PAM sequences and why the genome context matters

The spacer matches the genome on paper, the oligo arrives, and the locus comes back unedited. Often the missing piece was never the spacer. PAM sequences and why the genome context matters is the difference between a motif the protein will accept and a string of bases that merely looks complementary. The binding step is sketched in how CRISPR-Cas9 editing works in research. Proving what happened at the locus is in checking whether a genome edit worked. Short definitions sit beside this page in a glossary of genome-editing terms.

The motif is not the spacer

A protospacer adjacent motif is DNA. The protein reads it. The guide RNA's spacer is a different polymer and a different sequence. For the enzymes people teach first, the spacer is aimed at the protospacer next to the motif, and the motif itself is not base-paired by that spacer. Draw both, or you will order one string that tries to do two jobs.

Orientation matters as much as the letters. SpCas9's commonly taught NGG lies immediately on the 3-prime side of the protospacer, on the non-target strand as that enzyme is usually diagrammed. Cas12a enzymes are usually taught with a motif on the other side of the protospacer, and they cut in a different pattern. A cartoon copied from Cas9 and relabelled Cas12a points the guide the wrong way.

Design tools such as CRISPOR ask which enzyme you mean before they search. That question is the whole point of this page. Answer it with the protein in the freezer, not with the protein in the paper you read last.

Three motifs that teaching labs repeat

These are widely taught consensus motifs, not a promise that every site with the letters will cut, and not a complete catalogue of every orthologue.

Streptococcus pyogenes Cas9, SpCas9, is taught as NGG. N is any base. Guides designed for this enzyme are the default assumption in a large share of mammalian papers. A site that ends in NAG is sometimes weakly recognised by this protein in some contexts. Do not promote that footnote into a design rule unless the enzyme supplier and your own data say so. If you need NGG, require NGG.

Staphylococcus aureus Cas9, SaCas9, is taught as NNGRRT, where R is A or G. It is a smaller protein, often chosen when a delivery cargo has a size limit. An SpCas9 spacer next to NGG is not a gift to SaCas9. The protospacer length people use can differ as well. Treat length and motif as a pair that belongs to the enzyme.

Cas12a, also called Cpf1 in older papers, is taught with a T-rich motif often written TTTV, where V is A, C, or G, placed 5-prime of the protospacer in the usual diagram. It is not a Cas9 with a different PAM sticker. The protein, the guide scaffold, and the cut geometry differ. Ordering a Cas9 scaffold RNA for a Cas12a protein produces a tube that will not assemble into the complex you drew.

Near-PAMless proteins are another class

Engineered Cas9 variants that relax or nearly drop the motif requirement are published as new proteins, not as a mood of wild-type SpCas9. Their acceptable sites, their off-target tendencies, and the guides that suit them are part of that engineering. If your tube says a variant name, design with that name. If your tube says SpCas9, do not apply a near-PAMless paper's site list and hope the wild-type protein agrees.

Mixing classes shows up later as a blank validation. The sequencing is fine. The site was never a substrate. Record the protein name in the same notebook line as the spacer, using the habit in recording guides so an experiment can be repeated.

Chromatin can hide a perfect motif

A motif and a spacer can be correct on naked DNA and silent in a nucleus. Nucleosomes and compact chromatin change access. Design scores that look only at sequence do not open chromatin. They rank sequences. If a locus is known in your cell type to be silent and closed, a failed cut is a plausible outcome, and a second guide in the same closed region may fail too. A known-cutting guide at an accessible control locus separates "this delivery works" from "this chromatin does not". That control logic is in controls for a CRISPR experiment.

Do not invent an accessibility number you did not measure. Either cite an assay your laboratory ran, or say you did not measure access and the failure is still compatible with closure.

A SNP can delete the motif

Reference browsers, including the UCSC Genome Browser and Ensembl, show an assembly. Your line may differ by a single base that turns NGG into something SpCas9 will not use. The spacer can still match perfectly. The protein never starts. The same SNP can create a motif the reference lacked, which matters when you are explaining an unexpected cut.

Sequence the parental window, or inspect a variant file that truly came from this line, before a long redesign. Bacterial strains diverge from the accession you downloaded in the same way. "The E. coli genome" is not a strain.

Off-target candidate lists depend on the motif as well. A similar spacer with no acceptable motif is not a substrate for that enzyme. A SNP that creates a motif next to a near-match can create an off-target the reference search missed. Parental sequence is part of off-target hygiene, not a luxury. On-target sequencing still will not see those other sites. Ask for them separately if they are part of the claim, using the file list in what sequence data to request after an edit.

When the comparison itself goes wrong

The common failure is importing a guide sequence from a methods section and changing the protein later because a smaller plasmid was in the freezer. The spacer is now orphaned. Another failure is adding the PAM letters onto the oligo "so the sequence looks like the browser". The oligo then competes with the wrong target. A third is treating a near-PAMless paper as permission to point wild-type enzyme at every NNN site in an exon. Most of those sites are not substrates, and the ones that are weak are not a design.

If a site fails, branch. Confirm the enzyme name on the tube. Confirm the motif on the parent's sequence, not only on the assembly. Confirm a control guide cuts in the same delivery. Only then redesign. Reordering the same spacer with a new purification will not create a missing G.

Practical notes for the order

Write the enzyme class and the motif you required into the quote request, and attach the spacer without the motif bases. Nuclease and guide-RNA classes are in the molecular biology catalogue. If parental sequencing is how you will confirm the motif is really in the cells, frame that as a discussion through the CRISPR validation sequencing enquiry reference. It is an independent method reference. Ask whether a quotation is possible. The molecular biology pathway is the research context.

Heat and shipping do not change a motif, but a swapped label between an SpCas9 tube and a Cas12a tube does. Check the name on arrival, before resuspension. In a shared freezer, two proteins with similar caps are an easy mix-up. The genome context you carefully checked will not save a tube that contains the other enzyme.

This comparison is a research design note. It is not a clinical genotyping claim and not a biosafety approval for using any of these proteins.