Is understanding the structure-function of microbial community of sludge suitable using DNA-based molecular methods?

Analysis the similarities between total and active microbial community is crucial to evaluate whether understanding the structure-function of microbial community of different types sludge is suitable using DNA-based molecular methods. In this study, procaryotic communities in different types sludge samples (including primary sludge, excess sludge, mixed sludge, and digested sludge) were evaluated using DNA-based and RNA-based Illumina MiSeq. Results showed that the similarities between total and active procaryotic communities of all different types sludge were considerable high. Meanwhile, the similarity in running sludge higher than that in idle sludge, because of the Jarccard’s coefficient of digested sludge was the highest in all types sludge samples. This study indicates that the DNA-based molecular biology technology could be applicable to profile procaryotic community structure and function in sludge samples, and this study adds the understanding of total and active procaryotic communities of sludge samples.


Introduction
Activated sludge technology is currently the most broadly implemented biological method for biomass conversion in wastewater treatment plants (WWTPs) (Kim et al., 2003;Ye and Zhang, 2013;Korzeniewska and Harnisz, 2018). However, this process produced vast quantities of highly organic waste activated sludge, and this by-product mass continues to increase with the expansion of population and industry (Cui and Jahng, 2006;Zhang et al., 2010). Sludge disposal by landfill or incineration may no longer be feasible in the near future due to land scarcity, high waste charge and increasing stringent environmental control regulations (Weemaes et al., 2000;Wong et al., 2013;Wang et al., 2018). Therefore, the strategy for sludge management is shifting towards its reutilization as a potential source for renewable energy (Mossakowska et al., 1998;Rulkens, 2007). In this regard, the anaerobic digestion while producing renewable energy in the form of biogas (Riviere et al., 2009;Sundberg et al., 2013).
The processes of wastewater treatment using activated sludge technology and sludge disposal using anaerobic digestion depend on the synergistic interactions of microorganisms (Vanwonterghem et al., 2014;Wagner et al., 2002;Zakrzewski et al., 2012;Zhang et al., 2012). The culturing methods have been a very direct and effective way to characterize the microbial community (Neilson, 1978). However, most of the microorganism in complex environment cannot be cultured in an artificial medium in the laboratory and the diversity of the uncultured microorganism is quite considerable (Riviere et al., 2009). The molecular methods, such as polymerase chain reaction (PCR)-denaturing gradient gel electrophoresis (Supaphol et al., 2011;Ye and Zhang, 2011), terminal restriction fragment length polymorphism (Liu et al., 1997), cloning (Rincón et al., 2008;Schuppler et al., 1995), fluorescence in situ hybridization (Braguglia et al., 2012;Erhart et al., 1997) etc., greatly promoted our understanding of the microbial community. Typically, these molecular methods are gained through directly extracted DNA, but in the case of sludge samples, the problem is that this DNA contains not only the active microbial populations of interest but also genes of dormant, dead, and degraded cells. Because of these non-target pools that archive past and inactive genetic diversity, sludge DNA has been claimed to represent sludge history.
RNA, on the other hand, is more responsive with a higher turnover rate within cells.
RNA-based molecular methods can profile the structure-function links of microbial communities of environmental samples, which is a central theme of microbial ecology since its beginning (Brettar et al., 2012b;Sudhagar et al., 2018). However, RNA-based molecular methods are difficult to operate due to the easy degradability of RNA. If the two nucleic acid pools are extracted from the same environmental sample and analyzed with the same methodology, the structures of the DNA-based (likely including partly past) and RNA-based (likely currently viable and potentially more active) communities can be compared, and then evaluate the similarities between total and active microbial community. PCR-based community Illumina sequencing such as MiSeq is one popular high throughput sequencing systems due to its high quality and low cost. MiSeq of DNA-based and RNA-based has been successfully used in investigating total and active microbial community in various samples, such as forest soil (Baldrian et al., 2012a), land farming soil (Mikkonen et al., 2014), sediment from hydrothermal vent field (Lanz et al., 2011), and seawater (Brettar et al., 2012a). There are also several studies applying these methods in exploring total and active microbial community in anaerobic digestion reactors (Zakrzewski et al., 2012;Franchi et al., 2018). However, analysis of the similarities between total and active microbial community in different types sludge using Illumina MiSeq is still very limited.
The purpose of this study was to evaluate that whether understanding the structurefunction of microbial community of different types sludge using DNA-based molecular methods is suitable, according to analysis the similarities between total and active microbial community using DNA-based and RNA-based Illumina MiSeq. Procaryotic communities in different types sludge samples (including primary sludge, excess sludge, mixed sludge, and digested sludge) were evaluated in this study. For each sludge sample, we extracted both DNA and RNA, and then prepared and pyrosequenced PCR amplicons from both extractions.

Sludge digestion system description and sampling
In this study, all sludge samples were taken from a full-scale WWTP and its subsequent sludge anaerobic digestion system. Sampling was performed on May 28, and got 8 sludge samples. These samples were respectively primary sludge (PS) from primary settling tank, excess sludge (ES) from secondary settling tank, primary digested sludge I (PDS-I) from primary digester I, primary digested sludge II (PDS-II) from primary digester II, primary digested sludge III (PDS-III) from primary digester III, secondary digested sludge (SDS) from secondary digester, concentrated sludge (CS) (PS and ES mixed), and dewatering sludge (DS) (SDS and PS mixed, and then dewatered). All samples were collected in triplicated and were keep immediately in liquid nitrogen before being transported to the laboratory. Samples were stored at -80°C until DNA and RNA extraction.

Nucleic acid extraction and reverse transcription
DNA was extracted from 2 mL sludge (except DS, 0.5 g) using a PowerSoil DNA isolation kit (MoBio Laboratories, Carlsbad, CA, USA) according to the manufacturer's instructions. DNA was diluted 0-, 5-, and 10-fold for PCR, to adjust the DNA concentration to the proper condition for PCR, and to reduce the influence of the PCR inhibitors (e.g., humic substances).
RNA extraction was performed as previously (Wang et al., 2011). RNA was extracted from 2.0 g sludge using RNA PowerSoil total RNA isolation Kit (MoBio, Carlsbad, CA, USA). Extracted RNA was purified using Illustra MicroSpin Columns S-400 HR (GE Healthcare, Buckinghamshire, UK) to remove humic substances, and the remaining DNA was removed using a Turbo DNA-free kit (Applied Biosystems, Foster City, CA, USA). RNA was further purified and concentrated using an RNA Clean & Concentrator-5 kit (Zymo Research, Orange, CA, USA). The amount of purified RNA was measured using a NanoDrop 1000 (NanoDrop Products, Wilmington, DE, USA). The absence of DNA carryovers in the RNA samples was verified by PCR targeting the 16S rRNA gene without reverse transcription. Purified RNA was reverse transcribed using random hexamers and PrimeScript Reverse Transcriptase (Takara Bio, Otsu, Shiga, Japan) with 20 U of SUPERase-In RNase Inhibitor (Applied Biosystems).
Synthesized cDNA was diluted 10-fold for PCR. Replicate DNA and cDNA samples were pooled and used for Illumina MiSeq Sequencing.

Barcoge amplicon MiSeq sequenving
The V4 region of 16S rRNA gene was targeted and amplified for all cDNA and DNA libraries prepared using the universal forward primer 515F (5'-barcode-GTGCCAGCMGCCGCGGTAA-3') and the universal reverse primer 909R (5'-CCCCGYCAATTCMTTTRAGT -3'). The barcode is a sixteen-base sequence unique to each sample. PCR reactions were performed in triplicate in a 50-μL mixture containing 5 L 10×PCR buffer, 2.0 L 2.5 mM MgCl 2 , 2.0 L 2.5 mM dNTP, 2.0 L 10 mM forward primer, 2.0 L 10 mM reverse primer, 0.5 L TransStart Fast Pfu DNA Polymerase (TianGen, China), 10 ng DNA (or cDNA) and ddH 2 O. The PCR amplification program included initial denaturation at 94°C for 3 min, followed by 30 cycles of 94°C for 40 s, 56°C for 60 s, and 72°C for 60 s, and a final extension at 72°C for 10 min. Then, the PCR products of 3 replicated were mixed and purified with the EZNA ® Cycle-Pure Kit (Omega Bio-tek Inc, Doraville, GA, USA) and quantified using NanodropND-1000 UV-Vis Spectrophotometer (NanoDrop Technologies, Wilmington, DE, USA). All samples were pooled together with equal molar amount from each sample. The sequencing samples were prepared using TruSeq DNA kit according to manufacture's instruction. The purified library was diluted, denatured, re-diluted, mixed with PhiX (equal to 30% of final DNA amount) as described in the Illumina library preparation protocols, and then applied to an Illumina Miseq system for sequencing with the Reagent Kit v2 2×250 bp as described in the manufacture manual.

Processing and analyzing of sequencing data
Raw FASTQ files were de-multiplexed and quality-filtered using QIIME. The sequence data were processed using QIIME Pipeline-Version 1.7.0 (http://qiime.org/tutorials/tutorial.html) with the following criteria: (i) The 300-bp reads were truncated at any site that obtained an average quality score of <20 over a 10-bp sliding window, and the truncated reads shorter than 50 bp were discarded; (ii) exact barcode matching, two nucleotide mismatch in primer matching, and reads containing ambiguous characters were removed; and (iii) only overlapping sequences longer than 10 bp were assembled according to their overlapped sequence. Reads that could not be assembled were discarded. Sequences were clustered into operational taxonomic units (OTUs) at a 97% identity threshold. The aligned 16S rRNA gene sequences were used for chimera check using the Uchime algorithm (Edgar et al., 2011).
The rarefaction analysis based on Mothur v.1.21.1 (Schloss et al., 2009) in QIIME was conducted to reveal the diversity indices, including the Chao1, coverage, and Shannon diversity indices. The beta diversity analysis was performed using UniFrac (Lozupone et al., 2011) to compare the results of the principal component analysis (PCA) and cluster analysis (CA) using the Vengan and Permute package in R (x64 3.2.2). Jarccard's coefficient was used to indicate the similarity between active and total microbial communities in same sludge.

Sequencing results and diversity indices
As shown in Table 1 Results of CA (as shown in Figure 1) and PCA (as shown in Figure 2) showed that microbial community derived RNA and DNA of the same sludge sample are very similar, and aerobic environment samples (such as PS, ES, and CS) with anaerobic environment sample (e.g., PDS and SDS) were in different partitions. For example, PDS and SDS were clustered together due to all of them in anaerobic environment.

Prokaryotic community compositions
We applied Illumina Miseq to evaluate whether and how DNA-based (16S rRNA gene) and RNA-based (16S rRNA gene) microbial community derived from the same sludge sample differed, to evaluate differences and similarities in microbial community (DNAbased or RNA-based) derived from different sludge samples. As shown in Table 2, microbial communities derived RNA and DNA in sludge samples were predominated by bacteria, with abundances between 80.80% and 92.00%, while abundances of archaea were between 0.20% and 4.70%. The ratio of archaeal sequences was higher in digested sludge samples (mean 4.13% and 2.93% derived DNA and RNA) than in other sludge samples.

Prokaryotic community composition in PS
Total prokaryotic community and active prokaryotic community in PS were clustered together with high similarity (Jarccard's coefficient, 81.78%) at phylum level. As shown in Figure 3, Firmicutes (36.71% and 38.92%), Proteobacteria (35.20% and 40.52%), Bacteroidetes (10.58% and 8.96%), Actinobacteria (9.64% and 6.43%), Euryarchaeota (3.38% and 0.94%), and Chloroflexi (1.03% and 0.83%) were dominated in total prokaryotic community and active prokaryotic community. Among them, Firmicutes and Proteobacteria were highly abundant in two prokaryotic communities. Meanwhile, abundances of those 2 phyla in active prokaryotic community were higher than these in total prokaryotic community. However, abundances of other phyla in active prokaryotic community were lower than these in total prokaryotic community.
Meanwhile, abundances of those 3 phyla in active prokaryotic community were higher than these in total prokaryotic community. In addition to 3 phyla, TM7 and Chlorobi in active prokaryotic community were a little higher than these in total prokaryotic community. However, abundances of other phyla in active prokaryotic community were lower than these in total prokaryotic community.

Prokaryotic community composition in digested sludge
Total prokaryotic community and active prokaryotic community in digested sludge were clustered together with high similarity (Jarccard's coefficient, 81.81-85.68%) at phylum level. As shown in Figure 3, our results showed that prokaryotic communities were very similar in three parallel anaerobic digesters, suggesting that experimental results in this study are repeatable and credible. Moreover, prokaryotic community in secondary digester (normal temperature digestion) was similar to these in primary digesters (mesophilic digestion). Synergistetes in active prokaryotic community were higher than these in total prokaryotic community. However, abundances of other phyla in active prokaryotic community were lower than these in total prokaryotic community. Moreover, Verrucomicrobia, WWE1, Chloroflexi, Bacteroidetes, and Firmicutes were highly abundant in two prokaryotic communities.

Prokaryotic community composition in mixed sludge
At phylum level, the similarities between total prokaryotic community and active prokaryotic community in mixed sludge were lower than that in single sludge.
Jarccard's coefficient of CS and DS were respectively 74.41% and 72.10%. Just like at phylum level, the similarities between total prokaryotic community and active prokaryotic community in mixed sludge were lower than that in single sludge because that Jarccard's coefficient of CS and DS were respectively 44.37% and 41.72%.
Among them, Proteobacteria, Actinobacteria, Firmicutes, Chloroflexi, and Bacteroidetes were highly abundant in two prokaryotic communities. Meanwhile, abundances of Proteobacteria, Chloroflexi, Bacteroidetes, and Euryarchaeota in active prokaryotic community were higher than these in total prokaryotic community.

Discussion
The molecular analysis is now central to studies examining the diversity of microorganisms in the environment. Much of what is currently known about the microbial ecology of environment samples (such as soil, water, and sludge) has been inferred from studies targeting DNA. Even though there were studies on analysis of total and active bacterial communities in soil (Baldrian et al., 2012b;Griffiths et al., 2000), seawater (Brettar et al., 2012b) For this overall structure, the understanding of the core community with the core taxa performing all major functions is essential (Hofle et al., 2008).

The high similarity between total and active prokaryotic communities in sludge
A major goal of this study was to test the similarity between total and active prokaryotic communities in different types sludge samples. The Jarccard's coefficients of all sludge samples showed that the similarities between total and active procaryotic communities of them were considerable high in this present study. For example, Firmicutes and Proteobacteria were dominant phyla in PS's total and active procaryotic communities.
A previous study of analysis of bacterial core communities in the central Baltic by comparative RNA-DNA-based fingerprinting showed that large discrepancy between the different types (RNA-, DNA-based fingerprints) of fingerprints in surface water, whereas an unexpected high similarity of both types in the anoxic deep water due to a stable and active core community in this layer (Brettar et al., 2012b). This previous study revealed that low oxidative stress could be a conceivable mechanism explaining the increase of similarity between the RNA-and DNA-based community fingerprints in the anoxic zone (Brettar et al., 2012b). However, in this present study, low oxidative stress was not the conceivable mechanism. In sludge samples, although some of the present bacteria are not as active as others due to the high dynamics, the death rate of most procaryotes may be similar, which explained that the similarity was considerable high in the aerobic and anaerobic sludge samples. This was different from results in soil and surface seawater (Baldrian et al., 2012b;Brettar et al., 2012b). It may be because that sludge is controlled environmental sample, but soil and surface seawater are natural environment sample.
In this present study, procaryotic communities in different types sludge samples were very different, because different type sludge was in different environment and played different roles. For example, Actinobacteria, Proteobacteria, and Chloroflexi were dominant phyla in excess sludge, because excess sludge was similar to activated sludge. Previous studies have shown that those 3 phyla were dominant in activated sludge from wastewater treatment plants (Gao et al., 2016;Ju et al., 2014;Ju and Zhang, 2015). However, Verrucomicrobia and WWE1 were dominant phyla in digested sludge, which were showed in anaerobic reactor conditions (Limam et al., 2014;Riviere et al., 2009).
Moreover, the predominant phylum and genus in total and active procaryotic communities were same in each sludge, and their abundances in active procaryotic community were higher than those in total procaryotic community in all different types sludge samples. It may be because that the predominant phylum and genus are functional microorganism in sludge samples, which is easy to detect using RNA-based illumine MiSeq.

The similarity between total and active prokaryotic communities in running sludge higher than that in idle sludge
More intriguing in our study was that the similarity between the RNA and DNA-based procaryotic communities in running sludge higher than that in idle sludge due that the Jarccard's coefficient of digested sludge was the highest in all types sludge samples. It might be because that the running sludge (such as digested sludge and ES) played roles in the processes of wastewater treatment and sludge anaerobic digestion. Thus, abundant functional microorganisms existed in the running sludge. For example, Candidatus M. parvicella, which might have been responsible for phosphorus removal during sludge bulking in plants with similar configurations (Wang et al., 2014a;Wang et al., 2014b), and Thauera, which is affiliated with or closely related to denitrifying bacteria (Wang et al., 2014a), in ES's active procaryotic community were more abundant than these in ES's total procaryotic community. An important phenomenon was that abundance of Methanosaeta, in digested sludge's active procaryotic community was 3 times higher than that in digested sludge's total procaryotic community. However, in the idle sludge, the mortality most of procaryotes greatly increased due to some reasons such as the lack of nutrients.

Conclusion
Our study demonstrated the high similarity between active and total procaryotic communities in different types sludge samples using RNA-and DNA-based Illumina MiSeq. Meanwhile, the similarity between active and total procaryotic communities in running sludge higher than that in idle sludge. This study showed that the popular DNAbased molecular technology could be applicable to profile functional procaryotic community of sludge samples to a certain degree.