Journal of Experimental Botany, Vol. 70, No. 18 pp.

4949–4961, 2019
doi:10.1093/jxb/erz238 Advance Access Publication May 30, 2019

RESEARCH PAPER

Leaf anatomical adaptations have central roles in

photosynthetic acclimation to humidity
Qingjie Du†, , Tao Liu†, Xiaocong Jiao, Xiaoming Song, Jiayu Zhang and Jianming Li*

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

College of Horticulture, Northwest A&F University, Yangling, Shaanxi 712100, China

†
These authors contributed equally to this work.
* Correspondence: lijianming66@163.com

Received 5 November 2018; Editorial decision 7 May 2019; Accepted 8 May 2019

Editor: Tim Brodribb, University of Tasmania, Australia

Abstract
Rates of photosynthesis can be lower in plants grown under conditions of high leaf-to-air vapour pressure differ-
ence (VPD) than under low VPD. Leaf phenotype plasticity is a primary factor determining photosynthetic responses
to environmental stimuli. However, it remains unclear how changes in leaf anatomical traits drive photosynthetic
acclimation to high VPD. Here, we examined the role of leaf anatomy in the differing photosynthetic responses of
two tomato cultivars (Jinpeng and Zhongza) to long-term growth under high and low VPD. Photosynthesis was not
affected by VPD in Jinpeng. This was attributed to homeostasis in stomatal conductance (gs) and, to a lesser extent,
mesophyll conductance (gm). Disruption of synchronized changes to cell size in the epidermis and mesophyll meant
that growth under high VPD reduced stomatal density in Jinpeng, but minor vein density remained unchanged.
Thus, water supplied by the veins could support the increased transpirational demand, preventing stomatal closure.
Variation in VPD did not affect mesophyll cell structures, and therefore gm, in Jinpeng. By contrast, photosynthesis
in Zhongza was reduced under high VPD, which was primarily attributed to decreased gs and gm. The former was
mainly induced by decreased stomatal aperture. Thus, transpirational demand exceeded water supply in Zhongza.
This was likely due to coordinated decreases in stomatal and minor vein density driven by synchronized increases
in epidermal and mesophyll cell size under high VPD. Liquid-phase limitation was primarily responsible for the re-
duced gm in Zhongza under high VPD. High VPD induced an increase in liquid-phase resistance by reducing the
mesophyll surface area exposed to intercellular air spaces and increasing cytosolic resistance. These results sug-
gest that plasticity in epidermal and mesophyll cell size provides an efficient means of regulating photosynthesis
during acclimation to long-term high VPD.

Keywords: Leaf anatomy, mesophyll conductance, photosynthesis, stomatal conductance, tomato, vapour pressure difference.

Abbreviations: An, net assimilation rate; BL, biochemical limitation of An; Cc, chloroplast CO2 concentration; Ci, intercellular CO2 concentration; DL, diffusion limitation
of An; fias, fraction of mesophyll tissue occupied by intercellular air spaces; gias, CO2 diffusion conductance in gas phase; gliq, CO2 diffusion conductance in liquid
and lipid phases; gm, mesophyll conductance; gm-a, mesophyll conductance estimated from leaf anatomical traits; gs, stomatal conductance; gs-max, maximum
stomatal conductance; Kleaf, leaf hydraulic conductance; lcw, cell wall limitation of gm; lcyt, cytosol limitation of gm; lenv, chloroplast envelope limitation of gm; lias, gas-
phase limitation of gm; LMA, leaf dry mass per area; lpl, plasmalemma limitations of gm; lst, stroma limitations of gm; ML, mesophyll limitation of An; rcw, CO2 diffusion
resistance in cell wall; rcyt, CO2 diffusion resistance in cytosol; renv, CO2 diffusion resistance in chloroplast envelope; rpl, CO2 diffusion resistance in plasmalemma;
rst, CO2 diffusion resistance in stroma; Sc/S, surface area of chloroplast exposed to intercellular air spaces; Sc/Sm, ratio between chloroplast and mesophyll surface
area exposed to intercellular air space; SL, stomatal limitation of An; Sm/S, surface area of mesophyll exposed to intercellular air spaces per leaf area; Vc,max, max-
imum velocity of carboxylation; VPD, leaf-to-air vapour pressure difference.

© The Author(s) 2019. Published by Oxford University Press on behalf of the Society for Experimental Biology. All rights reserved.
For permissions, please email: journals.permissions@oup.com
4950 | Du et al.
Introduction
Photosynthesis is the key driver of primary metabolism in
plants and responds to many environmental variables (Valentini
et al. 1995; Lawlor and Tezara 2009). Leaf-to-air vapor pressure
difference (VPD) is one of the most important environmental
factors affecting photosynthesis (Leuschner 2002; Shirke and
Pathre 2004). Plants grown under high VPD usually exhibit re-
duced photosynthesis (Leuschner 2002; Lu et al. 2015). During
photosynthesis, CO2 diffuses from the atmosphere through sto-
mata and mesophyll to the site of carboxylation in the chloro-
plast (Flexas et al. 2002; Evans et al. 2009). Thus, a reduction
in stomatal (gs) and mesophyll (gm) conductance will lead to
low CO2 availability inside the leaf, thereby limiting photo-
synthesis. Furthermore, the pathway of CO2 diffusion implies
that leaf anatomical traits are involved in resistance to CO2 dif-
fusion (the inverse of conductance) (Evans et al. 1994; Franks
and Beerling 2009; Lawson and Blatt 2014). There is abundant
evidence that variation in VPD can trigger changes in stomata,
leaf vein, and mesophyll anatomy (Bongi and Loreto 1989;
Fanourakis et al. 2013). Therefore, changes in leaf anatomical
traits may account for the response of photosynthesis to vari-
ation in VPD by influencing gs and gm (Bongi and Loreto 1989;
Carins Murphy et al. 2014).
Many studies have suggested that high VPD induces low
intercellular CO2 concentration due to reduced gs (Lu et al.
2015; Zhang et al. 2017). It is well known that gs is determined
by stomatal density, size, and aperture (Franks and Beerling 2009;
Lawson and Blatt 2014). Long-term high VPD generally results
in small stomata (Fanourakis et al. 2013; Lu et al. 2015), but the
response of stomatal density to long-term VPD varies among
species. A decrease in stomatal density with increasing VPD has
been observed in tomato, rose, and fava bean (Fanourakis et al.
2013; Aliniaeifard et al. 2014; Lu et al. 2015), whereas the op-
posite response is found in Toona ciliata (Carins Murphy et al.
2014). Despite this, gs is typically reduced by growth under high
VPD. In fact, the response of stomatal aperture to varying VPD
is critical to determining gs (Fanourakis et al. 2013; Lawson and
Blatt 2014). Büssis (2006) reported that dynamic responses of
stomata can compensate for altered stomatal density. Stomatal
aperture can respond to leaf water status, which is determined by
the balance between water supply and evaporative demand (Liu
et al. 2015; Rodriguez-Dominguez et al. 2016; Brodribb and
McAdam 2017).This balance is generally achieved by regulating
the leaf vein density, which has a critical role in water supply
(Sack and Frole, 2006; Brodribb et al. 2007), and stomatal density,
which largely determines maximum transpiration (Franks and
Beerling 2009). The densities of leaf veins and stomata remain
coordinated in leaves grown under different light intensities be-
cause of a shared relationship with the expansion of surrounding
cells (Brodribb and Jordan 2011; Carins Murphy et al. 2012,
2016). Thus, cells of leaves grown under low light expand more
than cells of leaves grown under high light, which ‘passively di-
lutes’ veins and stomata, reducing their densities in parallel. This
coordination may not efficiently maintain leaf water status under
highVPD due to higher evaporative demand; under these condi-
tions, stomata would close and/or the relationship between leaf
vein and stomatal density would become uncoupled to prevent
leaf desiccation. Carins Murphy et al. (2014) found that dynamic
stomatal control maintained transpirational homeostasis during
growth under high and low VPD conditions, rather than disrup-
tion of the proportional relationship between vein and stomatal
density, in a woody angiosperm species. However, uncoupled
vein and stomatal initiation has been recently demonstrated in
herbaceous species (Carins Murphy et al., 2017; McAdam et al.
2017; Cardoso et al. 2018), suggesting that independent produc-
tion of veins and stomata could also maintain the balance be-
tween evaporative demand and water supply. More studies are
Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024
required to establish how changes in stomata and leaf vein drive
gs responses to different VPD.
Another physiological limitation on photosynthesis is gm.The
current literature indicates that gm declines (Bongi and Loreto
1989), increases (Perez-Martin et al. 2009), or is unchanged
(Warren 2008) in response to growth under highVPD.Although
Warren (2008) attributed this variability to differences among
high-VPD treatments, a more probable explanation may be
that anatomical traits related to gm respond to VPD in different
ways in different species (Perez-Martin et al. 2009). Extensive
information is available about the relationship between gm and
whole-leaf traits, such as leaf dry mass per area (LMA) (Galmés
et al. 2013; Peguero-Pina et al. 2016). The link between gm and
LMA is thought to be primarily driven by the two components
of LMA (leaf thickness and density) (Niinemets et al. 2009;
Perez-Martin et al. 2009; Tomás et al. 2013). However, CO2
molecules diffuse in the mesophyll across intercellular spaces,
cell wall, plasmalemma, cytoplasm, chloroplast envelope, and
stroma (Tosens et al 2012; Tomás et al. 2013). Thus, measure-
ments of changes in these individual anatomical components
are essential for estimating the responses of gm to environmental
stimuli. Variability in gm can result from changes to both the
gas and liquid phases of the CO2 diffusion pathway through
the mesophyll (Evans et al. 1994; Niinemets and Reichstein
2003). However, because the rate of CO2 diffusion is lower in
the liquid phase than in the gas phase, the resistance of the li-
quid phase largely determines CO2 diffusion in the mesophyll
(Evans et al. 1994; Tosens et al 2012). In the liquid phase, the
most important structural component is the surface area of
chloroplast exposed to intercellular air spaces (Sc/S), which de-
fines the surface area for CO2 dissolution (Galmés et al. 2013;
Tomás et al. 2013; Fini et al. 2016). Bongi and Loreto (1989)
reported that high VPD reduced Sc/S, and therefore decreased
gm. However, other studies have provided evidence that Sc/S is
not correlated with gm because other components, such as the
cell wall, cytosol, stroma, and cell packing, have important roles
in regulating gm (Hanba et al. 2004;Tomás et al. 2013; Muir et al.
2014). This may explain the different results in gm responses to
increasing VPD (Bongi and Loreto 1989; Warren 2008; Perez-
Martin et al. 2009).Therefore, knowledge of the responses of all
the components that contribute to gm is crucial to determine
the responses of gm to VPD.
Tomato (Solanum lycopersicum L.) is one of the most economic-
ally important vegetable crops in the world (Mulcahy et al 2013;
Baldantoni et al. 2016; de Marco et al. 2018). It has previously
been reported that photosynthesis is seriously restricted in tomato
 Leaf anatomy determines photosynthetic acclimation to VPD | 4951
plants growing under high VPD (Zhang et al. 2017; Du et al.
2018). Our previous work has identified two tomato cultivars
(Jinpeng and Zhongza) with differing photosynthetic responses to
long-term changes in VPD conditions, which provide an oppor-
tunity to examine the roles of gs and gm in photosynthetic acclima-
tion. Jinpeng maintains a similar rate of photosynthesis under high
and low VPD, while in Zhongza photosynthesis is lower in plants
grown under high VPD than under low VPD (see Supplementary
Table S1 at JXB online).Therefore, we hypothesize that (i) photo-
synthetic acclimation to long-term highVPD is mostly dependent
on the responses of gs and gm; (ii) the relationship between leaf
vein and stomatal density will determine the response of gs to
long-term changes in VPD; and (iii) VPD may influence gm not
only by changing Sc/S but also by changing cellular components.
Materials and methods
Plant material and growth conditions
Seeds of two tomato cultivars, Jinpeng and Zhongza, were germinated
on moist filter paper in the dark for 2 days at 28 °C. Seedlings were
transplanted to 0.5 l pots (1 plant per pot) containing a mixed peat–
perlite substrate and grown in a controlled-environment growth chamber
(RDN-1000D-4, Southeast Instrument Co. Ltd, Ningbo, China). The
photoperiod was 14 h at 300 μmol m–2 s–1 photon flux density. The tem-
perature was maintained at 30/20 °C (day/night) with a fluctuation of
±0.5 °C. The relative humidity was kept at 65/80% (day/night) with
a fluctuation of ±5% [VPD 1.48/0.47 kPa (day/night)] using a dehu-
midifier and humidifier inside the chamber. When plants displayed two
true leaves,VPD treatments were performed by setting different daytime
relative humidity levels. The daytime relative humidity was maintained
at 65±5% (VPD 1.48 kPa) for the low-VPD treatment and at 40±5%
(VPD 2.55 kPa) for the high-VPD treatment. For both VPD treatments,
the light, temperature, and nighttime relative humidity were kept the
same as those before the treatments. The temperature and relative hu-
midity were recorded by a datalogger (PDEI, Wuge Technology Co. Ltd,
Harbin, China). Plants were irrigated daily to near field capacity condi-
tions throughout the whole growth period. For each cultivar, 15 plants
per treatment were included. After 30 days, all measurements were per-
formed on new, fully expanded leaves.
Gas exchange and chlorophyll fluorescence measurements
Gas exchange and chlorophyll fluorescence were simultaneously meas-
ured using the LI-6800 portable photosynthesis system (Li-Cor Inc.,
Lincoln, NE, USA). Gas exchange and chlorophyll fluorescence param-
eters were estimated after reaching steady-state at a saturating photosyn-
thetic photon flux density (PPFD) of 1000 μmol m–2 s–1, atmospheric
CO2 concentration (Ca) of 400 μmol mol–1, and leaf temperature of
30 °C. The relative humidity in the leaf chamber was maintained at 65%
for low VPD and 40% for high VPD. Net assimilation rate (An), gs, inter-
cellular CO2 concentration (Ci), transpiration rate (E), steady-state fluor-
escence (Fs), and maximum fluorescence during a light-saturating pulse
of approximately 8000 μmol m–2 s–1 (Fm′) were recorded. All measure-
ments were recorded on two leaflets of the same leaf per individual plant,
using five plants per cultivar for each treatment. After the gas exchange
measurements, one leaflet was used for measurements of leaf water poten-
tial and leaf vein density, and another was used for observations of stomata
and to obtain semi-thin and ultrathin cross-sections.
Leaf hydraulic conductance measurements
Leaf hydraulic conductance (Kleaf) was calculated from the E and the
water potential gradient between leaflet (Ψl) and leaf rachis (Ψx) based
on Ohm’s law as follows:
E
Kleaf =
(1) ,
ψX − ψ1
where Ψl was measured on the leaflet used for gas exchange measure-
ments with a pressure chamber (PMS, Corvallis, OR, USA). Ψx was taken
as the water potential of a non-transpiring leaflet, which was obtained by
covering a leaflet with plastic film and aluminum foil on the night before
measurements were made.The non-transpiring leaflet was adjacent to the
two leaflets used for gas exchange measurements.
Leaf anatomy measurements
The impression method was used to analyse stomatal morphological
Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024
traits, as described by Xu and Zhou (2008). In order to estimate sto-
matal aperture, samples were obtained under high- or low-VPD con-
ditions during the light periods. Briefly, the abaxial and adaxial surfaces
of the central leaflet were smeared with nail varnish, avoiding the veins.
Then, the thin film was peeled from the leaflet and mounted upside
down on a glass slide, and observed under a light microscope (BX50,
Olympus, Tokyo, Japan). To determine stomatal density, three different
field-of-view regions per sampling area (removed from one leaflet from
each of five different plants per cultivar) were taken at a magnification
of ×20. Stomatal density was obtained by summing the number of ab-
axial and adaxial stomata per unit area. Stomatal aperture, length, and size
(defined as the combined area of a pair of guard cells) were measured at
×40 magnification in five randomly selected stomata within each field of
view using ImageJ software (National Institutes of Health, Bethesda, MD,
USA). Epidermal cell size (Sec) was calculated as:
[2 − (Ds × Ss )]
Sec =
(2)
Dec
where Ds is stomatal density, Ss is stomatal size, and epidermal cell density
(Dec) is the sum of the abaxial and adaxial values. gs-max was calculated as
described by Franks and Beerling (2009):
D × d × amax
gs−max =
(3) Ä s  ä
v × l + π2 × amax /π
where d is the diffusivity of water in air (m2 s−1), v is the molar volume
of air (m3 mol−1), l is the stomatal depth (m), which was measured on
semi-thin cross sections of leaf, and amax is the maximum area of the open
stomatal pore (m2), which was calculated as π(stomatal length/4)2.
Leaf segments (1 mm2) were cut from the central leaflet regions,
avoiding the veins, and then fixed with 2.5% glutaraldehyde (v/v) in
0.1 M phosphate buffer (pH 7.2) for 4 h. The leaf material was washed
twice in the same buffer and post-fixed in 2% osmium tetroxide for 2 h
at 4 °C. Then, the samples were dehydrated with a graded ethanol series
(30, 50, 70, 80, 90, and 100%), followed by gradually embedding the
samples in Spurr’s resin. Semi-thin (1 μm) and ultrathin (90 nm) cross-
sections were cut with an LKB ultramicrotome (LKB Co. Ltd, Uppsala,
Sweden). After staining with methylene blue, semi-thin cross-sections
were examined with a light microscope at ×40 magnification to measure
leaf thickness, palisade thickness, spongy thickness, mesophyll thickness,
total cross-sectional area of mesophyll cells (∑Ss), and stomatal depth.
Mesophyll cell size was estimated as the average cross-sectional area of
spongy and palisade mesophyll cell.The ultrathin cross-sections were im-
aged using a transmission electron microscope at magnifications of ×2000
to measure chloroplast distribution and ×15 000−40 000 to measure cell
wall and cytoplasm thickness. Five leaves from different plants of each
cultivar were analysed for each treatment. The fraction of intercellular air
spaces (fias), surface area of mesophyll exposed to intercellular air spaces
per leaf area (Sm/S), and Sc/S were calculated according to Syvertsen
et al. (1995) as follows:
SS
fias = 1 −
(4)
Tmes × W
4952 | Du et al.

Lmes 1
Smes /S =
(5) ×F gm -a = 1
(10) 1 RTk
W gias + gliq + H

Lc
Sc /S =
(6) × Smes /S
Lmes
where W is the width of the measured section, Lmes is the total length of
mesophyll cell exposed to the intercellular air space, Lc is the length of
chloroplasts exposed to the intercellular air space, and F is the curvature
correction factor, which was calculated from the ratio of width to height
of palisade and spongy cells (Evans et al. 1994).
Leaf vein density was measured in the detached leaflets used for Ψl
measurement. To measure the density of minor veins, 1 cm2 pieces were
where gias is the gas-phase conductance to CO2 diffusion across the inter-
cellular air spaces, gliq is CO2 diffusion conductance in liquid and lipid
phases, H is the Henry’s law constant (Pa m3 mol–1), R is the gas constant
(Pa m3 K–1 mol–1), and Tk is the absolute temperature (K). The dimen-
sionless form of Henry’s law constant (H/RTk) is used to convert gliq to
the corresponding gas-phase equivalent conductance because gm is de-
fined as gas-phase conductance (Niinemets and Reichstein 2003). Next,
gias was determined as follows:
Da × fias

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

gias =
(11)
cut from the center of each leaf (avoiding midrib and secondary veins) ∆Lias × ς
and prepared according to the method described by Kono et al. (1982).
Samples were boiled in 70% ethanol for 10 min and then transferred to where Da is the diffusivity of CO2 in the gas phase (m2 s–1), ς is the diffu-
boiling 85% lactic acid for 5 min. The samples were washed in distilled sion path tortuosity (m m–1), and ΔLias is the average gas-phase thickness
water and then mounted on glass slides for observation under a light (taken as half the mesophyll thickness). Then, gliq was obtained as the sum
microscope at ×4 magnification. of the inverse serial resistances, considering two different pathways of
The area of the leaf used for gas exchange measurements was calcu- CO2 inside the cell (Tomás et al. 2013):
lated based on the product of leaf length and width measured before Sm /S
starting the experiments (Supplementary Fig. S1). gliq =
(12)
(rcw + rpl + rcel,tot )
Estimation of gm
Sc /Sm (1 − Sc /Sm )
gm was calculated from measurements of gas exchange and chlorophyll gcel,tot =
(13) +
fluorescence according to Harley et al. (1992), as follows: rcyt ,1 + rst,1 + renv rcyt,2 + rst,2 + renv

An where rcw, rpl, and renv are the CO2 diffusion resistances of the cell wall,
gm =
(7) plasmalemma, and chloroplast envelope, respectively; rst,1 and rst,2 are stromal
Ci − Cc resistances in the perpendicular and parallel directions, respectively, rela-
tive to the cell wall; and rcyt,1 and rcyt,2 are cytosolic resistances from the
Γ∗ × [J + 8 × (An + Rd )] plasmalemma inner surface to the outer chloroplast surface and from the
Cc =
(8) interchloroplastic cell wall to the chloroplast, respectively. A constant value
J − 4 × (An + Rd )
of 0.0035 m s–1 was used for gpl and genv, as in previous reports (Evans et al.
where An and Ci were taken from the gas exchange measurements at 1994; Tosens et al. 2012). The conductances of other components in the
saturating light, Cc is the chloroplast CO2 concentration, J is the electron liquid-phase diffusion pathway were determined as follows:
transport rate, Rd is the daytime respiration rate, and Γ* is the chloroplast
CO2 compensation point. Rd and Γ* were estimated according to the 1 rf,i × Dw × pi
gi = =
(14)
method of Brooks and Farquhar (1985). The A–Ci curves were measured ri ∆Li
under three different light intensities (50, 100, and 200 μmol m–2 s–1).
The intersection of the initial linear regions of the three A–Ci curves where gi is the cell wall, cytosol, and stroma conductance, and Dw is the
represented Ci* (x-axis) and Rd (y-axis). Ci* was used as a proxy for Γ* aqueous phase diffusion coefficient for CO2 (m2 s–1). The diffusion path
according to the method of Warren (2006). J was determined as follows: length (ΔLi) in the corresponding component was obtained as described by
Tomás et al. (2013).The dimensionless factor (rf,i) was set to 1.0 for cell wall
J = ΦPSII × PPFD × α × β
(9) (Rondeau-Mouro et al. 2008) and 0.3 for cytosol and stroma (Niinemets
and Reichstein 2003). Effective porosity (pi) was set to 1.0 for cytosol and
where the effective quantum yield of photosystem II (ΦPSII) was calcu- stroma and estimated for cell wall according to Tosens et al. (2012).
lated as 1–Fs/Fm′ (Genty et al. 1989), α is the leaf light absorption, and
β is the fraction of light absorbed by PSII. The product α×β was deter-
mined using the relationship between ΦPSII and quantum efficiency of Quantitative limitation analyses of An
CO2 fixation (ΦCO2), which was estimated by varying the light intensity The limitations of gs (SL), gm (ML), and biochemical capacity (BL) on
under non-photorespiratory conditions (Valentini et al. 1995). An were investigated according to the approach proposed by Grassi and
The maximum velocity of carboxylation (Vc,max) was calculated from Magnani (2005). Relative change in An can be expressed in terms of par-
the An–Ci curves by fitting the data to the biochemical model of Farquhar allel relative changes in gs, gm, and biochemical capacity (Vc,max) using the
et al. (1980) as performed by Sharkey et al. (2007). Rubisco kinetic con- following equation:
stant values in the model were obtained from Bernacchi et al. (2002). To
calculate Vc,max, the values of Rd, Γ*, and gm were estimated as described dAn dgs dgm dgb
above. Five An–Ci curves per treatment and cultivar were estimated using = SL + ML + BL = ls × + lm × + lb ×
An gs gm gb
the leaves sampled for photosynthesis measurements at different Ca (400,
300, 200, 150, 100, 50, 400, 600, 800, 1000, and 1500 μmol mol–1).
(15)
gtot /gs × ∂A/∂Cc
ls =
(16)
Modeling mesophyll conductance on the basis of leaf anatomy gtot + ∂A/∂Cc
parameters
To estimate mesophyll conductance using leaf anatomical parameters (gm-a), gtot /gm × ∂A/∂Cc
the one-dimensional gas diffusion model of Niinemets and Reichstein lm =
(17)
gm + ∂A/∂Cc
(2003) was employed, as described by Tomás et al. (2013):
 Leaf anatomy determines photosynthetic acclimation to VPD | 4953

unaffected by VPD in Jinpeng (Fig. 1, Table 1). In both culti-

gtot vars, Vc,max was similar under high and low VPD. Ci was sig-
lb =
(18)gm + ∂A/∂Cc nificantly lower under high VPD than low VPD in Zhongza,
where gs is stomatal conductance to CO2, gtot is the total conductance but no difference was observed under high and low VPD in
to CO2 from the leaf surface to carboxylation sites (1/gtot=1/gs + 1/gm), Jinpeng. Consistently, Cc was reduced significantly in Zhongza
and ∂A/∂Cc is calculated as the slope of An–Cc response curves over a Cc under high VPD, whereas in Jinpeng it was similar under high
range of 50–100 μmol mol–1 (Tomás et al. 2013). and low VPD.
Then, the relative changes in An, gs, gm, and Vc,max were calculated as
follows: gs-max was lower under high VPD than low VPD in both
Jinpeng (18%) and Zhongza (33%) (Fig. 2A). Kleaf was signifi-
dAn Aref − An cantly higher under high VPD than low VPD in Jinpeng, but
= n ref
(19)
An An the opposite result was found in Zhongza (Fig. 2B).
The total photosynthetic limitation (sum of DL and BL) im-

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

dgs gref − g posed by high VPD was 26.73% in Zhongza, but only 6.57%
= s ref s
(20) in Jinpeng (Table 2). BL was similar in Zhongza (1.23%) and
gs gs
Jinpeng (2.06%), whereas DL was higher in Zhongza (25.5%)
than in Jinpeng (4.51%). Analysis of the components of DL in
dgm gref − g
= m ref m
(21) Zhongza indicated that SL (16.04%) exerted the strongest limi-
gm gm tation on An decline, followed by ML (9.46%).

dVc,max V ref − Vc,max

(22) = c,max ref
Vc,max Vc,max
where Anref, gsref, gmref, and Vc,maxref are reference maximum values for An,
gs, gm, and Vc,max, respectively.Values were obtained under low-VPD con-
ditions. Diffusional limitation (DL) was calculated as the sum of SL and
ML.
Quantitative analyses of anatomical limitations of gm
The limitations of gm were partitioned into components related to gas-
phase diffusion across the intercellular air spaces (lias) and liquid-phase
diffusion in the cell wall (lcw), plasmalemma (lpl), cytosol (lcyt), chloroplast
envelope (lenv), and stroma (lst) according to the method of Tomás et al.
(2013), as follows:
gm
lias =
(23) gias
gm
li =
(24)gi × (Sm /S)
where li is lcw, lpl, lcyt, lenv, and lst; and gi is the corresponding diffusion con-
ductance. The fractions of exposed cell-wall area lined with chloroplasts
or free of chloroplasts were used to determine diffusion conductance in
the cytosol, chloroplast envelope, and stroma.
Statistical analyses
All statistical analyses were performed using SPSS 19.0 (IBM Corp.,
Armonk, NY, USA). The significance of differences between mean
values was assessed by one-way analysis of variance (ANOVA) according
to Tukey’s test (P<0.05). The effects of cultivar, treatment, and their
interaction on all parameters were determined by two-way ANOVA.
Regression analyses were performed with SigmaPlot 12.5 (Systat
Software Inc., San Jose, CA, USA).
Results
Response of gas exchange parameters and Kleaf to
long-term high and low VPD
Variation in An, gs, and gm under different VPD were cultivar
specific. These parameters declined significantly in Zhongza
under high VPD compared with low VPD, whereas they were
Response of leaf anatomical characteristics to long-
term high and low VPD
In both cultivars, the stomatal density and size were markedly
reduced under high VPD compared with low VPD (Table 3,
Fig. 3). Stomatal density and size decreased by 12% and 15%,
respectively, in Jinpeng, and by 21% and 21%, respectively,
in Zhongza. In contrast, epidermal cell size increased under
high VPD in both cultivars. Stomatal aperture and minor vein
density decreased significantly under high VPD in Zhongza,
whereas both properties were unaffected by VPD in Jinpeng.
Leaf area was similar under high and low VPD in both cultivars.
Mesophyll thickness, fias, Sc/S, and Sm/S were reduced signifi-
cantly under high VPD in Zhongza, but they were not affected
by high VPD in Jinpeng (Table 4). By contrast, no significant
difference in the ratio between chloroplast and mesophyll sur-
face area exposed to intercellular air space (Sc/Sm) was found
under high and low VPD in either cultivar. Mesophyll cell size
in Zhongza was significantly higher under high VPD than low
VPD, whereas it was similar in Jinpeng under high and low VPD.
Increasing VPD induced a proportional decrease in minor
vein and stomatal density in Zhongza (R2=0.87, F(1,8)=53.60,
P<0.01), whereas the relationship between stomatal and
minor vein density was not significant in Jinpeng (R2=0.11,
F(1,8)=0.34, P>0.05) (Fig. 4A). Epidermal cell size and meso-
phyll cell size were strongly correlated in Zhongza (R2=0.55,
F(1,8)=9.97, P<0.01) but not in Jinpeng (R2=0.33, F(1,8)=4.02,
P>0.05) (Fig. 4B). The relationships between stomatal density
and epidermal cell size, and between minor vein density and
mesophyll cell size, were significant for both cultivars (Fig. 4C,
D). Pooled data across all treatments and cultivars indicated that
fias declined along with Sc/S (R2=0.70, F(1,18)=53.83, P<0.01)
and Sm/S (R2=0.68, F(1,18)=38.86, P<0.01) (Fig. 5).
Response of partial limitation of gm to long-term high
and low VPD
gm-a was tightly correlated with gm calculated according to the
method of Harley et al. (1992) (Fig. 6). To separate the contri-
butions of the total cellular resistance (rcw + rpl + rcyt + ren + rst)
4954 | Du et al.

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

Fig. 1. Net assimilation rate (An) (A), stomatal conductance (gs) (B), mesophyll conductance (gm) measured with the method of Harley et al. (1992) (C),
maximum velocity of carboxylation (Vc,max) (D), intercellular CO2 concentration (Ci) (E), and chloroplast CO2 concentration (Cc) (F) for tomato cultivars
Jinpeng and Zhongza grown under low (1.48 kPa) and high (2.55 kPa) VPD. Data are means ±SE (n=5 plants). Different letters denote statistically
significant differences between treatments (Tukey test, P<0.05).

Table 1. Results of two-way ANOVA for the effects of cultivar, treatment, and their interaction on net assimilation rate (An), stomatal
conductance (gs), mesophyll conductance measured with the method of Harley et al. (1992) (gm), maximum velocity of carboxylation
(Vc,max), intercellular CO2 concentration (Ci), chloroplast CO2 concentration (Cc), maximum stomatal conductance (gs-max), and leaf
hydraulic conductance (Kleaf), measured in tomato cultivars Jinpeng and Zhongza grown under low (1.48 kPa) and high (2.55 kPa) VPD.

An gs gm Vc,max Ci Cc gs-max Kleaf

ns ns ns ns
Cultivar 11.50** 3.86 0.09 17.79** 1.28 5.02* 1.56 13.08**
Treatment 8.40* 15.49** 7.68* 0.87ns 8.72** 9.45** 30.36** 1.29ns
Cultivar×Treatment 2.44ns 8.49* 2.40ns 0.02ns 2.45ns 2.44ns 2.66ns 29.74**

Data are values of F(1,16). **P<0.01, *P<0.05; ns, not significant.

and Sc/S on rliq (1/gliq), the change of log[(Sc/S)–1] and log(rcw
+ rpl + rcyt + ren + rst) were estimated (Fig. 7). The variation of
[(Sc/S)–1] in Zhongza spanned approximately 0.28 log10 unit,
larger than that of total cellular resistance. However, the variation
of log[(Sc/S)–1] and log(rcw + rpl + rcyt + ren + rst) had similar values
in Jinpeng. For both Jinpeng and Zhongza, rcw, rpl, renv, rst, and the
resistance to CO2 diffusion across intercellular spaces (rias) did not
differ under high or low VPD (Fig. 8). By contrast, the response
of rcyt to VPD was cultivar dependent. In Zhongza, rcyt was mark-
edly higher under high than low VPD, whereas it did not differ
under high or low VPD in Jinpeng.
Discussion
This study demonstrates that long-term high VPD induced
leaf anatomical acclimation, including stomata, leaf vein, and
 Leaf anatomy determines photosynthetic acclimation to VPD | 4955

Responses of An to long-term high and low VPD are

primarily determined by diffusion limitations
Efficient CO2 fixation is crucial for plant acclimation to envir-
onmental changes. In this study, An was not affected by VPD
treatment in the cultivar Jinpeng, whereas a significant decline
in An was found in Zhongza under high VPD (Fig. 1).Variation
in gs and gm were proportional to the changes in An. However,
Vc,max was similar under high and low VPD in both cultivars.
Thus, the reduction in An under high VPD in Zhongza resulted
from low CO2 availability due to high diffusion resistance (low
gs and gm), which was confirmed by the observed decline in Ci

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

and Cc. Furthermore, quantitative limitation analyses of An also
indicate that diffusion limitation of the atmosphere to carb-
oxylation sites has a major role in determining An (Table 2).
Although both gs and gm were important for explaining the re-
duced An of Zhongza under high VPD, gs was the predominant
factor. The larger limitation of CO2 diffusion through stomata
versus mesophyll under high VPD was consistent with findings
of a previous study of tomato (Warren 2008). In fact, a positive
relationship between gm/gs and intrinsic water use efficiency
(An/gs) has been reported in many studies (Galmés et al. 2011;
Giuliani et al. 2013;Tomeo and Rosenthal 2017), implying that
increasing gm/gs enhances An/gs.Therefore, a larger decline in gs
relative to gm could improve An/gs.

Responses of gs to long-term high and low VPD are

mediated by leaf anatomical traits

Fig. 2. Maximum stomatal conductance (gs-max) (A) and leaf hydraulic

Maximum stomatal conductance is constrained by stomatal
conductance (Kleaf) (B) for tomato cultivars Jinpeng and Zhongza grown morphology. In the present study, stomatal density and size were
under low (1.48 kPa) and high (2.55 kPa) VPD. Data are means ±SE (n=5 reduced under high VPD in both Jinpeng and Zhongza (Table
plants). Different letters denote statistically significant differences between 3), an observation that was consistent with previous studies in
treatments (Tukey test, P<0.05). tomato (Lu et al. 2015; Du et al. 2018), rose (Fanourakis et al.
2013), and fava bean (Aliniaeifard et al. 2014). The small sto-
mata facilitate a rapid response to excessive water losses (Drake
Table 2. Quantitative limitation analysis of net assimilation rate for et al. 2013; de Boer et al. 2016). However, stomatal density in-
tomato cultivars Jinpeng and Zhongza, as affected by high VPD creased with increasing VPD in T. ciliata (Carins Murphy et al.
(2.55 kPa), expressed as a percentage of reference maximum 2014) and woodland herbs (Leuschner 2002). This discrepancy
values obtained under low-VPD conditions (1.48 kPa). may be due to inherent species differences in responses to
increasing VPD. The reduction in stomatal density and size led
Cultivar TL SL ML BL DL to a decline in gs-max in both cultivars (Fig. 2). In spite of this,
Jinpeng 6.57 1.72 2.79 2.06 4.51
a significantly decreased gs was found in Zhongza but not in
Zhongza 26.73 16.04 9.46 1.23 25.50 Jinpeng under high VPD. In fact, the reduced vertical diffusion
of gas due to decreased stomatal density can be at least partly
Total photosynthetic limitation (TL) was partitioned into stomatal (SL), offset by increased lateral fluxes (Dow et al. 2014; Büssis et al.
mesophyll conductance (ML), and biochemical limitation (BL). Diffusional 2006). Stomatal density decreased by 11% in Jinpeng and 21%
limitation (DL) was calculated as the sum of SL and ML.
in Zhongza under high VPD, suggesting that the unaffected gs
in Jinpeng may be due to the small decline in stomatal density.
Furthermore, the different variations of gs in Jinpeng and
mesophyll cell structure. Changes operating at the level of Zhongza may also be attributed to specific stomatal dynamic
leaf anatomy play crucial roles in photosynthetic responses to responses to high VPD because the vertical fluxes were also
long-term high VPD, particularly in CO2 diffusion from the limited by stomatal aperture (Lawson and Blatt 2014). It was
atmosphere to the site of carboxylation in the chloroplast. observed in this study that stomatal opening was maintained
Furthermore, we confirmed that the homeostasis in gs was re- in Jinpeng under high VPD, while stomatal aperture decreased
lated to the relationship between stomata and minor veins, and in Zhongza.
that the regulation of gm was mainly dependent on Sm/S and The maintenance of open stomata is mainly driven by leaf
cytosolic resistance. water status, which is dependent on the balance between
4956 | Du et al.

Table 3. Stomatal density, stomatal size, stomatal aperture, epidermal cell size, minor vein density, and leaf area for tomato cultivars
Jinpeng and Zhongza grown under low (1.48 kPa) and high (2.55 kPa) VPD.

Cultivar Treatment Stomatal density Stomatal size Stomatal aperture Epidermal cell size Minor vein density Leaf area
(mm–2) (μm2) (μm) (μm2) (mm mm–2) (cm2)
Jinpeng Low VPD 275.5±7.8a 441.8±9.4a 1.34±0.11a 1764.2±53.5c 5.36±0.18a 70.3±6.4ab
High VPD 243.9±7.6b 376.6±9.8b 1.31±0.07a 1959.5±52.7ab 5.34±0.13a 63.5±3.7b
Zhongza Low VPD 250.7±7.3ab 481.0±13.8a 1.11±0.02b 1909.3±41.6bc 5.38±0.08a 84.5±5.3a
High VPD 197.9±5.6c 382.2±8.8b 0.96±0.03c 2081.9±65.8a 4.75±0.14b 73.7±3.9ab
Values of F(1,16) for two-way ANOVA
Cultivar 24.64** 4.42ns 17.22** 6.06* 4.12ns 6.16*
Treatment 34.94** 59.19** 1.49ns 11.53** 5.43* 3.17ns

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

Cultivar×Treatment 2.20ns 2.49ns 0.77ns 0.05ns 4.88* 0.17ns

Data are means ±SE (n=5 plants). Different letters within the same column denote statistically significant differences between treatments (Tukey test,
P<0.05). A two-way ANOVA was performed to test the effects of cultivar, treatment, and their interaction (**P<0.01, *P<0.05; ns, not significant).

Fig. 3. Representative micrographs of leaf veins (A−D), stomata on the abaxial epidermis (E−H), transverse leaf sections (I−L), mesophyll cell structure
(M−P), and chloroplast ultrastructure (Q−T) for tomato cultivars Jinpeng and Zhongza grown under low (1.48 kPa) and high (2.55 kPa) VPD. Scale bars:
500 μm for A−D; 50 μm for E−L; 10 μm for M−P; 1 μm for Q−T. (This fgure is available in colour at JXB online.)

leaf water supply and evaporative demand (Liu et al. 2015; (Sack and Frole 2006; Brodribb and Jordan 2011; Carins
Rodriguez-Dominguez et al. 2016; Brodribb and McAdam Murphy et al. 2012; Cardoso et al. 2018). In the present study,
2017). Previous studies have reported that this balance is linked the responses of Zhongza leaf minor vein density to VPD were
to plasticity in leaf minor vein density and stomatal density coordinated with those of stomatal density, whereas the leaves
 Leaf anatomy determines photosynthetic acclimation to VPD | 4957

Table 4. Mesophyll thickness, fraction of mesophyll tissue occupied by intercellular air spaces (fias), mesophyll surface area exposed to
intercellular air space (Sm/S), chloroplast surface area exposed to intercellular air space (Sc/S), ratio between chloroplast and mesophyll
surface area exposed to intercellular air space (Sc/Sm), and mesophyll cell size for tomato cultivars Jinpeng and Zhongza grown under
low (1.48 kPa) and high (2.55 kPa) VPD.

Cultivar Treatment Mesophyll thickness (μm) fias Sm/S (m2 m–2) Sc/S (m2 m–2) Sc/Sm Mesophyll cell size (μm2)
Jinpeng Low VPD 134.9±5.4a 0.51±0.03ab 10.6±0.4ab 8.7±0.5ab 0.83±0.03a 499.7±37.3a
High VPD 122.4±5.7a 0.47±0.01bc 9.6±0.4ab 7.8±0.3b 0.81±0.02a 519.1±33.6a
Zhongza Low VPD 123.6±5.2a 0.57±0.03a 11.9±1.0a 9.9±0.6a 0.84±0.02a 401.2±28.0b
High VPD 96.3±5.5b 0.42±0.01c 9.4±0.4b 7.5±0.4b 0.80±0.02a 496.9±32.5a
Values of F(1,16) for two-way ANOVA

Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024

Cultivar 11.81** 0.56ns 0.91ns 1.47ns 0.01ns 3.35ns
Treatment 13.35** 18.95** 8.31* 17.54** 1.54ns 3.05ns
Cultivar×Treatment 1.84ns 6.23* 1.81ns 3.34ns 0.12ns 1.34ns

Data are means ±SE (n=5 plants). Different letters within the same column denote statistically significant differences between treatments (Tukey test,
P<0.05). A two-way ANOVA was performed to test the effects of cultivar, treatment, and their interaction (**P<0.01, *P<0.05; ns, not significant).

Fig. 4. Relationship between stomatal and minor vein density (A), epidermal cell size and mesophyll cell size (B), stomatal density and epidermal cell
size (C), and vein density and mesophyll cell size (D) for tomato cultivars Jinpeng and Zhongza grown under low (1.48 kPa) and high (2.55 kPa) VPD.
Open circles, Jinpeng values under low VPD; closed circles, Jinpeng values under high VPD; open triangles, Zhongza values under low VPD; closed
triangles, Zhongza values under high VPD. When significant, regression lines are shown. Solid lines represent regressions for Jinpeng, and dashed lines
for Zhongza. For Jinpeng, stomatal density=475.83–0.12×epidermal cell size; vein density=7.05–0.0033×mesophyll cell size. For Zhongza, stomatal
density=69.82×minor vein density–129.38; epidermal cell size=1.31×mesophyll cell size+1406.30; stomatal density=575.91–0.17×epidermal cell size;
vein density=7.13–0.0046×mesophyll cell size.

of Jinpeng exhibited unchanged leaf minor vein density and
reduced stomatal density under high VPD (Fig. 4). Thus, the
decrease in stomatal aperture of Zhongza under high VPD may
be related to the proportional change in stomatal and leaf minor
vein density, because the reduced vascular system likely cannot
accommodate the increased evaporative demand despite a re-
duction in stomatal density and size (Sack and Holbrook 2006;
Carins Murphy et al. 2014). This supports the conclusion that
dynamic stomatal control maintained transpirational homeo-
stasis in T. ciliata plants grown under high and low VPD (Carins
Murphy et al. 2014). By contrast, the leaves of Jinpeng exhib-
ited unchanged minor vein density and leaf size and reduced
stomatal density under high VPD. The relative abundance of
veins compared with stomata could maintain water balance
and stomatal opening under high VPD. This adjustment maxi-
mizes photosynthetic returns for investment in stomatal and
vein infrastructure (Carins Murphy et al. 2017).
It has been suggested that coordinated changes in vein and sto-
matal density result from a common ‘dilution’ of veins and sto-
mata by expansion of the surrounding cells (Brodribb et al. 2013;
4958 | Du et al.
Fig. 5. Relationships between the fraction of mesophyll tissue occupied
by intercellular air spaces (fias) and chloroplast surface area exposed to
intercellular air spaces (Sc/S) (A) and mesophyll surface area exposed to
intercellular air spaces (Sm/S) (B) for tomato cultivars Jinpeng and Zhongza
grown under low (1.48 kPa) and high (2.55 kPa) VPD. Open circles,
Jinpeng values under low VPD; closed circles, Jinpeng values under high
VPD; open triangles, Zhongza values under low VPD; closed triangles,
Zhongza values under high VPD. The dashed lines indicate the total
regression for Jinpeng and Zhongza.
Fig. 6. Relationship between mesophyll conductance (gm) measured with
the method of Harley et al. (1992) and mesophyll conductance estimated
from leaf anatomical traits (gm-a) for tomato cultivars Jinpeng and Zhongza
grown under low (1.48 kPa) and high (2.55 kPa) VPD. Open circles,
Jinpeng values under low VPD; closed circles, Jinpeng values under high
VPD; open triangles, Zhongza values under low VPD; closed triangles,
Zhongza values under high VPD.
Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024
Fig. 7. Relationships between the inverse of chloroplast surface area
exposed to intercellular air space (Sc/S) and total resistance to CO2
diffusion (rcw + rpl + rcyt + ren + rst) for tomato cultivars Jinpeng and Zhongza
grown under low (1.48 kPa) and high (2.55 kPa) VPD. The axes are
log10-scaled. rcw, rpl, rcyt, ren, and rst represent CO2 diffusion resistance in
the cell wall, plasmalemma, cytosol, chloroplast envelope, and stroma,
respectively. rpl and ren are constants taken from previous work (Evans
et al. 1994; Tosens et al. 2012). Open circles, Jinpeng values under low
VPD; closed circles, Jinpeng values under high VPD; open triangles,
Zhongza values under low VPD; closed triangles, Zhongza values under
high VPD.
Carins Murphy et al., 2016). In the present study, the rela-
tionships between stomatal density and epidermal cell size, and
between minor vein density and mesophyll cell size, were sig-
nificant (Fig. 4), implying that synchronized changes to both
epidermal and mesophyll cell size are required to coordinate
vein and stomatal density. Moreover, mesophyll cell size was
significantly higher in Zhongza under high VPD than low
VPD, whereas it was similar in Jinpeng under both high and
low VPD. Epidermal cell size increased under high VPD in
both cultivars. Thus, the coordinated decrease in vein and sto-
matal density with increasing VPD in Zhongza is due to the
synchronized increase in epidermal and mesophyll cell size. In
contrast, the relationship between vein and stomatal density in
Jinpeng is decoupled because of disruption of synchronized
changes to cell size in the epidermis and mesophyll.
Additionally, a higher Kleaf, accompanied by unchanged minor
vein density, was observed in Jinpeng under high VPD, whereas
coordinated changes between Kleaf and minor vein density have
been found in Zhongza under high VPD in the present study
and in other studies (Sack and Frole 2006; Brodribb et al. 2007).
This contradiction may be ascribed to changes in the structural
characteristics of xylem conduits in the veins and/or aquaporin
expression in leaves (Sack and Holbrook 2006; Prado and Maurel
2013); this requires more detailed investigation in future work.
Regardless, an increased Kleaf is necessary to sustain high evap-
orative demand for the leaf to maintain open stomata and high
gs (Sack and Holbrook 2006).
Responses of gm to long-term high and low VPD are
mediated by leaf anatomical traits
A one-dimensional gas diffusion model was applied to esti-
mate mesophyll conductance based on leaf anatomical traits
 Leaf anatomy determines photosynthetic acclimation to VPD | 4959
Fig. 8. Resistance to CO2 diffusion across intercellular spaces (ias), cell wall (cw), cytoplasm (cyt), and stroma (st) in the leaves of tomato cultivars
Jinpeng (A) and Zhongza (B) grown under low (1.48 kPa) and high (2.55 kPa) VPD. *P<0.05 (Tukey test).
(Niinemets and Reichstein 2003; Tomás et al. 2013). The
strong linear correlation between gm-a and gm confirmed that
changes in gm under different VPD are primarily due to meso-
phyll architecture (Fig. 6). The variation of gm can be inter-
preted by dividing the CO2 diffusion pathway in mesophyll
into gas and liquid phases (Tomás et al. 2013).
Gas-phase CO2 diffusion is largely affected by fias and meso-
phyll thickness, which determine the path length of CO2 dif-
fusion from the stomatal cavities to the outer surfaces of cell
walls (Syvertsen et al. 1995; Niinemets and Reichstein 2003).
In this study, both fias and mesophyll thickness showed signifi-
cant decrease in Zhongza under high VPD, whereas in Jinpeng
they were unaffected by VPD (Table 4). Thus, it is not sur-
prising that rias was unaffected by VPD in Jinpeng, but the vari-
ation of rias was also non-significant in Zhongza. The pattern
in Zhongza—that the leaf with thinner mesophyll presented
lower fias—may help maintain conductance of CO2 diffusion
in the gas phase under stress conditions (Scippa et al. 2004;
Galmés et al. 2013). Nevertheless, lias accounted for only
3–4% of the total limitation in both cultivars under high VPD
(Supplementary Table S2), suggesting that liquid-phase limita-
tion, rather than gas-phase limitation, was primarily responsible
for the variation in gm, in agreement with previous findings (Lu
et al. 2016; Peguero-Pina et al. 2016).
The surface area for CO2 dissolution to reach the site of
carboxylation within the chloroplast is determined by Sc/S
(Niinemets and Reichstein 2003; Tomás et al. 2013; Ellsworth
et al. 2018). The changes of Sc/S generally result from adjust-
ments in Sc/Sm and Sm/S (Niinemets and Reichstein 2003).
However, no variations in Sc/Sm were observed in either
Jinpeng or Zhongza under different VPD conditions (Table 4).
It therefore appears that the influence of Sc/S on gm is medi-
ated by changes in Sm/S. Similar to the changes in Sc/S, Sm/S
was significantly reduced in Zhongza under high VPD, but it
was unaffected by VPD treatment in Jinpeng. It is noteworthy
that a significant correlation was found between fias and Sm/S
(Fig. 5). The leaves with lower fias had a higher leaf density and
ratio of palisade to spongy cell thickness (Supplementary Table
S3), which suggested that the low cell surface area in con-
tact with gas-phase CO2 may be due to denser cell packing
(Galmés et al. 2013).
Besides Sc/S, CO2 diffusion in the liquid phase is also influ-
enced by cellular components (Tomás et al. 2013). However,
Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024
cellular components seem to be less important than Sc/S for
explaining the responses of gm to VPD because log[(Sc/S)–1]
had a larger variation than log(rcw + rpl + rcyt + ren + rst) (Tosens
et al. 2012). In spite of this, rcyt was significantly increased in
Zhongza under high VPD (Fig. 8). Previous studies have attrib-
uted the increased rcyt to the increase of CO2 diffusion distance
in the cytosol (Tomás et al. 2013; Lu et al. 2016). The distance
from the chloroplasts to the cell wall (Lcyt) increased by 10% in
Jinpeng and 56% in Zhongza under high VPD (Supplementary
Table S4). Thus, the different Lcyt between Jinpeng and
Zhongza was responsible for the specific variation of rcyt in re-
sponse to high VPD. Sage and Sage (2009) suggested that Lcyt
was influenced by the proportion of the volume of each cell
occupied by the vacuole. Under high VPD, the low relative
volume of the vacuole might lead to increased Lcyt because
of water supply limitation in Zhongza, while chloroplasts are
pressed against the periphery of the cell by the large vacuole in
Jinpeng. Additionally, as argued by Evans et al. (2009), it is un-
certain whether the distance between the chloroplasts and cell
wall determined in those studies reflected the position of the
chloroplasts in vivo due to the possibility of fixation artefacts.
Future studies of cytosol resistance are needed to investigate its
function in gm variation under different VPD.
Conclusion
The results of this study suggested that the different photo-
synthetic responses to long-term high VPD in Jinpeng and
Zhongza were primarily due to differences in the responses of gs
and gm to high VPD, especially the former, which resulted from
differences in how leaf anatomy responds to high VPD. The gs
of Jinpeng under high VPD was maintained by regulating sto-
matal density independently from vein density, whereas the gs
in Zhongza was decreased primarily due to reduced stomatal
aperture induced by a proportional decrease in vein and sto-
matal density under high evaporative demand. Synchronized
changes to both epidermal and mesophyll cell size are required
to coordinate vein and stomatal density. Furthermore, no sig-
nificant differences were observed in gm and the mesophyll cell
structure of Jinpeng between high and low VPD. For Zhongza,
the changes in Sm/S and CO2 diffusion resistance in cytosol
were responsible for the reduction in gm under high VPD. This
4960 | Du et al.
study highlights the important role of plasticity in leaf anatomy
in photosynthetic responses to high VPD. The plasticity in leaf
anatomy could help optimize adaptation of plants to environ-
mental variations.
Supplementary data
Supplementary data are available at JXB online.
Table S1. Relative total dry mass, net assimilation rate, sto-
matal conductance, and mesophyll conductance for six tomato
cultivars grown under low and high VPD.
Table S2. Limitations of mesophyll conductance due to
intercellular space, cell wall, plasmalemma, cytoplasm, chloro-
plast envelope, and stroma for two tomato cultivars, Jinpeng
and Zhongza, grown under low and high VPD.
Table S3. Leaf mass per area, leaf thickness, leaf density, and
the ratio of palisade thickness to spongy thickness for two to-
mato cultivars, Jinpeng and Zhongza, grown under low and
high VPD.
Table S4. Chloroplast length, chloroplast thickness, cell wall
thickness, distance of chloroplasts from cell wall, distance be-
tween neighboring chloroplasts, and chloroplast number per
cell for two tomato cultivars, Jinpeng and Zhongza, grown
under low and high VPD.
Figure S1. Relationship between leaf area and the product
of leaf length and width for two tomato cultivars, Jinpeng and
Zhongza, grown under low VPD.
Acknowledgements
This work was financially supported by the Key Research and
Development Program of Shaanxi Province (2017ZDXM-NY-003),
China Agriculture Research System (CARS-23-C05), and National
Natural Science Foundation of China (31471916).
Conflicts of interest statement
The authors declare no conflicts of interest.
References
Aliniaeifard S, Malcolm Matamoros P, van Meeteren U. 2014. Stomatal
malfunctioning under low VPD conditions: induced by alterations in sto-
matal morphology and leaf anatomy or in the ABA signaling? Physiologia
Plantarum 152, 688–699.
Baldantoni D, Bellino A, Alfani A. 2016. Soil compost amendment en-
hances tomato (Solanum lycopersicum L.) quality. Journal of the Science of
Food and Agriculture 96, 4082–4088.
Bernacchi CJ, Portis AR, Nakano H, von Caemmerer S, Long SP.
2002. Temperature response of mesophyll conductance. Implications for
the determination of Rubisco enzyme kinetics and for limitations to photo-
synthesis in vivo. Plant Physiology 130, 1992–1998.
Bongi G, Loreto F. 1989. Gas-exchange properties of salt-stressed olive
(Olea europea L.) Leaves. Plant Physiology 90, 1408–1416.
Brodribb TJ, Feild TS, Jordan GJ. 2007. Leaf maximum photosyn-
thetic rate and venation are linked by hydraulics. Plant Physiology 144,
1890–1898.
Brodribb TJ, Jordan GJ. 2011. Water supply and demand remain
balanced during leaf acclimation of Nothofagus cunninghamii trees. New
Phytologist 192, 437–448.
Brodribb TJ, Jordan GJ, Carpenter RJ. 2013. Unified changes in cell
size permit coordinated leaf evolution. New Phytologist 199, 559–570.
Brodribb TJ, McAdam SAM. 2017. Evolution of the stomatal regulation of
plant water content. Plant Physiology 174, 639–649.
Brooks A, Farquhar GD. 1985. Effect of temperature on the CO2/O2 spe-
cificity of ribulose-1,5-bisphosphate carboxylase/oxygenase and the rate
of respiration in the light: Estimates from gas-exchange measurements on
spinach. Planta 165, 397–406.
Büssis D, von Groll U, Fisahn J, Altmann T. 2006. Stomatal aperture can
compensate altered stomatal density in Arabidopsis thaliana at growth light
conditions. Functional Plant Biology 33, 1037–1043.
Cardoso AA, Randall JM, Jordan GJ, McAdam SAM. 2018. Extended
differentiation of veins and stomata is essential for the expansion of large
leaves in Rheum rhabarbarum. American Journal of Botany 105, 1967–1974.
Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024
Carins Murphy MR, Dow GJ, Jordan GJ, Brodribb TJ. 2017. Vein
density is independent of epidermal cell size in Arabidopsis mutants.
Functional Plant Biology 44, 410–418.
Carins Murphy MR, Jordan GJ, Brodribb TJ. 2012. Differential leaf ex-
pansion can enable hydraulic acclimation to sun and shade. Plant, Cell &
Environment 35, 1407–1418.
Carins Murphy MR, Jordan GJ, Brodribb TJ. 2014. Acclimation to hu-
midity modifies the link between leaf size and the density of veins and sto-
mata. Plant, Cell & Environment 37, 124–131.
Carins Murphy MR, Jordan GJ, Brodribb TJ. 2016. Cell expansion not
cell differentiation predominantly co-ordinates veins and stomata within and
among herbs and woody angiosperms grown under sun and shade. Annals
of Botany 118, 1127–1138.
de Boer HJ, Price CA, Wagner-Cremer F, Dekker SC, Franks PJ,
Veneklaas EJ. 2016. Optimal allocation of leaf epidermal area for gas ex-
change. New Phytologist 210, 1219–1228.
de Marco I, Riemma S, Iannone R. 2018. Uncertainty of input param-
eters and sensitivity analysis in life cycle assessment: an Italian processed
tomato product. Journal of Cleaner Production 177, 315–325.
Dow GJ, Berry JA, Bergmann DC. 2014. The physiological importance
of developmental mechanisms that enforce proper stomatal spacing in
Arabidopsis thaliana. New Phytologist 201, 1205–1217.
Drake PL, Froend RH, Franks PJ. 2013. Smaller, faster stomata: scaling
of stomatal size, rate of response, and stomatal conductance. Journal of
Experimental Botany 64, 495–505.
Du QJ, Zhang DL, Jiao XC, Song XM, Li JM. 2018. Effects of atmos-
pheric and soil water status on photosynthesis and growth in tomato. Plant,
Soil and Environment 64, 13–19.
Ellsworth PV, Ellsworth PZ, Koteyeva NK, Cousins AB. 2018. Cell
wall properties in Oryza sativa influence mesophyll CO2 conductance. New
Phytologist 219, 66–76.
Evans JR, Caemmerer SV, Setchell BA, Hudson GS. 1994. The rela-
tionship between CO2 transfer conductance and leaf anatomy in transgenic
tobacco with a reduced content of Rubisco. Functional Plant Biology 21,
475–495.
Evans JR, Kaldenhoff R, Genty B, Terashima I. 2009. Resistances along
the CO2 diffusion pathway inside leaves. Journal of Experimental Botany 60,
2235–2248.
Fanourakis D, Heuvelink E, Carvalho SM. 2013. A comprehensive ana-
lysis of the physiological and anatomical components involved in higher
water loss rates after leaf development at high humidity. Journal of Plant
Physiology 170, 890–898.
Farquhar GD, von Caemmerer S, Berry JA. 1980. A biochemical model
of photosynthetic CO2 assimilation in leaves of C3 species. Planta 149,
78–90.
Fini A, Loreto F, Tattini M, Giordano C, Ferrini F, Brunetti C,
Centritto M. 2016. Mesophyll conductance plays a central role in leaf
functioning of Oleaceae species exposed to contrasting sunlight irradiance.
Physiologia Plantarum 157, 54–68.
Flexas J, Bota J, Escalona JM, Sampol B, Medrano H. 2002. Effects
of drought on photosynthesis in grapevines under field conditions: an evalu-
ation of stomatal and mesophyll limitations. Functional Plant Biology 29,
461–471.
Franks PJ, Beerling DJ. 2009. Maximum leaf conductance driven by CO2
effects on stomatal size and density over geologic time. Proceedings of the
National Academy of Sciences, USA 106, 10343–10347.
 Leaf anatomy determines photosynthetic acclimation to VPD | 4961
Galmés J, Conesa MÀ, Ochogavía JM, Perdomo JA, Francis DM,
Ribas-Carbó M, Savé R, Flexas J, Medrano H, Cifre J. 2011.
Physiological and morphological adaptations in relation to water use effi-
ciency in Mediterranean accessions of Solanum lycopersicum. Plant, Cell &
Environment 34, 245–260.
Galmés J, Ochogavía JM, Gago J, Roldán EJ, Cifre J, Conesa MÀ.
2013. Leaf responses to drought stress in Mediterranean accessions of
Solanum lycopersicum: anatomical adaptations in relation to gas exchange
parameters. Plant, Cell & Environment 36, 920–935.
Genty B, Briantais JM, Baker NR. 1989. The relationship between
the quantum yield of photosynthetic electron transport and quenching of
chlorophyll fluorescence. Biochimica et Biophysica Acta 990, 87–92.
Giuliani R, Koteyeva N, Voznesenskaya E, Evans MA, Cousins AB,
Edwards GE. 2013. Coordination of leaf photosynthesis, transpiration, and
structural traits in rice and wild relatives (genus Oryza). Plant Physiology
162, 1632–1651.
Grassi G, Magnani F. 2005. Stomatal, mesophyll conductance and bio-
chemical limitations to photosynthesis as affected by drought and leaf on-
togeny in ash and oak trees. Plant, Cell & Environment 28, 834–849.
Hanba YT, Shibasaka M, Hayashi Y, Hayakawa T, Kasamo K,
Terashima I, Katsuhara M. 2004. Overexpression of the barley aquaporin
HvPIP2;1 increases internal CO2 conductance and CO2 assimilation in the
leaves of transgenic rice plants. Plant & Cell Physiology 45, 521–529.
Harley PC, Loreto F, Di Marco G, Sharkey TD. 1992. Theoretical consid-
erations when estimating the mesophyll conductance to CO2 flux by analysis
of the response of photosynthesis to CO2. Plant Physiology 98, 1429–1436.
Kono Y, Nakata K, Tatsumi J. 1982. Observations of cross veins of the
second foliage leaf blade in the rice plant by use of a revised method for
clearing leaves. Japanese Journal of Crop Science 51, 445–454.
Lawlor DW, Tezara W. 2009. Causes of decreased photosynthetic rate
and metabolic capacity in water-deficient leaf cells: a critical evaluation of
mechanisms and integration of processes. Annals of Botany 103, 561–579.
Lawson T, Blatt MR. 2014. Stomatal size, speed, and responsiveness
impact on photosynthesis and water use efficiency. Plant Physiology 164,
1556–1570.
Leuschner C. 2002. Air humidity as an ecological factor for woodland
herbs: leaf water status, nutrient uptake, leaf anatomy, and productivity of
eight species grown at low or high VPD levels. Flora 197, 262–274.
Liu YY, Song J, Wang M, Li N, Niu CY, Hao GY. 2015. Coordination of
xylem hydraulics and stomatal regulation in keeping the integrity of xylem
water transport in shoots of two compound-leaved tree species. Tree
Physiology 35, 1333–1342.
Lu N, Nukaya T, Kamimura T, Zhang D, Kurimoto I, Takagaki M,
Maruo T, Kozai T, Yamori W. 2015. Control of vapor pressure deficit (VPD)
in greenhouse enhanced tomato growth and productivity during the winter
season. Scientia Horticulturae 197, 17–23.
Lu Z, Lu J, Pan Y, Lu P, Li X, Cong R, Ren T. 2016. Anatomical variation
of mesophyll conductance under potassium deficiency has a vital role in
determining leaf photosynthesis. Plant, Cell & Environment 39, 2428–2439.
McAdam SAM, Eléouët MP, Best M, et al. 2017. Linking auxin with
photosynthetic rate via leaf venation. Plant Physiology 175, 351–360.
Muir CD, Hangarter RP, Moyle LC, Davis PA. 2014. Morphological and
anatomical determinants of mesophyll conductance in wild relatives of to-
mato (Solanum sect. Lycopersicon, sect. Lycopersicoides; Solanaceae).
Plant, Cell & Environment 37, 1415–1426.
Mulcahy DN, Mulcahy DL, Dietz D. 2013. Biochar soil amendment in-
creases tomato seedling resistance to drought in sandy soils. Journal of Arid
Environments 88, 222–225.
Niinemets U, Díaz-Espejo A, Flexas J, Galmés J, Warren CR. 2009.
Role of mesophyll diffusion conductance in constraining potential photosyn-
thetic productivity in the field. Journal of Experimental Botany 60, 2249–2270.
Niinemets Ü, Reichstein M. 2003. Controls on the emission of plant
volatiles through stomata: a sensitivity analysis. Journal of Geophysical
Research: Atmospheres 108, 4211.
Peguero-Pina JJ, Sancho-Knapik D, Flexas J, Galmés J, Niinemets Ü,
Gil-Pelegrín E. 2016. Light acclimation of photosynthesis in two closely re-
lated firs (Abies pinsapo Boiss. and Abies alba Mill.): the role of leaf anatomy
and mesophyll conductance to CO2. Tree Physiology 36, 300–310.
Perez-Martin A, Flexas J, Ribas-Carbó M, Bota J, Tomás M,
Infante JM, Diaz-Espejo A. 2009. Interactive effects of soil water deficit
and air vapour pressure deficit on mesophyll conductance to CO2 in Vitis
vinifera and Olea europaea. Journal of Experimental Botany 60, 2391–2405.
Prado K, Maurel C. 2013. Regulation of leaf hydraulics: from molecular to
whole plant levels. Frontiers in Plant Science 4, 255.
Rodriguez-Dominguez CM, Buckley TN, Egea G, de Cires A,
Hernandez-Santana V, Martorell S, Diaz-Espejo A. 2016. Most stomatal
closure in woody species under moderate drought can be explained by sto-
matal responses to leaf turgor. Plant, Cell & Environment 39, 2014–2026.
Downloaded from https://academic.oup.com/jxb/article/70/18/4949/5506682 by Wageningen UR Library user on 02 July 2024
Rondeau-Mouro C, Defer D, Leboeuf E, Lahaye M. 2008. Assessment
of cell wall porosity in Arabidopsis thaliana by NMR spectroscopy.
International Journal of Biological Macromolecules 42, 83–92.
Sack L, Frole K. 2006. Leaf structural diversity is related to hydraulic cap-
acity in tropical rain forest trees. Ecology 87, 483–491.
Sack L, Holbrook NM. 2006. Leaf hydraulics. Annual Review of Plant
Biology 57, 361–381.
Sage TL, Sage RF. 2009. The functional anatomy of rice leaves: implica-
tions for refixation of photorespiratory CO2 and efforts to engineer C4 photo-
synthesis into rice. Plant & Cell Physiology 50, 756–772.
Scippa GS, Di Michele M, Onelli E, Patrignani G, Chiatante D,
Bray EA. 2004. The histone-like protein H1-S and the response of tomato
leaves to water deficit. Journal of Experimental Botany 55, 99–109.
Sharkey TD, Bernacchi CJ, Farquhar GD, Singsaas EL. 2007. Fitting
photosynthetic carbon dioxide response curves for C3 leaves. Plant, Cell &
Environment 30, 1035–1040.
Shirke PA, Pathre UV. 2004. Influence of leaf-to-air vapour pressure deficit
(VPD) on the biochemistry and physiology of photosynthesis in Prosopis
juliflora. Journal of Experimental Botany 55, 2111–2120.
Syvertsen JP, Lloyd J, McConchie C, Kriedemann PE, Farquhar GD.
1995. On the relationship between leaf anatomy and CO2 diffusion through
the mesophyll of hypostomatous leaves. Plant, Cell & Environment 18,
149–157.
Tomás M, Flexas J, Copolovici L, Galmés J, Hallik L, Medrano H,
Ribas-Carbó M, Tosens T, Vislap V, Niinemets Ü. 2013. Importance
of leaf anatomy in determining mesophyll diffusion conductance to CO2
across species: quantitative limitations and scaling up by models. Journal of
Experimental Botany 64, 2269–2281.
Tomeo NJ, Rosenthal DM. 2017. Variable mesophyll conductance among
soybean cultivars sets a tradeoff between photosynthesis and water-use-
efficiency. Plant Physiology 174, 241–257.
Tosens T, Niinemets Ü, Westoby M, Wright IJ. 2012. Anatomical
basis of variation in mesophyll resistance in eastern Australian sclero-
phylls: news of a long and winding path. Journal of Experimental Botany
63, 5105–5119.
Valentini R, Epron D, Deangelis P, Matteucci G, Dreyer E. 1995. In situ
estimation of net CO2 assimilation, photosynthetic electron flow and photo-
respiration in turkey oak (Q. cerris L) leaves: diurnal cycles under different
levels of water-supply. Plant, Cell & Environment 18, 631–640.
Warren C. 2006. Estimating the internal conductance to CO2 movement.
Functional Plant Biology 33, 431–442.
Warren CR. 2008. Soil water deficits decrease the internal conductance to
CO2 transfer but atmospheric water deficits do not. Journal of Experimental
Botany 59, 327–334.
Xu Z, Zhou G. 2008. Responses of leaf stomatal density to water status
and its relationship with photosynthesis in a grass. Journal of Experimental
Botany 59, 3317–3325.
Zhang D, Du Q, Zhang Z, Jiao X, Song X, Li J. 2017. Vapour pressure
deficit control in relation to water transport and water productivity in green-
house tomato production during summer. Scientific Reports 7, 43461.