Cumulative cellulolytic enzyme activities and initial litter quality in prediction of cellulose degradation in an alpine meadow of the eastern Tibetan Plateau

Abstract

Aims

Plant litter decomposition is a key ecosystem process that determines carbon and nutrient cycling in terrestrial ecosystems. As a main component of litter, cellulose is a vital energy source for the microbes associated with litter decomposition. The important role of cellulolytic enzymes in litter cellulose degradation is well understood, but seasonal patterns of cellulose degradation and whether cumulative enzyme activities and litter quality forecast cellulose degradation in an alpine meadow remain elusive, which limits our understanding of cellulose degradation in herbaceous plant litter.

Methods

A two-year field litterbag experiment involving three dominant species (Ajuga ovalifolia, Festuca wallichanica, and Pedicularis roylei) was conducted in an alpine meadow of the eastern Tibetan Plateau to explore the seasonal patterns of cellulose degradation and how cumulative cellulolytic enzyme activities and initial litter quality impact cellulose degradation.

Important findings

Our study demonstrates that cellulose degraded rapidly and exceeded 50% during the first year, which mainly occurred in the first growing season (31.9%–43.3%). At two years of decomposition, cellulose degradation was driven by cumulative endoglucanase (R² = 0.70), cumulative cellobiohydrolase (R² = 0.59) and cumulative 1,4-β-glucosidase (R² = 0.57). In addition, the concentrations of cellulose, dissolved organic carbon, total phenol, lignin and lignin/N accounted for 52%–78% of the variation in cellulose degradation during the two years of decomposition. The best model for predicting cellulose degradation was the initial cellulose concentration (R² = 0.78). The enzymatic efficiencies and the allocation of cellulolytic enzyme activities were different among species. The cellulolytic enzyme efficiencies were higher in the litter of F. wallichanica with relatively lower quality. For the complete cellulose degradation of the leaf litter, A. ovalifolia and F. wallichanica required 4-fold and 6.7-fold more endoglucanase activity, 3-fold and 4.5-fold more cellobiohydrolase activity and 1.2-fold and 1.4-fold more 1,4-β-glucosidase activity, respectively, than those required by P. roylei. Our results demonstrated that although microbial activity and litter quality both have significant impacts on cellulose degradation in an alpine meadow, using cellulose concentration to predict cellulose degradation is a good way to simplify the model of cellulose degradation and C cycling during litter decomposition.

摘要：

植物凋落物分解是决定陆地生态系统碳和养分循环的关键生态系统过程。作为凋落物的主要组成部分，纤维素是与凋落物分解相关的微生物的重要能量来源。纤维素酶在凋落物纤维素降解过程中的重要作用已为人们所熟知，然而纤维素降解的季节模式、累积酶活性和凋落物质量是否能预测高寒草甸的纤维素降解仍是一个未解之谜，这限制了我们对草本植物凋落物纤维素降解的认识。 为了探究纤维素降解的季节性模式以及累积纤维素分解酶活性和凋落叶初始质量对纤维素降解的影响，我们在青藏高原东部的高山草甸选取了三种优势种[圆叶筋骨草(Ajuga ovalifolia)、藏羊茅(Festuca wallichanica)和草甸马先蒿(Pedicularis roylei)]，进行了为期两年的凋落物网袋分解实验。 我们的研究发现，纤维素在第一年中迅速降解且降解率超过50%，而且主要发生在第一个生长季(31.9%–43.3%)在两年的分解过程中，纤维素降解由累积内切葡聚糖酶(R²                                  =                  0.70)，累积纤维二糖水解酶(R²                           =                0.59)和累积1,4-β-葡糖苷酶(R²        =       0.57)共同驱动。此外，在这两年的分解过程中，纤维素、可溶性有机碳、总酚、木质素的浓度和木质素/N可以解释纤 维素降解变异的52%–78%)。用初始纤维素浓度模型预测纤维素降解效果最佳(R² = 0.78)。在凋落物的分解过程中，酶效率和微生物对纤维素分解酶的分配因物种而异。藏羊茅凋落物中纤维素酶效率较高，但质量相对较低。与草甸马先蒿相比，完全降解圆叶筋骨草和藏羊茅的纤维素需要4倍和6.7倍的内切葡聚糖酶、3倍和4.5倍的纤维二糖水解酶、1.2倍和1.4倍的1,4-β-葡糖苷酶。我们的研究结果表明，虽然微生物酶活性和凋落物初始质量都对高山草甸纤维素降解有显著影响，但使用纤维素浓度来预测纤维素降解是简化凋落物分解过程中纤维素降解和C循环模型的好方法。

INTRODUCTION

Plant litter decomposition is a key ecosystem process that determines carbon and nutrient cycling in terrestrial ecosystems (Berg and McClaugherty 2014). Previous studies have used climatic factors (moisture and temperature parameters), initial litter quality (e.g. C:N, lignin:N) and extracellular enzyme activity to predict litter decay rates in many terrestrial ecosystems (Makkonen et al. 2012; Meentemeyer 1978; Melillo et al. 1982; Swift et al. 1979; Talbot and Treseder 2012; Zhang et al. 2008). Cellulose is the main structural component of the plant cell wall and the most abundant polysaccharide on earth; thus, cellulose is a vital energy source for the microbes associated with litter decomposition (Baldrian and Valášková 2008; Schwarz 2001; Sinsabaugh et al. 1981). However, few studies have addressed the predictions of cellulose degradation processes during litter decomposition in alpine meadows, which exhibit significant seasonal variability and have cold and relatively short growing seasons (Bliss 1971).

Cellulose exists in a number of crystalline and amorphous topologies, and due to its insolubility and heterogeneity, cellulose degradation is achieved mainly by three major types of enzymes: endoglucanases (EC 3.2.1.4), cellobiohydrolases (EC 3.2.1.91) and 1,4-β-glucosidases (EC 3.2.1.21), which hydrolyse cellulose polymers to smaller oligosaccharides, cellobiose and glucose (Criquet 2002; Schwarz 2001). In addition, cellulose degradation can be modulated significantly by litter species (Fioretto et al. 2005). For example, lignin can physically or chemically protect most of the cellulose and hemicellulose from microbial attack (enzymatic hydrolysis) during litter decomposition (Talbot and Treseder 2012). Previous studies have explored the relationships between litter mass loss and cellulolytic enzyme activities (Fioretto et al. 2000; Linkins et al. 1990; Sinsabaugh et al. 1994) rather than directly linking cellulolytic enzyme activities to cellulose degradation. However, as a main component of litter, whether the degradation of cellulose can be well predicted by litter quality (litter traits) or by cellulolytic enzyme remains unknown. Moreover, once enzymes have been secreted, enzyme activities have shown different efficiencies (Lashermes et al. 2016). The amount of enzyme activities required for cellulose decomposition can change significantly depending on litter type (Amin et al. 2014; Lashermes et al. 2016; Sinsabaugh et al. 2002). For example, hemicellulose content and its embedment in recalcitrant lignin linkages could primarily explain the efficiency of the C-acquiring enzymes (Lashermes et al. 2016). Enzyme efficiency may be negatively affected by high lignin contents (Sinsabaugh et al. 2002; Wickings et al. 2012). However, whether the efficiency of cellulolytic enzymatic decomposition differs among species remains to be elucidated.

Many ecosystems located at high latitudes and altitudes (e.g. alpine meadow) are subject to long periods of seasonal snow cover (Hobbie et al. 2000; Walker et al. 1999). During the winter, low temperature and freeze-thaw cycles could destroy a portion of microbial biomass and limit microbial activity, although studies have shown that microbial activity does not completely cease during the winter (Schimel and Clein 1996; Weintraub et al. 2007). During the growing season, the temperature increase due to solar radiation on the ground floor promotes the recovery of microbial activity, which enhances enzyme activity (A’Bear et al. 2014). Such seasonal variability can lead to large seasonal variation in cellulose degradation during litter decomposition in alpine forests and lotic ecosystems (He et al. 2015; Li et al. 2016; Yue et al. 2016a, 2016b). However, the environmental factors in alpine forest ecosystems and in lotic ecosystems could be different from those in alpine meadow ecosystems. For example, meadow ecosystems with high solar radiation exposure and low vegetation cover allow more direct contact with sunlight (Yanni et al. 2015). Moreover, the decomposability and chemistry of herbaceous species generally differ from those of woody species (Cornwell et al. 2008; Suseela et al. 2014). However, seasonal patterns of cellulose degradation in alpine meadow ecosystems have not been sufficiently documented, which limits our understanding of litter decomposition and C dynamics in different ecosystems.

To better understand the pattern and predictors of cellulose degradation, we hypothesized that (i) cellulose degradation varies with the season and that the degradation is higher in the growing season than in winter; (ii) cellulolytic enzyme activity and initial litter chemical traits could well predict cellulose degradation in alpine meadows; and (iii) cellulolytic enzyme efficiency during cellulose degradation differs among litters. To test these hypotheses, a 2-year field decomposition experiment using two forb species (Ajuga ovalifolia and Pedicularis roylei) and a graminoid species (Festuca wallichanica) with contrasting initial litter quality was conducted in an alpine meadow located on the eastern Tibetan Plateau in western China. Our work aims to provide insight into the patterns of cellulose degradation in alpine meadow ecosystems and to improve ecological forecasting of cellulose degradation and C cycling during litter decomposition.

MATERIALS AND METHODS

Site description

This experiment was conducted in an alpine meadow on Zhegu Mountain, which is an important riparian area located on the eastern edge of the Tibetan Plateau in Sichuan Province, China (31°51′428″N, 102°41′230″E). The alpine meadow is located at an altitude of 4200 m a.s.l., and the alpine desert is located at an altitude of 4500 m a.s.l. The alpine climate is cold during winter and cool during summer. The annual average temperature is 6–12°C, the average temperature in January is −8°C, and the annual temperature in July is 12.6°C. In addition, the annual precipitation and evaporation are 600–1100 mm and 1000–1900 mm, respectively, and the annual accumulated temperature is 1200–1400°C. The snow-covered season begins in November and lasts until the end of April (~6 or 7 months) in the alpine zone. The soils in the alpine meadow are classified as Histosols. The alpine meadow is dominated by A. ovalifolia (AO), P. roylei (PR), F. wallichanica (FW), Polygonum paleaceum and other species.

Litterbag experiment

A 50-m-wide transect (4200 to 4250 m a.s.l.) was placed along the contour line in an alpine meadow on Zhegu Mountain, from which three permanent sample plots were chosen. Freshly abscised leaves of AO, FW and PR were collected in October, which is when most of the litter fall occurs. The collected leaf litter was oven dried at 65°C for 48 h and mixed to obtain a homogeneous sample. Next, 10 g subsamples of the leaf litter were enclosed in nylon net bags (20 cm × 20 cm, 1.0 mm of surface-layer mesh and 0.5 mm of ground-layer mesh). For each species, 27 bags ((2 replicates × 4 harvest dates × 3 plots) for decomposition + 3 replicates for initial quality) were prepared, resulting in a total of 81 bags. Litter bags from each species were placed on top of the litter layer in the alpine meadows of the three fixed plots on 30 October 2012 and at least 5 cm apart. The litter decomposition temperatures in the litter bags were obtained every 3 h with thermometers (iButton DS1921-F5, Maxim/Dallas Semiconductor, Sunnyvale) and were automatically recorded. Six bags of each species in each plot were collected on 30 April 2013, 7 November 2013, 6 May 2014 and 4 November 2014 following the beginning of the decomposition experiment.

After each sampling, the soil and roots were carefully removed from the litters. Three litter bags were used to measure cellulose degradation. The other three bags were used to determine the moisture content and enzyme activity (a subsample of litter from each bag (30–40% fresh weight) was oven dried at 65°C to measure the moisture content, and the remaining litter in the three bags was stored at 4°C for enzyme extraction). The enzyme activities were measured within 1 week. The initial C, N, and P contents of the litter were determined as described by Lu (1999). The cellulose and lignin concentrations of the litter were determined using the acid detergent-lignin method (Vanderbilt et al. 2008). The extraction of dissolved organic carbon (DOC) was conducted according to Uselman et al. (2012), and the DOC concentration was analysed by using a Total Organic Carbon Analyzer (multi N/C 2100, Analytik Jena). The total phenol concentration was determined using the Folin–Ciocalteu method (Makkar et al. 2007).

Enzyme analyses

Enzymes were extracted according to the methods of Criquet (1999) with minor modifications. First, 4 to 9 g of freshly powdered litter (<0.5 mm) was extracted in 15 ml of a 0.1 M CaCl₂ solution with 0.05% Tween 80 and 0.40 g polyvinylpolypyrrolidone at 4°C overnight. The suspension was centrifuged at 12 000 ×g for 20 min at 4°C. The supernatant solution was subsequently dialyzed for 48 h at 4°C in 10-kDa molecular mass cut-off cellulose dialysis tubing against frequently exchanged 2 mM bis-tris (bis [2-hydroxyethyl] imino-tris [hydroxymethyl] methane) buffer (pH 6.0). The extracts that were boiled for 15 min served as controls for enzyme activity. Units of enzyme activities are expressed as µmol h⁻¹ per g of dry matter (DM).

The endoglucanase activity was determined by measuring the release of reducing sugars from appropriate substrates according to methods described in Criquet (2002). The reaction mixture, which contained 0.2 ml of the enzyme extract and 0.6 ml of 50-mM Na-acetate buffer (pH 6.0) containing 2% carboxymethylcellulose, was incubated for 1 h at 50°C. The amount of reducing sugars was determined using the dinitrosalicylic acid method (Miller 1959). The cellobiohydrolase and 1,4-β-glucosidase activities were measured according to the methods of Valášková et al. (2007). The reaction mixture contained 0.16 ml of 1.2 mM p-nitrophenyl-β-D-cellobioside and p-nitrophenyl-β-D-glucoside, respectively, in 50 mM sodium acetate buffer (pH 5.0) and 0.04 ml of enzyme extract. The reaction mixtures were incubated at 40°C for 40 min. The reaction was stopped by adding 0.1 ml of 0.5 M sodium carbonate, and the absorbance at 400 nm was determined. Enzyme activity was calculated using the molar extinction coefficient of p-nitrophenol (11 600 M⁻¹ cm⁻¹).

Calculations and statistical analyses

Cumulative cellulolytic enzyme activities were determined as described by Liu et al. (2009) and Waring (2013) and as follows:

where n is the duration (days) of the decomposition experiment, E_i is the mean enzyme activity (mmol g⁻¹ h⁻¹) between two successive measurements, and T_i is the time (h) between the two measurements. The cumulative enzyme activities are analogous to the degree days used to quantify the cumulative influence of temperature (Sinsabaugh et al. 1994, 2002).

The remaining cellulose content (MR) and cellulose degradation (L) of the litter were calculated as described by He et al. (2015) as follows:

where M_t is the mass of the remaining litter when sampled; C_t is the concentration of cellulose when sampled; MR_t and MR_{(t − 1)} represent the remaining cellulose contents between the current and previous sampling dates, respectively; and MR₀ is the initial cellulose content.

To characterize the temperatures during each period, we calculated the average temperature (AT), positive accumulated temperature (PAT) and negative accumulated temperature (NAT). In addition, the FFTC was calculated from the number of freeze-thaw cycles per period. The frequency of the freeze-thaw cycle was determined by considering that a cycle occurred when the temperature was <0°C for at least 3 h and then increased to >0°C for at least 3 h or when the temperature was >0°C for at least 3 h before decreasing to <0 °C for at least 3 h (Konestabo et al. 2007).

Repeated-measures ANOVA was used to evaluate whether species, decomposition period, and their interaction have significant effects on cellulose degradation and cellulolytic enzyme activities. We conducted one-way ANOVA (P < 0.05) and least significant difference tests (P < 0.05) for cellulose degradation and enzyme activities. Linear regression was performed to predict cellulose degradation based on the initial quality of litter and cumulative enzyme activities after 2 years of decomposition. To compare the allocation of cellulolytic enzyme activity and the efficiency of microbial decomposition among litter types, the relationships between cumulative enzyme activities and cellulose remaining in the litter (% initial) were fit using a first-order exponential decay model. The slope of these first-order exponential decay models is a rate constant (k: g mmol⁻¹) analogous to the first-order rate constant used to compare decomposition rates as a function of time (Sinsabaugh et al. 2002). To estimate and compare the complete cellulose degradation among litter types, turnover activities (the inverse of k) were calculated and expressed as mmol g⁻¹ (Sinsabaugh et al. 2002). When considering the first-order exponential decay model, larger slopes (k) and lower turnover activities corresponded with higher enzymatic efficiencies of cellulose degradation (Sinsabaugh et al. 2002). Moreover, Pearson correlation analysis was conducted to determine the relationship between cellulose degradation and enzyme activity and environmental factors. All statistical analyses were performed using SPSS Statistics for Windows (Version 25.0).

RESULTS

Environmental factors and initial litter quality

The average temperatures of the litter during decomposition and the positive accumulated temperatures during the winter periods were lower than the temperatures during the growing season, while the frequency of freeze-thaw cycles and negative accumulated temperatures during the winter periods were much higher than those during the growing season (Table 1). Overall, the moisture content was higher during the second year of litter decomposition (Table 1). The initial C concentrations in PR and FW were higher than those in AO (P < 0.05). The initial N and P concentrations were highest in FW, followed by PR and AO (P < 0.05). However, the C/N and C/P ratios of the three types of leaf litter decreased in the following order: AO>PR>FW (P < 0.05) (Table 2). The initial cellulose concentrations in AO and FW were higher than those in PR (P < 0.05). In contrast, the initial lignin concentration was greater in the PR litter than in the FW litter (P < 0.05). The DOC in PR was remarkably lower than that in the other species. The total phenol in FW was significantly lower than that in AO and PR (Table 2).

Cellulose degradation

From the results of repeated-measures ANOVA, cellulose concentration and degradation were significantly affected by species, decomposition period and their interaction (Table 3). Overall, the cellulose concentration decreased over time within species, and the cellulose concentration in AO and FW were always higher than those in PR (Fig. 1a). Cellulose degradation during the first growing season was the highest among the four decomposition periods, reaching nearly 31.9–43.3% (Fig. 1b). During the two studied years, 75.4–82.8% of the total cellulose was degraded, and during the first year, >50% of the total cellulose was degraded (Fig. 1c). After 2 years of decomposition, the amount of cellulose degradation in the three litter types decreased in the following order: FW>AO>PR (P < 0.05) (Fig. 1c).

Cellulolytic enzyme activity

The species, decomposition period and their interactions significantly impacted the cellulolytic enzymes except the impact of species on 1,4-β-glucosidase (Table 3). Endoglucanase activity was highest during the second winter, in contrast with cellobiohydrolase and 1,4-β-glucosidase activities (Fig. 2). Endoglucanase activity in FW was higher than that in AO and PR in four periods, with the exception of the second growing season. Cellobiohydrolase and 1,4-β-glucosidase activities in FW were higher than those in PR in four periods except the second growing season (Fig. 2).

Correlations

Initial litter qualities such as cellulose concentration (R² = 0.78), DOC (R² = 0.71), total phenol concentration (R² = 0.55), lignin concentration (R² = 0.54) and lignin/N (R² = 0.52) were good predictors of cellulose degradation (Fig. 3a). Additionally, cellulose degradation positively correlated with cumulative endoglucanase (R² = 0.70), cumulative cellobiohydrolase (R² = 0.59) and cumulative 1,4-β-glucosidase (R² = 0.57) activities (Fig. 3b). Based on the rule that larger slopes (k) and lower turnover activities corresponded with higher enzymatic efficiencies of cellulose degradation, cellulose degradation of PR was most efficient, followed by AO and then FW (Fig. 4). Specifically, cellulose degradation of AO and FW required 4- and 6.7-fold more endoglucanase activity than cellulose degradation of PR, respectively. Cellulose degradation of AO and FW required 3- and 4.5-fold more cellobiohydrolase activity than did cellulose degradation of PR, respectively. The requirements of 1,4-β-glucosidase activity of the three litter types were relatively close; that is, cellulose degradation of AO and FW required 1.2- and 1.4-fold more 1,4-β-glucosidase activity than cellulose degradation of PR, respectively (Fig. 4). Overall, cumulative cellulolytic enzymes were positively correlated with cellulose and DOC concentration, average temperature and moisture content. In contrast, cumulative cellulolytic enzymes were negatively correlated with lignin, lignin/N, total phenol concentration and frequency of the freeze-thaw cycle (Table 4).

DISCUSSION

In high-altitude and high-latitude ecosystems, seasonal snow cover and induced freeze-thaw cycles can strongly regulate litter decomposition (Aerts 2006; Kreyling et al. 2013). In line with our first hypothesis, cellulose degradation indeed varied with the season. Interestingly, we found that the proportion of cellulose degradation during the first growing season was the highest among the four decomposition periods, which is inconsistent with the result that the cellulose loss rate was the highest during the first winter in an alpine forest (He et al. 2015; Li et al. 2016). This could be explained by the different environmental factors in alpine meadow and alpine forest. For example, due to a temperature inversion phenomenon, the average daily temperature in alpine meadow was higher and snowmelt date was earlier than those in coniferous forest (Liu et al. 2016), which could promote higher cellulose degradation in the first growing season. In addition, after experiencing the winter season, litter decomposability could be increased, and the degradation of cellulose in the growing season could be further promoted (Wu et al. 2010) because recurrent freeze-thaw cycles can destroy the physical structure of leaf litter (Henry 2007). This suggests that cellulose degradation patterns are different even in varied alpine ecosystems, which should be considered in the establishment of litter decomposition models.

We found that a substantial amount of cellulose degradation (~20%) occurred during the winter despite the harsh environment, which was consistent with previous results showing that significant litter mass loss occurs during the winter (Brooks et al. 2005; Hobbie and Chapin 1996). After 2 years of litter decomposition, we found that >75% of the total cellulose was degraded within 2 years and that >50% of the cellulose was degraded within the first year, which implies that cellulose is a major source of carbon for decomposers, especially during early and intermediate stages of decomposition (Fioretto et al. 2007; Leitner et al. 2012).

Decomposition is a complex ecological process that is strongly affected by litter quality (Berg and McClaugherty 2014). The result supported our second hypothesis, that is, that cellulose degradation could be well predicted by litter chemical traits. For instance, cellulose degradation was predicted by concentrations of cellulose, DOC, total phenol, lignin and lignin/N, which can explain 52–78% of the variation after 2 years of litter decomposition. In accordance with the results of previously reported studies, cellulose degradation was affected by lignin and lignin/N in an alpine forest (Li et al. 2016) and alpine forest meta-ecosystem (Yue et al. 2016b). The underlying mechanism might be that the chemical and physical properties of the litter can alter the abundance and activity of decomposers and the effectiveness and stability of microbial enzymes, leading to different rates of decomposition due to plant species (Allison and Vitousek 2004; Melillo et al. 1982; Taylor et al. 1989). For instance, soluble and labile compounds (e.g. DOC) in decomposing litter can support rapidly growing microorganisms at the early decomposition stage (Swift et al. 1979). Furthermore, lignin can physically or chemically protect most of the cellulose and hemicellulose from microbial attack (enzymatic hydrolysis) during litter decomposition since cellulose is usually protected by lignin in plant litter (Talbot and Treseder 2012).

Considering the properties of cellulose, cellulose must be hydrolysed into smaller carbohydrates by extracellular enzymes (Criquet 2002). After 2 years of litter decomposition, the cellulose degradation rate was found to be significantly moderated by litter cumulative cellulolytic enzyme activities, which can explain 57–70% of the variation depending on enzyme types (Fig. 3b). In the present study, the cellulolytic enzyme activities were relatively low during the winter, especially during the first winter (Fig. 2). This could be because extracellular enzyme activity seems to be highly temperature limited in cold ecosystems (Wallenstein et al. 2009). For instance, greater decomposer activity may occur deeper in the soil profile where the temperature is greater than the temperature of the litter layer during the winter (Hobbie et al. 2000). Among litter quality and cellulolytic enzyme parameters, the concentration of cellulose is the most important factor to explain the variation in cellulose degradation after 2 years of decomposition. This suggests that cellulose concentration is a good litter trait predictor of cellulose degradation, which could simplify the litter decomposition model.

The result supported our third hypothesis, that the enzymatic efficiency and the allocation of the three cellulolytic enzyme activities differed among species (Fig. 4). Extracellular enzymes are the primary agents responsible for residue biodegradation in soil (Sinsabaugh et al. 2002). However, enzyme activity may not always reflect enzyme efficiency in natural environments such as soils (Amin et al. 2014). The differences in enzyme efficiencies, mean that the amount of enzyme activity required for substrate decomposition can vary markedly (Lashermes et al. 2016). Interestingly, in our study, the highest enzyme efficiency was found in litter PR, although enzyme activities of PR were lower than those of FW and AO (Figs 2 and 4). Several factors could influence enzyme efficiency in soils, such as accessibility to substrate, nonspecific interactions with minerals or substrates, and production of iso-enzymes by different microorganisms (Amin et al. 2014). In the present work, a higher enzymatic efficiency in litter corresponded to a lower litter quality (Fig. 4 and Table 2). This finding suggests that differences in the enzyme efficiencies might be explained by differences in litter quality, consistent with the result that enzyme efficiency was higher in the presence of a more recalcitrant litter than in the presence of a high-quality litter (Amin et al. 2014). This difference potentially occurred because of the different growth forms (Baptist et al. 2010), and the initial chemical properties of the plant control the decomposition process of the plant litter over short and long periods (Berg et al. 1996; Melillo et al. 1982). For example, the magnitude of the initial DOC concentration was opposite of the enzymatic efficiency (PR (2.28%)<AO (8.10%)<FW (9.47%) (Table 2)). This trend can be explained by the fact that DOC is the easiest carbon source for microbes to use (Clein and Schimel 1995). Additionally, the primary cell walls of grasses have a robust structure, making it difficult for microbes to access polysaccharides and proteins (Carpita 1996; Suseela et al. 2014).

CONCLUSIONS

Our study demonstrates that cellulose degradation mainly occurred during the early stages of litter decomposition and exceeded 50% during the first year in an alpine meadow in the eastern Tibetan Plateau. Moreover, cellulose degradation during the first growing season was the highest among the four decomposition periods, reaching nearly 31.9–43.3%. After 2 years of litter decomposition, cellulose degradation was well predicted by concentrations of cellulose, dissolved organic carbon, total phenol, lignin and lignin/N. Furthermore, cellulose degradation was driven by cumulative endoglucanase, cellobiohydrolase and 1,4-β-glucosidase activity. Among litter quality and cellulolytic enzyme parameters, the concentration of cellulose is the most important factor to explain the variation in cellulose degradation. The enzymatic efficiency and the allocation of the three cellulolytic enzymes were different among species, possibly due to the difference in initial litter quality. Our results demonstrated that although microbial activity and litter quality both have significant impacts on cellulose degradation in an alpine meadow, using cellulose concentration to predict cellulose degradation is a good way to simplify the model of cellulose degradation and C cycling during litter decomposition.

Funding

National Natural Science Foundation of China [31200345, 31570605 and 31370628]. China Scholarship Council (201706910039 to Y.C. (joint Ph.D. programme grant)).

Acknowledgements

We thank Dedong Zhou from the Forestry Bureau of Western Sichuan for his help with the field sampling work.

Conflict of interest statement. None declared.

REFERENCES