Cofactor-mediated protein promiscuity

To the Editor:

The review article by Nobeli et al.¹ in the February issue creates a conceptual framework for protein binding and functional promiscuity and delineates its significance for biotech. As the interpretation of protein promiscuity in the literature is itself promiscuous, this timely review comes to clarify important features of this phenomenon. The authors have proposed a comprehensive classification of the mechanisms responsible for protein recognition and functional promiscuity. We believe, however, that an important additional mechanism for protein promiscuity could be added to this classification.

Heme (ferroprotoporphyrin IX) is an ubiquitous macrocyclic compound present in all kingdoms of life, predominantly in a protein-bound form. The proteins that use heme as a cofactor are referred to as hemoproteins; these have pleiotropic functions not limited to redox chemistry². Heme is able to bind with a relatively high affinity to a great variety of proteins—from 'typical' hemoproteins, such as hemoglobin, myoglobin, peroxidases and cytochromes, to 'untypical' hemoproteins, such as albumin, prion protein³, amyloid beta (Aβ) peptide⁴, myosin⁵, growth hormone⁶, orphan nuclear receptors⁷ and many others. Indeed, it is a challenge to define a protein that is not able to show any heme-binding activity. The remarkable binding promiscuity of heme is broader than the promiscuity of any other organic compound of a similar size. What is most intriguing, however, is that the promiscuity of heme is not limited only to the molecule itself, but it could also spread to the proteins to which heme binds. This 'contagious promiscuity' is obvious in the case of hemoglobin—a protein that is able, thanks to its heme groups, to interact with diverse gaseous molecules—oxygen, carbon monoxide, nitric oxide, hydrogen sulfide, among others. In addition to recognition promiscuity, heme endows hemoglobin with a functional promiscuity, explaining its well-known activities in gas transport, in the reduction of the oxygen molecule as well as its peroxidase activity. An even more striking example is the interaction of heme with the Aβ peptide, which not only prevents the formation of amyloid plaques by conformational rearrangement of Aβ but also endows the same peptide with peroxidase activity. We have recently, documented yet another example of heme-mediated protein promiscuity concerning protein-protein interactions⁸. Approximately 20% of human antibodies bind heme as an interface cofactor, and the intrinsic binding promiscuity of this bound heme confers substantial antigen-binding promiscuity that could well be important for antibodies' role in the defense against pathogens.

The prominent promiscuity of heme may be useful in biotech. Looking carefully at the properties of heme as a prototype promiscuous molecule could help in understanding what determines the binding promiscuity of different compounds and could thus selectively guide drug designers in creating less promiscuous drugs—a challenge that was discussed in the Nobeli et al. review.