Protein Mixture Segregation at Coffee-Ring: Real-Time Imaging of

Subscriber access provided by HKU Libraries

Article

Protein Mixture Segregation at Coffee Ring: Real-Time Imaging of Protein Ring Precipitation by FTIR Spectromicroscopy Sun Choi, and Giovanni Birarda J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/acs.jpcb.7b05131 • Publication Date (Web): 10 Jul 2017 Downloaded from http://pubs.acs.org on July 16, 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.

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

Page 1 of 24

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

The Journal of Physical Chemistry

Protein Mixture Segregation at Coffee Ring: Real-time Imaging of Protein Ring Precipitation by FTIR Spectromicroscopy Sun Choi*,‡,1, Giovanni Birarda*,‡,2

1

Center for Urban Energy Research, Korea Institutes of Science and Technology, 5 Hwarang-ro 14-gil,

Seongbuk-gu, Seoul 02792, Republic of Korea 2

Elettra Sincrotrone Trieste, Strada Statale 14 - km 163,5 in AREA Science Park 34149 Basovizza,

Trieste, ITALY Corresponding Author

* Giovanni Birarda, PhD, Elettra-Sincrotrone Trieste, Strada Statale 14 - km 163,5 in AREA Science Park 34149 Basovizza, Trieste, ITALY

E-mail: [email protected].

During natural drying process, all solutions and suspensions tend to form the so-called “coffee-ring” deposits. This phenomenon, by far, has been interpreted by the hydrodynamics of evaporating fluids. However, in this study, by applying Fourier transform infrared imaging (FTIRI), it is possible to observe the segregation and separation of a protein mixture at the “ring”, hence we suggest a new way to interpret “Coffee ring effect” of solutions. The results explore the dynamic process that lead to the ring formation in case of model plasma proteins, such as BGG (Bovine γ Globulin), BSA (Bovine Serum Albumin), and Hfib (Human Fibrinogen) and also report fascinating discovery of the segregation at the ring deposits of two model proteins BGG (Bovine γ Globulin), BSA (Bovine Serum Albumin),

ACS Paragon Plus Environment

1


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

Page 2 of 24

which can be explained by an energy kinetic model, only. The investigation suggests that the coffee ring effect of solute in an evaporating solution drop is driven by an energy gradient created from change of particle-water-air interfacial energy configuration.

KEYWORDS: Coffee ring, Protein segregation, IR spectromicroscopy, Non-equilibrium processes, Evaporating kinetics Coffee ring deposition of particles in an evaporating drop of colloidal suspension or solution is a commonly observed phenomenon in daily life and its dynamics has been explored for past several decades 1-8. By far, mainly micro/nano-scale colloids have been studied with a wide-range of measurement technologies: from conventional optical microscopy 1,4,6,9-18 to fluorescence microscopy 19-23

from Raman spectroscopy 24-27, to X-ray spectroscopy 2,28,29. For the interpretation of colloidal

coffee rings, generally, hydrodynamic is applied to the particle constituency at the vicinity of threephase contact line 1,2,4,7,8,13,19,21,30-39. The system of interest can be solved either analytically 1,2,4,21,30,32,35,37,39 or with computational modeling 9,13,28,40-42. More importantly, to validate the proposed models, the data from the theory have to be compared with the experimental results. Micro-scale colloids are generally optically distinguishable from the medium and transient ring-growth profiles are to be obtained from visible microscopy imaging 1,4,9,11-18,22,42. However, this methodology cannot be applied to solutions: at first, the behavior of solutes-from dissolved to precipitated phases- do not follow conventional rules of hydrodynamics because as the solute precipitation continues, the interspacing between solute molecules reduces to less than the solutes dimension and the medium that carries it cannot be longer considered a fluid. This complexity may be the reason there are only few works on ring formation from protein solutions 19,20,24,26,27,43-45, blood 43,46-51 and other body fluids. 52,53 Hence, aiming to gain a better understanding of coffee ring formation for biologically relevant fluids, as plasma serum for example, we decided to study this phenomenon starting from solutions of individual plasma proteins, a simplified model and then make it more complex by mixing the proteins together.


2

Page 3 of 24

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


In this work, we developed a real-time imaging method capable to monitor protein precipitation at coffee ring region of an evaporating drop by utilizing Fourier Transform Infrared Imaging (FTIRI) (Figure 1a-b). Here, we, report also on the segregations of model proteins at different layers of the coffee ring region. Generally, it is thought that solvent-evaporation drives the coffee ring formation of the solutes in either solution or suspension, however, fluid-alone is not able to account for the separation behavior of different solutes in solution-phase. To explain the dynamics of protein segregation at the ring region, there is a need to apply different approach. To explain the experimental results, a new energy-based kinetic model is to be introduced, in order to understand transport behavior of multi-component systems in non-equilibrium processes.

Figure 1. a. Graphical representation of the experimental process. I) a 2 µl drop of proteins’ solution is deposited on the silicon surface, either with or without FOTS. II) The transmission measure is started and t0 is defined, at this stage the water is too thick and the signal s beyond the linear regime (above 1.0 a.u.) for the most part of the droplet but very close to the edge, from this moment a new map is acquired every ~10 seconds. The red square represents the Region of interest (ROI) measured III) The final step of the process, the formation and drying of the coffee ring, as can be seen in the spectra series below, all the signals are below 1.0 a.u. and there is no water, the colors match those in the inset. In the inset b) we can see an example of the spectrum of water (blue), water plus protein (green) and dry protein (orange). In dashed lines the baselines


3


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

Page 4 of 24

used for the water and protein signals integration. Arrow point to 2125 cm-1 water combination band and to amide II 1550 cm-1.

The experimental data will be presented both as time series and chemical images of the drying/precipitation process of the protein droplets (see Supplementary Figure S1). At the beginning, the signal of proteins is overwhelmed by the one water, so it is difficult to detect them. Then, as the water thickness diminishes with water evaporation, with the drop still pinned, protein becomes detectable, after that, we assist to a concentration of it at the edge, and that is considered the initial time of the formation of the coffee ring (see Supplementary Figure S2 and S3). At this stage, the ring is still connected to the whole droplet so there is exchange of material from and to it. Near the solid part, the solution is not so viscous to hinder this exchange. After this point, the growth rate increases rapidly and we assist to the formation of the real coffee ring, which is completed when the water droplet de-pins from it; the de-pinning process causes a slight decrease in the signal, and we hypothesize that it is due to the solubilization of a small amount of the protein from the ring.

Figure 2. Time evolution profiles of the ring growth for BSA (circles), BGG (diamonds) and Hfib (triangles). In solid lines the profiles for protein ring formation on hydrophobic surfaces, in dashed lines the temporal evolution on hydrophilic surfaces. The ordinates axis is in micron square since the value is the result of the sum of the protein thickness over the


4

Page 5 of 24

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


length of the droplet. The error bars represent the standard error over the different measures. Abscissa axis is time in seconds.

In Figure 2 we can observe the profiles obtained by averaging 5 runs of the drying processes of 2µL droplet BGG (Bovine γ Globulin), BSA (Bovine Serum Albumin), and Hfib (Human Fibrinogen) on a hydrophobic surface (solid lines). In dashed lines, the average profiles obtained during the drying process on a hydrophilic surface, i.e. the native silicon oxide on a silicon wafer. From the analysis of the graph we can observe that BGG always reaches the detection point, as defined in Supplementary Figure S2, before BSA, clue of a faster accumulation at the edge of the drop, nevertheless BSA growth rate is faster.

Figure 3. FPA images of the of the ring growth for BSA in D2O. Images were obtained by calculating the peak height of the second derivative spectra at 1545 cm-1, results have been multiplied per (-1) to obtain positive values. The bold dashed line marks the inner edge of the protein ring, the medium dashed line the center of the final ring that corresponds also to the outer part of the protein exchange area during the ring growth. The thin line is the outer edge of the coffee ring, where the precipitation starts.

In order to make more evident the precipitation process and the ring formation we performed an experiment with BSA in D2O. In this solvent, the Amide II signal is more easily discernable, and we can use it to map the protein concentration at the edge of the droplet. From Figure 3 we can see that in a few minutes before the starting of the precipitation there is more protein far from the edge then at the ACS Paragon Plus Environment

5


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

Page 6 of 24

interface, because the drop there is thinner. With time, the protein amount tends to become homogeneous, i.e. the concentration of protein increases at the edge, but the solution is not viscous, suddenly at t=10 min, for this experiment, there is the precipitation of the protein, red part. In the following minute we can see the growth of the ring as proteins are transported to the edge, until the droplet de-pins. For Hfib, a different behavior was observed. If compared to the BSA and BGG, it reached a detectable concentration at the edge before the other, but the signal had a slower increase rate during the accumulation phase. Then the major part of the protein is dragged away during the de-pinning process, hint of a weak-to-none interaction with the substrate. It implies that, unlike BSA and BGG, Hfib solution does not show strong coffee ring effect. The drying time scales and area almost linearly scale with the volume of the drop (its verification is contained in Supporting information with Figure S4). After single protein experiments we performed measures on mixtures. In Figure 4 a-b are presented the IR and optical images illustrating the drying process of a BSA-BGG mixture respectively on hydrophobic and hydrophilic surfaces. On both surfaces, we observe the segregation of the proteins at different layers of the coffee ring deposits. This “separation” though more defined boundaries on hydrophobic surfaces. It is also noteworthy that in presence of Fibrinogen in the mixture hinders the segregation, Figure 4c-d.


6

Page 7 of 24

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


Figure 4. a-b. RGB composite infrared and optical images of the drying and segregation process of the 2ul droplet of BSA, BGG proteins mixture respectively on hydrophobic surface (a) and hydrophilic one (b). The three moments presented are t0, right before the de-pinning, when a solid part of the ring is still on contact with a thin layer of water and the final dry ring situation. Red-pink color is the intensity of the signal of BSA 1657 cm-1, in second derivative multiplied per -1, to have positive values and the cyan-blue is the BGG signal at 1638 cm-1. In panel c-d there are the final points for the three proteins mixture BSA-BGG-Hfib, in this case no segregation can be observed. The optical images do not belong to the same experiment, since for technical limitation to have a good temporal resolution on IR repeated measures only the IR signal have to be collected. Scalebar is 50 µm.

In order to explain different precipitation rates of proteins and also the segregation behaviors of protein mixture solutions, a new model needs to be established. Since conventional hydrodynamics are not applicable to current system, a new approach that incorporates three-phase (solid, liquid, air), multibody dynamics is postulated. Inspired by the previous work that discusses colloid behaviors at air-water interface 54, a system of interest is defined (Figure 5a-b) and it is solved by assuming that it will evolve towards the minimization of total energy.


7


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

Page 8 of 24

In our analysis, it is assumed that over the course of protein segregation, there exist two areas: (1) the area, very thin, where the drop pins and (2) the rest of the drop. At the pinning point the evaporation is faster than the diffusion of molecules in solution; here we have a local increase in concentration of proteins, hence in viscosity, but as data in Figure 3 show, it stops as soon as the protein precipitates. Then we have the depletion of protein from the solution towards the edge to compensate the loss caused by the precipitation. This transport continues as the water evaporates, until the de-pinning of the droplet meniscus. (See Figure 3(D2O experiment).) As a result, we judge that viscosity may not play a major role in protein since there is always a redistribution of material mass from the solution and to the solid phase. A gain of Total Gibbs free energy of the system, ∆ , before and after ring precipitation is expressed as

∆ = ∆ − ∆ In this analysis, the role of entropy is at first, to differentiate account of the transport of globular proteins from that of non-globular proteins and apply different kinetics to each of them. In three-protein mixtures in our experiment, for instance, Fibrinogen, a non-globular protein, tends to segregate upward to air-water interface, to increase its degree of freedom (entropy), while the other two globular proteins move to the edge of the droplet, hence why we set aside drying dynamics of Fibrinogen to the later part of the analysis. For globular proteins, the enthalpy variation is more dominant than the entropy one (see reasoning in supplementary materials), the free energy gain is approximated as: ∆ ≈ ∆ = ∆


8

Page 9 of 24

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


Figure 5. a-b. Illustration of modelling domain with respect to key parameters (a) and illustration of protein segregation dynamics (b).

∆ is expressed as ∆ = 2 − / − 2 / − 2 / sin − Where: R: Radius of a sphere particle θ: Contact angle of air-water meniscus with respect to a substrate σ/: Surface energy of particle-air interface σ/ : Surface energy of particle-water interface σ/ : Surface energy of water-air interface

During the evaporation process if If ∆

0, proteins will form a ring until the evaporation is

completed, otherwise, if ∆ " 0, particles move inward as the contact-line of the droplet recedes. As the evaporation continues, at certain time, if ∆ intercepts zero point and becomes ∆

0, the ring

formation will begin. However, if ∆ " 0 throughout the all evaporation process, particles will be transported in the center and will form a cluster. From ∆ we can deduce the surface energy flux, ∆. (See derivation processes in supplementary materials.) ∆ can represent the amount of solute’s flux that will be generated from droplet center to its edge. ACS Paragon Plus Environment

9


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

Page 10 of 24

∆ is expressed as ∆ = ⁄ −

$%&'( $ )

⁄ −

)%$ )

/

(1)

From this expression, ∆ is created by an energy gradient from particle-water interfacial energy, that is the sum of water-air interfacial energy and particle-air interfacial energy. ∆ is determined by σ/ only(see reasoning in supplementary materials) and relative magnitudes of σ/ can be estimated by iso-electric point. For example, iso-electric point of BSA and BGG are 4.8, 6.1, respectively. In Deionized water (pH 7), both proteins show negative charge and the zeta-potential of BSA is higher than BGG. Which implies that, in BSA solutions, more water molecules will come to the constituency of protein molecules than that of BGG molecules. As a result, / * ∆|+,

+,

/ *

+--

and it leads to

∆|+--

Under this premises we can explain the earlier precipitation of BGG in respect to BSA and also the segregation behavior of BGG-BSA mixtures at the vicinity of the three-phase contact line in early stage of drying. In other words, BGG precipitation rate is faster than BSA in the beginning, but, then slows down as BGG solutes are depleted. For each protein solution, there exists a critical contact angle, / , that satisfies ∆ " 0. According to equation (1), higher ⁄ yields higher / . Hence, 0 |+-- " 0 |+, . As the protein concentration increases, the overall surface energy of air-liquid surface increases as the protein density at the interface increases, as a result, the contact angle increases. However, in our experiment, as shown in Figure 3. during the drying of protein solution, the contact line pins at the beginning of the drying, consequently, contact angle decreases because the pinned proteins prevents the meniscus from receding until it reaches down to a critical angle31. This contact angle decrease, actually promotes the segregation of proteins because it drives the selective “salt-out” of proteins of high surface energy. In BSA-BGG mixtures, as shown in Figure 4, if 12 " 13 |455 " 13 |467 (where 12 refers to initial contact angle of protein mixture solution, 13 |455 to critical contact angle of 8-globulin, 13 |467 to critical contact angle of BSA on a hydrophobic surface, respectively), when the contact angle of the


10

Page 11 of 24

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


protein mixture, 9, decreases from 12 to 13 |455, BGG will precipitate at the vicinity of the contact line while BSA is still remain dissolved. After 9 decreases down to 13 |467 , the rest of BGG & whole BSA are “squeezed” and as a result, forming a inner layer in the coffee ring. After the contact line recedes, and the protein precipitation is completed. In order to validate this hypothesis, the critical contact angle of pure protein solution and protein mixtures were extrapolated from the data of real-time drying processes (see Methods in suppplementary materials). The estimated critical contact angles of BSA, BGG, and mixtures of BSA and BGG on both hydrophilic and hydrophobic surfaces are summarized in Table 1. By analyzing the angle differences between BSA and BGG on both hydrophilic and hydrophobic we can foresee the BSA and BGG separation. The angle difference is calculated to be around 5.4º on hydrophobic surface while it is around 2.2º on hydrophilic surface. The angles for mixtures are almost the average value of those calculated for pure proteins, suggesting an internal consistency in the estimations. This results support the earlier precipitation of BGG in respect to BSA as well as the segregation of BSA-BGG mixture are driven by surface charge differences between BGG and BSA.


11


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

Materials

DI water

BSA

BGG

Hfib

Page 12 of 24

BSA-BGG

BSA-BGGHfib

Exp.#

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

Hydro

philic

phobic

philic

phobic

philic

phobic

philic

phobic

philic

phobic

philic

phobic

1

13.57

102.7

6.64

10.14

8.02

13.81

5.67

4.46

8.54

10.42

4.1

2.94

2

12.17

101.4

5.31

11.03

8.48

15.54

5.58

4.45

7.36

12.73

5.45

9.54

3

13.96

103.7

6.18

8.54

8.33

15.2

6.12

4.17

8.19

15.48

4.22

7.12

4

11.76

102.7

6.79

10.01

8.29

15.86

4.32

7.13

8.22

13.65

5.34

2.24

5

12.87

101.0

5.79

7.84

8.63

14.11

5.63

3.12

7.26

11.88

5.47

8.05

Average

12.87

102.3*

6.14

9.51

8.35

14.90

5.46

4.67

7.91

12.83

4.92

5.98

Table 1. Critical contact angle of DI water, BSA, BGG, Hfib, BSA-BGG mixtures, and BSA-BGGHfib mixtures. (*: measured by force tensiometer)

The drying dynamics of last protein of our study, fibrinogen (Hfib) is explored by separate experiments. We deposited Hfib solution and measured with the same procedure as applied to BSA and BGG. The observed drying dynamics are different from the globular proteins. A lower fraction of solutes was deposited at the contact line and the rest of the solutes was deposited inside the ring region. As shown in Figure 2, the mass accumulation was slower than the other two model proteins, and after the de-pinning of the droplet the signals decreases ~ 20 % in case of hydrophobic surface and ~ 80% for hydrophilic ones. This result was reinstated by the results in Table 1 to globular proteins. The critical contact angle of Hfib was measured to be around ~10° lower than the one of globular proteins and the discrepancy of the angle between on hydrophobic and hydrophilic is very low. This result implies that 12 ACS Paragon Plus Environment

Page 13 of 24

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


Hfib is less likely to be moved to the edge and concentrated solutes were identified by infrared only after significant time passes and the meniscus forms a small angle with the substrate. Reasoning for explaining drying dynamics of Hfib is contained in supporting information. The differences found in drying dynamics between globular and non-globular proteins alter the segregation patterns we observed in mixture-state. In Figure 3c-d, it is shown that Hfib layer across overlaps the coffee region in which BSA and BGG are deposited. According to the individual protein experiment, the time of the starting of the precipitation is Hfib

Protein Mixture Segregation at Coffee-Ring: Real-Time Imaging of

Recommend Documents