Using DoE, correct nutrient combinations were identified, along with the effects of their interactions with the other environmental parameters: plant growth regulators. Similar methods have recently been used for optimizing callus cultures of various plant species [35,37,38]. This method has two advantages over the more common factorial design: DoE allows us to investigate the effects of independent variables between the actual experimental data points and allows a researcher to easily increase the number of experimental variables to more than five, which is not practical in typical factorial designs. Nitrogen quantity and form have been the subject of many growth medium optimization studies [39,40,41,42,43]. The optimum nitrate concentration is reported to be 20–30 mM for both growth and taxol production in cell cultures of Taxus yunnanensis Cheng et al. [40]. In an experiment on the NH₄⁺/NO₃⁻ ratio in the culture medium, ginsenoside production in the adventitious roots of Panax ginseng C.A. Meyer was affected by NH₄⁺/ NO₃⁻ ratios in the culture medium, showing the greatest productivity at 18.5 mM NO₃⁻ without NH₄⁺ [39]. Biomass growth and azadirachtin production of Azadirachta indica (A. Juss) suspension cultures are significantly improved in a medium with a high NH₄⁺/NO₃⁻ ratio [41]. According to our data, the NH₄⁺/K⁺ ratio is a crucial factor for biomass accumulation and production of TPCs in the callus culture of R. rosea. The optimal response was seen at the NH₄⁺/K⁺ ratio of 0.33 and 20–40 mM NO₃⁻.

Plant growth regulators are one of the most important factors owing to their important regulating role in plant physiology and biochemistry [44]. An appropriate proportioning of cytokinins and auxins can maintain the balance between differentiation and dedifferentiation and achieve the goal of rapid proliferation for plant cells in vitro [45]. To obtain callus cultures of Rhodiola species, cytokinin BAP is most often employed in combination with various auxins. Compact callus aggregate suspension cultures of Rhodiola imbricata are obtained on the MS medium supplemented with 3 mg/L NAA and 3 mg/L BAP [28]. Rhodiola quadrifida (Pall.) calluses are obtained from hairy roots in the MS medium with the addition of hormones: 3 mg/L 2,4-D and 0.5 mg/L BAP [31]. Calluses of R. sachalinensis Boriss. can be successfully cultivated on the MS medium supplemented with 3 mg/L BAP and 0.3 mg/L NAA [27]. The medium containing 1 mg/L 2,4-D, 2 mg/L NAA, 0.5 mg/L BAP, and 0.1 mg/L kinetin proved to be the best for the induction of the callus from R. quadrifida (the induction rate was 83.3%); the optimized combination of plant growth regulators for callus subculture is 1 mg/L 2,4-D, 0.1 mg/L BAP, and 0.5 mg/L kinetin [23]. Several other combinations of plant growth regulators have also been found to be effective for callus induction in species of the genus Rhodiola. One research group revealed that callus induction of R. imbricata is frequently achieved in juvenile leaves (100% frequency) and roots (87.50%) in the MS medium supplemented with 0.5 mg/L thidiazuron and 1 mg/L NAA [30]. In another study, to obtain a callus culture of R. rosea, leaves were placed on the surface of a fresh MS medium supplemented with 3 mg/L of N⁶-(2-isopentenyl) adenine and 0.3 mg/L IAA [29]. For callus subcultures and subsequent cell suspension cultures of R. crenulata L., full-strength MS containing 0.5 mg/L thidiazuron and 0.5 mg/L NAA turned out to be the best [26]. This paper optimized plant growth regulator proportioning, and maximum biomass and production of phenolic compounds were obtained in the medium containing BAP/NAA ratios of 0.33–1.00, provided that the concentration of plant growth regulators was 30 μM. Here, we found that optimal growth of a callus culture of R. rosea requires BAP > 0. Other reports suggest that the synergistic combinations of auxin and cytokinin can significantly alter the production of secondary metabolites depending on plant species [46]. When compared with the initial medium (control medium), total plant growth regulators content has not been changed. However, the selected complex of optimal factors, in general, contributed to a significant increase in the growth parameters of R. rosea calluses.

Here, the callus obtained under optimal culture conditions (NH₄⁺/K⁺ 0.33, BAP/NAA 1.0, BAP + NAA 30 μM, and NO₃⁻ 40 mM) was semi-friable yellowish green. The increase in the growth of fresh biomass on this medium was up to 703%, which is 2.7 times higher than the growth rates on the standard MS medium. Nonetheless, the level of TPC biosynthesis on the nutrient medium that gave the highest increase in callus biomass was 13% lower than that on the standard MS medium. The highest content of TPCs was observed in calluses grown on media with the NH₄⁺/K⁺ ratio of 0.33 and BAP/NAA of 0.33 and a concentration of plant growth regulators of 30 μM. Data on the profile and concentrations of phenolic compounds in in vitro cultures vary and sometimes are contradictory because such results are influenced by various factors and stages of plant development [47]. Using RSM, it has been reported that higher KH₂PO₄ depletion and 75 μM m⁻²s⁻¹ light intensity favored the biosynthesis of anthocyanins and the other phenolic compounds and resulted in elevated antioxidant capacity in grape (Bogazkere Cv.) callus culture [48]. Through Plackett–Burman’s design and RSM, optimal proportions of plant growth regulators for a cell suspension culture of Siraitia grosvenorii were obtained. With the optimized plant growth regulators, the obtained cell biomass and polyphenols content were 32.18% and 13.86%, respectively, more than plant growth regulators proportioning before optimization [49]. Using HPLC, we determined that the profile and levels of phenolic compounds were similar between the above-ground part of intact plants and the callus culture. TPC concentration varied among callus cultures from 14.9 to 71.6 mg/g, the number of phenolic compounds from 10 to 20; in the above-ground part, these values were 73.1 mg/g and 21 phenolic compounds, respectively. Several new phenolic compounds were identified in the callus cultures: compounds No. 9, 11, 19, 26, and 27 (see Table S1).

Jasmonic acid is thought to be involved in the signal transduction pathway that induces the production of defense compounds in plants, such as alkaloids, terpenoids, and polyphenols [50]. MJ is an effective elicitor that participates in plant defense response pathways and triggers plant metabolite biosynthesis. Accordingly, MJ has been used for inducing metabolite production in plant cell cultures [33]. The treatment of callus cultures with an elicitor led to qualitative and quantitative changes in the profile of phenolic compounds. In this case, MJ concentration was of paramount importance. The use of 100 μM MJ was optimal and led to an increase in the TPC content up to 47.9 mg/g. The effectiveness of MJ as an elicitor has been demonstrated for many in vitro cultures, including those of Rhodiola species [28,51,52,53,54,55,56,57]. For example, for Rhodiola it is reported that the levels of bioactive compounds increase with MJ supplementation in a dose-dependent manner. The highest salidroside content (4.75 mg/g dry weight) is obtained during treatment with MJ at 125 μM [28].

Although salidroside (0.53%) was found in the leaves of R. rosea from Rila Mountain, Bulgaria [24] and in the leaves and stems of R. rosea cultivated in Poland (salidroside 0.12% and total rosavins 0.3%), the aerial parts of the plant grow anew every year and therefore their content is consistent each time [59]. Rattan et al. [30] found that rosavin and rosarin are present at the highest concentration in root-derived compact green calluses (0.15 mg/g dry weight) and root-derived friable green calluses (0.07 mg/g dry weight). Kurkin et al. [60] noted that in a suspension culture of R. rosea, the main phenolic compound is a phenylpropanoid called triandrin, while in a callus culture, the process of biosynthesis went further and, together with triandrin, the major phenolic compounds were dimeric phenylpropanoids: lignans; in other words, “ageing” of the biomass took place. Those authors emphasized the finding that neither salidroside nor phenylpropanoids—which are characteristic for the rhizomes of roseroot stonecrop (rosin, rosavin, and rosarin)—were found in the samples of biomass. In the in vitro cultures, a significant enhancement in the production of rosin and its derivatives was observed when the cultures were fed with the precursor: cinnamyl alcohol [5]. In R. rosea compact callus aggregate cultures, the observed rosin and rosarin content was even higher than that in field-cultivated plants, while the rosavin level was five times lower [2].