Introduction
Lodging is defined as persistent deviation of plant shoots from their vertical orientation due to different external (environmental) factors or internal plant characteristics (Ball et al., Reference Ball, Hanlan and Vandenberg2006; Bourmaud et al., Reference Bourmaud, Gibaud, Lefeuver, Morvan and Baley2015; Gomez et al., Reference Gomez, Muliana, Niklas and Rooney2017). It is dependent on the crop genotype, density and developmental stage, since these variables affect the stem mechanical and structural characteristics (Hall et al., Reference Hall, Sposaro and Chimenti2010). Lodging negatively affects yield, grain yield and quality, and harvest management and leads to economic losses, especially when it occurs during the grain-filling period (Kelbert et al., Reference Kelbert, Spaner, Briggs and King2004; Ball et al., Reference Ball, Hanlan and Vandenberg2006; Kong et al., Reference Kong, Liu, Guo, Yang, Sun, Li, Zhan, Cui, Lin and Zhang2013; Marcelo et al., Reference Marcelo, Tapic and Manangkil2017). Therefore, many research programmes are directed towards an improvement of lodging resistance and identifying plant characteristics that correlate with lodging and could be good indicators of lodging tolerance in early screening.
Stem morpho-anatomy, strength and weight distribution along the shoot exert a pivotal influence on lodging resistance (Hall et al., Reference Hall, Sposaro and Chimenti2010; Bourmaud et al., Reference Bourmaud, Gibaud, Lefeuver, Morvan and Baley2015; Mangieri et al., Reference Mangieri, Mantese and Schurmann2016; Zhang et al., Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016; Gui et al., Reference Gui, Wang, Xiao, Tu, Li, Li, Ji, Wang and Li2018). Stem height is positively correlated with increased lodging in many plant species, and height reduction is thus the main target of their breeding programmes (Zhang et al., Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016; Gomez et al., Reference Gomez, Muliana, Niklas and Rooney2017; Gui et al., Reference Gui, Wang, Xiao, Tu, Li, Li, Ji, Wang and Li2018). Lodging occurs in the basal stem region, which makes lodging susceptibility of the basal stem internodes especially important. Stem strength is determined by its morphology, anatomy and chemical composition. (Banniza et al., Reference Banniza, Hashemi, Warkentin, Vandenberg and Davis2005; Bourmaud et al., Reference Bourmaud, Gibaud, Lefeuver, Morvan and Baley2015). Shorter and thicker basal internodes, as well as greater stem thickness measured without the central cavity, lead to higher lodging resistance (Kelbert et al., Reference Kelbert, Spaner, Briggs and King2004; Beeck et al., Reference Beeck, Wroth and Cowling2006; Mangieri et al., Reference Mangieri, Mantese and Schurmann2016; Zhang et al., Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016; Muhammad et al., Reference Muhammad, Hao, Xue, Alam, Bai, Hu, Sajid, Hu, Samad, Li, Liu, Gong and Wang2020). In red fescue, an increase in internode length was found by Szczepanek et al. (Reference Szczepanek, Baczynski and Graczyk2021) to increase lodging, which decreased with an increase in the lower internode diameter. Conversely, Gomez et al. (Reference Gomez, Muliana, Niklas and Rooney2017) noted that internode strength was negatively correlated with diameter in sorghum, prompting them to posit that increased stem diameter reduces the stresses that are distributed across the stem. In wheat genotypes, stem thickness without central cavity, rather than total stem diameter, was found to be negatively correlated with lodging (Kelbert et al., Reference Kelbert, Spaner, Briggs and King2004; Kong et al., Reference Kong, Liu, Guo, Yang, Sun, Li, Zhan, Cui, Lin and Zhang2013). Yet, Singh and Srivastava (Reference Singh and Srivastava2015) failed to establish significant correlation between field pea stem diameter and lodging, whereas according to Beeck et al. (Reference Beeck, Wroth and Cowling2006) compressed stem thickness proved to be a better predictor of stem strength, compared to stem diameter.
The cell wall chemical composition, and the proportions of cellulose, hemicellulose and lignin in particular, determines the stem mechanical properties to a significant extent (Mengistie and McDonald, Reference Mengistie and McDonald2023). In legume stems, lignified tissues (e.g. xylem vessels, phloem and xylem fibres, sclerenchymatous parenchyma) contribute to the stem strength, since lignin provides strength to the secondary thickened cell walls (Engels and Jung, Reference Engels and Jung2005; Jung and Lamb, Reference Jung and Lamb2006; Zoric et al., Reference Zoric, Mikic, Cupina, Lukovic, Krstic and Antanasovic2014). Epidermal and collenchyma cells have thick, but not lignified primary cell walls, and therefore contribute less to the overall stem stiffness (Jung and Engels, Reference Jung and Engels2002). However, a positive effect of higher lignin content on stem lodging resistance in wheat is not always observed (Banniza et al., Reference Banniza, Hashemi, Warkentin, Vandenberg and Davis2005; Ball et al., Reference Ball, Hanlan and Vandenberg2006; Kong et al., Reference Kong, Liu, Guo, Yang, Sun, Li, Zhan, Cui, Lin and Zhang2013; Muhammad et al., Reference Muhammad, Hao, Xue, Alam, Bai, Hu, Sajid, Hu, Samad, Li, Liu, Gong and Wang2020).
Considering stem anatomy, enhanced lodging resistance is particularly associated with stems characterized by the presence of peripherally situated sclerenchyma tissue and vascular bundles, especially in grasses (Kaack et al., Reference Kaack, Schwarz and Brander2003; Kelbert et al., Reference Kelbert, Spaner, Briggs and King2004; Kong et al., Reference Kong, Liu, Guo, Yang, Sun, Li, Zhan, Cui, Lin and Zhang2013; Zhang et al., Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016; Marcelo et al., Reference Marcelo, Tapic and Manangkil2017; Gui et al., Reference Gui, Wang, Xiao, Tu, Li, Li, Ji, Wang and Li2018; Muhammad et al., Reference Muhammad, Hao, Xue, Alam, Bai, Hu, Sajid, Hu, Samad, Li, Liu, Gong and Wang2020). Anatomical trait-associated factors influencing stem lodging in cereals are cell wall thickness, number and diameter of vascular bundles and sclerenchyma thickness (Mengistie and McDonald, Reference Mengistie and McDonald2023). Moreover, stalk strength in cereal crops is closely associated with stalk geometry, structure and cell wall composition.
Sclerenchyma fibre bundles, their proportion, number, diameter and distribution of individual fibres, reinforce flax stem and contribute to lodging resistance (Bourmaud et al., Reference Bourmaud, Gibaud, Lefeuver, Morvan and Baley2015). A higher proportion of secondary xylem and sclerenchyma has also been reported to hinder sunflower stem lodging (Mangieri et al., Reference Mangieri, Mantese and Schurmann2016). Banniza et al. (Reference Banniza, Hashemi, Warkentin, Vandenberg and Davis2005) revealed a negative correlation between lodging and the proportion of supportive tissues, xylem, lignin, cellulose and fibre content in field pea stems. Ball et al. (Reference Ball, Hanlan and Vandenberg2006) posited that increased lodging associated with greater xylem proportion of lentil stems was probably due to increased stem brittleness. On the other hand, proportion of the pith cavity was not correlated with lodging. Despite the thin-walled nature of parenchyma cells, this tissue plays a crucial role in stabilizing the stem by facilitating energy transfer during bending, thus reducing the risk of lodging and stem breakdown (Kong et al., Reference Kong, Liu, Guo, Yang, Sun, Li, Zhan, Cui, Lin and Zhang2013). As parenchyma takes up most of the pea stem, variation in its proportion, number and size of the cells, rather than the total lignin content, may explain the differences in stem strength among pea genotypes (Beeck et al., Reference Beeck, Wroth and Cowling2006).
Selection of semi-dwarf plant varieties and use of plant growth regulators (PGRs), most of which are growth retardants, can minimize the effects of lodging on crop production (Miziniak et al., Reference Miziniak, Matysiak and Kaczmarek2017). Moddus is very popular PGR characterized by rapid foliar absorption. It is mostly used on grass grain crops, with trinexapac-ethyl (TE) being its main component (Anderson et al., Reference Anderson, Chastain, Garbacik, Silberstein and Young2011; Subedi et al., Reference Subedi, Karimi, Wang, Graf, Mohr, O'Donovan, Brandt and Beres2021). TE belongs to the cyclohexanone group, which stimulates the substrate of dioxygenases and inhibits biosynthesis of gibberellins and cell elongation. In the absence of gibberellins, the growth of internodes is retarded (Subedi et al., Reference Subedi, Karimi, Wang, Graf, Mohr, O'Donovan, Brandt and Beres2021). The effectiveness of growth retardants depends on several factors, weather conditions and precipitation in particular (Harasim et al., Reference Harasim, Wesolowski, Kwiatkowski, Rozylo and Maziarz2016; Miziniak et al., Reference Miziniak, Matysiak and Kaczmarek2017). In grasses, Moddus shortens internodes, which results in lodging reduction without significant change in yield performances, like in barley (Miziniak et al., Reference Miziniak, Matysiak and Kaczmarek2017) and wheat (Harasim et al., Reference Harasim, Wesolowski, Kwiatkowski, Rozylo and Maziarz2016). Again, internode diameters varied between years, but radial stem thickness, without central cavity, significantly increased at the fourth internode. These experiments also demonstrated that plant susceptibility to lodging is mostly dependent on the meteorological conditions and the retardant dose. More recently, Szczepanek et al. (Reference Szczepanek, Baczynski and Graczyk2021) reported third internode shortening and decreased lodging following the TE application on red fescue. However, TE is not commonly used in legume species. Strydhorst et al. (Reference Strydhorst, Yang, Gill and Bowness2019) applied different PGRs to improve field pea standability. PGR treatment had a negligible impact on standability, as plant height was reduced by 8–13% at only 66% of the sites/years, while the effects on seed yield, weight and protein content were inconsistent. Based on these responses, the authors could not recommend the use of growth regulators for the purpose of pea plant height reduction. They further noted that PGRs should be applied with caution, to avoid any potential negative impacts on the plants, as their effects can vary depending on the plant species, growing conditions, as well as application rates and dosage.
Vicia sativa L. and Vicia pannonica Crantz. are annual forage species that are widely used as grain crops or as a green forage (Naydenova and Aleksieva, Reference Naydenova and Aleksieva2014). As they are also affected by the stem lodging problem, in the present study, comparative and correlative analysis of stem morpho-anatomical characteristics and chemical composition of these two Vicia species was performed, before and after the treatment with different concentrations of the exogenous growth regulator TE. The hypotheses were that some of the vetch stem anatomical and morphological characteristics could contribute to lodging resistance and, therefore, could be used as markers for lodging resistance prediction; that higher lodging resistance could be expected from plants having higher proportions of lignin, acid detergent fibre (ADF), neutral detergent fibre (NDF) and tissues composed of thick-walled cells, as well as that the application of PGR (TE) could reduce lodging by reducing stem growth or affecting its anatomy and morphology. Therefore, the aim of this research was to (a) improve the understanding of the mechanisms of lodging in vetches by identifying stem anatomical and morphological components of lodging resistance; (b) determine and quantify the effect of different doses of Moddus growth retardant on Vicia stem morphology, anatomy and chemical composition, and their relationship to lodging resistance; (c) examine if stem mechanical properties and lodging resistance could be improved by the application of TE, without negatively affecting biomass and grain yield; and (d) establish if some of the anatomical or morphological characteristics could be used as markers for lodging resistance prediction.
Materials and methods
The plant species used in the current investigation, the widely cultivated forage crops V. sativa (common vetch) and V. pannonica (Hungarian vetch), were grown in 2020 and 2021 at the experimental field of the Institute of Field and Vegetable Crops at Rimski Šančevi, Northern Serbia (45°20′N, 19°51′E, 84 m a.s.l.). The trial was established as a randomized block design, with three replications. The plot size was 10 m2, and the seed sowing rate was 120 kg/ha, in accordance with the local practice. Sowing dates were 10 October 2020 and 15 October 2021. The following mechanical operations were conducted: stubble cultivation, ploughing to a depth of 25 cm, harrowing, seedbed preparation, sowing and rolling after sowing. Sowing was carried out with an inter-row spacing of 12.5 cm and at a depth of 4–5 cm with grain drill. No inoculation was performed. Sampling dates were 25 May 2020 and 21 May 2021. The soil was classified as slightly carbonated loamic chernozem, specifically Calcic Gleyic Chernozem (CH-cc.gl-lo), according to the WRB (2014). It had a pH value of 7.2 in KCl, with 2.2% of organic matter, 6.3% of CaCO3, 0.2% of nitrogen and contained 22.7 mg/100 g of K2O and P2O5. Bulk density was 1.35 g/cm3. No fertilizers were used.
There were six treatments with different TE concentrations, along with an untreated control. Plots C1, C2 and C3 were treated once with the following amounts of TE (trade name Moddus, Syngenta, 250 g/l of the active ingredient TE): 1.6, 2.4 and 3.2 l/ha, respectively. Plots C4, C5 and C6 received two treatments, with the second treatment applied 2 weeks after the first. The first treatment involved the same TE amounts as the single-treatment plots: 1.6, 2.4 and 3.2 l/ha, respectively. The second treatment applied 1.6 l/ha TE to all three plots.
The BBCH scale (Biologische Bundesanstalt, Bundessortenamt, and CHemical industry scale) is a standardized coding system that describes the developmental stages of plants. It was developed to provide a universal framework for identifying and recording growth stages in various plant species, which is crucial for agricultural practices, research and crop management. The first application was carried out at BBCH 3. According to the UPOV (International Union for the Protection of New Varieties of Plants) Guidelines (International Union for the Protection of New Varieties of Plants, 2013), the plants were at the principal growth stage 3: stem elongation, stage 30 – beginning of stem elongation. The second application took place 2 weeks later, when the internodes were visibly extended (principal growth stage 3: stem elongation, stage 39 – nine or more visibly extended internodes).
Lodging index was determined as a decrease in canopy height (LI = canopy height/plant height), with greater values signifying higher lodging resistance (Mikić et al., Reference Mikić, Ćupina, Rubiales, Mihailović, Šarūnaitė, Fustec, Antanasović, Krstić, Bedoussac, Zorić, Đorđević, Perić and Srebrić2015; Luo et al., Reference Luo, Wu, Fu, Dan, Yuan, Liang, Zhu, Hu and Wu2022).
For morphological analysis, ten randomly selected plants were cut at the full flowering stage, coinciding with the formation of the first visible pods (second half of May). For anatomical analyses, five plants from each group were cut in full bloom and were fixed in 50% ethanol (Ruzin, Reference Ruzin1999). For stereological investigations, cross-sections were made along the stem axis, according to the methodology developed by Kubinova (Reference Kubinova1991), as explained in Zoric et al. (Reference Zoric, Mikic, Cupina, Lukovic, Krstic and Antanasovic2014). Stem cross-sections were cut in positions sampled according to the principle of systematic uniform random sampling, along the stem axis, resulting in five segments per stem. The segments were numbered incrementally, starting from the top (1st) to the bottom (5th) part of the stem. All sections were obtained at −20°C using a cryostat Leica CM 1850 (Leica, Germany), at 40 μm cutting intervals. Light microscopy observations and measurements were performed using the image analysing system Motic Images Plus (Motic, Germany). The proportion of tissues was estimated by point-counting method, using the point grid test system of 391 test points. Approximately 1500–2000 points per stem were counted. Volume densities of tissues (Vv, in %) were calculated using the equation given by Kubinova (Reference Kubinova1993):
where n is the number of examined sections, Pj(x) (j = 1, …, n) – denoted the number of test points hitting the tissue (epidermis, collenchyma, cortex parenchyma, sclerenchyma, phloem, xylem, sclerenchymatous parenchyma, cylinder parenchyma and central cavity) within j-number of sections and Pj(y) (j = 1, …, n) – represents the number of test points hitting the entire stem cross-section area within j-number of sections.
Collected samples were also subjected to chemical analyses to determine the NDF, ADF and acid detergent lignin (ADL) content. These analyses were performed using an Ankom 2000 Fibre Analyser (Ankom Technology Corp., NY, USA) in the Laboratory for Soil and Agroecology at the Institute of Field and Vegetable Crops in Novi Sad.
Data were statistically processed, and the means, standard errors, coefficients of variation and correlation coefficients were calculated using STATISTICA software (TIBCO Software Inc., 2020). The significance of differences in the measured characteristics between treatment and control plants was determined using t-test (P ≤ 0.050). In graphs shown in Figs 1–3, only data for stem characteristics with significant differences compared to controls were presented.
Relative changes in morphological characteristics (%) in relation to control plants in V. sativa (a) and V. pannonica (b) as a results of PGR application (C1-C6, first year and C1-2-C6-2, second year; plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively; plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots). Only statistically significant changes are presented, compared to control plants, according to t-test.
Relative changes in volume density (Vv) of stem tissues (%) in V. sativa (a) and V. pannonica (b) as a results of PGR application (C1-C6, first year and C1-2-C6-2, second year; plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively; plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots). Only statistically significant changes were presented, compared to control plants, according to t-test. epid, epidermis; coll, collenchyma; cor par, cortex parenchyma; scl, sclerenchyma; phl, phloem; xyl + sclpar, xylem with sclerenchymatous parenchyma; cyl par, cylinder parenchyma; cav, central cavity; vasc bun, number of vascular bundles.
Relative changes, compared to control plants, in volume density (Vv) of stem tissues (%) from the middle (segment 3) to the bottom (segment 5) part of the stem in V. sativa (a–c) and V. pannonica (d–f), as a result of PGR application (C1-C6, first year and C1-2-C6-2, second year; plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively; plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots). Only statistically significant changes were presented, compared to control plants, according to t-test. epid, epidermis; coll. collenchyma; cor par, cortex parenchyma; scl, sclerenchyma; phl, phloem; xyl + sclpar, xylem with sclerenchymatous parenchyma; cyl par, cylinder parenchyma; cav, central cavity; vasc bun, number of vascular bundles.
Results
The values obtained for the analysed morphological and anatomical traits of the V. sativa and V. pannonica before and after the treatment are given in Tables 1–4. It is evident that, in the second experimental year, control plants of both Vicia species formed more intensively branched stems, with a significantly greater number of pods and seeds. Stem height and cross-section area did not differ significantly between the years for the V. sativa control plants, whereas V. pannonica stems were significantly taller in the second year and their thickness was significantly less in the third segment (P < 0.05). Significant phenotypic variation in the stem-related traits was recorded before, as well as after the treatment. PGR significantly reduced the stem height of both Vicia species in all applied concentrations (Fig. 1, Tables 1 and 2). Under the effect of PGR, stem and stem cross-section area without central cavity in the middle and bottom part was significantly reduced in V. sativa, especially in the second experimental year, while these changes were not observed in V. pannonica (P < 0.05). PGR application positively affected branching in the first, but not in the second experimental year. When applied in C4 concentration, it had positive effect on the number of pods and seeds and seed weight in V. sativa. However, in V. pannonica, PGR stimulated formation of pods and seeds in the first experimental year, especially when applied in higher concentrations, while having the opposite effect in the second year.
Stem morphological parameters of V. sativa control plants and changes following PGR application (mean ± standard error, coefficient of variation, D.F. = 48)
*Indicates that differences within one species were significant between the years, according to t-test, P < 0.05.
a C1-C6, first year and C1-2-C6-2, second year. Plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively. Plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots.
Stem morphological parameters of V. pannonica control plants and changes following PGR application (mean ± standard error, coefficient of variation, D.F. = 48)
*Indicates that differences within one species were significant between the years, according to t-test, P < 0.05.
a C1-C6, first year and C1-2-C6-2, second year. Plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively. Plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots.
Vv of stem tissues and the number of vascular bundles of V. sativa control plants and changes following PGR application (mean ± standard error, coefficient of variation, D.F. = 48)
*Indicates that differences within one species were significant between the years, according to t-test, P < 0.05.
a Vv – volume density of the tissue (%).
b C1-C6, first year and C1-2-C6-2, second year. Plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively. Plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots.
Vv of stem tissues and the number of vascular bundles of V. pannonica control plants and changes following PGR application (mean ± standard error, coefficient of variation, D.F. = 48)
*Indicates that differences within one species were significant between the years, according to t-test, P < 0.05.
a Vv – volume density of the tissue (%).
b C1-C6, first year and C1-2-C6-2, second year. Plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively. Plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots.
As shown in Fig. 4, the stem of both Vicia species is rounded in cross-section, winged and has primary anatomical structure. Epidermis is present at the surface, whereas primary cortex is composed of cortex parenchyma and peripherally positioned collenchyma. Central cylinder contains vascular bundles arranged in a circle, with sclerenchyma groups adjacent to the phloem and sclerenchymatous parenchyma located at the inner side of the xylem and between the bundles. Cylinder parenchyma cells separate to form a cavity in the central part of the stem. All Vv values exhibited relatively high variability, as indicated by the coefficients of variation reported in Tables 3 and 4.
Cross-sections of middle stem part of (a) V. sativa and (b) V. pannonica control plants.
After PGR application in different doses, rather inconsistent changes in the Vv of stem tissues were recorded, especially in V. sativa (Fig. 2a). Thus, only significant differences in the Vv of stem tissues between control and treated plants, which were confirmed in both experimental years, were considered as important. In V. sativa, in the first year, the Vv of cortex parenchyma significantly increased under C3 and C4 treatments, while in the second year all treatments, except C4, yielded this result (P < 0.05). The Vv of cylinder parenchyma increased under C2–C5 treatments in the first year and under C2 in the second. At the same time, the xylem proportion decreased significantly under treatments C3 and C4 in the first year, and under C5 treatment in the second year (P < 0.05). Proportions of collenchyma were significantly greater, and those of central cavity significantly less in most of the treatments, especially when higher PGR concentrations were applied, but only in the first experimental year (P < 0.05). Other recorded changes in Vv were variable and inconsistent in both years and were therefore not deemed significant. In general, the strongest effect was observed with the PGR application in C3 concentration.
PGR application induced fewer significant changes in the Vv of stem tissues of V. pannonica (Fig. 2b). Proportion of sclerenchyma was significantly reduced in the second experimental year, and that of phloem significantly increased in the first experimental year, compared to control (P < 0.05).
Stem segments closer to the ground are most responsible for lodging, since that is the point at which the stem bends. Accordingly, the effect of PGR on their structure was examined in more detail (Fig. 3, Tables S1–S6). In V. sativa PGR application induced a decrease in the Vv of sclerenchyma and increase in the Vv of cortex parenchyma in some doses on segment 4, and only in the second year on segment 5, compared to control plants. The C3, C5 and C6 concentrations were most effective. However, in most of the cases, the results were not consistent in both study years, so the changes in tissue proportions could not be assigned to the PGR effect only. In V. pannonica the changes in tissue proportions along the stem maturity gradient were not confirmed in both study years. In bottom segments, decreased proportion of sclerenchyma was recorded, under C1–C3 treatments, but only in the second year.
In both examined species lodging index increased with the application of PGR, which pointed to the higher lodging resistance of treated plants (Fig. 5). In V. sativa, the index values were the highest after PGR application in medium concentrations (C2, C5 and C6 in the first, and C2, C4 and C6 in the second year). In V. pannonica, the higher concentrations (C4–C6) were most effective in both years. Lodging index was significantly negatively correlated with stem height (rVS = −0.77, rVP = −0.71) and positively correlated with the number of branches (rVS = 0.31, rVP = 0.16) and seed weight (rVS = 0.46, rVP = 0.20) in both species. In V. sativa, lodging index was positively correlated with the number of pods (r = 0.42) and seeds (r = 0.51), whereas in V pannonica these correlations were negative and low. Unexpectedly, correlations of lodging index with stem cross-section area and stem cross-section area without central cavity were low, and were even negative in V. sativa (rVS = −0.14, −0.14; rVP = 0.16, 0.18).
Lodging index of V. sativa and V. pannonica before and after Moddus application (C-C6 – first year and C-2-C6-2 – second year; C – control plants; plots C1, C2 and C3 treated once with TE: 1.6, 2.4 and 3.2 l/ha, respectively; plots C4, C5 and C6 treated twice – the first treatment the same as previous, the second 1.6 l/ha TE to all three plots).
Considering anatomical characteristics, it seems that increased lodging resistance did not arise from the changes in stem tissue proportions, as even the correlations which appeared statistically significant were low (Figs 6 and 7). These results indicate that the origin of increased lodging resistance may lie among the changes in characteristics that are not of anatomical nature.
Correlations between lodging index and Vv of stem tissues in V. sativa (marked r values are significant at P < 0.050).
Correlations between lodging index and Vv of stem tissues in V. pannonica (marked r values are significant at P < 0.050).
As the highest NDF, ADF and ADL values were obtained for the control plants of both species, despite increasing the lodging resistance, Moddus application did not increase lignification of treated plants (Table 5). The obtained results point to lower lignin, cellulose and hemicellulose proportions in the PGR-treated V. sativa plants. The corresponding values were rather similar across all of the V. pannonica treatments and control plants. The percentage of ash increased in both species with the application of PGR. Increased protein and N content were recorded in V. sativa during the first year, and in V. pannonica during the second year only, compared to control plants.
Chemical composition of Vicia stems under different PGR treatments (%)
Discussion
Understanding the mechanisms of lodging and stem strength improvement should be one of the major strategies in the future vetch breeding programmes aimed at achieving lodging resistance. Since breeding for lodging resistance takes many years, finding alternative ways to enhance lodging resistance in a short time is challenging. One of the methods tested in this work was an application of the growth retardant Moddus on vetches, as this PGR gave good results in cereal crops and grasses (Harasim et al., Reference Harasim, Wesolowski, Kwiatkowski, Rozylo and Maziarz2016; Miziniak et al., Reference Miziniak, Matysiak and Kaczmarek2017; Szczepanek et al., Reference Szczepanek, Baczynski and Graczyk2021). The results varied between the two experimental years, which could be attributed to the environmental influences, as lodging susceptibility may change depending on the year and weather (Ball et al., Reference Ball, Hanlan and Vandenberg2006). Moreover, growth retardant effectiveness is highly dependent on the weather conditions (Harasim et al., Reference Harasim, Wesolowski, Kwiatkowski, Rozylo and Maziarz2016; Miziniak et al., Reference Miziniak, Matysiak and Kaczmarek2017). Under conditions that promote high vegetative development in Vicia, plants become highly susceptible to lodging. Yield loss in lodged crops occurs due to poor grain filling and pod loss. Lodged plants are also more susceptible to stem rotting and diseases, which can affect both forage and seed quality. Lodging at the later stages of crop development can compromise grain yield and quality, which is crucial for seed production. Souza et al. (Reference Souza, da Silva and Albrecht2022) found similar results, noting that TE reduced plant height and increased productivity in legumes. The application of TE has been studied extensively in small grains, with some research also conducted on beans, wheat and rapeseed (Ijaz et al., Reference Ijaz, Mahmood and Honermeier2015; Gomes et al., Reference Gomes, Freiria, Barbosa and Takahashi2018; de Faria et al., Reference de Faria, Silva and Lollato2022). In this context, our research represents pioneering work on the application of TE in vetches.
In both studied species, the lodging index increased upon the application of PGR, indicating an enhanced lodging resistance in the treated plants (Fig. 5). In V. sativa, Moddus doses C2, C5 and C6 were the most effective in the first year, and C2, C4 and C6 in the second year, whereas in V. pannonica higher doses (C4–C6) were the most effective in both years. It was important to discover the changes in which of the examined morpho-anatomical characteristics explained this increased lodging resistance.
In both examined species, in almost all applied doses, PGR significantly reduced stem height, which correlated with a decline in lodging. In V. sativa, a reduction of the stem height from 11.8 to 36.9% (23.7% on average) was achieved, whereas 12.1–40.1% (30.4% on average) decrease was noted in V. pannonica, depending on the applied dose (Fig. 1). These values were greater than those recorded by Strydhorst et al. (Reference Strydhorst, Yang, Gill and Bowness2019) for field pea (8–13%), Miziniak et al. (Reference Miziniak, Matysiak and Kaczmarek2017) for barley (5.6–16.5%) and Harasim et al. (Reference Harasim, Wesolowski, Kwiatkowski, Rozylo and Maziarz2016), but in all these studies lower TE doses were used. The same was true for lentil stems, where according to Ball et al. (Reference Ball, Hanlan and Vandenberg2006) plant height appeared to be the most influential on lodging, whereas stem diameter had an opposite effect. In extant literature stem diameter and stem thickness without central cavity are much more frequently recognized as reliable characteristics for the prediction of lodging resistance, emphasizing their positive correlation (Beeck et al., Reference Beeck, Wroth and Cowling2006; Muhammad et al., Reference Muhammad, Hao, Xue, Alam, Bai, Hu, Sajid, Hu, Samad, Li, Liu, Gong and Wang2020). Surprisingly, in the current study, the stem cross-section area and stem cross-section area without central cavity of V. sativa were significantly reduced following PGR application, and correlated negatively with lodging resistance (Fig. 1). PGR did not affect the stem cross-section area with or without central cavity, of V. pannonica. The increased lodging resistance could not be attributed to increased stem thickness, but to reduced stem height, only.
In findings reported by Ball et al. (Reference Ball, Hanlan and Vandenberg2006) for lentil stems, greater aboveground biomass was found to induce greater lodging. While these authors reported that the number of branches of lentil stems positively correlated with lodging, we recorded the opposite for the Vicia species. Positive correlation of lodging index with the number of branches and seed weight in Vicia species might be the result of proper weight distribution along the shoot, which stabilizes the stem. A lower sowing rate in vetch provides each plant more space, promoting stronger branching. In contrast, a higher sowing rate can reduce branching due to increased competition for resources. Although denser sowing may limit branching, its impact on lodging depends on factors like plant architecture, environmental conditions and specific management practices. In addition, strong, upright stems enable more efficient transport of water, nutrients and assimilates, which consequently leads to a higher grain yield. This assertion is supported by Singh and Srivastava's (Reference Singh and Srivastava2015) finding that decreased seed yield and weight loss in pea plants are correlated with lodging. It is important to note that, in V. sativa, Moddus had a positive effect on the seed weight and the number of seeds and pods when applied in C4 dose, and that increase correlated positively with the lodging index (Fig. 1). In V. pannonica, the effects were contradictory, since the PGR stimulated the formation of pods and seeds in the first experimental year, but inhibited it in the second year, especially when applied in higher doses, and the correlations with lodging index were negative and low. We thus concur with the conclusion reached by Strydhorst et al. (Reference Strydhorst, Yang, Gill and Bowness2019) that PGRs do not work well on the stability improvement of field pea, since the obtained results were inconsistent and differed slightly from those related to the control plants. These observations led Strydhorst et al. to assert that PGRs should not be used in field pea production.
The results obtained in the present study for the effect of PGR on the anatomical characteristics were highly variable both within and across the two study years. As no significant anatomical changes in V. pannonica with control plants were recorded in both experimental years, it is not plausible to attribute these outcomes solely to the application of Moddus, which exerted a more pronounced influence on V. sativa (Fig. 2).
Greater stem stability was expected to be positively correlated with the amount of strength-determining tissues, such as lignified sclerenchyma, xylem or sclerenchymatous parenchyma (Engels and Jung, Reference Engels and Jung2005; Jung and Lamb, Reference Jung and Lamb2006). However, all correlations between lodging index and volume densities of stem tissues were very low, including those few that were statistically significant (Figs 6 and 7). The obtained results did not validate the proposed hypothesis.
The key characteristics of most lodging-resistant stems are peripherally located, well-developed mechanical and vascular tissue and a greater number of vascular bundles (Kaack et al., Reference Kaack, Schwarz and Brander2003; Zhang et al., Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016). Sclerenchyma fibre bundles, their number, distribution and diameter of individual fibres were found by Bourmaud et al. (Reference Bourmaud, Gibaud, Lefeuver, Morvan and Baley2015) to positively affect the lodging resistance of flax stems. According to Mangieri et al. (Reference Mangieri, Mantese and Schurmann2016), well-developed secondary xylem and sclerenchyma decrease lodging of sunflower stems. Likewise, rice stems with better developed mechanical tissue, and higher lignin content, stem thickness without central cavity and proportion of vascular bundles are more resistant to lodging (Zhang et al., Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016; Gui et al., Reference Gui, Wang, Xiao, Tu, Li, Li, Ji, Wang and Li2018). Lodging-tolerant wheat genotypes examined by Kelbert et al. (Reference Kelbert, Spaner, Briggs and King2004) had thicker culms, measured without central cavity, greater number of vascular bundles and thicker sclerenchyma. This trend was not obvious in the vetch species in the present study. However, the proportion of subepidermal collenchyma significantly increased in V. sativa treated with Moddus, but only in the first study year (Fig. 2). The number of vascular bundles was not affected and the proportion of xylem with sclerenchymatic parenchyma even decreased significantly under the C3 and C4 doses in the first year and under C5 in the second year. Therefore, increased lodging resistance, which was evident after PGR application, could not be related to the proportions of mechanical or vascular tissue. The hypothesis about possible positive effect of the presence of tissues composed of thick-walled cells on lodging resistance was not supported.
A thick parenchyma layer increases lodging resistance, since it contributes to the stem stabilization and increases its thickness (Kaack et al., Reference Kaack, Schwarz and Brander2003). Parenchyma can absorb the energy imparted by external forces without mechanical damage to the plant. Moddus induced an increase in the Vv of cortex parenchyma (at C3 and C4 doses in the first year, and under all treatments except C4 in the second year) and cylinder parenchyma (under C2–C5 in the first year and C2 in the second year) in V. sativa (Fig. 2). These results are in line with those reported by Hall et al. (Reference Hall, Sposaro and Chimenti2010) who found that the epidermis and cortex thickness in sunflower stems contributed to higher resistance to failure. The size of the central cavity of hollow stems does matter, since compressed stem thickness was found by Beeck et al. (Reference Beeck, Wroth and Cowling2006) to be more strongly related to stem strength than total stem diameter in field pea. In rice, a smaller inner stem diameter was shown by Zhang et al. (Reference Zhang, Wu, Wu, Ding, Li, Li, Weng, Lu, Tang, Ding and Wang2016) to be helpful in enhancing mechanical strength. However, in the present study, PGR induced a significant decrease in the central cavity size in V. sativa only in the first experimental year (Fig. 2). Synergistic effect of an increased proportion of cortex and cylinder parenchyma and collenchyma, together with a reduced central cavity, might thus be a path to follow towards an increased lodging resistance in this species. While these results were not expected, increased lodging resistance without increased lignifications means that forage nutritive value was maintained in treated plants.
No significant correlations were found between lodging and anatomical traits. However, traits such as proportions of cortex and cylinder parenchyma, collenchyma and central cavity might still be useful in the selection of parent plants for breeding and crosses. The same lack of general association of stem morpho-anatomical characteristics, other than stem height, with lodging was previously reported for wheat genotypes (Kelbert et al., Reference Kelbert, Spaner, Briggs and King2004).
PGR, in all applied concentrations, decreased the amount of lignin, cellulose and hemicellulose in V. sativa, but had a lesser effect on these components in V. pannonica (Table 5). It is evident that, although the extent of lodging decreased after PGR application, this effect was not due to the additional stem lignification and our hypothesis was not supported. According to the meta-analysis conducted by Mengistie and McDonald (Reference Mengistie and McDonald2023), lignin content of the cereal stalks was positively correlated with lodging resistance in 63.9% of the reviewed studies, cellulose content in 53.9% and hemicellulose indicated no significant correlation with lodging in 52.9% of the cases. On the other hand, Muhammad et al. (Reference Muhammad, Hao, Xue, Alam, Bai, Hu, Sajid, Hu, Samad, Li, Liu, Gong and Wang2020) found that lignin content was not associated with the wheat stem breaking force. In lentils, Ball et al. (Reference Ball, Hanlan and Vandenberg2006) noted that lodging increased with increased fibre content (NDF and ADF), whereas in field pea, Banniza et al. (Reference Banniza, Hashemi, Warkentin, Vandenberg and Davis2005) established a negative correlation between fibre, lignin and cellulose content and lodging. Such contradictory data concerning lignin effect on lodging might suggest that the distribution of lignified tissues in the stem is more important for stem strength than total lignin or cellulose content. Actually, lignin may induce two opposite effects: higher amounts of lignin might give the stem additional strength and prevent lodging, or it might contribute to excessive weight of the aboveground part which would increase lodging.
We believe that 2 years of field experiments are sufficient to obtain valid results under real field conditions. However, one limiting factor is the variability in weather conditions, which can affect the timing of TE applications and the lodging index. Despite these challenges, our study simulates production conditions better than controlled laboratory environments. We conducted this study to evaluate the impact of TE on winter vetches, aiming to assess the feasibility and risks of using TE to mitigate lodging. Future research should focus on determining the effects of varying rates and application timings of TE on nodulation, stem structure, including the distribution of lignified tissues, plant height, grain yield and quality, as well as sowing density, and testing different varieties to ensure stable Vicia seed production.
Lodging resistance could be enhanced by selecting vetch species or genotypes with shorter stems. Application of PGR could contribute to the formation of such a phenotype without adversely affecting stem structure or grain yield. In the present study, lodging resistance enhancement observed in the PGR-treated plants did not arise due to stem structural changes, which were minimal and inconsistent, but was mostly attributable to the stem growth reduction. None of the examined morphological or anatomical characteristics, except stem height, could be used as a marker for lodging resistance prediction. Nonetheless, the obtained results encourage the use of Moddus growth regulator for the study of lodging resistance improvement in vetch species.
Conclusion
In both Vicia species, the lodging index increased following the application of Moddus, suggesting improved resistance to lodging in the treated plants. The correlations between lodging index and volume densities of stem tissues were very low and mostly insignificant. None of the investigated morpho-anatomical features, apart from stem height, proved viable for predicting lodging resistance. These findings suggest that employing Moddus to enhance lodging resistance in vetches by lowering stem height is a promising strategy, as it does not seem to adversely affect the stem structure or grain yield.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0021859624000704.
Author contributions
L. Z., Đ. K. and J. L. designed the study. L. Z., D. K. and D. Ž. conducted morphological and anatomical measurements. D. Ž. and Đ. K. conducted chemical analyses. All authors discussed the results. L. Z. and D. Ž. wrote the article.
Funding statement
The authors gratefully acknowledge the financial support of the Ministry of Science, Technological Development and Innovation of the Republic of Serbia (grants No. 451-03-66/2024-03/200125 and 451-03-65/2024-03/200125).
Competing interests
None.
Ethical standards
Not applicable.
Open access
