Impact of nicotine pathway downregulation on polyamine biosynthesis and leaf ripening in tobacco

Abstract Traditional breeding and molecular approaches have been used to develop tobacco varieties with reduced nicotine and secondary alkaloid levels. However, available low‐alkaloid tobacco varieties have impaired leaf quality likely due to the metabolic consequences of nicotine biosynthesis downregulation. Recently, we found evidence that the unbalanced crosstalk between nicotine and polyamine pathways is involved in impaired leaf ripening of a low‐alkaloid (LA) Burley 21 line having a mutation at the Nic1 and Nic2 loci, key biosynthetic regulators of nicotine biosynthesis. Since the Nic1 and Nic2 loci are comprised of several genes, all phenotypic changes seen in LA Burley 21 could be due to a mixture of genetics‐based responses. Here, we investigated the commercial burley variety TN90 LC and its transgenic versions with only one downregulated gene, either putrescine methyl transferase (PMT‐RNAi) or PR50‐protein (PR50‐RNAi). Nicotine levels of cured lamina of TN90 LC, TN90 PMT‐RNAi and TN90 PR50‐RNAi, were 70.5 ± 3.8, 2.4 ± 0.5, and 6.0 ± 1.1 mg/g dry weight, respectively. Low‐alkaloid transgenic lines showed delayed leaf maturation and impaired leaf quality. We analyzed polyamine contents and ripening markers in wild‐type TN90 control plants (WT) and the two transgenic lines. The ripening markers revealed that the PMT‐RNAi line showed the most pronounced impaired leaf maturation phenotype at harvest, characterized by higher chlorophyll (19%) and glucose (173%) contents and more leaf mesophyll cells per area (25%), while the ripening markers revealed that maturation of PR50‐RNAi plants was intermediate between PMT‐RNAi and WT lines. Comparative polyamine analyses showed an increase in free and conjugated polyamines in roots of both transgenic lines, this being most pronounced in the PMT‐RNAi plants. For PMT‐RNAi plants, there were further perturbations of polyamine content in the leaves, which mirrored the general phenotype, as PR50‐RNAi transgenic plants looked more similar to the WT than PMT‐RNAi transgenic plants. Activity of ornithine decarboxylase, the enzyme that catalyzes the committing step of polyamine biosynthesis, was significantly higher in roots and mature leaves of PMT‐RNAi plants in comparison to WT, while there was no increase observed for arginine decarboxylase. Treatment of both transgenic lines with polyamine biosynthesis inhibitors decreased the polyamine content and ameliorated the phenotype, confirming the intricate interplay of polyamine and nicotine biosynthesis in tobacco and the influence of this interplay on leaf ripening.


| INTRODUC TI ON
Extensive research based on traditional breeding and molecular biology approaches has been used to better understand the nicotine biosynthetic pathway in tobacco plants, how this pathway is regulated and the role that nicotine has in the plant. Reducing nicotine and related alkaloid levels has been a breeding target for tobacco scientists, and tobacco breeding lines with low-alkaloid content have been around for several decades (Legg et al., 1970). These plants were generated through traditional breeding and possess recessive alleles at the Nic1 and Nic2 loci (sometimes also named A and B loci), which have a synergetic effect of downregulating nicotine biosynthesis (Kidd et al., 2006;Legg & Collins, 1971;Shoji et al., 2010).
Unfortunately, the leaf ripening process is disturbed in these lowalkaloid (LA) tobacco varieties leading to inferior leaf quality (Chaplin & Burk, 1983;Chaplin & Weeks, 1976;Legg et al., 1970). Recently, we showed that the crosstalk between nicotine and polyamine biosynthesis is disturbed in these plants, and the accumulation of free and conjugated polyamines contributes to the impairment of leaf ripening (Nölke et al., 2018). Higher levels of polyamines have been shown to increase the longevity of tomato vines and to delay ripening and leaf senescence in transgenic tomato and salad plants (Mehta et al., 2002;Nambeesan et al., 2010;Serafini-Fracassini et al., 2010). Polyamines may act directly by stabilizing cell walls or through crosstalk with phytohormones such as ethylene, abscisic acid, cytokines, and gibberellins (Anwar et al., 2015). Treatment of LA Burley 21 plants with polyamine biosynthesis inhibitors or ethephon (Nölke et al., 2018) reduced the accumulation of polyamines and achieved a partial amelioration of the aberrant phenotype of the LA Burley 21. However, there is a series of genes (≥8) deleted in LA Burley 21, relative to Burley 21 (Shoji & Hashimoto, 2015;Shoji et al., 2010), and some of these genes that are not related to nicotine biosynthesis could also have an influence on the general physiology, stress response, and senescence of the tobacco plants (Chaplin & Weeks, 1976;Kidd et al., 2006). This makes it difficult to study the effect that the overly accumulating polyamines have on the ripening process in LA Burley 21 tobacco. Therefore, evaluating polyamines in transgenic plants, ones that have low-alkaloid content due to the downregulation of only one enzyme in the nicotine biosynthesis pathway, should give a clearer picture of the influence that polyamines have on the ripening process of LA tobacco.
There have been various transgenic approaches to reduce alkaloids in tobacco, e.g. downregulation of ornithine decarboxylase (ODC) (Dalton et al., 2016;DeBoer et al., 2013), berberine bridge enzyme-like (BBL) (Lewis et al., 2015) or putrescine methyltransferase (PMT) (Chintapakorn & Hamill, 2007) enzymes. While all of these strategies lead to reduced alkaloid content, most have drawbacks; for example, the plants with the downregulated ODC had not only lower alkaloid content but also lower polyamine content leading to a dwarfed phenotype (Nölke et al., 2005), while downregulation of the BBL gene family resulted in reduced yields of cured tobacco (Lewis et al., 2015). Thus, it seems that the ideal process by which LA tobacco varieties could be developed and used for commercial manufacture of LA tobacco products of high quality has still not been found.
Here, we describe our analysis of two transgenic Burley TN90 plants, the PMT-RNAi and PR50-RNAi plants (Kudithipudi & Hayes, 2018;Wang et al., 2000), which have a suppressed PMT gene or PR50 gene via RNA interference, resulting in lower nicotine content.
We conducted greenhouse experiments to compare the phenotype of these plants with WT plants using ripening markers (Nölke et al., 2018) and analyzed the polyamine content in the roots and leaves of both the transgenic and the WT plants during the ripening process. For PMT-RNAi plants, the activity of ODC and arginine decarboxylase (ADC), as the enzymes responsible for polyamine formation, was investigated.
The transgenic plants were furthermore treated with the polyamine inhibitor difluoromethylornithine (DFMO) to study if such treatment has an influence on the ripening phenotype of the plants. The data generated here give further insights into the relationship among nicotine biosynthesis, polyamine levels, and leaf/plant morphology. compared to the other three time points (before flowering, at topping, and 1 week post-topping) ( Figure 1a). However, the leaves of the PMT-RNAi and PR50-RNAi plants contained significantly (p < .05) higher levels of chlorophyll than the WT at harvest (16% and 12% more in the PMT-RNAi and PR50-RNAi, respectively).

| Morphological and biochemical differences between low-alkaloid tobacco varieties and WT during leaf ripening
Consequently, the leaves of PMT-RNAi and PR50-RNAi plants were greener than the WT (Figure 1b,c).
Evaluation of the size of the mesophyll cells in leaf 15 (numbered from base of the plant) revealed that the PMT-RNAi plants had smaller cells compared to wild type (Figure 2b). The number of cells per mm 2 leaf area was higher throughout the leaf ripening ranging from 21% more (before flowering, at topping, and 1 week post-topping (WPT)) to 25% more (at harvest) compared to the WT ( Figure 2a). The PR50-RNAi plants contained 17% more mesophyll cells per mm 2 leaf area at harvest, but no significant difference from the WT was measured at earlier time points (Figure 2a).
Monitoring of the free glucose concentration in leaf 11 (numbered from base of the plant) showed that both transgenic lines had significantly (p < .05) more glucose than the WT at topping (140% in PMT-RNAi line and 42% in PR50-RNAi) ( Table 1). The glucose content in PMT-RNAi increased more than in the WT during the whole leaf ripening process, reaching the highest levels at harvest (173% more than in WT). No glucose measurements were performed for PR50-RNAi after topping.

| Both transgenic genotypes accumulated higher levels of polyamines in roots
To investigate the impact of the PMT and PR50 downregulation on polyamine biosynthesis, we analyzed the levels of free and conjugated putrescine, spermidine, and spermine by liquid chromatography tandem mass spectrometry (LC-MS/MS). Time-course monitoring of the total polyamine content in leaves at the same developmental stage-that is, leaf 12 before flowering, leaf 19 at topping, leaf 22 at 1 WPT, and leaf 24 at harvest-revealed no significant differences  (Figure 3a,c). In contrast, polyamine analysis in roots showed that the PMT-RNAi plants accumulated significantly (p < .05) higher levels of total polyamines at topping (28%) and at harvest (57%), compared to the WT (Figure 3b). Significantly (p < .05) higher total polyamine content was also measured in the roots of PR50-RNAi plants at harvest (42%) than in the WT (Figure 3d).
Comparative analysis of the polyamine composition revealed that before flowering PMT-RNAi leaves contained significantly (p < .05) higher free spermine content (40%), while no changes in other polyamine fractions were observed (Figure 4a,b). Conversely, at topping the levels of free putrescine and spermine decreased by 46% and 62%, respectively, while the conjugated spermine increased by 73% (Figure 4a,b). No significant difference in the polyamine composition was observed at 1 week post-topping, and at harvest the free putrescine and spermine content increased significantly (p < .05) by 34% and 54%, respectively, compared to the WT. Free spermine was significantly higher in the leaves of PR50-RNAi line at topping and at harvest (90% and 54% higher, respectively), while no other differences in the polyamine composition compared to the WT were measured in the leaves of the PR50-RNAi plants (Figure 5a,b).
In the roots, the free putrescine fractions increased during ripening of the PMT-RNAi and PR50-RNAi plants compared to the WT.
The greatest relative increase in free putrescine was observed at harvest, that is, 320% increase in PMT-RNAi ( Figure 4c) and 296% increase in PR50-RNAi lines ( Figure 5c). The PMT-RNAi lines had significantly (p < .05) higher conjugated putrescine at topping (27%) and harvest (56%) (Figure 4d). Conjugated putrescine was also significantly higher in the roots of PR50-RNAi line at harvest (33%) compared to the TN90 control ( Figure 5d). Taken together, these data indicate that the PMT downregulation has a stronger impact on the polyamine biosynthesis pathway than the PR50 downregulation.

| Enzymatic activity of ODC varies significantly between PMT-RNAi and WT
To further investigate the effect of PMT downregulation on polyamine biosynthesis, the activities of ADC and ODC were analyzed in the leaves and roots of the PMT-RNAi plants that showed the strongest changes in the polyamine content, and the WT plants as well ( Figure 6). The ODC activity was significantly lower (−22%, p < .05) in the leaves of PMT-RNAi plants at topping, but was significantly higher (55%, p < .05) in leaves at harvest. Roots of the PMT-RNAi plants showed significantly higher ODC activity both at topping (100%, p < .05) and at harvest (239%, p < .05). While in roots the ADC activity remained low in accordance with earlier findings (Nölke et al., 2018), in leaves ADC activity followed the same pattern as ODC activity, although the differences between PMT-RNAi and WT were less pronounced and not statistically significant.

| D ISCUSS I ON
Low nicotine content tobacco has the side-effect of aberrant leaf maturation which in turn leads to inferior product quality. As analyzed previously (Nölke et al., 2018), polyamine biosynthesis is disturbed and upregulated in LA Burley 21 tobacco, which has lost eight or more genes, making analyses of cause-effect relationships difficult. In this study, we investigated two different transgenic TN90 plant lines where each has only one directly downregulated gene, either PMT (a key enzyme in nicotine biosynthesis) or PR50 (a 40S ribosomal protein S 12 homolog that is differentially expressed in roots of Nicotiana tabacum cv Burley 21 during the early stages of alkaloid biosynthesis) (Wang et al. 2000). Downregulation of PMT or PR50 genes led to lower nicotine levels in tobacco (Kudithipudi & Hayes, 2018). This work was conducted to provide insights into how the leaf ripening process is dis- The glucose concentrations in the PMT-RNAi leaves increased more rapidly and to higher values than in the WT so that the PMT-RNAi plants contain nearly three times as much glucose compared to the WT at harvest. Burley tobacco is characterized by low sugar concentrations in the leaves (Banožić et al., 2020) so an increase in glucose presents a potentially negative influence on the tobacco quality. It is known that while higher sugar concentration often increases the quality of a tobacco, there needs to be a balance among sugars, waxes, and resins (Mendell et al., 1984). It can be assumed that the inferior product quality of PMT-RNAi tobacco is partially due to the high glucose content in the leaves.
When total polyamines were measured in the transgenic and the WT plants, no significant differences were detected in leaves, while roots showed a significant increase in polyamines in the PMT-RNAi and PR50-RNAi plants compared to WT. The increase in total polyamines is more distinct in PMT-RNAi roots than in PR50-RNAi roots, which fits well with the data of the phenotypic ripening markers. A similar connection between a suppressed PMT gene and increased polyamines has been reported for Hyoscyamus niger (Geng et al., 2018). The higher concentration of polyamines in roots of the transgenic plants compared to roots of the WT became more distinct from topping to harvest, that is, during the leaf maturing process. Senescence that is induced by topping is developing normal in the WT but in the transgenic plants this senescence is counteracted by the higher concentrations of polyamines, leading to the aberrant phenotype with higher chlorophyll contents and smaller cells in the leaves.
Putrescine is both the precursor of other polyamines and the substrate of PMT, and thus also a precursor for alkaloids (Wang et al. 2000). As both ADC and ODC catalyze putrescine production, we checked the activities of these enzymes in PMT-RNAi plants.
At harvest, ODC activity is significantly higher in both leaves and roots of PMT-RNAi plants compared to the WT; for ADC, the trend is the same in leaves while there is very little activity in roots which is not unexpected, as ADC is mainly present in aerial parts of the plant (Bortolotti et al., 2004). The pattern seen for ODC activity fits well with other data that we collected; polyamines accumulate over time and mostly in roots. It seems illogical that the decrease in nicotine and the concomitant increase in polyamines happens in the roots, while the effects as a retarded senescence are seen in the leaves, however, similar effects have been observed before (Ruiz, Rios, et al., 2006;Sato et al. 2001) and it is known that polyamines are translocated within the plant from the roots to the upper parts and vice versa (Beraud et al. 1992;Caffaro et al. 1993;Rabiti et al. 1989).
The polyamine content in plants is tightly controlled, and previous studies have shown that conjugation of polyamines with biomacromolecules, for example, proteins, membranes, lignin, or hydroxycinnamic acid, is one of the mechanisms plants use to control the intracellular free polyamine content (Bassard et al., 2010;Nölke et al., 2018;Torras-Claveria et al., 2012). Accordingly, PMT-RNAi and PR50-RNAi plants with increased polyamines also show increased polyamine conjugation. There was an increase in conjugated polyamines with age in both leaves and roots in the WT and in the two transgenic lines, but it was in roots that a significant difference between the WT and the transgenic lines could be observed.
Thus, many of the phenotypical changes that we observed in the leaves of PMT-RNAi and the PR50-RNAi plants, for example, that The treatment of plants in the field with DFMO is not readily feasible, but if inhibition of polyamine biosynthesis is needed to restore the product quality of the low-alkaloid varieties, then other treatment approaches could be explored. Oligogalacturonides have been shown to reduce the expression of polyamine biosynthesisrelated enzymes (Falasca et al., 2008), and such a treatment may have the additional benefit of further reducing the nicotine content of the plants (Martinez et al., 2020) and helping the plants to defend themselves against pathogens in the absence of nicotine (Falasca et al., 2008;Ferrari et al., 2013). Based on the preliminary data reported here from greenhouse experiments, additional research is needed to better understand leaf quality of low-alkaloid tobacco lines grown in the field with such treatments.    (Kudithipudi & Hayes, 2018).

| Plant material and growth conditions
Transgenic, along with control seeds, were germinated in pots under greenhouse conditions at 27/23℃ day/night temperature and a 16hr photoperiod (~200 mmol/s m −2 ; λ = 400-700 nm) at 70% relative humidity. Five-week-old tobacco plantlets were transferred to 13 L pots with standard substrate (Einheitserde, Fröndenberg, Germany) and grown in the greenhouse for 4 additional weeks as previously described (Nölke et al., 2018).  Table S1.

| Chlorophyll and glucose measurements
The chlorophyll content was determined by measuring leaf absorbance in the red and infrared regions using a SPAD-502 Plus device (Minolta Camera Co.) as previously described (Nölke et al., 2018) from four randomly selected plants of each line. Measurements

| Determination of ODC and ADC activities
To determine enzymatic activities, 500 mg of tobacco leaf or root tissue collected from four different plants at topping (leaf 19, roots) and at harvest (leaf 24, roots) was ground in 1 ml HEPES extraction buffer (100 mM HEPES, 2 mM dithiothreitol (DTT), 1 mM EDTA, pH 7.5) and 100 mg of polyvinylpyrrolidone was added during grinding.

| Polyamine extraction and analysis
For polyamine analysis, 150 mg of leaf or root material was har- four biological replicates were used. Extraction of free and conjugated polyamines was performed as previously described (Nölke et al., 2018). The dansilation of free and conjugated polyamines was carried out with dansyl chloride as described by Flores and Galston (1982). The dansylated polyamines were measured by LC-MS/MS. All experiments were carried out on a 3200 QTRAP™ mass spectrometer (Sciex) coupled to an HPLC Agilent 1200 system as described before (Nölke et al., 2018).

| Statistical analysis
Significant differences between the genotypes were determined by applying one-way analysis of variance (ANOVA) followed by posthoc Bonferroni test using Excel software (Microsoft). Two-tailed t-tests were applied. A p-value <.05 was considered statistically significant.

ACK N OWLED G M ENT
The authors gratefully acknowledge Ibrahim Al Amehdi (RWTH Aachen University) for taking care of the plants in the greenhouse.

F I G U R E 8
Comparative analysis of free and conjugated polyamine content in leaves and roots of WT, PMT-RNAi, and PR50-RNAi untreated and treated plants with polyamine biosynthesis inhibitor DFMO at harvest. WT, PMT-RNAi, and PR50-RNAi plants were grown in the greenhouse in the absence (WT, PMT-RNAi, and PR50-RNAi) or presence (PMT-RNAi and PR50-RNAi) of polyamine biosynthesis inhibitor DFMO (2 mM). Treatment was performed three times per week from topping to harvest for a period of 4 weeks. Samples were collected from leaf 23 or roots of four biological replicates per genotype or treatment. The ratio between polyamine content from untreated or DFMO-treated PMT-RNAi (1. and 2. panel from left) or PR50-RNAi plants (3. and 4. panel) and WT is given in percent. Grey: No significant difference from the WT (=100%); red frame: significantly (p < .05) higher amount of polyamines compared to WT (>100%), the thickness of the box frames depicts the magnitude of the increase; and blue background: significantly (p < .05) lower amount of polyamines compared to WT (<100%)