Collagen Fibrils: Nature's Highly Tunable Nonlinear Springs - ACS

Subscriber access provided by - Access paid by the | UCSB Libraries

Collagen Fibrils: Nature’s Highly Tunable Nonlinear Springs Orestis G Andriotis, Sylvia Desissaire, and Philipp J Thurner ACS Nano, Just Accepted Manuscript • DOI: 10.1021/acsnano.8b00837 • Publication Date (Web): 12 Mar 2018 Downloaded from http://pubs.acs.org on March 13, 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 36 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

ACS Nano

Collagen Fibrils: Nature’s Highly Tunable Nonlinear Springs Orestis G. Andriotis†, Sylvia Desissaire† and Philipp J. Thurner† * †Institute of Lightweight Design and Structural Biomechanics, Vienna University of Technology, Getreidemarkt 9, 1060 Vienna, Austria *Correspondence to: Philipp J. Thurner Institute of Lightweight Design and Structural Biomechanics Vienna University of Technology Getreidemarkt 9 1060 Austria [email protected] Tel: +43 1 58801 31723 Fax: +43 1 58801 31799

ACS Paragon Plus Environment

1

ACS Nano 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 36

Abstract

Tissue hydration is well known to influence tissue mechanics and can be tuned via osmotic pressure. Collagen fibrils are nature’s nanoscale building blocks to achieve biomechanical function in a broad range of biological tissues and across many species. Intrafibrillar covalent cross-links have long been thought to play pivotal role for collagen fibril elasticity, but predominantly at large, far from physiological, strains. Performing nanotensile experiments of collagen fibrils at varying hydration levels by adjusting osmotic pressure in situ during atomic force microscopy experiments, we show the power the intrafibrillar non-covalent interactions have for defining collagen fibril tensile elasticity at low fibril strains. Nanomechanical tensile tests reveal that osmotic pressure increases collagen fibril stiffness up to 24-fold in transverse (nanoindentation) and up to 6-fold in the longitudinal direction (tension), compared to physiological saline in a reversible fashion. We attribute the stiffening to the density and strength of weak intermolecular forces tuned by hydration and hence collagen packing density. This reversible mechanism may be employed by cells to alter their mechanical microenvironment in a reversible manner. The mechanism could also be translated to tissue engineering approaches for customizing scaffold mechanics in spatially resolved fashion and it may help explain local mechanical changes during development of diseases and inflammation.

Keywords Collagen, hydration, osmotic pressure, mechanical properties, atomic force microscopy


2

Page 3 of 36 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

ACS Nano


3

ACS Nano 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 36

It is a well-accepted fact that tissue mechanics strongly depend on hydration. Consequently, factors affecting hydration in tissues namely osmolality, counter-ion condensation and osmotic pressure generated within tissues by proteoglycans and glycoproteins have been studied since the 1970s. While macroscopic mechanical properties and their dependence on hydration effects in the biomechanics community have been well-described,1-3 some of the underlying mechanisms remain unclear. Why is it in fact that tissues change their apparent mechanical properties, e.g. stiffness, depending on water content? Which level(s) of the tissue hierarchy are the structure and mechanics affected? Here we provide some insights to these questions investigating the basic structural building block of most tissues: the collagen fibril. Collagen fibrils are ultra-high aspect ratio rope-like structures, self-assembled from fibril-forming collagens, which are the most abundant molecules of the collagen protein family. Collagen fibrils are well preserved over a range of biological tissues and among a variety of species from the sea cucumber, 4 sea urchin 5 and teleost fish scales 6 to most loading-bearing tissues in vertebrae.7 This means, they are exposed to different environmental conditions and consequently mechanical requirements, dictating the need of mechanisms for adaption of mechanical competence to provide functionality to the supporting tissue. With time, collagen fibrils are chemically altered through enzymatic and non-enzymatic crosslinking. These modifications in living systems are normally slow taking weeks, months and years 8

resulting in irreversible covalent bonds between molecules i.e. the intermolecular cross-links. 9

To date, the general thought is that both enzymatic and non-enzymatic cross-links in collagen could explain changes in stiffness of the individual collagen fibril at physiologically relevant external deformations. This is based on studies at the tissue level of tendons and ligaments, where it has been deduced that changes in enzymatically-induced cross-linking 10 are responsible


4

Page 5 of 36 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

ACS Nano

for changes of tissue level elasticity. 11 This is an attractive theory but it neglects the multiple length scale levels, from the collagen fibril up to the tissue, that deformation mechanisms take place and contribute to the overall tissue mechanics. In addition, molecular dynamics simulations as well as experimental evidence suggest that naturally occurring enzymatic but also artificiallyinduced cross-links through glutaraldehyde treatment are mechanically activated from 5% to 20% of fibril extension. 12-14 This is far above from physiological regimes where the collagen fibril is suggested to extend up to 4%.15,16 Proteins, and in turn collagen fibrils, could also be considered as polyelectrolytes assuming the specific shape or folding as per their interaction with water and ions that are in solution. Therefore, important insights gained on polyectrolytes as well as on DNA 17 about such interactions and effects of counter-ion condensation 17 could also apply for collagen. More recently, Grant et al. showed an increase in the indentation modulus of about one order of magnitude (via AFM nanoindentation experiments) when exposing collagen fibrils from buffer to 1M KCl at pH7 as well as to increasing ionic strength. 18 Svensson et al. showed only a minor increase (~4%) of tensile elastic modulus when fibrils were exposed from buffer to 1M KCl at pH7. 19 The difference in magnitude of the effect between these two studies could partly be explained due to the different loading scenarios. Grant et al. conducted transverse indentation 18 and Svensson et al. tensile tests. 19 Given the structural anisotropy of collagen fibrils (transverse isotropic), the mechanical response of collagen fibrils to tension and transverse indentation is different and therefore also the response to changes in environmental conditions such as the ionic strength may be different. Since salts have therefore shown limited potential to influence collagen fibril mechanics, we hypothesize here that tuning of the mechanics of individual collagen fibrils can efficiently be


5

ACS Nano 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 36

achieved by adjusting fibril hydration through exerting osmotic pressure. In nature, osmotic pressure is achieved by co-localization of molecules with high negative charge densities, such as glycosaminoglycans (GAGs), proteoglycans (PGs) and lipids. 20 Compared to irreversible covalent bonds, the formation of non-covalent long-range interactions is achieved at shorter timescales, similar to the short time-life of mechanical signaling in cells, which happens over several microseconds.21,22 A number of studies have so far focused in the mechanics of individual collagen fibrils using mostly atomic force microscopy 14, 18,19, 23-33 as well as other devices.34-36 Many of these investigations were performed in dry environment 24, 27, 30, 34, 37 or in indentation mode, 18, 23-28, 30 both of which do not constitute physiological conditions, where tensile loads would mostly be expected. More recent studies investigated collagen fibrils under tensile loads in a hydrated environment,14, 19, 32, 36 and have delivered important insights investigating e.g. the influence of pH and salt concentration.19 To establish an understanding of the structural and mechanical effects of extrafibrillar osmotic pressure at the level of individual collagen fibrils, we further investigated the effect of hydration on the tensile properties of individual collagen fibrils, within a physiological strain regime. To achieve this, we conducted atomic force microscopy experiments on individual collagen fibrils at varying osmotic pressures in a physiological tensile loading case. Within this manuscript, we provide data detailing the achievable magnitude of structural and mechanical changes as well as an interpretation and explanation of the mechanism behind the effect seen.


6

Page 7 of 36 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

ACS Nano

Results and Discussion Collagen shrinking and stiffening. Hydration is well known to affect the mechanical properties of biological tissues.38-41 This is also true for collagen fibrils 18, 23, 26 but so far the mechanism of how exactly hydration influences hydration remain unknown. To experimentally test this, native collagen fibrils were fully submerged in an open fluid cell allowing experiments in aqueous solution (Figure S1) and were exposed to varying concentrations of a short chain poly-ethylene-glycol (PEG, MW=200 Dalton). Figure 1 shows consecutive atomic force microscopy (AFM) images of a native collagen fibril first air-dried, then hydrated in phosphate buffered saline (PBS, 10mM) and partially dehydrated upon increasing PEG concentration in a stepwise fashion. AFM images show that collagen fibrils are highly hygroscopic and the diameter of individual collagen fibrils increases by a factor of 1.5 – 2, when going from air-dried to a fully hydrated state in phosphate buffered saline (PBS) at pH 7.5 (figure 1a, see also figure S2). Gradual addition of the PEG solute, which cannot enter the intrafibrillar space, results in a concentration gradient of water. This drives water from the intrafibrillar to the extrafibrillar space which balances the concentration of water. This is counteracted by the affinity of collagen to water and eventually results in partial dehydration, i.e. shrinking, of the fibrils (figure 1b). Compared to other macromolecules collagens possess many donors and acceptor sites for Hbond formation. 42 At the dried state only structured water is tightly bound to the collagen molecule.43 But, water incorporates into the fibrilar structure when the collagen fibrils are submerged in an aqueous solution. Because water is polar with a high dielectric constant, it reduces the free energy between adjacent molecules resulting in an increase of the intermolecular


7

ACS Nano 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 36

distance 44 This interaction of collagen molecules with water explains the swelling of collagen fibrils. The reduction of the fibril diameter in 3.5M solution of PEG reached about 15% from the fully hydrated state. As such the ratio of volume of water to volume of collagen molecules within the fibril drops from about 1.6 to 0.9, which is a geometrical change in cross-section area of about 28% (figure 1c). Our collagen fibril shrinking results are well aligned with previous studies that measured Bragg spacing of collagen molecules upon addition of PEG.3, 45 Masic and coworkers reported up to 20% reduction in the lateral spacing between collagen molecules with relative changes in humidity from 95% to a completely dried state at 0%.3 Similarly Price et al. found that the intermolecular distance increased by 1.35-fold upon hydration based on X-ray diffraction data of tendon.46 What is striking from these and our results, is how much water fully hydrated collagen fibrils contain. Swelling by up to factor of 2 is common in collagen fibrils, which means that a fully hydrated fibril contains about 3 times more water than collagen ground substance, in a volumetric sense. Considering the volumetric changes of collagen fibrils resulting from osmotic pressure in our experiments, we expected a substantial effect. To probe this we first applied indentation-type AFM to collagen fibrils. Their diameters decrease linearly with PEG concentration, i.e. osmotic pressure, reaching a plateau at about 2.6M. Similarly, indentation modulus rises with increasing PEG concentration reaching a value of about 360 MPa at a PEG concentration of 3.5M (figure 1d), which is a 24-fold increase compared to the hydrated state in PBS with indentation modulus of about 15 MPa (which is the average modulus in PBS, 10mM and 1M PEG). Exposing the fibrils back to PBS after PEG exposure shows that rehydration is reversible and both height and indentation modulus return to their original values (Figure S3 and S4 in supporting information).


8

Page 9 of 36 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

ACS Nano

The magnitude of our indentation modulus results are corroborated by Grant et al. work where the authors documented an increase up to ~200 MPa of fibrils in ethanol.18 Beyond the effect of dehydration due to a water gradient presented here, pH and salts have also previously been shown to contribute to the mechanical behavior of collagen fibrils: With no salt added, Grant et al. measured a decrease in indentation modulus with increasing pH from 5 to 7, from 3.1 MPa to about 2 MPa, respectively.18 In our first trial, addition of PEG in the PBS solution resulted in an increase of pH from 7.5 (no PEG) to 8.8 (3.5M PEG). According to the findings by Grant et al., we would expect the increase in pH to contribute to lowering the indentation modulus. However, our results of overall 24-fold increase in indentation modulus suggest that the effect of dehydration prevails the effect of change in pH. In a second trial, we adjusted the pH to 7.5 and measured the height and indentation modulus of three collagen fibrils in PBS and in PBS with 3.5M PEG. We found that the height decreased by an average of 14% while the indentation modulus increased by an average of 36-fold, which confirms that the effect of dehydration is stronger than the effect of pH on the mechanics of collagen fibrils. With regards to salt, Grant et al. also measured an increase in indentation modulus with increasing ionic strength.18 Salts solutes may play a lesser role in our experiments but they would also condensate partly with PEG. We expect this to reduce osmotic pressure and leads to a change in salt concentration within the buffer and a change in pH at a certain PEG concentration (pH increased up to 8.8 with increasing concentration of PEG). Salt ions are small enough to enter collagen hence there is no boundary for creating osmotic pressure. Given the increasing pH with increasing concentration of PEG, counter-ion condensation could happen with the H+ or H3O+ ions, given a pH rise of about 1 we will have ten times more of the H+ than OH groups adhering to PEG.


9

ACS Nano 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 36

Tuning collagen pulling elasticity. Due to their rope-like structure and high structural anisotropy, collagen fibrils mostly provide tensile stiffness to tissues, as in ligaments and tendons but also in blood vessels and the airways. Hence, tensile loading cases are thought to better represent physiology and also the way cells feel forces imposed on these tissues. To investigate the effect of osmotic pressure in a tensile loading configuration we adapted the protocol (Figures S5, S6 and S7 in supporting information) by Svensson et al.. 14 After affixing collagen fibrils between a glass substrate (for details see Experimental Methods) and an AFM cantilever, repeated tensile tests were carried out in PBS, again with increasing concentrations of PEG to simulate the presence of proteoglycans. In these experiments we applied tensile forces in the order of hundreds of nN, similar to the scale of cell traction forces during ECM-cell interaction 47 as well as the magnitude of forces needed to activate cell mechanotransduction responses.48,49 Further, the corresponding fibril strains were up to 5% which is well within the physiological fibril strain range.50 Resulting force versus fibril displacement curves for a fibril are summarized in figure 2. Generally, the force increases non-linearly with displacement in the fully hydrated state and up to 1M PEG (figure 2b). Due to the material non-linearity both stiffness (slope of forcedisplacement curve) and tensile modulus (slope of stress-strain curve) are functions of strain and increase non-linearly with fibril strain (figure 2c-2f). There is up to a 6-fold increase (at 1.5% strain) in both the stiffness and the tensile modulus from the fully hydrated state in PBS to the partially dehydrated state in 2.6M PEG (figure 2e and 2f, respectively). From the stiffness and tensile modulus data (figure 2c-f) the evolution of stiffness changes qualitatively with increasing osmotic pressure between 0% and 1.5% strain, which may be explained by activation of different deformation pathways.


10

Page 11 of 36 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

ACS Nano

As mentioned above, the inclusion of PEG, results in partial dehydration of the fibrils, due to a concentration gradient of water, leading to a substantial increase in stiffness at the collagen fibril level. A recent study at the tissue level mechanics showed that the reduced GAG content resulted in reduced elasticity of the tissue matrix.51 Although the latter study was performed at the tissue level it well corroborates with our results of reduced stiffness and tensile modulus of individual collagen fibrils exposed to lower concentration or without PEG. The concentration of GAGs increases from tendon 52 to cartilage 53 as well as due to inflammation.54 The swelling pressure of tendon and cartilage is estimated to be 0.4 MPa3 to about 2 MPa,55 respectively. Given a back of the envelope calculation based on data found online (https://www.brocku.ca/researchers/peter_rand/osmotic/osfile.html; last accessed on 29 January 2018) and the fact that the concentration of GAGs in the cornea is higher than in cartilage 56 we could expect that the range from 0.5M PEG up to about 2M concentration of PEG may represent physiological variations in nanoscale tissue hydration. We expect this range to shift to higher PEG concentrations to represent the changes in nanoscale tissue hydration due to overexpression of GAGs in inflammation.54 Notably, the hydration of collagen is affected already at the loest concentration of PEG in our experiments as can be seen from the observable change in both the structural stiffness as well as the elastic modulus under tension. From the range of physiological osmotic pressures mentioned above, our results suggest that physiologically relevant changes may result in more than 2-fold increase in stiffness / elastic modulus. An even higher increase may be present in inflammation due to largely increase expression of GAGs. Collagen fibrilar and subfibrilar deformation pathways. The mechanical function of biological tissues is related to their hierarchical architecture, which extends across various length scales. Also collagen fibrils contain smaller structural elements


11

ACS Nano 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 36

such that their mechanical properties and deformation pathways depend on smaller length scale levels. The dependence of stiffness and tensile modulus on osmotic pressure reflects such a structure-material relationship. The basic structural unit of a fibril, the tropocollagen molecule, is a coiled-coil composed of three polypeptide alpha (α) chains 57 into a ~300 nm long and ~1.5 nm wide triple helix. Glycine (Gly) residues are found at every third position along the alpha chains, resulting in the repeating motif Gly-X-Y. At positions X and Y are positions of other bulkier amino acids, mostly (but not only) proline or hydroxyproline residues, respectively. The high abundance of the small Gly residues and their location towards the center of the collagen molecule promote the tight packing of the triple helix stabilized by inter-chain H-bonds.58 Studying molecular level deformation pathways of collagen has largely been accomplished using molecular dynamics. However, computational capacity is generally limiting the size of peptides that are studied to tens of amino acids, 59 whereas naturally collagen molecules are composed of ~1000 amino acids. Nevertheless, insights can be used to interpret our experimental results: the collagen molecule deforms because of its entropic (at low strains) and progressively energetic (higher strains) elasticity. When tested isolated under tension, the collagen molecule undergoes molecular rotation, straightening of the molecular kink formations 60,61 and molecular uncoiling 62,63

at strains lower than 5%. Based on persistence length measurements, the Young’s modulus

of a collagen molecule is one order of magnitude greater (~1-5 GPa) 64 than the tensile elastic modulus of the fibril. This infers that during fibril pulling the collagen molecules deform less, i.e. are less strained, than the collagen fibril. Because of the radial packing of collagen fibrils,65 the cross section of a 153 nm wide fibril (in air) consists of about 8895-9364 molecules. With a maximum force of 1.2 µN (figure 2b) at the fibril level, a maximum force of ~135 pN per collagen molecule (for 8895 molecules) is estimated to be applied, which is beyond the 30 pN


12

Page 13 of 36 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

ACS Nano

force inducing molecular uncoiling.63 Using the same number of molecules, the threshold for molecular uncoiling would be around 2% to 3% strain at the fully hydrated state in PBS. This threshold then gradually decreases down to 1% (2.6M PEG) with increasing osmotic pressure (figure 2). Therefore, the strain needed to initiate and propagate such deformation mechanisms is strongly related to the extent of dehydration induced by osmotic pressure. In a fully hydrated fibril the collagen molecules are further apart, with fewer and less strong non-covalent interactions and thus more freely to deform during pulling. As fibril dehydration increases with PEG concentration the average intermolecular distance decreases and the number and strength of non-covalent interactions increase. One could then envisage that stronger weak interactions (vdW, H-bonds, ionic bonds) a) lead to a quicker rise in force during tensile loading and b) hinder the molecular deformation mechanisms such that higher forces are needed for molecular rotation, uncoiling and straightening. This can explain why the stiffness vs. strain curve of a fibril is changing not only quantitatively but also qualitatively, with increasing osmotic pressure (figure 2c-f). Collagen time-dependent mechanics. As collagen fibrils are mechanically viscoelastic, increasing the loading speed directly affects their mechanical behavior. To test the time-dependent mechanics, we performed tensile force experiments at increasing loading speeds from 0.5 µms-1 to ~99.2 µms-1 (strain rates ranging from ~0.4%s-1 to ~78%s-1) measured in four environmental conditions with increasing osmotic pressure (Figure S8). Increasing loading speeds generally lead to higher forces needed to deform the fibril at a given strain. This is enhanced with osmotic pressure as depicted in Figure 3 in form of the stiffness and elastic modulus dependence on loading speed in PBS and 2.6M PEG. When hydrated in PBS, the fibril stiffness and elastic modulus highly depends on the fibril strain, as


13

ACS Nano 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 36

already shown above (figure 2c-f). The change in stiffness and elastic modulus with loading speed becomes greater when the fibril is exposed to 2.6M PEG (Figures S5). This infers that certain deformation mechanisms usually initiated and occurring at lower forces, such as molecular rotation, uncoiling and intermolecular sliding, are hindered by deformation mechanisms occurring at higher forces such as molecular stretching. Therefore one observes “stiffer” collagen fibrils at higher osmotic pressures and higher strain rates. This is consistent with findings from molecular dynamics simulations. When a molecule is pulled out of well hexagonally-ordered molecular arrangement, molecular sliding is hindered by a slip-stick motion due to intermolecular interlocking 66. Although this represents a simplified structural scenario, this effect was shown to decrease with increasing intermolecular distance.

Conclusions Nature’s finest structural tunable springs. The importance of osmotic pressure on the mechanical properties of connective tissue has been known since the 70’s.20 It is well known that osmotic pressure contributes to normal cartilage function while the collagen fibril network is competing this swelling pressure.55 Until now a mechanistic explanation, why and how the mechanics of individual collagen fibrils (under tensile load) are affected upon hydration is missing. Here we show how collagen fibrils, the smallest discernible structural elements of many connective tissues, respond to osmotic pressure, induced here by PEG. This response shows that the elastic properties of collagen fibrils within physiologically applied loads can be tuned by changing its chemical environment. Physically, this tuning is accomplished by adjusting the number and interaction strength of non-covalent intrafibrillar interactions. Figure 4 shows a simplistic interpretation of this mechanism in terms


14

Page 15 of 36 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

ACS Nano

of a non-covalent Lennard-Jones potential, describing the interaction between neighboring tropocollagen molecules. Our interpretation is based on similar effects on DNA.67,68 Upon partial dehydration at higher osmotic pressures the distance between tropocollagen molecules is decreased resulting in an increase of the potential-well-depth and a decrease of the equilibrium intermolecular distance. Our interpretation is that this subsequently results in higher potential gradients and even a higher number of interactions. As a result, higher forces are required to separate the molecules upon tensile loading. Such increase of electrostatic and/or Van der Waals interactions through osmotic pressure may be hindering deformation pathways that are otherwise present in the fully hydrated collagen fibril. It should be noted that in this study we have used samples from 6 months old mice. While these mice would be considered to be of mature age, collagen cross-linking is also known to be tissue specific.9 As reported by Svensson et al. murine tail tendons showed fewer mature cross-links compared to human patellar tendon.14 In this context, whether collagen fibrils with different/higher cross-linking are similarly susceptible to stiffness tuning via osmotic pressure remains to be shown. As discussed above, physiologically relevant changes in hydration are expected to be represented by our results between 0.5M to 2M PEG concentration and even higher in cases of inflammation. Putting our results into a translational clinical context it is notable that pathologies such as arthritis 69 or fibrosis 70 have been related to alterations in amount and type of osmotic pressure inducing peptides, such as proteoglycans and glycosaminoglycans. It may well be that changes in mechanical properties of tissues due to pathology result from increased expression of proteoglycans and/or glycoproteins influencing collagen mechanics in a swift manner. To test such hypotheses further research is required. In a first step a repeat of measurements with proteoglycans and glycosaminoglycans can be envisioned, which is beyond the scope of this


15

ACS Nano 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 36

study. Furthermore, an application of the effect found may be within tissue engineering, where the importance of mechanosensing and mechanoadaptation of cells is well known.71 Conceptually, a mixture of hydrated fibers (e.g. collagen fibrils) in combination with PEG could be used to 3D print scaffolds with spatially resolved mechanical properties.


16

Page 17 of 36 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

ACS Nano

Figure Legends Figure 1. Collagen fibril shrinking and stiffening with increasing osmotic pressure. (a) AFM height topography images of a collagen fibril in air-dried state (only bound water present), fully hydrated in phosphate buffered saline (PBS) and, subsequently, partly dehydrated in increasing concentrations of polyethylene glycol i.e. increasing osmotic pressure. (b) Corresponding height profiles of the same fibril in different aqueous solutions (colored line) and in air (black dashed line). (c) Data from two selected fibrils showing height decrease by 20% from fully hydrated state to highest concentration of PEG. (d) The indentation modulus increases from about 15 MPa in fully hydrated state to about 370 MPa at the highest concentration of PEG. Error bars in panel (c) and (f) represent the standard deviation of the shrinking taken from 150 points across the fibril crest. Error bars in panel (d) represent the standard deviation of indentation modulus across the fibril at a given PEG concentration.

Figure 2. Native collagen fibril under tension with osmotic pressure. (a) Illustration of a native collagen fibril under tensile loading. (b) Six force-displacement (fibril displacement) curves per solution for a 43 µm long and 230 nm wide (PBS) collagen fibril during gradual dehydration from phosphate buffer saline to 3.5M of PEG solutions. (c-f) Tensile structural stiffness and elastic modulus plotted against fibril strain (small strain approximation). (d, f) Comparison of structural stiffness and elastic modulus of the collagen fibril measured in PBS and 2.6M PEG. (cf) The error bars represent the standard deviation at the given fibril strain from six forcedisplacement curves.


17

ACS Nano 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 36

Figure 3. Collagen fibril stiffness and tensile modulus as a function of loading speed and fibril strain. Three dimensional illustrations show the dependence of collagen fibril (128 µm in length) stiffness (a) and elastic modulus (b) on both loading speed (ranging from 0.5 µm s-1 to about 99.2 µm s-1) and fibril strain in PBS and 2.6M PEG(semi-transparent). At lower fibril strains the changes in stiffness with displacement rate are small. However, at larger fibril strains the effect of displacement rate is more pronounced and it is further enhanced with increased osmotic pressure (c, d). The error bars represent the standard deviation of six force displacement curves. Figure 4. Physical properties of collagen fibrils are tunable via osmotic pressure. At low osmotic pressure the equilibrium intermolecular (interaxial) distance is larger compared to the one at higher osmotic pressure. Decreasing the intermolecular distance, the density of non-covalent interactions increases due to the increased packing density of collagen molecules at higher osmotic pressures. This results in a shift of the Lennard-Jones potential towards the situation with no water present, requiring higher forces for intermolecular movement such as lateral compression or sliding during indentation and tensile loading.


18

Page 19 of 36 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

ACS Nano

EXPERIMENTAL METHODS Chemicals Unless otherwise specified, all chemicals were purchased from Sigma Aldrich. Phosphate buffered saline was purchased in the form of tablets that were diluted in distilled water according to manufacturer’s instructions.

Collagen samples Collagen fibril samples were collected from tail-tendons of two wild type mice (male, 6 months old) similarly to a recent study,24 a tendon section is harvested with tweezers and a scalpel, immersed into 10mM phosphate buffered saline (PBS, pH7.5). A fascicle is pulled from the tendon section under a stereomicroscope (SZX10, Olympus) and with the use of sharp tweezers the fascicle is unwound while hydrated and spread onto a poly-l-lysine coated glass slide surface (Superfrost® Plus, Thermo Scientific) exposing individual collagen fibrils. The samples are then dried in air and stored in a desiccator until further use.

Atomic force microscopy Atomic force microscopy (AFM) experiments were conducted on a NanoWizard® Ultra Speed A (JPK Instruments, Berlin) mounted on an inverted optical microscope (Observer. D1, Zeiss). A polymethylmethacrylate ring of 2 cm inner diameter and 0.5 cm height was fixed with dental silicon (picodent twinsil®, picodent) on the glass slide enclosing the attached collagen fibrils allowing experiments in aqueous solution (Figure S1). Silicon nitride rectangular cantilevers (PNP-DB, Nanoworld, Switzerland) of 0.48 Nm-1 nominal spring constant with a 4-


19

ACS Nano 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 36

sided (quadratic tip) pyrex nitride probe of 10 nm nominal radius, were used for imaging in air and aqueous solution. The fibrils were first imaged in air, then in PBS and subsequently in PEG solutions with increasing concentration (0.1M, 0.5M, 1.0M, 1.5, 2M, 2.3M, 2.6M and 3.5M) based on repeated measures approach i.e. the exact same fibrils were tested in each solution. Each solution was replaced by removing the existing and adding the next one. Each solution was removed or added with a new pipette tip. This process was repeated 4-5 times for every replacement to minimize the amount of remnant solution that was replaced. The fibrils were left for 1.5 to equilibrate in each solution. In air, an overview region is imaged that includes individual collagen fibrils lying on the glass slide. Then the fibrils are imaged separately within a region of interest (ROI; 2.5 µm x 2.5 µm). Air-dried collagen fibrils were imaged in contact mode at 512 x 512 pixel resolution at 1 Hz scanning rate and hydrated collagen fibrils, either in PBS or PEG solutions, were imaged in Quantitatite Imaging mode (QITM, JPK Instruments) at 256 x 256 pixel resolution over a 2.5 µm x 2.5 µm area. During QITM mode the cantilever performs force-indentation curves at every pixel (i.e. in total 65536 force-indentation curves for 256 x 256 pixel resolution) minimizing the lateral forces from the tip onto the collagen fibrils. For every force-indentation curve the contact height between the tip and the sample is calculated at zero force and a height image is then constructed. This eliminates imaging artefacts originating from lateral forces during contact or tapping mode imaging between the tip and the soft sample and QITM mode increases the resolution and quality of height measurements in fluid. In the first trial the pH was not adjusted, but as we show further below in a second trial, adjusting the pH for the 3.5M of PEG results in the same exact effect of shrinking and indentation modulus stiffening of the collagen fibrils. This indicates that osmotic pressure due to


20

Page 21 of 36 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

ACS Nano

the concentration gradient of water is responsible for the structural and mechanical changes of collagen fibrils we observe in this study.

Swelling and shrinking measurements of collagen fibrils Height topography images were recorded from eight fibrils (four from each mouse) in air dried, in fully hydrated state (1mM PBS, pH7.5) as well as in eight different concentrations ranging from 0.1M to 3.5M of polyethylene glycol (average molecular weight 200 Da, SigmaAldrich). The height profiles of each fibril were compared between the different environments to calculate swelling and shrinking ratios. Assuming a circular cross-section for collagen fibrils, as it appears on transmission electron microscopy images,72 we chose height measurements instead of cross-sectional area from height profiles. Nevertheless, we also report stiffness data, to eliminate related errors from height measurements. Swelling ratio was defined fold increase in height when the fibril is fully hydrated in PBS compared to the air-dried state (HPBS/HAir ratio) and the shrinking was defined fold decrease in height when the fibril was exposed at a given concentration of PEG compared to the fully hydrated state in PBS (HPEG/HPBS ratio). The absolute height of collagen fibrils in different environments are shown in Figure S2. A separate experiment was performed to validate the rehydration of collagen fibrils. Three fibrils were scanned initially in air and then exposed to PBS, 3.5M PEG and finally rehydrated in PBS. The exact same fibrils were scanned in QITM mode in each solution. The pH of the PEG solution was adjusted to pH7.5 in all cases. To minimize the time going from one solution to the next, each solution was replaced by the next one using a pipette, i.e. the first PBS by 3.5M PEG and the latter by PBS again, without removing the glass slide from the AFM stage to detect the fibrils in each solution. Figure S3 shows height topography images in air, PBS and 3.5M PEG


21

ACS Nano 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 36

along with representative cross sectional height profiles. Figure S4 shows the fibril height and indentation modulus in each solution. The substitution of PEG by PBS at the rehydration step resulted in some remnant of PEG, which explains why the height and indentation modulus were not fully recovered. Nevertheless, Figure S4 shows that both the height and transverse stiffness are reversible with rehydration taking place within a relatively short time, i.e. 60 minutes.

AFM cantilever-based nanoindentation of individual collagen fibrils AFM was employed to perform both indentation (transverse) as well as tensile mechanical tests on individual collagen fibrils. The AFM cantilever is used as the load sensor in both indentation and tensile tests. Prior to mechanical tests all AFM cantilevers were calibrated for the inverse optical lever sensitivity in all environments by performing force-displacement curves on the glass surface. With the cantilever spring constant (kAFM) and its deflection (DAFM) known, the external force applied on the fibril can be determined by Force (nN) = kAFM (nN/nm) × DAFM (nm). In addition, the true fibril indentation and pulling displacement is found as the difference between the overall z-displacement (z-piezo displacement that drives the cantilever in the z-axis) and cantilever deflection (DAFM). AFM cantilever-based indentation employed to measure the transverse stiffness, similar to a recent study24 by indenting the surface of the fibril (Figure. S1d). Indentation measurements were load controlled at 2 nN and 18 µm/s cantilever displacement rate. Force curves from transverse indentation tests were analyzed with the Hertzian model assuming a quadratic (4-sided pyramid) AFM stylus and fitting the loading force-indentation (F-d) data with the function: = tan ( ) √2(1 − )

(1)


22

Page 23 of 36 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

ACS Nano

where, F is the applied external force (nN), α is the face angle of the four-sided pyramid, equal to 35 degrees, Ecol is the collagen fibril indentation modulus (MPa) and is used here as a fitting parameter, d is the indentation depth (nm) and v the Poisson’s ratio of the collagen fibril assumed to be 0.5. The resulting indentation moduli data carry quantitative errors due to the contact assumptions as well as the material properties assumptions, as previously discussed.24 However, we are able to compare the indentation moduli by employing the same AFM cantilever for a given collagen fibril tested under the different environmental conditions i.e. increasing concentrations of PEG.

Preparing collagen fibrils for AFM-based tensile tests Tensile experiments on individual collagen fibrils have been performed via microelectromechanical systems (MEMS) 34, 36 as well as atomic force microscopy (AFM). 14, 29, 32

Here, the collagen fibril is fixed between the glass slide and the AFM cantilever with micro-

sized epoxy droplets (2 components epoxy, UHU® Plus ENDFEST 300). The process of fixing is divided into 2 phases. At phase 1, two epoxy droplets are deposited onto two ends of the collagen fibril segment to be tested and are left to harden for at least 12 hours. At phase two one epoxy droplet is detached from the glass slide and fixed onto the AFM cantilever also using epoxy. Figure S5 shows the first three steps of phase 1 while Figure S6 and S8 illustrate the steps of phase 2. At the first step of phase 1, the collagen fibril is identified under an optical microscope. An AFM image is recorded to ensure that the sample in question is indeed an individual fibril rather than a fibril bundle or multiple fibrils next to each other. Then the two components of the epoxy are


23

ACS Nano 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 36

well mixed in a plastic tray. The tip of sharp tweezers is dipped into the epoxy glue creating a very small droplet of epoxy. This droplet is then deposited on the glass slide where the sample is located but away from the collagen fibrils as to avoid contamination of the area. At the second step of phase 1, the AFM cantilever is approached until in contact with the edge of the epoxy droplet and in that way a smaller droplet is adhered onto the edge of the tipless cantilever, while the epoxy still has a low viscosity . At the third step of phase 1, the cantilever is approached onto two locations on the collagen fibril, at about 100 µm contour length, depositing two micro-sized epoxy droplets. After overnight hardening of the epoxy droplets, we move to phase 2 of the fixing process. At the first step of phase 2, a stiff AFM cantilever (HQ:NSC15 silicon nitride probes of 20 Nm-1 to 40 Nm-1 and 10 nm tip radius, uMasch) is used to detach one of the micro-sized epoxy droplets from the glass slide by first indenting the edge of the epoxy located away from the gauge length (panel A Figure S5) until the epoxy doplet is substantially deformed. This helps the next step, i.e. to detach the fibril. The deformed epoxy is then imaged in contact mode using high line rates (10 Hz) and in direction of the long axis of the AFM cantilever pushing it towards the gauge length. This is a critical step and the AFM cantilever end must only image the very edge of the epoxy. The epoxy will substantially deform at this point. The AFM cantilever is then repositioned closer to the epoxy and another contact mode image is performed. Note that the contact mode image aims only to use the tipless AFM cantilever as a spatula to help detach the epoxy from the glass slide. After about two to three steps the epoxy will be removed and ready to be fixed on the end of the tipless AFM cantilever. Prior to the final step of fixing, the AFM cantilever is calibrated similarly as for the indentation measurements, i.e. the optical lever sensitivity is measured by performing force curves on the stiff glass slide and the spring constant is determined via the


24

Page 25 of 36 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

ACS Nano

thermal noise method. Finally, a second epoxy droplet is prepared and deposited onto the glass slide (Figure S6). The AFM cantilever is approached on the edge of the epoxy. Now the final critical step is to reduce the size of the epoxy on the AFM cantilever by performing subsequent approaches on clean spots on the glass slide. After two to three approaches on the glass slide, the AFM cantilever is approached on the detached end of the fibril and let to harden and set for about 12 hours. The waiting time depends on the setting period of the epoxy used.

Tensile tests of individual collagen fibrils Tensile tests were performed on the exact same fibrils that were assessed previously for swelling upon hydration, shrinking and stiffening upon dehydration with PEG. Tensile tests of native collagen fibrils were performed by fixing the collagen fibril between the glass slide and the edge of a stiff tipless cantilever (All-In-One, Budget Sensors 40 Nm-1 nominal spring constant) with micro-sized epoxy droplets (~ 50 µm in diameter droplets) as described in the experimental methods above. For reference, the gauge length, L0, of the fibril is initially determined with optical microscopy. The reference measurement helps the detaching process and to define the cantilever position from the bottom of the glass slide without damaging the fibril by measuring the extension of the stepper motors of the AFM head system. This process is performed in small steps while the fibrils is hydrated to avoid damage of the fibrils. First the cantilever is retracted by a 3 µm step and then moved towards the epoxy droplet that is fixed to the glass slide, avoiding crashing the fibril between the epoxy droplet and the glass slide. Once the cantilever is close to the epoxy droplet mounted on the glass slide, the cantilever is moved upwards at small steps of several 10 µm and then of 2 µm until the cantilever force reaches 50 nN (supplementary Movie S1). Repositioning the XY motorized stage also with small steps, while recoding the


25

ACS Nano 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 26 of 36

cantilever deflection signal helps further align the one end of the collagen fibril attached to the glass slide with that attached on the end of the cantilever. Due to instrument limitations the tensile tests were performed by controlling the displacement of the Z-sensor at 0.5 µm s-1 displacement rate. The initial gauge length of the fibril was determined by measuring the distance between the glass slide and the upper point of preloading at 50 nN. The tensile stiffness, S (Nm-1), is calculated by linear fitting of equally-spaced segments of the force-displacement curves. Knowing the cross-sectional area, , of the collagen fibrils at different environmental conditions and by using the small strain approximation we calculated the tensile elastic modulus, E (GPa), from the equation: = ( )

(2)

For displacement rate dependent experiments the displacement rate ranges from 0.5 – 100 µm s-1. Tensile force versus fibril strain for increasing loading speed are shown in Figure S8. Forcedisplacement data analysis for longitudinal stiffness measurements was performed using a custom Matlab script (Matlab, 2015b, The MathWorks, Inc.).

Supporting Information The supporting information includes a PDF file with supplementary figures and a movie.


26

Page 27 of 36 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

ACS Nano

Acknowledgments All authors contributed in writing the manuscript. OGA and PJT conceived the hypothesis and designed the experiments. OGA and SD performed the experiments and data analysis. All authors have given approval to the final version of the manuscript. All data are archived at the server of the Institute of Lightweight Design and Structural Biomechanics and are available upon request to the corresponding author. We gratefully like to acknowledge donation of murine tails by Peter Pietschmann, Medical University of Vienna and thank Rene Svensson for fruitful discussions. The authors declare no conflicts of interest related to the research and results in this manuscript.


27

ACS Nano 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 28 of 36

References 1. Sedlin, E. D.; Hirsch, C., Factors Affecting the Determination of the Physical Properties of Femoral Cortical Bone. Acta Orthop. Scand. 1966, 37, 29-48. 2. Jameson, M.; Hood, J.; Tidmarsh, B., The Effects of Dehydration and Rehydration on Some Mechanical Properties of Human Dentine. J. Biomech. 1993, 26, 1055-1065. 3. Masic, A.; Bertinetti, L.; Schuetz, R.; Chang, S.-W.; Metzger, T. H.; Buehler, M. J.; Fratzl, P., Osmotic Pressure Induced Tensile Forces in Tendon Collagen. Nat. Commun. 2015, 6. 4. Motokawa, T.; Tsuchi, A., Dynamic Mechanical Properties of Body-Wall Dermis in Various Mechanical States and Their Implications for the Behavior of Sea Cucumbers. Biol. Bull. 2003, 205, 261-275. 5. Benedetto, C. D.; Barbaglio, A.; Martinello, T.; Alongi, V.; Fassini, D.; Cullorà, E.; Patruno, M.; Bonasoro, F.; Barbosa, M. A.; Carnevali, M. D. C., Production, Characterization and Biocompatibility of Marine Collagen Matrices from an Alternative and Sustainable Source: The Sea Urchin Paracentrotus Lividus. Mar. Drugs 2014, 12, 4912-4933. 6. Gistelinck, C.; Gioia, R.; Gagliardi, A.; Tonelli, F.; Marchese, L.; Bianchi, L.; Landi, C.; Bini, L.; Huysseune, A.; Witten, P., Zebrafish Collagen Type I: Molecular and Biochemical Characterization of the Major Structural Protein in Bone and Skin. Sci. Rep. 2016, 6. 7. Myllyharju, J.; Kivirikko, K. I., Collagens, Modifying Enzymes and Their Mutations in Humans, Flies and Worms. Trends Genet. 2004, 20, 33-43. 8. Jackson, S.; Eckford, S.; Abrams, P.; Avery, N.; Tarlton, J.; Bailey, A., Changes in Metabolism of Collagen in Genitourinary Prolapse. The Lancet 1996, 347, 1658-1661. 9. Eyre, D. R.; Paz, M. A.; Gallop, P. M., Cross-Linking in Collagen and Elastin. Annu. Rev. Biochem. 1984, 53, 717-748. 10. Avery, N.; Bailey, A., Enzymic and Non‐Enzymic Cross‐Linking Mechanisms in Relation to Turnover of Collagen: Relevance to Aging and Exercise. Scand. J. Med. Sci. Sports 2005, 15, 231-240. 11. Torp, S.; Arridge, R.; Armeniades, C.; Baer, E., Structure-Property Relationships in Tendon as a Function of Age. Proc. Colston Conf 1975, 26, 197-221. 12. Depalle, B.; Qin, Z.; Shefelbine, S. J.; Buehler, M. J., Influence of Cross-Link Structure, Density and Mechanical Properties in the Mesoscale Deformation Mechanisms of Collagen Fibrils. J. Mech. Behav. Biomed. Mater. 2014. 13. Hansen, P.; Hassenkam, T.; Svensson, R. B.; Aagaard, P.; Trappe, T.; Haraldsson, B. T.; Kjaer, M.; Magnusson, P., Glutaraldehyde Cross-Linking of Tendon—Mechanical Effects at the Level of the Tendon Fascicle and Fibril. Connect. Tissue Res. 2009, 50, 211-222. 14. Svensson, R. B.; Mulder, H.; Kovanen, V.; Magnusson, S. P., Fracture Mechanics of Collagen Fibrils: Influence of Natural Cross-Links. Biophys. J. 2013, 104, 2476-2484. 15. Mo, J.; Prévost, S. F.; Blowes, L. M.; Egertová, M.; Terrill, N. J.; Wang, W.; Elphick, M. R.; Gupta, H. S., Interfibrillar Stiffening of Echinoderm Mutable Collagenous Tissue Demonstrated at the Nanoscale. Proc. Natl. Acad. Sci. 2016, 113, E6362-E6371. 16. Sasaki, N.; Odajima, S., Elongation Mechanism of Collagen Fibrils and Force-Strain Relations of Tendon at Each Level of Structural Hierarchy. J. Biomech. 1996, 29, 1131-1136. 17. Record, M. T.; Anderson, C. F.; Lohman, T. M., Thermodynamic Analysis of Ion Effects on the Binding and Conformational Equilibria of Proteins and Nucleic Acids: The Roles of Ion Association or Release, Screening, and Ion Effects on Water Activity. Q. Rev. Biophys. 1978, 11, 103-178.


28

Page 29 of 36 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

ACS Nano

18. Grant, C. A.; Brockwell, D. J.; Radford, S. E.; Thomson, N. H., Tuning the Elastic Modulus of Hydrated Collagen Fibrils. Biophys. J. 2009, 97, 2985-2992. 19. Svensson, R. B.; Hassenkam, T.; Grant, C. A.; Magnusson, S. P., Tensile Properties of Human Collagen Fibrils and Fascicles Are Insensitive to Environmental Salts. Biophys. J. 2010, 99, 4020-4027. 20. Wells, J., Salt Activity and Osmotic Pressure in Connective Tissue. I. A Study of Solutions of Dextran Sulphate as a Model System. Proc. R. Soc. Lond. B Biol. Sci. 1973, 183, 399-419. 21. Wang, N.; Tytell, J. D.; Ingber, D. E., Mechanotransduction at a Distance: Mechanically Coupling the Extracellular Matrix with the Nucleus. Nat. Rev. Mol. Cell Biol 2009, 10, 75-82. 22. Na, S.; Collin, O.; Chowdhury, F.; Tay, B.; Ouyang, M.; Wang, Y.; Wang, N., Rapid Signal Transduction in Living Cells Is a Unique Feature of Mechanotransduction. Proc. Natl. Acad. Sci. 2008, 105, 6626-6631. 23. Andriotis, O.; Chang, S.; Vanleene, M.; Howarth, P.; Davies, D.; Shefelbine, S.; Buehler, M.; Thurner, P., Structure–Mechanics Relationships of Collagen Fibrils in the Osteogenesis Imperfecta Mouse Model. J. Royal Soc. Interface 2015, 12, 20150701. 24. Andriotis, O. G.; Manuyakorn, W.; Zekonyte, J.; Katsamenis, O. L.; Fabri, S.; Howarth, P. H.; Davies, D. E.; Thurner, P. J., Nanomechanical Assessment of Human and Murine Collagen Fibrils Via Atomic Force Microscopy Cantilever-Based Nanoindentation. J. Mech. Behav. Biomed. Mater. 2014, 39, 9-26. 25. Baldwin, S. J.; Quigley, A. S.; Clegg, C.; Kreplak, L., Nanomechanical Mapping of Hydrated Rat Tail Tendon Collagen I Fibrils. Biophys. J. 2014, 107, 1794-1801. 26. Grant, C. A.; Brockwell, D. J.; Radford, S. E.; Thomson, N. H., Effects of Hydration on the Mechanical Response of Individual Collagen Fibrils. Appl. Phys. Lett. 2008, 92, 233902. 27. Heim, A. J.; Matthews, W. G.; Koob, T. J., Determination of the Elastic Modulus of Native Collagen Fibrils Via Radial Indentation. Appl. Phys. Lett. 2006, 89, 181902. 28. Uhlig, M. R.; Magerle, R., Unraveling Capillary Interaction and Viscoelastic Response in Atomic Force Microscopy of Hydrated Collagen Fibrils. Nanoscale 2017, 9, 1244-1256. 29. van der Rijt, J. A. J.; van der Werf, K. O.; Bennink, M. L.; Dijkstra, P. J.; Feijen, J., Micromechanical Testing of Individual Collagen Fibrils. Macromol. Biosci. 2006, 6, 697-702. 30. Wenger, M. P. E.; Bozec, L.; Horton, M. A.; Mesquida, P., Mechanical Properties of Collagen Fibrils. Biophys. J. 2007, 93, 1255-1263. 31. Spitzner, E.-C.; Röper, S.; Zerson, M.; Bernstein, A.; Magerle, R., Nanoscale Swelling Heterogeneities in Type I Collagen Fibrils. ACS nano 2015, 9, 5683-5694. 32. Quigley, A. S.; Veres, S. P.; Kreplak, L., Bowstring Stretching and Quantitative Imaging of Single Collagen Fibrils Via Atomic Force Microscopy. PLoS One 2016, 11, e0161951. 33. Gutsmann, T.; Fantner, G. E.; Kindt, J. H.; Venturoni, M.; Danielsen, S.; Hansma, P. K., Force Spectroscopy of Collagen Fibers to Investigate Their Mechanical Properties and Structural Organization. Biophys. J. 2004, 86, 3186-3193. 34. Eppell, S.; Smith, B.; Kahn, H.; Ballarini, R., Nano Measurements with Micro-Devices: Mechanical Properties of Hydrated Collagen Fibrils. J. Royal Soc. Interface 2006, 3, 117-121. 35. Snedeker, J. G.; Gautieri, A., The Role of Collagen Crosslinks in Ageing and Diabetesthe Good, the Bad, and the Ugly. Muscles Ligaments Tendons J. 2014, 4, 303. 36. Liu, Y.; Ballarini, R.; Eppell, S. J., Tension Tests on Mammalian Collagen Fibrils. Interface Focus 2016, 6, 20150080.


29

ACS Nano 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 30 of 36

37. Minary-Jolandan, M.; Yu, M.-F., Nanomechanical Heterogeneity in the Gap and Overlap Regions of Type I Collagen Fibrils with Implications for Bone Heterogeneity. Biomacromolecules 2009, 10, 2565-2570. 38. Bembey, A.; Oyen, M.; Bushby, A.; Boyde, A., Viscoelastic Properties of Bone as a Function of Hydration State Determined by Nanoindentation. Philos. Mag. 2006, 86, 5691-5703. 39. Werth, A. J.; Harriss, R. W.; Rosario, M. V.; George, J. C.; Sformo, T. L., Hydration Affects the Physical and Mechanical Properties of Baleen Tissue. Royal Soc. Open Sci. 2016, 3, 160591. 40. Meek, K.; Leonard, D.; Connon, C. J.; Dennis, S.; Khan, S., Transparency, Swelling and Scarring in the Corneal Stroma. Eye 2003, 17, 927-936. 41. Hara, M.; Verkman, A., Glycerol Replacement Corrects Defective Skin Hydration, Elasticity, and Barrier Function in Aquaporin-3-Deficient Mice. Proc. Natl. Acad. Sci. 2003, 100, 7360-7365. 42. Berendsen, H.; Migchelsen, C., Hydration Structure of Fibrous Macromolecules. Ann. N. Y. Acad. Sci. 1965, 125, 365-379. 43. Bella, J.; Eaton, M.; Brodsky, B.; Berman, H. M., Crystal and Molecular Structure of a Collagen-Like Peptide at 1.9 a Resolution. Science 1994, 266, 75-81. 44. Fullerton, G. D.; Rahal, A., Collagen Structure: The Molecular Source of the Tendon Magic Angle Effect. J. Magn. Reson. Imaging 2007, 25, 345-361. 45. Leikin, S.; Rau, D.; Parsegian, V., Direct Measurement of Forces between SelfAssembled Proteins: Temperature-Dependent Exponential Forces between Collagen Triple Helices. Proc. Natl. Acad. Sci. 1994, 91, 276-280. 46. Price, R. I.; Lees, S.; Kirschner, D. A., X-Ray Diffraction Analysis of Tendon Collagen at Ambient and Cryogenic Temperatures: Role of Hydration. Int. J. Biol. Macromol. 1997, 20, 23-33. 47. Maruthamuthu, V.; Sabass, B.; Schwarz, U. S.; Gardel, M. L., Cell-Ecm Traction Force Modulates Endogenous Tension at Cell–Cell Contacts. Proc. Natl. Acad. Sci. 2011, 108, 47084713. 48. Weber, G. F.; Bjerke, M. A.; DeSimone, D. W., A Mechanoresponsive Cadherin-Keratin Complex Directs Polarized Protrusive Behavior and Collective Cell Migration. Dev. Cell 2012, 22, 104-115. 49. Wang, C.; Chowdhury, S.; Driscoll, M.; Parent, C. A.; Gupta, S.; Losert, W., The Interplay of Cell–Cell and Cell–Substrate Adhesion in Collective Cell Migration. J. Royal Soc. Interface 2014, 11, 20140684. 50. Puxkandl, R.; Zizak, I.; Paris, O.; Keckes, J.; Tesch, W.; Bernstorff, S.; Purslow, P.; Fratzl, P., Viscoelastic Properties of Collagen: Synchrotron Radiation Investigations and Structural Model. Philos. Trans. R. Soc. Lond. B Biol. Sci. 2002, 357, 191-197. 51. Schneider, K. H.; Aigner, P.; Holnthoner, W.; Monforte, X.; Nürnberger, S.; Rünzler, D.; Redl, H.; Teuschl, A. H., Decellularized Human Placenta Chorion Matrix as a Favorable Source of Small-Diameter Vascular Grafts. Acta Biomater. 2016, 29, 125-134. 52. Yoon, J. H.; Halper, J., Tendon Proteoglycans: Biochemistry and Function. J. Musculoskelet. Neuronal Interact. 2005, 5, 22-34. 53. Price, F.; Levick, J.; Mason, R., Glycosaminoglycan Concentration in Synovium and Other Tissues of Rabbit Knee in Relation to Synovial Hydraulic Resistance. J. Physiol. 1996, 495, 803-820.


30

Page 31 of 36 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

ACS Nano

54. Gandhi, N. S.; Mancera, R. L., The Structure of Glycosaminoglycans and Their Interactions with Proteins. Chem. Biol. Drug Des. 2008, 72, 455-482. 55. Maroudas, A., Balance between Swelling Pressure and Collagen Tension in Normal and Degenerate Cartilage. Nature 1976, 260, 808-809. 56. Pacella, E.; Pacella, F.; De Paolis, G.; Parisella, F. R.; Turchetti, P.; Anello, G.; Cavallotti, C., Glycosaminoglycans in the Human Cornea: Age-Related Changes. Ophthalmol. Eye Dis. 2015, 7, OED. S17204. 57. Ramachandran, G., Structure of Collagen. Nature 1956, 176, 710-711. 58. Jenkins, C. L.; Raines, R. T., Insights on the Conformational Stability of Collagen. Nat. Prod. Rep. 2002, 19, 49-59. 59. Buehler, M. J., Atomistic and Continuum Modeling of Mechanical Properties of Collagen: Elasticity, Fracture, and Self-Assembly. J. Mater. Res. 2006, 21, 1947-1961. 60. Chang, S. W.; Shefelbine, S. J.; Buehler, M. J., Structural and Mechanical Differences between Collagen Homo- and Heterotrimers: Relevance for the Molecular Origin of Brittle Bone Disease. Biophys. J. 2012, 102, 640-648. 61. Orgel, J. P. R. O.; Irving, T. C.; Miller, A.; Wess, T. J., Microfibrillar Structure of Type I Collagen in Situ. Proc. Natl. Acad. Sci. 2006, 103, 9001-9005. 62. Gautieri, A.; Buehler, M. J.; Redaelli, A., Deformation Rate Controls Elasticity and Unfolding Pathway of Single Tropocollagen Molecules. J. Mech. Behav. Biomed. Mater. 2009, 2, 130-137. 63. Buehler, M. J.; Wong, S. Y., Entropic Elasticity Controls Nanomechanics of Single Tropocollagen Molecules. Biophys. J. 2007, 93, 37-43. 64. Lovelady, H. H.; Shashidhara, S.; Matthews, W. G., Solvent Specific Persistence Length of Molecular Type I Collagen. Biopolymers 2014, 101, 329-335. 65. Hulmes, D.; Wess, T. J.; Prockop, D. J.; Fratzl, P., Radial Packing, Order, and Disorder in Collagen Fibrils. Biophys. J. 1995, 68, 1661-1670. 66. Gautieri, A.; Pate, M. I.; Vesentini, S.; Redaelli, A.; Buehler, M. J., Hydration and Distance Dependence of Intermolecular Shearing between Collagen Molecules in a Model Microfibril. J. Biomech. 2012, 45, 2079-2083. 67. Record, M. T.; Lohman, T. M.; De Haseth, P., Ion Effects on Ligand-Nucleic Acid Interactions. J. Mol. Biol. 1976, 107, 145-158. 68. Stevens, M. J., Simple Simulations of DNA Condensation. Biophys. J. 2001, 80, 130-139. 69. Stolz, M.; Gottardi, R.; Raiteri, R.; Miot, S.; Martin, I.; Imer, R.; Staufer, U.; Raducanu, A.; Duggelin, M.; Baschong, W., Early Detection of Aging Cartilage and Osteoarthritis in Mice and Patient Samples Using Atomic Force Microscopy. Nat Nanotechnol 2009, 4, 186-92. 70. Herrera, J.; Henke, C. A.; Bitterman, P. B., Extracellular Matrix as a Driver of Progressive Fibrosis. J. Clin. Invest. 2018, 128, 45-53. 71. Discher, D. E.; Janmey, P.; Wang, Y.-l., Tissue Cells Feel and Respond to the Stiffness of Their Substrate. Science 2005, 310, 1139-1143. 72. Screen, H. R.; Shelton, J. C.; Chhaya, V. H.; Kayser, M. V.; Bader, D. L.; Lee, D. A., The Influence of Noncollagenous Matrix Components on the Micromechanical Environment of Tendon Fascicles. Ann. Biomed. Eng. 2005, 33, 1090-1099.


31

ACS Nano 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

Graphical abstract 36x15mm (300 x 300 DPI)


Page 32 of 36

Page 33 of 36 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

ACS Nano

Figure 1. Collagen fibril shrinking and stiffening with increasing osmotic pressure. (a) AFM height topography images of a collagen fibril in air-dried state (only bound water present), fully hydrated in phosphate buffered saline (PBS) and, subsequently, partly dehydrated in increasing concentrations of polyethylene glycol i.e. increasing osmotic pressure. (b) Corresponding height profiles of the same fibril in different aqueous solutions (colored line) and in air (black dashed line). (c) Data from two selected fibrils showing height decrease by 20% from fully hydrated state to highest concentration of PEG. (d) The indentation modulus increases from about 15 MPa in fully hydrated state to about 370 MPa at the highest concentration of PEG. Error bars in panel (c) and (f) represent the standard deviation of the shrinking taken from 150 points across the fibril crest. Error bars in panel (d) represent the standard deviation of indentation modulus across the fibril at a given PEG concentration. 100x56mm (600 x 600 DPI)


ACS Nano 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

Figure 2. Native collagen fibril under tension with osmotic pressure. (a) Illustration of a native collagen fibril under tensile loading. (b) Six force-displacement (fibril displacement) curves per solution for a 43 µm long and 230 nm wide (PBS) collagen fibril during gradual dehydration from phosphate buffer saline to 3.5M of PEG solutions. (c-f) Tensile structural stiffness and elastic modulus plotted against fibril strain (small strain approximation). (d, f) Comparison of structural stiffness and elastic modulus of the collagen fibril measured in PBS and 2.6M PEG. (c-f) The error bars represent the standard deviation at the given fibril strain from six force-displacement curves. 95x67mm (300 x 300 DPI)


Page 34 of 36

Page 35 of 36 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

ACS Nano

Figure 3. Collagen fibril stiffness and tensile modulus as a function of loading speed and fibril strain. Three dimensional illustrations show the dependence of collagen fibril (128 µm in length) stiffness (a) and elastic modulus (b) on both loading speed (ranging from 0.5 µm/s to about 99.2 µm/s) and fibril strain in PBS and 2.6M PEG(semi-transparent). At lower fibril strains the changes in stiffness with displacement rate are small. However, at larger fibril strains the effect of displacement rate is more pronounced and it is further enhanced with increased osmotic pressure (c, d). The error bars represent the standard deviation of six force displacement curves. 117x118mm (300 x 300 DPI)


ACS Nano 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

Figure 4. Physical properties of collagen fibrils are tunable via osmotic pressure. At low osmotic pressure the equilibrium intermolecular (interaxial) distance is larger compared to the one at higher osmotic pressure. Decreasing the intermolecular distance, the density of non-covalent interactions increases due to the increased packing density of collagen molecules at higher osmotic pressures. This results in a shift of the Lennard-Jones potential towards the situation with no water present, requiring higher forces for intermolecular movement such as lateral compression or sliding during indentation and tensile loading. 89x92mm (300 x 300 DPI)


Page 36 of 36

Collagen Fibrils: Nature's Highly Tunable Nonlinear Springs - ACS

Recommend Documents