Formation and Stoichiometry of CRISPR-Cascade Complexes with Varying Spacer Lengths Revealed by Native Mass Spectrometry.

Sabine Wittig,Inga Songailiene,Carla Schmidt

doi:10.1021/jasms.9b00011

Abstract

The adaptive immune system of bacteria and archaea against viral DNA is based on clustered, regularly interspaced, short palindromic repeats (CRISPRs) which are encoded in the host genome and translated into CRISPR RNAs (crRNAs) containing single spacer sequences complementary to foreign DNA. crRNAs assemble with CRISPR-associated (Cas) proteins forming surveillance complexes that base-pair with viral DNA and mediate its degradation. As specificity of degradation is provided by the crRNA spacer sequence, genetic engineering of the CRISPR system has emerged as a popular molecular tool, for instance, in gene silencing and programmed DNA degradation. Elongating or shortening the crRNA spacer sequence are therefore promising ventures to modify specificity toward the target DNA. However, even though the stoichiometry of wild-type complexes is well established, it is unknown how variations in crRNA spacer length affect their stoichiometry. The CRISPR-associated antiviral defense surveillance complexes of Streptococcus thermophilus (StCascade complexes) contain crRNA and five protein subunits. Using native mass spectrometry, we studied the formation and stoichiometry of StCascade complexes assembled on a set of crRNAs with different spacer lengths. We assigned all relevant complexes and gained insights into the stoichiometry of the complexes as well as their preferred assembly. We found that stable complexes, which incorporate or lose a (Cas7)2(Cse2)1-module, assemble on crRNA varied in length by 12-nucleotide units, while varying crRNA length in six-nucleotide units results in heterogeneous mixtures of complexes. Combining our results from the various variants, we generated an assembly pathway revealing general features of I-E type Cascade complex formation.

Highlights

The host transcription machinery produces CRISPR RNAs which contain single spacer sequences flanked by repeat sequences. crRNAs assemble with one or more Cas proteins to form surveillance complexes that base-pair with complementary foreign DNA followed by degradation of the invader, for instance, by Cas[3] helicase-nuclease in Type I systems.[2,5,6]
We investigated StCascade complexes assembled on crRNAs extended or shortened by odd-numbered 6-nucleotide units when compared with their 12-nucleotides counterparts (−18, −6, +6, and +18 nucleotides)
We show that native mass spectrometry is well-suited to study large and heterogeneous protein assemblies such as those assembled on crRNAs of varying lengths

Summary

■ INTRODUCTION

Bacteria and archaea developed an adaptive immune system against viral DNA encoded by clustered, regularly interspaced, short palindromic repeats (CRISPRs) and CRISPR-associated proteins (Cas proteins).[1,2] After infection with foreign DNA, DNA spacer sequences are inserted into the CRISPR locus of the host genome.[3,4] The host transcription machinery produces CRISPR RNAs (crRNAs) which contain single spacer sequences flanked by repeat sequences. crRNAs assemble with one or more Cas proteins to form surveillance complexes that base-pair with complementary foreign DNA followed by degradation of the invader, for instance, by Cas[3] helicase-nuclease in Type I systems.[2,5,6] Selection of spacer sequences from invader DNA after initial infection is guided by the presence of protospacer adjacent motifs (PAMs) in the foreign genome.[7]. CrRNAs assemble with one or more Cas proteins to form surveillance complexes that base-pair with complementary foreign DNA followed by degradation of the invader, for instance, by Cas[3] helicase-nuclease in Type I systems.[2,5,6] Selection of spacer sequences from invader DNA after initial infection is guided by the presence of protospacer adjacent motifs (PAMs) in the foreign genome.[7] The CRISPR locus of the host is lacking these sequences and is protected against selfdegradation. Cascade complexes with elongated or shortened crRNA spacers have been reported for E. coli,[19,20] Shewanella putrefaciens,[21] and S. thermophilus.[22] A recent study employing crRNA spacers up to +57 nucleotides showed that elongated spacer length increased DNA binding affinities and supports R-loop formation, DNA degradation does. The accurate molecular weight was determined from LC-MS/MS experiments. (B) Fragment spectrum of the N-terminal peptide of Cas[6] containing N-terminal methionine. y- and b-ion series are labeled (red and yellow). (C) Fragment spectrum of the N-terminal peptide of Cse[2] lacking N-terminal methionine. y- and b-ion series are labeled (red and yellow)

■ METHODS

■ DISCUSSION

■ CONCLUSIONS

■ ACKNOWLEDGMENTS

■ REFERENCES