Effect of hydrophobic mutations in the H2-H3 subdomain of prion protein on stability and conversion in vitro and in vivo

Abstract

Prion diseases are fatal neurodegenerative diseases, which can be acquired, sporadic or genetic, the latter being linked to mutations in the gene encoding prion protein. We have recently described the importance of subdomain separation in the conversion of prion protein (PrP). The goal of the present study was to investigate the effect of increasing the hydrophobic interactions within the H2-H3 subdomain on PrP conversion. Three hydrophobic mutations were introduced into PrP. The mutation V209I associated with human prion disease did not alter protein stability or in vitro fibrillization propensity of PrP. The designed mutations V175I and T187I on the other hand increased protein thermal stability. V175I mutant fibrillized faster than wild-type PrP. Conversion delay of T187I was slightly longer, but fluorescence intensity of amyloid specific dye thioflavin T was significantly higher. Surprisingly, cells expressing V209I variant exhibited inefficient proteinase K resistant PrP formation upon infection with 22L strain, which is in contrast to cell lines expressing wild-type, V175I and T187I mPrPs. In agreement with increased ThT fluorescence at the plateau T187I expressing cell lines accumulated an increased amount of the proteinase K-resistant prion protein. We showed that T187I induces formation of thin fibrils, which are absent from other samples. We propose that larger solvent accessibility of I187 in comparison to wild-type and other mutants may interfere with lateral annealing of filaments and may be the underlying reason for increased conversion efficiency.

Figure 1. V175, T187 and V209 in murine prion protein can be substituted for isoleucine without large influence on protein secondary structure.

A) Schematic representation of amino-acid residues selected for substitution with more hydrophobic isoleucine. V175 and T187 reside in H2 and V209 lies in H3. B) CD spectra of wild-type (▪), V175I (⋄), T187I (▴), and V209I (▿) reveal a high content of α-secondary structure as expected for mPrP. C) Thermal denaturation curves of wild-type and mutants (symbols used as above).

Figure 2. In vitro conversion of isoleucine mutants.

A) In vitro fibrillization of wild-type (▪), V175I (⋄), T187I (▴), and V209I (▿) mPrPs followed by fluorescence intensity of thioflavin T. A representative of four experiments is shown. B) Presence of fibrils was confirmed by TEM. Bar represents 250 nm. C) Final thioflavin T fluorescence intensity of T187I is significantly higher than in the samples of other mPrPs (p<0.0005). The differences among wild-type, V175I and V209I are not significant.

Figure 3. AFM reveals the presence of thin fibrils in addition to mature fibrils in conversion reactions of T187I.

Conversion reactions of wild-type (WT), V175I, V209I (upper row) and T187I (middle, bottom row) were observed under the atomic force microscope. In all samples fibrils with approximate height around 7 nm were observed. Cross-section of such fibril from sample T187I is shown (bottom row right, full line). In the T187I sample in addition to such fibrils, fibrils with less than 1 nm in height were also present (middle row). Image 187B shows a close-up of selected part in image 187A. Cross-section of thin fibril is shown in the bottom row (bottom row right, broken line). Bar represents 500 nm. Arrows in images 187B and 187C indicate positions where cross-sections were taken. Images have not been corrected for the width and shape of the AFM tip.

Figure 4. Cell line HpL3-4 expressing T187I produces more PrP^res than cell lines expressing wild-type and other mutants when infected by mouse prion strain 22L.

A) Stably transduced HpL3–4 bulk populations expressing wild-type and mutant PrP proteins were exposed to mouse-adapted prion strain 22L and subsequently passaged in the absence of inoculum for 4 passages. B) Cell lysates of HpL3–4 cell populations expressing wild-type mPrP or mPrP V209I were tested for PrP^res content. While mutant V209I expressed well in HpL3–4 (non-digested by proteinase K, -PK), only weak bands of PrP^res are present in this cell population (+PK). C) Wild-type mPrP, mPrP V175I and mPrP T187I-expressing HpL3–4s were infected by 22L prions (5 parallels each). Upper part shows PrP^res (+PK), middle part non-digested cell lysate (-PK) and bottom line GAPDH loading control. D) Analysis of averaged normalized intensity of PrP^res reveals significantly higher amount of PrP^res present in cell lysates of HpL3–4s expressing T187I (p<0.0001).