Plant materials
All the materials grew outdoors in Huazhong Agricultural University, Wuhan, China (114o22’E, 30o29’N). A rice (Oryza sativa L.) cultivar Huanghuazhan (HHZ) growing in paddy field between 50~70 days after transplant was selected. All the materials were well watered and fertilized, free of diseases, pests, and weeds. Meanwhile, camphor (Cinnamomum camphora L.) grew in campus was investigated to test the relationships of two transpiration values obtained respectively by gas exchange instrument and balance, because camphor has relatively small leaves which can be fully covered in gas exchange chamber.
Harvest time and sample storage
Rice tillers were cut off in the paddy field under water in the early morning (between 5:30 am and 6:00 am) or previous day night (between 18:30 pm and 19:00 am). The fresh cut of tiller was soaked in degassed pure water, and leaves were covered with black plastic bags. The samples collected at night were conserved in ordinary ultrapure water and sterile ultrapure water, and half the tillers were randomly selected and recut in the morning to test the effect of blockage at the cut surface (Fig.S3). Longer than 0.5 m camphor branches were sampled in the early morning and conserved in ordinary ultrapure water. It took 10 to 15 min to transfer the samples to the laboratory. Branches and tillers were recut under pure water. Their cut ends were soaked in water, and other parts were covered with bags for at least 1 hour.
Equipment settings
Leaf hydraulic conductance (Kleaf) was determined using evaporative flux method (EFM) reported previously [5,37]. To reduce system error, the water evaporation in the graduated cylinder under multiple conditions was quantified as blank control. The system transpiration was measured under the following four conditions: the water without any addition, water with addition of liquid wax, the balance chamber with high humidity by putting wet tissue papers, and the combination of liquid wax addition and high humidity in the balance chamber. The combination was adopted in the subsequent leaf experiments due to its superior capacity to prevent water loss. In order to avoid ion deposition in leaf, ultra-pure water rather than ionic solution was adopted [53,54]. Ultra-pure water was vacuumed for 8 hours to remove bubbles or directly stored overnight for measuring.
One end of low-resistance transparent tube (inner diameter = 2mm, Oupli campany, Shanghai, China) filled water was connected to the graduated cylinder with water on a balance (±0.01mg; Mettler MS205DU, Mettler-Toledo GmbH, Greifensee, Switzerland). Finally, water volume in cylinder was adjusted to ensure that leaves were placed 2 cm below, 2 cm above, or as high as the meniscus of the water in the cylinder.
Changes in environmental factors
After sample preparation and system setting, three environmental factors – light intensity, air temperature, and airflow around leaf – were individually changed. The environment conditions were as follows: the temperature was set as 37±1oC or 27±1 oC; the photosynthetic photon flux density (PPFD) at leaf surface was set as 500, 1000, or 1500 µmol m-2 s-1; and the relative humidity was set as 50%.
Kleaf measurement
The newly- and fully-expanded target leaves with 2 cm sheathes or petioles were cut from tiller under airless distilled water. Leaf sheathes of rice leaves or the petioles of camphor leaves were connected to the water pipe using a hose tape and cork to achieve seamless connection between the petiole and tube. Leaf was lifted higher than water surface to detect whether bubbles occurred in the connection. After leaf was placed on fish line net, leaf surface was wiped with tissue paper and irradiated by lamp (600W, Weichuang Company, Wuhan, China). A box fan (Comfort Zone 20 Inch Box Fan, the factory Depot Advantages, Inc, USA) was used to minimize the boundary resistance (wind speed marked on the fan was 1 m s-1 or 0 m s-1). At the same time, water weight in the graduated cylinder and flow rate (F) were recorded every 3 seconds.
The water flow rate into the leaves was recorded until it was stable for a period of time. The detail identification of steady state was described in statistical analysis section below. The temperature of the blade middle was determined as leaf temperature by thermocouple (XimaAS877, Wanchuang electronic products Co., Ltd, Dongguan, China). Afterwards, leaf area (LA) and final leaf water potential () were measured. Kleaf was calculated according to the following formula:
Where E referred to evaporation rate, and it was calculated as water flow rate (F) divided by leaf area (LA). All the Kleaf values were normalized into those at 25°C considering that water viscosity varied with temperature [55]. The measurements were performed from 8:00 am to 18:00 pm since there was no correlation between measuring time and Kleaf or E (data not shown).
Leaf water potentials
Upper and lower leaves adjacent to the target leaf used for hydraulic conductance measurement were cut from the tiller before Kleaf measurement, quickly put in an exhaled double-layer zip-lock bag, and placed in a foam box for water equilibration. Subsequently, leaf initial water potential (ψintial) was detected in pressure chamber (PMS Instrument Company, Albany, OR, USA). Constant slow pressurization rate was maintained during measurement. After flow measurement, the finial leaf water potential (ψfinal) was determined, as described above. In addition, the water equilibration time of leaves in foam box were divided into 10, 20, 30, 60, 90, 120, 150, and180 min. Since there was no difference in ψ under 10~60 min equilibration, we selected 30 min equilibration time in this study.
Liquid flow and gas flow comparison
In order to test the consistency of liquid flow rate and gas flow rate obtained by balance and near infrared gas analyzer LI-6800 portable photosynthesis system (LI-COR Inc., Lincoln, NE, USA), leaf was quickly clamped by 6*6 cm transparent chamber (LI-6800-13, LI-COR Inc., Lincoln, NE, USA) in the middle of rice leaf or the whole leaf blade of the camphor after being put on net. The chamber environment was as coincident with ambient environment as possible. Auto-log was conducted with 30-second interval until a steady state was reached.
Statistical analysis
As the flow data was typically dynamic, the judgment on whether the flow rate has stabilized is challenge. An effective method to restrict segment lengths of given flow data is to explicitly allow high variance of segments, and segment length restriction was achieved via the break-point penalty parameter P in ‘dpseg’ package. In our analysis, high P value will allow high variance of the individual segments to produce long segments. F - time curve was segmented according to P-value, and the obtained segments were marked with different color. Different segments separated by a few outliers produced within 1 min were regarded as one segment (Fig.4). The time corresponding to the last segment was required to be longer than 15 min, and the t-test P-value of the curve slope corresponding to the last 15 min (about 300 points) was required to be larger than 0.001.
One-way analysis of variation (ANOVA) and multiple comparisons (least significant ranges, ‘agricolae’ package) were conducted at level of 0.05. The correlation between traits was fitted using ‘ggpmisc’ package. All figures were plotted using ‘tidyverse’ package. All of the statistical analyses were performed in R version 3.6.1 (https://cran.r-project.org).