Interplay of buried histidine protonation and protein stability in prion misfolding

Misofolding of mammalian prion proteins (PrP) is believed to be the cause of a group of rare and fatal neurodegenerative diseases. Despite intense scrutiny however, the mechanism of the misfolding reaction remains unclear. We perform nuclear Magnetic Resonance and thermodynamic stability measurements on the C-terminal domains (residues 90–231) of two PrP variants exhibiting different pH-induced susceptibilities to aggregation: the susceptible hamster prion (GHaPrP) and its less susceptible rabbit homolog (RaPrP). The pKa of histidines in these domains are determined from titration experiments, and proton-exchange rates are measured at pH 5 and pH 7. A single buried highly conserved histidine, H187/H186 in GHaPrP/RaPrP, exhibited a markedly down shifted pKa ~5 for both proteins. However, noticeably larger pH-induced shifts in exchange rates occur for GHaPrP versus RaPrP. Analysis of the data indicates that protonation of the buried histidine destabilizes both PrP variants, but produces a more drastic effect in the less stable GHaPrP. This interpretation is supported by urea denaturation experiments performed on both PrP variants at neutral and low pH, and correlates with the difference in disease susceptibility of the two species, as expected from the documented linkage between destabilization of the folded state and formation of misfolded and aggregated species.


Computing the ΔG op values of the 'pure' protonated form of hamster PrP and the protonated and neutral forms of rabbit PrP.
We start from Eq. (7) of the main text (Methods section) that describes the experimentally derived per--residue opening free energy in the multi--state model: We then recast this equation for the case where the PrP system samples only 2 sub-states corresponding to those where His 187/186 is either neutral or protonated respectively and the per--residue opening free energies are derived from experiments performed at pH7 and pH5 respectively: !" ] (6) ΔGj op (pH) are the experimentally derived opening free energies per residue at the set pH and p1 and p2 are the populations of species with neutral His residues at pH7 and pH5 respectively. ΔGj,N op and ΔGj,P op are the opening free energies of the 'pure' (de--mixed) species that are typically computed in protein simulations.
The populations p1 (and p2 ) of neutral His species at a given pH denoted here as [His] can be obtained from the chemical equilibrium equation and the measured pKa values: = ! !! !" !" ! !!" (7) Introducing the values of p1 and p2 derived from Eq. (7) into Eq.
(2) then allows to derive the values of the per--residue opening free energies of the 'pure' species: For the solution of Eqs (8) and (9) to exist the values of the expressions whose logarithm is computed must be positive. This in turn imposed the following restrictions on the value of the difference ΔG(pH7) -ΔG (pH5): For rabbit PrP, Eqs (8) and (9)  When the proton exchange occurs only through the protonated sub--state as for hamster PrP one of the inequalities in Eq. (10) becomes an equality. Formally this can be expressed as: Using Eq. (11), Eq. (9) and populations from Eq. (7) computed with pKa=4.9, we then derive the ΔG op p values for the hamster protein (Fig. 4a of the main text):

Deriving the values for the unfolding free energies of the neutral and protonated forms of GHaPrP and RaPrP, from urea denaturation experiments carried out at pH4 and pH7.
Urea denaturation curves were measured for GHaPrP and RaPrP , at pH7 and pH4 respectively, following the experimental procedures described in Methods (main text). The raw data are listed in Table S9.
Non--linear least squares fits were used to derive the unfolding free energy ΔG of the neutral and protonated forms of the PrP proteins. To this end it was considered that the GHaPrP and RaPrP, solutions at each pH represent a mixture of 2 forms: a neutral form PrPN , and a protonated form PrPP corresponding to the forms of PrP in which buried H187/186 is in its neutral and protonated state, respectively. We assume that each of these forms follows a two--state denaturation reaction.
Since the pKa of His 187/186 depends on its environment in the protein, the relative population of the neutral and protonated forms of the His residue, and hence of the PrP protein, change as a function of pH, and of the concentration of denaturant (urea), as the unfolded form of the protein accumulates.
Thus we have: [!] [!] = 10 (!"!!" ! ) (13) Where [N] and [P] are the concentration of the neutral and protonated forms of the His residue and PrP protein, and the pKa is that of the H186/187, residues. 20 Using Eq (13) we derive the free energy difference between the two protein forms in the folded and unfolded states, respectively: The right hand sides of Eq. (14)  It is convenient to express the free energies in terms of the difference relative to the free energy of the neutral folded state: Where the superscript x designates the protein state (folded, f, or unfolded, u), and the subscript y designates the protein form (neutral, N, or protonated P).
The thermodynamic processes linking the free energies of the different states and forms are represented by the thermodynamic cycle of Fig. S9 Taking into account the pH dependence of the free energy values (Eq. 14) and their dependence on the urea concentration denoted as C, where the ! ! , and ! values represent the slopes of this dependence) we have the following relationships: Where A(C,pH) and A(0,pH) are the measured responses as a function of urea concentration C (and in absence of urea) and pH, using the following additional 2 assumptions and settings: 1--The measured response (ellipticity) for the different states and forms varies linearly with urea concentration: Where the meaning of x and y is the same as in Eq. (15), C is the urea concentration and ! ! is the slope of the measured ellipticity as a function of urea concentration.
2--The protonated and neutral forms of the protein give rise to the same response when in the native, or the unfolded state, respectively.
3-- The pKa of the neutral form was set to the experimental values of 5.1 for RaPrP and 4.9 for GHaPrP.
On the basis of the above equations and assumptions the values of the following 7 parameters were derived by least squares fit of the expression in Eq. (18) (detailed in Eq. 16) to the experimental data for the two PrP proteins measured at pH 4 and pH7, respectively: The resulting fit is displayed in Fig. 5 of the main text, and the values of the various parameters are listed in Table S10 and Table 2 of the main text.

Estimating the ΔΔG op j values as a function of pH, and His187 pKa:
We use the well--known definitions of the pH and pKa, we estimate the ΔΔG op j values as a function of pH solely on the basis of the measured His187 pKa values as follows: respectively, we derive the fraction of the population with protonated His residues as a function of pH: Assuming that exchange occurs only through the His + sub--state ( Fig. 3a--b) we have: The ratio of the populations with [His + ], at pH 5 and pH 7, can then be obtained from Eq. (23), and the corresponding free energy difference can be directly calculated from Eq. (7) of the main text, as follows:

Possible role of acidic residues protonation
As stated in the text we did not examine the role of acidic residues (Glu and Asp).
Inspection of the rabbit and hamster PrP solution structures indicates that none of the 13 acidic residues were sufficiently shielded from solvent in the NMR conformational ensembles to feature a shifted pKa. This suggests in turn that protonation of these residues would not destabilize the monomeric PrP fold. The role of the other 4 His residues of the considered PrP domain (residue 90--231) was not considered for the same reason.
A recent study (Singh & Udgaonkar, 2016) finds indeed, that the pH--induced misfolding transition of H186F moPrP (a stabilizing mutation, of the single buried His residue in PrP) has an apparent pKa of 3.8. This pKa value is virtually identical to the pKa of solvated acidic residues, confirming our initial conclusion that the pKa of these residues is not shifted in the folded protein. Yet, rather surprisingly, the same study also shows that two single mutants of mouse PrP D177N and D201N, which mimic the neutral form of the Asp side chains, are essentially misfolded over a wide pH range (pH 2--6). The large apparent effect of the D177N and D201N mutation on PrP misfolding is therefore intriguing. High salt conditions are known to foster PrP oligomerization in denaturation experiments as mentioned in the main text, likely by screening net charges. The D-->N replacements may act similarly, and their effect may hence stabilize PrP oligomers rather than destabilize the monomer.
In any case, the observed behavior of the asp mutants does not apply to the wt Asp residues. Their protonated from will be significantly populated only at low pH (<5), but remains too low (≤0.1%, according to Eq. (23)) to contribute to the unfolding reaction of monomeric PrP at neutral pH. Asp protonation should also marginally affect the stability of monomeric PrP, measured at low pH, and instead significantly stabilize the oligomeric form, thereby depleting the monomeric form and driving the aggregation reaction.