The Pedagogical Value of Linderstrøm-Lang's Protein Ontology

Abstract

In his 1951 Lane Medical Lectures, Kaj Ulrich Linderstrøm-Lang (LL) organized protein structure into a three-tiered hierarchy, with the amino acid sequence comprising the primary structure, local-in-sequence interactions stabilizing the secondary structure, and long-range interactions stabilizing the tertiary structure. Here we call this ‘the LL ontology.’ Extensions of this ontology to quaternary, quinary, and even senary structural levels have been proposed in the literature, with the quaternary level being nearly universally adopted in undergraduate curriculums. We argue that, since the microscopic nature of protein-protein interactions are identical for a ternary versus a quaternary interface, the inclusion of quaternary structure is not edifying in the context of an undergraduate curriculum. Protein-protein interactions can be treated as general phenomena that are mediated by protein conformational states without an essentialist invocation of quaternary structure. The LL ontology also carries with it the implication that secondary structure is always ontologically prior to tertiary structure, or that tertiary structure is always prior to quaternary structure. These generalizations are disproven by the existence of chameleon sequences (primary structural elements that occur in different secondary structural contexts) and by the existence of intrinsically disordered proteins (proteins lacking a stable three-dimensional fold in the absence of a binding partner.) We argue that, in terms of the pedagogy of protein structure, students are best served by studying the first three tiers of the LL ontology to provide a still-useful descriptive language for all proteins. Folding should not be presented as a simple ascension of this hierarchy unless a framework mechanism is being explicitly discussed. Ultimately we favor the energy landscape as a universal descriptor for the interplay between protein folding and function.