Salt-tolerant plants are able to grow and complete their life cycle on a substrate that contains high concentrations of soluble salt. Plants have developed a wide spectrum of mechanisms to adapt to salt stress, including regulating ion uptake and compartmentalization (selective accumulation or exclusion, compartmentalization in cells or organs, control of uptake by roots), osmotic adjustment (accumulation of compatible solutes), enhancing antioxidant capacity (free radical scavenging), hormone modulation (whole metabolism regulation), and stomatal closure (increasing water use efficiency) (Hasegawa et al., 2000; Parida and Das, 2005; Munns and Tester, 2008; Krasensky and Jonak, 2012; van Zelm et al., 2020).

Recent work suggests that red plant pigments can also play an important role in salt tolerance (Polturak et al., 2018). Concentrations of anthocyanin, a type of red pigment commonly occurring in plants, often increase substantially in vegetative shoots exposed to salt stress, and this has been associated with improved salinity tolerance. Salt treatment increased anthocyanin accumulation in wheat leaves, and wheat genotypes with high anthocyanin content were able to maintain significantly higher dry matter production after salt stress treatment (Mbarki et al., 2018). Ectopic expression of an Arabidopsis thaliana uridine diphosphate (UDP) glycosyltransferase (UGT79B2/B3) significantly increased anthocyanin concentration and enhanced salt tolerance, with increased plant antioxidant capacity thought to be the mechanism (Li et al., 2017). Similarly, overexpression of the Antirrhinum majus anthocyanin pathway transcriptional activators Delila or Rosea1 enhanced anthocyanin production in Nicotiana tabacum leaves and flowers, which improved salt tolerance by augmenting radical-scavenging activities (Naing et al., 2017, 2018).

Betalains are a group of water-soluble pigments that include the red betacyanins and the yellow betaxanthins. In plants, they occur solely among certain families in the Caryophyllales, where they are considered to replace the anthocyanins. Their base moiety is derived from tyrosine and the core pathway is comprised of only a few enzymatic steps (Supplementary Figure 1), the genes for which have been identified (Strack et al., 2003; Christinet et al., 2004; Sasaki et al., 2005; Hatlestad et al., 2012; Polturak et al., 2016; Polturak and Aharoni, 2018). The accumulation of betacyanin pigments in vegetative tissue under salt stress has been observed in a range of species including Amaranthus hypochondriacus, Portulaca oleracea, Alternanthera philoxeroides, and Mesembryanthemum crystallinum (Casique-Arroyo et al., 2014; Ribeiro et al., 2014; Mulry et al., 2015; Oh et al., 2015), and has been correlated with induction of betalain biosynthetic gene expression (Casique-Arroyo et al., 2014). One of the most studied species is Disphyma australe, a member of the Aizoaceae (iceplant) family. The red coloration in D. australe results from the production of betacyanin (Chinnock, 1971). In D. australe “red” and “green” vegetative morphs exist naturally, with the red morphs being more salt tolerant (Jain and Gould, 2015; Jain et al., 2015). Furthermore, the induction of betacyanins in the green morph by feeding the pathway intermediate L-DOPA reduced the detrimental effects of salt treatment, suggesting the benefit is caused specifically by betacyanin production (Jain et al., 2015).

The mechanism(s) through which betacyanins help with salt tolerance are not fully understood. Given the capacities of these red pigments to absorb green and yellow quanta, they should be able to effectively reduce excitation pressure on chloroplasts for which the light harvesting properties have been compromised by salt stress. Accordingly, the D. australe red morph shows less evidence of photo-oxidative damage than does the green one, and induced betacyanin production in the green morph provided photoprotection during salt stress (Jain and Gould, 2015; Jain et al., 2015). Photoprotection from anthocyanins has also been implicated in enhancing salt tolerance (Narbona et al., 2018; Zheng et al., 2019; Lo Piccolo et al., 2020).

Salt stress causes over production of reactive oxygen species (ROS), which leads to cellular damage and even plant cell death. Betacyanins are noted for their strong antioxidant properties in in vitro assays (Belhadj Slimen et al., 2017), and may have a role in planta as ROS scavengers during stress conditions. However, at least for D. australe, the accumulation of the pigments is restricted to the epidermis, arguing against such a role in salt stress responses in this species (Jain et al., 2015).

Recently, researchers have successfully transferred the genes required for betacyanin biosynthesis into non-betacyanic species (Polturak et al., 2016, 2017), allowing model systems to be developed to test causative roles for betacyanin pigments in abiotic and biotic stress tolerance. To further investigate the proposed contribution of betacyanin pigmentation to salt stress tolerance, we produced betacyanic N. tabacum, a species that does not naturally produce these pigments. We hypothesized that: (1) the benefit that betacyanins confer for plant salt tolerance could be transferred to non-betacyanic plants; (2) the production of betacyanin would deliver photoprotection and ROS scavenging benefit to plants under salt stress.

We demonstrate the beneficial effect of betacyanins on N. tabacum photosynthesis protection, ROS scavenging capability and seedling survival under salt stress. This study also illustrates a new approach for conferring increased salt tolerance to plants, which may prove useful for generating more resilient crop plants.

Introducing three betalain biosynthetic genes, BvCYP76AD1, MjcDOPA5GT, and BvDODA1, resulted in heterologous betalain production in N. tabacum. These three betalain genes have been previously used to introduce betalain production into N. tabacum and other solanaceous species (anthocyanic rather than betalainic plants) as well as microorganisms (Polturak et al., 2016, 2017). The red leaf pigments in our BtOE plants were confirmed as betacyanins.

When plants are expose to salt stress, their photosynthetic activity is significantly depressed, which leads to problems with excessive light energy (Takahashi and Murata, 2008; van Zelm et al., 2020). The excess excitation energy within the photosynthetic apparatus can cause impairment of chloroplast performance, revealed by a rapid decline in F_v/F_m. Therefore, light screening may provide a significant benefit to plants undergoing salt stress under relatively high light conditions, and this is supported by the results for our betacyanic N. tabacum. The BtOE lines had improved F_v/F_m values under stress conditions compared with control lines, and recovered significantly faster (Figure 7). Notably, betacyanin pigment was abundant throughout leaf tissue (Figure 3B), and so could reduce the photosynthetically active light passing to photosynthetic cells beneath (Li et al., 2019). Covering WT and EV leaf disks with a red filter simulating betacyanins had a similar effect on F_v/F_m as in planta betacyanin production (Figure 6), supporting light screening as one of the principal benefits of betacyanin production. Betacyanins have similar spectral properties as anthocyanins (Neill and Gould, 1999; Li et al., 2019), and there is abundant evidence that anthocyanins assist in protecting chloroplasts from potentially damaging effects of supernumerary photons directly by abating incident quantum fluxes (Landi et al., 2015).

The photoprotective function of betacyanins has also been reported in studies on betalainic species. In Amaranthus cruentus, photoinhibition was greater in green leaves than in betacyanic leaves and it was suggested that betacyanins reduced the excitation pressure on PSII by attenuating the harmful excess light (Nakashima et al., 2011). Suaeda salsa seedlings that accumulated betacyanins not only showed a higher resistance to photoinhibition but also recovered more quickly after light stress (Wang and Liu, 2007). In D. australe, photoprotection was indicated as part of the mechanism by which betacyanins increased salt stress tolerance, as pigmented plants had greater PSII quantum yields and photochemical quenching ability and recovered swifter after salt treatment than the green plants (Jain and Gould, 2015; Jain et al., 2015). Moreover, betacyanic D. australe leaves absorbed more green and yellow light (Jain and Gould, 2015). A photoprotective role for betacyanins could explain their relative abundance in species found in stressful environments such as salt marshes and deserts (Ehrendorfer, 1976; Jain and Gould, 2015).

In the presence of abiotic stresses, such as drought or salinity, seedling establishment is a critical process during the life cycle of a plant (Bohnert et al., 1995; Adolf et al., 2013). At this life stage, halophytes have a particular advantage over glycophytes on saline soils (Malcolm et al., 2003; Debez et al., 2004; van Zelm et al., 2020). BtOE seedlings had a greater survival rate after exposure to high salinity (Figure 8B). Similar to as has been observed for seeds of some naturally betalanic plants, such as Chenopodium quinoa (Escribano et al., 2017), the BtOE seeds contain betacyanins (Figures 2A–C), and leaves, stem and roots of the germinated seeds had a strong red-violet hue (Figures 2D,E). The higher survival rate of BtOE seedling than EV and WT (Figure 8B) could be because of the photoprotection benefit delivered by the presence of betacyanins in leaves and/or increased ROS scavenging capacity throughout the seedling. Increased production of the more common vacuolar red pigments of plants, anthocyanins, has also been shown to improve abiotic stress tolerance. For example, overproduction of anthocyanin significantly enhanced Arabidopsis seedling survival under salt stress, while reduction of anthocyanin production decreased survival rates (Oh et al., 2011).

In addition to photoprotection, the production of betacyanins in N. tabacum may also have contributed to improved salt stress tolerance through increased capacity for ROS scavenging, as the ABTS assay showed that BtOE lines had significantly increased total antioxidant capacity under either control or stress conditions (Figure 9). The ABTS assay has been showed to give a good measure of antioxidant capacity in N. tabacum, with both the ABTS and 2,2-diphenyl-1-picrylhydrazyl (DPPH) assays giving very similar results (Naing et al., 2017, 2018). Both light and salt stress can induce the formation of supernumerary ROS (van Zelm et al., 2020), such as singlet oxygen and H₂O₂, which can cause oxidative damage to cell membranes, proteins, and DNA (Gill and Tuteja, 2010). Both betanin and isobetanin, two of the three types of betacyanin identified in BtOE leaves, have been shown to have ROS scavenging properties in vitro that are approximately three times greater than an equivalent amount of ascorbic acid (Cai et al., 2005). Also, purified betacyanin from Suaeda japonica could prevent H₂O₂-induced protein oxidation (Hayakawa and Agarie, 2010). Similar to the BtOE results for salt stress, transgenic betacyanic tomato (Solanum lycopersicum) plants had increased antioxidant capacity (60% increase was measured in a fruit extract); and increased ROS scavenging was suggested as contributing to the enhanced resistance against phytopathogenic Botrytis cinerea in transgenic betacyanic N. tabacum (Polturak et al., 2017). An antioxidant benefit associated with betacyanins has also been suggested for naturally betacyanic species. For example, betacyanin production is induced by H₂O₂ application to S. salsa roots (Wang and Liu, 2007) or Berberis vulgaris leaves (Sepúlveda-Jiménez et al., 2004). As with light screening, anthocyanins as well as betacyanins have been shown to have ROS scavenging activity. In an in vivo study, anthocyanins were identified as one of the major low molecular weight antioxidants in leaves, significantly enhanced H₂O₂ scavenging (Gould et al., 2002). Transgenic N. tabacum and Arabidopsis thaliana plants with increased anthocyanin production had higher total ROS scavenging activity than control plants (Li et al., 2017; Naing et al., 2017, 2018).

The relative benefits from anthocyanins of direct light screening compared with ROS scavenging is of much debate (Agati et al., 2012, 2013, 2020), and the same arguments can be applied to betalains (Davies et al., 2018). A common observation, supporting the proposition that they principally have a screening function, is that betacyanins are predominantly located in epidermal cells [e.g., in D. australe (Jain et al., 2015)], while ROS generation is primarily in sub-epidermal photosynthetic palisade cells. However, for BtOE N. tabacum, the promoters of the transgenes allowed less restricted pigmentation, with betacyanin accumulating throughout the leaf including palisade and spongy mesophyll cells (Figure 3B). Thus, betacyanins could have provided an extra benefit in this case even if they do not normally have this role. In addition to epidermal-localization, another argument that could be raised against betacyanins functioning as key antioxidants is that they are localized within the vacuole while principal sources of ROS are chloroplasts, mitochondria and peroxisomes (Apel and Hirt, 2004). Although the full sub-cellular localization of betalains has not been confirmed for betalain over-production transgenics, they appeared to be accumulated throughout the vacuole at least (Figures 3E,F). However, there is evidence that H₂O₂ can diffuse easily between cellular compartments and so can come into contact with the secondary metabolites in the vacuole (Gould et al., 2002; Agati et al., 2012). Moreover, detailed arguments for an effective role of flavonoids as antioxidants and/or activity within the REDOX system have been recently presented (Chapman et al., 2019; Agati et al., 2020; Gayomba and Muday, 2020), and many of these proposals could also apply to betacyanins. An ability to reduce H₂O₂ concentrations is also supported by the finding that in Catharanthus roseus 90% of total leaf class III peroxidase (POX) activity was localized in the vacuoles (Ferreres et al., 2011), and the vacuolar-located POX and secondary metabolites (SMs) were proposed to have a key role in the homeostasis of H₂O₂ content (Ferreres et al., 2011). An indirect observation supporting a ROS scavenging role by betacyanins in the BtOE N. tabacum is the increased carotenoid content, compared with WT or EV lines, under both salt and light stress (Figure 5). Carotenoids are thought to be important ROS scavengers that are degraded during the reaction, especially in association with singlet oxygen scavenging during photo-oxidative stress (Burke et al., 2000; Ramel et al., 2012; Foyer, 2018). Therefore, the increased carotenoid content may reflect reduced ROS amounts. However, it is difficult to separate the possible ROS scavenging role from a reduction in ROS resulting from reduced photo-oxidative stress following direct light screening by the betacyanins.

The WT (but not EV) plants had a small, but statistically significant, higher amount of carotenoids and chlorophyll under non-stress conditions. Previous studies have found the similar trends. For example, betacyanins were negatively correlated with chlorophyll and carotenoids content in shoots of Salicornia brachiate (Parida et al., 2018). The same trend has also been reported for anthocyanin and chlorophyll content (e.g., Ren et al., 2019), suggesting the increased light-screening can lower absolute amounts of photosynthetic pigments under non-stress conditions.

Scavenging of ROS depends on a range of enzymatic components activity within the cell (Das and Roychoudhury, 2014). Transgenic N. tabacum plants with increased anthocyanin production (through overexpression of two anthocyanin-related transcription factors) also had increased transcript abundance for several important antioxidant enzymes and a salt tolerance-related genes such as POX, SOD, CAT, and osmotin (Naing et al., 2017, 2018). However, we found that introduction of betacyanin biosynthesis into N. tabacum using biosynthetic genes did not cause increased transcript abundance for POX, SOD, CAT, or osmotin (Supplementary Figure 6), nor changed phenylpropanoids profile (Supplementary Figure 4). It is possible that the transcription factor transgenes used for increasing anthocyanin biosynthesis also activate other stress-related pathways, and this does not occur with the specific betalain biosynthetic transgenes used. Further analysis on oxidative damage parameters and cell membrane function in the future study will be needed in order to elucidate the antioxidant effect of betacyanins in plant in vivo.

Anthocyanins and betacyanins have similar light screening properties. If light screening is the only mechanism by which salinity tolerance is enhanced by these pigments, then betacyanins may not offer better salt-tolerance properties than anthocyanins; yet they are commonly produced in halophytes such as D. australe. It could be that betacyanins contribute to salinity tolerance when produced in D. australe in additional ways that are not transferred across to N. tabacum. Knowledge is scant on many aspects of betalain production in comparison to our understanding for anthocyanins. Generally, little is known about betalain intra- or inter-cellular transport and accumulation mechanisms, whether other compounds are produced from the intermediates within the betalain biosynthetic pathway, or if transcriptional regulators of the pathway also regulate other metabolic responses. Any one of these could also contribute to improved salinity responses, and highlights the need for further research on the betalain pathway. The evolution of betalain pigmentation and loss of anthocyanin pigmentation in the Caryophyllales is intriguing. The ability to make transgenic plants of a single species that accumulate either type of red pigment, or both together, should aid studies to elucidate their comparative advantages under different stresses.

In addition to photoprotection and/or ROS scavenging, betacyanins could potentially contribute to salt tolerance by reducing uptake of Na ions or causing the Na ions to be sequestered away from sensitive cellular components. The activity of an H⁺-ATPase (V-ATPase) involved in Na⁺ vacuolar transport increased more in red betacyanic leaves than green leaves of Suaeda salsa when plants were exposed to salt stress (Wang and Liu, 2007), suggesting betacyanins could alter Na⁺ sequestration. However, we found no difference in either Na⁺ concentration (Supplementary Figure 7, Supplementary Methods) or distribution (Supplementary Figure 8, Supplementary Methods) between leaves of WT or BtOE plants with or without 400 mM NaCl treatment. Thus, in the transgenic N. tabacum, light screening and/or ROS scavenging are the more probable mechanism for conferring salt stress tolerance. It is possible that in the genetic background of halophytes such as Suaeda salsa and Disphyma australe there are additional components that act with the betacyanins to alter Na⁺ transport.

In conclusion, we demonstrated a causative role for betacyanins in salt tolerance and showed it is possible to transfer advantages of betacyanin pigmentation to a non-betalainic species. The betacyanins provided photoprotective light screening and potentially functioned as ROS-scavengers. Finally, as soil salinity is one of the most important agricultural problems, recently estimated to cause US$ 27.3 billion in global agricultural losses (Qadir et al., 2014), introducing betacyanin biosynthesis could be a new approach to increase the salt tolerance of crops for production on salinizing soils.

The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.

KG, KS, KD, and KR conceived the project. YZ made the transgenics and conducted the experiments. TK and YZ performed photosynthesis measurement and sodium ion concentration and distribution analysis. DL, SA, and TM conducted metabolite analyses. YZ, KS, and KD contributed to the experimental design, interpretation of the data, and wrote the manuscript. KG and KR contributed to the manuscript editing. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.