RESPONSE
A response to this article has been published. To view the response, click here.

Errors in Crystal structure of HINT from Helicobacter pylori

Inaccuracies in the article, Crystal structure of HINT from Helicobacter pylori by Tarique et al. [(2016) Acta Cryst. F72, 42–48 ] are presented, and a brief history of HINT nomenclature is discussed.

The article, Crystal structure of HINT from Helicobacter pylori by K. F. Tarique et al. (2016) has come to my attention as containing several factual errors in the Introduction regarding the nature and annotation of HINT enzymes. As an author cited by this paper, I feel that it is necessary to provide corrections for the benefit of interested readers.

The HINT proteins are well established as performing hydrolysis on a wide variety of phosphoramidates and acyl-nucleotides (Bieganowski et al., 2002; Chou et al., 2007; Chou & Wagner, 2007). Interest in these enzymes has renewed recently as HINT1 is responsible for the cellular activation of blockbuster antiviral phosphoramidate nucleotide prodrug sofosbuvir by cleaving the bond between the phosphorous and nitrogen (Murakami et al., 2010). Tarique et al. mischaracterize the catalysis as hydrolysis of the adenine nucleobase from the ribose and identify the substrate of the HINT proteins as phosphoramidites, an error in substrate oxidation state.

The structure of the protein now known as HINT1 (also hHint1) was first solved in 1996 at Columbia by the Hendrickson group. The bovine homolog of this protein had been identified biochemically as an inhibitor of protein kinase C (PKC) (McDonald & Walsh, 1985), giving rise to the initial annotation of protein kinase C interacting protein 1 (PKCI-1) in three associated PDB structures (1kpa , 1kpb , and 1kpc ). However, even in this first paper, there was doubt about the veracity of the PKC activity, `…other investigators have characterized a protein similar to bovine PKCI-1 from other organisms and have not been able to demonstrate inhibitory effect on PKC activity… In addition, some doubt has been cast on the physiological relevance of PKC inhibition by bovine PKCI-1' (Lima et al., 1996). A year later, an enzymatic function of nucleotide diphos­phate hydrolysis was demonstrated (associated structures 1av5 , 1kpe , and 1kpf ) (Lima et al., 1997).

That same year, the Petsko group at Brandeis also purified and crystallized rabbit HINT (associated structures 3rhn , 4rhn , 5rhn , 6rhn ), saying, `HINT is nearly identical to proteins that have been given the designation protein kinase C inhibitor-1 (PKCI-1). Bovine PKCI-1 was so named because it was present in brain cytosol fractions that inhibited a mixture of PKC isoforms (McDonald & Walsh, 1985). It has not been possible to reproduce this inhibition with HINT from rabbit heart or with purified recombinant HINT.' (Brenner et al., 1997) It was this publication that gave HINT proteins their name.

In 1999, Brenner et al. devoted an entire section of a review article to the pervasive annotation of HINT as PKCI-1, insisting that `Hint is not a PKC inhibitor' (Brenner et al., 1999). They noted that this enzyme is highly conserved, even in organisms that do not have PKC homologs, such as most bacteria. In fact, the appellation of PKCI-1 as an inhibitor of protein kinase C had been downgraded from heat-stable to heat-labile in 1991 (Fraser & Walsh, 1991), and there had been no success in replicating this activity, as mentioned above. The resurgence of this annotation is attributed, by Brenner et al., to yeast two-hybrid experiments that expressed fragments of the human HINT cDNA sequence, which matched the defunct PKCI-1 references in protein databases. One of the yeast two-hybrid screens identified PKC-β as a binding partner for HINT, but further analysis showed that this was only an artifact of using a HINT fragment, rather than the full sequence (Brenner et al., 1999). This 1999 article presented a clear history of why the PKCI-1 annotation had been used and why it is inappropriate to continue using it.

Since this time, the structural literature has been consistent in identifying analogous proteins using the HINT nomenclature (Bardaweel et al., 2010; Ozga et al., 2010; Dolot et al., 2012, 2013; Wang et al., 2012; Maize et al., 2013) until Crystal structure of HINT from Helicobacter pylori (Tarique et al., 2016), which not only reintroduces the abandoned PKCI-1 annotation, but also asserts that PKCI-1 is a distinct family member, rather than just another name for HINT1. It is unclear where the authors got their information, as the papers that they cite in the introduction clearly express either doubt or outright condemnation that HINT inhibits protein kinase C (Lima et al., 1996; Brenner et al., 1999).