Super-Resolution Fluorescence Imaging of Spatial Organization of

Subscriber access provided by UNIVERSITY OF TOLEDO LIBRARIES

Article

Super-resolution Fluorescence Imaging of Spatial Organization of Proteins and Lipids in Natural Rubber Jinrong Wu, Wei Qu, Guangsu Huang, Siyuan Wang, Cheng Huang, and Han Liu Biomacromolecules, Just Accepted Manuscript • Publication Date (Web): 02 May 2017 Downloaded from http://pubs.acs.org on May 6, 2017

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 free 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 accessible to all readers and 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.

Biomacromolecules 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 20 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

Biomacromolecules

Super-resolution Fluorescence Imaging of Spatial Organization of Proteins and Lipids in Natural Rubber Jinrong Wu 1, ‡, Wei Qu1, ‡, Guangsu Huang1,*, Siyuan Wang2, Cheng Huang1, Han Liu1 1

State Key Laboratory of Polymer Material Engineering, College of Polymer Science and

Engineering, Sichuan University, Chengdu 610065, People’s Republic of China. 2

Howard Hughes Medical Institute, Department of Chemistry and Chemical Biology,

Department of Physics, Harvard University, Cambridge, MA 02138, USA.

ABSTRACT. Natural rubber (NR) with proteins and lipids has superior mechanical properties to its synthetic counterpart, polyisoprene rubber. However, it is a challenge to unravel the morphology of proteins and lipids. Here we used two-color stochastic optical reconstruction microscopy (STORM) to directly visualize the spatial organization of proteins and lipids in NR. We found that the proteins and lipids form an inter-dispersed stabilizing layer on the surface of NR latex particles. After drying, the proteins and lipids form aggregates of up to 300 nm in diameter. The aggregates physically interact with the terminal groups of polyisoprene chains, leading to the formation of a network, which contributes to the high elasticity and mechanical property of NR. If we remove proteins in NR, the large phospholipid aggregates disintegrate into small ones. However, it does not decompose the network but rather reduces the effective crosslinking density, thus the deproteinized NR is still elastic-like with decreased mechanical property. Removing both proteins and lipids wholly decomposes the network, thus results in a liquid-like

ACS Paragon Plus Environment

1

Biomacromolecules 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 20

behavior of the rubber. The STORM measurements in this paper enable more insight into the structureproperty relationship of NR, which also shows a great potential of STORM in studying the fine structure of polymeric materials and nanocomposites.

Introduction Biomacromolecules usually form unique hierarchical structures that are critical in determining their mechanical properties and functionalities. An elegant example is natural rubber (NR) bio-synthesized by rubber tree, Hevea brasiliensis. NR possesses excellent mechanical properties, such as high tensile and tear strength, high crack growth resistance, good elasticity and minimal heat buildup1, 2. These properties, however, cannot be fully replicated in synthetic polyisoprene rubber (IR), although the latter can achieve a similar molecular weight and chain configuration to the rubber chains in NR3. As a result, NR is still indispensible for heavy-duty applications, especially tires for aircrafts and heavy trucks, and rubber bearings for seismic isolation. NR consists of approximately 94% cis-1, 4-polyisoprene (cis-PIP) and 6% non-rubber components such as proteins, phospholipids, glycolipids, fatty acids, etc4. The non-rubber components, especially phospholipids and proteins, contribute significantly to the excellent mechanical properties of NR, since removing phospholipids and proteins greatly deteriorates the mechanical properties5, 6. Therefore, unraveling how non-rubber components organize in the rubber matrix and how they contribute to the mechanical properties of NR is a central issue in understanding the structureproperty relationship of NR, and designing new biomimetic material with similar properties. Proteins and lipids are immiscible with cis-PIP molecules, thus they have a high propensity to phase separate from cis-PIP and form aggregates embedded in the rubber matrix. Currently, it is widely accepted that the aggregates of proteins interact with ω terminals of cis-PIP chains, while the aggregates of phospholipids (and other lipids) interact with the α terminals of cis-PIP chains, leading to the formation of a network; this is the main reason for the better mechanical properties of NR7-9. To explore the morphology of the proteins and lipids in the network, small-angle neutron scattering (SANS), small-angle


2

Page 3 of 20 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

Biomacromolecules

X-ray scattering (SAXS), scanning and transmission electron microscopy, and atomic force microscopy (AFM) measurements were performed10-14. SANS and SAXS tests suggests that proteins form aggregates of ≥ 20 nm, while AFM tests suggest that the aggregates of protein are quite irregular with size up to hundreds of nanometers10, 11. Conventional optical microscope and confocal fluorescence microscope were also used to investigate the morphology of proteins and lipids11, 12. However, these techniques can only observe aggregates well above the diffraction barrier of light microscopy, and moreover one cannot differentiate proteins from lipids due to the nonspecific labeling by fluorescent probes. As a consequence, the spatial organization of proteins and lipids has not been observed directly, and the interaction between them in NR remains an open question. Resolving these questions relies on imaging tools with both high spatial resolution and molecular specificity. Recent advances in superresolution fluorescence microscopy enable resolution down to 10 nm with molecular specificity; this provides a promising approach to solve the above questions. Among various superresolution techniques, stochastic optical reconstruction microscopy (STORM) is particularly powerful for probing proteins and lipids15. By using photo-switchable probes, STORM activates a sparse subset of optically resolvable fluorophores at any moment, thereby allowing localization of individual fluorescent molecules with localization precision as high as several nanometers15. Over the course of multiple activation cycles, the coordinates of a sufficient number of fluorophores are determined and used to construct a high-resolution STORM image with a routine resolution of 20nm 15, and in some special implementations a sub 10-nm resolution16, 17. The development of multicolor STORM depending on photo-switchable probes with distinct colors further enables the investigation of spatial distribution and molecular interaction between proteins and lipids on the nanometer scale18. Here we exploited STORM to directly visualize the morphology of NR, thereby we were able to unravel the aggregate structure of proteins and phoslipids on both NR latex particles and dry NR. By combining with the measurements of solid state nuclear magnetic resonance and rheology, we confirmed that the aggregates physically interact


3


Page 4 of 20

with the terminal groups of rubber chains leading to the formation of a network, which contributes to the high elasticity and mechanical properties of NR. Experimental methods Sample preparation. Triton X-100 and protease (P-5380) were provided by Sigma-Aldrich. Concentrated natural rubber latex from clone RRIM600, provided by Chinese Academy of Tropical Agricultural Sciences (China), was preserved at a solid fraction of 60% (wt/wt) with the presence of 0.6% ammonia for nearly 3 months before experiments. The concentrated natural rubber latex was diluted to a concentration of 30% (wt/wt) and then centrifuged at 1000 rpm to remove impurities other than latex particles. The latex particles were re-dispersed in deionized water to form purified NR latex with a concentration of 30% (wt/wt). The purified NR latex was then casted in a petri dish and dried in a vacuum oven at 40 oC for 3 days to obtain dry NR. Deproteinization of the purified NR latex with a solid fraction of 30 % (wt/wt) was carried out by incubating it with 0.08 % (wt/wt) protease in the presence of 0.15% (vol/vol) Triton X-100 and 1 % (wt/wt) aqua ammonia at 37 oC for 12 h. The incubated latex was centrifuged at 10000 rpm for 40 min to obtain a cream. The cream was redispersed in deionized water and washed for three times by repeated centrifugation and dispersion. The washed cream was cast on a Teflon mold and then placed in a vacuum oven at 40 oC for 3 days to acquire dry DPNR. Measurement of the nitrogen content of DPNR was performed using a nitrogen elemental analyzer (CARLO ERBA 1106, Italian). The nitrogen content was reduced from 0.5% to 0.07% after deproteinization. To prepare transesterified DPNR (TEDPNR), 6 g of dry DPNR was dissolved in 400 ml toluene. The solution was added with 6 ml sodium methoxide in methanol (14% wt/vol). Transesterification of DPNR was carried out at room temperature for 8 h. The resulting TEDPNR was purified by precipitation of the solution in excess methanol. The precipitated TEDPNR was then placed in a Teflon mold and dried in a vacuum oven at 40 oC for 3day. Measurement of the phosphorus content in TEDPNR was performed


4

Page 5 of 20 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

Biomacromolecules

using a phosphorus elemental analyzer (IRIS 1000 ICPAES, Thermo Electron Co. USA). The results showed that the phosphorus content was reduced from 224 ppm of NR to 65 ppm of TEDPNR after transesterification. We also added a protein back inside DPNR. The protein in dry form, provided by Seebio in Shanghai, is a surface protein originally found in the cell wall of staphylococcus aureus. It is hydrophilic with a

molecular weight around 42 kDa. The protein is labeled by a cyanine dye (Cy5) with Nhydroxysuccinimide (NHS) ester, which was attached to the protein via its lysine residue. The protein was first dissolved in deionized water and then blended with DPNR latex under agitation for 30min to form a weight fraction of 2% protein in dry rubber. The latex was casted on a Teflon mold and then dried in a vacuum oven at 40 oC for 3 days to form a DPNR film with added Cy5 labeled protein. To prepare samples for STORM imaging, dry NR, DPNR, TEDPNR and DPNR with added Cy5 labeled protein were dissolved in toluene with a concentration of 2% (wt/vol) under strong agitation. The solutions were spin coated onto glass slides to form thin films about 5 µm in thickness. The thin films of NR, DPNR and TEDPNR on the glass slides were incubated in a protein-staining solution containing 25µM Cy5 NHS ester (VWR) in dimethyl sulphoxide (DMSO) for 2 hours; during this process the Cy5 fluorophores can be covalently coupled with the protein molecules by reacting the NHS groups with the amine groups19. The thin films were then rinsed with fresh DMSO for five times. Subsequently, the Cy5labeled thin films of NR, DPNR and TEDPNR and the thin film of DPNR with added Cy5 labeled protein were incubated in phospholipid-staining solution containing 20 µM DiI (VWR) in DMSO for 5–10 min. Lipophilic DiI molecules are structurally analogous to phospholipids in terms of their positively-charged polar head groups and dual nonpolar acyl chains. The DiI molecules label the cell membrane by organizing the polar head groups parallel to the surface of the cell and imbedding the acyl chains in the bilayer parallel to the phospholipid acyl chains.20 In natural rubber, since phospholipids assembly with other lipids to form a micelle-like structure,9, 21 we believe that DiI labels lipids in the same way as that in the cell membrane. The DiI labeled thin films then repeatedly washed with fresh DSMO for 10 times.


5


Page 6 of 20

Purified NR latex was first diluted to have a concentration about 1% (wt/wt); the diluted NR latex was deposited on a glass slide by spin coating. Spinning at 300 rpm for 1 min resulted in dried particles fixed on the glass slide, which then underwent the same labeling process as dry NR, DPNR and TEDPNR thin films. After staining, the glass slides with thin films or fixed latex particles were immediately used for imaging. STORM imaging. The imaging buffer, which was used to improve photo-switching property of fluorophores, contained 2 mg/ml glucose, 0.5 mg/ml glucose oxydase, 40 µg/mL catalase, 1% (v/v) βmercaptoethanol and 10% (v/v) PBS.22 We carried out STORM imaging on an Olympus IX-71 inverted microscope. A 405-nm laser (Coherent) was used for photoactivation, and 561- and 640-nm lasers (Coherent) were used to excite and image the DiI and Cy5 probes, respectively. The power densities were automatically controlled by an acousto-optic tunable filter (AOTF PCAOM NI VIS; Crystal Technology) for the 405-, 561- and 640-nm lines. Before acquiring STROM images, we used relatively weak 561- and 640-nm lasers to illuminate the sample and record the conventional fluorescence image. We then used simultaneous illumination with both the activation and imaging lasers, which resulted in the stochastic switching of the reporter fluorophores between the fluorescent and dark states. An EMCCD camera was used to acquire images continuously at a frame rate of 60 Hz. An Insight 3 software was used to process the superresolution images to obtain STORM images. For each sample, at least three independent thin films or arrays of deposited latex particles were imaged to assure the reproducibility of STORM imaging. 1

H double-quantum solid-state NMR measurements. 1H DQ NMR measurements were performed on a

Bruker Advance III 300M solid-state NMR spectrometer operating at a Larmor frequency of 300.2 MHz. All experiments were conducted under static condition in a 4 mm MAS probe using 90° pulses of 3µs. The sample temperature was regulated at 40 ± 0.1 oC, and the rotors were prevented from spinning using Teflon-tape wrapping around the drive turbines. The Saalwachter’s NMR sequence [15] was used to obtain DQ buildup ( I DQ ) curves with DQ evolution time ( τ DQ ). The crosslink density was acquired by the method in the Supporting Information.


6

Page 7 of 20 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

Biomacromolecules

Rheological measurements. Rheological measurements were performed using a HAAKE PXR800 rheometer. Frequency dependence of the storage modulus was measured in the frequency range from 0.01 to 100 Hz under a deformation amplitude of 1%; this deformation was in the range of linear viscoelastic response. The geometry of parallel plates was employed with a plate diameter of 20 mm and a sample thickness of about 1 mm. Mechanical property measurement. Mechanical property was measured with an Instron-5567 material testing machine at 298K with the deformation rate of 12mm/min. The original sample length was 20mm, so the initial rate of deformation was 0.01s-1. Results and Discussion Ideally, to investigate the morphology of proteins and lipids in NR with fluorescence microscopy, one needs two different dyes with distinct fluorescence colors to selectively label proteins and lipids. Here, we used a cyanine dye (Cy5) coupled with N-hydroxysuccinimide (NHS) ester and an indocarbocyanine dye (DiI) to label proteins and lipids respectively with high specificity (Supporting Figure S1 and Figure S2)15, 23. Then, we performed two-color STORM imaging to visualize the spatial distribution of proteins and lipids. In the two-color STORM imaging, we used a 405 nm laser to activate the Cy5 and DiI probes, and used 640-nm and 561-nm lasers to image the activated Cy5 and DiI probes, respectively. We first performed two-color STORM imaging on NR latex, which was harvested from rubber trees. The NR latex is composed of dispersed colloidal particles, which have a core of cis-PIP and a surface membrane composed of proteins and lipids12, 24. Measurements by electron microscopes have revealed that the surface membrane has a thickness of about 2-4 nm13, 25, however, the spatial organization of proteins and lipids in the membrane is still unclear. Here the colloidal particles of NR latex were deposited and fixed on a glass slide by spin coating and measured by two-color STORM imaging. It clearly reveals that the colloidal particles are quite polydisperse (Figure 1a). On the surface of the particles, both proteins and lipids are segregated into domains of tens to hundreds of nanometers in


7


Page 8 of 20

diameter (Figure 1b). These domains are randomly inter-dispersed, suggesting that the proteins and lipids form a mixed layer, which acts as surfactants in stabilizing the latex particles. The diameter of the particles is in the range of 0.1 to 3 µm with a distribution center at 400nm (Figure 1c), which agrees well with the values measured by light scattering and AFM12, 26.

Figure 1. Morphology of NR latex particles measured by two-color STORM imaging. (A) Two-color image of a large area of NR latex deposited on a glass slide. The image was taken by focusing on the bottom surface of the spheres. The localizations of Cy5 and DiI are colored with red and green, respectively. (B) Magnified view corresponding to the boxed region in (A) shown at a higher magnification. (C) Particle size distribution of NR latex obtained by analyzing 100 particles. In dry NR, the proteins form aggregates which interact with ω terminals of cis-PIP chains, and phospholipids assembly with long-chain fatty acid ester groups of α terminals to form a micelle-like structure.9, 21 To visualize the morphology of proteins and lipids in dry NR, we coat the dry NR onto a


8

Page 9 of 20 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

Biomacromolecules

glass slide to form a film of about 5 µm; this thickness allows the film to have a morphology approaching that of the bulk state since the aggregate sizes of proteins and lipids are generally smaller than this film thickness. The film is annealed at 90 °C for 5 min to remove any possible residual stress generated by spin coating. Conventional fluorescence imaging on the film shows that both proteins and lipids form segregated big domains with diameters from 1 to 10 µm in NR (Figure 2a&b). This morphology agrees well with the observation by conventional optical microscopy 11. A rough estimation based on the areas of the big domains suggests that the volume percentages of proteins and lipids in NR are about 10% and 7%, respectively. However, the actual weight fractions of proteins and lipids in NR are only about 2 wt% and 3 wt%, respectively.11, 27 This significant discrepancy indicates that the low resolution of conventional fluorescence microscopy can lead to false interpretation of the morphology of proteins and lipids in NR. In contrast to the conventional fluorescence images, two-color STORM images clearly reveal that the aggregate sizes of proteins and lipids are much smaller, up to 300 nm, as shown in Figure 2c. Moreover, the higher resolution allows us to observe the fine structure in the micron-sized big domains in Figure 2a&b. We find that these domains contain not only proteins and lipids, but also rubber matrix. The proteins and lipids form relatively large aggregates embedded in the rubber matrix, and some of the aggregates are 200 to 300 nm in diameter (Figure 2d). Outside the big domains, we observe mainly dispersed and small aggregates of proteins and lipids up to 100 nm in diameter (Figure 2e). The aggregate sizes inside the big domains are on the same scale as that observed by AFM, while outside the big domains our results agrees well with the SANS and SAXS measurements10, 11. Therefore, the twocolor STORM imaging provides a more comprehensive morphology for NR than conventional imaging and scattering technologies.


9


Page 10 of 20

Figure 2. Morphology of proteins and lipids in dry NR coated on a glass slide. (A) Conventional fluorescence image of Cy5-labeled proteins in NR. (B) Conventional fluorescence image of DiI-labeled lipids of the same area. (C) Two-color STORM image of the same area in dry NR. The localizations of Cy5 and DiI are colored with red and green, respectively. (D and E) STORM images corresponding to the boxed region in (C) shown at a higher magnification. (F) Ratio of proteins to lipids within and outside the big domains. A closer observation reveals two more features on the two-color STORM images (Figure 2c-d). One feature is that both the aggregates of proteins and lipids are not randomly distributed throughout the cis1,4-polyisoprene matrix, instead they form a mesh-like morphologies in many areas. This morphology is favorable to the formation of the naturally occurring network, which distinguishes NR from synthetic


10

Page 11 of 20 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

Biomacromolecules

rubbers7-9. Another feature is that there are more proteins than lipids in the big domains, while fewer proteins are observed outside the big domains. To quantify this difference, we calculate the ratio of proteins to lipids based on the areas of the two components. We find that the ratios are 1.5 and 0.8 within and outside the big domains respectively (Figure 2f). The richness of proteins in some areas can be used to explain a widely observed phenomenon that removing proteins from NR leads to the disappearance of big domains10, 11. To further explore the morphology of proteins and lipids, we also prepared deproteinized NR (DPNR) and transesterified DPNR (TEDPNR). Deproteinization with a proteolytic enzyme removes most of the proteins in DPNR, while transesterification of DPNR with sodium methoxide further removes most of the lipids for TEDPNR (see details in Methods). Single-color STORM imaging was performed for DiI labeled DPNR and TEDPNR, and the results were compared with that of NR. We find that the phospholipid aggregates in NR are quite inhomogeneous with a wide size range, in which the upper limit can reach 100 to 200 nm (Figure 3a&d). After removing proteins, the relatively large aggregates disappear, and the distribution becomes rather homogeneous in DPNR (Figure 3b&e). Thus, the existence of proteins can promote the aggregation of lipids, and leads to formation of the relatively large aggregates. Although the protein aggregates are removed upon deproteinization (Supporting Figure S1), the lipid aggregates alone can compose a network, which can be confirmed by rheological measurement. The storage modulus (Gˊ) of NR shows a rubbery plateau at room temperature (Figure 3f), suggesting its elastic nature due to the existence of the naturally occurring network. DPNR, although has lower Gˊ values than NR, still shows the rubbery plateau; this indicates that deproteinization only leads to reduced effective crosslinking density yet still retains the naturally occurring network. Further transesterification nearly removes all lipids (Figure 3c). As a consequence, Gˊ of TEDPNR is further reduced and decreases remarkably with decreasing frequency (Figure 3f). The lack of a rubbery plateau suggests that TEDPNR


11


Page 12 of 20

is a viscous liquid rather than an elastomer. Thus, removing both proteins and lipids wholly decomposes the naturally occurring network in TEDPNR.

Figure 3. Morphology, rheological and mechanical properties, and network structure of NR, DPNR and TEDPNR. (A) Single-color DiI image of NR from a part of the area in Figure 2c. (B) Single-color DiI image of DPNR. (C) Single-color DiI image of TEDPNR. (D) Images with different magnifications selected from (A). (E) Images with different magnifications selected from (B). (F) The frequency dependent curves of storage modulus, Gˊ, for NR, DPNR and TEDPNR at 25 °C. (G) The stress-strain curves of NR, DPNR, TEDPNR and IR measured at room temperature. (H) Normalized 1H-DQ NMR buildup curves for NR, DPNR and TEDPNR at 40 °C, and the inset shows the effective crosslinking density.


12

Page 13 of 20 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

Biomacromolecules

To quantify the changes in the network structure upon deproteinization and transesterification, we perform 1H-double quantum nuclear magnetic resonance measurements (1H-DQ NMR) on NR, DPNR and TEDPNR. The initial slope of the normalized DQ buildup (InDQ) curve decreases with the removal of proteins and lipids, as shown in Figure 3h. From the InDQ curves, we can calculate the effective crosslinking density (v) of NR, DPNR and TEDPNR (see details in Supporting Information). The v includes both the relatively strong crosslinks generated by the naturally occurring network and the instantaneous physical crosslinks generated by chain entanglements28. We find that the v value decreases upon deproteinization, and further decreases after transesterification, as shown in the inset of Figure 3h. Since TEDPNR is made of linear polymer chains8, 9, the v value of TEDPNR can only be attributed to the physical crosslinks of chain entanglements. Thus, subtracting the v value of TEDPNR from that of NR will give us an effective crosslinking density generated by the naturally occurring network, which is about 1.5×10-5mol/g. The change in the naturally occurring network upon deproteinization and transesterification has a remarkable effect on the mechanical properties of NR, as shown in Figure 3g. The existence of naturally occurring network leads to remarkable strain-induced crystallization during the tensile test, which contributes significantly to the high tensile stress and elongation of NR, as revealed by wide angle X-ray diffraction (WAXD) in the Supporting Figure S3. The lower effective crosslinking density in DPNR reduces the crystallinity of strain-induced crystallization, which decreases the strain-induced crystallinity and the mechanical property. TEDPNR without proteins and lipids does not show any strain-induced crystallization, thus its mechanical property is decreased to a very low level, comparable to that of uncrosslinked IR. The network formation in NR mainly takes advantage of the interactions between the terminal groups of cis-PIP molecules and the aggregates of proteins and lipids. However, it is still unclear that whether the interactions are reversible physical associations or non-reversible covalent bonds. Several reports suggest that the interactions involve a main portion of hydrogen bonds and a small portion of ionic bonds5, 8, 9.


13


Page 14 of 20

These reversible interactions allow the aggregates of proteins and lipids to act as crosslinking points, which are stronger than the instantaneous crosslinking points formed by entanglement. However, these interactions are not permanent and susceptible to thermal fluctuation, thus increasing temperature could lead to disintegration of the network and reorganization of the proteins and lipids. To confirm the reversible and physical nature of the network, we thermally anneal NR at a temperature of 150 °C for 15min. This high-temperature can lead to the dissociation of the reversible interactions and increase the chain diffusivity of both rubber and non-rubber molecules. As a result, the network disintegrates into linear molecules, rendering NR behaving like a viscous liquid, as demonstrated by the decreasing of Gˊ with decreasing frequency at 150 °C (Figure 4a). In the viscous-liquid state at the high temperature, the segregation driving force of proteins and lipids is greatly reduced, which may lead to the disintegration of the aggregates of proteins and lipids. A subsequent quenching process by rapid cooling will immobilize the disintegrated aggregates. Indeed, STORM measurements for NR quenched after thermal annealing at 150 °C demonstrate that the large aggregates of both proteins and lipids disappear. As a result, the size distribution of the aggregates becomes more homogeneous compared to that in the pristine NR, as shown in Figure 4b&c. Thus, the high-temperature thermal annealing mainly affects the reorganization of large aggregates of proteins and lipids, as schematically described in Figure 4d. In addition to the spatial organization, the interaction between proteins and lipids is also an interesting subject, which remains elusive in dry NR up to date but has already been explored on the particle surface of NR latex29, 30. Here STORM observations suggest that there is some sort of interaction between proteins and lipids in dry NR. The first observation is that there are some locations, particularly the big domains, in which the aggregates of proteins and lipids overlap with each other (Figure 2c&f), suggesting that these two components can co-assembly together. The second one is that deproteinization not only removes proteins, but also leads to the disappearance of the relatively large aggregates of lipids (Figure 3b&e), which indicates that the existence of proteins can induce the aggregation of lipids. The third one is that if we incorporate a Cy5-labeled protein back inside the DPNR, it forms aggregates up to


14

Page 15 of 20 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

Biomacromolecules

300 nm in diameter; more importantly, the large aggregates of lipids with size up to 200 nm show up again upon addition of the Cy5-labeled protein (Figure 5). This phenomenon once again suggests the aggregation of lipids can be promoted by introducing proteins.

Figure 4. Reorganization of morphology of proteins and lipids in dry NR during thermal annealing at 150 °C and after being quenched from 150 °C. (A) The frequency dependent curves of storage modulus of NR at 25 and 150 °C. (B) Two-color STORM image of a large area in NR quenched after thermal annealing at 150 °C for 15min. The localizations of Cy5 and DiI are colored with red and green, respectively. (C) Image of the boxed region in (B). (D) Schematic description of reorganization of morphology of proteins and lipids during thermal annealing at 150 °C and after being quenched from 150 °C.


15


Page 16 of 20

Figure. 5. STORM imaging of DPNR with incorporation of Cy5 labeled proteins. (A) Two color STORM image of a large area. (B) and (C) STORM images of higher magnifications. Conclusion As a unique biomass, NR consists of not only polyisoprene but also proteins, lipids, carbohydrates and inorganic constituents. The existence of proteins and lipids is believed to be the essential reason why NR has superior mechanical properties to its synthetic counterpart, polyisoprene. However, the spatial organization of proteins and lipids has not been unraveled directly. Here we use STORM imaging to visualize the morphology of proteins and lipids in both NR latex and dry NR. We find that the proteins and lipids form separate domains, which are inter-dispersed on the surface of NR latex. After drying, the proteins and lipids segregate to form aggregates in dry NR. The aggregates are relatively large in some domains, which have more distributed proteins than lipids; while outside these domains with more distributed lipids, the aggregates are smaller and more dispersed. The aggregates of proteins and lipids interact with the terminal groups of cis-PIP chains through physical interactions, which leads to the formation of a naturally occurring network. This network contributes to the elastic behavior and the high mechanical property of the uncrosslinked NR. After deproteinization, the large aggregates of lipids disappear, leading to a narrower size distribution of the aggregates. Deproteinization reduces the effective crosslinking density of NR, yet the naturally occurring network is still retained. As a consequence, DPNR shows decreased mechanical property yet is still elastic-like. If both proteins and lipids are removed, the naturally occurring network is completely decomposed. Consequently, TEDPNR shows a viscous-liquid-


16

Page 17 of 20 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

Biomacromolecules

like behavior, and its mechanical property decreases to the same level as that of synthetic polyisoprene. The successful application of STORM in revealing the morphology of NR suggests a great future for this technique in studying the fine structure of various polymeric materials and their composites and providing more insightful understanding toward the structure-property relationships.

ASSOCIATED CONTENT Supporting Information. The Supporting Information is available free of charge on the ACS Publications website. Specific labeling of Cy5 NHS ester and DiI (Figures S1&S2), 1H-double quantum nuclear magnetic resonance measurements and in-situ tensile X-ray measurement (Figure S3) (PDF) AUTHOR INFORMATION Corresponding Author *Phone and Fax: 86-28-85463433; Email: [email protected] (G.H.). Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ‡These authors contributed equally. Notes The authors declare no competing financial interest. ACKNOWLEDGMENT This work was supported by National Natural Science Foundation of China (grant No.: 51333003), Special Fund for Agro-scientific Research in the Public Interest (201403066-1) and


17


Page 18 of 20

Sichuan University (grant No.: 2013SCU04A01). We appreciate Prof. Xiaowei Zhuang at Harvard University for providing the STORM instruments. REFERENCES 1. Tang, C.; Yang, M.; Fang, Y.; Luo, Y.; Gao, S.; Xiao, X.; An, Z.; Zhou, B.; Zhang, B.; Tan, X., The rubber tree genome reveals new insights into rubber production and species adaptation. Nat. Plants 2016, 2, 16073. 2. Roberts, A. D., Natural rubber science and technology. Oxford University Press: 1988. 3. Gent, A.; Kawahara, S.; Zhao, J., Crystallization and Strength of Natural Rubber and Synthetic cis-1, 4-polyisoprene. Rubber Chem. Technol. 1998, 71, (4), 668-678. 4. Tanaka, Y.; Tarachiwin, L., Recent advances in structural characterization of natural rubber. Rubber Chem. Technol. 2009, 82, (3), 283-314. 5. Amnuaypornsri, S.; Sakdapipanich, J.; Toki, S.; Hsiao, B. S.; Ichikawa, N.; Tanaka, Y., Strain-induced crystallization of natural rubber: effect of proteins and phospholipids. Rubber Chem. Technol. 2008, 81, (5), 753-766. 6. Zhou, Y.; Kosugi, K.; Yamamoto, Y.; Kawahara, S., Effect of non‐rubber components on the mechanical properties of natural rubber. Polym. Adv. Technol. 2017, 28, (2), 159-165. 7. Tanaka, Y., Structural characterization of natural polyisoprenes: solve the mystery of natural rubber based on structural study. Rubber Chem. Technol. 2001, 74, (3), 355-375. 8. Tarachiwin, L.; Sakdapipanich, J.; Ute, K.; Kitayama, T.; Bamba, T.; Fukusaki, E.-i.; Kobayashi, A.; Tanaka, Y., Structural characterization of α-terminal group of natural rubber. 1. Decomposition of branch-points by lipase and phosphatase treatments. Biomacromolecules 2005, 6, (4), 1851-1857. 9. Tarachiwin, L.; Sakdapipanich, J.; Ute, K.; Kitayama, T.; Tanaka, Y., Structural characterization of α-terminal group of natural rubber. 2. Decomposition of branch-points by phospholipase and chemical treatments. Biomacromolecules 2005, 6, (4), 1858-1863. 10. Karino, T.; Ikeda, Y.; Yasuda, Y.; Kohjiya, S.; Shibayama, M., Nonuniformity in Natural Rubber As Revealed by Small-Angle Neutron Scattering, Small-Angle X-ray Scattering, and Atomic Force Microscopy. Biomacromolecules 2007, 8, 693-699. 11. Toki, S.; Burger, C.; Hsiao, B. S.; Amnuaypornsri, S.; Sakdapipanich, J.; Tanaka, Y., Multi‐scaled microstructures in natural rubber characterized by synchrotron X‐ray scattering and optical microscopy. J. Polym. Sci. Part B: Polym. Phys. 2008, 46, (22), 2456-2464. 12. Nawamawat, K.; Sakdapipanich, J. T.; Ho, C. C.; Ma, Y.; Song, J.; Vancso, J. G., Surface nanostructure of Hevea brasiliensis natural rubber latex particles. Colloids Surf., A 2011, 390, (1), 157-166. 13. Rochette, C. N.; Crassous, J. J.; Drechsler, M.; Gaboriaud, F.; Eloy, M.; de Gaudemaris, B.; Duval, J. F., Shell structure of natural rubber particles: evidence of chemical stratification by electrokinetics and cryo-TEM. Langmuir 2013, 29, (47), 14655-65. 14. Kosugi, K.; Kawahara, S., Natural rubber with nanomatrix of non-rubber components observed by focused ion beam-scanning electron microscopy. Colloid Polym. Sci. 2015, 293, (1), 135-141. 15. Rust, M. J.; Bates, M.; Zhuang, X., Sub-diffraction-limit imaging by stochastic optical reconstruction microscopy (STORM). Nat. Meth. 2006, 3, (10), 793-796.


18

Page 19 of 20 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

Biomacromolecules

16. Xu, K.; Babcock, H. P.; Zhuang, X., Dual-objective STORM reveals three-dimensional filament organization in the actin cytoskeleton. Nat. Meth. 2012, 9, (2), 185-188. 17. Vaughan, J. C.; Jia, S.; Zhuang, X., Ultrabright photoactivatable fluorophores created by reductive caging. Nat. Meth. 2012, 9, (12), 1181-1184. 18. Bates, M.; Huang, B.; Dempsey, G. T.; Zhuang, X., Multicolor super-resolution imaging with photo-switchable fluorescent probes. Science 2007, 317, (5845), 1749-1753. 19. Terpetschnig, E.; Szmacinski, H.; Ozinskas, A.; Lakowicz, J. R., Synthesis of squaraineN-hydroxysuccinimide esters and their biological application as long-wavelength fluorescent labels. Anal. Biochem. 1994, 217, (2), 197-204. 20. Axelrod, D., Carbocyanine dye orientation in red cell membrane studied by microscopic fluorescence polarization. Biophys. J. 1979, 26, (3), 557-573. 21. Tangpakdee, J.; Tanaka, Y., Characterization of sol and gel in Hevea natural rubber. Rubber Chem. Technol. 1997, 70, (5), 707-713. 22. Dempsey, G. T.; Vaughan, J. C.; Chen, K. H.; Bates, M.; Zhuang, X., Evaluation of fluorophores for optimal performance in localization-based super-resolution imaging. Nat Meth 2011, 8, (12), 1027-1036. 23. Shim, S.-H.; Xia, C.; Zhong, G.; Babcock, H. P.; Vaughan, J. C.; Huang, B.; Wang, X.; Xu, C.; Bi, G.-Q.; Zhuang, X., Super-resolution fluorescence imaging of organelles in live cells with photoswitchable membrane probes. Proc Natl Acad Sci USA 2012, 109, (35), 13978-13983. 24. Hasma, H., Proteolipids of natural rubber particles [Hevea brasiliensis]. Short communication. J. Natural Rubber Res. 1987. 25. Sansatsadeekul, J.; Sakdapipanich, J.; Rojruthai, P., Characterization of associated proteins and phospholipids in natural rubber latex. J. Biosci. Bioeng. 2011, 111, (6), 628-634. 26. Tanaka, Y.; Sakdapipanich, J. T., Chemical structure and occurrence of natural polyisoprenes. In Biopolymers, Koyama, T.; Steinbuchel, A., Eds. Wiley-VCH: Germany, 2001; Vol. 2, pp 1-25. 27. Eng, A.; Ong, E., Plastics Engineering: Handbook of Elastomers. 2000; Vol. 61, p 29-59. 28. Toki, S.; Hsiao, B. S.; Amnuaypornsri, S.; Sakdapipanich, J., New insights into the relationship between network structure and strain-induced crystallization in un-vulcanized and vulcanized natural rubber by synchrotron X-ray diffraction. Polymer 2009, 50, (9), 2142-2148. 29. Berthelot, K.; Lecomte, S.; Estevez, Y.; Zhendre, V.; Henry, S.; Thévenot, J.; Dufourc, E. J.; Alves, I. D.; Peruch, F., Rubber particle proteins, HbREF and HbSRPP, show different interactions with model membranes. Biochim. Biophys. Acta -Biomembr. 2014, 1838, (1, Part B), 287-299. 30. Wadeesirisak, K.; Castano, S.; Berthelot, K.; Vaysse, L.; Bonfils, F.; Peruch, F.; Rattanaporn, K.; Liengprayoon, S.; Lecomte, S.; Bottier, C., Rubber particle proteins REF1 and SRPP1 interact differently with native lipids extracted from Hevea brasiliensis latex. Biochim. Biophys. Acta -Biomembr. 2017, 1859, (2), 201-210.


19


Page 20 of 20

Table of Contents Graphic


20

Super-Resolution Fluorescence Imaging of Spatial Organization of

Recommend Documents