Self versus non-self discrimination during CRISPR RNA-directed immunity

Abstract

All immune systems must distinguish self from non-self to repel invaders without inducing autoimmunity. Clustered, regularly interspaced, short palindromic repeat (CRISPR) loci protect bacteria and archaea from invasion by phage and plasmid DNA through a genetic interference pathway. CRISPR loci are present in approximately 40% and approximately 90% of sequenced bacterial and archaeal genomes, respectively, and evolve rapidly, acquiring new spacer sequences to adapt to highly dynamic viral populations. Immunity requires a sequence match between the invasive DNA and the spacers that lie between CRISPR repeats. Each cluster is genetically linked to a subset of the cas (CRISPR-associated) genes that collectively encode >40 families of proteins involved in adaptation and interference. CRISPR loci encode small CRISPR RNAs (crRNAs) that contain a full spacer flanked by partial repeat sequences. CrRNA spacers are thought to identify targets by direct Watson-Crick pairing with invasive 'protospacer' DNA, but how they avoid targeting the spacer DNA within the encoding CRISPR locus itself is unknown. Here we have defined the mechanism of CRISPR self/non-self discrimination. In Staphylococcus epidermidis, target/crRNA mismatches at specific positions outside of the spacer sequence license foreign DNA for interference, whereas extended pairing between crRNA and CRISPR DNA repeats prevents autoimmunity. Hence, this CRISPR system uses the base-pairing potential of crRNAs not only to specify a target, but also to spare the bacterial chromosome from interference. Differential complementarity outside of the spacer sequence is a built-in feature of all CRISPR systems, indicating that this mechanism is a broadly applicable solution to the self/non-self dilemma that confronts all immune pathways.

Figure 1. Protection of nes target by spacer flanking sequences

a, Direct repeats (DR1–4, purple boxes) and spacers (1–3, colored boxes) of the S. epidermidis RP62a CRISPR locus were cloned into pC194 generating pCRISPR(wt) and its deletion variant, pCRISPR(del). b, pNes(wt) and pNes(mut) contain wild-type and mutated nes target sequence of pG0400 (highlighted in yellow, with mutations in red). In pNes(5′DR,8-1) and pNes(3′DR,15) nes target flanks were replaced by repeat sequences present upstream and downstream of spc1 (highlighted in purple), respectively. c, Individual nucleotides upstream of nes target were replaced by those present upstream of spc1 (highlighted in purple). The G at position −2 was also changed to C and T (in red). The AGA sequence (underlined) is shared by both the nes target and spacer 5′ flanking sequences. All plasmids were transformed into S. epidermidis RP62a and its isogenic Δcrispr mutant. The average of at least three independent measures of the transformation efficiency (determined as cfu/μg DNA) is reported and error bars indicate 1 s.d.

Figure 2. Complementarity between crRNA and target DNA flanking sequences is required for protection

a, Schematic of the complementarity between the flanking sequences of crRNA (top, highlighted in purple) and target DNA (bottom). The red box indicates the nucleotides mutated in the experiments shown in b. b, c, Conjugation assays of pG0400 and its mutant variants, using as a recipient the Δcrispr strain harboring different pCRISPR plasmids. Mutations are shown in red. Conjugation efficiency was determined as transconjugant cfu/recipient cfu; the average of at least 3 independent experiments is reported.

Figure 3. Mutations in upstream flanking sequences of CRISPR spacers elicit autoimmunity

a, Deletions were performed in the flanking repeats of spc1: the 3′ half of DR2, the 5′ half of DR2, and all of DR1 were deleted from pCRISPR(wt) in pDR2(3′del), pDR2(5′del) and pDR1(del), respectively. b, c, Substitutions (red) were introduced in the 5′ flanking sequence (highlighted in purple) of spc1 (highlighted in yellow), generating different pDR1 variants that were tested by transformation. All plasmids were transformed into S. epidermidis RP62a and Δcrispr strains. The average of at least three independent measures of the transformation efficiency (determined as cfu/μg DNA) is reported and error bars indicate 1 s.d.

Figure 4. Requirements for targeting and protection during CRISPR immunity

a, In S. epidermidis, CRISPR interference is enabled by mismatches between target DNA and crRNA sequences upstream of the spacer. Formation of at least 3 base pairs at positions −4, −3 and −2 eliminates targeting. b, Complementarity between the S. epidermidis CRISPR locus and the crRNA 5′ terminus protects it from interference. Disruption of base pairing at positions −4,−3 or −3,−2 eliminates protection. c, General model for the prevention of autoimmunity in CRISPR systems. The ability of crRNA termini (5′, 3′, or both) to base pair with potential targets enables discrimination between self and non-self DNA during CRISPR immunity.