Bacterial community network complexity and role of stochasticity decrease during primary succession

In microbial ecology,there is limited understanding of the mechanisms governing patterns in community structure across space and time.Here,we studied the changes of bacterial co-occurrence network structure over four primary successional soils after glacier retreat,including a sand dune system and three glacier foreland series,varying in timescale from centuries to millennia.We found that in all series,network structure was most complex in the earliest stages of succession,and became simpler over time.Richness and abundance of keystone species and network stability also declined over time.It appears that with less productive,nutrient poor and physiologically extreme conditions of early succession,closer interactions among bacterial species are ecologically selected for.These may take the form of consortia(with positive interactions)or stronger niche exclusion(with negative interactions).Additionally,we quantified the relative roles of different structuring processes on bacterial community using a bin-based null model analysis(iCAMP).With each successional series,community composition was initially governed by stochasticity,but as succession proceeded there was a progressive increase in deterministic selection over time,correlated with decreasing pH.Overall,our results show a consistency among the four series in long-term processes of community succession,with more integrated networks and greater stochasticity in early stages.