CRISPR Comparisons

CRISPR systems are widespread throughout bacteria and archaea, and they are also diverse. The three best-characterized types of systems are Types I, II, and III. Here are some key areas of similarity and contrast between the three system types.

Distribution and diversity

10 microbial cells: 9 have CRISPR systems icons, and 10th is empty, with a phage on it. Of cells with CRISPR systems, 2 cells have a Type II Cas9 system, 3 have a Type I system, 1 has a Type III system, and 2 have both a Type I and Type III system.

CRISPR systems are found throughout the bacterial and archaeal kingdoms – about 40% of bacterial genome sequences and over 80% of archaeal genome sequences contain CRISPR genes. CRISPR systems are diverse, but that diversity isn’t as simple as following the relationship of their bacterial and archaeal hosts. CRISPR system components appear to have been lost and gained often, so closely related bacterial species may have different types of systems, while distantly related species may have similar systems. And cells aren’t limited to a single system – some contain multiple clusters of CRISPR associate (cas) genes.

CRISPR systems are categorized based on their components, into classes and then types. There are currently six recognized CRISPR types, and research has continued to expand the number of known CRISPR systems. The features of Type I, II and III systems are listed below.

Type I systems are the most abundant systems in nature and are found in bacteria and archaea. Their signature gene is cas3, a nuclease that is, in most cases, not part of the CRISPR complex but is rather recruited during interference.

Type II systems are almost exclusively found in bacteria; most process the crRNA using a bacterial enzyme, RNase III, for processing the crRNA, that is absent in archaea. Their signature gene is cas9, the single protein component of the Type II CRISPR complex.

Type III systems are found in bacteria and archaea, and are commonly found in cells with at least one other CRISPR system. Their signature gene is cas10, the large subunit of the Type III CRISPR Complex, with domains that are responsible for the DNase and cOA production activities during interference.

CRISPR complex composition

The composition of the CRISPR complex used for interference determines the “class” of CRISPR systems–the first level of classification, before they are divided into Types. Class 1 systems have complexes made up of multiple protein subunits and include Types I and III, while Class II systems have a single protein and include Type II. All the complexes include a CRISPR RNA (crRNA) that is derived from the spacers in the CRISPR array.

Labeled Cascade, crRNA & Cas3. Cascade: grey elongated arch of overlapping ovals, containing crRNA: a thick line with short black tail, center green arch & black hairpin. Cas3 is separate, a red bean with paired triangles near the top & ovals lower. — **Type I systems**

use complexes known as Cascade (CRISPR-associated complex for antiviral defense), which contain multiple protein subunits and a crRNA. Cascade has a “seahorse” structure, with the protein subunits generally arranged into two twisted filaments. It has a backbone of repeating subunits spanning the head and body of the seahorse shape, with other subunit proteins flanking and making up the seahorse shape’s belly and tail. Cascade does not have DNA cleavage activity and recruits a separate protein, Cas3, to cleave target DNA that matches the crRNA.

Blue bilobed structure, Cas9, containing tracrRNA & crRNA. tracrRNA: a thick gray line, starts horizontal & folds back and forth, with bonds between parallel sections. crRNA: thick straight line, green (left) and black, bound to bottom of tracrRNA — **Type II systems**

have complexes made of a single protein (Cas9), a crRNA and a tracrRNA (trans-activating crRNA). The protein forms a bilobed structure, with the target DNA sitting in between the “jaws” of the nuclease and recognition lobes. The tracrRNA is complementary to a portion of the crRNA and binds to it in the complex; in many gene editing tools these two RNAs are combined into a single guide RNA. Two domains in the nuclease lobe each make a single cut, one in target strand and the other in the non-target strand, resulting in the target DNA being cleaved with blunt ends.

Type III Complex, a beige elongated arch of overlapping oval, containing crRNA: a thick line, a green arch with a short black tail. Csm6 is separate, a light purple oval with indents at the center of each long edge & 1 at the lower rounded edge. — **Type III systems**

have complexes contain multiple protein subunits and a crRNA. The complex forms a worm-like structure, with two curved lines of protein subunits generally arranged into two curved lines, including a backbone of subunit repeats. Type III systems have multiple cleavage activities. In the complex itself, each backbone repeat makes a cut in the target RNA, and another subunit has non-specific DNA cleavage activity. That subunit also produces secondary messengers that activate Csm6-family proteins with RNA cleavage activity.

Surveillance and targeting

Though the binding of a complementary sequence to a crRNA is a core commonality across CRISPR systems, many of the steps in CRISPR system surveillance and targeting involve interacting other features of the target nucleic acid, including two types sequences beyond the protospacer: the Protospacer Adjacent Motif (PAM) and Protospacer Flanking Sequence (PFS).

Systems that target double-stranded DNA face a major challenge in the early steps of surveillance: dsDNA is abundant in cells and must be unwound in order for the complex to test for crRNA complementarity. Simply reading every stretch of dsDNA encountered would be far too inefficient to identify the rare foreign DNA target. Instead, the CRISPR complexes of both Type I and Type II systems first scan dsDNA for a specific protospacer adjacent motif, or PAM, and then only bind to and unwind the dsDNA at those locations. Therefore, the targets are limited to regions near a PAM.

In contrast, these first steps are more simple for Type III system complexes that bind single-stranded RNA: ssRNA is uncommon in cells other than in nascent RNA transcripts, and the bases are already available for binding, no unwinding required. Therefore, simply binding any ssRNA strand encountered is a feasible way to read a large enough portion of the possible targets to find the rare invaders. Type III systems also recognize a region next to the protospacer – the Protospacer Flanking Sequence, or PFS. However, it affects later steps in interference: if an ssRNA matching the crRNA also matches the PFS, DNA cleavage and cOA production are blocked, though the ssRNA is still cleaved.

PAM recognition & crRNA DNA base pairing for Type I & II systems. A CRISPR complex sits on DNA with a protospacer next to a short PAM: PAM is on the left for Type I, right for Type II. Next, the complexes’ crRNA binds the protospacer but not PAM

Prevention of autoimmunity

PAM

These short sequences are targeted by the CRISPR machinery that capture foreign DNA sequences to insert into the CRISPR array, but the PAM is removed before the spacer is added to the array. Each system has a preferred PAM sequence; some are strict and others are flexible. In Type I systems the PAM is 5’ of the protospacer on the target sequence, and in Type II systems it is 3’ of the protospacer. The requirement of a PAM before beginning to unwind and checking if the crRNA sequence matches also reduces potential autoimmunity, as the spacers in the CRISPR array do not have PAMs and therefore would not be read by the CRISPR complex.

2 CRISPR complexes: Type I with a green crRNA & Type II with a red crRNA. Arrows covered by red Xs point to a CRISPR array, black stretches of repeats alternating with green and red spacers, lacking a PAM and not recognized by the complexes

PFS

The PFS is found 3’ of the protospacer in target sequences. When a target RNA is bound to a crRNA, the PFS location corresponds to the 5’ handle of the crRNA, which is derived from the CRISPR array repeat sequence. Because the PFS is from the array repeat, it is unlikely to occur in a foreign nucleic acid sequence in the same location relevant to the spacer sequence. However, an RNA transcript that matches both the crRNA spacer and handle sequences could be produced if the CRISPR array was transcribed in the incorrect direction.

Type III complex binds non-self phage RNA, with green glow and ATP converted to cOA & cleavage of phage ssDNA. Below: complex binds self RNA with a blue PFS base paired with 5’ handle of the crRNA - red glow, no cOA production or ssDNA cleavage

Interference process

Different CRISPR systems have different targets and methods of cleaving those targets. For all systems, during interference the CRISPR complex interacts with a target nucleic acid to identify sequences that match the crRNA and destroy them. But many details vary, including the type of nucleic acids targeted, how they are cleaved and impact of certain sequences outside the protospacer. For full details, check out our primer sections on Interference.

A strand of dsDNA; at the center is a short blue PAM region to the left of a protospacer, labeled green on the top strand. An arrow down post to broken DNA fragments for the PAM and region to the left, and an intact protospacer and DNA to the right

Type I systems target DNA, which is cleaved in a processive manner, resulting in fragments. The target DNA must contain both a PAM and a region complementary to crRNA for cleavage to occur.

A strand of dsDNA; at the center is a protospacer, labeled green on the top strand, with a short red PAM region to the right. An arrow down post to the DNA broken into two, in a blunt cut within the protospacer, just a few bases away from the PAM

Type II systems target DNA and both strands of target DNA are cleaved in a single, sequence-specific location resulting in a blunt cut. The target DNA must contain both a PAM and a region complementary to crRNA for cleavage to occur.

RNA with a green protospacer. Arrow points to 3 outcomes: cleavage of the phage RNA (regular cuts in protospacer), cleavage of phage ssDNA (dsDNA with bubble of unpaired strands; cuts in the top strand), & cleavage of host RNA: RNA curve with 2 cuts

Type III systems target RNA and are able to cleave DNA in a transcription-dependent manner. They are capable of three separate cleavage activities: the sequence-specific cleavage of target RNA by the repeating subunits of the complex backbone, resulting in multiple cuts six bases apart; non-specific cleavage of ssDNA, resulting in variable-sized fragments; and non-specific RNA cleavage by Csm6 family proteins is triggered by the cOA production of a complex subunit, which results in variable-sized fragments. The cleavage activities have different requirements: all require the RNA target to have a region complementary to crRNA spacer region, but the non-specific ssDNA and RNA cleavage activities are blocked if the RNA target also has a PFS that is complementary to the crRNA handle.