A strategy for optimizing the monodispersity of fusion proteins: application to purification of recombinant HPV E6 oncoprotein

Yves Nomine1, Tutik Ristriani2, Cecile Laurent2, Jeannew protein studied. Moreover, in many cases, the protein or Francois Lefevre1, Etienne Weiss2 and Gilles Trave1,3 protein domain studied does not display any known activity at all. A classical alternative to biological assays consists in 1Laboratoire de RMN, UPR 9003 du CNRS and 2Laboratoire evaluating the percentage of soluble protein produced. Provided d’Immunotechnologie, UPRES 1329, Ecole Superieure de Biotechnologie de that the protein of interest is cloned as a genuine unfused Strasbourg, 67400 Illkirch, France 3To whom correspondence should be addressed. sequence, solubility may be regarded as a universal check for E-mail: trave@esbs.u-strasbg.fr proper folding, since misfolded proteins generally precipitate. However, recombinant proteins or domains are very often Recombinant production of HPV oncoprotein E6 is notorifused to carrier proteins such as maltose-binding protein ously difficult. The unfused sequence is produced in inclu(MBP), glutathione-S-transferase (GST) or thioredoxin (Trx) sion bodies. By contrast, fusions of E6 to the C-terminus (LaVallie and McCoy, 1995). Fusion proteins have become of carrier proteins such as maltose-binding protein or crucial tools for recombinant protein production. They are gluthatione-S-transferase are produced soluble. However, especially useful for affinity purification of recombinant proit has not yet been possible to purify E6 protein from such teins and detection of their specific interactions with biological fusion constructs. Here, we show that this was due to the ligands, including other proteins or peptides. Many workers biophysical heterogeneity of the fusion preparations. We in the field tend to extend to fusion proteins the paradigm that find that soluble MBP-E6 preparations contain two subpopsolubility is a guarantee of proper folding and biological ulations. A major fraction is aggregated and contains exclusively misfolded E6 moieties (‘soluble inclusion activity. However, several reports have contradicted this view, bodies’). A minor fraction is monodisperse and contains by showing that soluble preparations of a fusion protein may the properly folded E6 moieties. Using monodispersity display low biological activity or no activity at all, as compared as a screening criterion, we optimized the expression with the native protein from natural sources (Louis et al., conditions, the purification process and the sequence of 1991; Saavedra-Alanis et al., 1994; Lorenzo et al., 1997; E6, finally obtaining stable monodisperse MBP-E6 preparaSachdev and Chirgwin, 1998, 1999). These findings stressed tions. In contrast to aggregated MBP-E6, these preparations the need for a universal criterion, other than solubility, which yielded fully soluble E6 after proteolytic removal of MBP. would allow one to evaluate the quality of fusion proteins. Once purified, these E6 proteins are stable, folded and E6 is an oncoprotein produced by the ‘high-risk’ human biologically active. The first biophysical measurements on papillomaviruses (HPVs) responsible for 95% of cervical pure E6 were performed. This work shows that solubility cancers (Scheffner et al., 1990). More than 100 E6-encoding is not a sufficient criterion to check that the passenger DNA sequences have been published since 1985 (Chan et al., protein in a fusion construct is properly folded and active. 1995). Despite this fact, there is little information about the By contrast, monodispersity appears as a better quality molecular properties of the protein. The 3D structure is not criterion. The monodispersity-based strategy presented known and even easy biophysical measurements concerning here constitutes a general method to prepare fusion proteins the optical properties of the protein (UV absorption, UV with optimized folding and biological activity. fluorescence, circular dichroism) are missing. This lack of