Nodulation does not alter root hydraulic conductance in pea Authors: Jiacan Sun, Timothy J Brodribb, Ibrahim Bourbia, Eloise Foo Published: 2026 ABSTRACT AbstractRoot hydraulic conductance (K r) normalised by root dry mass (including nodules) under hydrated and drought conditions in Pisum sativum wild‐type (Frisson, Nod+), non‐nodulator (sym19, Nod−), and supernodulator (Psrdn1, Nod++).Keywords: drought, nodulation, non‐nodulating mutant, Pisum sativum L., root hydraulic conductance, root water uptake, supernodulating mutant Disclaimer The New Phytologist Foundation remains neutral with regard to jurisdictional claims in maps and in any institutional affiliations. Nodulation is the symbiosis between legume roots and nitrogen‐fixing bacteria, which contributes substantially to biological nitrogen inputs in agricultural ecosystems (Herridge et al., 2008). In legumes, nodule initiation occurs in a susceptible zone of the root some distance from the elongating root zone or root tip (Stougaard, 2000). Nodulation is important for plant nitrogen acquisition and overall growth, but the potential impact of nodulation on root water relations, specifically root water uptake, has received little attention. Nodulation requires the formation of root nodules, which transforms a portion of the absorptive root system into nodular tissue. This therefore could reduce the surface area available for water uptake, potentially altering the whole‐root water uptake efficiency. Root hydraulic conductance (K r) serves as an integrated measure of the hydraulic efficiency of the root system and reflects the ease with which water moves from root surface to the stem (Bourbia et al., 2021). K r integrates both radial and axial hydraulic pathways and is determined by the morphological, anatomical, and molecular properties of roots. Previous work has shown that K r does not increase linearly with root length (Baca Cabrera et al., 2024), indicating that not all parts of the root system contribute equally to water uptake. Furthermore, previous studies report that nonsuberised regions immediately behind the root meristem have greater K r than mature, suberised root regions (Zwieniecki et al., 2002; Gambetta et al., 2013). Given that nodules occupy only part of the root system and typically develop in mature root zones, variation in nodule density might therefore be expected to have only a limited impact on whole‐root K r. Previous studies have reported contrasting effects of nodulation on plant water relations, showing that compared with a soybean genotype that could not form nodules, a supernodulating soybean plant displayed a lower K r when expressed per unit leaf area, but not when K r was normalised by root dry mass (g). This suggests that supernodulation primarily alters whole‐plant hydraulic balance (root supply relative to canopy demand) through changes in biomass allocation and root system size, rather than reducing hydraulic conductance per unit root tissue. Also, a recent study reported reduced daily root water uptake in supernodulating Lotus japonicus mutants compared with wild‐type (WT) plants (Kawade et al., 2024). Nevertheless, it remains unknown whether the absence or excess of nodulation is associated with differences in K r under both well‐watered and drought conditions. To examine whether contrasting nodulation phenotypes are associated with differences in K r we tested whether variation in nodule density affects K r in Pisum sativum (pea) both under well‐watered and soil‐drying conditions. We compared three pea genotypes with contrasting nodulation phenotypes: the WT cultivar Frisson (Nod+), the non‐nodulating mutant sym19 (Nod−), and the supernodulating mutant Psrdn1 (Nod++). This experimental design enabled us to examine whether K r varies systematically across plants with absent (Nod−), normal (Nod+), or excessive nodulation (Nod++). Our experimental conditions aimed to achieve comparable plant size and nitrogen status in the different pea genotypes. The WT pea Frisson (Nod+) and the supernodulating mutant Psrdn1 (Nod++) form nodules when supplied with no nitrogen (Fig. 1a). The non‐nodulating mutant sym19 (Nod−) was supplied with 10 mM nitrogen to compensate for the absence of symbiotic nitrogen fixation and to achieve biomass comparable to the WT Frisson (Nod+). However, as the Psrdn1 (Nod++) develops an excessive number of nodules, a large proportion of assimilated carbon is allocated to nodule formation rather than vegetative growth, resulting in significantly lower biomass in the Psrdn1 (Nod++) line compared with Frisson (Fig. 1b) (Kawade et al., 2024). Although these three pea genotypes show significant differences in nodule density (c. 2910 and 14 999 nodules g−1 root dry mass in Frisson and Psrdn1, respectively, and no nodules in sym19) and projected leaf area (m2), the shoot:root ratio is similar across genotypes (Fig. 1a,c,d). However, total biomass still differed in genotypes, particularly between Frisson (nod+) and Psrdn1 (nod++) (Fig. 1b). Thus, while the nitrogen treatments allowed comparison among viable plants without severe nitrogen starvation, differences in plant size remained across genotypes. Root hydraulic conductance (K r) normalised by root dry mass (including nodules) under hydrated and drought conditions in Pisum sativum wild‐type (Frisson, Nod+), non‐nodulator (sym19, Nod−), and supernodulator (Psrdn1, Nod++). (a) Nodule density (nodule number per g root mass). (b) Biomass of nodule dry weight (g), root dry weight excluding nodules (g), and shoot dry weight (g) for three pea genotypes. (c) Ratio of shoot to root (including nodules). (d) Projected leaf area (m2). (e) Box plot showing K r normalised by root dry mass (including nodules) under hydrated conditions (−Ψstem = 0 MPa). The centre line represents the median; box limits indicate the upper and lower quartiles; whiskers extend to 1.5× the interquartile range; points represent individual biological replicates, with points beyond the whiskers considered outliers. (f) Scatter plot showing K r normalised by root mass (including nodules) under drought conditions. The solid, dotted, and dashed lines represent Nod+, Nod++, and Nod−, respectively, lines indicate fitted linear regressions and shaded areas indicate 95% confidence intervals. For (a–f) data points represent individual replicate samples (n = 10–12), and error bars indicate ± SE. Asterisks indicate significant differences between genotypes (t‐test, **, P < 0.01; *, P < 0.05). Plants with absent (Nod−), normal (Nod+), or excessive nodulation (Nod++) all exhibited similar K r (Fig. 1, Supporting Information Fig. S1). Under hydrated conditions, K r did not differ significantly among genotypes regardless of whether values were normalised by root dry mass including nodule mass (Fig. 1e) or by projected leaf area (Fig. S1a). During soil drying, K r was observed to decline in a linear fashion, with K r approaching zero at a −Ψstem (MPa) between 1.35 and 1.7 MPa across genotypes regardless of nodule density. Model comparison using delta Akaike Information Criterion (ΔAICc) confirmed that models excluding genotype provided the best fit for these hydraulic decline curves (Table S1). Furthermore, the proportion of variance explained by each predictor in the fitted models was quantified using ANOVA (Table S2). Neither the slope nor the intercept of this relationship differed significantly among genotypes, regardless of whether K r was normalised by root dry mass or projected leaf area (Figs 1f, S1b). Taken together, these results indicate that, despite the likely indirect effects of the mutations on shoot and root size, as well as the effects of the contrasting nitrogen regimes, K r did not differ in three pea genotypes. Thus, under our experimental conditions, neither the absence nor the excess of nodulation, nor the traits associated with these genotypes, resulted in a detectable change in K r when normalised by projected leaf area or root dry mass. This suggests either that not all parts of the root system contribute equally to water uptake, such that nodulated regions have only a limited influence on K r, or that changes in root architecture or local hydraulic properties compensate for any reduction in hydraulic function associated with nodulation. These alternative explanations cannot be resolved from the present data as K r was measured as an integrated property of the entire root system. Previous work in soybean reported reduced water transport in a supernodulating mutant compared with a non‐nodulating soybean mutant (Caroline Silva Lopes et al., 2019). However, several aspects of the experimental design differed from this study and may partly explain the contrasting outcomes. In the soybean study, K r was measured exclusively using pressure‐based measurements that required destructive sampling, whereas here K r was assessed using noninvasive approaches that allowed continuous measurements under both well‐watered and soil drying conditions. In addition, the non‐nodulating plants in the soybean study would have been under nitrogen starvation, while supernodulating mutants could acquire nitrogen from nodules. These contrasting nitrogen conditions may have altered plant growth and root system development, potentially generating differences in hydraulic traits between genotypes that were not directly attributable to nodulation. Another recent related study reported reduced daily water uptake and shoot water content in supernodulating L. japonicus mutants, suggesting a trade‐off between excessive nodulation and whole‐plant performance (Kawade et al., 2024). The difference between that study and the present results reflects differences in how water relations were evaluated. Daily water loss normalised by root biomass was estimated in their study, whereas we calculated K r by accounting for the soil‐to‐stem water potential gradient. Therefore, excessive nodulation may reduce actual daily water uptake or alter whole‐plant water relations. We did not assess the effects of overall biomass, root system size, or carbon allocation, on normalised K r, focusing rather on K r normalised by the projected leaf area or root dry mass. Although this metric assesses tissue‐level transport efficiency, it does not rule out variation due to plant size. Our findings indicate that increased nodulation does not impose a detectable penalty on K r normalised by projected leaf area or root dry mass and that mature root zones bearing nodules may not contribute significantly to root water uptake. This suggests that symbiotic nitrogen fixation and root hydraulic function are probably functionally separated within the root system. This may reduce potential competition for resources between symbiotic nitrogen fixation and water uptake, allowing legumes to sustain metabolically demanding nodules without compromising water transport efficiency. Our results also suggest that the carbon costs of nodulation are mainly associated with investment in nodule formation and maintenance rather than changes in root traits influencing water uptake. Our study has implications for crop improvement, suggesting that increasing nodulation may be achievable without compromising K r in pea. For example, increasing nodulation through manipulation of autoregulation of nodulation pathway, such as CLAVATA3/Embryo Surrounding Region‐Related (CLE) signalling (Zhong et al., 2024), may not necessarily compromise hydraulic function. Materials and Methods Plant material and growth conditions Three genotypes of Pisum sativum L. (pea) were used in this experiment: WT Frisson (normal nodulation, Nod+), corresponding non‐nodulator mutant sym19 (non‐nodulating, Nod−), and supernodulating mutant P. sativum root determined nodulation 1 (Psrdn1) (Nod++). WT Frisson formed nodules in the presence of rhizobia. The sym19 line is mutated in SYMRK/DMI2 gene, which encodes a leucine‐rich repeat receptor‐like kinase critical for the perception of nodulation factors (Paszkowski, 2006). This receptor is essential for activating the signalling pathway required to establish nodulation, and sym19 is unable to form nodules even in the presence of rhizobia (Geurts & Bisseling, 2002; Bersoult et al., 2005). The Psrdn1 mutant line was derived from the pea WT cultivar Frisson (Duc & Messager, 1989) and shows impaired autoregulation of nodulation signalling due to a mutation in PsRDN1, formerly known as nod3 (Scott et al., 2025). PsRDN1 encodes the hydroxyproline O‐arabinosyltransferase RDN1, which is required for the post‐translational modification of CLE peptides involved in feedback signalling (Schnabel et al., 2023). Plants were grown in 0.5‐l punnet pots containing steam‐sterilised premium potting mix (HALS Horticulture and Landscape Supplies, Australia), which does not contain controlled‐release fertiliser. The substrate was steam‐sterilised at c. 80°C for 1 h before use. Pea plants were inoculated 7 d after planting with Rhizobium leguminosarum bv viciae (RLV248; 20 ml per punnet pot). Bacteria were cultured for 3 d in yeast–mannitol broth at 25°C with shaking at 125 rpm. The culture was then diluted to 10% with tap water and applied to the pots (Foo et al., 2013). Frisson, Psrdn1 line, and sym19 line were supplied weekly with a modified Long Ashton nutrient solution 5 mM NaH2PO4 as the phosphorus source, applied once per week (Hewitt, 1965). WT (Frisson, Nod+) and supernodulator (Psrdn1, Nod++) received a nitrogen‐free solution (0 mM), whereas non‐nodulator (sym19, Nod−) was provided with 4 mM KNO3 and 3 mM Ca (NO3)2·4H2O (equivalent to 10 mM total nitrogen). The nitrogen treatments were designed to match plant size between three genotypes by compensating lack of nitrogen fixation in sym19 mutant. Plants were grown in cabinet (light intensity c. 3.7 μmol m−2 s−1 at pot height) at 25°C : 22°C day : night with an 18‐h photoperiod. Pea was used for the hydraulics experiments 3 wk after inoculation. After measurements of K r (mmol m−2 MPa−1 s−1; mmol g−1 MPa−1 s−1) (outlined below), roots were harvested from each plant and nodules were counted. Leaf area of each replicate was scanned and quantified. Shoots were collected and dried to determine shoot dry mass. Nodules were removed from the roots and each dried separately to determine dry mass of root and nodules. For dry mass determination, plant tissues were dried in an oven at 70°C until constant weight was reached, which typically required c. 3 d. Nodule number was normalised to root dry mass. Measurements of K r under well‐hydrated conditions Under well‐hydrated conditions (soils at field capacity), K r was determined using the evaporative flux method under steady state (Eqn 1) (Tsuda & Tyree, 2000; Sun et al., 2026): Briefly, average canopy transpiration (E) was measured gravimetrically by weighing each plant using a balance (VIBRA ALE‐6202, 6200 G × 0.01 G) at 30‐min intervals during midday, when E and the stem water potential (Ψstem, MPa) were at steady state in three genotypes. Each genotype included 10–12 biological replicates. At the end of the experiment, the projected whole‐plant leaf area (m2) and root dry mass (g) were measured for each replicate and used to normalise E. K r was normalised by projected leaf area to express water transport relative to shoot transpirational demand, and by root dry mass to express water transport relative to belowground biomass investment. Projected leaf area was quantified from scanned images of the leaves using the imagej software. K r normalised by projected leaf area was used to express root water transport capacity relative to potential shoot transpirational demand, whereas K r normalised by root dry mass was used to express root water transport capacity relative to belowground biomass investment. These two normalisations provide complementary interpretations and allowed us to assess whether contrasting nodulation phenotypes affected K r when expressed relative to either shoot demand or root biomass. Immediately after E measurements, Ψstem was determined using a pressure chamber. For Ψstem measurements, leaves were covered with moist paper towels and aluminium foil for at least 2 h before E measurements to maintain a nontranspiring state and allow leaf and stem water potentials to equilibrate. The soil was maintained near field capacity rather than saturation, so the soil water potential (Ψsoil, MPa) was assumed to be close to zero but slightly negative for the calculation of K r. Measurements of K r under soil drying conditions Relaxation kinetics of Ψstem were analysed under non‐steady‐state conditions to assess the response of K r to progressive soil drying. This approach minimised soil hydraulic resistance so that K r remained the primary limitation to rehydration. Rehydration experiments were performed on intact plants that had been dehydrated to varying levels of water stress (Brodribb & Holbrook, 2004; Bourbia et al., 2021; Sun et al., 2026). Each genotype included 10–12 biological replicates. Water deficits were imposed by withholding watering for varying durations while Ψstem was continuously monitored with optical dendrometers. Optical dendrometers were attached to pea tendrils to continuously record changes in tissue width, which served as a proxy for Ψstem dynamics. A consistent linear relationship between Ψstem and tendril width has been reported previously (Sun et al., 2026). For each plant, measurements of tendril width obtained with the optical dendrometer were calibrated against Ψstem determined at regular intervals on covered leaves using a Scholander pressure chamber (PMS Instruments, Albany, OR, USA) as soil drying progressed. Once plants had reached the desired level of water stress, they were kept in the dark under high humidity (> 80%) for at least 2–3 h before rewatering to promote stomatal closure and allow Ψstem and Ψsoil to equilibrate. Following this equilibration period in the dark, Ψsoil was assumed to be equal to Ψstem before rewatering. Following rewatering, plants were kept under the same conditions until Ψstem recovered to values close to 0. K r was derived from the kinetics of stem water potential recovery during plant rehydration. The plant was treated as a capacitor‐resistor system, in which water potential increases towards zero as stored water is released. An exponential curve was fitted to the initial 20 min of Ψstem recovery after rewatering, and K r was calculated using Eqn 2 (Brodribb & Holbrook, 2004; Bourbia et al., 2021): where Ψ0 represents the stem water potential before rehydration, Ψfinal is the water potential measured after 20 min of rehydration, and C plant represents whole‐plant capacitance (mmol m−2 MPa−1; mmol g−1 MPa−1). Plant capacitance measurement (C plant) Whole‐plant capacitance (C plant) was required to calculate K r during soil drying. Ψstem and whole‐plant mass were measured simultaneously during a dry‐down from c. −0.3 to −1.75 MPa (Bourbia et al., 2021; Sun et al., 2026). After the rehydration experiment, plants were removed from the pots and their roots were carefully washed to remove residual soil. Excess water was removed from the roots using absorbent paper, and plants were then placed on a balance (VIBRA ALE‐6202, 6200 g × 0.01 g). To monitor Ψstem during soil drying, an optical dendrometer was attached to a pea tendril. Plant mass (g) and Ψstem were measured simultaneously every 10 min as the plants progressively dried. Plants were kept in darkness and allowed to dehydrate until Ψstem declined to c. −2 MPa, which took c. 24 h. For each nodulation phenotype, C plant was determined as the absolute value of the slope of the linear relationship between whole‐plant mass and Ψstem during dehydration. Genotype‐specific C plant values were then used in the corresponding K r calculations. C plant was subsequently normalised by whole‐plant projected leaf area or root dry mass, as appropriate. Sample sizes for C plant normalised by projected leaf area were 11 Nod+, 15 Nod−, and 12 Nod++ plants, whereas those for C plant normalised by root dry mass were 4 Nod+, 7 Nod−, and 4 Nod++ plants. Statistical analysis To test whether genotype influenced the relationship between K r and −Ψstem during soil drying, model selection was performed using an information‐theoretic approach based on the (ΔAICc (Symonds & Moussalli, 2011)). Briefly, linear models with and without genotype effects on the slope and intercept were compared using the ΔAICc. Candidate models were generated using the dredge function from the MuMIn library (Bartoń, 2026) and ranked according to ΔAICc corrected for small sample sizes. The global model was specified as: where genotype represents different pea genotypes (WT pea cultivar Frisson/non‐nodulating pea mutant sym19/supernodulating pea mutant Psrdn1). For each normalisation method in pea, model support was evaluated using AICc, with the model showing the lowest AICc value identified as the best‐supported model and models with ΔAICc ≤ 2 considered to have comparable support (Table S1) (Symonds & Moussalli, 2011). Models that did not include genotype (Models 1 and 3) best describe the data (Table S1). To quantify the relative contribution of each model term, we calculated the percentage of variance explained from the ANOVA sum of squares for each fitted model (Table S2). The percentage of variance explained by each model term was calculated from the ANOVA sum of squares (Table S2). −Ψstem explained most of the variation in both models, accounting for 62.8% in Model 1 and 61.7% in Model 3, while residual variation accounted for 37.2% and 38.3%, respectively. The normality of the K r dataset was tested for each genotype under hydrated conditions. As the data followed a normal distribution, differences in K r among the three genotypes were tested using independent t‐tests. Similarly, C plant and physiological parameters, including projected leaf area, root dry mass, and ratio of shoot‐to‐root, nodule numbers per root gram, were also compared using t‐tests. All statistical analyses were performed in R (R Core Team, 2026). Competing interests None declared. Author contributions EF, TB, IB and JS conceived and designed this study. EF and TB managed this project. JS performed experiments and data analysis. JS wrote the manuscript with contributions from EF, TB and IB. Supporting information Dataset S1 Raw data of stem water potential (Ψstem) and root hydraulic conductance (K r) under well‐watered and soil drying conditions. Dataset S2 Physiological indices across replicates for the three genotypes of Pisum sativum . Dataset S3 Plant capacitance across the three genotypes of Pisum sativum . Fig. S1 Root hydraulic conductance (K r) normalised by projected leaf area under hydrated and drought conditions in Pisum sativum wild‐type (Frisson, Nod+), non‐nodulator (sym19, Nod‐), and supernodulator (Psrdn1, Nod++). Table S1 Model selection results for root hydraulic conductance (K r) normalised by leaf area and root dry mass (including nodule dry mass) in relation to stem water potential (−Ψstem) across different genotypes of Pisum sativum . Table S2 ANOVA summary table for the model explaining variation in root hydraulic conductance (K r). Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Acknowledgements EF, TB, IB and JS acknowledge support from the Australian Research Council including the Centre of Excellence for Plant Success (grant no.: CE200100015). JS acknowledges support from the Tasmanian Graduate Research Scholarship. Open access publishing facilitated by University of Tasmania, as part of the Wiley ‐ University of Tasmania agreement via the Council of Australasian University Librarians. Contributor Information Ibrahim Bourbia, Email: bourbia.ibrahim@utas.edu.au. Eloise Foo, Email: eloise.foo@utas.edu.au. Data availability The data supporting the findings of this study are available as Supporting Information (Datasets S1–S3; Tables S1–S2). References Associated Data Supplementary Materials Dataset S1 Raw data of stem water potential (Ψstem) and root hydraulic conductance (K r) under well‐watered and soil drying conditions. Dataset S2 Physiological indices across replicates for the three genotypes of Pisum sativum . Dataset S3 Plant capacitance across the three genotypes of Pisum sativum . Fig. S1 Root hydraulic conductance (K r) normalised by projected leaf area under hydrated and drought conditions in Pisum sativum wild‐type (Frisson, Nod+), non‐nodulator (sym19, Nod‐), and supernodulator (Psrdn1, Nod++). Table S1 Model selection results for root hydraulic conductance (K r) normalised by leaf area and root dry mass (including nodule dry mass) in relation to stem water potential (−Ψstem) across different genotypes of Pisum sativum . Table S2 ANOVA summary table for the model explaining variation in root hydraulic conductance (K r). Please note: Wiley is not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Data Availability Statement The data supporting the findings of this study are available as Supporting Information (Datasets S1–S3; Tables S1–S2).