Processivity

Processivity describes how many successive catalytic steps an enzyme performs on one substrate during a binding encounter before separating from it. A DNA polymerase can add successive nucleotides; an exoribonuclease can remove them. The average run length and its distribution describe a different property from the number of steps performed per second.

An original probability model makes that distinction precise. After one completed step, suppose the enzyme continues to another step with a constant probability $p$, independently at each decision, and otherwise releases the substrate. If $N$ counts all completed steps including that first one, thenThis is a geometric distribution. With $p=0.9$, the mean is 10 steps. Making each step take twice as long leaves that mean unchanged in the stipulated model. It changes speed, while changing the continuation probability changes processivity.

A structural experiment on Xrn1 connected substrate recognition to repeated cleavage. Mutations near its terminal-nucleotide recognition site changed the degradation pattern of labeled RNA, including accumulation of intermediates. The study combined structural information with denaturing-gel readouts rather than inferring enzyme behavior from the protein structure alone.^[ref]

The probability model assumes a constant continuation probability and an indefinitely long substrate. A finite chain, a structured region or changing contacts can truncate or alter the run-length distribution. Likewise, a long final product can result from several binding encounters unless the assay distinguishes continued action from release followed by rebinding.