Mycorrhizas alter sucrose and proline metabolism in trifoliate orange exposed to drought stress

Arbuscular mycorrhizal fungi (AMF) can enhance drought tolerance in plants, whereas little is known regarding AMF contribution to sucrose and proline metabolisms under drought stress (DS). In this study, Funneliformis mosseae and Paraglomus occultum were inoculated into trifoliate orange (Poncirus trifoliata) under well watered and DS. Although the 71-days DS notably (P < 0.05) inhibited mycorrhizal colonization, AMF seedlings showed significantly (P < 0.05) higher plant growth performance and leaf relative water content, regardless of soil water status. AMF inoculation significantly (P < 0.05) increased leaf sucrose, glucose and fructose concentration under DS, accompanied with a significant increase of leaf sucrose phosphate synthase, neutral invertase, and net activity of sucrose-metabolized enzymes and a decrease in leaf acid invertase and sucrose synthase activity. AMF inoculation produced no change in leaf ornithine-δ-aminotransferase activity, but significantly (P < 0.05) increased leaf proline dehydrogenase activity and significantly (P < 0.05) decreased leaf both Δ1-pyrroline-5-carboxylate reductase and Δ1-pyrroline-5-carboxylate synthetase activity, resulting in lower proline accumulation in AMF plants under DS. Our results therefore suggest that AMF strongly altered leaf sucrose and proline metabolism through regulating sucrose- and proline-metabolized enzyme activities, which is important for osmotic adjustment of the host plant.

mycorrhizal inoculation did not affect AI activities, but increased SS activities 40 days after seed germination 15 . Such changes in sugar accumulation and sucrose-metabolized enzyme activities under mycorrhization highlight AMF's ability in OA to enhance drought tolerance of host plants.
In addition to sugars, proline, an important osmolyte in plants, plays an important role in the process of OA under DS 16 . Indeed, proline accumulation is likely to a key indicator, which provides plants with an osmotic mechanism to maintain a favourable osmotic potential for water uptake, therefore, alleviating the injury of DS 16 . In plants, proline is synthesized by the glutamate and/or ornithine pathways. The glutamate pathway is the main pathway in proline synthesis, in which glutamate first converts into glutamate-semialdehyde (GSA) by the Δ 1 -pyrroline-5-carboxylate synthetase (P5CS), then spontaneously converts to Δ 1 -pyrroline-5-carboxylate (P5C), and then transforms into proline by Δ 1 -pyrroline-5-carboxylate reductase (P5CR) [16][17][18] . In the ornithine synthetic pathway, proline can be synthesized from ornithine in mitochondria, where it is first transaminated by ornithine-δ -aminotransferase (OAT), producing GSA and P5C, and then transformed into proline by P5CR 16,19 . Proline catabolism occurs in mitochondria, which outlines that proline is first catabolized by proline dehydrogenase (ProDH) producing P5C, and then converted to glutamate by P5C dehydrogenase (P5CDH) 16 . AMF inoculation often induces proline accumulation 17,19,20 or proline decrease 5,21,22 in drought-stressed plants. Wu et al. 5 concluded that AMF seedlings accumulated less proline than non-AMF seedlings in leaves of citrus exposed to DS, due to either greater drought resistance or less injury in AMF seedlings under DS. Similar findings were reported for other species, such as Antirhinum majus 3 , Erythrina variegate 6 , Cyclobalanopsis glauca 21 and Ocimum gratissimum 22 . In contrast, higher proline level in AMF plants exposed to DS was also found in Oryza sativa 17 and Macadamia tetraphylla 19 . However, the effect of AMF on proline metabolism is still not fully understood.
Citrus, one of the most important fruit crops grown in many regions of the world, is highly sensitive to soil drought stress (DS). Moreover, citrus plants are strongly dependent on the AM symbiosis 4 . Trifoliate orange (Poncirus trifoliata L. Raf.), a close relative to Citrus, is widely used as the main rootstock of citrus plantations in Asia, including China, India, and Japan. Studies confirm that AMF inoculation enhanced drought tolerance in citrus plants 5,[23][24][25] . Nevertheless, little is known regarding the contribution of AMF to sucrose-and proline-metabolized enzyme activities in trifoliate orange exposed to DS, despite that the changes of sugars and proline concentration induced by AMF have been reported 5 . In this background, the objectives of the present study are to elucidate the effects of AMF on organic solute contents in trifoliate orange subjected to well watered (WW) and DS conditions and to analyze the changes in sucrose-and proline-metabolized enzyme activities.

Results
Root AMF colonization. Mycorrhizal colonization was not observed in roots of non-AMF seedlings regardless of WW and DS. Root mycorrhizal colonization in inoculated plants varied from 34.2% to 57.5% with F. mosseae and 49.7% to 70.5% with P. occultum, respectively (Table 1). DS treatment significantly (P < 0.05) decreased root colonization by F. mosseae and P. occultum. In addition, mycorrhizal colonization by P. occultum was markedly (P < 0.05) higher than by F. mosseae, regardless of soil water status. There was a significant interaction between soil water treatments and AMF treatments (P < 0.01) ( Table 1).
Plant growth performance. Compared with non-AMF treatment, F. mosseae and P. occultum treatments significantly (P < 0.05) increased plant height, stem diameter, leaf number, and leaf, stem, and root dry weight under both WW and DS, whilst inoculation with P. occultum showed better effects on stimulating plant growth than with F. mosseae (Table 1).
Leaf relative water content (RWC). AMF seedlings by F. mosseae and P. occultum showed significantly (P < 0.05) higher RWC than non-AMF seedlings under WW and DS conditions, respectively (Fig. 1). No significant interaction between water treatments and AMF treatments was observed ( Table 2). Leaf carbohydrate concentrations. The seedlings colonized by P. occultum and F. mosseae recorded significantly (P < 0.05) higher leaf fructose, glucose, and sucrose concentrations than non-AMF colonized seedlings, regardless of WW or DS (Fig. 2a-c; Table 2). Significantly higher fructose, glucose, and sucrose levels were found in leaves of AMF seedlings inoculated with P. occultum compared with those inoculated with F. mosseae, irrespective of soil water status (Fig. 2a-c).

Activities of sucrose-metabolized enzymes in leaves.
Under WW conditions, treatment with F. mosseae significantly (P < 0.05) increased leaf AI and SPS activity by 23.5% and 69.2%, and inoculation with P. occultum increased leaf SPS activity by 92.5%, compared with non-AMF treatment ( Fig. 3a-d). AMF seedlings colonized by F. mosseae and P. occultum also showed 25.6% and 20.7% significantly (P < 0.05) lower leaf SS activity than non-AMF seedlings under WW condition. Under DS condition, mycorrhization with F. mosseae and P. occultum markedly (P < 0.05) increased leaf NI activity by 69.2% and 45.8% and leaf SPS by 198.1% and 282.0% respectively. They also decreased leaf AI by 57.0% and 20.2% and leaf SS activity by 83.5% and 88.5%, respectively ( Fig. 3a-d). AMF seedlings colonized by F. mosseae and P. occultum showed notably (P < 0.05) higher net activity of sucrose-metabolized enzymes regardless of soil water status (Fig. 3e). There was the significant interaction in leaf AI, NI, SPS, and net activity of sucrose-metabolized enzymes between soil water condition and AMF treatment (P < 0.01) ( Table 2).  Proline concentration. AMF seedlings colonized by F. mosseae and P. occultum showed significantly (P < 0.05) lower leaf proline level under WW and DS conditions (Fig. 4). There was no significant difference in leaf proline concentration between F. mosseae and P. occultum (Fig. 4) and a significant (P < 0.05) interaction between soil water treatments and AMF treatments ( Table 2).

Activities of proline-metabolized enzymes in leaves.
In gereral, F. mosseae-and P. occultum-treated seedlings showed significantly (P < 0.05) lower leaf P5CS and P5CR activity than non-AMF-treated seedlings under WW and DS activity (Fig. 5a,b; Table 2). Only P. occultum treatment significantly (P < 0.05) increased leaf OAT activity compared with non-AMF treatment under WW condition ( Fig. 5c; Table 2). Treatment with F. mosseae and P. occultum significantly (P < 0.05) increased leaf ProDH activity compared to non-AMF treatment under both WW and DS ( Fig. 5d; Table 2). A significant (P < 0.05) interaction between soil water treatments and AMF treatments was observed for ProDH activity (Table 2).
Correlation studies. Under WW condition, leaf SS activity was significantly (P < 0.05) negatively correlated with leaf glucose concentration, and activity of leaf SPS and net activity of sucrose-metabolized enzymes were significantly (P < 0.01) and positively correlated with leaf fructose, glucose, and sucrose concentrations ( Table 3). Under DS condition, leaf AI activity was significantly (P < 0.05) negatively correlated with leaf glucose  conditions. Data (mean ± SD, n = 5) followed by different letters above the bars among treatments indicate significant differences at 5% level.
concentration, and leaf SS activity was significantly (P < 0.01) and negatively correlated with leaf fructose, glucose, and sucrose concentration. Activity of leaf NI, SPS and net activity of sucrose-metabolized enzymes were significantly (P < 0.05 or P < 0.01) and positively correlated with leaf fructose, glucose, and sucrose concentrations (Table 3). Under WW conditions, leaf proline concentration was significantly (P < 0.01) positively correlated with leaf P5CR and P5CS activity (Table 3). Under DS condition, leaf proline concentration was significantly (P < 0.05) positively correlated with leaf P5CR activity and negatively with leaf ProDH activity (Table 3).

Discussion
Mycorrhizal roles in plant growth under drought stress. This study showed a significant (P < 0.05) decrease in root AMF colonization by DS. Similar results had been reported in other plant species, such as sweet potato 26 , Cyclobalanopsis glauca 21 and Cucumis melo 27 . The negative effects of drought stress on root AMF colonization might be due to the germination of spores and the spread of hyphae in soils being inhibited by DS 21,27 .
Although the decrease of root colonization under DS was represented, such AMF colonization still significantly promoted plant growth parameters of trifoliate orange grown under DS conditions. This result is consistent with previous works using Erythrina variegate 6 and Poincianella pyramidalis 28 . Enhancement of growth and biomass and higher leaf RWC in AM plants could be due to improved water and nutrient uptake assisted by mycorrhizal hyphae 3,6,28 . In addition, P. occultum had markedly (P < 0.05) greater plant growth performance than F. mosseae under DS, which is closely related with considerably higher root mycorrhizal colonization under Data (mean ± SD, n = 5) followed by different letters above the bars among treatments indicate significant differences at 5% level.

Figure 5. Effects of Funneliformis mosseae and Paraglomus occultum on activity of proline-metabolized enzymes in leaves of trifoliate orange seedlings under well watered (WW) and drought stress (DS)
conditions. Data (mean ± SD, n = 5) followed by different letters above the bars among treatments indicate significant differences at 5% level.
Scientific RepoRts | 7:42389 | DOI: 10.1038/srep42389 P. occultum than F. mosseae. It also suggests that growth improvement of trifoliate orange by AMF is strongly dependent on AMF species.

Mycorrhizal roles in sucrose metabolism under drought stress.
Carbohydrates are considered as the important compatible solutes for OA in plants under DS 7 . In the present study, AMF colonization by F. mosseae and P. occultum significantly increased leaf sucrose, fructose, and glucose concentrations under DS conditions, which would act as osmolytes to protect and stabilize plant macromolecules and structures from drought damage, thereby enhancing the drought tolerance of the host plant by OA 9 . A relatively higher RWC was observed in AM trifoliate orange seedlings than in non-AM seedlings under DS, further suggesting greater water status in AM plants subjected to DS. These results are in agreement with earlier studies 3 .
In this study, leaf SPS activity was significantly higher in AMF seedlings than in non-AMF seedlings exposed to WW and DS. This is in agreement with the findings of Zhu et al. 29 , who reported a higher SPS activity in G. tortuosum-colonized Zea mays under low temperature (15 °C for 2 weeks). Moreover, our results further indicated that leaf SPS activity was significantly (P < 0.01) and positively correlated with leaf fructose, glucose, and sucrose concentrations under DS conditions. Higher leaf carbohydrate accumulation in AMF seedlings was caused by an AMF-induced increase in leaf SPS activity. This shows that AM symbiosis can modulate SPS activity to induce sugar accumulation.
In general, sucrose needs to be cleaved by sucrose-cleaving enzymes (AI, NI, and SS) into glucose and fructose, while glucose can be absorbed directly by mycorrhizal formations 30 . The present study showed that leaf AI activity under WW was increased by F. mosseae but was reduced by F. mosseae and P. occultum under DS. Similarly, root colonization by F. mosseae and P. occultum did not alter leaf NI activity under WW, but significantly increased leaf NI activity under DS. These results indicated that soil water strongly affects the behavior of AMF on leaf AI and NI activity. Significantly lower leaf SS activity was found in AMF seedlings than in non-AMF seedlings under DS, irrespective of F. mosseae or P. occultum. Similar results of the AMF-induced SS changes were also found in Citrus tangerina colonized by F. mosseae 4 and trifoliate orange colonized by F. mosseae grown in 3 mM P level of the growth substrate 31 .
Sugar accumulation is relative to dynamic balance between the activity of sucrose synthetic enzymes (SPS) and sucrose-cleaving enzymes (AI, NI and SS), and thus net activity of sucrose-metabolized enzymes plays a dominant role in sugar accumulation 13 . In this study, AMF inoculation with F. mosseae and P. occultum strongly stimulated an increase in net activity of sucrose-metabolized enzymes, regardless of WW or DS. Net activity of sucrose-metabolized enzymes was significantly and positively correlated with sucrose, glucose, and fructose concentration in leaves. This result suggested that AMF-induced sugar accumulation might be associated with an increase in the net activity of sucrose-metabolizing enzymes. It seems that AM symbiosis might modulate net activity of sucrose-metabolized enzymes to induce sugar accumulation, which is beneficial to OA. AMF-induced sucrose-metabolized enzyme changes will need to be studied by analyzing their gene regulation through qRT-PCR.

Mycorrhizal roles in proline metabolism under drought stress.
Data from the present study showed a lower accumulation of proline in leaves from AMF trifoliate orange plants than non-AMF plants under DS. This result is in agreement with previous findings in soybean 32 , Erythrina variegate plants 6 and Ocimum gratissimum 24 under DS. As stated by Augé and Moore 33 , lower accumulation of proline caused by mycorrhization may be due to less strain by DS, because of greater water status in AMF plants. The present work revealed drought stress induced higher proline concentrations accompanied by an increase of P5CR and P5CS activity, a decrease of OAT activity and no difference of ProDH in leaves of trifoliate orange, suggesting that proline accumulation in AMF and non-AMF trifoliate orange was derived from the enhancement of the glutamate synthetic pathway of proline but not the ornithine synthetic pathway of proline. This is in accordance with previous studies of Zou et al. 34 . In this study, AMF seedlings showed significantly lower P5CR and P5CS activity but substantially higher ProDH activity than non-AMF seedlings, irrespective of WW or DS conditions. This means that a decrease in proline accumulation in AMF seedlings is potentially associated with an AMF-modulated decrease of glutamate synthetic pathways and an increase of proline catabolism, which will be still studied by checking the relevant gene expression.
Scientific RepoRts | 7:42389 | DOI: 10.1038/srep42389 mycorrhization, which is beneficial to OA in the host plant. Nevertheless, the underlying molecular mechanisms in AMF plants still need to be further examined by RNA-seq.

Methods
Experimental set-up. Approximately 1000 spores/pot of each AM fungus were applied at transplanting of the seedlings. Non-AMF treatment also received the equivalent quantity sterilized inocula and 2 mL inocula filtrate (25 μ m filter) to keep similar microbial communities except the AM fungus. All the seedlings were grown in a glass greenhouse with the characteristics of photosynthetic photon flux density of 728-965 μ mol/m 2 /s, day/night temperature 20-35/15-26 °C, and relative humidity 70-95%.
Water treatments. Water treatments began 87 days after seedling transplanting. Half of the seedlings were subjected to WW by gravimetrically maintaining 70% of the maximum water holding capacity of the substrate for 71 days. The other seedlings were exposed to DS via maintaining 50% of the maximum water holding capacity of the substrate for 71 days. Water status in each pot was maintained daily by weighing and any loss of water was resupplied to maintain the target soil relative water content. The location of pots was shifted weekly to avoid environmental differences.
Experimental design. The experiment was designed using a 3 × 2 randomized complete block design with three inoculations with AMF (F. mosseae, P. occultum, and non-AMF) and two soil water regimes (WW and DS). Each treatment was replicated five times, requiring a total of 30 pots for this study.
Variable determinations. After 71 days of water treatments, the seedlings were harvested and growth parameters such as plant height, stem diameter, and leaf number per plant were recorded. The seedlings were divided into shoots and roots and the dry weight of each was measured after baking at 75 °C for 48 h.
A portion of fresh roots were cut into 1-cm long root segments, cleared in 10% (w/v) KOH at 90 °C for 1.5 h, acidified in 20 mM HCl for 10 min, and finally stained with 0.05% trypan blue in lactophenol 35 . Root AMF colonization was expressed as the percentage of infected root lengths by AMF against observed total root lengths.
Leaf relative water content (RWC) was assessed by the method employed by Huang et al. 24 . Fructose, glucose and sucrose concentration in leaves was determined by the protocol outlined by Wu et al. 31 . A 0.2-g fresh leaf sample was homogenized in a chilled mortar with 4 mL 100 mM Hepes-NaOH buffers (pH 7.5), containing 20 mM EDTA, 1 mM NaF, 1 mM benzamidine, 20 mM cysteine and 1% polyvinyl pyrrolidone. The mixture was centrifuged at 10,000 × g for 30 min, and the supernatants were dialyzed with 100 mM Hepes-NaOH buffer (pH 7.5) in a 21-mm dialysis bag for 12 h at 4 °C. The activity of AI and NI was determined by the protocol described by Wu et al. 4 . To determine SPS activity, 40 μ L dialyzed supernatant was added into 100 μ L reaction mixtures including 5 mM fructose-6-phophate, 10 mM uridine diphosphate glucose, 15 mM MgCl 2 , 15 mM glucose-6-phophate, 1 mM EDTA, and 50 mM Hepes-KOH buffer (pH 7.5). The 140 μ L mixtures were incubated at 30 °C for 30 min, terminated with 0.2 mL 5 mM NaOH at 100 °C for 10 min, mixed with 3.5 mL anthrone solution (0.15 g anthrone + 100 mL 81% H 2 SO 4 ) at 40 °C for 20 min, and then measured according to Hubbard et al. 15 . The activity of SS (degradative direction) was determined according to Lowell et al. 36 . Mixtures containing 80 mM Mes buffers (pH = 5.5), 5 mM NaF, 100 mM sucrose and 5 mM UDP were incubated at 30 °C for 30 min, terminated with DNS at 100 °C for 5 min, and the absorbance of the mixtures was recorded at 540 nm. Net activity of sucrose-metabolized enzymes was calculated according of Hubbard et al. 13 with the following formula: net activity of sucrose-metabolized enzymes = SPS activity − (AI + NI + SS) activity.
Proline concentration in leaves was determined using the ninhydrin method of Troll and Lindsley 37 . The activity of leaf P5CS, OAT, and ProDH was assayed according to Zou et al. 34 . The activity of leaf P5CR was determined according to the method of Chilson et al. 38 with minor modifications. Briefly, 0.2 g of fresh leaf samples was homogenized with 5 mL 100 mM Tris-HCl buffer (pH 7.5) containing 10 mM MgCl 2 , 1 mM EDTA, 10 mM β -mercaptoethanol, 2 mM phenylmethanesulfonyl fluoride and 2% (w/v) polyvinylpolypyrrolidone, and then centrifuged at 20,000 × g for 20 min at 4 °C. The enzyme activity was assayed with a final volume of 1.0 mL reaction mixture containing supernatant, 200 mM glycine buffer (pH 10.3), 15 mM NAD + , and 20 mM proline. The absorbance at 340 nm was recorded and one unit of P5CR was defined as the enzyme amount of 1 μ mol NADH during 1 min (U/g FW).
Statistical analysis. Data (means ± SD, n = 5) was performed using the two-way analysis of variance (ANOVA) with SAS software (8.1 v, SAS Institute Inc., Cary, NC, USA), and the significant differences between the treatments were compared with the Duncan's multiple range test at P < 0.05. Pearson correlation coefficients between variables were tested by the CORR procedure based on SAS software.