Introduction

The Orchidaceae is one of the largest and most diverse families of flowering plants. With approximately 28,000 species spanning 736 genera, this family is widely distributed in various ecosystems throughout the world, except at the two Poles and in desert regions (Dressler 1993; Christenhusz and Byng 2016). However, because of their great economic importance to floral and pharmaceutical industries, as well as recent habitat losses, many of those species are now only locally distributed and becoming rare (Crain and Tremblay 2014). The photosynthetic potential is lower for most members of this family than for other angiosperm species (Assmann and Zeiger 1985; Hew and Yong 2004; Moreira et al. 2009; Zhang et al. 2015). Orchid plants usually grow slowly and have a long vegetative period (Shefferson 2006). This slower growth might be attributed to their reduced photosynthetic capacity (Obeso 2002). Although this low potential in some orchids might be due to crassulacean acid metabolism (Kerbauy et al. 2012), some C3 orchids also have this characteristic (Assmann and Zeiger 1985; Zhang et al. 2015). Therefore, the reasons for the low photosynthetic efficiencies in orchids remain unclear, especially our understanding of the correlation between leaf internal structure and photosynthetic capacity.

Although both Cypripedium and Paphiopedilum belong to the diandrous orchid subfamily Cypripedioideae (Cox et al. 1997; Cameron et al. 1999) and are closely related phylogenetically (Cox et al. 1997), the photosynthetic capacity of species in the former is approximately three to four times greater than that of species in the latter (Zhang et al. 2010; Chang et al. 2011). Both genera also have obvious differences in their leaf characters and geographical distribution. Cypripedium is a genus with ~ 50 species; they are perennial geophytes which are dormant in the winter. China is the main distribution area with 32 species and one variant at altitudes of above 1800 m in southwest China, including 24 endemic species. These species usually grow in humus-rich soils with good properties for absorbing and holding water in alpine grasslands, subalpine woodlands, and understory areas. However, Paphiopedilum are evergreen plants; there are 66 species, with 27 known species in China. Most of these species grow in karst limestone areas below an altitude of 2000 m with relatively poor soil and low water availability (Guan et al. 2011; Zhang et al. 2012). Therefore, these wide contrasts in morphology and photosynthesis, accompanied by their close relatedness, make Cypripedium and Paphiopedilum a valuable system for exploring the association of leaf structure with photosynthetic capacity in Cypripedioideae.

Investigating leaf morphology and anatomy is crucial to understanding why Orchidaceae often has such low photosynthetic potential, because those traits somewhat affect plant resource acquisition, gas diffusional resistance, physiological functions, and adaptations to environmental changes (Holbrook and Putz 1996). Leaf photosynthesis is determined by the concentration of CO2 inside the photosynthetic carboxylation sites and by the efficiency of plants to assimilate CO2 (Evans and von Caemmerer 1996; Muir et al. 2014). However, before atmospheric CO2 arrives at the carboxylation site, it must pass through a series of ‘physical barriers’, including the stomata, intercellular airspaces, cell walls, plasma membrane, cytosol, and chloroplast envelopes and stroma (Evans and von Caemmerer 1996; Flexas et al. 2012; Tomas et al. 2013; Muir et al. 2014). Hence, net CO2 assimilation is directly affected by CO2 diffusion resistance from ambient air to carboxylation sites in the chloroplasts (Farquhar and Sharkey 1982; Ball et al. 1987; Peguero-Pina et al. 2012). It is possible that the low photosynthetic rate found with Orchidaceae may be attributed to diffusional limitations that involve stomatal conductance (g s, the conductance of CO2 from the atmospheric environment into stomata) and mesophyll conductance (g m, the conductance of CO2 from the substomatal cavity to the carboxylation sites) to CO2 (Grassi and Magnani 2005; Gago et al. 2013). See the list of all abbreviations provided at the beginning of the paper. At each step in the diffusion pathway, leaf morphology and anatomy impact CO2 conductance (Flexas et al. 2006; Niinemets and Sack 2006; Muir et al. 2014). Because stomata represent important channels for this diffusion into the leaf, the pattern of stomatal distribution on the abaxial and adaxial leaf surfaces as well as stomatal density, aperture, and size are key anatomical traits that influence g s (Franks and Beerling 2009; Giuliani et al. 2013). For example, the unique stomatal structure and lack of chloroplasts in the guard cells can limit stomatal opening in Paphiopedilum, resulting in low photosynthetic performance (Zhang et al. 2010). Stomatal conductance is also correlated with leaf morphological traits, including leaf density (as leaf dry mass per unit area divided by leaf thickness) and thickness, leaf area, and abaxial and adaxial epidermis thicknesses (Dunbar-Co et al. 2009; Zhang et al. 2010, 2012). One potential strategy for improving performance is to increase stomatal density and reduce pore size because smaller stomata have greater sensitivity to environmental change (Franks and Farquhar 2007). Although researchers already recognize that orchids have lower stomatal density and larger stomata than other angiosperms (Guan et al. 2011; Zhang et al. 2015), it is still unclear how those traits affect photosynthesis in orchids.

The role that mesophyll conductance has in photosynthesis is a new focus of research that has produced results suggesting that g m is finite and variable, and is a major limiting factor (Grassi and Magnani 2005; Niinemets et al. 2005, 2009; Muir et al. 2014; Gago et al. 2016; Han et al. 2016). Resistance to mesophyll diffusion due to specific anatomical structures can be quantified in terms of leaf dry mass per unit area, leaf thickness, leaf density, surface area of the mesophyll cells exposed to intercellular airspace per unit of leaf area, and the fraction of the mesophyll occupied by intercellular airspace (Oguchi et al. 2003; Peguero-Pina et al. 2012; Tosens et al. 2012). However, the most powerful sources of g m variation are the amount of chloroplast surface area exposed to intercellular airspace per unit of leaf area and the cell wall thickness (Tholen and Zhu 2011; Carriqui et al. 2015; Tosens et al. 2016). This influence of leaf anatomy on g m differs by species (Flexas et al. 2014), but the impact of such anatomical parameters on g m and photosynthesis has rarely been tested in orchids.

Photosynthetic capacity can also be affected by biochemical factors, such as leaf N content, Rubisco activity, and N-partitioning into photosynthetic components. Being widely distributed in organelles that have photosynthetic functioning, Rubisco is the most abundant naturally occurring protein on earth. Consequently, it can directly affect net photosynthetic production, and its content and activity can modulate the rate of CO2 assimilation in plants (Sawada et al. 2003; Yamori et al. 2012). Nitrogen is an important raw material for chlorophyll and carboxylase, and leaf photosynthesis in numerous plants, including eucalyptus and orchid, is closely related to their leaf N contents (Laclau et al. 2004; Yamori et al. 2011).

Leaf photosynthetic performance is generally regulated through a combination of stomatal, mesophyll, and biochemical limitations (Grassi and Magnani 2005; Flexas et al. 2012; Tomas et al. 2013). For example, the biochemical limitation (L b) is the largest factor depressing photosynthesis in seven angiosperms and eight wild relatives of Solanum sect. Lycopersicon, followed by mesophyll conductance (L mc) and stomatal (L s) limitations, while L mc is the most important factor controlling fern photosynthesis, followed by L s and L b (Muir et al. 2014; Carriqui et al. 2015). Those findings indicate that the extent to which biochemical and diffusional factors affect photosynthetic performance varies across species (Terashima et al. 2006; Flexas et al. 2008, 2012). Despite these scientific advances, however, no investigations have been reported on the relative contribution of those three types of limitations to the low photosynthetic capacity of orchids.

We selected six closely related C3 orchid species (three each from Cypripedium and Paphiopedilum) and compared their rates of photosynthesis, as well as key anatomical and physiological traits. Our main objectives were to improve our understanding of (1) how leaf anatomical structures affect stomatal and mesophyll conductance and photosynthetic performances and (2) the relative roles of stomatal, mesophyll, and biochemical limitations to orchid photosynthesis.

Materials and methods

Plant materials and growing conditions

We investigated the association between photosynthesis and orchid leaf anatomy, diffusional resistance, and biochemistry, using three Cypripedium species (C. flavum, C. tibeticum, and C. yunnanense) and three Paphiopedilum species (P. armeniacum, P. micranthum, and P. dianthum). The photosynthetic type of these six selected species is C3 photosynthesis (Guan 2010). The aim of selecting those closely related species was to alleviate the effect of large morphological and physiological differences caused by phylogeny (Peguero-Pina et al. 2017). The ecological, habitat, and leaf characteristics of the six species are shown in Appendix S1. Because the optimal growth conditions are different between Paphiopedilum and Cypripedium (Guan et al. 2010; Chang et al. 2010), they were cultivated at two sites to ensure their best growth and physiological performances. The Paphiopedilum species were grown in a greenhouse in Kunming Botanical Garden (elevation 1990 m; 102°410′E, 25°01′N). The growth conditions included an air temperature of 20–25 °C (day) and 10–15 °C (night), 70–80% of relative humidity (RH), and with a maximum mid-day PPFD of approximately 800 µmol photons m−2 s−1. The three Cypripedium species were grown in Shangrila Alpine Botanical Garden (elevation 3260 m; 99°50′E, 27°48′N) with a temperature of 15–23 °C at day and approximately 10 °C at night, and RH ranging from 60 to 85% during the growing period. The maximum light intensity received by the plants was approximately 1000 µmol photons m−2 s−1. All sampling and measurements were conducted during their respective flowering periods in 2016, i.e., Paphiopedilum in September and Cypripedium in June.

Monitoring of photosynthetic gas exchange

All measurements of leaf gas exchange were performed on clear days by clamping fully expanded and healthy leaves from each species into 2-cm2 cuvettes in an open infrared gas exchange system with an integrated fluorescence chamber (LI-6400-40; Li-Cor, Lincoln, NE, USA). To ensure that the results from these two genera would be comparable, mature leaves (the species of Cypripedium at a leaf age of 60 days and Paphiopedilum at a leaf age of 2 years) were used for photosynthetic measurements under the optimum conditions established previously (Chang et al. 2011; Zhang et al. 2006). For photosynthetic measurements, RH was maintained at approximately 60%, vapor pressure deficit (VPD) at approximately 1.5 kPa, and leaf temperatures at 25 °C for Paphiopedilum and 20 °C for Cypripedium (see Guan et al. 2011; Chang et al. 2011 for temperatures used in previous studies). The leaf steady-state conditions of these two genera were induced for 25–30 min prior to the measurement period, using a chamber CO2 concentration of 380 µmol mol−1 and a saturating photosynthetic photon flux density of 400 µmol photons m− 2 s− 1 for Paphiopedilum and 800 µmol photons m−2 s−1 for Cypripedium.

Photosynthetic CO2 response curves were tested with CO2 concentrations ranging from 0 to 2000 µmol mol−1. The automatic protocol system for the LI-6400-40 was used to maintain a light intensity of 400 µmol photons m−2 s−1 for Paphiopedilum and 1200 µmol photons m−2 s−1 for Cypripedium (Zhang et al. 2006; Guan et al. 2011; Chang et al. 2011). Photosynthetic rates and chlorophyll fluorescence were recorded within a steady state by setting the wait time of each concentration at 3 min. Five leaves per species were sampled from different plants.

The photosynthetic response curves for CO2 were fitted by Photosyn Assistant software V1.1 (Dundee Scientific, UK), and the maximum rate of carboxylation (V c,max) was determined as described by Long and Bernacchi (2003). Using the A N–C i curves, mesophyll conductance was calculated according to the method of Harley et al. (1992):

$${g_{\text{m}}}~=\frac{{{A_{\text{N}}}}}{{~{{{C_{\text{i}}}~ - ~{\Gamma ^*}\left[ {ETR~+~8\left( {{A_{\text{N}}}~+~{R_{\text{d}}}} \right)} \right]} \mathord{\left/ {\vphantom {{{C_{\text{i}}}~ - ~{\Gamma ^*}\left[ {ETR~+~8\left( {{A_{\text{N}}}~+~{R_{\text{d}}}} \right)} \right]} {\left[ {ETR~ - 4\left( {{A_{\text{N}}}~+~{R_{\text{d}}}} \right)} \right]}}} \right. \kern-0pt} {\left[ {ETR~ - 4\left( {{A_{\text{N}}}~+~{R_{\text{d}}}} \right)} \right]}}~}},$$

where A N, C i, and ETR are the net photosynthetic rate, the CO2 concentration in substomatal cavities, and the rate of electron transport at the ambient CO2 concentration, respectively. Synchronized with the measurement of photosynthetic CO2 response curve, the value of ETR was determined by a portable photosynthesis analysis system with a LI-6400-40 fluorescence chamber. R d is the dark respiration rate and was measured after 30-min dark adaptation at daytime. The CO2 concentration at which net CO2 fixation offsets CO2 loss from photorespiration (Г *) was calculated from Rubisco specificity factor at a temperature of 20 °C in Cypripedium species and 25 °C in Paphiopedilum using the following equation (von Caemmerer 2000; Sharkey et al. 2007):

$${\Gamma ^*}~=~\frac{{0.5 \cdot \left[ {{{\text{O}}_2}} \right]}}{{{S_{{\text{c}}/{\text{o}}}}}},$$

where S c/o, the Rubisco specificity factor, was derived from Rubisco kinetics as

$${S_{{\text{c}}/{\text{o}}}}~=~\frac{{{V_{{\text{c}},\hbox{max} }}{K_{\text{o}}}}}{{{V_{{\text{o}},\hbox{max} }}{K_{\text{c}}}}},$$

where V o,max, the maximum RuBP saturated rate of oxygenation, equals 1/2 V c,max. K c and K o are Michaelis constants for CO2 and O2, respectively. Because K o and K c are temperature dependent, their values at given temperatures for each species (20 °C for Cypripedium species and 25 °C for Paphiopedilum) were calculated by an exponential function (Harley et al. 1992; Bernacchi et al. 2001; Long and Bernacchi 2003; Sharkey et al. 2007). The values of Rubisco specificity factor were 23.95 µmol mol−1 for Cypripedium species and 32.21 µmol mol−1 for Paphiopedilum species.

The value for chloroplast CO2 concentration (C c) was estimated by the following equation (Harley et al. 1992; Yamori et al. 2011):

$${C_{\text{c}}}~=~~{C_{\text{i}}}~ - ~\frac{{{A_{\text{N}}}}}{{{g_{\text{m}}}}}.$$

Analysis of quantitative photosynthetic limitations

According to the method proposed by Grassi and Magnani (2005), the quantitative limitations on photosynthesis can be partitioned into stomatal conductance, mesophyll conductance, and biochemical capacity, i.e., L s + L mc + L b = 1. Thus, the relative importance of stomatal diffusion, mesophyll diffusion, and photosynthetic biochemistry to the photosynthetic potential of each species was calculated as follows:

$${L_{\text{s}}}~=~\frac{{{g_{{\text{tot}}}}/{g_{\text{s}}} \cdot \partial {A_{\text{N}}}/\partial {C_{\text{c}}}}}{{{g_{{\text{tot}}}}~+~\partial {A_{\text{N}}}/\partial {C_{\text{c}}}}},$$
$${L_{{\text{mc}}}}~=~\frac{{{g_{{\text{tot}}}}/{g_{\text{m}}} \cdot \partial {A_{\text{N}}}/\partial {C_{\text{c}}}}}{{{g_{{\text{tot}}}}~+~\partial {A_{\text{N}}}/\partial {C_{\text{c}}}}},$$
$${L_{\text{b}}}~=~\frac{{{g_{{\text{tot}}}}}}{{{g_{{\text{tot}}}}~+~\partial {A_{\text{N}}}/\partial {C_{\text{c}}}}},$$

where g tot is the total conductance of CO2 from ambient air to carboxylation sites (1/g tot = 1/g s + 1/g m) (Grassi and Magnani 2005). The photosynthetic limitation was calculated using the values of A N and C c at the ambient CO2 concentration.

Further, we have calculated the limitations of the two biochemical processes (RuBP carboxylation or RuBP regeneration) on photosynthesis (Yamori et al. 2011). Chloroplast CO2 concentration at which the transition from Rubisco to RuBP regeneration limitation occurs (C transition) was calculated as

$${C_{{\text{transition}}}}~=~\frac{{{K_{\text{c}}}\left( {1~+O/{K_{\text{o}}}} \right){J_{\hbox{max} }}/4{V_{{\text{c}},\hbox{max} }}~ - 2{\Gamma ^*}}}{{1~ - {J_{\hbox{max} }}/4{V_{{\text{c}},\hbox{max} }}}},$$

where O and J max are the intercellular O2 concentration and the maximum rate of electron transport, respectively.

Leaf anatomical traits

After the gas exchange parameters were examined, the middle parts of the same leaves used in those measurements were fixed in FAA (95% ethanol:distilled water:formaldehyde:glacial acetic acid, 10:7:2:1, v/v/v/v) for at least 24 h. These samples were cleaned with water prior to anatomical analysis. Transverse sections (18–40 µm) were made with a Microtome Cryostat (CM3050S; Leica, Germany) and then examined and photographed with a light microscope (U-CMAD3; Olympus Inc., Tokyo, Japan).

For further investigation, 1 × 1 mm pieces were excised from the portion of the leaf that was enclosed in the chamber. Those pieces were quickly infiltrated in the fixative 2.5% glutaraldehyde for at least 12 h and then post-fixed in 1% osmic acid for 2 h before being further dehydrated in an ethanol and acetone series. After the samples were embedded in LR White resin for 1 h, the cross sections (40 nm) were obtained with an ultramicrotome (Leica-R) and then stained with uranyl acetate and lead citrate. The ultrathin sections were viewed at the magnifications of ×1800 to ×40,000 by a JEOL JEM-1011 transmission electron microscope (JEOL Ltd. Tokyo, Japan) and photographed with an Olympus-SIS Megaview digital camera (Olympus Soft Imaging Solutions GmbH, Münster, Germany).

The images were used for determining the following parameters, using the ImageJ software package: leaf thickness (LT), mesophyll layer thickness (MT), adaxial and abaxial cuticle thicknesses (CTad and CTab), adaxial and abaxial epidermis thicknesses vertically (ETad and ETab), mesophyll cell length and width, chloroplast thickness (T chlor), cell wall thickness (T cw), total length of mesophyll cell perimeter facing the intercellular airspace (L mes), and total length of chloroplast perimeter facing the intercellular airspace (L c). Afterward, the mesophyll surface area exposed to intercellular airspace per unit leaf area (S mes/S) and chloroplast surface area exposed to intercellular airspace per unit of leaf area (S c/S) were calculated by applying the method of Tosens et al. (2012):

$${S_{{\text{mes}}}}/S~=~\frac{{{L_{{\text{mes}}}}}}{W}\gamma ,$$
$${S_{\text{c}}}/S~{\text{=~}}\frac{{{L_{\text{c}}}}}{{{L_{{\text{mes}}}}}}{S_{{\text{mes}}}}/S,$$

where W is the width of the sections measured and γ is the curvature correction factor for each species, calculated as described by Thain (1983).

The fraction of intercellular airspace (ƒias) was determined as

$${f_{{\text{ias}}}}~=~1 - ~~\frac{{\mathop \sum \nolimits^{} {S_{\text{s}}}}}{{{\text{MT}} \cdot {\text{W}}}},$$

where \(\sum {\text{S}}_{\text{s}}\) is the sum of cross-sectional areas for mesophyll cells.

For a given section, all parameters were determined from at least eight fields of view, and at least five sections per species were analyzed.

Leaf morphology

The abaxial epidermis from the middle part of a mature leaf was coated with a thin layer of colorless, transparent nail polish. After the polish dried, the film was gently torn away from the leaf surface with tweezers. These films were mounted on a microscope slide and images were taken with the light microscope. Randomly selected images (40 per species) were used to obtain the stomatal number and stomatal density (SD). Stomatal length (SL) and width (SW), referring to the length and width of the guard cells, were measured for 80 stomata, from which the stomatal aperture area (SA) could then be calculated as 1/4 × π × SL × SW (James and Bell 2001). Pore depths were determined by examining those transverse sections (Lawson et al. 1998).

Leaf areas of the mature leaves used for monitoring gas exchange were measured with a leaf portable area meter (LI-3000A; Li-Cor). Afterward, the samples were oven-dried at 80 °C to a constant weight. The leaf dry mass was then weighed on a Mettler-Toledo analytical balance (ME 204; China) so that leaf dry mass per unit area (LMA) could be calculated. Then the leaf total nitrogen content (Nmass and Narea) was measured using an elemental analyzer (Elementar Analysensysteme GmbH, Vario EL III, Hanau, Germany). And the photosynthetic nitrogen use efficiency (PNUE) was determined as

$${\text{PNUE~=~}}\frac{{{A_N}}}{{{N_{{\text{area}}}}}},$$

where A N is the maximum photosynthetic rate at 380 µmol mol−1 CO2 concentration and Narea is the leaf nitrogen content per unit area.

Statistical analysis

Statistical analyses were performed with the IBM SPSS 21.0 software package (SPSS Inc., Chicago, IL, USA). Differences in leaf anatomical traits, photosynthetic parameters, and biochemical traits among species were determined by one-way ANOVA and Tukey’s multiple comparison tests. A Pearson correlation analysis was used to evaluate the associations between parameters. Graphic images were produced using the Sigma Plot 10.0 package (Systat Software Inc., Richmond, CA, USA).

Results

Leaf morphologies and anatomies of Cypripedium and Paphiopedilum

Most of the leaf morphological and anatomical characteristics differed significantly between our selected species of Cypripedium and Paphiopedilum (Table 1, Appendix S2). Compared with Cypripedium, Paphiopedilum species had thicker leaves (LT = 1062 ± 85 µm) due to their larger adaxial epidermis cells (Ead) and mesophyll thickness (MT). The values for ETad and MT of Paphiopedilum were sixfold and threefold higher, respectively, than that for Cypripedium. Stomatal density (SD) and substomatal cavity area (Asc) did not differ between the two genera. However, pore depth (PD) and substomatal cavity depth (SCD) were significantly larger for Paphiopedilum than for Cypripedium, while the area of individual stomata (As) was 2.2-fold higher.

Table 1 Differences in leaf anatomical and photosynthetic characteristics among Cypripedium and Paphiopedilum species based on t tests of independent samples
Full size table

Leaf anatomical traits were significantly different between genera (Table 1, Appendix S2, Figs. 1, 2). For example, the cell wall thickness (T cw) was approximately 2.6-fold higher for Paphiopedilum than for Cypripedium, whereas the values calculated for the surface of mesophyll cells (S mes/S) and the chloroplast (S c/S) intercellular airspaces per leaf area were much larger for Cypripedium than for Paphiopedilum. Although the proportion of exposed chloroplast to mesophyll surface areas (S c/S mes) was 62% for Cypripedium but 52% for Paphiopedilum, no significant difference between genera was found for the intercellular airspaces (ƒ ias).

Fig. 1
figure 1

Transmission electron micrographs of transverse sections from adaxial mesophyll cells of Cypripedium and Paphiopedilum: a C. flavum, b C. tibeticum, c C. yunnanense, d P. armeniacum, e P. micranthum, and f P. dianthum. C chloroplast, IAS intercellular airspace, SG starch grains

Full size image
Fig. 2
figure 2

Transmission electron micrographs of transverse sections from cell walls of Cypripedium and Paphiopedilum: a C. flavum, b C. tibeticum, c C. yunnanense, d P. armeniacum, e P. micranthum, and f P. dianthum. CW cell wall, C chloroplast, SG starch grains

Full size image

The leaf N content (Nmass) of Cypripedium was about twofold higher than that of Paphiopedilum. Conversely, the value of Narea was 1.20 ± 0.05 g m−2 for Cypripedium, but 1.90 ± 0.14 g m−2 for Paphiopedilum. In addition, there were obvious differences in leaf dry mass per unit area (LMA) and photosynthetic nitrogen use efficiency (PNUE) between Cypripedium (47.4 ± 1.4 g m−2 and 7.68 ± 0.89 µmol s−1 CO2 g−1 N, respectively) and Paphiopedilum (151.5 ± 15.6 g m−2 and 1.50 ± 0.28 µmol s−1 CO2 g−1 N, respectively) (Table 1).

Photosynthetic performances of Cypripedium and Paphiopedilum

At ambient CO2, these six species differed greatly in their photosynthetic capacities (Table 1; Fig. 3, Appendix S3), with the net CO2 assimilation rate being approximately 3.3-fold higher for members within Cypripedium. In particular, the A N values range from 10.4 µmol CO2 m−2 s−1 for C. flavum down to only 1.94 µmol CO2 m− 2 s− 1 for P. dianthum (Fig. 3). Similarly, the electron transport rate (ETR) was approximately 2.2-fold higher for Cypripedium (Fig. 4b). However, the maximum carboxylation rate (V c,max) did not differ significantly between genera. Stomatal conductance was much higher for Cypripedium and the value of mesophyll conductance was approximately fivefold higher in that genus. Finally, mean values calculated for the substomatal and chloroplastic CO2 concentrations were slightly higher in species of Cypripedium.

Fig. 3
figure 3

Light-saturated rate of CO2 assimilation at 380 µmol mol−1 CO2 concentration (A N) for six species from Cypripedium and Paphiopedilum. Values are means ± SE (n = 5)

Full size image
Fig. 4
figure 4

Response of CO2 assimilation rate (A N) (a) and electron transport rate (ETR) (b) to incident intercellular CO2 concentration (C i) in Cypripedium (black) and Paphiopedilum (white). Values are means ± SE (n = 5)

Full size image

Correlations between photosynthesis and leaf morphology and anatomy

Across species, several significant correlations were found between leaf structural traits, diffusional conductance, and photosynthetic rates (Figs. 5, 6, and Appendix S4). For example, A N was strongly and positively correlated with g s (r = 0.997; Fig. 5a) and g m (r = 0.996; Fig. 6a). In addition, A N was significantly associated with leaf morphological and anatomical traits. For example, it was correlated negatively with LT, ETad, ETab, MT, PD, SCD, and LMA, but positively with As, S mes/S, and S c/S. We also found that g s was strongly correlated with As and PD (r = 0.95 and − 0.96, respectively), and g m was strongly correlated with S mes/S, S c/S, and T cw (r = 0.99, 0.98, and − 0.81, respectively). In addition, g s was negatively related to LT, ETab, and LMA, while g m had a negative correlation with LT, MT, and LMA.

Fig. 5
figure 5

Correlations between CO2 assimilation rate, stomatal conductance, and leaf structure: a stomatal conductance (g s) versus CO2 assimilation rate (A N); b g s versus leaf thickness (LT); c g s versus stomatal density (SD); d g s versus stomatal apparatus area (As); e g s versus pore depth (PD); and f g s versus substomatal cavity depth (SCD). Values are means ± SE (n = 5) for Cypripedium (black dots) and Paphiopedilum (white dots)

Full size image
Fig. 6
figure 6

Correlations between CO2 assimilation rate, mesophyll conductance, and leaf structure: a mesophyll conductance (g m) versus CO2 assimilation rate (A N); b g m versus leaf thickness (LT); c g m versus mesophyll layer thickness (MT); d g m versus cell wall thickness (T cw); e g m versus surface area of mesophyll cells exposed to airspace per unit of leaf area (S mes/S); and f g m versus chloroplast surface area exposed to intercellular airspace per unit of leaf area (S c/S). Values are means ± SE (n = 5) for Cypripedium (black dots) and Paphiopedilum (white dots)

Full size image

We conducted a photosynthetic quantitative limitation analysis to examine the relative impacts of stomatal, mesophyll conductance, and biochemical limitations on the photosynthetic potential of two genera (Fig. 7). Among those three, L mc had the greatest influence on Cypripedium (55.9% of the total) and Paphiopedilum (59.7%), while L s had the least impact, i.e., 12.3% for Cypripedium and 19.8% for Paphiopedilum. Finally, the total diffusional limitation (L s + L mc) on photosynthesis was larger for Paphiopedilum (79.5%) than for Cypripedium (68.2%). We also analyzed the limitations of two biochemical processes on photosynthesis and found that C c for A 380 (A N at 380 mmol mol−1 CO2 concentrations, Table 1) was less than the C transition (197.2 µmol mol− 1 for Paphiopedilum, but 116.9 µmol mol−1 for Cypripedium), indicating that CO2 assimilation was limited by RuBP carboxylation in both genera.

Fig. 7
figure 7

Relative stomatal (L s), mesophyll conductance (L mc), and biochemical (L b) limitations to photosynthesis in Cypripedium and Paphiopedilum

Full size image

Discussion

We believe that this study is the first report to explore why orchids have low photosynthetic potential by using an integrated approach. Using six species of Cypripedium and Paphiopedilum, we demonstrated that leaf anatomy and morphology play key roles in regulating leaf diffusional conductance, thereby affecting photosynthetic performance. Our results indicated that many of the leaf morphological, anatomical, and physiological traits measured here differed significantly between these closely related genera. Compared with Cypripedium, Paphiopedilum species had much thicker leaves and cell walls, expanded adaxial epidermal cells, very large mesophyll cells, greater pore depth, and higher leaf dry mass per unit area, but smaller area for the stomatal apparatus, less surface area for mesophyll cells, and reduced intercellular airspaces in the chloroplasts per leaf area. Meanwhile, Cypripedium had a reduced T cw and larger S mes/S and S c/S, but significantly higher stomatal conductance, mesophyll conductance, leaf nitrogen content, photosynthetic nitrogen use efficiency, electron transport rate, and photosynthetic capacity. Some of these findings are in accordance with the results reported previously (Chang et al. 2011; Guan et al. 2011). These obvious differences in leaf traits between the closely related genera reflect the adaptation to their habitats. The natural growth conditions of Cypripedium are usually characterized by nutrient-rich soils and high soil moisture content during the growing season. Conversely, the natural habitats of Paphiopedilum in the karst limestone area are accompanied by low availability of substrates for storing water and nutrients. Paphiopedilum has features which may enable growth in moderate xerophytic conditions due to periodic shortages of water; this includes large epidermal cells, a thicker layer of mesophyll cells, a thicker cuticle, small stomata, high leaf dry mass per unit area, a relatively low leaf N content, and photosynthetic nitrogen use efficiency. These features could prevent excessive transpiration, support storage of water, with investment of more resources (C and N) into the construction and maintenance of leaves and less into the photosynthetic apparatus. We believe that the lower photosynthetic capacity of Paphiopedilum was caused by the lower values of Nmass, PNUE, ETR, g s, and g m compared with Cypripedium. And all of these observations provide evidence that differences among species in their leaf structures and photosynthetic performances are indicative of their abilities to adapt to their environments.

Our results showed that leaf morphological and anatomical structures have crucial roles in regulating diffusional resistance and photosynthetic performance for Cypripedium and Paphiopedilum. This may occur because such traits control the diffusion of gas between the outside and the inside of the leaf (Niinemets and Sack 2006; Boyer 2015). Stomata are keys in determining internal CO2 concentrations (Farquhar and Sharkey 1982; Araujo et al. 2011; Taylor et al. 2012). We found that the g s value was lower in Paphiopedilum than in Cypripedium, and g s was positively correlated with A N across the six species. This relationship is in accordance with the theoretical basis of the Ball–Berry model (Ball et al. 1987). Our analysis of photosynthetic limitations confirmed that stomatal factors were stronger in Paphiopedilum. A positive correlation between maximum A N and g s has been observed in other species and various habitats (Patakas et al. 2003; Gago et al. 2013, 2016). The lower g s of Paphiopedilum means that this genus displays greater stomatal resistance that reduces the concentration of CO2 in the chloroplasts and leads to a lower photosynthetic rate. It is commonly thought that g s is mainly affected by both stomatal density/size and whether the stomata are open (Franks and Beerling 2009; Ocheltree et al. 2012; de Boer et al. 2016). In contrast, our results indicated that g s is strongly linked to As, PD, SCD, and LMA, but is not significantly correlated with SD. Therefore, we might conclude that As and PD are the most conspicuous stomatal features that determine the degree of stomatal opening and the distance that CO2 diffuses through the stomata, respectively, thereby jointly affecting the leaf internal CO2 concentration. Some previous studies have also shown that As is closely correlated with g s (Giuliani et al. 2013; Brodribb et al. 2016). Overall, our results suggest that stomatal area and pore depth, rather than stomatal density, are the most important factors that influence g s in our selected species.

Mesophyll conductance plays an important role in determining the CO2 concentration at the sites of carboxylation inside the chloroplasts. Other researchers have also demonstrated that g m is an important factor for the limitation of photosynthesis (Flexas et al. 2008, 2012; Sagardoy et al. 2010) and is positively correlated with the photosynthetic rate (Loreto et al. 2003; Singsaas et al. 2004; Grassi and Magnani 2005; Warren and Adams 2006). For example, g m is responsible for the lower photosynthetic capacity in ferns when compared with angiosperms (Carriqui et al. 2015). We also noted the role that g m has in controlling photosynthesis in our tested species and found that Paphiopedilum has a higher mesophyll limitation than does Cypripedium. The positive relationship between photosynthetic rate and mesophyll conductance for Cypripedium and Paphiopedilum growing in their natural distribution areas was also confirmed by our study (Appendix S5) and a previous study (Chang et al. 2011), respectively. This has also been revealed elsewhere. For example, Tomas et al. (2013) reported that g m is highly correlated with S c/S and T cw in 15 fern species, but that the effects of leaf anatomical features on g m vary among species. For plants with mesophytic leaves, membrane permeability, the cytosol, and stromal conductance dictate g m, while T cw is the most important limitation for plants with sclerophytic leaves. In fact, some models have estimated that approximately 25–50% of the variability in g m can be explained by differences in cell wall thickness (Evans et al. 2009; Tosens et al. 2012). However, other researchers have suggested that the majority of g m variation is due to the combination of three parameters: S mes/S, S c/S, and T cw (Fini et al. 2016; Xiong et al. 2016). Our evaluation of six orchid species indicated that g m increased linearly with S mes/S and S c/S but decreased linearly with T cw. We were surprised to learn that g m was also related to LT, MT, and LMA. While leaf thickness and cell wall thickness affect the length of the CO2 diffusion path, S mes/S and S c/S determine the number of photosynthetic carboxylation sites adjacent to the intercellular airspace (Flexas et al. 2012; Tosens et al. 2012). These relationships collectively demonstrate that the leaf anatomical structure is associated with the internal diffusion of CO2 conductance and photosynthetic performances of species within Paphiopedilum and Cypripedium.

In summary, the combination of stomatal, mesophyll, and biochemical factors help determine the photosynthetic potentials of Paphiopedilum and Cypripedium. However, photosynthetic capacity is more strongly influenced by diffusional limitations than by biochemical limitations. Mesophyll conductance plays a vital role in the performance of the six orchids tested here. Both stomatal apparatus area and pore depth are the key anatomical traits that affect stomatal conductance, while mesophyll conductance is modulated by the surface area of mesophyll cells exposed to intercellular airspace per unit of leaf area, the chloroplast surface area exposed to intercellular airspace per unit of leaf area, and cell wall thickness. Therefore, the lower photosynthetic rate in Paphiopedilum species can largely be explained by diffusional resistance, particularly internal resistance which is imposed by their unique leaf anatomy and morphology. These findings provided new insight into the evolution of leaf structures, photosynthetic physiology, and slow growth rates by orchids. However, the phenotypes of plants are a consequence of differences in their genotypes and response to their environments; thus, further studies on structural and biochemical differences between representative species in these genera growing in their natural habitats will be of interest.