Leaf hydraulic traits are considered the key determinants of gas exchange and therefore affect species distributions along environmental gradients, but the patterns of leaf hydraulic traits and their associations with gas exchange across altitudinal gradients remain largely unknown. Here, we measured leaf hydraulic traits, gas exchange, leaf anatomical traits and plant size traits in two dominant alpine shrubs (Caragana jubata and Salix gilashanica) across an altitudinal gradient from 3100 to 3700 m. The findings indicated that with increasing altitude, both shrub species exhibited an increase in leaf hydraulic safety (more negative K_leaf P₅₀), a decrease in leaf hydraulic efficiency (K_leaf-max) and an increase in intrinsic water use efficiency (WUEi), thus allowing them to adapt to higher altitude habitats. The more negative K_leaf P₅₀ was associated with a greater ratio of major to minor vein density (VLA_maj/VLA_min), the lower K_leaf-max was associated with a lower minor vein density (VLA_min), and greater increase in WUEi arose from the small decrease in the photosynthetic rate relative to the stomatal conductance. However, C. jubata was consistent with the ‘hydraulic safety strategy’ with a great decrease in K_leaf P₅₀, a small decrease in K_leaf-max, and a small increase in WUEi along with decreasing plant height and leaf area with increasing altitude. Whereas S. gilashanica was consistent with the ‘photosynthetic efficiency strategy’ with a small decrease in K_leaf P₅₀, a greater decrease in K_leaf-max, and a greater increase in WUEi along with an increasing plant height and unchanged leaf area with increasing altitude. These findings provide new insights to improve understanding how shift in leaf hydraulic traits and their associations with gas exchange and plant size allow plants to adapt to high-altitude habitats.

Study site and shrub species

This study was conducted in the Pailugou catchment of the Qilian Mountains (100°17´E, 38°24´N), Gansu Province, Northwest China. The altitude of this catchment ranges from 2700 m to 3700 m. Meteorological variables, including the growing season temperature (GST, °C), growing season precipitation (GSP, mm) and relative humidity (RH, %) were recorded by an automatic forest ecological station (the four monitoring points are at 2700 m, 2900 m, 3300 m and 3700 m) from 2018-2021. The vapor pressure deficit (VPD, kPa) was calculated via the following equations based on the meteorological data:VPD=0.61078×e^{(17.27×GST)/ (GST+237.3)} ×(1-RH)(Tang et al., 2023). The GST decreased from 11.5°C to 6.0°C, the GSP decreased from 337.1 mm to 153.1 mm, the RH ranged from 57.4% mm to 64.6% and the VPD ranged from 0.58 kPa to 0.33 kPa from 2700 m to 3700 m during the growing season (see Figure S1 in Supporting Information).

The vegetation types are semiarid mountain steppe, mountain forest and alpine shrub meadow from 2700 m to 3700 m. Shrubbery is mainly distributed in the ecologically fragile area at an altitude of 3100–3700 m, accounting for two-thirds of the vegetation area in the Qilian Mountains. This area plays an important role in water conservation and maintaining ecological security and regional high-quality development in the Hexi Corridor. The two investigated shrub species (C. jubata and S. gilashanica) are naturally widespread in the Qilian Mountains and are distributed at altitudes ranging from 3100 m to 3600 m and 3100 m to 3700 m, respectively.

Experimental design

Seven sampling sites from 3100 m to 3700 m at 100 m intervals were set in July 2021, and three sampling plots (5 m × 5 m) were set up at each sampling site. All plots were similar in slope (32.6±1.9) and aspect (north slope). In three sampling plots at each sample site, ten fully expanded sun leaves were selected from ten individuals of C. jubata and S. gilashanica for gas exchange measurements at each altitude. The five leaves close to the measured leaf gas exchange were collected and placed in a formaldehyde-acetic acid-90% ethanol fixative (5:5:90, vol: vol: vol) for morphological and anatomical structure measurements. Five to ten leafy branches from each sampling plot were randomly selected and cut under pure water at nightfall, wrapped with black plastic bags to avoid leaf transpiration, and immediately transported to the forest ecological station, it is a national station, allowing field investigation and sampling. All excised branches were rehydrated for 2–4 hours and then used to construct leaf hydraulic vulnerability curves.

Gas exchange and leaf water potential measurements

The saturated photosynthesis (A_sat) and stomatal conductance (g_s) were measured between 09:00 and 11:00 a.m. in ten replicate individuals per sample site with a portable open gas exchange system (LI-6800; LI-COR, Lincoln, NE, USA). The leaf chamber was supplied with 1200 μmol m^-2 s^-1 photosynthetic photon flux density, the leaf temperature was set at 25°C, the CO₂ concentration was set at 400 μmol mol⁻¹, the VPD was set at approximately 1.1–1.4 kPa, and the relative humidity was maintained at 50%. The intrinsic water use efficiency (WUEi) was calculated as the ratio between A_sat and g_s according to previous studies (Klein et al., 2013; Yang et al., 2021).

Leaf water potential (ψ_gas) was measured on the same individuals from leaves close to those used for the photosynthetic measurement between 09:00 and 11:00 via a pressure chamber (Model 1000, PMS Instrument Company, Albany, OR). Briefly, leaves were collected and immediately placed in a sealed black plastic bag, previously wrapped in soaked paper towel to create a humid environment. Following 20 min of equilibration, ψ_gas was measured (Rodriguez-Dominguez et al., 2022).

Leaf hydraulic trait measurements

Leaf hydraulic conductance (K_leaf) was measured via the two-point rehydration method (Nardini et al., 2012; Ocheltree et al., 2016), which is based on the following formula:

K_leaf=Cln(ψ_0/ψ_f)/t

where C is the leaf capacitance, ψ₀ is the initial leaf water potential, ψ_f is the final leaf water potential and t is the duration of rehydration. The air temperature and relative humidity were 22–26℃ and 42%–55%, respectively. Briefly, the leafy branches were cut into segments with at least three leaves under ultrapure water and allowed to desiccate slowly before being carefully bagged to prevent water loss and maintain water potential equilibrium throughout the segments. The initial leaf water potential (ψ₀) was determined by measuring leaves neighboring the sample leaf with a Scholander pressure chamber (PMS, Albany, OR, USA). The sample leaf was then cut under water and allowed to rehydrate for a period of between 15 and 300 s depending on ψ₀, after which their petioles were immediately dabbed dry, and the leaf was wrapped in moist paper towel and double bagged for 10 min to allow water potentials to equilibrate throughout the leaf before ψ_f determination. For each sample, K_leaf was calculated via the corresponding formula. Leaf capacitance (C) was measured from five leaves of each species via the slope of the pressure–volume relationship for each species. The leaf hydraulic vulnerability curve was generated by plotting K_leaf against ψ₀, K_leaf–max was calculated when ψ₀ was equal to -0.3 MPa, and K_leaf P₅₀ was calculated from the leaf hydraulic vulnerability curve as the ψ_leaf value at which 50% of K_leaf was lost from K_leaf–max (Nardini et al., 2012; Scoffoni et al., 2016).

Leaf vein density and stomatal density measurements

The leaf vein density (VLA) was measured according to the conventional method (Xiong & Flexas, 2022). The densities of the 1st–3rd veins are considered major VLAs (VLA_maj), and veins higher than the third order are considered minor VLAs (VLA_min) (Kawai & Okada, 2016). Briefly, the leaf samples were boiled for 20 min and then placed in 5% NaOH in aqueous solution with water bath heating (keeping the leaf in it for 40 min) until the leaf cuticle had dissolved. The leaves were then flushed with water until NaOH was vacuumed off until they appeared transparent. If the leaves exhibited any color, 5% NaClO was added to the leaves to bleach them for 20 min, after which they were rinsed twice with water. These leaves were brought into EtOH dilution series solutions (30%, 50%, 70%, 100%). Each series lasted 3–5 min. After the 100% EtOH stage, the leaves were soaked in 1% safranin (1 g safranin/100 ml of 100% EtOH) for 15–30 min and gently rinsed with 100% EtOH. The stained leaves were scanned with a scanner (Epson-V800, Japan) at 9,600–12,800 dpi to measure the length of the 1st–3rd veins. Next, the leaves were imaged with a light microscope (Zeiss LSM 880, Germany) at 40× magnification to measure the length of higher-order veins. Leaf areas were measured from leaf images, and the lengths of all veins were measured via ImageJ (version 1.42q; National Institutes of Health, USA) to calculate the VLA_min. Stomatal density (SD) was analyzed from images obtained via light microscopy at 40× magnification of five leaves per species at each sampling site following methods described by Zhao et al., (2016).

Plant and leaf morphology measurements

Thirty leaves of each species in each altitude were randomly selected to measure lead area and leaf mass per area (LMA). Leaves were scanned with a digital scanner at 300 DPI, and leaf area was calculated with ImageJ. Leaf dry mass was determined after dehydration at 80°C for 72 h, and LMA was calculated as the ratio of dry mass to its leaf area (Chen et al., 2021). Plant height was recorded in ten replicates at each altitude from the base of the plant to the tip of the tallest leaf (Lehrer & Hawkins, 2023).

Turgor loss point measurement

The turgor loss point (π_tlp) was measured following the osmometer method (Bartlett et al., 2012). In short, five branches were randomly sampled from each sampling plot and kept humid in opaque bags, recut underwater and then rehydrated overnight to ensure that all measurements were made at full hydration. Several exposed, healthy, fully expanded leaves were randomly sampled from five excised branches and immediately stored in liquid nitrogen for measurement of the leaf osmotic potential at full turgor (ψ_o). ψ₀ was measured at room temperature with a vapor pressure osmometer (VAPRO 5520, Wescor Inc., Logan, Utah) and used to calculate π_tlp with the following equation: π_tlp=0.832ψ₀-0.631(Fuchs et al., 2021). The residual turgor (RT) was calculated with the following equation: RT=ψ_gas-π_tlp (Nardini et al., 2014)_.

Data analysis

The functions fitplc and fitcond were used to extract K_leaf-max and K_leaf P₅₀ from leaf vulnerability curves and to calculate their associated 95% confidence intervals (CIs). The package “fitplc” was used to assess differences in K_leaf-max and K_leaf P₅₀ along the altitudinal gradient (Duursma & Choat, 2017). Differences in plant height (10 individuals at each altitude), leaf area (30 leaves from ten individuals at each altitude), as well as leaf physiological and structural traits (5-10 leaves from five to ten individuals at each altitude) were analyzed using one-way ANOVA (Duncan's test). Prior to this analysis, data were tested for normal distribution and homoscedasticity using SPSS 19.0 (SPSS Inc., Chicago, IL, USA). Trait–altitude correlations were tested to examine how the traits shift with increasing altitude, and then trait–trait correlations were tested to examine the trade-offs between K_leaf-max and K_leaf P₅₀, and also to determine the structural traits behind the shift of K_leaf-max and K_leaf P₅₀ via simple linear functions (SigmaPlot; version 14.0; Systat Software, San Jose, CA, USA). For those analysis, each best regression function was used with the highest r² determined from the original data plots. All data analyses were considered highly significant if P <0.05.

References

Bartlett, M. K., Scoffoni, C., Ardy, R., Zhang, Y., Sun, S., Cao, K., & Sack, L. (2012). Rapid determination of comparative drought tolerance traits: Using an osmometer to predict turgor loss point. Methods in Ecology and Evolution, 3(5), 880–888. https://doi.org/10.1111/j.2041-210X.2012.00230.x

Chen, Y. J., Choat, B., Sterck, F., Maenpuen, P., Katabuchi, M., Zhang, S. Bin, Tomlinson, K. W., Oliveira, R. S., Zhang, Y. J., Shen, J. X., Cao, K. F., & Jansen, S. (2021). Hydraulic prediction of drought-induced plant dieback and top-kill depends on leaf habit and growth form. Ecology Letters, 24(11), 2350–2363. https://doi.org/10.1111/ele.13856

Duursma, R., & Choat, B. (2017). fitplc - an R package to fit hydraulic vulnerability curves. Journal of Plant Hydraulics, 4, e002. https://doi.org/10.20870/jph.2017.e002

Fuchs, S., Leuschner, C., Mathias Link, R., & Schuldt, B. (2021). Hydraulic variability of three temperate broadleaf tree species along a water availability gradient in central Europe. New Phytologist, 231(4), 1387–1400. https://doi.org/10.1111/nph.17448

Kawai, K., & Okada, N. (2016). How are leaf mechanical properties and water-use traits coordinated by vein traits? A case study in Fagaceae. Functional Ecology, 30(4), 527–536. https://doi.org/10.1111/1365-2435.12526

Klein, T., Shpringer, I., Fikler, B., Elbaz, G., Cohen, S., & Yakir, D. (2013). Relationships between stomatal regulation, water-use, and water-use efficiency of two coexisting key Mediterranean tree species. Forest Ecology and Management, 302, 34–42. https://doi.org/10.1016/j.foreco.2013.03.044

Lehrer, M. A., & Hawkins, J. S. (2023). Plant height shapes hydraulic architecture but does not predict metaxylem area under drought in Sorghum bicolor. Plant Direct, 7(5), 1–10. https://doi.org/10.1002/pld3.498

Nardini, A., & Luglio, J. (2014). Leaf hydraulic capacity and drought vulnerability: Possible trade-offs and correlations with climate across three major biomes. Functional Ecology, 28(4), 810–818. https://doi.org/10.1111/1365-2435.12246

Nardini, A., Pedà, G., & Rocca, N. La. (2012). Trade-offs between leaf hydraulic capacity and drought vulnerability: Morpho-anatomical bases, carbon costs and ecological consequences. New Phytologist, 196(3), 788–798. https://doi.org/10.1111/j.1469-8137.2012.04294.x

Ocheltree, T. W., Nippert, J. B., & Prasad, P. V. V. (2016). A safety vs efficiency trade-off identified in the hydraulic pathway of grass leaves is decoupled from photosynthesis, stomatal conductance and precipitation. New Phytologist, 210(1), 97–107. https://doi.org/10.1111/nph.13781

Rodriguez-Dominguez, C. M., Forner, A., Martorell, S., Choat, B., Lopez, R., Peters, J. M. R., Pfautsch, S., Mayr, S., Carins-Murphy, M. R., McAdam, S. A. M., Richardson, F., Diaz-Espejo, A., Hernandez-Santana, V., Menezes-Silva, P. E., Torres-Ruiz, J. M., Batz, T. A., & Sack, L. (2022). Leaf water potential measurements using the pressure chamber: Synthetic testing of assumptions towards best practices for precision and accuracy. Plant Cell and Environment, 45(7), 2037–2061. https://doi.org/10.1111/pce.14330

Scoffoni, C., Chatelet, D. S., Pasquet-Kok, J., Rawls, M., Donoghue, M. J., Edwards, E. J., & Sack, L. (2016). Hydraulic basis for the evolution of photosynthetic productivity. Nature Plants, 2(6), 1–8. https://doi.org/10.1038/nplants.2016.72

Xiong, D., & Flexas, J. (2022). Safety–efficiency tradeoffs? Correlations of photosynthesis, leaf hydraulics, and dehydration tolerance across species. Oecologia, 200(1–2), 51–64. https://doi.org/10.1007/s00442-022-05250-4

Yang, Y. J., Bi, M. H., Nie, Z. F., Jiang, H., Liu, X. D., Fang, X. W., & Brodribb, T. J. (2021). Evolution of stomatal closure to optimize water-use efficiency in response to dehydration in ferns and seed plants. New Phytologist, 230(5), 2001–2010. https://doi.org/10.1111/nph.17278

Zhao, W. L., Chen, Y. J., Brodribb, T. J., & Cao, K. F. (2016). Weak co-ordination between vein and stomatal densities in 105 angiosperm tree species along altitudinal gradients in Southwest China. Functional Plant Biology, 43(12), 1126–1133. https://doi.org/10.1071/FP16012