Restriction enzymes cut double-stranded DNA* at specific locations based the pattern of bases found at those locations. These enzymes predictably cut both strands because the sequences they recognize are palindromic. That is the recognition sequences are short string of identical bases on both DNA strands.
Palindromic sequences are similar to language palindromes, but follow a distinct set of rules. Any string of bases can be made into a palindromic sequence by following these rules.
In English, the term palindrome refers to a string of letters that have the same meaning written in both directions some classic English palindrome are kayak, civic, noon, and racecar.
As a set of paired sequences (one on each of the strands of a double strand of DNA), the palindromes recognized by restriction enzymes follow a slightly different set of rules. They are probably more properly referred to as palindromic sequences to distinguish them from language palindromes. Palindromic sequences are a short run of bases (typically 3 to 5 in length), follow by their complementary bases in reverse order. For example the recognition sequence for BamHI is GGATCC.
Note the first three bases GGA are followed by the complement of those three bases in reverse order: TCC. The complement to the whole six base strand is CCTAGG, read backwards (as it would be when reading from 5’ to 3’ on the complementary strand) is GGATCC, an exact match for the original strand.
This pattern makes it possible to reconstruct a palindromic sequence from one-half of one strand. For example, a six-base recognition sequence (e.g. TAGCTA) can be reconstructed from just knowing the first three bases on one strand:
- Starting with the original sequence - TAG
- Calculate the reverse complement of the sequence - ATC
- Reverse the order of the reverse complement (CTA) and add it to the end of the forward strand - TAGCTA
- Calculate the reverse complement of the whole forward strand to finish the reverse strand: ATCGAT
Final result: TAGCTAATCGAT
Having short stretches of DNA that read the same on both strands of double-stranded DNA allow restriction enzymes to cut both strands in the same place.
Note having palindromic sequences along short the short stretches over which restriction enzymes function is not common.
To prove this take any random stretch of DNA such as AGTCCGATCCGT
find its reverse complement: TCAGGCTAGGCA
flip it to the proper 5’-3’ orientation for the complementary strand: ACGGATCGCCACT
and you don’t get the same sequence back ACGGATCGCCACT ≠ AGTCCGATCCGT
- Problem Sets