Evaluation of beneficial and inhibitory effects of nitrate on nodulation and nitrogen fixation in common bean (Phaseolus vulgaris)

The effects of applied nitrate on symbiotic nitrogen fixation in legumes are complex. Both inhibition and promotion of nodulation by nitrate have been observed in a dose‐dependent manner. The objectives of this study were to determine the effects of nitrate at different concentrations on root nodulation in different genotypes in common bean (Phaseolus vulgaris). Six genotypes were inoculated with the same rhizobial strain and grown hydroponically in growth pouches in a growth chamber and exposed to six nitrate concentrations, including 0, 2.5, 5, 10, 15, and 20 mM for 4 weeks. The tested genotypes included three recombinant inbred lines (RILs, 25, 46, and 70) that differed in their responses to nitrogen (based on observations of one field growing season), their parents (Mist and Sanilac—registered varieties), which are different in N‐fixing abilities, and one nonnodulating mutant (R99). Our results showed that small amounts of nitrate (2.5 and 5 mM) promoted nodule formation and increased nodule biomass, compared with plants in the 0 nitrate control treatment. In contrast, nitrate concentrations over 10 mM inhibited nodulation, resulting in reductions in nodule number and nodule biomass. Nodulation was completely inhibited by 15‐mM nitrate in all the genotypes. Regression analyses indicated that 5‐mM nitrate is the optimum concentration for promoting nodulation as measured by the total number of nodules formed, the number of effective nodules formed, and the nodule biomass formed. In contrast, nitrogen fixation was inhibited by all levels of nitrate. No genotypic differences were observed in nodulation among the three RILs and their parental cultivars, but all were significantly different than R99, a nonnodulating mutant.


| INTRODUCTION
Symbiosis between legume plants and rhizobia is of ecological and economic importance as this process produces a large amount of nitrogen (N) that enters both natural and agricultural systems. Symbiotic nitrogen fixation (SNF) is the process of converting atmospheric N 2 into alternative nitrogenous compounds used by the host plant (Loomis & Connor, 1992). SNF in legumes occurs in nodules that are specialized plant organs attached to roots (for review, see Ferguson et al., 2019). Vegetative cells inside nodules house bacteroids that differentiate from free-living rhizobia and synthesize nitrogenase (Meakin et al., 2007). Nitrogenase is the enzyme that reduces dinitrogen to ammonia, a process that requires large quantities of ATP and low partial oxygen pressures (Meakin et al., 2007). Roots release flavonoid molecular signals into the rhizosphere that attract rhizobia to root hair surfaces (Ferguson et al., 2019). The rhizobia invade roots, travel to root cortex cells, and cause them to divide and form nodules.
Bacteria receive nutrients and energy from plants. Small nodules are visible with the naked eye about 10 days after infection in soybean (Ohyama et al., 2011). Under field conditions, small nodules are visible within 2-3 weeks of planting. SNF is initiated when nodules become larger and turn pink or reddish in color. The pink or red color is caused by leghemoglobin (Lb), a nodule-specific high-affinity carrier protein that controls oxygen flow to rhizobia (Meakin et al., 2007). At the pod-filling stage, nodules of annual legumes generally lose their ability to fix N 2 . Factors affecting nodulation performance include weather, legume species, degree of nodulation by effective strains, the supply of mineral N in the soil, and plant density (Loomis & Connor, 1992).
Bacteria do not usually fix N 2 in the presence of mineral N. A high level of combined N usually inhibits SNF, whereas a small amount sometimes promotes nodule development (Ferguson et al., 2019;Streeter, 1988). A small application of combined N (1-2 mM) has been claimed to be needed for maximum growth and nodule formation in legumes (Streeter, 1988). For example, a long-term supply of 1-mM nitrate promotes nodulation in soybean root nodules (Yashima et al., 2005). An N application of 20-30 kg ha −1 , applied as starter application, improved the growth and productivity of field pea (Erman, Ari, Togay, & Cig, 2009) and groundnuts (Sulfab, Mukhtar, Hamad, & Adam, 2011). The above observations likely indicate that the low levels of N that were used in the studies promoted plant health but did not exceed levels that would inhibit SNF (Ferguson et al., 2019).
Nitrate effects on nodule growth are complex and variable; the effects are either beneficial or inhibitory, depending on nitrate concentration, exposure period, and growth medium (Cabeza et al., 2014;Saito et al., 2014). Nitrate in soils limits root infection, nodule development, and nitrogenase activity (Dwivedi et al., 2015). High concentrations of nitrate reduce the binding of rhizobia to root hairs, decrease the number of infection threads, increase the number of aborted infection events, and inhibit Lb synthesis (Bonilla & Bolaños, 2010;Streeter, 1988). Sixteen bean cultivars experienced a reduction of nodule weight and visual nodulation scores when combined N increased from 0 to 3.5 mM, with continuing reductions being observed as N levels increased to 10 mM (Park & Buttery, 1989).
Exposure to 5-mM nitrate for 1 day almost completely depresses the increase of soybean nodule size, due to the cessation of cell expansion in nodules (Fujikake et al., 2003). However, nodule growth is able to recover quickly after nitrate is removed (Fujikake et al., 2003). Nitrate not only inhibits nodule initiation and formation but also depresses functions of existing nodules (Vessey & Waterer, 1992). After exposure to nitrate for several days, soybean nodules lost their activity (Schullerf, Minchinp, & Gresshoff, 1988). Nodule-specific nitrogenase activity, CO 2 evolution, the proportion of [ 14 C]-labeled photosynthate translocated to nodules, respiration in nodules, and the concentration of nodule starch are significantly decreased in soybean plants when exposed to 10-mM nitrate for 48 h (Vessey, Walsh, & Layzell, 1988).
Dry weight per nodule and the rate of acetylene reduction decreased when white clover (Trifolium repens) root nodules were exposed to more than 7-mM nitrate for 2-3 days and no new nodules developed at high concentrations of nitrate (Davidson & Robson, 1986).
Autoregulation of nodulation (AON) is the mechanism that regulates the number of nodules formed in leguminous plants; AONimpaired mutants are partially tolerant to nitrate, and possess a hypernodulating phenotype (Ferguson et al., 2019;Reid, Ferguson, & Gresshoff, 2011). In a study conducted on Medicago truncatula, it was reported that nodule number per plant was reduced, and nodule initiation was inhibited with nitrate concentrations greater than 2.5 mM; however, root hair curling remained unaffected (van Noorden et al., 2016). The inhibitory effects of nitrate on nodulation involve modifications of flavonoid and defense metabolism, as well as changes to redox (van Noorden et al., 2016). Nitrate treatment (2 mM) induced GmN1C1, a candidate CLE peptide-encoding gene, which regulated nodule numbers; nodulation was inhibited in the roots of transgenic soybean plants through ectopic overexpression of the CLE peptideencoding gene (Reid et al., 2011). ACC SYNTHASE 10 (ACS10), an ethylene biosynthesis gene, is responsible for nitrate inhibition of nodulation in the leguminous plant, M. truncatula (van Zeijl et al., 2018).
Common bean (Phaseolus vulgaris) is the most important grain legume for human consumption. Its center of origin is believed to be Mesoamerica (Bitocchi et al., 2012). Common bean has generally been considered to be a weak N 2 fixer compared with other legumes (Heilig, Beaver, Wright, Song, & Kelly, 2017). Previous research showed that approximately 75% of total N in faba bean, 62-94% of N in soybean, groundnut, pea, and lentil, and 54-58% of N in cowpea, chickpea, and pigeon pea was derived from SNF, whereas only 39% of N in common bean was derived from SNF (Dwivedi et al., 2015).
However, there is significant genetic variability for SNF among common bean genotypes (Farid & Navabi, 2015;Kamfwa, Cichy, & Kelly, 2015), indicating that it is possible to improve SNF through breeding efforts. Early flowering bean genotypes were generally inferior in their SNF ability (Chaverra & Graham, 1992). Either promoting earlier nodulation or delaying nodule senescence can improve overall SNF (Chaverra & Graham, 1992). Common bean varieties with a longer vegetative duration generally have greater SNF ability (Farid, 2015). Climbing-type beans are superior in SNF compared with bush-type beans (Rennie & Kemp, 1983). Nodule number is positively correlated with N fixed in common bean (Pereira, Miranda, Attewell, Kmiecik, & Bliss, 1993). Park and Buttery (1989) identified genotypes with superior nodulation and N 2 fixation characteristics that they believed would be useful for improving the nitrate-tolerant nodulating characteristics of beans. Farid, Earl, and Navabi (2016) and Farid, Earl, Pauls, and Navabi (2017) compared yields in SNF-dependent versus N-fertilizerdependent environments for 140 F 4 -derived F 5 recombinant inbred lines (RILs), developed from a cross between a low SNF bean genotype "Sanilac" and a high SNF bean genotype "Mist" (Farid & Navabi, 2015). The Nitrogen Management Yield Differential Indices (YDIs) that they calculated from the yields in the SNF-dependent versus N-fertilizer-dependent environments identified lines that were as productive in environments where their nitrogen requirements were met by SNF as in the N-fertilizer-dependent environments, including Mist (Farid et al., 2017). However, other lines in the population were less productive in SNF-dependent versus N-fertilizer-dependent environments, including Sanilac.
The current study utilized lines with contrasting YDIs from the previous study to determine the effects of nitrate on SNF, including the determination of critical thresholds of the beneficial and inhibitory effects of nitrate on SNF-related traits. In addition, the study examined relationships among nodule development and plant function traits.
A growth chamber experiment was conducted using a split-plot and a completely randomized design (CRD) arrangement with three replications, with nitrate concentration as the main-plot factor and genotype as the subplot factor. Seeds of the six genotypes were surface sterilized using 1% sodium hypochlorite for 3 min, rinsed five times with sterile distilled water, and germinated on wetted filter paper in petri dishes at room temperature, in the dark. After 2 days, germinated seeds were treated with a commercial peat-based inocu-  et al. (2010). Each growth pouch was considered to be one replication.
The plants within each whole-plot factor (nitrate concentration) of 18 plants (six genotypes × three replications) were randomly assigned.
A second inoculation (4 × 10 5 Rhizobium cells per seedling) was made by applying 2.5 mL of the above inoculant broth to the root region of each plant using a syringe 1 week after the first inoculation to ensure a sufficient number of rhizobia for symbiosis. The moisture levels in growth pouches were monitored daily and nutrient solutions were added when necessary to avoid drought stress. The plants were grown in a growth chamber at 25 C/18 C in light/dark conditions, respectively, with a 16/8 h photoperiod, at a light intensity of 400 mmol m −2 s −1 flux of photosynthetically active radiation, and a relative humidity of 70%.
Two, 3, and 4 weeks after the first inoculation with rhizobia, chlorophyll content of the first fully expanded leaf on the main stem was AB), following the protocol described by Shearer and Kohl (1993).
R99 was used a nonnodulating reference plant to estimate percent of nitrogen derived from the atmosphere (%Ndfa) through the natural 15 N abundance method (Buttery & Park, 1993

| RESULTS
Plants became more robust with greener and larger leaves when nitrate concentrations increased from 0 to 20 mM after 2-3 weeks of planting (Figure 1a-f). At zero or low nitrate (2.5 mM) levels, genotypes differed in leaf greenness (chlorophyll content). In particular, R99 had abnormal yellow leaves, whereas the other five genotypes (Mist, Sanilac, RIL25, RIL46, and RIL70) that were able to fix nitrogen had greener leaves. Plants of these genotypes developing in relatively low nitrate (2.5 and 5 mM) solutions produced larger and more pink root nodules compared with those in nitrate free or at high nitrate (more than or equal to 10 mM) concentrations (Figure 1g versus Figure 1h; Table 1).
The main effect of nitrate concentration was significant on all traits under investigation, except for leaf chlorophyll content after 4 weeks of rhizobia inoculation and the number of white nodules.
After 2 and 3 weeks of rhizobia inoculation, leaf chlorophyll content increased with an increase in nitrate concentration (   Note: Means with a common letter within each column and each main effect did not differ at P < 0.05. Abbreviations: δ 13 C, carbon isotope discrimination; %Ndfa, percent of nitrogen derived from atmosphere; ns, non-significant; SPAD, Soil-Plant-Analysis-Development chlorophyll meter; wt, weight.

| Inhibitory or beneficial effects of nitrate application
Both inhibitory and beneficial effects of additional nitrate application were observed in the present study, depending on nitrate concentration. Low levels of nitrate stimulated nodule growth, due to an increase in plant vigor (Ferguson et al., 2019;Streeter, 1988). Beneficial effects of nitrate were also observed in certain legumes between the pod-filling stage and maturity, as the demand for C and N is relatively high at that stage (Becana & Sprent, 1987). Our results showed that low levels of nitrate (2.5 and 5 mM) increased the number of effective nodules and nodule biomass, and nitrate concentrations greater than 10 mM and above inhibited nodulation.
Interestingly, no beneficial effect of nitrate, even at low concentrations, was observed for the amount of nitrogen fixed by the 4-weekold plants. These results suggest that the elaboration of the nodules needed for nitrogen fixation and the biological and biochemical activity that carry out that function are separately and differently affected by nitrate. Previous research showed that nitrate (more than 2-3 mM) inhibited nitrate reductase activity, decreased the content of Lb and soluble nodule protein, and accelerated the senescence of nodules in soybean (Becana & Sprent, 1987). Using RNAseq, researchers found that all genes related to Lb were down-regulated in M. truncatula when exposed to nitrate continuously (Cabeza et al., 2014). Inhibitory effects of nitrate on nodulation were also associated with cellular iron allocation and mitochondrial ATP synthesis. Genes related to nodule senescence were differentially expressed between control and nitrate-treated nodules (Cabeza et al., 2014).
Only long-term exposure (4 weeks) to nitrate effects was evaluated in the present study. Previous studies showed that short-term exposure to nitrate had a reversible effect on nodule activity, whereas SNF ability was irreversibly lost when exposed to nitrate for a long period of time (Becana & Sprent, 1987). How the irreversible effects were related to carbohydrate deprivation or the accumulation of NO 2 − was not clear in the 1980s (Becana & Sprent, 1987). More energy is needed to fix N 2 compared with the utilization of NO 3 − (Ferguson et al., 2019;Sprent & Raven, 1985). Therefore, in the presence of nitrate, plants are able to detect levels of nitrate and adjust SNF accordingly.

| Genotypic effect on SNF traits
Common bean has been noted in several studies to be a weak N 2 fixer in comparison with other leguminous plants; however, genotypic effects (genetic variability) play a large role in common beans ability to fix nitrogen as several P. vulgaris genotypes have been reported to have higher SNF capability when compared with other genotypes. (Farid & Navabi, 2015;Kamfwa et al., 2015;Farid et al., 2016;Farid et al., 2017;Wilker et al., 2019). It is possible that, compared with other legumes, common bean can take better advantage of low N quantities present in the soil under field conditions. Therefore, we recommend that more work is needed to be done to explore the underlying reason why common bean is considered to be a weak N 2 fixer in comparison with other legumes.
Plant genotype is a key factor determining the efficiency of SNF in legumes, such as in field pea (Bourion et al., 2010) (Farid et al., 2016;Farid et al., 2017). Perhaps, there are some additional components in the field, that condition the genotype effect, such as native rhizobia that are missing in the laboratory experiment.
However, the laboratory method for studying the nitrate effect on nitrogen fixation within genotypes is reproducible and very convenient for studying different developmental times and nitrate concentration effects on nodule formation and nitrogen fixation processes within genotypes.

| Relationship among SNF-related traits
Our results showed that the amount of shoot N from SNF was most closely correlated with %Ndfa among the traits measured,  (Adams, Turnbull, Sprent, & Buchmann, 2016). Only a weak relationship between δ 13 C and %Ndfa, with the correlation coefficient of 0.239, was observed in our study, because plants in the current study were not exposed to drought stress. According to Knight, Verhees, Van Kessel, and Slinkard (1993), the negative correlation between δ 13 C and %Ndfa is only observed under drought stress conditions.
Plants with the 0-mM nitrate application had lower δ 13 C discrimination compared with plants receiving 2.5-to 20-mM nitrate, T A B L E 2 Correlation matrix between symbiotic nitrogen fixation (SNF)-related traits in common bean indicating that plants without additional nitrate supply had greater WUE, most likely because these plants were generally less robust with smaller leaf area and thus experienced less water loss from evaporation. The nonnodulating genotype, R99, also had lower δ 13 C value (greater WUE) compared with other N-fixing genotypes that had N availability from both SNF and nitrate application and thus greater leaf area and more evapotranspiration than R99.

| CONCLUSION
A small amount of nitrate (2.5 and 5 mM) promoted nodulation in common bean, by increasing the number of total nodules and effective nodules and nodule biomass. On the contrary, a high amount of nitrate (greater than 10 mM) inhibited nodulation. No significant difference of nitrate tolerance was observed among three RILs and their parental lines. In the future, research that is directed to uncovering the intricate mechanisms underlying beneficial and inhibitory effects of nitrate on nodulation is recommended.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interest.

ETHICAL STATEMENT
This article does not contain any studies with human or animal subjects.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request.