JIA-2019-11

2615 CHEN Xu et al. Journal of Integrative Agriculture 2019, 18(11): 2605–2618 that the substrates were insufficient to increase the PLFA abundance, so the fungi/bacteria ratios were similar (Qin et al. 2010). We also analyzed all 27 PLFAs at days 3 and 60 of incubation due to the significant differences between GL- RAD and GL-IAD from day 35 to 60. Our data demonstrate that the changes in fungi at day 3 were driven by changes in 18:1ω9c associated with CK and the various additions after aggregate-size reduction. The differences in bacterial PLFAs were largely driven by changes in i16:0 and a17:0 in RAD. The number of PLFAs driving the changes increased from 2 to 6. In contrast, the data for glucose and alanine additions indicated a trend opposite to that for CK. As discussed above, these findings imply that adding easily available substrates (glucose and alanine) to soil provides resources that microorganisms use rapidly within a few days (Blagodatskaya and Kuzyakov 2008). Therefore, these inputs probably represented the available energy that caused the shift from a dormant to an active state, and then back to dormancy following resource depletion. This shift may be followed by the activation of specific microorganisms that decompose substrates with low availability (e.g., the organic materials released after macroaggregate reduction) (Landi et al. 2006) or by the activation of slow-growing K-strategists that degrade organic matter (Fontaine et al. 2003). 4.3. Responses of aggregate-linked enzymatic activities to substrate addition and aggregate-size distribution Extracellular enzymes are unevenly distributed in soil and may be sensitive to environmental changes depending on their location (Nannipieri et al. 2002), which could strongly regulate the turnover of organic matter in different aggregate sizes (Schnecker et al. 2014). Our study found that macroaggregate reduction increased the activity of microbial enzymes and that enzymatic activities were significantly correlated with the mass of macroaggregates and the SOC content in GL. Our results also showed that the majority of the variation in enzymatic activities was likely related to PLFA abundance. Our study nevertheless identified a pattern of decreasing activity with increasing nutrient availability (Sinsabaugh et al. 1993). For example, BG activity typically decreases with glucose addition, and the activity of the N-hydrolyzing enzyme chitinase decreases with inorganic-N addition (Olander and Vitousek 2005). The addition of C, though, can induce a positive relationship between C and C-hydrolyzing enzymes (Fontaine et al. 2003), likely due to the production of microbial biomass or to a transient effect that does not occur under equilibrium conditions due to the accumulation of intermediates (Allison et al. 2007). The general pattern of soil enzymatic activities is often dominated by the amount and quality of substrates and by various physical and chemical mechanisms of protection (Allison and Jastrow 2006). Enzymatic activities and PLFA abundance in our study were not significantly higher in the FL- and BF-RAD soils with or without substrate addition. The activity of enzymes involved in the degradation of labile C (BG) was similar in BF and FL, regardless of aggregate treatment. The activity of NAG, which is also involved in the degradation of recalcitrant C (e.g., chitin), however, was higher in FL than in BF. These differences may be ascribed to the different characteristics of the enzymes and or the availability of substrates. Substrate limited microorganisms switch to recalcitrant C when the easily available C has been consumed (Dungait et al. 2012), so the FL soil is expected to have a predominance of recalcitrant C, similar to the BF soil, which has not received any C inputs for decades. The higher NAG activity may also be due to the production of enzymes by microbes that degrade the chitin-rich cell walls of dead fungi (Guggenberger et al. 1999), which were more abundant in FL than in BF. It seems that controls on enzyme activity can be assessed on a microbial biomass basis (Hassett and Zak 2005) if microbial biomass changes are occurring simultaneously with shifts in enzyme activity. 5. Conclusion The reduction in size of macroaggregates (WSA >2 mm ) significantly influenced the microbial community in GL and increased enzymatic activities. This finding was supported by the strong relationships between the PLFA profiles and SOC content, enzymatic activity and SOC content. The IAD and RAD treatments did not significantly affect the microbial communities or the enzymatic activities in the BF and FL soils. The addition of easily available C tended to stimulate microbial groups in FL and BF but not in GL. The addition of available N had little to no impact on the microbial communities and enzymatic activities, regardless of aggregate treatment or history of soil management. In summary, the reduction of aggregates into smaller size classes had clear effects on the microbial communities and enzymatic activities in soil with larger aggregates and higher C contents. In contrast, soils with low SOC contents and smaller aggregates responded more strongly to easily available C substrates such as root exudates or fresh litter input than to the SOC contained within the different aggregate-size fractions. Acknowledgements This study was funded by the National Key Research and Development Program of China (2016YFD0300806-1,