Lysozyme Treatment Makes 16S rRNA Amplicon Sequencing Results Less Biased


 Background

Amplicon sequencing is widely applied in gut bacteria structure analysis. However, the proportion of Gram-positive bacteria may greatly affect the results of microbial community analysis. Lysozyme is an effective agent to extract DNA of Gram-positive bacteria. In this study, we assessed the influence of lysozyme treatment on results of Bactrocere dorsalis rectal bacteria structure.
Result

The results indicated that the total bacteria content can be significantly increased in lysozyme treated samples. Moreover, rectal bacteria diversity was significantly higher in lysozyme treated samples. A detail analysis revealed that abundance of Gram-positive bacteria significantly increased in samples treated with lysozyme.
Conclusion

This study indicates that lysozyme treatment before DNA extraction is an effective way to reduce bias in bacteria structure analysis, especially for samples with high proportion of Gram-positive bacteria.


Result
The results indicated that the total bacteria content can be signi cantly increased in lysozyme treated samples. Moreover, rectal bacteria diversity was signi cantly higher in lysozyme treated samples. A detail analysis revealed that abundance of Gram-positive bacteria signi cantly increased in samples treated with lysozyme.

Conclusion
This study indicates that lysozyme treatment before DNA extraction is an effective way to reduce bias in bacteria structure analysis, especially for samples with high proportion of Gram-positive bacteria.

Background
The structure of gut bacteria can be shaped by various factors [1]. Diet can shape the gut bacteria of human [2]. On the contrary, gut bacteria can also affect the metabolism and behavior of their hosts [3,4].
Studies have indicated that even obesity and lean of human can be affected by the gut bacteria [5,6].
With the development of sequencing technique, cost of 16s rRNA amplicon sequencing is dramatically reduced and functions of more and more gut bacteria are investigated in different hosts [7]. Insects and other terrestrial arthropods are closely associated with symbiotic microorganisms [8][9][10]. Among these special hosts, gut bacteria are pervasive. Thanks to the development of the 16s rRNA amplicon sequencing method, breakthroughs are made by identifying and classifying the gut bacteria of more and more insect hosts [11].
Previous study has shown that gut bacteria sequencing results may be affected by various factors [12][13][14]. DNA extraction method is one of the most vital factors that can in uence the sequencing results besides the primers and sequencing platform [15]. Normal extracting steps may lead to so huge bias on the result of sequencing that it leads to completely different conclusions. For example, some studies have showed that Proteobacteria (most are Gram-negative bacteria) occupied the dominant status in B. dorsalis guts, while others have claimed that the Firmicutes (most are Gram-positive bacteria) is the major class [5,16]. Such difference may be caused by Gram-positive bacteria, since different DNA extraction method can result in a huge difference in abundance of Gram-positive bacteria. Therefore, it is urgent to optimize the method of DNA extraction to reduce the bias in gut bacteria analysis.
With the invention of Gram staining, bacteria can be divided into two groups by the staining color, Grampositive and Gram-negative bacteria [17,18]. The cell wall structures of the Gram-positive and Gramnegative bacteria are totally different. For Gram-positive bacteria, the cell wall is thicker with more layers of membrane and peptidoglycan and less lipid. Lysozyme can destroy the β-1,4-glycosidic bond between n-acetyl intracytoic acid and N-acetyl glucosaccharide which is an important component in the cell wall of Gram-positive bacteria. With β-1,4-glycosidic being destroyed, the insoluble mucosaccharide in cell wall can be decomposed into soluble glycopeptides, resulting in the rupture of cell wall and the escape of the contents and nally causing the dissolution of bacteria [19,20]. Therefore, it is an effective way to extract the DNA of Gram-positive bacteria by incubating with lysozyme.
In this study, we analyzed the 16s rRNA amplicon sequencing results of B. dorsalis rectum bacteria for which DNA was extracted by Bacterial genome kits adding with lysozyme or not. The results showed that the rectal bacteria diversity and Gram-positive bacteria abundance in lysozyme treatment was signi cantly higher than none-lysozyme treatment.

Results
Lysozyme incubation affects the total bacteria content extracted from rectum qPCR was used to test if the total bacteria content in rectum can be affected by lysozyme incubation. The results showed that total bacteria content could be signi cantly affected by lysozyme incubation. For newly emerged male, bacteria content in rectum was signi cantly higher if the DNA was extracted without lysozyme treatment. However, bacteria contents were signi cantly lower if the DNA was extracted without lysozyme treatment in 3-and 6-day-old male. For 9-and 12-day-old male, there was no difference for total bacteria contents ( Figure 1).

Rectum bacteria diversity was signi cantly affected by lysozyme incubation
To test if lysozyme incubation would affect bacteria diversity, 16S rRNA amplicon sequencing was used to investigate rectum bacteria diversity. We assessed the impact of lysozyme incubation on sequencing data. The results showed that both sequencing tag and OTU numbers were signi cantly higher in lysozyme incubation samples (Figure 2A and 2B). Furthermore, α diversity and β diversity represented by Shannon indices and PCoA analysis were assessed. The results were in agreement with the sequencing data evaluation results: Shannon indices were signi cantly higher in lysozyme incubation samples ( Figure 2C). PCoA plots of non-lysozyme treatment samples showed that the rectum samples collected at different time points were highly clustered, while distinct separations were showed in lysozyme treatment samples ( Figure 2D and 2E). These results indicate lysozyme treatment before DNA extraction is vital for investigating the bacteria diversity in rectum accurately.
Lysozyme treatment signi cantly affected the taxonomic results Taxonomic compositions of the rectum bacteria were compared to test whether lysozyme treatment can affect the bacteria community composition. For lysozyme treated samples, rectum bacteria changed signi cantly at different time points. The dominant taxa in 0d samples was Gammaproteobacteria and Bacilli, while Bacilli was the only dominant taxa in 3d, 6d, 9d, and 12d samples. However, bacteria composition patterns were signi cantly different in non-lysozyme treated samples.
Gammaproteobacteria, Bacilli and Clostridia were the dominant taxa in rectum at all the time points and the abundance of the same bacteria group uctuated greatly in the repeated samples, which indicated that DNA extraction e ciency of the rectum bacteria was very unstable (Figure 3).
Since lysozyme can effectively improve DNA extraction e ciency of Gram-positive bacteria [20], We divided the rectum bacteria into two groups (Gram-negative and Gram-positive bacteria) and compared their abundance. In the top six groups in term of abundance, Bacilli, Clostridia and Actinobacteria were mainly Gram-positive bacteria, while Gammaproteobacteria, Bacteroidia and Alphaproteobacteria were mainly Gram-negative bacteria ( Figure 3). For 0d rectum samples, none of Gram-positive and Gramnegative bacteria were signi cantly different in abundance between samples treated with lysozyme and control ( Figure 4). However, Gram-positive bacteria abundance were signi cantly higher in samples treated with lysozyme for 3d, 6d, 9d and 12d rectum and the abundance of Gram-negative bacteria were signi cantly lower (Figure 4). These results indicated that lysozyme can signi cantly increase the abundance of Gram-positive bacteria extracted in rectum.

Discussion
In this study, bacteria content and structure in B. dorsalis rectum were compared between rectum DNA extracted with lysozyme treatment and none-lysozyme treatment. Our results indicated that lysozyme treatment before DNA extraction could signi cantly increase the extracted bacteria content and broad the range of the rectal bacteria communities.
Lysozyme treatment before DNA extraction could signi cantly increase the content of bacteria in B. dorsalis rectum at the early stages after eclosion. With the development of the host insects, the gut bacteria abundance may be affected by metabolism of host insect [21]. Studies have indicated that the bacteria content of insect is relatively low at the early stages after elosion [21,22]. Thus, the bacteria content may be signi cantly lower in rectum for which Gram-positive bacteria were dominant and DNA was extracted without lysozyme treatment. Though the bacteria content showed no difference at the later stages, the bacteria structures were signi cantly different. Physically grinding and thermal incubation could increase the e ciency of DNA extraction, but it wasn't enough to destroy the cell wall of Grampositive bacteria entirely [23]. Thus, extracting DNA without lysozyme treatment may lose the Grampositive bacteria, which may lead to bias in bacteria structure [24].
In B. dorsalis rectum, Bacilli, Gammaproteobacteria and Alphaproteobacteria were three dominant taxa in the lysozyme treatment while most bacteria in none-lysozyme treatment rectum were Gammaproteobacteria. Though some Gram-positive bacteria could be identi ed in none-lysozyme treatment samples, their abundance was signi cantly lower than the lysozyme treatment. Without lysozyme treatment, the abundance of Gram-positive bacteria had great uctuation within the same groups. Thus, those Gram-positive bacteria may be one of the factors that resulted in bias in the sequencing results.
Recent years, the 16s rRNA amplicon sequencing technology have become a fundamental technology in bacteria community analysis. Re ecting a real composition and diversity of the bacteria is the basis for analyzing the function of the related bacteira [25][26][27]. To reduce the bias of sequencing results, researchers continued to optimize the DNA extraction method before sequencing in different dimension including the extraction protocols, sampling methods and surface materials in different cases [7,28]. Some researchers had noticed the importance of lysozyme used in the DNA extraction before sequencing, while others relied solely on the DNA extraction for the convenient, which could lead to a dramatic difference in bacteria communities [29][30][31].

Conclusion
Our research had showed that lysozyme treatment before DNA extraction was rather necessary to obtain the accurate information. Combining lysozyme treatment and DNA extraction kit together can reach a more complete composition of the bacteria, especially for those samples in which Gram-positive bacteria abundance was high.

Rectal sample collection
Male adult rectums (0, 3, 6, 9 and 12 days old) were dissected and soaked in absolute ethyl alcohol, respectively. Eight rectums were dissected and put into one centrifuge tube as one sample, 10 samples were collected for each stage (0, 3, 6, 9 and 12 days old ies). Then the samples were divided into two groups, one group containing 5 samples were sent for DNA extraction with Lysozyme treatment, the other group was used for DNA extraction without lysozyme treatment.

DNA extraction
The rectal DNA was extracted with the Bacterial Genome Extraction Kit (Tiangen, Beijing, China). Brie y, the samples in the centrifuge tubes were grinded in a grinder. Then one group of grinded samples was incubated in 180 μl lysozyme (20 mg/ml) for 50 minutes under 37℃. Then the samples were used for DNA extraction with the Bacterial Genome Extraction Kit following the manufacturer's protocols. The other group of samples without lysozyme incubation was used as control.

Total bacteria content quanti cation
The total bacteria content in the collected rectal samples was evaluated by real-time quantitative PCR. The 16S rRNA gene fragment was ampli ed using the universal primer of bacteria: 338F (5'-ACTCCTACGGGAGGCAGCAG -3') and 518R(3'-GGTCGTCGGCGCCATTA-5') [32]. Then the genomic DNA of E. coli was extracted for ampli cation with the primers. The ampli ed fragment was then cloned into the pMD 18-T vector, which was then transferred into E. coli DH5α to reproduce. The reproduced vector was then extracted with a plasmid extraction kit, and diluted in a series of 10-fold dilutions to obtain 7 different plasmid concentrations (measured by Nanodrop spectrophotometer). A standard curve for qPCR was then generated by amplifying the 16S rRNA of the plasmid. The absolute abundance of bacteria in the rectal was determined by referring to the standard curve.
16S rRNA amplicon sequencing for rectal bacteria V3-V4 region of the 16S rRNA gene were ampli ed by PCR using primers 341F CCTACGGGNGGCWGCAG; 806R GGACTACHVGGGTATCTAAT. PCR reactions were performed in triplicate 50 μL mixture containing 5 μL of 10×KOD Buffer, 5 μL of 2.5 mM dNTPs, 1.5 μL of each primer (5 μM), 1 μL of KOD Polymerase, and 100 ng of template DNA. Then the amplicons were extracted from 2% agarose gels and puri ed using the DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, U.S.) and quanti ed using QuantiFluor-ST (Promega, U.S.). Puri ed amplicons were pooled in equimolar and paired-end sequenced (2×250) on an Illumina platform according to the standard protocols. The raw reads were deposited into the NCBI Sequence Read Archive (SRA) database (Accession Number: PRJNA669205). Then the sequenced raw reads for each sample were cleaned and analyzed according to the standards of our previous studies [33].

Declarations
Ethics approval and consent to participate Ethical approval was not required for this work as no human or animal samples and/or data were analysed in this study. B. dorsalis used in this study were collected from a carambola (Averrhoa carambola) orchard in Guangzhou, Guangdong Province and reared in the laboratory under the following conditions: 25 ± 1 °C, 16:8 h light: dark cycle, 70-80% RH. Adult ies were fed with yeast and sugar mixture.

Consent for publication
Not applicable.
Availability of data and materials Sequence data are available from the NCBI Short Read Archive. The accession numbers are PRJNA669205 (lysozyme treated samples) and PRJNA780089 (non-lysozyme treated samples) .     Gram-positive (A) and Gram-negative (B) bacteria relative abundance comparison. Asterisks indicate signi cant differences (*** P < 0.001, NS: no signi cance).