Zinc-Induced Conformational Transitions in Human Islet Amyloid

Oct 9, 2018 - *Michael T. Bowers, Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106, USA. E-mail: ...
0 downloads 0 Views 6MB Size
Subscriber access provided by UNIV TEXAS SW MEDICAL CENTER

B: Biophysics; Physical Chemistry of Biological Systems and Biomolecules

Zinc Induced Conformational Transitions in Human Islet Amyloid Polypeptide and Their Role in the Inhibition of Amyloidosis Alexandre Iourievich Ilitchev, Maxwell J. Giammona, Jürgen Nicholas Schwarze, Steven K. Buratto, and Michael T. Bowers J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.8b06206 • Publication Date (Web): 09 Oct 2018 Downloaded from http://pubs.acs.org on October 10, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Zinc Induced Conformational Transitions in Human Islet Amyloid Polypeptide and Their Role in the Inhibition of Amyloidosis Alexandre I. Ilitchev†, Maxwell J. Giammona†, Jurgen N. Schwarze†, Steven K. Buratto†, Michael T. Bowers†*

† Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106, USA

 

1   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

ABSTRACT Type-2 diabetes mellitus (T2DM) is a disease hallmarked by improper homeostasis within the islets of Langerhans of the pancreas. The most critical species affected is insulin, which is produced by the β-cells of the islets, but there are a number of other species co-packaged and co-secreted within the insulin granules. This includes zinc, which exists in high (millimolar) concentrations within the β-cells, and islet amyloid polypeptide (IAPP) which is an amyloid peptide thought to induce β-cell apoptosis through self-association into toxic amyloid oligomers. Zinc is essential in the packaging of crystalline insulin within the vesicles but it can also bind and interact with IAPP. This implies a complex relationship between all three species and diabetes, particularly in the structure and function of toxic IAPP aggregates. Atypical (low or high) concentrations of zinc generally appear to correlate with increased hIAPP aggregation, whereas physiological zinc concentrations have an inhibitory effect. To better understand how zinc ions alter the monomer and oligomer structure of hIAPP in vitro, we employ a combination of ion mobility mass spectrometry and atomic force microscopy. We observe an increase in the extended β-hairpin conformation of hIAPP when it is bound to zinc. With sufficiently low concentrations of zinc this could result in an association site for zinc-free hIAPP, promoting amyloid aggregation. At high zinc concentrations we see the appearance of a secondary zinc association site whose coordination could account for the loss of inhibition at high zinc concentrations. Generally, it appears that zinc preferentially stabilizes the β-hairpin conformation of hIAPP and the population of zincbound hIAPP in solution determines what effect this has on amyloid aggregation.

 

2   ACS Paragon Plus Environment

Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

INTRODUCTION The interaction of metal ions with amyloid peptides is a well characterized phenomenon and a critical direction in the study of amyloid aggregation inhibition.1-2 The 37 amino acid islet amyloid polypeptide (IAPP) is co-secreted with insulin from pancreatic β-cells, and its aggregation is implicated as one of the primary causes of pancreatic apoptosis in type-2 diabetes mellitus (T2DM).3-6 IAPP is packaged in insulin rich vesicles called insulin granules, within which insulin is stored in densely packed crystals. The packaging of crystalline insulin necessitates two zinc ions per insulin hexamer,7-8 which results in β-cells containing the highest in vivo concentration of zinc.9 While this relationship is broadly understood there is evidence that hIAPP also coordinates with zinc in the insulin granules.10-12 This seems to indicate that the presence of zinc not only has a critical role to play in the packaging of insulin but the inhibition of IAPP aggregation as well. Despite these observations, the relationship between zinc transporter disorder (which lowers the concentration of zinc within the granules) and diabetes has proven difficult to deconvolute, despite the transporter’s evident effect on the efficiency of insulin packaging.13 Differing zinc concentrations have been shown to either enhance or inhibit the prevalence of both type-1 and type 2 diabetes.14-18 This could be directly linked to the disparate effects zinc has on hIAPP aggregation, as several studies have found correlation between the loss of fibril formation and zinc-hIAPP coordination.19-20 The standard Thioflavin-T assay reveals decreased β-sheet character in solutions containing quantities of zinc up to 2000:1 ZnCl2:hIAPP.20 Whether zinc binding also inhibits toxic early oligomer formation is still a matter of contention.21-23 Some have found that low zinc

 

3   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

concentrations inhibit IAPP aggregation indirectly via the generation of free insulin, which acts as an inhibitor to the amyloid. Others have focused on the direct interaction between zinc and hIAPP and have found that the ability of 1:1 hIAPP:zinc adducts to exhibit monomer-monomer amyloid interaction is greatly restricted.20, 23 It appears as if very low (< 2:1 Zn2+:hIAPP) and high (> 100:1 Zn2+:hIAPP) concentrations have a promoting effect on aggregation whereas intermediary amounts inhibit self-association.20, 23

Zinc binding to hIAPP occurs primarily at the His18 binding site, which has a strong

affinity for the metal ion, although several studies have pointed to a potential secondary binding site of indeterminate location within the hIAPP sequence.20, 24 This binding site activates at particularly high zinc:hIAPP ratios and may account for the loss of inhibition at those concentrations. Clearly there is need for further clarification of the complex role zinc plays in the self-association of hIAPP, particularly in the structure and assembly of early oligomers. Here we employ a combination of ion mobility mass spectrometry (IMS-MS) and atomic force microscopy (AFM) to view early aggregates as well as larger macroscopic features. Ion mobility mass spectrometry with a soft electrospray ionization source is ideal for analyzing ligand complexes of identical composition but differing conformations.25-27 Atomic force microscopy, meanwhile, has been used to analyze the nature of larger assemblies that form beyond the scope that is observable in IMS-MS.28-30 This combination of techniques proves particularly fruitful in the study of early assemblies as it allows us to ascertain what conformations zinc induces in vitro and how these oligomers inform the resulting macroscopic fibril assemblies.

 

4   ACS Paragon Plus Environment

Page 5 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

MATERIALS AND METHODS Peptide Sample Preparation. Human islet amyloid polypeptide with an amidated Cterminus (3903.3 amu) was purchased from Genscript (Piscataway, NJ) and used without further purification (>95% purity via HPLC). The sample was dissolved in 100% hexafluoroisopropanol (HFIP) (Sigma-Aldrich, St. Louis, MO) and 10µL aliquots were taken and dried under vacuum overnight. Dry samples were redissolved in 1:1 methanol:water (pH = 7.4) to a final concentration of 50 µM. Zinc chloride (ZnCl2) was purchased from EMD Millipore (Billerica, MA) and used to create a stock solution of 1mM ZnCl2. Aliquots of the stock were combined with hIAPP to generate the desired Zn2+: 50µM hIAPP ratio.

Ion Mobility Experiments. Ion mobility and mass spectra were collected on a homebuilt instrument which has been described in detail previously but will be summarized briefly here.31 The instrument consists of a nano-electrospray ionization source (nanoESI) connected via an ion funnel to a 4.5 cm drift cell filled with ~3.5 Torr Helium buffer gas. Ions generated at the source are pulled through the drift cell by the application of a tunable weak electric field. Ions leaving the drift cell are mass filtered by a quadrupole and detected by a conversion-dynode coupled electrode multiplier. Ions can be introduced continuously into the drift cell to generate a mass spectrum or pulsed at discreet intervals, with the time required to reach the detector recorded as the analyte’s drift time yielding an arrival time distribution (ATD).32 These ATDs can be analyzed by comparing them to the flux of single cross section species to indicate whether one or more families of structures are involved.33 Details are given in the supporting information.

 

5   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

Additional high resolution mass spectra were collected on a Bruker MaXis (Billerica, Massachusetts) outfitted with an off-line NanoElectrospray source.34

Atomic Force Microscopy. Atomic force microscopy (AFM) images were collected using a MFP-3D Atomic Force Microscope (Asylum Research, Goleta, CA) operated in tapping mode in air. The silicon probe had a cantilever spring constant of 7 N/m and a resonant frequency of 155 kHz (MikroMasch USA, Lady’s Island, SC). Samples used for the IMS-MS experiments were incubated for one hour and diluted 10x before being deposited on freshly cleaved atomically flat V1-grade mica (TedPella, Redding, CA, USA). Once deposited the samples were dried under vacuum overnight. Height values were extracted for features in the AFM images using an algorithm built into the Asylum AFM software. A mask was applied to areas in the image with a height greater than 0.4 nm and the maximum height value for each of the masked areas of the image was extracted. Analysis of the particle height data was performed using the KernelMixtureDistribution function included in the Mathematica 11.2 package.35 To begin the analysis an initial estimate of the kernel bandwith, h, which determines the sensitivity of the histogram, is required. These initial estimates for h were made using Silverman’s normal distribution approximation, given by: ℎ ≈ 1.06𝜎𝑛!!

!

Where σ is the standard deviation of the data and n is the population size.36 A probability density function is then generated from the mixture distributions of the histograms and populations are identified based on the 1st and 2nd derivative of the probability density

 

6   ACS Paragon Plus Environment

Page 7 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

function. Populations could not be identified for features with a height of > 3 nm due to the low number of features observed in this height regime.

Figure 1. (A) Comparison of the 50µM hIAPP mass spectra obtained on the IMS-MS instrument at varying zinc concentrations. As zinc is divalent there is a shift toward higher charge states when the ratio of zinc increases. (B) Arrival time distributions of 993 m/z peak corresponding to a Zn2+ adduct to hIAPP. Generally, as zinc concentration increases the amount of extended conformer (746 Å2) also increases. Each peak is fit using the procedure described in the supporting information. Illustrations of structures adapted from Ref. 37.

 

7   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

RESULTS AND DISCUSSION Addition of Zinc Produces Primary and Secondary Aqua-Zinc Adducts. The addition of zinc appears to have a concentration dependent effect on hIAPP aggregation.20, 23 Hence we began by first ascertaining the effect that zinc addition has on the mass spectral peak distribution of hIAPP. We began with 1:1 ZnCl2:hIAPP and increased the concentration up to 100:1 with higher concentrations prohibited by the relative insolubility of zinc, as it began to precipitate out as Zn(OH)2 (Fig. 1). The +4 charge monomer peak shows the most association with zinc (Fig. 1A), likely due to the fact that the +2 charge zinc adducts add an additional charge to the +3 native state. Previous studies have found that zinc is displaced when Zn-bound hIAPP monomers aggregate,24 so it is perhaps not surprising that the 2/+5 dimer peak does not appear to have an associated Zn. The first adduct peak to appear is hIAPP+Zn2+ which predominates even at higher concentrations. This is followed by hIAPP+Zn(H2O)52+ which begins to appear at 10:1 Zn2+:hIAPP. The addition of hydrated zinc has not been described in other experiments but is not entirely surprising as the octahedral Zn(H2O)62+ complex is the principal form of zinc in aqueous solutions.38-39 At 25:1 Zn2+:hIAPP we begin to see the formation of an additional zinc adduct state corresponding to hIAPP+2Zn(H2O)52+. No hIAPP+2Zn2+ or hIAPP+Zn2++Zn(H2O)52+ state is observed. In Figure 1B the ATDs for the hIAPP+Zn2+ adduct are given for several Zn2+ concentrations. The fact that the extended β-hairpin conformer becomes more prevalent as the concentration of Zn increases is noteworthy. This is an interesting result which indicates that dynamics are occurring in solution with Zn2+ attaching and detaching from IAPP in a quasi-steady-state manner. The data indicate that Zn2+ binds more strongly to

 

8   ACS Paragon Plus Environment

Page 9 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

the extended β-hairpin conformer, and as the Zn2+ concentration increases the dynamics select this conformer preferentially until saturation sets in at ~50:1 Zn2+:hIAPP. Whether or not multiple Zn2+ ions attach in solution but are lost in the ESI process is not possible to determine.

High Resolution Mass Spectrometry Corroborates Peak Distribution. While our small drift cell instrument allows us to obtain ion mobility data with tunable drift cell injection parameters to determine the stability of our complexes and conformations, the mass spectrum resolution is low and results in some ambiguity in the assignment of peaks, particularly for a peak distribution as complex as the one seen in the 100:1 Zn2+:hIAPP spectrum. To verify our initial assignments we employed a higher resolution ion mobility mass spectrometer with a quadrupole time of flight mass analyzer which allows for isotopic MS resolution.34 As is seen in Figure 2A, peak positions and distributions match well with those obtained on the smaller instrument, with additional resolution revealing potassium adduct peaks paired with each primary peak, which is common in positivemode electrospray.40-41 As the m/z = 1055 peak is set apart from the remaining +4 charge monomer peaks, there was some concern that it was unrelated to the remaining distribution. However, it is evident that the mass spacing corresponds to a +4 charge state (Fig. 2B) grouping it with the remaining zinc adduct peaks.

 

9   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

Figure 2. High resolution mass spectra of the 1/+4 monomer of hIAPP (A) at 100:1 Zn2+:hIAPP. Lower resolution mass spectrum obtained on the IMS-MS instrument at identical sample conditions is overlaid in grey for comparison. Each primary highresolution peak corresponding to hIAPP and its zinc adducts is coupled with a corresponding potassium adduct peak. Close up of the 1051.5 m/z peak is inset (B) to show the isotope spacing, which corresponds to a +4 charge state. IMS-MS instrument peaks annotated with (*) indicate gas-phase deamination of the sample within the source.

 

10   ACS Paragon Plus Environment

Page 11 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 3. Monomeric 1/+4 hIAPP arrival time distributions with varying zinc adducts at 50:1 Zn2+:hIAPP. Anhydrous zinc (B) shifts the distribution toward the extended βhairpin conformation. Hydrated zinc adducts produce a more pronounced shift (C, D).   Each peak is fit using the procedure described in the supporting information. Illustrations of structures adapted from Ref. 37.

 

11   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

Zinc Adducts Stabilize the Amyloidogenic β-Sheet Conformation of hIAPP. The extended β-hairpin conformation of IAPP is unique to amyloidogenic species and has been identified as a precursor to toxic oligomers.37, 42-44 One avenue of IAPP aggregation inhibition has therefore focused on prohibiting the transition of the helix-coil monomer to the extended β-hairpin. What we have found, however, is that zinc actively promotes the extended β-hairpin conformation, as seen in Figure 3. The addition of the first aqua-zinc complex significantly shifts the distribution toward the extended conformation and this effect becomes more pronounced as the second aqua-zinc appends to the peptide. It is likely that the Zn2+ or Zn(H2O)52+ ions attach at the histidine at position 18. This attachment may well destabilize the helix-coil (more compact) form of hIAPP relative to the hairpin. It is also likely that His18 is more accessible in the β-hairpin conformer leading to preferential attachment of Zn2+ or Zn(H2O)52+ in solution. As seen in Figure 4, the conformational distributions only moderately and inconsistently change with injection energy with and without zinc, strongly supporting these are solution and not gas-phase structures.37 To further investigate the role these stable hIAPP adducts play on the aggregation propensity of hIAPP, atomic force microscopy (AFM) was employed to analyze the bulk solution for early proto-fibril aggregates.

 

12   ACS Paragon Plus Environment

Page 13 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Figure 4. Representative injection energy studies of 1/+4 hIAPP monomer peaks with (B) and without (A) zinc adducts at 500:50µM Zn2+:hIAPP. The conformational distribution between extended and compact monomer remains fairly static regardless of the energy contributed to the system, indicating that these distributions are solution conformations and are not a result of gas-phase conformational transitions. Each peak is fit using the procedure described in the supporting information.

 

13   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

Atomic Force Microscopy Shows Inhibition of Fibril Formation by Zinc Ligands. To connect the role of zinc-bound hIAPP to the assembly of fibrillar structures, the same solutions used for IMS-MS were diluted 10x and pipetted onto atomically flat muscovite mica and analyzed via AFM. Figure 5 shows representative AFM images collected from samples containing from 1:1 to 100:1 Zn2+:hIAPP, as well as a pure hIAPP sample for reference. The overall trend of the AFM data corroborates what has been previously observed in the literature that higher zinc concentrations progressively inhibit the formation of fibril and proto-fibril assemblies. As the concentration of Zn2+ ions in solution increases relative to hIAPP two overall trends are observed.

Figure 5. Representative AFM images of human islet amyloid polypeptide (hIAPP) with varying concentrations of zinc ion in solution. All samples were incubated at 50µM hIAPP for one hour before being diluted 10x for imaging. As the ratio of zinc increases the quantity of proto-fibril assemblies in solution decreases.

 

14   ACS Paragon Plus Environment

Page 15 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

First, starting at 10:1, an increasing number of globular aggregates appear between extended proto-firbils. As the concentration of zinc increases, more of these nonamyloidogenic features are seen relative to the amount of proto-fibrillar material. This is surprising given the ion-mobility results show that increasing the concentration of Zn2+ increases the relative amount of extended β-hairpin hIAPP monomer (Fig. 1). From this we can conclude that the Zn2+ and Zn(H2O)52+ associated β-hairpin conformers are less amyloidogenic than free β-hairpin hIAPP in solution.

Figure 6. Particle height histogram and resulting probability distribution function of the sum of several AFM images collected from samples prepared at 500:5 Zn2+ to hIAPP. Total particle count N = 250. Five distinct peaks are seen, corresponding to stacking molecular layers of protein. Features above six layers are likely not seen as the structures can no longer stand vertically and lay flat on the mica surface as a single layer instead.

 

15   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

The second change of note is an increased spacing between features in the AFM images as the concentration of Zn2+ increases. This may be due to increased coulombic repulsion as aggregates gain more positive charge. The charge in the Zn2+ and Zn(H2O)52+ adducts is now more localized at His18. The tightly bound water ligands likely inhibit the rearrangement of β-hairpin to β-sheet necessary for amyloid aggregation. Previously we have also noted that monomer, when unable to aggregate, can weakly associate and form a film on the mica surface which further explains the greater feature separation.30   Finally, the distribution of heights exhibited by the globular aggregates may be analyzed to determine how these structures interact. The resulting particle height histogram and fit peaks are given in Figure 6. The largest peak is at ~0.5 nm and can be assigned as single “molecular layer” of protein. This would include both single layer globular aggregates as well as single filament proto-fibrils lying flat against the mica substrate. Taller peaks are apparent at 2, 4, 5, and 6 molecular layers. Based on this progression, it is likely that this layer by layer addition continues to longer structures, but that after 6 layers, these structures lay down flat on the mica and appear within the large single molecular layer peak. As each layer is added, there is some compaction as oligomers accommodate each other, resulting in slightly less than a doubling in height from peaks A to B and peaks B to D. Only a minor population is apparent at 3 molecular layers suggesting that the 4 layer tall features consist of two double layer oligomers, or that the 3 layer tall species is itself unstable without the addition of a 4th layer. It remains unclear if this is the normal progression of growth for hIAPP or if this represents a new mode of aggregation induced by high zinc concentration.

 

16   ACS Paragon Plus Environment

Page 17 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Based on the high amounts of β-hairpin monomer present in the IMS, it seems likely that we are observing the layer-by-layer association of individual β-hairpin monomers. This appears to indicate that, consistent with the IMS-MS measurements, higher zinc concentrations stabilized β-hairpin structures. These stabilized hairpins are less amyloidogenetic than the free conformation but continue to exhibit some propensity for aggregation.

CONCLUSIONS Comparison of concentration dependent hIAPP zinc ligands has revealed several important aspects of zinc mediated aggregation inhibition. The first is that the penta-aqua zinc ion, formed from the displacement of one water molecule from the solution-stable hexa-aqua zinc complex, forms a complex with hIAPP. The fact that these water ligands are retained during electrospray indicates that they are thermodynamically stable within the complex and must play some role in the binding of the complex to hIAPP. Anhydrous zinc-hIAPP ligands are observed but only for the single zinc adduct, suggesting that hIAPP competes with H2O ligands in solution and the IAPP solvation of the first Zn2+ makes further attachment unfavorable. Second, we were surprised to find that addition of zinc ligands caused the monomer distribution to shift from the compact helix-coil state to the extended β-hairpin, which has been identified as the amyloidogenic conformation.37 A single zinc or penta-aqua zinc adduct causes a dramatic shift to the β-hairpin extended conformation either due to disruption of the compact helix-coil conformer or due to accessibility of the His18 adduct in the β-hairpin conformer. The anhydrous zinc ATD favors the β-hairpin less than the

 

17   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

penta-aqua zinc complex, indicating that the ligand waters play a role in the stabilization of the extended structure. The zinc adducts stabilize the amyloidogenic conformation, suggesting that very low concentrations of zinc promote β-hairpin aggregation. However, at moderate concentrations more hIAPP is coordinated with either Zn2+ or Zn(H2O)52+ and these complexes may inhibit isomerization of the β-hairpin to β-sheet dominated oligomers and resultant proto-fibrils. We see moderate decreases in the proto-fibril formation via AFM until zinc concentration is greater than 50:1 Zn2+:hIAPP at which point essentially all hIAPP is coordinated with Zn2+ and β-sheet aggregation is eliminated. Overall, we have observed a number of features of the zinc-hIAPP complex which serves to elucidate the metal ion’s complex effect on hIAPP aggregation. The hydration state of zinc has a marked influence on the conformational stability of the zinc-hIAPP complex. The primary zinc adduct site has been shown to associate with both aqueous and anhydrous zinc ligands, but the secondary zinc site only appends to the pentaaqueous zinc ligand, and only when the primary site is also bound to a penta-aqua complex. Additionally, zinc adducts favor the amyloidogenic extended β-hairpin conformation but they only inhibit the formation of hIAPP oligomers and proto-fibrils when the hIAPP-Zn2+ complex dominates in solution. Simply observing the conformational preference of β-hairpin is insufficient to establish hIAPP amyloidogenesis. The fact that the extended conformers are ligated with Zn2+ and Zn(H2O)52+ must be considered, and in the case of these ligands the extended β-hairpin conformer becomes non-amyloidogenic.

 

18   ACS Paragon Plus Environment

Page 19 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

ASSOCIATED CONTENT Supporting Information. Additional ion mobility experimental details, arrival time distributions of human IAPP in the presence of varying concentrations of zinc, and mass spectra and arrival time distributions of 100:1 Zn2+:hIAPP in different solvent conditions. This information is available free of charge via the Internet at http://pubs.acs.org

AUTHOR INFORMATION Corresponding Author *Michael T. Bowers, Department of Chemistry and Biochemistry, University of California, Santa Barbara, California 93106, USA. E-mail: [email protected]. Phone: 805-893-2893. Fax: 805-893-8703.

Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript.

ACKNOWLEDGMENT Support from the National Science Foundation under grant CHE-1565941 (M.T.B.), is gratefully acknowledged.

 

19   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 24

REFERENCES (1)  Alexandrescu,  A.  T.  Amyloid  Accomplices  and  Enforcers.  Protein  Sci.  2005,  14  (1),   1-­‐12.   (2)  Bush,  A.  I.;  Tanzi,  R.  E.  The  Galvanization  of  Beta-­‐Amyloid  in  Alzheimer's  Disease.   Proc.  Natl.  Acad.  Sci.  U.S.A.  2002,  99  (11),  7317-­‐7319.   (3)  Bram,  Y.;  Frydman-­‐Marom,  A.;  Yanai,  I.;  Gilead,  S.;  Shaltiel-­‐Karyo,  R.;  Amdursky,   N.;  Gazit,  E.  Apoptosis  Induced  by  Islet  Amyloid  Polypeptide  Soluble  Oligomers  Is   Neutralized  by  Diabetes-­‐Associated  Specific  Antibodies.  Sci.  Rep.  2014,  4.   (4)  Guardado-­‐Mendoza,  R.;  Davalli,  A.  M.;  Chavez,  A.  O.;  Hubbard,  G.;  Dick,  E.;  Majluf-­‐ Cruz,  A.;  Tene-­‐Perez,  C.  E.;  Jimenez-­‐Ceja,  L.  M.;  Casiraghi,  F.;  Kamath,  S.,  et  al.   Pancreatic  Islet  Amyloidosis  and  Hyperglycemia  Are  Associated  to  Beta  Cell   Apoptosis,  Alpha  Cell  Proliferation  and  Hyperglucagonemia  in  Non-­‐Human   Primates.  Diabetes  2009,  58,  A404-­‐A404.   (5)  Guardado-­‐Mendoza,  R.;  Davalli,  A.  M.;  Chavez,  A.  O.;  Hubbard,  G.  B.;  Dick,  E.  J.;   Majluf-­‐Cruz,  A.;  Tene-­‐Perez,  C.  E.;  Goldschmidt,  L.;  Hart,  J.;  Perego,  C.,  et  al.   Pancreatic  Islet  Amyloidosis,  Beta-­‐Cell  Apoptosis,  and  Alpha-­‐Cell  Proliferation   Are  Determinants  of  Islet  Remodeling  in  Type-­‐2  Diabetic  Baboons.  Proc.  Natl.   Acad.  Sci.  U.S.A.  2009,  106  (33),  13992-­‐13997.   (6)  Lin,  C.  Y.;  Gurlo,  T.;  Kayed,  R.;  Butler,  A.  E.;  Haataja,  L.;  Glabe,  C.  G.;  Butler,  P.  C.   Toxic  Human  Islet  Amyloid  Polypeptide  (H-­‐Iapp)  Oligomers  Are  Intracellular,   and  Vaccination  to  Induce  Anti-­‐Toxic  Oligomer  Antibodies  Does  Not  Prevent  H-­‐ Iapp-­‐Induced  Beta-­‐Cell  Apoptosis  in  H-­‐Iapp  Transgenic  Mice.  Diabetes  2007,  56   (5),  1324-­‐1332.   (7)  Blundell,  T.  L.;  Cutfield,  J.  F.;  Cutfield,  S.  M.;  Dodson,  E.  J.;  Dodson,  G.  G.;  Hodgkin,   D.  C.;  Mercola,  D.  A.;  Vijayan,  M.  Atomic  Positions  in  Rhombohedral  2-­‐Zinc  Insulin   Crystals.  Nature  1971,  231  (5304),  506-­‐&.   (8)  Dunn,  M.  F.  Zinc-­‐Ligand  Interactions  Modulate  Assembly  and  Stability  of  the   Insulin  Hexamer  -­‐  a  Review.  Biometals  2005,  18  (4),  295-­‐303.   (9)  Foster,  M.  C.;  Leapman,  R.  D.;  Li,  M.  X.;  Atwater,  I.  Elemental  Composition  of   Secretory  Granules  in  Pancreatic-­‐Islets  of  Langerhans.  Biophys.  J.  1993,  64  (2),   525-­‐532.   (10)  DeToma,  A.  S.;  Salamekh,  S.;  Ramamoorthy,  A.;  Lim,  M.  H.  Misfolded  Proteins  in   Alzheimer's  Disease  and  Type  II  Diabetes.  Chem.  Soc.  Rev.  2012,  41  (2),  608-­‐ 621.   (11)  Westermark,  P.;  Li,  Z.  C.;  Westermark,  G.  T.;  Leckstrom,  A.;  Steiner,  D.  F.  Effects   of  Beta  Cell  Granule  Components  on  Human  Islet  Amyloid  Polypeptide  Fibril   Formation.  FEBS  Lett.  1996,  379  (3),  203-­‐206.   (12)  Tomasello,  M.  F.;  Sinopoli,  A.;  Pappalardo,  G.  On  the  Environmental  Factors   Affecting  the  Structural  and  Cytotoxic  Properties  of  Iapp  Peptides.  J.  Diabetes   Res.  2015.   (13)  Xavier,  G.  D.;  Bellomo,  E.  A.;  McGinty,  J.  A.;  French,  P.  M.;  Rutter,  G.  A.  Animal   Models  of  Gwas-­‐Identified  Type  2  Diabetes  Genes.  J.  Diabetes  Res.  2013.   (14)  Yi,  B.;  Huang,  G.;  Zhou,  Z.  G.  Different  Role  of  Zinc  Transporter  8  between  Type   1  Diabetes  Mellitus  and  Type  2  Diabetes  Mellitus.  J.  Diabetes  Invest.  2016,  7  (4),   459-­‐465.    

20   ACS Paragon Plus Environment

Page 21 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(15)  Lemaire,  K.;  Ravier,  M.  A.;  Schraenen,  A.;  Creemers,  J.  W.  M.;  Van  De  Plas,  R.;   Granvik,  M.;  Van  Lommel,  L.;  Waelkens,  E.;  Chimienti,  F.;  Rutter,  G.  A.,  et  al.   Insulin  Crystallization  Depends  on  Zinc  Transporter  Znt8  Expression,  but  Is   Not  Required  for  Normal  Glucose  Homeostasis  in  Mice.  Proc.  Natl.  Acad.  Sci.   U.S.A.  2009,  106  (35),  14872-­‐14877.   (16)  Nicolson,  T.  J.;  Bellomo,  E.  A.;  Wijesekara,  N.;  Loder,  M.  K.;  Baldwin,  J.  M.;   Gyulkhandanyan,  A.  V.;  Koshkin,  V.;  Tarasov,  A.  I.;  Carzaniga,  R.;  Kronenberger,   K.,  et  al.  Insulin  Storage  and  Glucose  Homeostasis  in  Mice  Null  for  the  Granule   Zinc  Transporter  Znt8  and  Studies  of  the  Type  2  Diabetes-­‐Associated  Variants.   Diabetes  2009,  58  (9),  2070-­‐2083.   (17)  Taylor,  C.  G.  Zinc,  the  Pancreas,  and  Diabetes:  Insights  from  Rodent  Studies  and   Future  Directions.  Biometals  2005,  18  (4),  305-­‐312.   (18)  Sladek,  R.;  Rocheleau,  G.;  Rung,  J.;  Dina,  C.;  Shen,  L.;  Serre,  D.;  Boutin,  P.;  Vincent,   D.;  Belisle,  A.;  Hadjadj,  S.,  et  al.  A  Genome-­‐Wide  Association  Study  Identifies   Novel  Risk  Loci  for  Type  2  Diabetes.  Nature  2007,  445  (7130),  881-­‐885.   (19)  Lee,  S.  J.  C.;  Choi,  T.  S.;  Lee,  J.  W.;  Lee,  H.  J.;  Mun,  D.  G.;  Akashi,  S.;  Lee,  S.  W.;  Lim,   M.  H.;  Kim,  H.  I.  Structure  and  Assembly  Mechanisms  of  Toxic  Human  Islet   Amyloid  Polypeptide  Oligomers  Associated  with  Copper.  Chem.  Sci.  2016,  7  (8),   5398-­‐5406.   (20)  Brender,  J.  R.;  Hartman,  K.;  Nanga,  R.  P.  R.;  Popovych,  N.;  Bea,  R.  D.;   Vivekanandan,  S.;  Marsh,  E.  N.  G.;  Ramamoorthy,  A.  Role  of  Zinc  in  Human  Islet   Amyloid  Polypeptide  Aggregation.  J.  Am.  Chem.  Soc.  2010,  132  (26),  8973-­‐8983.   (21)  Nedumpully-­‐Govindan,  P.;  Ding,  F.  Inhibition  of  Iapp  Aggregation  by  Insulin   Depends  on  the  Insulin  Oligomeric  State  Regulated  by  Zinc  Ion  Concentration.   Sci.  Rep.  2015,  5.   (22)  Flannick,  J.;  Thorleifsson,  G.;  Beer,  N.  L.;  Jacobs,  S.  B.  R.;  Grarup,  N.;  Burtt,  N.  P.;   Mahajan,  A.;  Fuchsberger,  C.;  Atzmon,  G.;  Benediktsson,  R.,  et  al.  Loss-­‐of-­‐ Function  Mutations  in  Slc30a8  Protect  against  Type  2  Diabetes.  Nat.  Genet.   2014,  46  (4),  357-­‐+.   (23)  Nedumpully-­‐Govindan,  P.;  Yang,  Y.;  Andorfer,  R.;  Cao,  W.  G.;  Ding,  F.  Promotion   or  Inhibition  of  Islet  Amyloid  Polypeptide  Aggregation  by  Zinc  Coordination   Depends  on  Its  Relative  Concentration.  Biochemistry  2015,  54  (50),  7335-­‐7344.   (24)  Salamekh,  S.;  Brender,  J.  R.;  Hyung,  S.  J.;  Nanga,  R.  P.  R.;  Vivekanandan,  S.;   Ruotolo,  B.  T.;  Ramamoorthy,  A.  A  Two-­‐Site  Mechanism  for  the  Inhibition  of   Iapp  Amyloidogenesis  by  Zinc.  J.  Mol.  Biol.  2011,  410  (2),  294-­‐306.   (25)  Wyttenbach,  T.;  Bowers,  M.  T.  Gas-­‐Phase  Conformations:  The  Ion  Mobility/Ion   Chromatography  Method.  Top.  Curr.  Chem.  2003,  225,  207-­‐232.   (26)  Baker,  E.  S.;  Bernstein,  S.  L.;  Bowers,  M.  T.  Structural  Characterization  of  G-­‐ Quadruplexes  in  Deoxyguanosine  Clusters  Using  Ion  Mobility  Mass   Spectrometry.  J.  Am.  Soc.  Mass  Spectrom.  2005,  16  (7),  989-­‐997.   (27)  Bleiholder,  C.;  Do,  T.  D.;  Wu,  C.;  Economou,  N.  J.;  Bernstein,  S.  S.;  Buratto,  S.  K.;   Shea,  J.  E.;  Bowers,  M.  T.  Ion  Mobility  Spectrometry  Reveals  the  Mechanism  of   Amyloid  Formation  of  a  Beta(25-­‐35)  and  Its  Modulation  by  Inhibitors  at  the   Molecular  Level:  Epigallocatechin  Gallate  and  Scyllo-­‐Inositol.  J.  Am.  Chem.  Soc.   2013,  135  (45),  16926-­‐16937.  

 

21   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 24

(28)  Adamcik,  J.;  Jung,  J.  M.;  Flakowski,  J.;  De  Los  Rios,  P.;  Dietler,  G.;  Mezzenga,  R.   Understanding  Amyloid  Aggregation  by  Statistical  Analysis  of  Atomic  Force   Microscopy  Images.  Nat.  Nanotechnol.  2010,  5  (6),  423-­‐428.   (29)  Smith,  J.  F.;  Knowles,  T.  P.  J.;  Dobson,  C.  M.;  MacPhee,  C.  E.;  Welland,  M.  E.   Characterization  of  the  Nanoscale  Properties  of  Individual  Amyloid  Fibrils.   Proc.  Natl.  Acad.  Sci.  U.S.A.  2006,  103  (43),  15806-­‐15811.   (30)  Economou,  N.  J.;  Giammona,  M.  J.;  Do,  T.  D.;  Zheng,  X.  Y.;  Teplow,  D.  B.;  Buratto,  S.   K.;  Bowers,  M.  T.  Amyloid  Beta-­‐Protein  Assembly  and  Alzheimer's  Disease:   Dodecamers  of  a  Beta  42,  but  Not  of  a  Beta  40,  Seed  Fibril  Formation.  J.  Am.   Chem.  Soc.  2016,  138  (6),  1772-­‐1775.   (31)  Wyttenbach,  T.;  Kemper,  P.  R.;  Bowers,  M.  T.  Design  of  a  New  Electrospray  Ion   Mobility  Mass  Spectrometer.  Int.  J.  Mass  Spectrom.  2001,  212  (1-­‐3),  13-­‐23.   (32)  Gidden,  J.;  Ferzoco,  A.;  Baker,  E.  S.;  Bowers,  M.  T.  Duplex  Formation  and  the   Onset  of  Helicity  in  Poly  D(Cg)(N)  Oligonucleotides  in  a  Solvent-­‐Free   Environment.  J.  Am.  Chem.  Soc.  2004,  126  (46),  15132-­‐15140.   (33)  Mason,  E.  A.;  McDaniel,  E.  W.,  Transport  Properties  of  Ions  in  Gases.  Wiley:  New   York,  1988;  p  xvi,  560  p.   (34)  Wyttenbach,  T.;  Kemper,  P.  R.;  Baykut,  G.;  Park,  M.  A.;  Bowers,  M.  T.  A  New   Instrument  with  High  Mass  and  High  Ion  Mobility  Resolution.  Int.  J.  Mass   Spectrom.  Submitted  2018.   (35)  Mathematica,  Version  11.2;  Wolfram  Research:  Champaign,  IL,  2017.   (36)  Silverman,  B.W.  Density  Estimation  for  Statistics  and  Data  Analysis.  In   Monographs  on  Statistics  and  Applied  Probability;  Chapman  and  Hall:  London,   U.K.  1986.   (37)  Dupuis,  N.  F.;  Wu,  C.;  Shea,  J.  E.;  Bowers,  M.  T.  Human  Islet  Amyloid  Polypeptide   Monomers  Form  Ordered  Beta-­‐Hairpins:  A  Possible  Direct  Amyloidogenic   Precursor.  J.  Am.  Chem.  Soc.  2009,  131  (51),  18283-­‐18292.   (38)  Mhin,  B.  J.;  Lee,  S.;  Cho,  S.  J.;  Lee,  K.;  Kim,  K.  S.  Zn(H2o)6(2+)  Is  Very  Stable   among  Aqua-­‐Zn(Ii)  Ions.  Chem.  Phys.  Lett.  1992,  197  (1-­‐2),  77-­‐80.   (39)  Trzaskowski,  B.;  Adamowicz,  L.;  Deymier,  P.  A.  A  Theoretical  Study  of  Zinc(Ii)   Interactions  with  Amino  Acid  Models  and  Peptide  Fragments.  JBIC,  J.  Biol.  Inorg.   Chem.  2008,  13  (1),  133-­‐137.   (40)  Keller,  B.  O.;  Suj,  J.;  Young,  A.  B.;  Whittal,  R.  M.  Interferences  and  Contaminants   Encountered  in  Modern  Mass  Spectrometry.  Anal.  Chim.  Acta  2008,  627  (1),   71-­‐81.   (41)  Tong,  H.;  Bell,  D.;  Tabei,  K.;  Siegel,  M.  M.  Automated  Data  Massaging,   Interpretation,  and  E-­‐Mailing  Modules  for  High  Throughput  Open  Access  Mass   Spectrometry.  J.  Am.  Soc.  Mass  Spectrom.  1999,  10  (11),  1174-­‐1187.   (42)  Dupuis,  N.  F.;  Wu,  C.;  Shea,  J.  E.;  Bowers,  M.  T.  The  Amyloid  Formation   Mechanism  in  Human  Iapp:  Dimers  Have  Beta-­‐Strand  Monomer-­‐Monomer   Interfaces.  J.  Am.  Chem.  Soc.  2011,  133  (19),  7240-­‐7243.   (43)  Mirecka,  E.  A.;  Feuerstein,  S.;  Gremer,  L.;  Schroder,  G.  F.;  Stoldt,  M.;  Willbold,  D.;   Hoyer,  W.  Beta-­‐Hairpin  of  Islet  Amyloid  Polypeptide  Bound  to  an  Aggregation   Inhibitor.  Sci.  Rep.  2016,  6.  

 

22   ACS Paragon Plus Environment

Page 23 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

(44)  Liu,  N.;  Duan,  M.  J.;  Yang,  M.  H.  Structural  Properties  of  Human  Iapp  Dimer  in   Membrane  Environment  Studied  by  All-­‐Atom  Molecular  Dynamics  Simulations.   Sci.  Rep.  2017,  7.    

 

23   ACS Paragon Plus Environment

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 24

TOC Graphic

 

24   ACS Paragon Plus Environment