# Source paper archive

- Source title: Integrating Biometric, Metabolic, and Photochemical Traits to Identify Candidate Physiological Indicators of Salinity Tolerance and Forage Quality in Alfalfa Cultivars
- Author(s): Barbieri, G., Cornacchione, M., Luna, D. F., Annicchiarico, P., Rodriguez, M. S.
- Source landing URL: https://www.biorxiv.org/content/10.64898/2026.11.18.767659v1
- PDF URL: https://www.biorxiv.org/content/10.64898/2026.11.18.767659v1.full.pdf
- PDF SHA-256: 1aa8ac7e3e8385e56cd489f1671600822307db0390955b184e76f1fcbbeb682b
- Converter: pymupdf4llm 0.0.27
- Limits: Automated PDF text extraction; reading order, equations, tables and formatting may be incomplete or incorrect. Images are omitted, and no OCR was performed. Markup and links are neutralized for safe downloading. Consult the original PDF for the authoritative version.

---

&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


1 Research paper


2 **Integrating Biometric, Metabolic, and Photochemical Traits to Identify Candidate**


3 **Physiological Indicators of Salinity Tolerance and Forage Quality in Alfalfa Cultivars**


4 Barbieri Giuliano &#91;1&#93;, Cornacchione Mónica &#91;2&#93;, Luna Dario Fernando &#91;1;3&#93;, Annicchiarico Paolo &#91;4&#93;,


5 Rodriguez Marianela S. &#91;1;3&#93;     

6


7 1 Unidad de Estudios Agropecuarios (UDEA- INTA-CONICET) Camino 60 cuadras km 5.5


8 X5020ICA, Córdoba, Argentina.


9 2 Instituto Nacional de Tecnología Agropecuaria (INTA), Estación Experimental Santiago del


10 Estero, Jujuy 850, Santiago del Estero, 4200, Argentina.


11 3 Instituto de Fisiología y Recursos Genéticos Vegetales, Ing Agr Victorio Segundo Trippi


12 (IFRGV), Centro de Investigaciones Agropecuarias (CIAP), Instituto Nacional de Tecnología


13 Agropecuaria (INTA), Camino 60 cuadras km 5.5, X5020ICA, Córdoba, Argentina.


14 4 Consiglio per la Ricerca in Agricoltura e l'Analisi dell'Economia Agraria (CREA), Centro di


15 Ricerca Zootecnia e Acquacoltura, Lodi, Italy


16 Giuliano Barbieri: barbieri.giuliano@inta.gob.ar, orcid.org/0009-0004-8543-0444, Mónica


17 Cornacchione: cornacchione.monica@inta.gob.ar, orcid.org/0000-0002-6303-8407, Fernando


18 Luna: luna.fernando@inta.gob.ar, orcid.org/0000-0001-5550-0828, Paolo Annicchiarico:


19 paolo.annicchiarico@crea.gov.it, orcid.org/0000-0001-8027-1035,


20 *Corresponding author; Marianela S. Rodriguez, rodriguez.marianela@inta.gob.ar;


21 marianelasolrodriguez@gmail.com, orcid.org/0000-0003-4869-004X


22


23 Keywords: alfalfa; chlorophyll fluorescence; ion homeostasis; photochemical acclimation;


24 salinity tolerance; forage quality


25


26 **Running title:** Biomarkers of Alfalfa Salt Tolerance


27


28 **Highlight**


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


29 Time-resolved chlorophyll fluorescence dynamics, combined with leaf �� &#91;�&#93; /� &#91;�&#93; homeostasis


30 and carbohydrate partitioning, provide physiological candidate indicators for early salinity


31 tolerance screening in alfalfa.


32


33 **Abstract**


34 � Salinity limits alfalfa ( _Medicago sativa_ L.) productivity, yet integrative physiological


35 markers of cultivar tolerance remain poorly defined. This study aimed to identify candidate


36 physiological indicators of salt tolerance by integrating growth, metabolic, and time

37 resolved photosynthetic traits across diverse alfalfa cultivars.


38 � We exposed nine alfalfa cultivars to a 200 mM NaCl hydroponic shock and monitored their


39 physiological responses over 14 days. We assessed shoot and root growth, leaf Na � /K �


40 homeostasis, carbohydrate partitioning, dynamic chlorophyll a fluorescence (ChlF), and


41 NIRS-based forage quality.


42 � Salinity elicited cultivar-dependent responses. Tolerant cultivars-maintained shoot biomass,


43 restricted foliar Na+ accumulation, preserved leaf starch reserves and progressively


44 recovered Phi2 and linear electron flow (LEF) following initial declines. In contrast, less


45 tolerant cultivars showed persistent growth inhibition, reduced leaf starch, and persistent


46 photochemical impairment. At 7 and 14 dpt, PI total and F � ’/F � ’ correlated positively with


47 final biomass, whereas PhiNPQ and PhiNO showed negative correlations. Notably, the


48 association between regulated energy dissipation (PhiNPQ) and biomass retention inverted


49 over time—shifting from an early photoprotective mechanism at 1 dpt to a sign of chronic


50 photoinhibitory strain by 7 and 14 dpt. These responses were accompanied by cultivar

51 specific changes in Na � /K � balance, soluble sugars, starch, and forage quality.


52 � Tracking time-resolved PSII photochemistry reveals that maintaining �� total and F � ’/F � ’


53 under saline stress offers an accurate, non-invasive proxy for evaluating cultivar resilience.


54 Their integration into high-throughput phenotyping and breeding programs could facilitate


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


55 the selection of resilient cultivars. Field validation should consider both biomass production


56 and forage nutritional quality.


57


58 Abbreviations. ADF, acid detergent fiber; ADL, acid detergent lignin; ChlF, chlorophyll _a_


59 fluorescence; CP, crude protein; dpt, days post treatment; DW, dry weight; NDF, neutral


60 detergent fiber.


61


62 **1. Introduction**


63 Salinity is one of the major abiotic limitations on agricultural productivity, particularly


64 in irrigated and semi-arid regions where salt accumulation is intensified by limited rainfall,


65 freshwater scarcity and irrigation practices (Munns &amp; Millar 2023). The most recent global


66 assessment by FAO estimates that salt-affected soils cover approximately 1.38 billion ha,


67 representing 10.7% of the global land area, with around 10% of both irrigated and rainfed


68 cropland affected (FAO, 2024). Salinity causes both osmotic and ionic stresses. The early


69 osmotic phase restricts water uptake and carbon assimilation, whereas prolonged exposure


70 promotes Na � accumulation, nutrient imbalance and oxidative stress, ultimately reducing plant


71 growth and yield (van Zelm _et al._ 2020; Ma _et al_ . 2026). Because these effects develop over


72 time, salinity tolerance depends not only on the initial stress response but also on the capacity to


73 re-establish physiological homeostasis during acclimation.


74 Alfalfa ( _Medicago sativa L._ ) is the most widely grown perennial legume worldwide


75 (Annicchiarico _et al._ 2015) and represents a cornerstone of global livestock systems, owing to


76 its superior biomass production and nutritional profile. Its cultivation is increasingly constrained


77 by soil salinity and the reliance on marginal irrigation water. Salinity tolerance is already a key


78 factor, along with drought tolerance, for cultivar targeting in various regions, such as the


79 Mediterranean (Annicchiarico et al., 2011). Alfalfa exhibits moderate, cultivar-dependent salt


80 tolerance, offering avenues for genetic improvement (Zhang _et al_ . 2024; Pecetti _et al._ 2024).


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


81 However, the physiological mechanisms coordinating ion homeostasis, carbon metabolism, and


82 growth among alfalfa cultivars under salinity remain incompletely understood. Moreover, while


83 the deleterious effects of salinity on biomass production are well established, its effects on


84 forage quality, including crude protein (CP), fiber fractions (NDF, ADF, ADL), and non

85 structural carbohydrates (NSC), remain comparatively less documented. In general, forage


86 quality is highly variable and deeply influenced by stress severity and genetic background


87 (Moore &amp; Hatfield 1994; Van Soest 1994). Salinity-induced changes in forage quality are


88 driven by carbon reallocation toward soluble sugar accumulation for osmotic adjustment,


89 thereby altering NSC composition (Sandhu _et al_ . 2017). Salinity may influence forage quality


90 through multiple mechanisms, including osmotic effects on plant growth and developmental


91 progression, which can alter fiber accumulation and leaf-to-stem allocation. Similar responses


92 have been documented in drought-stressed alfalfa, where changes in growth and plant


93 morphology were associated with shifts in forage composition (Halim _et al_ ., 1989). In


94 particular, negative associations between biomass production and forage quality have been


95 reported across diverse alfalfa accessions under water deficit (Zhang _et al_ ., 2018). Drought

96 induced increases in leaf-to-stem ratio, digestibility, and crude protein content have also been


97 documented (Carter &amp; Sheaffer, 1983; Pecetti _et al_ ., 2017). Although the underlying


98 mechanisms and magnitude of these responses may differ between water deficit and salinity,


99 these findings suggest that stress-induced changes in plant growth and biomass partitioning can


100 have important consequences for forage nutritional value. Therefore, assessing salinity tolerance


101 in alfalfa requires considering forage quality alongside biomass production.


102 Ion homeostasis is a major component of salt tolerance. Under saline conditions,


103 excessive Na � accumulation in leaves disrupts enzymatic activity, membrane function and


104 nutrient balance, particularly by interfering with K � -dependent metabolic processes (Munns &amp;


105 Millar 2023). Tolerant cultivars often restrict Na � transport to aerial tissues or maintain a more


106 favorable Na � /K � ratio, thereby reducing ionic toxicity (Smethurst _et al_ . 2008; Cornacchione


107 &amp; Suarez 2015). However, ion regulation alone does not fully explain differences in cultivar


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


108 performance, because growth under salinity also depends on the maintenance of photosynthetic


109 activity and carbon supply. Therefore, carbon metabolism adds another layer to salinity


110 acclimation. For instance, soluble sugars can contribute to osmotic adjustment, cellular


111 protection, redox regulation and stress signaling, whereas starch metabolism reflects changes in


112 carbon storage and mobilization between source and sink tissues (Saddhe _et al._ 2021; Ribeiro _et_


113 _al._ 2022). Under salinity, sustained carbon supply is essential for root maintenance, ion


114 transport, and osmotic regulation. Thus, carbohydrate dynamics may help explain why some


115 cultivars recover photosynthetic function and maintain growth, whereas others experience


116 carbon limitation and growth inhibition.


117 Photosynthesis is one of the earliest physiological processes affected by salinity


118 (Chaves _et al._ 2009). Reduced CO � assimilation increases excitation pressure within the


119 photosynthetic apparatus, leading to lower photosystem II (PSII) efficiency, altered electron


120 transport and activation of energy dissipation mechanisms that protect against photodamage


121 (Baker 2008). Chlorophyll a fluorescence (ChlF) therefore provides a sensitive approach to


122 detect stress-induced changes in photosynthetic function before visible damage becomes evident


123 (Murchie &amp; Lawson 2013). Measurements over time can distinguish transient photoprotective


124 adjustment from persistent photoinhibition. This distinction is particularly relevant for cultivar


125 screening, because tolerance may depend more on photosynthetic stability than on the


126 magnitude of the initial decline.


127 Developing alfalfa cultivars with enhanced salt tolerance and forage quality is critical


128 for sustaining agricultural production in increasingly saline environments. In this study, we


129 evaluated nine alfalfa cultivars previously characterized and selected for their diverse responses


130 to salinity (Pecetti et al., 2024). We hypothesized that i) photosynthetic acclimation following


131 an acute salt shock, rather than initial baseline resistance, determines final biomass preservation


132 across cultivars; and ii) salinity decouple from overall forage nutritional value due to non

133 structural carbohydrate alteration and cell wall lignification. Specifically, we investigated


134 whether cultivars that maintain biomass under salt stress also maintain PSII efficiency following


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


135 an initial decline, and whether this stability correlates with distinct leaf Na � /K � ratios or


136 carbohydrate partitioning. A secondary objective was to determine the impact of salinity on key


137 forage quality traits. To achieve these aims, we analyzed biometric parameters, leaf ion


138 homeostasis (Na � and K � ), shoot and root soluble sugar and starch profiles, and chlorophyll


139 fluorescence (ChlF) dynamics. By integrating these physiological dimensions, we aimed to


140 identify physiological and photosynthetic indicators integrated into predictive multivariate


141 models for salt tolerance in alfalfa, ultimately contributing to the selection of more resilient


142 cultivars. To this end, we applied an acute 200 mM NaCl treatment under hydroponic


143 conditions to elicit contrasting physiological responses among cultivars and facilitate the


144 identification of discriminative traits. This approach provides a framework for rapid phenotypic


145 screening that could support alfalfa breeding programs targeting saline environments.


146


147 **2. Materials and Methods**


148 **2.1. Plant material and experimental setup**


149 Seeds of nine alfalfa ( _Medicago sativa_ L.) cultivars were used, including Ameristand


150 801S and Salado bred in the USA, the Argentinian cultivars Kumen and Monarca, the Italian


151 cultivar Prosementi, the Tunisian landrace Chenini, and CREA’s selections MSI036, MSI037


152 and MSI038. All cultivars except Prosementi underwent a previous field evaluation in either or


153 both Northern Argentina under saline soil and Tunisia under irrigation with saline water (Pecetti


154 et al., 2024), showing salt tolerance ranging from fairly modest to high partly as a function of


155 the test region. Prosementi represented a salt less tolerant reference cultivar based on results in


156 Annicchiarico _et al._ (2011) and complementary results in Pecetti _et al._ (2024).


157 Seeds of each cultivar were germinated on mesh supports containing moist vermiculite


158 positioned over 3-L containers filled with five-fold diluted B&amp;D nutrient solution (Broughton &amp;


159 Dilworth, 1971), supplemented with NH � NO � to a final concentration of 5 mM under


160 continuous aeration. The nutrient solution was adjusted to pH 6.5 and renewed weekly. Plants


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


161 were grown in a controlled-environment growth chamber under a 16 h light/8 h dark


162 photoperiod, a photosynthetic photon flux density of 300 μ mol m � ² s � ¹, and a temperature of


163 23 ± 2 °C.


164 Three independent experiments were conducted. Salt stress was imposed 20 d after


165 sowing by adding a shock of NaCl 200 mM. The acute 200 mM NaCl shock was selected based


166 on preliminary screening assays establishing this concentration as the optimal threshold to


167 maximize inter-cultivar physiological variance and induce rapid metabolic reprogramming


168 without causing unspecific osmotic collapse within 14 days. Control plants remained in nutrient


169 solution without NaCl. Leaf and root samples were collected 14 days post treatment (dpt).


170 Samples or biochemical analyses were immediately frozen in liquid nitrogen and stored at −80


171 °C until further analysis (Fig. S1).


172


173 **2.2. Biometric parameters**


174 Biometric measurements were recorded for both control and salt-treated plants at the


175 end of the experimental period, based on three independent experiments with thirty plants per


176 treatment (n = 3). The following variables were assessed: shoot length (cm), main root length


177 (cm), and total leaf number per plant. For biomass determination, shoots and roots were


178 separately harvested and oven-dried at 40 °C for 72 h until constant weight.


179


180 **2.3. Forage quality parameters**


181 Forage nutritional quality was assessed using near-infrared reflectance spectroscopy


182 (NIRS, DS2500 FOSS). Samples were obtained from three independent experiments, each


183 comprising 30 plants per treatment pooled into a single sample (n = 3). Ground samples (1-mm


184 particle size) were scanned in reflectance mode over the 400–2500 nm wavelength range at 2

185 nm intervals, with 32 co-added scans per spectrum. Calibration equations were developed using


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


186 reference samples for which crude protein (CP; n = 861), ash (n = 424), neutral detergent fiber


187 (NDF; n = 594), acid detergent fiber (ADF; n = 1035), and acid detergent lignin (ADL; n = 167)


188 concentrations had been determined by wet-chemistry methods according to AOAC (1990).


189 Spectral data were pre-processed using standard normal variate (SNV) and detrending


190 transformations, and calibration equations were developed by partial least squares (PLS)


191 regression using leave-one-out cross-validation. All experimental samples were analyzed in


192 duplicate, and predicted values were expressed as a percentage of dry matter (% DM).


193


194 **2.4. Ion concentration**


195 Leaf Na � and K � were extracted from dried leaf powder in 30% (v/v) methanol and


196 quantified by non-suppressed ion chromatography with conductivity detection (Shim-pack IC

197 C3 column; 3 mM oxalic acid as the mobile phase; 40 °C; 1.2 mL min � ¹), following Pantsar

198 Kallio &amp; Manninen (1995). Na � and K � were determined by HPLC (Shimadzu Prominence


199 Modular HPLC, Kyoto, Japan). Chromatographic runs were performed at 40 °C with a flow rate


200 of 1.2 mL min � ¹ for 18 min under non-suppressed conditions, and ions were detected by


201 conductivity. Chromatography was registered and analyzed using LabSolutions (ver. 5.6)


202 software (Pantsar-Kallio &amp; Manninen 1995). Three plants per treatment were pooled in each of


203 three independent experiments (n = 3).


204


205 **2.5. Soluble Sugars (SS) content**


206 Total soluble sugars (SS) were quantified by the anthrone method (Fales 1951), adapted


207 to a final reaction volume of 200 µL. Ethanolic leaf and root extract was mixed with anthrone


208 reagent, heated at 100 °C for 5 min, rapidly cooled on ice, and absorbance was measured at 620


209 nm. Glucose was used for calibration. Three plants per treatment were pooled in each of three


210 independent experiments (n = 3).


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


211


212 **2.6. Starch content**


213 Starch content was determined in leaf and root pellets by quantifying the reducing


214 sugars released following hydrolysis of the pellet with α -amyloglucosidase (Sumner &amp; Somers


215 1944), using glucose as the standard. Values were expressed as mg glucose equivalents g � ¹


216 DW. Three plants per treatment were pooled in each of three independent experiments (n = 3).


217


218 **2.7. Chlorophyll-** _**a**_ **fluorescence and photosynthetic performance measurements**


219 ChlF and photosynthetic performance measurements were conducted at 1, 7, and 14 dpt.


220 Fast ChlF transients were recorded using a Pocket-PEA fluorometer (Plant Efficiency Analyzer;


221 Hansatech Instruments Ltd., King’s Lynn, Norfolk, UK). The third fully expanded leaf of 12


222 plants per cultivar and treatment in each of three independent experiments (36 plants per


223 cultivar and treatment in total) were dark-adapted for at least 30 min using leaf clips to ensure


224 complete oxidation of the PSII reaction centers. Leaves were subsequently exposed to a


225 saturating light pulse of 3500 μ mol photons m � ² s � ¹ for 1 s. The OJIP transients were analyzed


226 according to the JIP-test (Strasser et al., 2004). Derived parameters included the absorption flux


227 per active PSII reaction center (ABS/RC), electron transport flux beyond Q � per active PSII


228 reaction center (ET � /RC), energy dissipation flux per active PSII reaction center (DI � /RC),


229 electron flux reducing terminal electron acceptors at the PSI acceptor side per active PSII


230 reaction center (RE � /RC), and the total performance index (PI ����� ). Definitions of all


231 calculated parameters are provided in Table S1.


232 Complementary measurements were conducted at midday on the third fully expanded


233 leaves using a MultispeQ V2.0 handheld device connected to the PhotosynQ platform (Kuhlgert


234 _et al_ . 2016), following the publicly available “Photosynthesis Rides 500” protocol. The


235 parameters obtained included the maximum efficiency of PSII in the light-adapted state


236 (F � ’/F � ’), effective quantum yield of PSII photochemistry (Phi2), estimated linear electron


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


237 flow (LEF), the fraction of open PSII reaction centers (qL), proton conductivity of the thylakoid


238 membrane (gH � ), steady-state proton flux across the thylakoid membrane (vH � ), non

239 photochemical quenching (NPQt), and the quantum yields of regulated (PhiNPQ) and non

240 regulated (PhiNO) energy dissipation (Table S1).


241


242 **2.8. Statistical Analysis**


243 All experiments were performed with at least three biological replicates. Statistical


244 significance among means was assessed by analysis of variance (ANOVA), followed by


245 Tukey’s post hoc test ( _P_ &lt; 0.05). Data processing and statistical analyses were conducted using


246 InfoStat software (Di Rienzo _et al.,_ 2019). Spearman correlation analyses were performed using


247 GraphPad Prism version 10.0.0 for macOS (GraphPad Software, Boston, MA, USA).


248 Percentage changes relative to the control were calculated as &#91;(treatment mean − control


249 mean)/control mean&#93; × 100, with positive and negative values indicating increases and


250 decreases, respectively.


251 For the integrative analyses, data were summarized at the cultivar level (n = 9). Shoot


252 dry weight values from the six biological replicates available for each cultivar and treatment


253 were averaged, and biomass retention under salinity was calculated as the ratio between mean


254 shoot dry weight under NaCl treatment and mean shoot dry weight under control conditions


255 (NaCl/Control). This ratio was used as a quantitative indicator of cultivar performance under


256 salinity.


257 Multivariate analyses were performed in Python using the pandas, SciPy, and scikit

258 learn libraries. Principal component analysis (PCA) was conducted using the 52 physiological,


259 chlorophyll fluorescence (ChlF), ionic, and biochemical variables measured in the study. For


260 each variable, the salinity response was expressed as a log � fold change (NaCl/Control), and


261 the resulting data matrix was standardized prior to PCA. The association between PC1 scores


262 and biomass retention was assessed using Spearman’s rank correlation.


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


263 Pairwise Spearman correlations were also calculated among the 52 measured variables


264 and biomass retention across the nine cultivars, resulting in 1,378 unique pairwise comparisons.


265 P-values were adjusted for multiple testing using the Benjamini–Hochberg false discovery rate


266 (FDR) procedure, with statistical significance defined as an FDR-adjusted P &lt; 0.05.


267


268 **3. Results**


269 **3.1. Salinity induces contrasting shoot growth preservation and root biomass allocation**


270 **across alfalfa cultivars**


271 The aerial response to 200 mM NaCl varied among the nine evaluated cultivars (Table


272 1, Fig. S2). Ameristand 801S, Salado, MSI038, Kumen, and Chenini exhibited better ability to


273 maintain shoot growth under stress relative to control conditions, with small, nonsignificant


274 reductions in the measured traits (P &gt; 0.05). Consistent with these responses, shoot biomass


275 retention under salinity was highest in Kumen and Ameristand 801S (Fig. 1). In contrast,


276 Prosementi exhibited the greatest overall susceptibility to salinity, suffering high and significant


277 (P &lt; 0.05) reductions in shoot DW (-32.0%), leaf number and shoot length relative to the


278 control. Monarca and MSI037 exhibited a significant decrease in all growth traits, whereas


279 MSI036 showed a decrease in shoot DW combined with a relatively stable shoot length.


280 The analysis of the root traits revealed cultivar responses that were partly different from


281 those for shoot traits (Table 1). The accumulation of root biomass was highly sensitive in


282 specific cultivars, yet longitudinal root elongation remained unaffected across the board. Root


283 DW was significantly reduced in Monarca and Prosementi by 40.0% and 33.3%, respectively,


284 relative to their corresponding cultivar controls (P &lt; 0.05). In contrast, no statistically


285 significant reductions in root length were detected for any of the other cultivars. These


286 dynamics indicate that the reduced root biomass accumulation observed in less tolerant cultivars


287 occurred without a detectable inhibition of primary root elongation.


288


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


289 **3.2. Salinity Alters Alfalfa Forage Quality through Cultivar-Specific Shifts in Protein,**


290 **Structural Fiber, and Non-Structural Carbohydrates**


291 NIRS-based estimates revealed shifts in forage quality, with significant cultivar

292 dependent responses to salt stress (Table 2). For example, total Ash% content declined


293 significantly (P &lt; 0.05) across all cultivars, with reductions ranging from -16.6% in Ameristand


294 801S to -28.6% in Monarca. CP content exhibited significant stress-induced increases (P &lt; 0.05)


295 in six cultivars and especially in MSI038 (+50.3%), Salado (+48.3%), MSI037 (+43.6%), and


296 MSI036 (+40.8%). CP concentration did not differ significantly from the corresponding control


297 in Ameristand 801S, Prosementi, and Monarca. The structural carbohydrate matrix, which


298 dictates forage digestibility and voluntary intake limits, demonstrated complex, divergent


299 responses among cultivars. ADL increased significantly in MSI037 (+19.5%), MSI038


300 (+16.3%), Ameristand 801S (+11.3%), and MSI036 (+11.0%). In contrast, ADF decreased


301 significantly in Monarca (−12.1%), Prosementi (−10.4%), and Ameristand 801S (−9.4%). NDF


302 decreased significantly in Ameristand 801S (−6.2%) but increased significantly in MSI036


303 (+8.2%) and MSI038 (+3.8%). Non-structural carbohydrate (NSC) content, calculated as the


304 sum of independently measured soluble sugars and starch, was lower under salinity in all


305 cultivars, with decreases ranging from 6.4% in Chenini to 57.2% in MSI036. The largest


306 reductions occurred in MSI036 (−57.2%), Monarca (−48.7%), and Prosementi (−43.6%).


307


308 **3.3. Disruption of Leaf Ion Homeostasis and Na** � **Accumulation Are Associated with Less**


309 **Tolerant Cultivars**


310 Leaf Na � concentration increased significantly in all alfalfa cultivars under salt stress


311 compared with their respective control (Fig. 2A). However, the extent of this ion accumulation


312 revealed statistically significant cultivar variations. The more salt-tolerant group based on the


313 biomass results (Chenini, Ameristand 801S, Kumen, and MSI038) showed comparatively lower


314 foliar Na � accumulation. In contrast, Prosementi, Monarca, and MSI036 accumulated


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


315 substantially higher foliar Na � concentrations. Several cultivars also exhibited significant


316 increases in K � concentration under salinity (P &lt; 0.05) (Fig. 2B). While Ameristand 801S,


317 MSI038, and Chenini showed no significant changes in K � concentration; Salado, Kumen,


318 MSI036, Monarca, and Prosementi showed significant increases. The cultivar MSI037 showed a


319 distinct ionic response, combining high Na � accumulation with a marked and significant


320 increase in K � concentration. Some cultivars showed comparatively smaller increases in the


321 Na � /K � ratio under salinity (Fig. 2C). This group included Kumen, Chenini, Ameristand 801S,


322 and MSI037. Cultivars MSI038 and Salado occupied intermediate positions. Na � /K � under


323 NaCl was highest in MSI036, Monarca and Prosementi (Fig. 2C; Table S2).


324


325 **3.4. Salinity Elicits Cultivar-Specific Reprogramming of Source–Sink Carbohydrate**


326 **Partitioning and Soluble Sugar-to-Starch Ratios**


327 Whole-plant carbohydrate dynamics (Table S3) provided insights into the relationship


328 between carbon assimilation, carbohydrate allocation, and biomass accumulation, helping


329 interpret the observed phenotypic responses. Salt stress induced cultivar-specific shifts in


330 carbon homeostasis, as evidenced by changes in soluble sugar and starch pools, reflecting


331 adjustments in readily available carbohydrates involved in transport and signaling and in


332 transient carbon and energy reserves, respectively. Within the foliar source tissues, soluble


333 sugar concentrations exhibited divergent, statistically significant responses (P &lt; 0.05) to


334 salinity. A marked accumulation of soluble sugars was observed in MSI037 (+79.7%), MSI038


335 (+46.6%), and Chenini (+38.7%). In contrast, Monarca, Kumen, Prosementi, and MSI036


336 showed decrease in their foliar soluble sugar, ranging from -43.6 to -24.2%. Foliar starch


337 content was severely and significantly (P &lt; 0.05) decreased in MSI036 (-63.8%), Monarca (

338 49.4%), Prosementi (-46.7%), and MSI037 (-45.9%). This collective metabolic adjustment


339 altered the foliar soluble sugar-to-starch ratio (Soluble sugar/Starch) (Fig. 3A), with a


340 particularly high increase of the ratio observed in MSI037 (+232.31%), MSI036 (+109.60%),


341 and MSI038 (+96.11%). The increased soluble sugar/starch ratio was consistent with a stress

&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


342 induced shift in carbohydrate partitioning toward soluble sugars relative to starch, potentially


343 contributing to osmotic adjustment.


344 The partitioning of carbohydrates to the root system, the primary sink during osmotic


345 stress, further outlined the different cultivar adaptive strategies for stress tolerance. Soluble


346 sugar content in roots increased significantly (P &lt; 0.05) in Prosementi (+77.8%), MSI037


347 (+64.3%), Kumen (+25.7%), and Chenini (+22.9%), suggesting an enhanced requirement for


348 active osmolytes or metabolic signaling molecules within the root organ to sustain growth under


349 saline conditions. In contrast, Monarca and MSI036 exhibited a substantial and significant


350 decline in root soluble sugars (-50.0% and -46.6%, respectively). Root starch content also


351 showed cultivar-dependent responses. Salado and Monarca significantly increased root starch


352 by 52.8% and 46.9%, respectively, whereas Kumen, Chenini, Ameristand 801S, and MSI037


353 showed significant decrease of root starch reserves, ranging from approximately -29.1 to 

354 38.1%. Consequently, the root soluble sugar/starch ratio revealed contrasting carbon allocation


355 patterns across cultivars (Fig. 3B), with MSI037 (+131.8%) and Kumen (+103.3%) showing a


356 shift toward soluble sugars relative to stored non-structural carbohydrate, whereas Monarca


357 (−65.9%) and MSI036 (−63.7%) showed the opposite pattern, characterized by a marked


358 decrease in soluble sugars relative to their starch pools (Table S3). Figure 3B shows that the


359 MSI036 control (ratio ~0.6) was substantially higher than those of the other cultivars (~0.1–


360 0.25); therefore, the −63.7% reduction observed in this cultivar is largely influenced by an


361 atypical control value.


362


363 **3.5. Time-resolved ChlF measurements**


364 Salt stress induced distinct cultivar-specific temporal shifts in photochemical efficiency


365 respect to control (Fig. S3). At 1 dpt, genotypic differences were primarily evident in the


366 magnitude and nature of the early photochemical response. Ameristand 801S showed an


367 increase in PhiNPQ together with only minor decreases in ETo/RC, REo/RC, and Phi2,


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


368 indicating relatively limited early perturbation of photochemical performance. In contrast,


369 MSI038, Salado, Kumen, and Chenini exhibited a broader early response characterized by


370 reductions in Phi2, ETo/RC, REo/RC, and PI total, accompanied by rapid increases in NPQt and


371 PhiNPQ, consistent with enhanced regulated energy dissipation. Monarca and MSI037


372 exhibited immediate photochemical impairment across several parameters, including F � ’/F � ’,


373 Phi2, LEF, and PI total, whereas Prosementi displayed no detectable alterations at this initial


374 stage.


375 By 7 dpt, physiological differences respect to control among cultivars became more


376 apparent. Ameristand 801S maintained marked photochemical stability, exhibiting only


377 localized reductions in REo/RC. MSI038 and Chenini demonstrated partial photochemical


378 stability, with residual impairment largely restricted to electron-transport-related parameters.


379 Salado and Kumen maintained enhanced regulated energy dissipation together with moderate


380 reductions in ETo/RC and LEF. In contrast, Prosementi, Monarca, and MSI037 exhibited


381 persistent photochemical impairment, characterized by reductions in PI total, F � ’/F � ’, Phi2, and


382 LEF, together with increases in ABS/RC and DI � /RC, sustained regulated energy dissipation


383 (NPQt and PhiNPQ), and increased non-regulated energy dissipation (PhiNO).


384 By 14 dpt, cultivar-specific differences in photochemical acclimation became more


385 pronounced. Chenini and Kumen showed the strongest preservation of photochemical


386 performance, with values approaching those of their respective controls, whereas Ameristand


387 801S and MSI038 exhibited only minor residual impairments. Salado displayed an intermediate


388 response, with persistent reductions in ETo/RC and LEF. In contrast, Prosementi and Monarca,


389 and to a lesser extent MSI037 and MSI036, exhibited sustained declines in PSII photochemical


390 efficiency and electron transport, as indicated by reductions in F � ’/F � ’, Phi2, LEF, and


391 REo/RC, together with increased non-photochemical energy dissipation. These alterations were


392 associated with greater biomass penalties under salinity.


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


393 Consistent with these endpoint differences, the temporal analysis of Phi2 and PhiNPQ


394 between 1 and 14 dpt revealed distinct photosynthetic response patterns among cultivars. Under


395 salinity, Kumen, Ameristand, and Chenini maintained relatively stable Phi2, with no significant


396 changes over time. In contrast, the remaining cultivars exhibited significant declines,


397 particularly MSI037 (−31.11%), followed by Salado (−22.22%), MSI038 (−19.57%), and


398 MSI036 (−19.51%).


399


400 **3.6. Integrative Trait Network Analysis Links Temporal Chlorophyll Fluorescence**


401 **Dynamics and Foliar Na** � **to Biomass Maintenance**


402 To broadly map multi-trait interaction networks without a priori functional assumptions,


403 pairwise Spearman correlations were performed across the complete 52-variable dataset (1,378


404 unique pairwise comparisons; Fig. S4). Given the extreme statistical penalty inherent to


405 Benjamini-Hochberg FDR adjustments over high-dimensional exploratory matrices, nominally


406 significant correlations (126 pairs, P &lt; 0.05) were utilized strictly for pattern discovery rather


407 than univariate hypothesis testing. To resolve biologically relevant trait clusters, a targeted


408 correlation sub-matrix (Fig. 4) was constructed focusing on 12 core physiological traits that


409 demonstrated consistent covariance and strong alignment with principal component axes.


410 Among the strongest observed associations, several involved photochemical


411 performance and energy dissipation. At 7 dpt, F � ’/F � ’ was positively associated with PI total ( ρ


412 = 0.93), whereas PhiNPQ was negatively associated with PI total ( ρ = −0.93). NPQt at 7 dpt was


413 strongly negatively associated with F � ’/F � ’ at 14 dpt ( ρ = −0.97). The strongest positive


414 association was observed between PhiNPQ at 7 and 14 dpt ( ρ = 0.983), suggesting that cultivar

415 specific patterns of energy dissipation were maintained across these sampling times. At 14 dpt,


416 PI total was negatively associated with TRo/RC ( ρ = −0.93). Among early responses, gH � and


417 vH � at 1 dpt were strongly positively associated ( ρ = 0.97). In addition, foliar Na �


418 concentration was negatively associated with starch content ( ρ = −0.95).


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


419 Associations between ChlF responses at 1, 7, and 14 dpt and final shoot biomass


420 retention were also examined. Several relationships showed nominal associations with biomass


421 retention; however, these did not remain significant after correction for multiple comparisons


422 across the complete correlation matrix. At 1 dpt, PhiNPQ showed a positive association with


423 biomass retention ( ρ = 0.72). At 7 dpt, PI total and F � ’/F � ’ showed strong positive associations


424 with biomass retention, whereas PhiNPQ and PhiNO showed negative associations (Fig. 5).


425 NPQt and gH � also displayed negative trends. At 14 dpt, F � ’/F � ’ and Phi2 showed positive


426 associations with biomass retention, while PhiNPQ remained negatively associated.


427 Notably, the direction of the PhiNPQ–biomass retention relationship changed over time,


428 from positive at 1 dpt to negative at 7 and 14 dpt. This temporal pattern may reflect differences


429 in the role of energy dissipation during the progression of salt stress. Early increases in PhiNPQ


430 could be associated with an initial photoprotective response, whereas sustained or elevated


431 energy dissipation at later stages may indicate persistent stress and a greater limitation in


432 maintaining photosynthetic performance. Nevertheless, these relationships should be interpreted


433 as exploratory associations rather than evidence of a causal mechanism. Consequently, while


434 individual univariate correlations within the high-dimensional matrix serve as exploratory


435 associations, the definitive validation of these trait relationships requires dimension-reducing


436 multivariate models that bypass pairwise multiple-testing constraints.


437


438 **3.7. Multivariate Trait Integration and Principal Component Analysis of Alfalfa Cultivar**


439 **Responses to Salinity**


440 Integration of the measured traits revealed distinct cultivar-specific response profiles


441 under salinity (Fig. 6). Cultivars with greater biomass retention, particularly Kumen,


442 Ameristand 801S, Chenini, and MSI038, generally exhibited smaller or transient alterations in


443 photochemical performance and electron transport, together with comparatively limited changes


444 in growth traits. In contrast, cultivars with lower biomass retention, particularly Monarca and


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


445 Prosementi, showed broader and more persistent reductions in photochemical performance,


446 accompanied by stronger changes in non-photochemical energy dissipation and growth-related


447 traits. Salado, MSI036, and MSI037 displayed intermediate or contrasting response patterns


448 depending on the physiological process considered. Variation in ion accumulation, non

449 structural carbohydrate pools, and forage-quality traits further demonstrated that cultivar


450 responses to salinity involved changes across multiple physiological levels. Overall, the


451 integrated profiles were consistent with the cultivar ranking based on final biomass retention.


452 To further resolve the multivariate structure of these responses, a principal component


453 analysis (PCA) was performed using the salinity responses of the 52 measured physiological,


454 photochemical, ionic, and biochemical variables (Fig. 7A). The first principal component (PC1)


455 explained 34.1% of the total variance, while PC2 explained 21.1%, together accounting for


456 55.2% of the total variation. Cultivars were distributed broadly across the PC1–PC2 space,


457 indicating substantial differences in their integrated responses to salinity. Kumen and


458 Ameristand 801S, which exhibited the highest biomass retention, were positioned toward the


459 positive PC1 region, whereas Monarca and Prosementi, which showed lower biomass retention,


460 were located toward the negative side of PC1. MSI038 also showed a relatively high positive


461 PC1 score, while the remaining cultivars occupied intermediate positions. This distribution


462 indicates that the principal multivariate axis captured a broad, multi-trait physiological gradient


463 underlying salinity response, which fundamentally aligns with and explains overall biomass


464 preservation.


465 PC1 scores were positively associated with final shoot biomass retention across


466 cultivars (Spearman’s ρ = 0.800, _P_ = 0.0096, _n_ = 9; Fig. 7B), with higher scores corresponding


467 to greater biomass retention under salinity. This relationship linked the multivariate


468 physiological response profiles to cultivar differences in biomass maintenance.


469


470 **4. Discussion**


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


471 This study provides new insights into the physiological basis of intraspecific variation


472 in salt tolerance in alfalfa and its implications for forage quality. The cultivar responses


473 characterized through shoot and root traits under hydroponic conditions were largely consistent


474 with field performance previously reported by Pecetti _et al._ (2024) in saline environments in


475 Argentina and Tunisia. This agreement supports the suitability of the hydroponic system as a


476 rapid phenotyping approach for characterizing genotypic variation in salt tolerance and


477 identifying promising candidates for breeding programs. Specifically, our results corroborate


478 the superior salt tolerance of cv. Kumen (previously evaluated as experimental line SISA14;


479 Cornacchione &amp; Suarez, 2017). An exception to this field-hydroponic correlation was cv.


480 MSI037. Although MSI037 exhibited consistently high long-term biomass yield in saline fields,


481 it did not rank among the most tolerant cultivars in our acute stress assay. This discrepancy


482 suggests that short-term physiological responses may not fully capture cultivar resilience under


483 chronic field salinity. Nevertheless, rather than segregating cultivars into strictly tolerant or


484 sensitive categories, our findings reveal a phenotypic continuum in which different cultivars


485 deploy distinct combinations of adaptive traits. For instance, despite its lower overall biomass


486 ranking in this study, MSI037 exhibited K � accumulation coupled with enhanced soluble sugar


487 accumulation in both leaves and roots likely supporting osmotic adjustment. Furthermore,


488 maintaining K � homeostasis is proposed as a critical determinant of salinity tolerance in alfalfa,


489 particularly under severe NaCl stress (150–200 mM), where ionic imbalances drive severe


490 growth inhibition and loss of vegetative vigor (Wang _et al_ ., 2026). Ultimately, these genotype

491 specific responses underscore the mechanistic diversity of alfalfa salt adaptation, demonstrating


492 that distinct physiological strategies can optimize performance under stress without necessarily


493 converging on a single, broad-spectrum tolerance phenotype.


494 Our results indicate that salt tolerance in alfalfa is not determined by the absence or


495 magnitude of the initial stress response, but rather by the capacity to sustain physiological


496 functioning during prolonged salt exposure. Although all cultivars exhibited physiological


497 adjustments in response to salinity, more tolerant cultivars, including Ameristand 801S,


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


498 MSI038, Salado, Kumen, and Chenini, better maintained growth and biomass accumulation,


499 whereas Prosementi and Monarca showed more pronounced growth inhibition. These


500 contrasting phenotypic responses were accompanied by differences in photochemical stability,


501 ion homeostasis, and organ-specific carbohydrate dynamics. Collectively, our findings suggest


502 that intraspecific variation in alfalfa salt tolerance arises from the coordinated contribution of


503 photosynthetic performance, ionic regulation, and carbohydrate metabolism (Daud _et al_ ., 2026).


504 Beyond biomass preservation, our results revealed substantial cultivar-specific changes


505 in forage nutritional quality under salinity. Salinity can reduce biomass yield even when forage


506 quality is maintained or modestly improved (Anderson _et al_ ., 2023; Díaz _et al_ ., 2018; Pérez

507 Reverón _et al_ ., 2024). CP content tended to increase under salt stress, consistent with previous


508 reports under mild salinity (Anderson _et al_ ., 2023; Ferreira _et al_ ., 2015). Such increases may


509 partly result from a concentration effect associated with reduced biomass accumulation and


510 changes in leaf-to-stem allocation, given the generally higher protein content of leaves


511 compared with stems (Pecetti _et al_ ., 2017). However, CP content did not increase significantly


512 in Ameristand 801S, Prosementi, or Monarca, despite their contrasting growth responses. While


513 Ameristand 801S largely maintained growth under salinity, Prosementi and Monarca exhibited


514 fewer leaves and shorter stems, indicating that changes in plant architecture did not necessarily


515 translate into increased CP concentration. Changes in structural fiber fractions also varied


516 among cultivars, with potential implications for forage digestibility and intake (Van Soest,


517 1994). The significant increases in NDF and lignin (ADL) observed in MSI036 and MSI038


518 suggest cultivar-specific alterations in cell wall composition that could negatively affect forage


519 digestibility (Le Gall _et al_ ., 2015; Vanholme _et al_ ., 2019). Conversely, ADF decreased in


520 Ameristand 801S, Monarca, and Prosementi, while NDF also declined in Ameristand 801S.


521 These contrasting responses indicate that changes in individual nutritional components do not


522 necessarily reflect a consistent improvement or deterioration in overall forage quality.


523 Furthermore, lower NSC content in the less tolerant cultivars (MSI036, MSI037, Monarca, and


524 Prosementi) may represent an additional nutritional and metabolic consequence of salt stress.


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


525 Collectively, these findings indicate that increased CP content alone should not be interpreted as


526 an overall improvement in forage nutritional quality. Rather, salinity induced cultivar-specific


527 changes in protein, fiber, and non-structural carbohydrate fractions, which may have different


528 implications for forage utilization. Therefore, cultivar selection for saline environments should


529 integrate salt tolerance, biomass production, and a comprehensive assessment of forage


530 nutritional quality to identify genotypes capable of sustaining both productivity and nutritional


531 value under salinity stress.


532 Comparative evaluations of plant germplasm have shown that salt- tolerant genotypes


533 tend to limit Na � accumulation and maintain higher foliar K � concentrations, thereby


534 preserving ionic homeostasis under salinity (Wang et al., 2026; Feng et al., 2025; Yin et al.,


535 2023; Li et al., 2022). This pattern is consistent with findings reported in different crops,


536 including pearl millet ( _Pennisetum glaucum_ L.), where genotypic variation in ionic regulation


537 has been associated with differences in salinity tolerance (Rashid et al., 2026). Similarly, studies


538 investigating stress mitigation through exogenous silicon application have highlighted the


539 importance of maintaining a favorable leaf K � /Na � ratio as a component of improved salt


540 tolerance (Meng et al., 2020). Consistent with these established mechanisms, the differential


541 capacity for photochemical stability observed among cultivars studied here was intimately


542 linked to their distinct leaf ion accumulation profiles. Although foliar Na � increased across all


543 evaluated cultivars, the excessive Na � load accumulated by specific cultivars likely drove the


544 persistent photochemical impairment recorded at 14 dpt. Because toxic Na � accumulation is


545 known to directly disrupt chloroplast function and compromise thylakoid membrane integrity


546 (Deinlein _et al._, 2014; van Zelm _et al._, 2020), this ion-induced damage provides a clear


547 mechanistic basis for the sustained decline in PSII efficiency observed in the less tolerant lines.


548 Our results suggest that salt tolerance was more closely associated with the maintenance of


549 relative ionic balance (Na � /K � ratio) than with K � accumulation alone (Chen _et al_ . 2007;


550 Shabala &amp; Cuin 2008). In this regard, Monarca and Prosementi exhibited substantial increases


551 in K � concentration, but their inability to restrict Na � accumulation resulted in unfavorable


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


552 Na � /K � ratios. In contrast, the exceptionally high K � accumulation observed in MSI037


553 appeared sufficient to partially offset Na � accumulation and maintain an ionic balance. These


554 differences in ionic homeostasis were accompanied by contrasting carbohydrate profiles.


555 Kumen and Chenini maintained or increased root soluble sugars while preserving growth,


556 whereas Monarca and MSI036 showed marked reductions in root soluble sugars and biomass


557 under salinity. Moreover, the reduction in root starch observed in Kumen, Chenini, and


558 Ameristand 801S may reflect mobilization of carbohydrate reserves under stress (Thalmann &amp;


559 Santelia 2017; Dong &amp; Beckles 2019). Together, these results suggest that salt tolerance


560 involves coordinated adjustments in ionic homeostasis and carbohydrate metabolism,


561 potentially contributing to osmotic adjustment and the maintenance of growth under salinity


562 (Krasensky &amp; Jonak 2012).


563 Previous multivariate analyses in _Arabidopsis thaliana_ identified non-photochemical


564 quenching as an important contributor to early salt-stress responses, whereas quantum yield was


565 more strongly associated with long-term performance (Awlia _et al_ ., 2016). Consistent with an


566 early perturbation of photosynthetic performance, our time-resolved ChlF measurements


567 revealed reductions in Phi2, ETo/RC, and LEF at 1 dpt in several alfalfa cultivars. Although


568 most of these early photochemical changes were not significantly associated with final biomass,


569 PhiNPQ measured at 1 dpt showed a positive correlation with subsequent biomass


570 accumulation. This association is consistent with a potential photoprotective role of regulated


571 energy dissipation during the initial phase of salt stress, as proposed for photosynthetic


572 acclimation mechanisms (Ruban 2016). By 7 dpt, clearer associations emerged between


573 photochemical performance and final biomass (Fig. 5). At this stage, higher PI total and F � ’/F � ’


574 values were positively associated with final biomass, indicating that the preservation of


575 photosynthetic performance was linked to greater biomass retention under salinity. Conversely,


576 less tolerant cultivars showed greater photochemical impairment, accompanied by sustained


577 increases in regulated and non-regulated energy dissipation, as indicated by elevated NPQt,


578 PhiNPQ, and PhiNO. Collectively, these findings demonstrate that dynamic ChlF parameters—


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


579 specifically intermediate �� total (7 dpt) and late � � �/� � � (14 dpt)—do not act as isolated


580 univariate drivers, but as core functional nodes within a multivariate response network. Their


581 strong alignment with the biomass-predictive PC1 axis ( ��0.800, ��0.0096 ) validates these


582 non-destructive photochemical traits as robust candidate indicators for salinity tolerance


583 screening (Feng _et al._ 2025).


584 Although this study provides insights into the interplay among photosynthetic


585 performance, ionic homeostasis, and carbon partitioning in alfalfa under salinity, it was


586 conducted under controlled experimental conditions. As expected, the cultivar ranking for salt


587 tolerance observed here did not fully match that reported in multi-year field experiments


588 (Pecetti _et al_ ., 2024), highlighting the influence of environmental conditions on genotype


589 performance. Therefore, further studies under field conditions, where salinity and other


590 environmental factors fluctuate, are needed to assess the consistency of these responses. In


591 particular, the potential of dynamic ChlF traits, including PI total, LEF, and F � ’/F � ’, as


592 indicators of salt tolerance should be evaluated across larger and more diverse germplasm


593 collections under both managed and naturally saline conditions.


594 Overall, our findings indicate that variation in alfalfa salt tolerance is associated with


595 the coordinated maintenance of photochemical performance, ionic balance, and organ-specific


596 carbohydrate adjustments over the course of stress exposure. Ameristand 801S, MSI038,


597 Chenini, and Kumen displayed the most robust short-term response profiles, combining better


598 biomass maintenance with greater photochemical stability. Our integrative findings demonstrate


599 that intermediate photochemical stability (7 dpt) provides a reliable physiological signature,


600 distinguishing tolerant germplasm prior to the onset of severe visual symptoms. Integrating


601 these physiological indicators into broader phenotyping strategies could improve the


602 identification and selection of alfalfa germplasm with enhanced adaptation to saline


603 environments, while field validation remains an essential step toward their practical application


604 in breeding programs.


605


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


606 **Supplementary data**


607 Fig. S1. Experimental set-up.


608 Fig. S2. Contrasting shoot phenotypes of alfalfa cultivars under salinity.


609 Fig. S3. Temporal photochemical responses of nine alfalfa cultivars under salinity.


610 Fig. S4. Correlation matrix of physiological, photochemical, ionic and biochemical variables.


611 Table S1. Chlorophyll fluorescence parameters.


612 Table S2. Effect of salt stress on leaf ion accumulation and Na � /K � ratio in alfalfa cultivars.


613 Table S3. Organ-specific sugar and starch partitioning under salinity.


614


615 **Acknowledgments**


616 We thank Daniela Ortiz (INTA Anguil) for generating the reference data, performing the wet


617 chemistry analyses, and developing the NIRS calibration models for alfalfa. We also thank


618 Mónica Cornacchione, Ariel Odorizzi, and Luciano Pecetti for providing alfalfa seeds, and


619 Paola Suarez for technical assistance.


620


621 **Author contributions**


622 MR conceptualization. GB and MR designed and performed the experiments. GB, MC


623 methodology. GB, MC, FL, PA and MR formal analysis. GB and MR created and edited the


624 final Figures. GB, MR writing - original draft. GB, MC, FL, PA, and MR writing - review &amp;


625 editing. MR: supervision; MR: funding acquisition.


626


627 **Conflict of interest**


628 No conflict of interest declared.


629


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


630 **Funding**


631 This work was supported by Agencia Nacional de Promoción Científica y Tecnológica


632 (ANPCyT, grant PICT2021-00229), Consejo Nacional de Investigaciones Científicas y


633 Técnicas (CONICET grant PUE-UDEA2018), Instituto Nacional de Tecnología Agropecuaria


634 (grant INTA-2023-PD-I084; INTA-2023-PD-I100).


635


636 **Data availability**


637 The data that support the findings of this study are available from the corresponding author


638 upon reasonable request.


639


640 **6. References.**


641 1. Anderson A.W., Gull U., Benes S.E., Singh S., Hutmacher R.B., Brummer E.C.,


642 Putnam D.H. (2023) Salinity and cultivar effects on alfalfa forage yield and nutritive


643 value in a Mediterranean climate. _Grassland_ _Research_, **2**, 153–166.


644 https://doi.org/10.1002/glr2.12061


645 2. Annicchiarico P., Barrett B., Brummer E.C., Julier B., Marshall A.H. (2015)


646 Achievements and challenges in improving temperate perennial forage legumes.


647 _Critical_ _Reviews_ _in_ _Plant_ _Sciences_, **34**, 327–380.


648 https://doi.org/10.1080/07352689.2014.898462


649 3. Annicchiarico P., Pecetti L., Abdelguerfi A., Bouizgaren A., Carroni A.M., Hayek T.,


650 M’Hammadi Bouzina M., Mezni M. (2011) Adaptation of landrace and variety


651 germplasm and selection strategies for lucerne in the Mediterranean basin. _Field Crops_


652 _Research_, **120**, 283–291. https://doi.org/10.1016/j.fcr.2010.11.003


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


653 4. AOAC (1990) _Official Methods of Analysis of the Association of Official Analytical_


654 _Chemists_ . 15th ed. Helrich K. (Ed.). Association of Official Analytical Chemists,


655 Arlington, VA, USA.


656 5. Awlia M., Nigro A., Fajkus J., Schmoeckel S.M., Negrão S., Santelia D., Trtílek M.,


657 Tester M., Julkowska M.M., Panzarová K. (2016) High-throughput non-destructive


658 phenotyping of traits that contribute to salinity tolerance in Arabidopsis thaliana.


659 _Frontiers in Plant Science_, **7**, 1414. https://doi.org/10.3389/fpls.2016.01414


660 6. Baker N.R. (2008) Chlorophyll fluorescence: a probe of photosynthesis in vivo. _Annual_


661 _Review_ _of_ _Plant_ _Biology_, **59**, 89–113.


662 https://doi.org/10.1146/annurev.arplant.59.032607.092759


663 7. Broughton W.J., Dilworth M.J. (1971) Control of leghaemoglobin synthesis in snake


664 beans. _Biochemical Journal_, **125**, 1075–1080. https://doi.org/10.1042/bj1251075


665 8. Carter P.R., Sheaffer C.C. (1983) Alfalfa response to soil water deficits. I. Growth,


666 forage quality, yield, water use, and water-use efficiency. _Crop Science_, **23**, 669–675.


667 https://doi.org/10.2135/cropsci1983.0011183x002300040016x


668 9. Chaves M.M., Flexas J., Pinheiro C. (2009) Photosynthesis under drought and salt


669 stress: regulation mechanisms from whole plant to cell. _Annals of Botany_, **103**, 551–


670 560. https://doi.org/10.1093/aob/mcn125


671 10. Chen Z., Zhou M., Newman I.A., Mendham N.J., Zhang G., Shabala S. (2007)


672 Potassium and sodium relations in salinised barley tissues as a basis of differential salt


673 tolerance. _Functional Plant Biology_, **34**, 150–162. https://doi.org/10.1071/FP06237


674 11. Cornacchione M.V., Suarez D.L. (2015) Emergence, forage production, and ion


675 relations of alfalfa in response to saline waters. _Crop Science_, **55**, 444–457.


676 https://doi.org/10.2135/cropsci2014.01.0062


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


677 12. Cornacchione M.V., Suarez D.L. (2017) Evaluation of alfalfa ( _Medicago sativa_ L.)


678 populations’ response to salinity stress. _Crop_ _Science_, **57**, 137–150.


679 https://doi.org/10.2135/cropsci2016.05.0371


680 13. Daud M., Li W., Qiao H., Adil M., Ashraf M.A., Khan M., Ijaz H., Haq Z.U., Bibi N.,


681 Xu S. (2026) Assessment of alfalfa varieties under salt stress through morpho

682 physiological and biochemical traits using a comprehensive membership function


683 approach. _BMC Plant Biology_, **26,** 1097. https://doi.org/10.1186/s12870-026-09157-7


684 14. Deinlein U., Stephan A.B., Horie T., Luo W., Xu G., Schroeder J.I. (2014) Plant salt

685 tolerance mechanisms. _Trends_ _in_ _Plant_ _Science_, **19**, 371–379.


686 https://doi.org/10.1016/j.tplants.2014.02.001


687 15. Díaz F.J., Grattan S.R., Reyes J.A., De La Roza-Delgado B., Benes S.E., Jiménez C.,


688 Dorta M., Tejedor M. (2018) Using saline soil and marginal quality water to produce


689 alfalfa in arid climates. _Agricultural_ _Water_ _Management_, **199**, 11–21.


690 https://doi.org/10.1016/j.agwat.2017.12.003


691 16. Di Rienzo J.A., Casanoves F., Balzarini M.G., Gonzalez L., Tablada M., Robledo C.W.


692 (2019) InfoStat versión 2019. Grupo InfoStat, Facultad de Ciencias Agropecuarias,


693 Universidad Nacional de Córdoba, Córdoba, Argentina.


694 17. Dong S., Beckles D.M. (2019) Dynamic changes in the starch-sugar interconversion


695 within plant source and sink tissues promote a better abiotic stress response. _Journal of_


696 _Plant Physiology_, **234–235**, 80–93. https://doi.org/10.1016/j.jplph.2019.01.007


697 18. Fales F.W. (1951) The assimilation and degradation of carbohydrates by yeast cells.


698 _Journal of Biological Chemistry_, **193**, 113–124. https://doi.org/10.1016/s0021

699 9258(19)52433-4


700 19. FAO (2024) _Global Status of Salt-Affected Soils: Main Report_ . Food and Agriculture


701 Organization of the United Nations, Rome, Italy. https://doi.org/10.4060/cd3044en


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


702 20. Feng Y., Chen Z., Chen L., Han M., Liu J., Liu Y., Chai R., Wang J., Sun S., Fan J.,


703 Yan X., Guo Y. (2025) Comprehensive evaluation of physio-morphological traits of


704 alfalfa ( _Medicago sativa_ L.) varieties under salt stress. _Physiologia Plantarum_, **177**,


705 e70044. https://doi.org/10.1111/ppl.70044


706 21. Ferreira J., Cornacchione M., Liu X., Suarez D. (2015) Nutrient composition, forage


707 parameters, and antioxidant capacity of alfalfa ( _Medicago sativa_ L.) in response to


708 saline irrigation water. _Agriculture_, **5**, 577–597.


709 https://doi.org/10.3390/agriculture5030577


710 22. Halim R.A., Buxton D.R., Hattendorf M.J., Carlson R.E. (1989) Water-deficit effects on


711 alfalfa at various growth stages. _Agronomy_ _Journal_, **81**, 765–770.


712 https://doi.org/10.2134/agronj1989.00021962008100050014x


713 23. Krasensky J., Jonak C. (2012) Drought, salt, and temperature stress-induced metabolic


714 rearrangements and regulatory networks. _Journal of Experimental Botany_, **63**, 1593–


715 1608. https://doi.org/10.1093/jxb/err460


716 24. Kuhlgert S., Austic G., Zegarac R., Osei-Bonsu I., Hoh D., Chilvers M.I., Roth M.G.,


717 Bi K., TerAvest D., Weebadde P., Kramer D.M. (2016) MultispeQ Beta: a tool for


718 large-scale plant phenotyping connected to the open PhotosynQ network. _Royal Society_


719 _Open Science_, **3**, 160592. https://doi.org/10.1098/rsos.160592


720 25. Le Gall H., Philippe F., Domon J.-M., Gillet F., Pelloux J., Rayon C. (2015) Cell wall


721 metabolism in response to abiotic stress. _Plants_, **4**, 112–166.


722 https://doi.org/10.3390/plants4010112


723 26. Li J., Ma M., Sun Y., Lu P., Shi H., Guo Z., Zhu H. (2022) Comparative physiological


724 and transcriptome profiles uncover salt tolerance mechanisms in alfalfa. _Frontiers in_


725 _Plant Science_, **13**, 931619. https://doi.org/10.3389/fpls.2022.931619


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


726 27. Ma L., Li J., Li J., Huo Y., Yang Y., Jiang C., Guo Y. (2026) Plant salt-tolerance


727 mechanisms: classic signaling pathways, emerging frontiers, and future perspectives.


728 _Molecular Plant_, **19**, 538–570. https://doi.org/10.1016/j.molp.2025.12.009


729 28. Meng Y., Yin Q., Yan Z., Wang Y., Niu J., Zhang J., Fan K. (2020) Exogenous silicon


730 enhanced salt resistance by maintaining K � /Na � homeostasis and antioxidant


731 performance in alfalfa leaves. _Frontiers_ _in_ _Plant_ _Science_, **11**, 1183.


732 https://doi.org/10.3389/fpls.2020.01183


733 29. Moore K.J., Hatfield R.D. (1994) Carbohydrates and forage quality. In: Fahey G.C. Jr.


734 (Ed.), Forage Quality, Evaluation, and Utilization. American Society of Agronomy,


735 Crop Science Society of America, Soil Science Society of America, Madison, WI,


736 USA: 229–280. https://doi.org/10.2134/1994.foragequality.c6


737 30. Munns R., Millar A.H. (2023) Seven plant capacities to adapt to abiotic stress. _Journal_


738 _of Experimental Botany_, **74**, 4308–4323. https://doi.org/10.1093/jxb/erad179


739 31. Murchie E.H., Lawson T. (2013) Chlorophyll fluorescence analysis: a guide to good


740 practice and understanding some new applications. _Journal of Experimental Botany_, **64**,


741 3983–3998. https://doi.org/10.1093/jxb/ert208


742 32. Pantsar-Kallio M., Manninen P.K.G. (1995) Determination of sodium, potassium,


743 calcium and magnesium cations by capillary electrophoresis compared with ion


744 chromatography. _Analytica Chimica Acta_, **314**, 67–75. https://doi.org/10.1016/0003

745 2670(95)00266-3


746 33. Pecetti L., Annicchiarico P., Scotti C., Paolini M., Nanni V., Palmonari A. (2017)


747 Effects of plant architecture and drought stress level on lucerne forage quality. _Grass_


748 _and Forage Science_, **72**, 714–722. https://doi.org/10.1111/gfs.12272


749 34. Pecetti L., Tlahig S., Confalonieri M., Cornacchione M.V., Hayek T., Angueira S.P.,


750 Annicchiarico P. (2024) A comparison of procedures for evaluating and selecting alfalfa


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


751 landrace germplasm for tolerance to salinity. _Crop Science_, **64**, 2143–2157.


752 https://doi.org/10.1002/csc2.21258


753 35. Pérez-Reverón R., Grattan S.R., Perdomo-González A., Pérez-Pérez J.A., Díaz-Peña


754 F.J. (2024) Marginal quality waters: adequate resources for sustainable forage


755 production in saline soils? _Agricultural Water Management_, **305**, 109142.


756 https://doi.org/10.1016/j.agwat.2024.109142


757 36. Rashid M., Amjad N., Shah K.H. (2026) Chlorophyll a fluorescence kinetics uncover


758 differential photosystem II response and salinity tolerance mechanisms in pearl millet


759 (Pennisetum glaucum L.). _BMC Plant Biology_, **26**, 277. https://doi.org/10.1186/s12870

760 026-08104-w


761 37. Ribeiro C., Stitt M., Hotta C.T. (2022) How stress affects your budget—stress impacts


762 on starch metabolism. _Frontiers_ _in_ _Plant_ _Science_, **13**, 774060.


763 https://doi.org/10.3389/fpls.2022.774060


764 38. Ruban A.V. (2016) Nonphotochemical chlorophyll fluorescence quenching: mechanism


765 and effectiveness in protecting plants from photodamage. _Plant Physiology_, **170**, 1903–


766 1916. https://doi.org/10.1104/pp.15.01935


767 39. Saddhe A.A., Manuka R., Penna S. (2021) Plant sugars: homeostasis and transport


768 under abiotic stress in plants. _Physiologia_ _Plantarum_, **171**, 739–755.


769 https://doi.org/10.1111/ppl.13283


770 40. Sandhu D., Cornacchione M.V., Ferreira J.F.S., Suarez D.L. (2017) Variable salinity


771 responses of 12 alfalfa genotypes and comparative expression analyses of salt-response


772 genes. _Scientific Reports_, **7**, 42958. https://doi.org/10.1038/srep42958


773 41. Shabala S., Cuin T.A. (2008) Potassium transport and plant salt tolerance. _Physiologia_


774 _Plantarum_, **133**, 651–669. https://doi.org/10.1111/j.1399-3054.2007.01008.x


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


775 42. Smethurst C.F., Rix K., Garnett T., Auricht G., Bayart A., Lane P., Wilson S.J., Shabala


776 S. (2008) Multiple traits associated with salt tolerance in lucerne: revealing the


777 underlying cellular mechanisms. _Functional_ _Plant_ _Biology_, **35**, 640–650.


778 https://doi.org/10.1071/FP08030


779 43. Strasser R.J., Tsimilli-Michael M., Srivastava A. (2004) Analysis of the chlorophyll _a_


780 fluorescence transient. In: Papageorgiou G.C., Govindjee (Eds), _Chlorophyll a_


781 _Fluorescence: A Signature of Photosynthesis_ . Advances in Photosynthesis and


782 Respiration, Vol. 19. Springer, Dordrecht, The Netherlands: 321–362.


783 https://doi.org/10.1007/978-1-4020-3218-9_12


784 44. Sumner J.B., Somers G.F. (1944) _Laboratory Experiments in Biological Chemistry_ .


785 Academic Press, New York, NY, USA.


786 45. Thalmann M., Santelia D. (2017) Starch as a determinant of plant fitness under abiotic


787 stress. _New Phytologist_, **214**, 943–951. https://doi.org/10.1111/nph.14491


788 46. Vanholme R., De Meester B., Ralph J., Boerjan W. (2019) Lignin biosynthesis and its


789 integration into metabolism. _Current Opinion in Biotechnology_, **56**, 230–239.


790 https://doi.org/10.1016/j.copbio.2019.02.018


791 47. Van Soest P.J. (1994) _Nutritional Ecology of the Ruminant_ . 2nd ed. Cornell University


792 Press, Ithaca, NY, USA.


793 48. Van Zelm E., Zhang Y., Testerink C. (2020) Salt tolerance mechanisms of plants.


794 _Annual Review of Plant Biology_, **71**, 403–433. https://doi.org/10.1146/annurev-arplant

795 050718-100005


796 49. Wang Y., Wu W., Liang F., Zhao T., Li S., Yang M., Leng F., Zhu X., Wang X. (2026)


797 Integrated multi-omics reveals divergent salt stress response mechanisms in three alfalfa


798 (Medicago sativa L.) cultivars. _Plant_ _Science_, **370**, 113216.


799 https://doi.org/10.1016/j.plantsci.2026.113216


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


800 50. Yin Y.L., Fan S.G., Li S., Amombo E., Fu J.M. (2023) Involvement of cell cycle and


801 ion transferring in the salt stress responses of alfalfa varieties at different development


802 stages. _BMC Plant Biology_, **23**, 343. https://doi.org/10.1186/s12870-023-04335-3


803 51. Zhang T., Kesoju S., Greene S.L., Fransen S., Hu J., Yu L.-X. (2018) Genetic diversity


804 and phenotypic variation for drought resistance in alfalfa ( _Medicago sativa_ L.)


805 germplasm collected for drought tolerance. _Genetic Resources and Crop Evolution_, **65**,


806 471–484. https://doi.org/10.1007/s10722-017-0546-9


807 52. Zhang Y., Zhang Y., Wang C., Xiao J., Huang M., Zhuo L., Zhang D. (2024)


808 Enhancement of salt tolerance of alfalfa: physiological and molecular responses of


809 transgenic alfalfa plants expressing _Syntrichia caninervis_ -derived ScABI3. _Plant_


810 _Physiology_ _and_ _Biochemistry_, **207**, 108335.


811 https://doi.org/10.1016/j.plaphy.2024.108335


812


813 **Tables**






















&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)









































814 **Table 1** . **Effect of salt stress on growth traits of alfalfa cultivars.** Leaf number, shoot dry


815 weight, shoot length, root dry weight, root length and root:shoot ratio measured in nine alfalfa


816 cultivars grown under control conditions and exposed to 200 mM NaCl. Values represent means


817 ± SE. Asterisks indicate significant differences between plants under saline treatment and their


818 respective controls, based on three independent experiments with thirty plants per treatment


819 (n=3), analyzed using ANOVA followed by Tukey’s test (p &lt; 0.05).


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


820 **Table 2.** Salinity-induced shifts in forage nutritive value among alfalfa cultivars. Crude protein


821 (CP), ash, acid detergent lignin (ADL), acid detergent fiber (ADF), and neutral detergent fiber


822 (NDF) were estimated by near-infrared reflectance spectroscopy (NIRS) in nine alfalfa cultivars


823 grown under control conditions or exposed to 200 mM NaCl. Non-structural carbohydrates


824 (NSC) were calculated as the sum of independently measured soluble sugar and starch contents.


825 Values represent means ± SE from three independent experiments, each comprising 30 plants


826 per treatment pooled into a single sample (n = 3). Asterisks indicate significant differences


827 between NaCl-treated plants and their respective controls, as determined by ANOVA followed


828 by Tukey’s test ( _P_ &lt; 0.05).


829


830 **Figure legends**


831 **Fig. 1. Biomass retention of alfalfa cultivars under salinity stress.** Biomass retention was


832 calculated as the ratio between shoot dry weight under 200 mM NaCl and the corresponding


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


833 control condition (NaCl/Control), expressed as a percentage. Cultivars are ranked from highest


834 to lowest biomass retention. The dashed horizontal line indicates the control level (100%),


835 representing no reduction in shoot biomass under salinity. Higher biomass retention indicates a


836 greater ability to maintain biomass production under salt stress. The gradient from green to red


837 represents decreasing tolerance to salinity. Values were calculated from three biological


838 replicates per cultivar and treatment.


839 **Fig. 2. Effect of salinity on leaf ion concentrations in alfalfa cultivars.** (A) Sodium (Na � )


840 concentration, (B) potassium (K � ) concentration, and (C) Na � :K � ratio in leaves of nine


841 alfalfa cultivars under control conditions and 200 mM NaCl. Data are presented as means ± SE


842 from three independent biological experiments, with each biological replicate consisting of a


843 pool of three plants. Asterisks indicate significant differences between control and NaCl


844 treatments within each cultivar, whereas different lowercase letters indicate significant


845 differences among cultivars under NaCl treatment. Statistical differences were determined by


846 two-way ANOVA with treatment and cultivar as factors, followed by Tukey’s post-hoc test ( _P_ &lt;


847 0.05).


848 **Fig. 3. Organ-specific carbohydrate adjustments in alfalfa cultivars under salinity.** Soluble


849 sugar-to-starch ratio in (A) leaves and (B) roots of nine alfalfa cultivars grown under control


850 conditions or exposed to 200 mM NaCl for 14 days. Values represent means ± SE. Asterisks


851 indicate significant differences between plants under saline treatment and their respective


852 controls, based on three independent experiments with three plants per treatment, grouped into


853 one pool (n=3), analyzed using ANOVA followed by Tukey’s test (p &lt; 0.05).


854 **Fig. 4. Spearman correlation matrix of variables involved in the most robust associations**


855 **detected across cultivars.** Spearman correlation matrix showing the relationships among the 12


856 variables involved in the seven pairwise associations that remained significant after Benjamini–


857 Hochberg FDR correction in the complete correlation analysis comprising 52 measured traits


858 and shoot biomass retention (53 variables in total). The matrix includes relationships among the


859 selected variables, allowing visualization of the broader correlation structure. Correlation


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


860 coefficients ( ρ ) are represented by a diverging color scale centered at zero, with blue indicating


861 negative and red indicating positive associations. Significance annotations correspond to FDR

862 adjusted _P_ values.


863 **Fig. 5. Correlation between photochemical traits at 7 dpt and final shoot biomass under**


864 **salinity.** Spearman rank correlations between the response ratio (RR; NaCl/Control) of shoot


865 dry weight per plant and the RR of (A) PItotal, (B) F � ’/F � ’, (C) PhiNPQ, and (D) PhiNO


866 measured at 7 dpt. Each point represents one alfalfa cultivar (n = 9). Response ratios were


867 calculated as the mean value under 200 mM NaCl divided by the corresponding control mean


868 for each cultivar. Spearman’s ρ and associated _P_ -values are indicated in each panel.


869 **Fig. 6. Integrated physiological responses of nine alfalfa cultivars to salinity.** Heatmap


870 summarizing cultivar responses to 200 mM NaCl across photochemical performance and


871 electron transport, non-photochemical energy dissipation, growth traits, ion concentrations, non

872 structural carbohydrates, and forage-quality parameters. Heatmap values represent log � fold


873 changes (NaCl/Control), with blue indicating decreases and red indicating increases relative to


874 the corresponding control. Photochemical parameters are shown at 1, 7, and 14 dpt where


875 applicable. Horizontal bars indicate final shoot biomass retention under salinity, expressed as a


876 percentage of the corresponding control. Cultivars are ordered from highest to lowest biomass


877 retention to facilitate comparison between integrated physiological responses and salinity


878 tolerance. Asterisks indicate significant differences between NaCl-treated plants and their


879 corresponding controls within each cultivar and sampling time, as determined by ANOVA


880 followed by Tukey’s post hoc test (P &lt; 0.05).


881 **Fig. 7. Principal component analysis of cultivar-level salinity responses and association**


882 **with biomass retention in alfalfa.** **(A)** Principal component analysis (PCA) of 52


883 physiological, photochemical, ionic, and biochemical variables across nine alfalfa cultivars.


884 Salinity responses were expressed as log � fold changes (NaCl/Control), and variables were


885 standardized before analysis. PC1 and PC2 explained 34.1% and 21.1% of the total variance,


886 respectively, together accounting for 55.2%. Cultivar positions reflect similarities and


&#91;bioRxiv preprint doi: https://doi.org/10.64898/2026.11.18.767659; this version posted November 19, 2026. The copyright holder for this preprint&#93;(https://doi.org/10.64898/2026.11.18.767659)

(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


887 differences in their integrated responses to salinity. **(B)** Relationship between PC1 scores and


888 final shoot biomass retention, calculated as shoot dry weight under 200 mM NaCl divided by


889 the corresponding control value. The association was assessed using Spearman’s rank


890 correlation coefficient ( ρ ). Each point represents one cultivar ( _n_ = 9). The fitted line is included


891 to visualize the overall trend.


&#91;available under a CC-BY-NC-ND 4.0 International license.&#93;(http://creativecommons.org/licenses/by-nc-nd/4.0/)


