Plant phenols contents and their changes with nitrogen availability in peatlands of northeastern China

Abstract

Aims

Climate warming and increasing nitrogen (N) deposition have influenced plant nutrient status and thus plant carbon (C) fixation and vegetation composition in boreal peatlands. Phenols, which are secondary metabolites in plants for defense and adaptation, also play important roles in regulating peatland C dynamics due to their anti-decomposition properties. However, how the phenolic levels of different functional types of plants vary depending on nutrient availability remain unclear in boreal peatlands.

Methods

Here, we investigated total phenols contents (TPC) and total tannins contents in leaves of 11 plant species in 18 peatlands of the Great Hing’an Mountains area in northeastern China, and examined their variations with leaf N and phosphorus (P) and underlying mechanisms.

Important Findings

Shrubs had higher TPC than graminoids, indicating less C allocation to defense and less uptake of organic N in faster-growing and nonmycorrhizal graminoids than in slower-growing and mycorrhizal shrubs. For shrubs, leaf TPC decreased with increasing N contents but was not influenced by changing leaf phosphorus (P) contents, which suggested that shrubs would reduce the C investment for defense with increasing N availability. Differently, leaf TPC of graminoids increased with leaf N contents and decreased with leaf P contents. As graminoids are more N-limited and less P-limited, we inferred that graminoids would increase the defensive C investment under increased nutrient availability. We concluded that shrubs would invest more C in growth than in defense with increasing N availability, but it was just opposite for graminoids, which might be an important mechanism to explain the resource competition and encroachment of shrubs in boreal peatlands in the context of climate warming and ever-increasing N deposition.

摘要

气候变暖和大气氮沉降增加会改变北方泥炭地的养分状态，从而影响其植被的物种组成和固碳功能。酚类物质是植物用于 防御植食性动物和适应环境的次生代谢产物，由于其具有抗分解的特性，在调节泥炭地碳动态方面也起着重要的作用。然而，北方 泥炭地不同功能型植物的酚含量及其如何随氮有效性变化尚不清楚。本论文通过测定中国东北大兴安岭地区18个泥炭地共11 种植物的叶片总酚含量(Total Phenols Contents, TPC)，研究了它们随叶片氮、磷含量的变化关系，并探讨了其潜在机制。结果表明，灌木叶片TPC高于草本植物，说明生长较快且无菌根的草本植物比生长较慢且具有菌根的灌木对防御的碳投入较少。灌木叶片TPC随叶片氮含量增 加而降低，表明其防御碳投入随氮有效性增加而减少。相反，草本植物叶片TPC随氮含量增加而增加，随磷含量增加而减少，由于草本植物相对于灌木具有较强的氮限制和较弱的磷限制，草本植物防御碳投入可能随养分有效性增加而增加。我们的结果表明，泥炭地在氮有效性随气候变暖和氮沉降增加而增加的背景下，灌木将投入相对较多的碳用于生长而非防御，而草本植物则与之相反。这些发现将有助于对北方泥炭地灌木入侵及其资源竞争机制的理解。

INTRODUCTION

Peatlands are one of the most important terrestrial ecosystems, largely distributed in the boreal regions (Limpens et al. 2008). Due to low temperature and strong nitrogen (N) limitation conditions in these regions, boreal peatlands have commonly slower decomposition rate than inputs of organic matter from vegetation productivity (Yu 2012), which allows them to accumulate 15%–30% of the global soil carbon (C) pool, despite covering less than 3% of global land area (Limpens et al. 2008). In recent years, climate warming and increasing N deposition are threatening the stability of the C pool in boreal peatlands, even accelerating the decomposition of peat and leading to unpredictable greenhouse gas emissions, thus resulting in positive feedback on climate warming (Breeuwer et al. 2008; Dorrepaal et al. 2009). However, considering the increase in vegetation productivity due to warming and increased N availability, there is still great uncertainty about C dynamics in boreal peatlands (Dise 2009; Moore et al. 1998; Turunen et al. 2004).

Phenolic compounds (phenols) play an important role in the stability of C in peats because of their inhibitory effects on microbial activity and decomposition of peat organic matter (Freeman et al. 2001). Phenols are products of plant secondary metabolism, mainly coming from plant litter input and leaching (Aerts et al. 1999; van Breemen 1995). Plant phenols can help plants resist harsh environments, including defense against herbivores (Coley et al. 1985), adaptation to low temperature (Albert et al. 2009) and resource limiting conditions (Northup et al. 1995). Plant phenols are also important allelochemicals that can inhibit the growth of other plants nearby and the activity of microbes (Watson 2018). Accordingly, quantifying plant phenols is important for understanding the defensive strategies of plants, litter decomposability and even peat C dynamics (Freeman et al. 2001). Although there have several studies on the secondary metabolism and litter chemistry of plants in boreal peatlands, relatively few studies focused on plant phenols (Dorrepaal et al. 2005; Turetsky et al. 2008; Wiedermann et al. 2007).

Plant phenols contents vary greatly and are mainly regulated by the trade-off between growth and defense, which reflects the interactions of biotic and abiotic factors during plants adapting to growing environments (Bryant et al. 1983; Coley et al. 1985; Herms and Mattson 1992). Many theories have been proposed to explain the interspecies differences in phenols contents, and plant functional types (PFTs, such as graminoids and shrubs) are widely used to predict the differences in plant communities (Dorrepaal 2007; Dorrepaal et al. 2005; Mao et al. 2018). For example, the ‘apparency theory’ speculated that long-lived species have higher phenols contents because they are apparent for both generalist and specialist herbivores, while short-lived species can evade specialists and only need to defense generalists, because they are unapparent and ephemeral in time and space, thus having lower phenols contents (Feeny 1976). In addition, the ‘growth rate hypothesis’ assumed that fast-growing species produce fewer phenols, because the negative impacts of losing leaf area for them were low as they could replace lost leaves more quickly, thus investing fewer resources in defense (Endara and Coley 2011). Although these theories are often used to explain the interspecies differences in plant phenols contents, few have been proven to be general across species and ecosystems.

Nutrient availability is also an important factor in regulating phenols production, and it is generally believed that the higher nutrient availability will promote plant growth and reduce plant defensive investment, thus leading to a lower phenols contents (i.e. C–nutrient balance, Bryant et al. 1983; Jones and Hartley 1999). So far, decreased leaf total phenols contents (TPC) with increased nutrient contents was observed for a variety of PFTs and species (Kraus et al. 2004; Price et al. 1989), although insignificant or positive correlations occurred in a few cases (Zhang et al. 2012). Moreover, these findings were mostly based on the test of plant phenols contents under nutrient-addition experiments or along soil nutrient gradients. Given that soil nutrient availability is not equal to plant nutrient availability due to differing nutrient uptake abilities and competition among different PFTs and species (Liu et al. 2018), the effects of plant nutrient status on phenols production should be further explored. Furthermore, plant N utilization strategies are also related to their phenols contents. Mycorrhizal plants that can utilize soil organic N usually have higher phenols contents (Hättenschwiler and Vitousek 2000), Northup et al. (1995) speculated that this may be the adaptation of these plants to N-limited conditions. Plant phenols (especially tannins) could help plants effectively ‘monopolize’ N contained in the litter by immobilizing N into the phenol–protein complex that could only be used by mycorrhizal plants, which made these plants have a competitive N acquisition advantage (Kraus et al. 2003; Northup et al. 1995). However, current research on the relationship between plant N utilization strategies and phenols contents is still weak.

Based on above background and questions, this study measured TPC and total tannins contents (TTC) in leaves of different plants in peatlands of the Great Hing’an Mountains (GHM) area in northeastern (NE) China, and examined their relationships with leaf N, leaf phosphorus (P), contributions of different soil N sources to leaf N and environmental factors. The objectives of this work were (i) to explicitly examine differences and variabilities of phenols contents among PFTs and species and (ii) to explore effects of N, P availability and plant N utilization strategies on leaf TPC among the peatland plants. We predicted that (i) leaf TPC are lower in graminoids than in shrubs, (ii) leaf TPC decrease with the increases of N, P contents for all species and PFTs and (iii) species with higher contributions of soil organic N to leaf N have higher leaf phenols contents.

MATERIALS AND METHODS

Study area

The GHM (50°10′–53°33′ N, 121°12′–127°00′ E) are located in NE China and on the southern margin of the Eurasian permafrost zone (Fig. 1). The elevation of the GHM is 180–1528 m, averaging 573 m. The climate of the GHM area is characterized by a cold temperate monsoon type. The mean annual temperature (MAT) was about −4.9°C, with the lowest temperature of −52.3°C. The mean annual precipitation (MAP) was about 403 mm, with approximately 80% of precipitation fall from June to September. During Pleistocene, the GHM area experienced three interglacial epochs and no peatlands had initiated until the Holocene Megathermal. Currently, the GHM runs from northeast to southwest, with high-latitude permafrost zones, high and cold erosion lands, low-middle mountains and hills as major land forms (Jin et al. 2007). Ombrotrophic peatlands are largely distributed in intermontane basins of the GHM area, and are similar in vegetation and water conditions (Song et al. 2009; Wang et al. 2010). Dwarf shrubs, graminoids and bryophytes (mainly Sphagnum) are typical vegetation of peatlands in the GHM area. The common and dominant graminoids include Deyeuxia angustifolia, Eriophorum vaginatum and Carex globularis. The dominant shrubs are Ledum palustre, Chamaedaphne calyculata, Rhododendron capitatum, Vaccinium vitis-idaea, Salix rosmarinifolia, Salix myrtilloides, Betula fruticosa and Vaccinium uliginosum in peatlands of the GHM area.

Sample collection

Leaf sampling was conducted from 22 July to 5 August 2014 (during the growing season of plants), and dominant plant species of peatlands were collected at 18 sites of the GHM area (Fig. 1). Selected study sites were far from human habitations and sampling sites of plants had open and abundant sunshine. Plant species collected at each site were based on their presence listed in Table 1, for a total of 11 species, which included three graminoids, four evergreen shrubs and four deciduous shrubs. At each site, three plots of about 10 m × 10 m (with a distance of about 10 m from each other) were set up, common and dominant plant species were sampled. For each species, healthy, mature and sunlit leaves were collected from several plant individuals and were combined into one plot-based sample. Some plant species had only 1–2 replicates at some sites due to their presence in only 1–2 plots (Table 1). The total 267 plant samples included 79 for graminoids, 65 for evergreen shrubs and 123 for deciduous shrubs (Table 1). In addition, soil samples were collected in the same plots of plant leaf samples, three plot-based soil samples were collected for each peatland site. In this paper, the mean values of soil N and P contents of each peatland are given in Table 1 to show the nutrient status of these peatlands. The geographical and climatic information of study sites were also detailed in Table 1 and plant mycorrhizal types, leaf shapes and heights of collected plants are listed in Supplementary Table S1.

Chemical analyses

Fresh samples of plant leaves were rinsed with deionized water and ultrapure water (18.2 MΩ), and then dried at 75°C in an oven to constant weights. Dried leaf samples were ground to fine powders with a ball mill (MM200, Retsch, Haan, Germany). Leaf C, N and P contents (% dry weight (dw)) were measured and have been reported in Li et al. (2018) (Supplementary Fig. S1), and leaf δ ¹⁵N values were measured and used to calculate the contributions of soil nitrate (NO₃⁻), ammonium (NH₄⁺) and hydrolyzable amino acids (HAA) to leaf N. Please see supporting information for the specific calculation method of the contributions of these three N sources to leaf N.

Concentrations of water-extractable phenolic compounds were measured as free TPC to indicate phenolic levels in plants (Hobbie 1996; Sundqvist et al. 2012). To extract phenolic compounds in leaf samples, the mixture of 10 mg leaf powder and 10 ml ultrapure water were shaken for 6 h and were centrifuged for 15 min at 4500 r·min⁻¹. Then the supernatant was filtered by syringe filters (Φ = 0.45 µm), and 0.5 ml of filtered extracts were mixed with 0.5 ml of Folin–Ciocalteu reagent (Sigma-Aldrich, 1 N) and 2.5 ml of 10% sodium carbonate in a glass tube, and incubated for 1 h in the dark before the absorbance measurement at the wavelength of 760 nm on a UV spectrophotometer (UV-2700, Shimadzu, Kyoto, Japan) (Jassey et al. 2011). Ultrapure water and gallic acid standards were treated and measured in the same way as leaf samples for data calibrations, TPC values were expressed in the percentage content of dry weight (%, dw, GAE, gallic acid equivalent) (Fig. 2). Three replicates of individual leaf samples showed a mean analytical precision of ±0.3%. To measure leaf TTC, TPC in the supernatant after the polyvinylpyrrolidine treatment (for removing tannins) were determined. TTC of the sample was calculated as: TTC = TPC − non-tannins TPC (Supplementary Fig. S2).

Statistical analyses

Prior to the statistical analysis, all data were tested for normality. Comparisons of leaf TPC and leaf TPC:C (the ratio of leaf TPC and leaf C) among PFTs and species were conducted using one-way analysis of variance, multiple comparisons were performed with least significant difference post hoc test (Fig. 2). The effects of species, sites and their interactions on leaf TPC were analyzed by General Linear Models (GLMs, normal distribution) (Table 2). Coefficients of variations (CV values, %) of leaf TPC among sites or species were calculated as corresponding ratios of standard deviation (SD) to arithmetical mean values (Supplementary Fig. S3). In this study, we defined CV values of <10%, 10%–100% and >100% as weak, moderate and strong variations, respectively. Stepwise multiple regressions were used to analyze proportional contributions of environmental variables (MAT, MAP and elevation), soil variables (soil N and P) and leaf variables (leaf C, N and P) to leaf TPC variations (Table 3). Correlations among leaf TPC, leaf TPC:C and C, N or P variables were examined by linear regressions (Supplementary Table S2 and Fig. 3), we also examined the relationships between leaf TTC and the contributions of soil NO₃⁻, NH₄⁺ and HAA to leaf N (Supplementary Fig. S4). Statistical analyses were performed by using the SPSS software (SPSS for Windows, Version 20.0, Chicago, IL, USA). Statistical significance was set as P < 0.05 in this study.

RESULTS

Leaf TPC of peatland plants ranged between 0.8% and 16.4% and averaged 7.4% ± 4.1% in the GHM area (Fig. 2a). Leaf TPC differed significantly among three PFTs, showing an order of graminoids (2.7% ± 1.1%) < evergreen shrubs (8.0% ± 3.1%) < deciduous shrubs (9.7% ± 3.4%) (Fig. 2a). When compared between species in the same functional type, D. angustifolia, E. vaginatum and C. globularis had similar leaf TPC levels among graminoids. Among evergreen shrubs, leaf TPC differed significantly and showed an order of L. palustre < C. calyculata < R. capitatum < V. vitis-idaea. For deciduous shrubs, S. rosmarinifolia and S. myrtilloides had lower leaf TPC levels, and other species showed no significant differences (Fig. 2a). In addition, leaf TPC:C ranged from 0.02 to 0.42 (averaged 0.18 ± 0.10) and the pattern was the same as leaf TPC (Fig. 2b), leaf TTC also showed a similar pattern as leaf TPC among different PFTs (Supplementary Fig. S2).

For graminoids, site differences explained 61% of leaf TPC variations, while site (12.4%), species (42.7%) differences and their interaction (13.1%) together contributed to leaf TPC variations for evergreen shrubs, and site (25.2%), species (23.1%) differences and their interaction (27.9%) for deciduous shrubs (Table 2). Besides, leaf TPC showed mean intersite CV values of 35%, 33%, 19% and interspecies CV values of 13%, 27% and 28% for graminoids, deciduous shrubs and evergreen shrubs, respectively (Supplementary Fig. S3). In addition, MAT, MAP, leaf N and leaf P contributed 17.4%, 6.8%, 41.2% and 34.6% of the TPC variations for graminoids, respectively (Table 3). MAP, elevation and leaf N contributed 4.9%, 20.2% and 74.9% of the TPC variations for evergreen shrubs, respectively (Table 3). Leaf N contributed total TPC variations for deciduous shrubs (Table 3).

Correlations between leaf TPC, leaf TPC:C and C, N or P variables differed among PFTs and species (Supplementary Table S2 and Fig. 3). For graminoids, leaf TPC increased with N, decreased with P, thus increased with N:P values (Supplementary Table S2 and Fig. 3a and b). Differently, leaf TPC decreased with increasing N and showed no clear variations with P (Fig. 3a and b) for evergreen and deciduous shrubs, and leaf TPC decreased with N:P values for evergreen shrubs (Supplementary Table S2). The relationships between leaf TPC:C and N, P were the same as leaf TPC (Fig. 3c and d). Furthermore, at the community level, leaf TTC decreased with the increased contribution of NH₄⁺ to leaf N, but increased with the increased contribution of HAA, and had no relationship with the contribution of NO₃⁻ (Supplementary Fig. S4).

DISCUSSION

In this study, we measured the leaf TPC of 11 plant species in boreal peatlands and examined the effects of PFTs and environmental factors, especially nutrient availability. Interestingly, we found substantial differences in the leaf TPC patterns and their correlations with nutrient availability among PFTs. But due to the lacking of measuring structural defense compounds, our findings of leaf TPC allowed us to discuss chemical defense more exactly, instead of structural defense or overall plant defense (Dudt and Shure 1994; Sundqvist et al. 2012; Watson 2018).

Leaf phenolic levels among PFTs and species

We found that leaf TPC of peatland plants in the GHM area (7.4% ± 4.1%) were similar to those of subarctic peatland plants (6.3% ± 4.4%) in Europe (Dorrepaal et al. 2005), and the pattern of lower leaf TPC in graminoids than in shrubs (Fig. 2a) was consistent with our first prediction, which reflected that different leaf TPC may be attributed to the different PFTs and their different responses under changing environmental conditions (De Long et al. 2016; Dorrepaal et al. 2005; Mao et al. 2018). On the one hand, the differences in longevity determined their different defense strategies. Long-lived shrubs would be bound to be found by generalist and specialist herbivores and would be under higher herbivory pressure, thus investing more resources in defense for effective defense (Endara and Coley 2011; Feeny 1976). In contrast, short-lived graminoids were unapparent and it would be difficult for herbivores to specialize on them, so graminoids could evade specialists and invest fewer resources in defense (Endara and Coley 2011; Feeny 1976).

On the other hand, the growth rate is also an important factor in regulating defense investment (Coley et al. 1985). Fast-growing graminoids would relatively invest fewer resources in defense and allocate more photosynthetic C to sustain the growth than slow-growing shrubs (Coley et al. 1985) (Fig. 4). Lower leaf TPC:C in graminoids than in shrubs (Fig. 2b) also demonstrated the different C allocation strategy (Fig. 4). In other words, graminoids would recover more quickly via faster growth after being damaged by herbivores, thus leading to higher cost–benefit ratios of defense investment than shrubs (Endara and Coley 2011). McKane et al. (2002) found graminoids such as E. vaginatum had similar aboveground biomass with every shrub, but accounted for the most of the aboveground net primary production in Arctic tundra, reflecting higher growth rate but lower longevities of graminoids. Besides, lower leaf C, N and P but higher leaf C:N and C:P values in graminoids than shrubs (Supplementary Fig. S1, Li et al. 2018) also supported physiologically higher use efficiency of N and P for faster growth of graminoids in the GHM area (Li et al. 2018).

In addition, we also found that evergreen shrubs had slightly lower leaf TPC than deciduous shrubs, although the evergreen had longer life-spans (Díaz and Cabido 1997), this result might reflect the differences in allocation to chemical or structural defense between evergreen and deciduous shrubs, which could be supported by higher lignin contents in leaves of evergreen shrubs (Dorrepaal et al. 2005). Besides, this may be attributed to the fact that the several species we collected could not represent all the plants of the same functional type in our study area, and the few repetition numbers of these species also made our results somewhat accidental. However, lower leaf TPC:C in evergreen shrubs (Fig. 2b) did reflect to a certain extant lower C allocation to chemical defense than deciduous shrubs. Higher N and P resorptions supporting photosynthesis and growth (Aerts et al. 1999) potentially favored higher leaf C but lower TPC levels in evergreen shrubs than deciduous shrubs.

At the species level, our results reflected that three species of graminoids invested similar resources in defense, causing similar leaf TPC and leaf TPC:C (Fig. 2). Among evergreen shrubs, we found that plant species with higher leaf N (L. palustre > C. calyculata > R. capitatum > V. vitis-idaea, Supplementary Fig. S1) invested more C in growth and therefore had lower leaf TPC (Fig. 2) (Bryant et al. 1983). In our study, the lower leaf TPC of Salix could be attributed to the higher leaf N and P than other species (Fig. 2 and Supplementary Fig. S1). According to the carbon–nutrient balance, i.e. more C investment in growth and less in defense with higher N availabilities (Bryant et al. 1983), and the lanceolate leaf shape also reflected that S. rosmarinifolia have stronger physical defense ability, thus leading to lower chemical defensive level (Supplementary Table S1 and Fig. 2) (Ibanez et al. 2013; Sanson et al. 2001). Our results of GLMs showed the proportion of leaf TPC variations explained by site and species varies greatly across PFTs (Table 2). For graminoids, a higher proportion of leaf TPC variations were explained by site rather than species, which indicated that their defense strategies have stronger environmental adaptability and smaller interspecies difference than shrubs (Chapin et al. 1996), and highest intersite CV values (Supplementary Fig. S3) of graminoids also supported this result. However, it was also possible that the low/high variance explained by the species was an artifact of the low number of species sampled per PFT.

Effects of nutrient availability on leaf phenolic variations

In our study, no clear and consistent relationships were observed across three PFTs between leaf TPC and corresponding MAT, MAP, elevation, soil N and P among the study sites (Table 3). Based on available biotic parameters, significant correlations were found between leaf TPC and leaf N, P contents and their correlations differed among PFTs (Table 3 and Fig. 3), which indicated nutrient availability plays an important role in regulating phenolic levels. Evergreen and deciduous shrubs showed the same pattern of the relationship between leaf TPC and N, P contents, while graminoids were just the opposite, this result reflected the difference between woody and herbaceous plants, and the similarity within the woody plants.

For woody shrubs, negative correlations between leaf TPC or TPC:C and N contents were resistant with the second prediction (Fig. 3a–c), which clearly reflected the mechanism of C–nutrient balance, i.e. more C investments in growth and less in defense with higher N availability (Bryant et al. 1983). Physiologically, our results also reflected that the production of phenols and proteins competed for the common precursor (Fig. 4). Increased plant N uptake and availability potentially increased protein production, leading to a reduction of phenylalanine into phenols production (Jones and Hartley 1999). Accordingly, our finding also explained why many shrubs could grow and expand rapidly under increased N availability and form encroachments into boreal regions (Vowles and Björk 2019). Differing from the effect of leaf N contents, no clear relationships were found between leaf TPC and P contents (Fig. 3b–d). This could simply be attributed to the fact that P was not directly involved in the biosynthesis of phenols, although it could influence plant growth and metabolism (Jones and Hartley 1999). However, Wright et al. (2010) contrasted leaf TPC of plants growing in P-rich and P-depleted soil in southern New Zealand and showed that higher P availability did correspond to lower phenols, although N availability is a more important determinant of plant investment in defensive compounds than P availability. These results coindicated that the production of plant phenols have lower plasticity and sensitivity to P than to N for shrubs (De Long et al. 2016; Wright et al. 2010).

Inconsistent with the second prediction, leaf TPC and TPC:C increased with increasing leaf N contents for graminoids, suggesting higher C allocation to defense with higher N availability (Figs 3 and 4). This result may be because graminoids had more defense demand to effectively defend against herbivory with the increased N availability (Keinänen et al. 1999; Shan et al. 2018), or higher N availability increased the meristematic activities of graminoids, which would facilitate the syntheses of complex molecules including phenols (Bryant et al. 1983; Herms and Mattson 1992). In northern Sweden, Sundqvist et al. (2012) also observed that leaf TPC increased for most plant species with increased leaf nutrient concentrations along an elevational gradient, which suggested that there can be considerable differences in defense strategies among coexisting species in response to changes of nutrient availability (Hamilton et al. 2001). Besides, concentrations of some flavonoids in the stem of Scots pine seedlings increased with increasing N in Ghimire et al. (2018), while the results of another research were just the opposite (Keski-Saari and Julkunen-Tiitto 2003). These results indicated that C allocation into individual phenols differs among species (Sundqvist et al. 2012) and highlighted the importance of considering the properties and composition of individual phenols (Hättenschwiler et al. 2011). In addition, the effects of P were opposite to N, showing decreased leaf TPC and TPC:C with increased leaf P contents (Fig. 3b–d), which could be explained by C–nutrient balance, these results could reflect that both N and P are involved in the synthesis of phenols, but affect different metabolic pathways for graminoids (Keski-Saari and Julkunen-Tiitto 2003).

Consistent with our third prediction, leaf TPC were positively related to the contribution of HAA, but negatively related to the contribution of NH₄⁺ at the whole community level, which suggested the higher NH₄⁺ assimilations but lower tannins levels of nonmycorrhizal graminoids (Supplementary Fig. S4), as opposed to mycorrhizal shrubs assimilated more organic N and produced more tannins (Supplementary Fig. S4). Our results indicated that plants with higher contributions of organic N tend to have higher phenolic levels (Supplementary Fig. S4), which reflected the plant adaption mechanisms to N limitation that shrubs monopolized N contained in the litter by producing more phenols–organic N complexes and utilizing litter N sources through mycorrhiza (Hodge et al. 2001), thereby having a competitive advantage against graminoids (Northup et al. 1995).

Ecological implications and remarks

In our study, different leaf TPC levels among peatland plants revealed that PFTs are a key indicator of phenols production (Fig. 4). Higher leaf TPC of shrubs demonstrated that shrubs are more resistant to pests and pathogen infections, and have stronger growth inhibition on surrounding plants than graminoids. As the pattern of leaf TPC levels can influence that of litter TPC levels via leaf fall, higher TPC in shrubs indicate the lower decomposability of shrub litter than graminoids due to the anti-decomposition properties of phenolic compounds (Freeman et al. 2001). However, with the intensified global change, climate warming, elevated N deposition and subsequently increased N availability will deeply affect the plant growth and physiological characteristics, community structure and functions in boreal peatlands. According to our results, shrubs will invest more resources in growth than in defense with increasing N availability, while graminoids will oppositely prefer defense to growth (Fig. 4), thus making shrubs evolve higher competition advantages, which may help to understand the shrub encroachment in boreal and subarctic regions (Myers-Smith et al. 2015). In addition, the differing responses of leaf TPC to increasing N availability between shrubs and graminoids will also affect their litter decomposability in different ways, which has an impact on the key processes of C cycling in peatlands. Accordingly, our results have important implications for understanding future changes in plant community compositions and C dynamics in boreal peatlands. Moreover, as the structural defense also plays an important role in plant defense, litter decomposability and nutrient cycling in boreal peatlands, future research will focus on the structural and chemical defense of plants more comprehensively for a better explanation on the role of TPC in these aspects.

Supplementary Material

Supplementary material is available at Journal of Plant Ecology online.

Table S1: Plant mycorrhizal types, leaf shapes and plant heights in the Great Hing’an Mountains peatlands (Li et al. 2018).

Table S2: Correlation coefficients between leaf TPC and C, N, P variables of peatland plants in the Great Hing’an Mountains area, northeastern China.

Figure S1: Leaf C, N and P contents (a–c), C:N, C:P and N:P values (d–f) of different peatland plants in the Great Hing’an Mountains area, northeastern China.

Figure S2: Leaf TTC of different peatland plants in the Great Hing’an Mountains area, northeastern China.

Figure S3: Coefficients of (a) intersite and (b) interspecies variations (CV values, %) of leaf TPC of peatland plants in the Great Hing’an Mountains area, northeastern China.

Figure S4: Variations of leaf TTC with the contributions (%) of NO₃⁻ (a), NH₄⁺ (b) and HAA (c) to leaf N of different peatland plants in the Great Hing’an Mountains area, northeastern China.

Funding

This work was supported by the National Key Research and Development Program of China (2016YFA0600802), the National Natural Science Foundation of China (41730855, 41522301) and the Open Project Foundation in Key Laboratory of Geographical Processes and Ecological Security of Changbai Mountains, Ministry of Education (GPES201904). Xue-Yan Liu was also supported by the 11th Recruitment Program of Global Experts (the Thousand Talents Plan) for Young Professionals granted by the central budget of China.

Acknowledgements

We thank Drs Zhong-Cong Sun, Tian-Yi Ma and Hao Huang for their help on the experiments, and Wei Song for helpful comments on this manuscript.

Conflict of interest statement. None declared.

REFERENCES

Author notes