Improvement in Microbiota Recovery Using Cas-9 Digestion of Mānuka Plastid and Mitochondrial DNA

Abstract

Understanding host-microbe interactions in planta is an expanding area of research. Amplicon sequencing of the 16S rRNA gene is a powerful and common method to study bacterial communities associated with plants. However, the co-amplification of mitochondrial and plastid 16S rRNA genes by universal primers impairs the sensitivity and performance of 16S rRNA sequencing. In 2020, a new method, Cas-16S-seq, was reported in the literature to remove host contamination for profiling the microbiota in rice, a well-studied domestic plant, by engineering RNA-programmable Cas9 nuclease in 16S rRNA sequencing. For the first time, we tested the efficiency and applicability of the Cas-16S-seq method on foliage, flowers, and seed of a non-domesticated wild plant for which there is limited genomic information, Leptospermum scoparium (mānuka). Our study demonstrated the efficiency of the Cas-16S-seq method for L. scoparium in removing host contamination in V4-16S amplicons. An increase of 46% in bacterial sequences was found using six guide RNAs (gRNAs), three gRNAs targeting the mitochondrial sequence, and three gRNAs targeting the chloroplast sequence of L. scoparium in the same reaction. An increase of 72% in bacterial sequences was obtained by targeting the mitochondrial and chloroplast sequences of L. scoparium in the same sample at two different steps of the library preparation (DNA and 1st step PCR). The number of OTUs (operational taxonomic units) retrieved from soil samples was consistent when using the different methods (Cas-16S-seq and 16S-seq) indicating that the Cas-16S-seq implemented for L. scoparium did not introduce bias to microbiota profiling. Our findings provide a valuable tool for future studies investigating the bacterial microbiota of L. scoparium in addition to evaluating an important tool in the plant microbiota research on other non-domesticated wild species.

Keywords: 16S rRNA gene amplicon sequencing; Cas-16S-seq; Host contamination; Microbiota profiling; Plant microbiome.

Fig. 1

In vitro cleavage using RNP complex on V4-16S rRNA gene amplicons. gRNA showed cleavage efficiency of Leptospermum scoparium mitochondrial and chloroplast sequences. Six gRNAs (mt-88, mt-91, mt-107, cp-81, cp-131, and cp-189) were used alone (lanes 1 to 6) or mixed (lanes 7 to 10 and 11 to 15) to evaluate their cleavage performance. Lane 1 is mt-88, lane 2 is mt-91, lane 3 is mt-107, lane 4 is cp-81, lane 5 is cp-131, lane 6 is cp-189, lane 7 is Mix-1 (= mt-88 + cp-81), lane 8 is Mix-2 (= mt-88 + cp-131), lane 9 is Mix-3 (= mt-88 + cp-189), lane 10 is Mix-4 (= mt-91 + cp-81), lane 11 is Mix-5 (= mt-91 + cp-131), lane 12 is Mix-6 (= mt-91 + cp-189), lane 13 is Mix-7 (= mt-107 + cp-81), lane 14 is Mix-8 (= mt-107 + cp-131), and lane 15 is Mix-9 (= mt-107 + cp189). Control samples (lanes C) underwent the same cleavage step without RNP complex

Fig. 2

Implementation of the Cas-16S-seq method on Leptospermum scoparium ‘Crimson Glory’ floral DNA. A Evaluation of the gRNA combinations on eliminating plastid and mitochondrial sequences. Box plot representing the ratio bacterial reads on total reads for each sample. Samples are grouped by condition, 16S-seq = library preparation without the use of CRIPSR-Cas9. Control = samples underwent the same cleavage step without RNP complex, Mix-1 = mt-88 + cp-81, Mix-2 = mt-88 + cp-131, Mix-3 = mt-88 + cp-189, Mix-4 = mt-91 + cp-81, Mix-5 = mt-91 + cp-131, Mix-6 = mt-91 + cp-189, Mix-7 = mt-107 + cp-81, Mix-8 = mt-107 + cp-131, Mix-9 = mt-107 + cp-189, Mix-10 = all gRNAs, Mix-11 = mt-88 + mt-91, Mix-12 = mt-88 + mt-107, Mix-13 = mt-107 + mt-91, Mix-14 = cp-81 + cp-131, Mix-15 = cp-81 + cp-189, and Mix-16 = cp-131 + cp-189. T-test results using Control samples as reference are shown as non-significant (ns) or significant (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001). B Evaluation of the enzymatic cleavage at the different steps of library preparation on eliminating plastid and mitochondrial sequences. Box plot representing the ratio bacterial reads on total reads for each sample. Samples are grouped by condition, 16S-seq = library preparation without the use of CRIPSR-Cas9, Control DNA = samples underwent the same cleavage at DNA step without RNP complex, Control Amp = samples underwent the same cleavage at 1st step PCR amplicon step without RNP complex, Control DNA + Amp = samples underwent the same cleavage at DNA and 1st step PCR amplicon step without RNP complex, DNA = library preparation with cleavage step on DNA, Amp = library preparation with cleavage step on 1st step PCR amplicon, DNA + Amp = library preparation with cleavage step on DNA and 1st step PCR amplicon. In vitro cleavage used RNP including mix-3 (gRNAs mt-88 and cp189). T-test results using Control-DNA samples as reference are shown as non-significant (ns) or significant (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001)

Fig. 3

Number of bacterial OTUs retrieved in four soil samples using 16S-seq and Cas-16S-seq methods. Evaluation of the gRNAs on introducing bias in bacterial OTUs recovery from soil samples. Box plot representing the number of OTUs of four soil samples per condition. Control = samples underwent the same cleavage step without RNP complex, Mix-1 = mt-88 + cp-81, Mix-5 = mt-91 + cp-131, and Mix-9 = mt-107 + cp-189. Wilcoxon-test results using Control samples as reference are shown as non-significant (ns) or significant (*p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001)

Fig. 4

Number of OTUs from L. scoparium samples processed with 16S-seq and Cas-16S-seq. This represents a subset of samples (n = 20) from the comparison of the Cas-16S-seq method to the standard 16S rRNA amplicon sequencing (16S-seq) performed on a total of 262 DNA samples. The lines connect the sample that was process with both method for direct comparison of the number of OTUs