genomic level, H. triflora shows strong similarity to Impatiens, with comparative chloroplast genome analyses revealing highly conserved genome structure, gene content, and gene order. The chloroplast genome of H. triflora comprises 115 genes (81 protein-coding, 30 tRNA, and 4 rRNA genes) and is remarkably consistent in size (154,188–154,189 bp) across studies [4, 10]. Owing to this close yet distinct relationship, H. triflora is widely used as an outgroup in phylogenetic studies of Impatiens and Balsaminaceae [2, 4, 10, 11]. Phylogenetic analyses indicate that Hydrocera diverged from Impatiens approximately 30.7 million years ago in the Late Eocene; however, unlike the extensive radiation observed in Impatiens, Hydrocera has remained monospecific, likely due to its restricted geographic distribution and less efficient seed dispersal [12].

H. triflora occurs in parts of South and Southeast Asia, but in mainland China its distribution is highly restricted to Hainan Province (Lingshui and Sanya) [3]. Ecologically, it inhabits shallow (40–70 cm), stable, mesotrophic freshwater habitats along lake margins, is highly sensitive to water-level fluctuations, and is often associated with aquatic taxa such as Blyxa, serving as an indicator of past lacustrine environments [13]. Due to habitat degradation and environmental changes, H. triflora now exhibits a highly restricted distribution range and has reached the level of regional extinction in some areas [14]. Although the species possesses high ornamental value due to its extended flowering period and attractive flower coloration, natural regeneration remains severely constrained. Seed propagation is the primary reproductive strategy in the Balsaminaceae family, and H. triflora is mainly propagated by seeds [15]. However, seed collection in the wild is challenging because mature seeds are easily shed, and the species exhibits low seed germination rates and slow propagation speed, all of which significantly hinder population recovery [15].

H. triflora is a phylogenetically important, ecologically sensitive, and morphologically unique species within Balsaminaceae, serving as a key taxon for understanding the evolutionary history of Balsaminaceae and related plant groups. Its endangered status, narrow habitat range, and limited natural regeneration make in situ conservation challenging. To date, no in vitro regeneration procedure has been reported for this species. This study aims to establish a reliable and efficient regeneration system using stem explants, providing a foundation for large-scale propagation, ex situ conservation, and future genetic, physiological, and evolutionary studies of this evolutionarily and ecologically significant species.

Materials and methods

Plant material and explant source

Plant materials were collected in 2023 from wetlands, including paddy fields and roadside ditches near General Hill, Haikou (GCJ02, 19.78537325° N, 110.28773911° E). Healthy and disease-free Hydrocera triflora plants at a similar vegetative growth stage were selected to ensure uniformity. A total of five individual plants were sampled, and nodal stem segments (approximately 3 cm in length) from the middle portion of the stem were excised and used as explants. The collected materials were transported to the laboratory in clean, moist conditions and processed on the same day of collection to minimize contamination and tissue deterioration. The species was identified based on morphological characteristics following standard taxonomic references. All chemicals and culture media used in the experiments were sourced commercially: 6-benzylaminopurine (6-BA), α-naphthaleneacetic acid (NAA), indole-3-butyric acid (IBA), indole-3-acetic acid (IAA), and 2,4-dichlorophenoxyacetic acid (2,4-D) were obtained from Biosharp Biotechnology Co., Ltd.; sodium hypochlorite and anhydrous ethanol were supplied by Xilong Scientific Co., Ltd.; Murashige and Skoog (MS) culture medium was purchased from Guangdong Zhongshan Baiwei Biotechnology Co., Ltd.; and carrageenan was obtained from Beijing Kangbeisi Technology Co., Ltd.

Disinfection of explants

The stem segments of H. triflora were cut into approximately 3 cm long pieces containing buds and initially cleaned using a commercial laboratory detergent (washing powder) under gentle manual agitation for 10 min to remove dust and debris. In this study, eight surface sterilization methods were evaluated, as detailed in Table 1. All disinfection steps, including initial cleaning with detergent and subsequent surface sterilization, were conducted under sterile conditions in a laminar flow hood to prevent contamination. The explants were washed five times with sterile water and inoculated into MS [16] medium (30 g L⁻¹ sucrose, 6.5 g L⁻¹ carrageenan, pH 5.8). Each explant was cultured individually in a separate vial and considered an independent biological replicate. For each treatment, a total of three independent replicates were used.

Table 1.

Different sterilization methods for explants of H. triflora

Treatment	Sterilization Protocol
A1	Rinse explants under running tap water for 30 min, wash 3× with sterile water, then perform 5 additional sterile water rinses before inoculation on MS medium.
A2	Same initial rinses as A1, followed by 1 min treatment with 75% ethanol, 5 sterile water washes, then inoculation.
A3	Initial rinses and ethanol sterilization (1 min) as in A2, followed by 10 min treatment with 0.1% mercuric chloride + 1 drop Tween, 5 sterile water washes, then inoculation.
A4	Same as A3, but mercuric chloride exposure increased to 15 min before final washes and inoculation.
A5	Same as A3, with mercuric chloride exposure extended to 20 min before final washes and inoculation.
A6	Initial rinses and 1 min ethanol sterilization, followed by 10 min treatment with 20% sodium hypochlorite, 5 sterile water washes, then inoculation.
A7	Initial rinses, brief 30 s ethanol sterilization, followed by 15 min 20% sodium hypochlorite treatment, 5 sterile water washes, then inoculation.
A8	Same as A7, but sodium hypochlorite treatment extended to 20 min before final washes and inoculation.

Explants were cultured on semi-solid Murashige and Skoog (MS) medium supplemented with vitamins and plant growth regulators as specified in each experiment. The medium was solidified with 0.8% (w/v) carrageenan and dispensed at a volume of 50 mL per culture vessel. Cultures were initially maintained in the dark at 26 °C for 7 days to support explant establishment, followed by growth at 26 °C under LED illumination with a 10 h light/14 h dark photoperiod and a light intensity of 150 µmol m⁻² s⁻¹. The contamination rate was recorded after seven days. The contamination rate was expressed as a percentage, obtained by (contaminated explants / total explants inoculated) × 100. The browning rate was expressed as a percentage of explants exhibiting browning relative to the total number of inoculated explants [17].

Screening of inducing hormones from H. triflora explants

Uncontaminated stem explants obtained after 7 days of sterilization were transferred to MS basal medium (B1) or MS media (B2–B6) supplemented with 6-BA alone or in combination with different auxins, including NAA, IBA, IAA, or 2,4-D, as shown in Table 2. For each treatment, ten culture vessels were prepared, each containing seven explants. Individual explants were considered independent biological replicates for data collection and statistical analysis. Explants were maintained in the dark at 26 °C for 7 days, followed by culture at 26 °C under LED illumination with a 10 h light/14 h dark photoperiod and a light intensity of 150 µmol m⁻² s⁻¹. Shoot induction efficiency was evaluated by recording the induction rate after 15 days of culture. The induction rate (%) was calculated as the number of explants producing adventitious shoots divided by the total number of inoculated explants, multiplied by 100.

Table 2.

Treatments of induction media with different hormones for stem segments of H. triflora . NAA, IBA, IAA, 2,4-D, and 6-BA represent α-naphthaleneacetic acid, indole-3-butyric acid, indole-3-acetic acid, 2,4-dichlorophenoxyacetic acid, and 6-benzylaminopurine, respectively

Treatments	induction media
B1	MS
B2	MS + 6-BA1.0 mg L⁻¹
B3	MS + 6-BA1.0 mg L⁻¹+NAA0.2 mg L⁻¹
B4	MS + 6-BA1.0 mg L⁻¹+IBA0.2 mg L⁻¹
B5	MS + 6-BA1.0 mg L⁻¹+IAA0.2 mg L⁻¹
B6	MS + 6-BA1.0 mg L⁻¹+2,4-D 0.2 mg L⁻¹

Screening of propagation medium for regeneration buds of H. triflora

Stem segments obtained from regenerated juvenile shoots were transferred to four distinct proliferation media (C1–C4), each based on Murashige and Skoog (MS) medium supplemented with a constant concentration of indole-3-butyric acid (IBA, 0.2 mg L⁻¹) and increasing levels of 6-benzylaminopurine (6-BA). The 6-BA concentrations tested were 0.5, 1.0, 1.5, and 2.0 mg L⁻¹ for C1, C2, C3, and C4, respectively. For each treatment, ten culture vessels were prepared, each containing seven explants. Individual explants were treated as independent biological replicates for data collection and statistical analysis. All cultures were incubated at 26 °C under LED illumination with a 12 h light/12 h dark photoperiod and a light intensity of approximately 150 µmol m⁻² s⁻¹. Shoot proliferation was evaluated after 20 days of culture.

Effects of different media on rooting of H. triflora

Uniform and robust tissue-cultured seedlings were selected, and individual plantlets were excised and transferred to different rooting media. The treatments were as follows: D1, MS medium as the control; D2, MS medium supplemented with 0.2 mg L⁻¹ α-naphthaleneacetic acid (NAA); and D3, MS medium supplemented with 0.2 mg L⁻¹ indole-3-butyric acid (IBA). For each treatment, ten culture bottles were prepared, each containing five plantlets. Individual plantlets were considered independent biological replicates for data collection and statistical analysis. Cultures were maintained in darkness at 26 °C for 5 days, followed by growth at 26 °C under LED illumination with a 16 h light/8 h dark photoperiod and a light intensity of 150 µmol m⁻² s⁻¹. Rooting performance was assessed after 15 days of culture. Different culture conditions were applied depending on the developmental stage.

Statistical analysis

All treatments were conducted in a Completely Randomized Design (CRD). Data were analyzed using R software (version 4.4.0). One-way ANOVA was performed using the base R function aov to evaluate differences among treatments, and normality was confirmed using the Shapiro–Wilk test (shapiro.test in base R). Means were compared by Duncan’s multiple range test using the agricolae package. Data are expressed as mean ± standard deviation (SD), and means followed by different letters differ significantly at p < 0.05.

Results

Influence of different disinfection methods on the contamination and browning of H. triflora explants

The effects of eight sterilization methods (A1–A8) on the contamination and browning rates of H. triflora stem explants are illustrated in Fig. 1, with significant differences observed among treatments at p < 0.05. The control treatment (A1) exhibited a 100% contamination rate with no browning. Similarly, A2, which included 75% ethanol treatment for 1 min, also showed 100% contamination, although 8% of explants exhibited browning. Methods using mercuric chloride (HgCl₂) significantly reduced contamination: A3 (0.1% HgCl₂ for 10 min) showed a contamination rate of 66% and a browning rate of 12%; A4 (0.1% HgCl₂ for 15 min) reduced contamination to 35% with 13% browning; and A5 (0.1% HgCl₂ for 20 min) further reduced contamination to 28% but increased browning to 45%. In contrast, treatments using sodium hypochlorite (NaClO) were less effective. A6 (20% NaClO for 10 min) resulted in 100% contamination with 40% browning, while A7 and A8 (15 and 20 min NaClO, respectively) showed contamination rates of 55% and 40% with extensive browning (80% and 100%). These results indicate that HgCl₂ is more suitable than NaClO for disinfecting H. triflora stem segments, providing a better balance between low contamination and minimal tissue browning.

Fig. 1.

Effect of different sterilization methods on the contamination and browning rate of explants of H. triflora. Error bars denote the standard deviation (SD), and means labeled with different letters are significantly different according to Duncan’s multiple range test at p ≤ 0.05

Among the tested methods, A4—rinsing in tap water for 30 min, washing three times with sterile water, sterilizing in 75% ethanol for 1 min, washing five times with sterile water, followed by 0.1% HgCl₂ with one drop of Tween for 15 min, and a final wash five times with sterile water before inoculation—was identified as the most effective sterilization protocol, achieving low contamination (35%) with minimal browning (13%).

Influence of various plant growth regulators on shoot induction in H. triflora explants

Impact of combined plant growth regulator treatments on shoot initiation from stem explants is summarized in illustrated in Figs. 2 and 3 with significant differences observed among treatments at p < 0.05. In the control MS medium (B1), only 10% of explants produced 1–2 shoots, and no callus formation was observed. Supplementation with 1.0 mg L⁻¹ 6-BA (B2) increased the induction rate to 50%, with 1–2 short shoots and minimal callus. The addition of 6-BA in combination with 0.2 mg L⁻¹ NAA (B3) further increased the induction rate to 55.7%, producing 2–3 shoots per explant with few calli, numerous roots, and yellowish-green regenerated shoots. The most effective induction medium was 6-BA combined with 0.2 mg L⁻¹ IBA (B4), which achieved a 96% induction rate, generating 3–12 strong, green shoots with minimal callus and well-developed shoot clusters, while root formation remained limited. Other hormone combinations were less effective. Medium B5 (6-BA + 0.2 mg L⁻¹ IAA) induced only 30% of explants to produce 1–2 shoots, accompanied by abundant callus and yellowish-green shoots. Medium B6 (6-BA + 0.2 mg L⁻¹ 2,4-D resulted in only 5% of explants forming a single shoot, with extensive callus and shoots too short to form clusters. These findings demonstrate that treatment B4, comprising 6-BA in combination with IBA, is the most effective for promoting the formation of multiple, vigorous shoots from stem segments of H. triflora, yielding robust, green, and well-organized shoot clusters with negligible callus production.

Fig. 2.

Effects of different plant growth regulator treatments (B1–B6) on shoot induction from stem explants of Hydrocera triflora. Representative images showing differences in shoot formation and callus development after 15 days of culture

Fig. 3.

Influence of various plant growth regulators on shoot induction frequency in H. triflora explants. Error bars represent standard deviation (SD), and means followed by different letters differ significantly according to Duncan’s multiple range test at p ≤ 0.05

Effects of different 6-BA doses on the proliferation of regenerated shoots from H. triflora

The effects of varying 6-BA concentrations on the proliferation of regenerated shoots of H. triflora are summarized in Table 3, with morphological characteristics illustrated in Figs. 4 and 5. Significant differences were observed among treatments (p < 0.05). All four tested media (C1–C4) supported shoot multiplication, producing 2–12 shoots per explant, with proliferation coefficients greater than 4. Medium C2 (MS + 1.0 mg L⁻¹ 6-BA + 0.2 mg L⁻¹ IBA) was the most effective, generating 3–12 shoots per explant. Shoots were the tallest (4.0–6.0 cm) and showed low callus incidence (15–20%), indicating optimal shoot vigor and minimal callus formation. Medium C1 (MS + 0.5 mg L⁻¹ 6-BA + 0.2 mg L⁻¹ IBA) produced 3–6 shoots per explant, with moderate shoot length (2.0–3.0 cm) and low-to-moderate callus incidence (14–21%). Medium C3 (MS + 1.5 mg L⁻¹ 6-BA + 0.2 mg L⁻¹ IBA) yielded 2–8 shoots per explant, with intermediate shoot length (2.5–4.0 cm) and higher callus formation (21–28%), reflecting less efficient morphogenesis. Medium C4 (MS + 2.0 mg L⁻¹ 6-BA + 0.2 mg L⁻¹ IBA) was the least favorable, producing only 2–4 shoots per explant with the shortest shoots (1.5–2.5 cm) and extensive callus (42–49%), indicative of poor shoot development and excessive callus formation. Based on the number of regenerated shoots, shoot length, and callus incidence, the ranking of media for shoot proliferation was C2 > C3 > C1 > C4. These results confirm that moderate concentrations of 6-BA combined with IBA optimize shoot multiplication while minimizing callus formation, highlighting the importance of hormonal balance in directing organized morphogenesis in H. triflora.

Table 3.

Influence of different culture media formulations on shoot proliferation in regenerated H. trifloral

Proliferation Medium	Number of Regenerated Shoots	Shoot Length (cm)	Callus Incidence (%)
C1	3–6	2.0–3.0	14–21
C2	3–12	4.0–6.0	15–20
C3	2–8	2.5–4.0	21–28
C4	2–4	1.5–2.5	42–49

Fig. 4.

Shoot regeneration responses of H. triflora to different 6-BA concentrations (C1–C4). Treatments correspond to 0.5, 1.0, 1.5, and 2.0 mg L⁻¹ 6-BA, respectively

Fig. 5.

Impact of different 6-benzylaminopurine (6-BA) levels on shoot multiplication in regenerated explants of H. triflora. The standard deviation is shown by bars, while Duncan’s multiple range test at the 5% probability level indicates significant variations between means by various letters

Morphologically, regenerated shoots on C2 were robust, green, well-developed, and formed minimal callus, with few roots. C1 produced green shoots with limited roots and moderate growth compared to C2. Shoots on C3 were shorter, slower-growing, and accompanied by some callus and roots, reducing subculture efficiency. C4 produced poorly developed shoots with extensive callus, weak growth, and limited root formation, resulting in the least efficient proliferation. Considering both multiplication rate and seedling vigor, the overall ranking for shoot growth was C2 > C1 > C3 > C4. These results indicate that C2 medium provides the optimal balance between high proliferation rate and vigorous seedling development for regenerated shoots of H. triflora.

Influence of various culture media formulations on in vitro rooting

Tissue culture seedlings of H. triflora were inoculated onto media D1–D3 to evaluate the effects of different hormone treatments on seedling vigor and rooting, with significant differences observed among treatments (p < 0.05; Figs. 6; Table 4). Quantitatively, seedlings cultured on D3 medium (MS + 0.2 mg L⁻¹ IBA) showed the best rooting performance, with roots appearing as early as day 5, an average of 16.1 roots per plantlet, and a maximum root length of 6.3 cm. Seedlings on D2 (MS + 0.2 mg L⁻¹ NAA) exhibited moderate rooting, with roots emerging on day 9, averaging 8.4 roots per plantlet and shorter roots, while seedlings on the control D1 medium (MS without hormones) showed the weakest rooting, with roots appearing on day 10, 3.8 roots per seedling, and a root length of 5.9 cm. Rooting rate and all measured parameters were significantly higher in D3 compared to the other treatments (p < 0.05).

Fig. 6.

In vitro rooting of Hydrocera triflora plantlets on different media: D1 (MS control), D2 (MS + 0.2 mg L⁻¹ NAA), and D3 (MS + 0.2 mg L⁻¹ IBA)

Table 4.

Effect of different culture media on seedling strengthening and rooting of tissue-culture-derived H. triflora. IBA = indole-3-butyric acid; NAA = α-naphthaleneacetic acid. ERT = earliest rooting time (days); MRL = mean root length (cm); MNR = mean number of roots per plantlet. Results are presented as mean ± SD, and distinct letters indicate statistically significant differences among treatments based on Duncan’s multiple range test at the 5% significance level

Treatments	Rooting Medium	ERT	MRL	MNR	Rooting Rate (%)	Growth Characteristics
D1	MS	10 ± 1a	5.9 ± 0.2a	3.8 ± 0.1c	75 ± 3c	Slender, few roots; moderate seedling growth
D2	MS + 0.2 mg L⁻¹ NAA	9 ± 1a	2.1 ± 0.1b	8.4 ± 0.4b	85 ± 3b	Thick, short roots; moderately strong seedlings
D3	MS + 0.2 mg L⁻¹ IBA	5 ± 1b	6.3 ± 0.3a	16.1 ± 0.3a	100 ± 7a	Slender, numerous roots; robust seedling growth

Morphologically, seedlings on D3 produced numerous slender roots and exhibited robust growth and uniform seedling development. Seedlings on D2 had thicker but shorter roots, and overall growth was less vigorous. In D1, seedling growth was uneven, with fewer roots and lower rooting rates (75%). Overall, the effectiveness of the media in promoting rooting and seedling vigor followed the order: D3 > D2 > D1. These results identify MS medium supplemented with 0.2 mg L⁻¹ IBA (D3) as the optimal rooting medium for H. triflora seedlings.

Discussion

Explant disinfection as a critical bottleneck for H. triflora in vitro culture

Explant disinfection represents a major bottleneck in establishing an in vitro culture system for H. triflora, particularly due to its aquatic habitat and constant exposure to humid, microbe-rich environments. Sterilization is a fundamental step in plant tissue culture, as it aims to eliminate microbial contaminants and provide aseptic conditions necessary for explant survival and development [17, 18]. Ineffective sterilization can severely reduce culture success by impairing explant growth, regeneration potential, and overall productivity [19]. In the present study, conventional sterilization approaches using ethanol alone or sodium hypochlorite were largely ineffective compared to HgCl₂, resulting in high contamination rates and, in several treatments, severe explant browning. Sodium hypochlorite, a widely applied oxidizing disinfectant, disrupts microbial membranes, but its efficacy depends strongly on microbial load, explant type, and tissue sensitivity [18, 20, 21], and similar limitations have been reported in species with high endogenous microbial populations [20, 22]. In contrast, mercuric chloride (HgCl₂) substantially reduced contamination while maintaining explant viability due to its protein-denaturing activity and inhibition of essential microbial enzymes [18, 22]. Its superior performance relative to NaClO has been documented in species characterized by persistent microbial contamination [17, 20, 21]. The present results further demonstrate that sterilization efficiency in H. triflora is highly dependent on exposure duration. An intermediate treatment with 0.1% HgCl₂ for 15 min (A4) provided the optimal balance between contamination control and minimal tissue browning, highlighting the importance of optimizing both disinfectant concentration and exposure time [19, 22]. Prolonged exposure reduced contamination further but caused severe oxidative browning, reflecting chemical injury rather than improved explant quality [21, 23]. Microbial contaminants can release proteins and secondary metabolites that alter the nutrient profile and pH of the culture medium, negatively affecting explant survival and morphogenesis [20, 21, 23]. Browning observed during explant sterilization is primarily caused by oxidative stress and the accumulation of phenolic compounds. Phenolics are oxidized by enzymes such as polyphenol oxidase (PPO), peroxidase (POD), and phenylalanine ammonia lyase (PAL), producing quinones that polymerize into brown pigments [17, 23]. The enzymatic formation of cytotoxic quinones damages cells, inhibits growth and regeneration, and contributes to reduced culture success during the establishment phase [23]. Effective disinfection reduces microbial metabolic activity and establishes conditions conducive to successful tissue establishment and regeneration [21].

Overall, these findings confirm that explant disinfection is a critical limiting step in H. triflora tissue culture, and that careful optimization of sterilant type and exposure duration is essential for successful culture initiation. The effectiveness of HgCl₂ observed here aligns with its reported use at concentrations of 0.05–0.1% for short exposure periods in sensitive plant tissues [17–19]. Despite identifying A4 as the best method among those tested, the relatively high contamination (35%) and browning (13%) rates indicate that establishment remains a significant bottleneck for practical micropropagation. Although mercuric chloride (HgCl₂) was effective in reducing contamination, its toxicity and environmental risks are well recognized [19]. Reliance on HgCl₂ is a limitation for conservation-focused applications, and future studies should explore safer or lower-toxicity alternatives for explant sterilization.

Auxin–cytokinin interactions in shoot and root regeneration of H. triflora

Efficient in vitro regeneration of H. triflora is strongly dependent on the precise balance between cytokinins and auxins throughout shoot induction, shoot proliferation, and rooting stages. Basal MS medium alone was insufficient to support organized shoot induction or reliable rooting, confirming the necessity of exogenous plant growth regulators for successful morphogenesis in this species [16–19]. Among all tested hormone combinations, MS medium amended with 1.0 mg L⁻¹ 6-BA and 0.2 mg L⁻¹ IBA consistently produced superior outcomes, yielding the highest shoot induction rate (96%), well-developed shoot clusters (3–12 shoots per explant), minimal callus formation, and vigorous green shoots. Other cytokinin–auxin combinations were less effective, with synthetic or stronger auxins such as NAA, IAA, and especially 2,4-D promoting excessive callus formation, abnormal shoot morphology, and reduced shoot quality [24, 25]. These results indicate that the effectiveness of 6-BA for shoot induction depends on the type of auxin used in combination, highlighting the importance of cytokinin–auxin balance for shoot development.

The superior performance of IBA in combination with 6-BA highlights its unique physiological role in regulating cytokinin–auxin interactions. IBA is widely regarded as one of the most effective auxins for organogenesis due to its chemical stability, low cytotoxicity, efficient conversion to indole-3-acetic acid, and targeted accumulation in morphogenetically active tissues [26–28]. Previous studies on water lilies and related taxa have similarly reported high shoot induction efficiencies when IBA is incorporated into the culture medium [17, 19], supporting the general applicability of this auxin for aquatic and semi-aquatic species. Shoot proliferation further confirmed the importance of cytokinin dosage optimization rather than maximization. A moderate concentration of 6-BA (1.0 mg L⁻¹) in combination with IBA resulted in the highest multiplication coefficient and best overall shoot vigor, whereas lower cytokinin levels limited proliferation and higher concentrations increased callus formation and compromised shoot quality. Excessive cytokinin has been reported to inhibit shoot elongation and disrupt normal development despite increasing shoot numbers [29–31]. These observations indicate that careful hormonal balance is crucial for physiological normality, and species-specific responses may occur; thus, validation under acclimatized conditions is recommended to ensure consistent performance.

Rooting responses further reinforced the central role of IBA in the regeneration system of H. triflora. Root induction and seedling strengthening were strongly auxin-specific, with MS medium amended with 0.2 mg L⁻¹ IBA outperforming both NAA-supplemented and hormone-free media. IBA promoted earlier root emergence, higher rooting rates, greater root numbers, and longer roots, producing slender, well-distributed root systems that supported superior seedling vigor and are likely to enhance survival during acclimatization. In contrast, NAA induced thicker but shorter roots, which may compromise transplant success, while hormone-free MS medium was inadequate for reliable root formation. These results indicate that exogenous auxin promotes rooting in H. triflora, with IBA showing the most favorable response, in agreement with observations reported for other woody and herbaceous species [27, 32, 33, 34, 35].

Taken together, the present study demonstrates that a regeneration system based on MS medium supplemented with 6-BA and IBA provides a coherent, stage-compatible hormonal regime for shoot induction, proliferation, and rooting of H. triflora. The repeated superiority of IBA across all developmental stages underscores its pivotal role in coordinated morphogenesis and highlights the importance of cytokinin–auxin balance rather than single-hormone optimization for successful in vitro regeneration of this evolutionarily and ecologically significant species.

Conclusion

The present work establishes a complete in vitro regeneration system for H. triflora, a rare and endangered aquatic species lacking previous protocols, while acknowledging that some steps, such as explant sterilization, remain limiting. This study fills a key gap in the biotechnological and conservation literature for H. triflora and the family Balsaminaceae. The optimized protocol integrates effective surface sterilization, shoot induction, shoot proliferation, and rooting, enabling reliable plant regeneration. Successful regeneration depended on precise optimization tailored to the high biological sensitivity of H. triflora. Mercuric chloride–based disinfection provided a balance between contamination control and explant survival. Shoot induction was maximized using a synergistic combination of the cytokinin 6-benzylaminopurine (6-BA) and the auxin indole-3-butyric acid (IBA), producing morphologically normal shoots with minimal callus formation. Moderate 6-BA concentrations supported shoot proliferation, while rooting on Murashige and Skoog medium supplemented with IBA promoted early root initiation and higher rooting frequency with well-developed roots. Overall, these results underscore the importance of hormonal balance rather than absolute hormone concentration in directing organized morphogenesis. Contamination during establishment remains a significant bottleneck, and safer, low-toxicity sterilization methods should be explored in future work. Beyond H. triflora, the established regeneration system provides a foundation for ex situ conservation, large-scale micropropagation, and biotechnological applications for other aquatic or recalcitrant plant species, as well as for future genetic, physiological, and evolutionary studies of phylogenetically significant taxa.

Authors’ contributions

Fei Lin conducted the experiments and drafted the initial manuscript. Yong Kang performed the data analysis. Yamei Li and Yuhua Guo participated in the analysis of the experimental results. Junmei Yin, Afrah E. Mohammed and Fenling Tang conceived and designed the project and guided the experimental design. Mamdouh A. Eissa and Afrah E. Mohammed revised the manuscript. All authors read and approved the final version of the manuscript.

Funding

This work was supported by the earmarked fund for CARS (CARS-23-G60) and by the Hainan Province Key Research and Development Project (ZDYF2023XDNY029). This research has been supported by the Princess Nourah bint Abdulrahman University Researchers Supporting Project number (PNURSP2026R740), Princess Nourah bint Abdulrahman University, Riyadh, Saudi Arabia.

Data availability

All data supporting the findings of this study are available within the paper and its Supplementary Information.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.