A Deeper Insight in Metal Binding to the hCtr1 N-terminus Fragment: Affinity, Speciation and Binding Mode of Binuclear Cu2+ and Mononuclear Ag+ Complex Species

Ctr1 regulates copper uptake and its intracellular distribution. The first 14 amino acid sequence of the Ctr1 ectodomain Ctr1(1-14) encompasses the characteristic Amino Terminal Cu2+ and Ni2+ binding motif (ATCUN) as well as the bis-His binding motif (His5 and His6). We report a combined thermodynamic and spectroscopic (UV-vis, CD, EPR) study dealing with the formation of Cu2+ homobinuclear complexes with Ctr1(1-14), the percentage of which is not negligible even in the presence of a small Cu2+ excess and clearly prevails at a M/L ratio of 1.9. Ascorbate fails to reduce Cu2+ when bound to the ATCUN motif, while it reduces Cu2+ when bound to the His5-His6 motif involved in the formation of binuclear species. The histidine diade characterizes the second binding site and is thought to be responsible for ascorbate oxidation. Binding constants and speciation of Ag+ complexes with Ctr1(1-14), which are assumed to mimic Cu+ interaction with N-terminus of Ctr1(1-14), were also determined. A preliminary immunoblot assay evidences that the anti-Ctr1 extracellular antibody recognizes Ctr1(1-14) in a different way from the longer Ctr1(1-25) that encompasses a second His and Met rich domain.


Introduction
Copper is an essential element for life that is needed for all cellular systems of both prokaryotic and eukaryotic organisms [1] Cu + /Cu 2+ redox process can be detrimental to cells, as it may result in the production of different reactive oxygen, nitrogen and carbon species that are thought to be the cause of different disorders [2,3].
A defective absorption of the metal ion is linked to a series of pathologies, while an excessive copper accumulation may cause degenerative disorders [4,5].
Thus, it is not a surprise that copper homeostasis is strictly controlled in all eukaryotic organisms. To this end, a protein network is specifically devoted to the regulation of intracellular Cu + levels. The membrane copper transporter 1 (Ctr1), which belongs to the SLC31 family of solute carriers [6] and possesses high specificity and affinity for Cu + , is the most important protein transporting copper into the cell [7] Ctr1 in concert with Cu + -transporting ATPase pumps (ATP7A and ATP7B) [8], which are localized in the intracellular membrane and harmonize copper trafficking by regulating copper uptake, intracellular distribution and eventually its export from the cell [9]. Human Ctr1 (hCtr1) consists of 190 amino acids and shares a large homology with yeast Ctr1 (yCtr1), though of Cu 2+ or Cu + results in the coordination of up to three or six Cu 2+ or Cu + ions, respectively [33]. The log K value determined for Cu + -hCtr1  is of the same order of magnitude of that obtained for Cu + -hCtr1   [30] (14.79 vs. 14.92), while it is some six orders of magnitude larger than that of Cu + -hCtr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) [23]. Such a pronounced difference strongly suggests that the residues contained in the segment 15-46 contribute to the binding of the metal ion and account for the different stability constants found, which, in turn, may be ascribed to the involvement of His and Met motifs in the longer peptide segment (hCtr1 (1-46) ) [33]. Additionally, Cu K-edge X-ray absorption spectroscopy results support a structure containing a tetrahedrally arranged Cu + with a 2N2S set of donor atoms. Tandem ESI-MS results indicate that the coordination sites are provided by the first His-rich and the second Met-rich motifs of the segment, though different Cu + -binding arrangements may still be possible for the full-length protein that forms a trimeric cone-shaped pore [13].
Defining the arrangement around copper is of crucial importance, as it would impact on the ease of reduction of Cu 2+ ions coordinated to the N-terminus ATCUN motif. It has been shown that Cu 2+ reduction by ascorbate is faster when hCtr1 model peptides contain the bis-His motif [23] and is even significantly faster for the trimer containing three MDHSHH motifs than for the analogous monomer containing a single MDHSHH unit [34]. Interestingly, Cu 2+ ions bound to MDHS motifs and lacking the bis-His units are also less susceptible to reduction. This is in keeping with recent reports [35,36] that indicate that Cu 2+ bound to ATCUN via a 4N coordination mode is not reduced, though it is also suggested that a species where the metal ion involves the binding of an amine and an imidazole nitrogen is responsible for the reduction process [36]. Remarkably, by competition experiments with bicinchoninic acid, it has been shown that the trimer containing the MDHSHH units are much better competitors than those with the MDHS motifs, thus indicating that the bis-His sequence is mainly responsible for Cu + binding.
Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) has seven sites that can be protonated (i.e., two carboxylate side chains, three nitrogen atoms from imidazole rings, the terminal NH 2 group and the phenolate group of Tyr side chain); accordingly, the fully protonated ligand is defined as [H 7 L]. The protonation constants values determined by us (Table S1) are in good agreement with those obtained in a previous work [17]. Cu 2+ affinity for Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) was investigated potentiometrically by exploring a Cu 2+ /Ctr1 (1-14) concentration ratio ranging from 0.9 to 1.9; such a concentration range is wider than that investigated previously by others [17]. The 'best' set of stability constants is reported in Table 1 together with the values determined by Stefaniak et al. [17], while the species distribution resulting from our data at different Cu 2+ /Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) ratios is shown in Figure 1. Since hCtr1 is also found in some intracellular areas (e.g., lysosomes) where pH is acidic, we started our study from pH values as low as 3 ( Figure 1) [14,38]. The set of mononuclear species detected in the present study as well as the values of their stability constants mirror those reported by Stefaniak et al. [17], who restricted their investigation to the 0.4-0.9 Cu 2+ /Ctr1 (1-14) range, however; the coordination mode of mononuclear species has been discussed previously [17]. Noteworthy is the fact that, in addition to mononuclear species, we have also detected the formation of copper binuclear species, the percentage of which is not negligible even in quasi-equimolar concentrations of Cu 2+ and Ctr1 (1-14) (Figure 1b). Likely, this arises from the wider concentration ratio explored in our study. In order to give a better perception of their significance, we have computed the total concentration of both mono-and binuclear species in the conditions reported in the captions of Figure 1a-d. Figure 2 depicts the trend of mono-and binuclear complexes as we go from a solution with a small excess of ligand ( Figure 2b) to a solution with an excess of metal ion (Figure 2d). concentration ratio is 0.9, 1.2, 1.6 and 1.9 for (a-d), respectively.
Since hCtr1 is also found in some intracellular areas (e.g., lysosomes) where pH is acidic, we started our study from pH values as low as 3 ( Figure 1) [14,38]. The set of mononuclear species detected in the present study as well as the values of their stability constants mirror those reported by Stefaniak et al. [17], who restricted their investigation to the 0.4-0.9 Cu 2+ /Ctr1(1-14) range, however; the coordination mode of mononuclear species has been discussed previously [17]. Noteworthy is the fact that, in addition to mononuclear species, we have also detected the formation of copper binuclear species, the percentage of which is not negligible even in quasi-equimolar concentrations of Cu 2+ and Ctr1(1-14) (Figure 1b). Likely, this arises from the wider concentration ratio explored in our study. In order to give a better perception of their significance, we have computed the total concentration of both mono-and binuclear species in the conditions reported in the captions of Figure 1a-d. Figure 2 depicts the trend of mono-and binuclear complexes as we go from a solution with a small excess of ligand ( Figure 2b) to a solution with an excess of metal ion (Figure 2d).
What catches the eye immediately is that the homobinuclear species have a relevant weight (25%) at physiological pH values even when Cu 2+ and Ctr1(1-14) are employed in a 1.2/1 ratio (Figure 2b), while they prevail when the metal ion exceeds the transporter (Figure 2d). Mononuclear complex species are mainly formed with small Cu 2+ /Ctr114 ratios (Figure 1a,b), while binuclear species prevail as the Cu 2+ /Ctr114 ratio increases (Figures 1c,d  and 2c,d). Although small spectroscopic differences have been detected between the group of species that is detected in acidic conditions (CuLH2, CuLH) and the one that is formed in the alkaline region (CuL, CuLH-1, CuLH-2), all of these species have been reported to have a 4N coordination environment and to be characterized by the Cu 2+ -AT- What catches the eye immediately is that the homobinuclear species have a relevant weight (25%) at physiological pH values even when Cu 2+ and Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) are employed in a 1.2/1 ratio (Figure 2b), while they prevail when the metal ion exceeds the transporter (Figure 2d).
Mononuclear complex species are mainly formed with small Cu 2+ /Ctr1 14 ratios (Figure 1a,b), while binuclear species prevail as the Cu 2+ /Ctr1 14 ratio increases (Figures 1c,d and 2c,d). Although small spectroscopic differences have been detected between the group of species that is detected in acidic conditions (CuLH 2 , CuLH) and the one that is formed in the alkaline region (CuL, CuLH -1 , CuLH -2 ), all of these species have been reported to have a 4N coordination environment and to be characterized by the Cu 2+ -ATCUN motif. We will restrict the discussion to binuclear species (Table 2) since mononuclear species have been discussed by others [17]. Table 2. Spectroscopic parameters obtained from the titration of the Cu 2+ -Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) . The potentiometric analysis evidences the formation of a binuclear species, formally [Cu 2 L], which would be better represented as [Cu 2 (HL)H -1 ], since tyrosine is not deprotonated at these pH values. The deconvoluted UV-vis bands suggest that the formation of two different Cu 2+ binding sites have different coordination environments. The UV-vis band is broad due to the contribution of two distinct absorption maxima quite away from each other. The first maximum is centered at 523 nm and is diagnostic of an ATCUN binding characterized by a {N NH2 , 2N -, N Im } coordination environment localized in the MDH domain of the Ctr1 (1- 14) . The second maximum is centered at 704 nm and a value typically observed for a 1N1O coordination mode in which, beside an imidazole nitrogen of a histidine residue (His5 or His6), a carbonyl or carboxyl oxygen is bound to the metal ion [39]. The presence of two distinct binding sites for Cu 2+ is corroborated by the spectroscopic values obtained through the deconvolution of CD spectra. The negative band at 272 nm provides evidence for the concurrent contributions of the N Im → Cu 2+ and N NH2 → Cu 2+ CT bands. The presence of an ATCUN coordination motif and a second binding site for Cu 2+ causes an increase of the ∆ε value of the N Im → Cu 2+ CT band as well as an inversion of the sign of the dichroic signal. The increase of the ∆ε value of the Namide → Cu 2+ CT band at 314 nm supports the coordination of a second nitrogen atom of a deprotonated amide in the ATCUN site, the typical spectroscopic parameters of which [40] are summarized in Table 3. The species resulting from the deprotonation of [Cu 2 L], formally [Cu 2 LH -1 ] but basically [Cu 2 (HL)H -2 ], reaches the formation maximum at pH 6.2; its pK value (pK 21-1 = log β 21-1 − log β 210 = 5.91) suggests the involvement of a further imidazole nitrogen of a histidine residue (His5 or His6) in the second binding site. This pK value is very close to that found by La Mendola et al. [41] for Cu 2+ bound to two adjacent histidine residues in a peptide with a sequence similar to that of Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) .

Species
At physiological pH values (Figure 1d), [Cu 2 LH -2 ] (equivalent to [Cu 2 (HL)H -3 ]) is the main species. Its pK value (pK 21-2 = log β 21-2 − log β 21-1 = 6.55) indicates the deprotonation of a further amide nitrogen atom [42]. The spectroscopic data strongly suggest that such a deprotonation increases the copper affinity of the second binding site. The PeakFit analysis reveals two bands with λ max at 522 nm (ACTUN site) and 590 nm (second binding site); likely, the second Cu 2+ binding site has a 3N1O (2N Im , N -, O COO -) coordination environment in agreement with the literature data for similar Cu 2+ peptide sequences [41,42]. In addition, an increase in the ∆ε value of the Namide → Cu 2+ CT band (314 nm) of [Cu 2 LH -2 ] further supports the Cu 2+ binding mode suggested above, which impacts on the overall conformation of the system. In [Cu 2 LH -2 ] the [Cu 2 L] CD band at 272 nm splits in a N Im → Cu 2+ CT band (λ max = 260 nm) with a positive ∆ε value and a negative N NH2 → Cu 2+ CT band at 281 nm.
A further pH increase causes the formation of [Cu 2 LH -3 ] (basically [Cu 2 (HL)H -4 ]); its pK value (pK 21-3 = logβ 21-3 − logβ 21-2 = 7.80) is consistent with a 4N {2N Im , 2N -} [41] coordination mode resulting from the involvement of an additional deprotonated amide nitrogen. For [Cu 2 LH -3 ], the overall UV-vis band at 545 nm of the deconvoluted (Hyper-Quad) UV-vis spectrum has a symmetrical shape. The PeakFit analysis indicates a first band with λ max at 522 nm for the ACTUN motif and a second band with λ max at 560 nm for the 4N {2N Im , 2N -} coordination environment of the second binding site (Table 2). Both a further increase in ∆ε value and a slight red shift of the Namide → Cu 2+ CT band centered around 320 nm, obtained by deconvoluting the [Cu 2 LH -3 ] CD spectrum, suggest a Cu 2+ new coordination environment for the second binding site.
Between pH 9 and 11, the OH group of tyrosine deprotonates results in the formation of [Cu 2 LH -4 ] and [Cu 2 LH -5 ], in which both copper(II) ions are coordinated to the Ctr1 (1- 14) , each through four nitrogen atoms. The pK value for [Cu 2 LH -4 ] (pK 21-4 = log β 21-4 − log β 21-3 = 8.72) is consistent with the coordination of a further amide nitrogen with a 4N {N Im , 3N -} coordination environment. The deconvoluted UV-vis spectrum for [Cu 2 LH -4 ], while maintaining the same shape found for the previous complex species, is characterized by both a further slight blue shift and an increase in ε max . The overall UV-vis band is centered at 535 nm; the PeakFit analysis evidences both the band ascribed to the ATCUN motif (λ max at 522 nm) and the blue-shifted band related to second metal binding site (λ max at 540 nm). No further changes in the ∆ε value for the N NH2 → Cu 2+ CT band at 284 are observed in the deconvoluted CD spectrum for [Cu 2 LH -4 ], whereas the intensity of the Namide → Cu 2+ CT band centered around 320 nm increases. The pK value of [Cu 2 LH -5 ] (pK 21-5 = log β 21-5 − log β 21-4 = 10.72) is consistent with the deprotonation of the phenolic group of the tyrosine residue, which takes no part in the coordination to the Cu 2+ . The similarity between the [Cu 2 LH -5 ] and [Cu 2 LH -4 ] spectroscopical parameters calculated strongly supports the idea that the deprotonation of [Cu 2 LH -4 ] involves no change in the metal centers coordination environment.

Cu 2+ Complexes with Ctr1 (1-14) Oxidize Ascorbic Acid When Binuclear Species Form
The oxidation kinetics of ascorbate has been extensively investigated employing both free Cu 2+ and Cu 2+ complexes [43]; these data show that the different Cu 2+ binding environments drive the kinetics and mechanisms of the ascorbate oxidation. Some studies report that different Cu 2+ binding domains in the same protein or polypeptides display different ascorbate oxidation abilities [40,44].
In order to compare the oxidation ability of the Cu 2+ bound to the Ctr1 (1-14) ATCUN and secondary binding site, we measured the changes in the rates and levels of ascorbate consumption with changes of Ctr1 (1-14) /Cu 2+ ratios. Figure 3a shows that the addition of one equivalent of Cu 2+ to the buffer causes the largest and fastest ascorbate consumption (red line). The addition of one equivalent of Ctr1 (1-14) to a buffered Cu 2+ solution inhibits the Cu 2+ -mediated ascorbic acid oxidation (blue line) due to the formation of copper(II) ion complex with the MDH sequence of the N-terminus (ATCUN motif). Increasing the metal to peptide ratio, that is, favoring binuclear species formation (Figure 4), causes an increase in the kinetics of ascorbate consumption (magenta line) that, however, is not comparable to that measured for a buffered solution containing Cu 2+ only (red line). Noteworthy is the fact that, if we push the metal/ligand ratio further up, we detect a fast ascorbate consumption (green line) similar to that found when one equivalent of only Cu 2+ is added to the buffer (red line). This indicates that the excess Cu 2+ that is no longer complexed by Ctr1 (1-14) causes a reduction of ascorbate comparable to that detected for the solution containing no ligand. Figure 3b depicts the UV-vis spectra for the Cu 2+ -Ctr1 (1- 14) system obtained with a 1:1 (blue line) and 2:1 (magenta line) Cu 2+ /Ctr1 (1-14) ratio in the same experimental conditions employed for the Cu 2+ ascorbate-dependent reduction. This figure clearly shows that λ max shifts towards greater wavelenghts as binuclear species prevail ( Figure 4). The difference spectrum (brown line), obtained by subtracting the spectral contribution characteristic of the ATCUN binding site (blue line) from the spectrum obtained with a greater Cu 2+ /Ctr1 (1- 14) ratio (magenta line), gives an idea of the characteristics of the Cu 2+ second binding site, further indicating that such a binding site is weaker than the ATCUN motif. Ctr1(1-14) causes a reduction of ascorbate comparable to that detected for the solution containing no ligand.   Figure 3b depicts the UV-vis spectra for the Cu 2+ -Ctr1(1-14) system obtained with a 1:1 (blue line) and 2:1 (magenta line) Cu 2+ /Ctr1(1-14) ratio in the same experimental conditions employed for the Cu 2+ ascorbate-dependent reduction. This figure clearly shows that λmax shifts towards greater wavelenghts as binuclear species prevail (Figure 4). The difference spectrum (brown line), obtained by subtracting the spectral contribution characteristic of the ATCUN binding site (blue line) from the spectrum obtained with a greater Cu 2+ /Ctr1(1- 14) ratio (magenta line), gives an idea of the characteristics of the Cu 2+ second binding site, further indicating that such a binding site is weaker than the ATCUN motif.
The above results support the idea that it is just the Cu 2+ bound to the second site containing the His-His motif that is reduced by ascorbate. This changes the previous reduction model according to which Cu 2+ coordinated to the ATCUN binding site would first be reduced by ascorbate and the resulting Cu + would then be transferred to the His- Ctr1(1-14) causes a reduction of ascorbate comparable to that detected for the solution containing no ligand.   Figure 3b depicts the UV-vis spectra for the Cu 2+ -Ctr1(1-14) system obtained with a 1:1 (blue line) and 2:1 (magenta line) Cu 2+ /Ctr1(1-14) ratio in the same experimental conditions employed for the Cu 2+ ascorbate-dependent reduction. This figure clearly shows that λmax shifts towards greater wavelenghts as binuclear species prevail (Figure 4). The difference spectrum (brown line), obtained by subtracting the spectral contribution characteristic of the ATCUN binding site (blue line) from the spectrum obtained with a greater Cu 2+ /Ctr1(1- 14) ratio (magenta line), gives an idea of the characteristics of the Cu 2+ second binding site, further indicating that such a binding site is weaker than the ATCUN motif.
The above results support the idea that it is just the Cu 2+ bound to the second site containing the His-His motif that is reduced by ascorbate. This changes the previous reduction model according to which Cu 2+ coordinated to the ATCUN binding site would first be reduced by ascorbate and the resulting Cu + would then be transferred to the His- The above results support the idea that it is just the Cu 2+ bound to the second site containing the His-His motif that is reduced by ascorbate. This changes the previous reduction model according to which Cu 2+ coordinated to the ATCUN binding site would first be reduced by ascorbate and the resulting Cu + would then be transferred to the His-His diade that is able to stabilize the reduced state of the metal ion [23,27,28,34]. Noteworthy is the fact that the literature data reveal that our model for the reduction of the Cu 2+ -Ctr1 (1-14) has been also suggested for HSA that contains an unspecific second metal binding domain [44]. This additional coordination site, characterized by an affinity to Cu 2+ lower than that of the Cu 2+ binding to the ATCUN motif of the protein, is thought to be responsible for ascorbate oxidation. In addition, a second binding site (sometimes termed 'the secondary site') in which Cu 2+ is coordinated to His13 and His14 has been evidenced for Aβ (4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16) ; as found for the second site HSA and Ctr1 (1-14) that binds Cu 2+ , this non-ATCUN Aβ-Cu 2+ species has a micromolar affinity constant and oxidizes ascorbate [45].

Ag + Forms Mononuclear Complex with Ctr1 (1-14)
Cu + binding affinity to the Ctr1 N-terminus has been studied using peptides with different sequences as well as different experimental approaches [23,30,33,34]. Initially, the Met-rich motif of yCtr1 was investigated [25]; the study was then extended to hCtr1 that also contains a His-rich sequence [23]. For the latter system, the conditional stability constants were determined by competition between Cu + bound to Ctr1 (1-14) copper(I) complexes and bicinchoninic acid, the affinity constants of which are precisely known; experiments were also extended to polypeptides encompassing additional His and Met motifs. These experiments were carried out in anaerobic conditions to avoid the oxidation of [Cu(MeCN) 4 ]PF 6 . A narrow metal to peptide ratio was explored in these competition experiments; the peculiar experimental conditions employed might be the cause of some contrasting data.
Since Cu + cannot be studied straightforwardly in aqueous solutions and Ag + is considered a good surrogate for monovalent copper [13,20], we investigated by potentiometry different Ag + /Ctr1 (1-14) ratios (1:1, 1:2 and 2:1) to determine the binding constants and to obtain the species distribution of Ag + complexes with Ctr1 (1- 14) . The treatment of the potentiometric data, which also included Ag + hydrolytic species [46], invariably converged towards the set of species listed in Table 4; the distribution diagram of these species is shown in Figure 5. [AgLH 4 ] starts to form around pH 4 and reaches its maximum percentage of formation close to pH 6, where the carboxylic groups are deprotonated; log K* (log K* 114 = log β 114 − log β 014 = 33.45 − 30.47 = 2.98) value suggests that at least one imidazole nitrogen of a deprotonated His residue is involved in the binding, while the remaining histidine moieties and the amino and tyrosine residues are protonated. As pH increases, [AgLH 4 ] loses a proton yielding [AgLH 3 ], the log K* of which (log K* 113 = log β 113 − log β 013 = 27.40 − 24.18 = 3.22) is consistent with a 2N {2N Im } binding environment resulting from the coordination of an additional imidazole nitrogen of a second deprotonated His residue. The log K* (log K* 112 = log β 112 − log β 012 = 21.05 − 17.27 = 3.78) of the species that prevails at physiological pH ([AgLH 2 ]) is consistent with a third deprotonated His.
In this context, in an effort to understand whether the His or Met residue is involved in the binding, a systematic study of metal complexes with mutated Ctr1 (1- 14) , in which the His and Met residues have been replaced by alanine or norleucine residues, has been carried out [23]. The affinity constant values reported by this study evidence that the simultaneous substitution of His5 and His6 as well as the replacement of His3 result in a significant decrease of the log K found for the un-mutated peptide, suggesting a possible involvement of the histidine residue of the ATCUN motif in Cu + binding. On the other hand, the replacement of all Met residues also affects the metal affinity, though the decrease of log K is smaller than that displayed by the His mutated peptides. His3 involvement in Cu + binding to monomeric MDHSHH and its trimeric derivative is also supported by an NMR investigation showing that Met1 is also involved in Cu + coordination to these Ctr1 (1-14) metal binding models [27]. In addition, NMR findings show that in the wild-type peptide in the presence of Cu + , the most significant chemical shifts are detected for the His and Met residues, suggesting that His3, Met7, Met9 and Met12 are the most relevant residues for Cu + binding [29]. However, converging evidence from XAS, computational and NMR results supports the involvement of His5 and His6 in Cu + -Ctr1 (1-14) complexes with a {2N im , 1O, 1S} coordination mode [27,47]. In this context, in an effort to understand whether the His or Met residue is involved in the binding, a systematic study of metal complexes with mutated Ctr1 (1-14), in which the His and Met residues have been replaced by alanine or norleucine residues, has been car ried out [23]. The affinity constant values reported by this study evidence that the simul taneous substitution of His5 and His6 as well as the replacement of His3 result in a signif icant decrease of the log K found for the un-mutated peptide, suggesting a possible in volvement of the histidine residue of the ATCUN motif in Cu + binding. On the other hand the replacement of all Met residues also affects the metal affinity, though the decrease o log K is smaller than that displayed by the His mutated peptides. His3 involvement in Cu binding to monomeric MDHSHH and its trimeric derivative is also supported by an NMR investigation showing that Met1 is also involved in Cu + coordination to these Ctr1(1-14 metal binding models [27]. In addition, NMR findings show that in the wild-type peptide in the presence of Cu + , the most significant chemical shifts are detected for the His and Met residues, suggesting that His3, Met7, Met9 and Met12 are the most relevant residues for Cu + binding [29]. However, converging evidence from XAS, computational and NMR results supports the involvement of His5 and His6 in Cu + -Ctr1(1-14) complexes with a {2Nim 1O, 1S} coordination mode [27,47].
Above the physiological pH range, only one Ag + complex species ([AgLH]) that grad ually makes way for Ag + hydroxylated species is detected; the precipitation of this latter species prevents a reliable investigation of the effect resulting from the (possible) depro tonation of the tyrosine OH group. AgLH formation is associated with a pK value (pK11 = log β112 − log β111 = 7.33) that recalls the value of the proton dissociation constant of the terminal α-amino group, which is not able to bind silver ion.

The Magnetic Parameters Values Obtained by Electro Spin Resonance (ESR) Measurements Indicate That the Two Binding Sites of Ctr1(1-14) Experience A Different Coordination Environment
Studies of the native peptide as well as shorter fragments or mutated peptides have reported the magnetic parameters of the complex of Cu 2+ with the N-terminus of Ctr1(1-14 Above the physiological pH range, only one Ag + complex species ([AgLH]) that gradually makes way for Ag + hydroxylated species is detected; the precipitation of this latter species prevents a reliable investigation of the effect resulting from the (possible) deprotonation of the tyrosine OH group. AgLH formation is associated with a pK value (pK 111 = log β 112 − log β 111 = 7.33) that recalls the value of the proton dissociation constant of the terminal α-amino group, which is not able to bind silver ion.

The Magnetic Parameters Values Obtained by Electro Spin Resonance (ESR) Measurements Indicate That the Two Binding Sites of Ctr1 (1-14) Experience a Different Coordination Environment
Studies of the native peptide as well as shorter fragments or mutated peptides have reported the magnetic parameters of the complex of Cu 2+ with the N-terminus of Ctr1 (1- 14) involving an ATCUN motif [17,[27][28][29]32]. It is generally accepted that the metal binding to the N-terminal domain of Ctr1 is characterized by a 4N (an amino nitrogen, an imidazole nitrogen and two deprotonated amide nitrogens) coordination mode. The g and A values obtained for the Cu 2+ mononuclear complex with a 0.9 metal-to-ligand ratio were used for the attribution of the parameters of the analogous 4N complex found for the Cu 2+ complexes with Ctr1 (1-14) detected with a 1.8 metal to ligand ratio over a large pH range. For the mononuclear complex (0.9 metal-to-ligand ratio), about one equivalent of 63 Cu 2+ was added to a peptide solution adjusting the pH value, which was then adjusted to 4.0. The ESR spectrum run on this frozen solution shows at least two patterns of signals beside those due to Cu 2+ hexaaquo ion ( Figure 6, Table 5); for a better visualization of the pattern of the complex species, the features of the hexaaquo ion (spectrum obtained at pH 3 not shown) were subtracted from the overall spectrum. The Hamiltonian parameters of the species were obtained directly from the resulting spectrum and used as initial values for the simulation of the subtracted spectrum. The simulation yields the following values: g = 2.307, A = 172 × 10 −4 cm −1 and g = 2.245, A = 178 × 10 −4 cm −1 . As the pH is raised to 4.5, the spectral contribution of latter chromophore predominates though the ESR spectrum does not change significantly. A further pH increase (5.0-5.5) causes a sizable change in the ESR spectra. The Hamiltonian parameters obtained by simulating the spectra collected over this pH range are: g = 2.177, A = 205 × 10 −4 cm −1 and g = 2.230, A = 185 × 10 −4 cm −1 . The first set of parameters is typically obtained for ATCUN Cu 2+ complexes [37]. The second set of Hamiltonian parameters indicates the formation of a different second coordination site. This site can be assigned to the coordination of a histidinic imidazole nitrogen and two deprotonated amide nitrogen atoms [41]. species were obtained directly from the resulting spectrum and used as initial values fo the simulation of the subtracted spectrum. The simulation yields the following values: g = 2.307, A‖ = 172 × 10 −4 cm −1 and g‖ = 2.245, A‖ = 178 × 10 −4 cm −1 . As the pH is raised to 4. the spectral contribution of latter chromophore predominates though the ESR spectrum does not change significantly. A further pH increase (5.0-5.5) causes a sizable change i the ESR spectra. The Hamiltonian parameters obtained by simulating the spectra collecte over this pH range are: g‖ = 2.177, A‖ = 205 × 10 −4 cm −1 and g‖ = 2.230, A‖ = 185 × 10 −4 cm The first set of parameters is typically obtained for ATCUN Cu 2+ complexes [37]. The se ond set of Hamiltonian parameters indicates the formation of a different second coord nation site. This site can be assigned to the coordination of a histidinic imidazole nitroge and two deprotonated amide nitrogen atoms [41]. Above pH = 6.6, only the ATCUN features are detected in the ESR spectra as they ar also found by other groups [37]; however, the Hamiltonian parameters do not signif cantly change as we move towards more alkaline regions. Above pH = 6.6, only the ATCUN features are detected in the ESR spectra as they are also found by other groups [37]; however, the Hamiltonian parameters do not significantly change as we move towards more alkaline regions.
The ESR spectrum runs on a solution of Ctr1 (1-14) containing 1.8 equivalents of 63 Cu 2+ at pH = 4 and mainly shows the features of the hexaaquo metal ion. Over the pH range 5-7, despite the strong decrease of the signals of hexaaquo copper ion that progressively fade to disappear above pH 6, the spectra are almost featureless in the parallel part, while they show a very broad signal in the perpendicular part. The broadening likely results from the formation of binuclear complex species characterized by similar coordination environments.
Since the magnetic interaction of the two copper ions strongly depends on their mutual distance, [Cu 2 LH -2 ] and [Cu 2 LH -3 ] were modeled by HyperChem (HyperChem, Hypercube, 2000) [48,49] to have an estimate of the Cu-Cu distance. The carboxylate oxygen is most probably provided by the Asp2 residue, and this would bring the two copper ions into close proximity (i.e., about 5.7 Å) in the frozen solution, thus causing the broadening of the ESR signal due to the dipolar interaction [50]. Aggregation effects might also be involved in the severe broadening occurring at low temperatures in the ESR spectra, as observed by Shenberger et al. [29].
As we move to more alkaline regions, the further deprotonation leads to a 4N {2N Im , 2N -} coordination mode, with a loss of the coordinated Asp. The two Cu 2+ ions move apart (i.e., about 7.3 Å), thus weakening the dipolar interaction. Consequently, above pH 8, the parallel features of the ESR spectra become visible, although some broadening still persists. A signal at 340 mT is also detected, which may be ascribed to the ATCUN spectrum recorded in alkaline regions for solutions containing one Cu 2+ equivalent ( Figure 6). The magnetic parameters in alkaline regions were obtained by simulation using the parameters of the {NH 2 , N Im , 2N -} and the {2N Im , 2N -} coordination modes, i.e., g = 2.181, A = 205 × 10 −4 cm −1 and g = 2.210, A = 185 × 10 −4 cm −1 , respectively, and introducing in the simulation larger linewidth values, since some of the spectra show line broadening.

Immunoreactivity of Ctr1 Peptides of Different Length with Anti-Ctr1 Extracellular Domain Antibody
The 2N1O1S coordination mode of the Cu + complex with Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) is functional both to the N-terminus reductase ability and to the transport of metal ion via the ectodomain C-terminus region responsible for cytoplasm metal uptake. The labile complex formation would favour Cu + exchanges involving a series of labile sites down a concentration gradient and/or a thermodynamic stability gradient, as suggested by Pushie et al. [27]. A recent report emphasizes the role played by Cu + binding to hCtr1  in the peptide folding that is in turn responsible for its interactions with cell membranes; noteworthy is the fact that apo-hCtr1 (1- 46) does not interact with the membrane. The conformation change has been ascribed to the metal ion coordination to some residues of the first and second His-rich and Met-rich domain of Ctr1  , which encompasses different copper binding sites such as His23-His6, His6-His33 and His23-Met7 [33].
Inspired by these findings and by the models proposed so far, we have performed preliminary assays to compare the molecular recognition by an anti-Ctr1 extracellular antibody of the different ectodomains present in Ctr1 (1-14) and a longer peptide Ctr1  in the absence and presence of copper to gain further information on the interaction with the membrane.
The dot blot analysis shows that the antibody recognizes the generic epitope associated with Ctr1 (1-25) , while it does not recognize the shorter Ctr1 (1-14) (Figure 7). Both peptides (1 mM in phosphate buffer) were also pre-incubated for 15 min with copper in different ratios (L/ Cu 2+ 1:1, 1:2 and 1:3). The dot blot analysis shows a decrease of antibody affinity for Ctr1 (1-25) following pre-incubation, with copper in the samples containing the metal ion at a ratio of 1:1 and no interactions with antibodies following incubation with higher metal-to-ligand ratios (L/Cu 2+ 1:2 and 1:3). The incubation was carried out in the presence of 1% SDS in order to verify the influence of the native peptide conformation on antibody binding. Noteworthy is the fact that the antibody does not recognize Ctr1 (1)(2)(3)(4)(5)(6)(7)(8)(9)(10)(11)(12)(13)(14) when the peptide is used without adding copper nor when it is preincubated with copper.

Antibody
The 2N1O1S coordination mode of the Cu + complex with Ctr1(1-14) is functional both to the N-terminus reductase ability and to the transport of metal ion via the ectodomain C-terminus region responsible for cytoplasm metal uptake. The labile complex formation would favour Cu + exchanges involving a series of labile sites down a concentration gradient and/or a thermodynamic stability gradient, as suggested by Pushie et al. [27]. A recent report emphasizes the role played by Cu + binding to hCtr1(1-46) in the peptide folding that is in turn responsible for its interactions with cell membranes; noteworthy is the fact that apo-hCtr1(1-46) does not interact with the membrane. The conformation change has been ascribed to the metal ion coordination to some residues of the first and second His-rich and Met-rich domain of Ctr1(1-46), which encompasses different copper binding sites such as His23-His6, His6-His33 and His23-Met7 [33].
Inspired by these findings and by the models proposed so far, we have performed preliminary assays to compare the molecular recognition by an anti-Ctr1 extracellular antibody of the different ectodomains present in Ctr1(1-14) and a longer peptide Ctr1(1-25) in the absence and presence of copper to gain further information on the interaction with the membrane.
The dot blot analysis shows that the antibody recognizes the generic epitope associated with Ctr1(1-25), while it does not recognize the shorter Ctr1(1-14) (Figure 7). Both peptides (1 mM in phosphate buffer) were also pre-incubated for 15 min with copper in different ratios (L/ Cu 2+ 1:1, 1:2 and 1:3). The dot blot analysis shows a decrease of antibody affinity for Ctr1(1-25) following pre-incubation, with copper in the samples containing the metal ion at a ratio of 1:1 and no interactions with antibodies following incubation with higher metal-to-ligand ratios (L/Cu 2+ 1:2 and 1:3). The incubation was carried out in the presence of 1% SDS in order to verify the influence of the native peptide conformation on antibody binding. Noteworthy is the fact that the antibody does not recognize Ctr1(1-14) when the peptide is used without adding copper nor when it is preincubated with copper.    Our results evidence that the choice of peptide length is crucial to obtaining a model that appropriately represents Ctr1. hCtr1 N-terminal extracellular domain contains two Met-rich as well as two His-rich motifs involved in the coordination of copper ions, which have been extensively studied by using different model peptides (full N-terminal domain, Ctr1 (1-14) , Ctr1 (1-46) ) [23,30]. Future and deeper investigation of a longer peptide model of Ctr1 (namely, Ctr1 (1-25) ) will hopefully help to better clarify how the extracellular localiza-tion and the copper binding capacity of the hCTR1 N-terminus allow the protein to recruit copper ions from the extracellular environment.

Concluding Remarks
hCTR1 ectodomain has two Met-rich motifs, i.e., the two 7 MXMXXM and 41 MMMXM sequences, and two His-rich motifs arranged as 1 MDHXHH and 22 HHH. Initially, the findings highlighted that removing the Met motifs nullified the CTR1-assisted uptake of copper [51]. Later on, studies on a short peptide fragment (Ctr1 (1-14)) [23] as well as on longer peptides (Ctr1 (1-46) [33] and Ctr1 (1-55) [30]) emphasized the role of the histidine residues of both the first and second His-rich motifs in the binding of the metal ion. The incubation of long peptides with different Cu 2+ /Ctr1 ratios [33] showed that the extracellular part of hCtr1 can bind up to three Cu 2+ and six Cu + , while Ctr1 (1-14) binds one Cu 2+ through the MDH sequence; the study also showed that His5 and His6 residues contribute to the reduction of Cu 2+ in the presence of ascorbate [34].
Our study unveils the formation of Cu 2+ binuclear species, the second binding site of which involves the two imidazole nitrogen atoms of the His5 and His6 residues. The second binding site is characterized by an affinity lower than that of the first Cu 2+ bound to the Ctr1 (1-14) ATCUN motif and is also responsible for ascorbate oxidation. These new results contribute to a better description of the features of hCtr1 (1-14) as a reductase model of the extracellular N-terminus copper binding portion. In light of the role played by the second His-rich motif, we are currently investigating a longer peptide (Ctr1 (1-25) ) that encompasses the 22 HHH sequence by using the same approach employed in the present study.
Since, as also found for some analogous shorter peptides [32], Ctr1 (1-14) may contain water and the accurate determination of the thermodynamic parameters requires concentrations to be precisely known, the water content of stock solutions was checked by potentiometric (protonation) titrations, as routinely done in our group for newly synthesized ligand as well as for ligands that tend to absorb water from the surrounding environment [52][53][54]. The water content was double checked by thermogravimetric analysis (TGA). The samples (1-2 mg) were analyzed (TA Instruments Q500, New Castle, DE, USA) under a nitrogen flow (60 mL·min −1 ) using a heating rate of 5 • C/min and 10 • C/min in 30-200 and 200-800 • C, respectively (sensitivity = 2.5 × 10 −6 g). The TGA analysis shows that, within the experimental error, the ligand contains the same amount of water (ca. 25%) determined by potentiometric titrations and undergoes complete decomposition in the 200-300 • C range; no further weight loss was detected up to 800 • C, indicating that the ligand is free from inorganic impurities. The above figure was considered to have the correct analytical concentration for potentiometric as well as spectroscopic experiments.

Electron Spin Resonance Spectroscopy (ESR)
A Bruker CW-ESR spectrometer (Elexsys E500, Bruker, Karlsruhe, Germany) driven by a PC running the XEpr program and equipped with a Super-X microwave bridge (ER 049X) operating at 9.3-9.5 GHz and a SHQE probe head were used throughout the work. All frozen solution ESR spectra of Cu 2+ complexes were recorded at 150 K by means of an ER4131VT variable temperature unit. A small amount of methanol (up to 10%) was added to the Cu 2+ -Ctr1 samples in order to increase spectral resolution. Parallel spin Hamiltonian parameters were taken directly from the experimental spectra and were always calculated from the 2nd and 3rd line to get rid of errors coming from second order effects [55]. The resolution of the ESR spectra [56] was improved by using isotopically pure 63 Cu(NO 3 ) 2 (0.05 M). In order to achieve a better determination of the magnetic parameters, some of the experimental spectra were simulated by the program Monoclin [57,58], which is able to discriminate one or more species [57]. Instrumental settings of frozen solution ESR spectra were as follows: number of scans, 4; microwave frequency, 9.425-9.429 GHz; modulation frequency, 100 kHz; modulation amplitude, 0.7 mT; time constant, 163 ms; sweep time, 2.8 min; microwave power, 20 mW; receiver gain, 60 dB.
The initial pH was always adjusted to 2.4. To avoid systematic errors and check for reproducibility, the electromotive force (EMF) values of each experiment were taken at different time intervals. Hyperquad [59] and HySS [60] were used to calculate complexation constants from potentiometric data and species distribution diagrams, respectively.

Ultraviolet-visible (UV-vis) Measurements
UV-vis spectra were recorded at 25 • C using an Agilent 8453 (Agilent Technologies, Santa Clara, CA, USA) or a Jasco V-670 (Jasco, Easton, MD, USA) spectrophotometer. The concentrations of the peptide and Cu 2+ used to record absorption spectra were the same as those for the potentiometric titrations.
Combined spectroscopic and potentiometric metal-complex titrations were performed into a 3 mL quartz cuvette with a 1 cm path length to obtain the spectrum in the visible region at each pH value. These experiments were replicated at least three times for each copper-peptide system. Spectroscopic data were processed using Hyperquad [58]. UV-vis spectra were deconvoluted using PeakFit 4.0 (Jandel Scientific Software).

Circular Dichroism (CD) Experiments
CD spectra were obtained at 25 • C under a constant flow of nitrogen in the 280-800 nm range with a model 810 Jasco spectropolarimeter (Jasco, Easton, MD, USA) at a scan rate of 50 nm min −1 , a resolution of 0.1 nm and a 1 cm path length. The spectra were recorded as an average of 3-5 scans. Calibration of the instrument was performed with a 0.06% aqueous solution of ammonium camphor sulfonate. The CD spectra of Cu 2+ complexes as a function of the pH were obtained in both the 190-250 and 250-800 nm regions. All solutions were freshly prepared using twice distilled water. Cu 2+ and peptide concentrations used for CD spectra in the visible region were identical to those used in the potentiometric titrations. Results are reported as ∆ε (molar dichroic coefficient) in M −1 cm −1 .
A stock solution (20 mM) of ascorbate was prepared in milli-Q water just before the experiment and was used immediately. Owing to the instability of ascorbate solutions, fresh solutions were prepared for each experiment. Cu(NO 3 ) 2 stock solutions were prepared and standardized with ethylenediaminetetraacetic acid, as reported elsewhere [61].