Effect of Bursaphelenchus xylophilus infection on leaf photosynthetic characteristics and resource‐use efficiency of Pinus massoniana

Abstract Pine wilt disease (PWD) is considered as the most destructive forest‐invasive alien species in China. We measured gas exchange parameters and foliar carbon isotope ratios (δ13C) of different infection phases of Masson pine in order to investigate the effect of Bursaphelenchus xylophilus infection on photosynthetic responses and resource‐use efficiency. The results showed that net photosynthetic rate (P n), transpiration rate (T), stomatal conductance (g s), and internal CO 2 concentrations (C i) decreased in the infested trees at photosynthetic photon flux density (PPFD) levels from 0 to 2,000 μmol m−2 s−1 compared with controls. The maximum net photosynthetic rate (P max) was significantly declined in the infected trees than in controls (p < .05). There also exist significant differences in dark respiration rate (R d) among different infection phases (p < .05), but the value is highest in the middle infection phase, followed by the control and then the terminal infection phase. This indicates that Pinus massoniana plants need to consume more photosynthetic products during the middle infection phase in order to defend against pine sawyer beetle feeding and PWD infection. Isotopic analysis revealed a significant decrease of the foliar δ13C (p < .05), as much as 2.5‰ lower in the infected trees. The mean leaf N content was about 12.94% less in the middle infection phase and 27.06% less in the terminal infection phase, causing a significant increase of the foliar C:N ratio in infested trees. Both of the net photosynthetic rates and foliar δ13C were linearly correlated with the foliar N content. We also found a significant decrease (p < .05) of resource‐use efficiency in PWD‐induced P. massoniana plants, which can be attributed to the closure of stomatal pores and the inactivation or loss of both Rubisco and other key Calvin cycle enzymes. This study highlights the impact of photosynthetic characteristics, foliar carbon isotope ratios, and resource‐use efficiency of PWD‐induced trees, which can help identify PWD infestations at the photosynthetic and physiological levels so as to better facilitate management actions.

ration rate (T), stomatal conductance (g s ), and internal CO 2 concentrations (C i ) decreased in the infested trees at photosynthetic photon flux density (PPFD) levels from 0 to 2,000 μmol m −2 s −1 compared with controls. The maximum net photosynthetic rate (P max ) was significantly declined in the infected trees than in controls (p < .05).
There also exist significant differences in dark respiration rate (R d ) among different infection phases (p < .05), but the value is highest in the middle infection phase, followed by the control and then the terminal infection phase. This indicates that Pinus massoniana plants need to consume more photosynthetic products during the middle infection phase in order to defend against pine sawyer beetle feeding and PWD infection. Isotopic analysis revealed a significant decrease of the foliar δ 13 C (p < .05), as much as 2.5‰ lower in the infected trees. The mean leaf N content was about 12.94% less in the middle infection phase and 27.06% less in the terminal infection phase, causing a significant increase of the foliar C:N ratio in infested trees. Both of the net photosynthetic rates and foliar δ 13 C were linearly correlated with the foliar N content.
We also found a significant decrease (p < .05) of resource-use efficiency in PWDinduced P. massoniana plants, which can be attributed to the closure of stomatal pores and the inactivation or loss of both Rubisco and other key Calvin cycle enzymes. This study highlights the impact of photosynthetic characteristics, foliar carbon isotope ratios, and resource-use efficiency of PWD-induced trees, which can help identify PWD infestations at the photosynthetic and physiological levels so as to better facilitate management actions.

K E Y W O R D S
Bursaphelenchus xylophilus, carbon isotope ratios, gas exchange, photosynthesis, Pinus massoniana

| INTRODUCTION
Pine wilt disease, which is caused by the pine wood nematode Bursaphelenchus xylophilus (Steiner and Buhrer) Nickle (Nematoda: Aphelenchoididae), was first introduced to China in 1982 and identified as the leading forest pest and disease (Shi et al., 2006;Wan, Zheng, & Guo, 2005). What is worse, nearly all the native pine species are highly susceptible to the invasion of B. xylophilus within the current PWD distribution areas in China. Masson pine (Pinus massoniana Lamb), belonging to the Pinaceae family, is a native species in China and can grow well in dry, sandy soils and arid climates (Zhao, Futai, Sutherland, & Takeuchi, 2008). Pinus massoniana is considered as the pioneer tree species for afforestation and is widely distributed across 19 provinces in the central and southern parts of China (Gao, Shi, Huang, Wang, & Luo, 2015). Green needles of P. massoniana trees affected by PWD turn to red and wilt (Figure 1), leading to the rapid death of the host tree. In China, it is difficult to completely eradicate this disease because pines will die within 2-3 months after being infected by B. xylophilus. The problem is exacerbated by a lack of effective control measures for the spread of pine wood nematodes (Gao et al., 2015;Hu et al., 2012). Therefore, the B. xylophilus invasion has caused huge economic losses in the timber industry as well as important ecological consequence for the forest ecosystem (Gao et al., 2015;Shi et al., 2007;Yu, Xu, & Ding, 2011).
In China, the pine sawyer beetle (Monochamus alternatus Hope) is the vector of B. xylophilus and can facilitate the nematode to spread from infected host trees to the healthy ones. When it arrives on a healthy tree, B. xylophilus can enter the resin canals through the broken canals in the cortex and xylem caused by the feeding pine sawyer beetles. Then, they can rapidly spread inside the infected hosts, up to 150 cm per day (Zhao et al., 2008). In addition, B. xylophilus can also exude cellulases and pectinases in order to kill the contents of living cells in the host trees (Kikuchi, Jones, Aikawa, Kosaka, & Ogura, 2004).
This rapid diffusion combined with the fast reproduction of B. xylophilus leads to the dysfunction of the xylem and tracheid, which blocks the sap ascent and water transportation from the roots to the crown (Zhao et al., 2008).
In plants, infection by pests or diseases leads not only to the induction of defense systems but also to changes in photosynthetic activity.
The reaction of host photosynthesis is an important physiological parameter and very sensitive to external stimulation, such as the invasion of pests, disease, and climate changes. To better understand the host reaction to the development of pine wilt disease, measurements of the changes in physiology and photosynthetic characteristics of the host are important (Zhao et al., 2008). Previous studies have demonstrated that the host's resistance mechanism to the B. xylophilus is a change in photosynthesis (Fukuda, 1997;Melakeberhan, Toivonen, Vidaver, Webster, & Dube, 1991;Woo et al., 2010). Woo et al. quanti-fied decreases in the instantaneous gas exchange rate in the seedlings of Pinus densiflora inoculated with B. xylophilus. However, quantification of the continuous photosynthetic responses and resource-use efficiency with certain environmental variables, such as CO 2 or photosynthetic photon flux density (PPFD), can certainly provide a clearer picture of PWD-induced photosynthetic activity.
The plant resource-use efficiencies, which are influenced by light, temperature, humidity, water stress, and environmental stimulation (Field, Merino, & Mooney, 1983), change dramatically during the various life stages of an individual plant, as well as between different plant species. In addition, Dawson et al. (2002) suggested that carbon, water, and nitrogen are three of the most important resources that have a significant positive correlation with plant function, growth, and the biogeochemical cycles in which plants participate. In the case of PWD, relatively little is known about the impact of the infection on resource-use efficiency of the host trees.
Although many studies have been conducted on the infection symptom, wilting mechanism, vector, and the biology and physiology of B. xylophilus (Gao et al., 2015;Kim et al., 2010;Mamiya, 1988;Shi, Chen, Luo, Wang, & Xie, 2013;Yoshimura et al., 1999), relatively little is known about the underlying impact of photosynthetic characteristic and resource-use efficiency on PWD-induced tree mortality. The primary objective of this study was to further determine the effects of the invasion of B. xylophilus on the photosynthetic characteristic and resource-use efficiency of P. massoniana trees. In this study, our hypothesis was that P. massoniana trees infected with PWD would undergo a decrease in photosynthesis and resource-use efficiency decreases during the infection by PWD compared with uninfected trees.
The photosynthetic response of P. massoniana with various PWD infection phases were measured to determine the changes in CO 2 assimilation, transpiration behavior, stomata condition, stable carbon isotopic composition, and resource-use efficiency.

| METHODS
This study was taking place in the Yiling District of China, an eastern part of the Three Gorges reservoir region. With an eastern midsubtropical monsoon climate, the Yiling District has a mean annual precipitation and mean annual temperature of 997-1,370 mm and 16.6°C, respectively. P. massoniana is the primary coniferous tree species in this district. Since its first occurrence in 2006, B. xylophilus has spread rapidly in Yiling District.
The research was conducted in a 3-year PWD-infected Masson pine forest plots, which the stand density was 1,304 tree/ha and the basal area is 14.96 m 2 /ha. All P. massoniana trees in our plots were surveyed for B. xylophilus infection using Zhao's PWD rating system (Zhao et al., 2008), which is based on the external symptoms and internal changes of the B. xylophilus-infected pine trees; it includes the following phases: control, initial infection phase, early infection phase, middle infection phase, serious infection phase, and terminal infection phase. For this project, we measured the control (healthy Masson pine), middle infection (discoloration of old needles, decrease of oleoresinosis, partial necrosis of cells, low conductivity of sap ascent, and high population of pine wood nematode), and terminal infection phases (discoloration of both old and young needles, none of oleoresinosis, wide-area necrosis of cells, completely stop of sap ascent, and extensive propagation of pine wood nematode) of PWD-induced P. massoniana trees. A total of nine large trees (three for each treatment, Table 1) were selected for measurements of photosynthetic gas exchange characteristics and carbon isotope ratios.

| Leaf gas exchange
The measurements were conducted during the sunny days in August in 2014. Two branches were selected for each sampled Masson pine tree, and three replicate photosynthetic measurements were conducted per branch. The P n , T, g s , and C i were measured with a portable photosynthesis system equipped with a red/blue LED source and CO 2 injector (LI-6400 XT, Li-Cor Inc, USA). A clamp-on leaf cuvette that exposed 6 cm 2 of leaf area was equipped to the LI-6400. When used, the instrument was zeroed and the chemicals replaced (Meinzer et al., 2004). All measurements were conducted with the cuvette temperature set at 25°C and relative humidity at 60%.

| Photosynthetic light-response curves
Photosynthetic light-response curves were calculated by gradually decreasing PPFD in 15 levels from 2,000 to 0 μmol m −2 s −1 . In the meantime, the cuvette CO 2 partial pressure was fixed at a constant level of 400 μmol/mol with a high-pressure liquefied CO 2 cartridge source.
There was a 2-min waiting period between each step, and measurements were taken when stability was achieved. Two light-response curves were determined for each P. massoniana trees in the selected damaged stages for a total of 18 curves. P max , dark respiration rate, light compensating points (LCP), and light saturation points (LSP) were calculated from these light-response curves using nonlinear regression techniques (Cavatte et al., 2012;Ögren & Evans, 1993). In addition, we calculated light-use efficiency (LUE), instantaneous water-use efficiency (WUE i ), instantaneous carboxylation efficiency (CE i ), and nitrogen-use efficiency (NUE i ), which are related to photosynthetic capacity and resource availability (Field et al., 1983;Hsu et al., 2015).
The formulas are as follows: where P n is the net photosynthetic capacity (μmol CO 2 m −2 s −1 ), P max is the greatest net photosynthetic value (μmol CO 2 m −2 s −1 ), PPFD is the photosynthetic photon flux density(μmol m −2 s −1 ), T is the transpiration (μmol H 2 O m −2 s −1 ), C i is the internal CO 2 concentrations, and N mass is foliar N content.

| Carbon and nitrogen isotope ratio
After gas exchange was measured, foliar tissue (one sample for each branch) on which gas exchange was measured was collected, dried, and ground to a fine powder (Meinzer et al., 2004), and analyzed for δ 13 C, C%, and N% with an isotope ratio mass spectrometer (IsoPrime100, IsoPrime Corporation, Manchester, UK). The formula is as follow: where δ sample is the value of δ 13 C of the measured foliar tissue samples, R sample is the 13 C/ 12 C value of the measured sample, and R standard is the 13 C/ 12 C value of the standard sample. The standard sample for 13 C is Pee Dee Belemnite (PDB).

| Statistical analyses
Photosynthetic data were collected and analyzed statistically using one-way ANOVA. The Fisher's least significant difference (LSD) test with an alpha value of p < .05 level was used to compare the means of the photosynthetic characteristics, isotope abundance, and resourceuse efficiency of P. massoniana trees. A least squares linear regression model (again at p < .05) was used to assess the relationship between the foliar N content and δ 13 C, as well as P max , for each measured sample. All of the statistical analyses were performed using SPSS 22.0 for Windows (SPSS Inc., Chicago, IL, USA) and GraphPad Prism 6.0 (GraphPad Software, La Jolla, CA, USA).

| Effect on photosynthetic characteristics
From Figure 2a, we can clearly see that the P n decreased in the infested plants at PPFD levels from 0 to 2,000 μmol CO 2 m −2 s −1 . In the F I G U R E 2 Changes in net photosynthetic rates (a), transpiration rate (b), stomatal conductance (c), and internal CO 2 concentrations (d) among different phases of pine wilt disease-infected Masson pine across the PPFD levels infected trees, the P n in the middle infection phase was higher than those in the terminal infection phase, for which the values were less than 0 across all PPFD levels. Similarly, the values of the T, g s , and C i also decreased significantly (p < .05) in the infested P. massoniana plants as PPFD increased from 0 to 2,000 μmol m −2 s −1 when compared with the controls. (Figure 2b-d).
The values of P max (F = 190.07, p < .01) and LSP (F = 26.906, p < .01) were significant decrease in the leaves of infested P. massoniana plants than those in controls (Table 1). However, the value of LCP was significantly higher in the infested treatment (F = 8.141, p < .05,

| Effect on carbon and nitrogen isotopic composition
Pinus massoniana trees show a significantly higher C:N in infested treatments (F = 14.114, p < .01, Table 1). The difference was due to changes in foliar N mass concentration in infested Masson pines, for which the mean foliar N content was about 12.94% less in the middle infection phase and 27.06% less in the terminal infection phase when compared with those in control P. massoniana trees (Table 2).
Conversely, there is little variation about the mean foliar C content ( Table 2).
The foliar δ 13 C values were decreased about 2.5‰ in the infested trees when compared with the control P. massoniana trees. The foliar δ 13 C did not differ between the middle and terminal infection phases.
The P max estimated from the light-response curves at 400 μmol/mol CO 2 -saturated irradiance were 18.71, 7.73, and −0.66 μmol m −2 s −1 for control, middle infection, and terminal infection phases, respectively.

| Effect on resource-use efficiency
Compared with controls, the LUE decreased significantly in the infested P. massoniana trees (p < .05), and the value for the terminal infection phase is less than 0 (Figure 4a). The value of WUE for the infested P. massoniana trees decreased as PPFD increased from 0 to 2,000 μmol m −2 s −1 when compared with the controls, but there was no significant difference between the control and middle infection phases (Figure 4b). Similarly, there was no significant decrease in CE for the middle infection phase trees. However, the CE for the terminal infection phase, which was steady and less than 0 at every PPFD, was significantly decreased (p < .05) in comparison with the controls (Figure 4c). The foliar N content was significantly lower (p < .05) in photosynthetic leaves of infested P. massoniana trees than those in control trees (Table 1). Meanwhile, the NUE markedly decreased in the infested P. massoniana trees (Figure 4d).

| DISCUSSION
In PWD researches, few studies have focused on the relationship between PWD and photosynthetic response and resource-use efficiency. The present work provides new insights into the significant physiological changes that B. xylophilus infestation causes as measured by leaf gas exchanges, foliar carbon isotope ratios, and resourceuse efficiency in P. massoniana trees.
In this study, the values of P n , T, g s , and C i across PPFD levels in P. massoniana leaves were significantly lower in the trees infected with B. xylophilus than in controls, and the corresponding value in the terminal infection phase decreased even more markedly (Figure 2).
In addition, the decreasing P n and T in the infected P. massoniana treatments were associated with a reduction in g s . Therefore, we can presume that the decreases in P n and T are caused by the closure of stomatal pores in infested P. massoniana leaves. When a healthy tree is infected, B. xylophilus can cause dysfunction in the xylem and tracheid, blocking the sap ascent and water transportation from the roots to the crown (Zhao et al., 2008). With the dehydration of the mesophyll cells, the stomatal pores will partially or completely close, inhibiting the exchange of water and CO 2 between external and internal foliar cells (Bigot, Fontaine, Clement, & Vaillant-Gaveau, 2007;Hsu et al., 2015), and leading to a decrease in T and C i . Meanwhile, with a water deficit in the infected foliar cells, photosynthetic enzymatic activity can be inactivated or experience a loss of Rubisco and other key Calvin cycle enzymes (Bigot et al., 2007), which in turn reduces the total net photosynthetic rates. These results were consistent with previous studies, which reported that P n , transpiration, and g s are known to covary in a consistent manner in many species (Bigot et al., 2007;Farquhar & Caemmerer, 1982;Hsu et al., 2015;Meinzer et al., 2004). In addition, from Figure 2, we can see the closure of stomatal pores has a greater effect on CO 2 (C i ) than H 2 0 (transpiration), which is mainly due to the host foliar cells showing additional resistance associated with diffusion of CO 2 than H 2 0 (Zhang, Feng, Cregg, & Schumann, 1997).
Because PWD infection has a significant influence on photosynthetic enzymatic activities, significant differences in P max, LCP, and LSP across light levels exist among three PWD-induced treatments (Table 1). At the same time, R d in the middle infection phase is higher than those in the control and terminal infection phases, indicating that most of the photosynthetic products in the middle infection phase were consumed by plant respiration. This may be due to the P. massoniana plants needing more energy to support their defense system in these infection phases to prevent pine sawyer beetle feeding and PWD infection (Zhao et al., 2008). This assumption is supported by many previous findings, which reported changes in respiration and oxidative enzyme levels in response to environmental stress (Bigot et al., 2007;Lopes & Berger, 2001;Zangerl, Arntz, & Berenbaum, 1997). Zangerl et al. (1997) noted that respiration rates increased 19% in pierced-damaged leaves of wild parsnip 2 h after treatment. Lopes and Berger (2001) also found that the dark respiration rate increased on diseased leaves, which were damaged by rust and anthracnose. Bigot et al. (2007) found glycolate oxidase and cleavage T protein levels were raised by herbicide stress on grapevine.
Light-use efficiency and carboxylation efficiency are two important parameters for estimating plant productivity (Akmal & Janssens, 2004). In our research, LUE (P n /PPFD) and CE (P n /C i ) were significantly lower in the PWD-induced P. massoniana plants, which contributed to the significant decline in P n . These results can be explained by the fact that P n is an instantaneous measure of Rubisco enzyme activity (Makoi, Chimphango, & Dakora, 2010), which is affected by the invasion of PWD. Therefore, we can conclude that the reduction of LUE and CE in the PWD-induced trees is caused by the disruption of Rubisco performance. Also, this assumption is confirmed by previous studies (Nabity, Heng-moss, & Higley, 2006;Walia et al., 2010).
In our experiment, the foliar C:N ratio was significantly higher in PWD-infected P. massoniana trees (Table 1), which is a result of a significant decrease in foliar N mass concentration in the infested Masson pines. The significant loss of foliar N content may be due to a lack of N uptake by the trees, caused by the closure of stomata. Similar results were found in dwarf mistletoe-induced western hemlock, in which the leaf nitrogen content was 35% lower in infected trees (Meinzer et al., 2004). In 2007, Cabrera-Bosquet also reported that plant nitrogen content was very susceptible to loss due to environmental stimulation (Cabrera-Bosquet, Molero, Bort, Nogués, & Araus, 2007). In addition, the significant loss of foliar N content can influence the activity of the Rubisco enzyme that causes the reduction of photosynthetic capacity in infested P. massoniana trees (Clearwater & Meinzer, 2001;Walia et al., 2010;Warren, Dreyer, & Adams, 2003). In our data, P max values were linearly correlated with the foliar N content of each measurement ( Figure 3), which is consistent with existing literature. Similarly, NUE (P max /N mass ) was also significantly lower in infected P. massoniana trees ( Figure 4d).  (Dawson et al., 2002;Flower et al., 2013). In our research, the foliar δ 13 C level decreased as the PWD damage increased (Figure 3), which can be explained by a reduction of g s and a decrease in the value of intercellular to ambient CO 2 concentration (Cabrera-Bosquet et al., 2007;Farquhar et al., 1989;Walia et al., 2010). This result was consistent with published studies, which reported a marked decrease in foliar δ 13 C values for conifers during a dwarf mistletoe infection (Meinzer et al., 2004;Sala, Carey, & Callaway, 2001).
Water is one of the most critical environmental factors that can affect the growth and performance of plants (Cavatte et al., 2012).
WUE is defined as the amount of carbon gained per unit water loss (Sinclair, Tanner, & Bennett, 1984;Walia et al., 2010), which can be demonstrated directly by dividing P n by T and indirectly by determining the δ 13 C value of foliar tissues (Zhang et al., 1997). In this study, the value of WUE for the infested P. massoniana trees decreased as PPFD increased from 0 to 2,000 μmol m −2 s −1 (Figure 4b); a significant decrease in foliar δ 13 C also occurred (Table 1), especially for the trees in the terminal infection stage. This result may be due to the fact that P n did not change proportionally to PWD-induced changes in transpiration, which have a more relative decrease of P n and (or) a more relative increase of T in infested trees. These results were supported by previous studies, which reported the reduction of WUE in other infected hosts as indicated by a decline in foliar δ 13 C (Meinzer et al., 2004;Pataki, Oren, & Phillips, 1998).
In conclusion, our results demonstrate that P. massoniana trees undergo distinct photosynthetic changes and decrease in resource-use efficiency with infection by PWD compared with uninfected trees. Our study suggests that the significant decreases in photosynthetic capacity and resource-use efficiency of PWD-infected P. massoniana trees are caused by the closure of stomatal pores and possibly also by the inactivation or loss of Rubisco and other key Calvin cycle enzymes. The results of this study will help forest managers to identify PWD infestations at the photosynthetic and physiological levels so as to better facilitate management actions.