Supercharged proteins and polypeptides for advanced materials in chemistry and biology
Ma, Chao
IMPORTANT NOTE: You are advised to consult the publisher's version (publisher's PDF) if you wish to cite from it. Please check the document version below.
Document Version
Publisher's PDF, also known as Version of record
Publication date: 2019
Link to publication in University of Groningen/UMCG research database
Citation for published version (APA):
Ma, C. (2019). Supercharged proteins and polypeptides for advanced materials in chemistry and biology. University of Groningen.
Copyright
Other than for strictly personal use, it is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), unless the work is under an open content license (like Creative Commons).
Take-down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately and investigate your claim.
Downloaded from the University of Groningen/UMCG research database (Pure): http://www.rug.nl/research/portal. For technical reasons the number of authors shown on this cover page is limited to 10 maximum.
49
Chapter 2
Tuning Ice Nucleation with Supercharged Polypeptides
This chapter has been published:
Huige Yang,+Chao Ma,+ Kaiyong Li, Kai Liu, Mark Loznik, Rosalie Teeuwen, Jan C. M. van Hest, Xin Zhou, Andreas Herrmann, and Jianjun Wang
50
The ability to tune ice nucleation via external interventions is of importance to wide-ranging fields as diverse as cloud seeding, deicing of airplanes or cryopreservation of cells and organs. However, it is still a challenge to control ice formation and the parameters influencing this process remain elusive. In this manuscript, we present a way of how to precisely tune ice nucleation on surfaces with the aid of supercharged unfolded polypeptides (SUPs). These materials are produced by genetic engineering, allowing absolute control over nature of charges, chain length and charge density in addition to the monodisperse character of the resulting protein polyelectrolytes. By variation of the nature of charges and charge density, we show that positively charged SUPs facilitate ice nucleation, while negatively charged SUPs suppress it. Moreover, ice nucleation can be further tuned by adjusting different charge densities within SUPs. Regarding the underlying mechanism, we propose that the tuning of ice nucleation with SUPs is achieved via the structural change of interfacial water caused by the local electric field near SUPs, which consequently alters the energy barrier of ice nucleation. From a materials point of view, SUPs are biodegradable and exhibit low toxicity, which might qualify them as cryo-protectants in a medical context in the future.
J.J.W. and A.H. conceived the idea. C.M., K.L. and M.L. fabricated and expressed the supercharged proteins as well as ELP90. R.T. and J.C.M.H. provided the plasmid of ELP90. H.Y., K.Y.L.and X.Z. performed the surface characterization. The introduction, results and discussion parts relating to SUPs were written by C.M. and A.H..
51
1. INTRODUCTION
The ability to tune ice nucleation via external interventions[1] is of importance to wide-ranging fields as diverse as cloud seeding, cryopreservation of cells, tissues, and organs.[2] Among various external auxiliaries, electric field affects ice nucleation in both naturally occurring situations and industrial environments.[3] A large number of studies have indicated that local electric field near charged surfaces can reorient water molecules and thus may affect ice nucleation.[4] Recently, it was demonstrated that ice nucleation can be promoted or inhibited on positively or negatively charged surfaces.[5] On the other hand, it is widely believed that charged surfaces exhibit a good wettability, which is generally a favorable factor for ice nucleation, as indicated by the classical nucleation theory (CNT).[6] Thus, understanding of ice nucleation on differently charged surfaces remains elusive.
Herein, we report the tuning of ice nucleation through systematic control of both surface charge and charge density, which is achieved via modifying solid surfaces with supercharged unfolded polypeptides (SUPs).[7] Compared to their chemically synthesized counterparts, SUPs are of low toxicity and are biodegradable.[7a] Moreover, their fabrication procedure based on genetic engineering allows absolute control over nature of charges, chain length, and charge density, combined with a monodisperse character.[8] These structural attributes of SUPs open up new possibilities for the study of ice nucleation. The results show that surfaces modified with positively charged SUPs facilitate ice nucleation, while surfaces modified with negatively charged SUPs suppress it. The theoretical analysis based on CNT shows that SUPs change the difference in the electrostatic energy density between ice nucleus and the same volume of water under an electric field as well as surface tensions of ice nucleus and water with the substrate,[9] which consequently vary the energy barrier of ice nucleation, and thus explains our experimental observations. Supercharged unfolded elastin-like polypeptides (ELP) consist of repetitive sequences of five amino acids (GXGVP)n, where G is glycine, V is valine, P is proline, and X denotes either positively charged lysine (K) or negatively charged glutamic acid (E). The chain length and charge density were adjusted by cloning and recombinant synthesis. A series of SUPs with various charges and ELP90 without any charges in the repetitive motif[10] were expressed in Escherichia coli (Table 1). The number in the SUP name indicates the number of glutamic acid or lysine residues within the structure.[7]
52
Table 1. The name, isoelectric point, primary structure, and molecular weight of
SUPs used in this work.
2. RESULTS AND DISCUSSION
Via mussel-inspired surface chemistry,[11] 3,4-dihydroxyphenylalanine (DOPA) was first polymerized to form an adherent deposit on silicon wafers, then these SUPs were tethered onto the DOPA-functionalized surfaces due to the reactivity of the catechol group of polydopamine toward nucleophiles. Two oppositely charged SUP
Name of SUPs Isoelectric point (PI)
Formula Molecule Weight (Da) E18 4.16 GAGP[(GVGVP)(GEGVP)9]2GWPH6 10193 E36 3.78 GAGP[(GVGVP)(GEGVP)9]4GWPH6 18922 E72 3.45 GAGP[(GVGVP)(GEGVP)9]8GWPH6 36512 Half-charged E35 (HC-E35) 3.79 GQ [(GVGVPGEGVP)5]7GWH6C 31217 Zwitterionic 16 (KE16) 7.17 GQ[GVGVP(GKGVPGEGVP)4]4GWL H10C 17709 K18 11.37 GAGP[GVGVP (GKGVP)9]2GWPH6 10176 K36 11.54 GAGP[GVGVP (GKGVP)9]4GWPH6 18888 K72 11.85 GAGP[GVGVP (GKGVP)9]8GWPH6 36446 Half-charged K30 (HC-K30) 11.54 GQ [(GVGVPGKGVP)5]6GWH6C 26944 ELP90 7.25 GSSHHHHHHSSGLVPRGSHMLEKR EAEAGPVP(GGGVPGVGVPGVGVP GGGVPGLGVPGVGVPGVGVPGVG VPGGGVPGLGVP)9GGGA 39671
53
random coils, i.e., K36 and E36 were selected to investigate the effects of charge signs of SUPs on ice nucleation (Figure 1a).
As presented in Figure 1b, the pristine silicon wafer exhibits an average surface roughness (Ra) of 0.53 ± 0.14 nm. Surface functionalization with DOPA and the subsequent modification with SUPs lead to slightly rougher surfaces with Ra of 1.43 ± 0.24, 1.00 ± 0.15, and 1.61 ± 0.29 nm for the DOPA-functionalized surface, K36- and E36-modified surfaces, respectively (Table S2, Supplemetary Data). The pristine silicon wafer shows a contact angle of 49.8° ± 1.7°. After polydopamine functionalization, the substrate becomes slightly more hydrophilic with a decreased contact angle of 38.8° ± 1.3°. Finally, the contact angles increase to 66.4° ± 0.4° and 67.6° ± 1.8° after the immobilization of SUPs, K36 and E36, respectively (see Figure 1c). The variation of surface roughness (Table S2, Supplemetary Data), the values of contact angle and contact angle hysteresis (Figure 1c), and surface chemistry characterized by X-ray photoelectron spectroscopy (XPS) (Figure S6, Supplemetary Data) verify the successful modification with polydopamine and afterward with SUPs.
Figure 1. a) Scheme of modifying solid surfaces with SUP random coils. The charged
54
residues K and E in SUPs are represented by blue and red colors, respectively. b) Atomic force microscopy height images (1 μm × 1 μm) of the pristine silicon wafer, the DOPA-functionalized surface, and the surfaces modified with K36 and E36. c) The contact angles and contact angle hystereses of the pristine silicon wafer, the DOPA-functionalized surface, E36- and K36-modified surfaces.
The ice nucleation temperature was determined by a homebuilt apparatus. [12] It was determined by measuring the freezing temperatures of condensed microdroplets on SUP modified surfaces. Thereby, the samples were placed in a closed chamber with a relative humidity of 100%. Measuring the freezing temperatures of condensed water droplets on solid surfaces has been adopted by many groups for the investigation of the effect of solid surfaces on ice nucleation [5] although the freezing of condensed water microdroplets is a complex process and can be triggered by the freezing of proximal ones. [13]
We extract sequential images of water microdroplets condensed onto the DOPA-functionalized surface (top), K36- (middle) and E36-modified surfaces (bottom) before (left column) and after freezing (right column) (Figure 2a). The insets of Figure 2a show magnified images of water microdroplets encircled by red lines to show the sudden variation in the opacity before and after freezing, which indicates the occurrence of ice nucleation. It should be pointed out that ice formation in microdroplets is divided into two stages, i.e., nucleation and growth.[14] As compared with the slower ice nucleation, the rapid growth leads to the instant freezing of water microdroplets upon the formation of ice nuclei.[15] Thus, the sudden onset of freezing we observed can be identified as ice nucleation.
Figure 2b shows that ice nucleation occurs at −21.0 ± 0.5 °C on the DOPA-functionalized surface with a cooling rate of 2 °C min−1. In contrast, ice nucleation temperatures on the surfaces modified by SUPs (K36 and E36) differ from each other and from that on the surface without SUPs. Ice nucleation occurs at −19.2 ± 0.7 °C on the surface modified with K36, whereas the nucleation temperature on the surface modified with E36 is −22.8 ± 0.6 °C. The results demonstrate that ice nucleation is facilitated on the surface modified with the positively charged SUP,
55
while it is suppressed on the surface modified with the negatively charged SUP. Moreover, investigation of the freezing of single macroscopic droplets on SUP-modified surfaces was carried out to verify the data obtained with microdroplets. Figure S7 (Supplemetary Data) shows a significant difference of freezing temperatures on E36- and K36-modified surfaces. This consolidates the effects of SUPs on ice nucleation.
Figure 2. a) Optical microscopy images of water microdroplets condensed onto the
DOPA-functionalized surface, K36- and E36-modified surfaces before (left column) and after freezing (right column). Cooling rate: 2 °C min−1; scale bar: 100 μm. b) Temperatures of ice nucleation on the DOPA functionalized surface, K36- and E36-modified surfaces. Cooling rate: 2 °C min−1. c) Cooling rate dependence of ice nucleation temperature on K36- and E36-modified surfaces.
Figure 2c indicates that the ice nucleation temperature decreases with increasing the cooling rate. In our experiment, the droplets are subjected to a constant cooling rate γ = dT/dt. When the surface temperature is decreased gradually, the fraction (x) of frozen droplets increases. As previously reported by our group[12] and consolidated
56
in Figure S8 (Supplemetary Data), ice nucleation initiates at the liquid–solid interface. We define a temperature Txf at which a fraction (x) of the droplets is frozen according to [16]
(1)
where J(T) is the ice nucleation rate. This relationship shows that an increase in the cooling rate results in a decrease of Tx
f, providing a reasonable explanation for the observed cooling-rate dependence of the ice nucleation temperature, which agrees with the results reported by others.[16b] The extrapolation to zero cooling rate results in the freezing temperatures of −19.0 and −22.5 °C for the surfaces modified by positively charged K36 and negatively charged E36, respectively (see Figure S9 and the detailed discussion given in the Supplemetary Data). The minute difference between the freezing temperatures obtained at zero cooling rate and the ones we adopted (2 °C min−1) verifies the validity of our method.
We expressed two series of SUPs including positively charged SUPs (K18, K36, and K72) and negatively charged SUPs (E18, E36, and E72), of which each has an equal charge density but different chain length, to probe the effect of molecular weight on ice nucleation. The temperatures of ice nucleation on the surfaces modified with positively or negatively charged SUPs show very little variation within each series (Figure 3a), indicating that the effect of chain length on ice nucleation is negligible. Moreover, these experiments show that a slight variation in surface roughness does not affect ice nucleation (Table S2, Supplemetary Data).
dT
fxd
g
=
ln(1
-
x)
s
´
J
s(T
fx)
57
Figure 3. a) Temperatures of ice nucleation on the surfaces modified with two series of
SUPs, and each series possesses an equal charge density but different chain lengths. b) Charge density dependence of ice nucleation temperature on the surfaces modified with a series of positively and negatively charged SUPs.
Subsequently, K36, HC-K30, KE16, ELP90, HC-E35, and E36 with different charge densities (HC stands for halfcharged) and charge signs are utilized to demonstrate the ability of SUPs to tune ice nucleation. Considering the neutral nature of freshly condensed water, the net charge density subjects to the order: K36 = E36 > HC-K30 = HC-E35 > KE16 > ELP90. Figure 3b shows that the K36-modified surface exhibits the highest ice nucleation temperature. Conspicuously, a
58
decrease of charge density of positively charged SUPs leads to a decrease in the temperature of ice nucleation. In contrast, a decrease in charge density of negatively charged SUPs results in an increase of ice nucleation temperature. For positively charged SUPs, the results indicate that increasing charge density further promotes ice nucleation, whereas for negatively charged SUPs increasing charge density further inhibits it. The repeatability of SUPs in tuning ice nucleation is further confirmed by measuring the freezing temperatures of macroscopic droplets being placed on surfaces modified by K36, HC-K30, HC-E35, and E36 (Figure S10, Supplemetary Data).
Figure 4. a) Optical microscopy images of ice nucleation on the surfaces modified with
SUPs, K36 (top) and E36 (bottom), at −19.0 °C. Scale bar: 100 μm. b) Delay times of ice nucleation on the surfaces modified by K36, HC-K30, HC-E35, and E36.
To further consolidate the ability of SUPs to tune ice nucleation, we measured the delay time of ice nucleation. It is defined to be the difference between starting time
59
(the moment when a surface reaches a targeted temperature) and the time taken for ice nucleation of the microdroplet to occur. After a delay time of 54 s, ice nucleation occurs on the surface modified with K36 (Figure 4a). Notably, the delay time is greatly increased up to 1974 s on the surface modified with E36. Figure 4b indicates that a decrease in charge density of positively charged SUPs leads to an increase of the delay time, while a decrease in charge density of negatively charged SUPs leads to a decrease of the delay time. The nucleation delay time is inversely proportional to the nucleation rate, which depends on the supercooling as expressed by the following equation [17]
J=K(T)exp(-ΔG(T)/kBT) (2)
where J is the nucleation rate and K(T) is a kinetic factor representing the attraction of free water molecules to a forming ice embryo. kB is the Boltzmann constant. The nucleation energy barrier under an electric field is [18]
(3)
where is the shape factor of a spherical
cap-shaped ice nucleus with a contact angle θ on the SUP modified surface, γ is the surface tension between ice nucleus and water, ΔT is supercooling below the equilibrium melting temperature, ΔS is the entropy of fusion, and W is the difference in the electrostatic energy density between ice nucleus and the same volume of water under an electric field. Therefore, the delay time indicates that the nucleation energy barrier on four surfaces increases in the sequence of K36 < HC-K30 <HC-E35 < E36. The sequence of the increase in the nucleation energy barrier agrees well with that obtained from the investigation of freezing temperatures, which again demonstrates the validity of our experimental results.
2 3
)
Δ
Δ
-3(
16
)
(
=
Δ
T
S
W
πγ
θ
f
G
2 ) -)(1 + (2 4 1 = ) (θ cosθ cosθ f60
Ice nucleation on solid surfaces is heterogeneous and is therefore always affected by the structure of the interfacial water, i.e., the transition layer of a few nanometers between the substrate and ice nucleus/water.[12,17,19] Previously, W was estimated from polarization energy of ice nucleus under an uniform electric field, and it was concluded that positively and negatively charged surfaces with the same electric intensity raise equally ice nucleation temperature.[18] This contradicts with our experimental observations (Figure 3b). We propose that ice nucleus and water atop of the interfacial water are polarized by SUPs with different magnitudes. In addition, due to the asymmetric environment, the polarization of interfacial water is asymmetric on positively and negatively charged SUP surfaces. According to Equation (3), the ice nucleation energy barrier is mainly controlled via W and f(θ). W on the SUP-modified surface can be expressed as W = −aσ2 − bσ∆P0, where a and b are constants and ∆P0
is the difference in the spontaneous polarization between ice nucleus and water on the neutral surface. σ is the surface charge density of the SUPmodified surface. The contact angle of ice nucleus on the interfacial water and f(θ) are dependent on charge density of the substrate. For small σ, which is the case for the present experiments, we have
(4)
where θ0 is the contact angle of ice nucleus when σ is zero. Then we similarly have
f(θ) = f(θ0)(1 + cσ +···) by Taylor expansion, which indicates that the change of σ
leads to the variation in the energy barrier of ice nucleation. Our experimental results show that ice nucleation becomes easier with an increase of charge density on positively charged surfaces, while ice nucleation is suppressed by the increase of charge density on negatively charged surfaces. These findings indicate that the linear terms of W and f(θ) dominate the effect of surface charge in the current study.
3. CONCLUSION
In this chapter, ice nucleation was tuned by precise variation of the nature of charges and charge density of SUPs. This tailoring of freezing was enabled by the exact structural control of SUPs facilitated by genetic engineering. We show that positively charged SUPs facilitate ice nucleation, while negatively charged SUPs
+
+
+
)
(
-)
(
=
cosθ
0kσ
pσ
2γ
σ
γ
σ
γ
cosθ
w i, i s, w s,≈
61
suppress it. Ice nucleation was further tailored by varying the charge density of SUP backbone. The molecular weight of SUPs does not influence ice nucleation. These observations can be rationalized by considering the asymmetric polarization of interfacial water on positively and negatively charged surfaces. To make a better comparison of SUPs and other anti-icing systems, a table is prepared in Table S3. In the future, we will exploit SUPs for new anti-icing strategies and cryopreservation of biological matter.
4. EXPERIMENTAL PART AND SUPPLEMENTARY DATA
The preparation of samples for the ice nucleation investigation was carried out in a Class II Type A2 biosafety cabinet and the whole lab was air-conditioned to maintain a constant 25 °C. The tethering of SUPs on the DOPA-modified silicon surface follows the method used by Lee et al.[11] Quartz crystal microbalance with dissipation investigation shown in Figure S11 (Supplemetary Data) confirmed that no dissociation of DOPA or polypeptide occurred during the ice nucleation investigation. Ice nucleation examination was carried out in a small closed cell with a relative humidity of 100%. An average ice nucleation temperature was first obtained by the freezing of condensed water microdroplets in the window of observation under the optical microscope, and then it was again averaged over six different samples. The delay time of ice nucleation of macroscopic water droplets was also averaged over six different samples.
Cloning/Gene oligomerization
The building block of the Elastin like polypeptide (ELP) gene was ordered from Entelechon (Regensburg, Germany) and was delivered in the pEN vector. Gene sequences and respective amino acid sequences of monomers are shown in Figure S1. As the recognition sites of the restriction enzymes PflMI and BglI had to be preserved, one Valine residue per ten pentapeptide repeats was incorporated instead of a Glutamic acid or Lysine residue. The ELP gene was excised from the pEN vector by digestion with EcoRI and HinDIII and run on a 1% agarose gel in TAE buffer (per 1L, 108 g Tris base, 57.1 mL glacial acetic acid, 0.05 M EDTA, pH
62
8.0).The band containing the ELP gene was excised from the gel and purified using the QIAGEN spin column purification kit. pUC19 was digested with EcoRI and HinDIII and dephosphorylated. The vector was purified by agarose gel extraction following gel electrophoresis. The linearized pUC19 vector and the ELP-encoding gene were ligated and transformed into chemically competent DH5α cells (Stratagene, Cedar Creek, TX) according to the manufacturer’s protocol. Cells were plated and colonies were picked and grown overnight in LB medium supplemented with 100 µg/mL Ampicillin, and plasmids were isolated using the GeneJET Plasmid Miniprep kit. Positive clones were verified by plasmid digestion with PflMI and BglI and subsequent gel electrophoresis. The DNA sequence of putative inserts was further verified by DNA sequencing (GATC, Konstanz, Germany). Gene oligomerization was performed as described by Chilkoti and co-workers. [20]
63
Figure S1. Genes and corresponding polypeptide sequences of ELP E9 (a,
containing nine glutamic acids), ELP K9 (b, containing nine Lysines), Half charge series ELP E5(c, containing five glutamic acids), Half charge series ELP K5(d, containing five Lysines) and Zwitterionic KE series (e). Restriction sites flanking the insert gene are PflMI and BglI.
Expression vector construction
The expression vector pET25b(+) was modified by cassette mutagenesis, for incorporation of a unique SfiI recognition site and an affinity tag consisting of six histidine residues at the C-terminus (shown in Figure S2), as described before. [7a] Briefly, The modified pET25b(+) expression vector was digested with SfiI, dephosphorylated and purified using a micro-centrifuge spin column kit. The
64
repetitive ELP gene was excised from the pUC19 cloning vector using PflMI and BglI, and purified by agarose gel extraction following gel electrophoresis. The linearize vector and ELP-encoding gene were ligated with T4 ligase (Thermo Scientific), transformed into DH5α competent cells and screened as described above. For the construction of Halfcharged series as well as Zwitterionic KE16, the ELP gene was excised from the cloning vector using EcoRI and NdeI and ligated to pET25b(+) which was digested using the same restriction enzymes before, and transformed into DH5α competent cells. The constructs of pET25b-ELP were verified first by EcoRI and NdeI digestion (shown in Figure S3), and then sent for DNA sequencing.
Figure S2. Insert sequence of modified expression vector pET-25b (+)-SfiI-H6. The
vector contains a unique SfiI recognition site for inserting ELP genes and sequence encoding for a hexa-histidine (H6) tag at the C-terminus of the expressed protein via affinity chromatography purification.
65
Figure S3. Gel electrophoresis of ELP genes from expression vector pET25b. The
vectors containing the ELP gene were digested with EcoRI and NdeI and separated on a 1% agarose gel. DNA bands were visualized by ethidium bromide staining. Digestion produced the pET25b vector fragment of 5,451bp and the insert ELP gene fragment. Lane1, digested pET25b vector with an insert K18 of 330bp. Lane2, digested pET25b vector with an insert K36 of 630bp. Lane3,digested pET25b vector with an insert K72 of 1350bp. Lane4, digested pET25b vector with an insert Half charge K30 of 1080bp. Lane5, digested pET25b vector with an insert E18 of 330bp. Lane6, digested pET25b vector with an insert E36 of 630bp. Lane7,digested pET25b vector with an insert E72 of 1350bp. Lane8, digested pET25b vector with an insert Half charge E30 of 1260bp. Lane9, digested pET25b vector with an insert Zwitterionic KE16 of 600bp. M=DNA ladder.
Polypeptide expression and purification
E.coli BLR (DE3) cells (Novagen) were transformed with the pET25b expression vectors containing the respective ELP genes. For polypeptide production, Terrific Broth medium (for 1 L, 12 g tryptone and 24 g yeast extract) enriched with phosphate buffer (for 1 L, 2.31 g potassium phosphate monobasic and 12.54 g potassium phosphate dibasic) and glycerol (4 mL per1 L TB) and supplemented with 100 µg/mL ampicillin, was inoculated with an overnight starter culture to an initial optical density at 600 nm (OD600) of 0.1 and incubated at 37°C with orbital agitation at 250 rpm until OD600 reached 0.7. Polypeptide production was induced by a temperature shift to 30°C. Cultures were then continued for additional 16h post-induction. Cells were subsequently harvested by centrifugation (7,000 x g, 30 min, 4 ºC), re-suspended in lysis buffer (50 mM sodium phosphate buffer, pH 8.0, 300 mM NaCl, 20 mM imidazole) to an OD600 of 100 and disrupted with a constant cell disrupter (Constant Systems Ltd., DaventryNorthants, UK). Cell debris was removed by centrifugation (25,000 x g, 30 min, 4 ºC). Polypeptides were purified from the supernatant under native conditions by Ni-sepharose chromatography. Product-containing fractions were pooled and dialyzed against ultrapure water and then purified by anion exchange chromatography using a Q HP column. Product-containing fractions were dialyzed extensively against ultrapure water. Purified products were frozen in liquid nitrogen, lyophilized and stored at -20ºC until further use. For the expression of ELP90, protocol was provided by
66 Prof. Jan C. M. van Hest.[10]
Protein Characterization
The concentrations of the purified polypeptides were determined by measuring absorbance at 280 nm using a spectrophotometer (Spectra Max M2, Molecular Devices, Sunnyvale, CA). Product purity was determined by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on a 10% polyacrylamide gel. Afterwards, gels were stained with Coomassie staining solution (40% methanol, 10% glacial acetic acid, 1 g/L Brilliant Blue R250). Photographs of the gels after staining were taken with a LAS-3000 Image Reader (Fuji Photo Film GmbH, Dusseldorf, Germany). The result is shown as Figure S4, where the supercharged polypeptides exhibit different electrophoretic mobilities which is consistent with previous report. [7a]
67
Figure S4. Supercharged polypeptides sequences and 10% SDS-PAGE stained with
Coomassie brilliant blue R250. a) Amino acid in one letter code of polypeptides used in this work. b) Supercharged polypeptides separated in a 10% SDS-PAGE. Lane 1, E18(10192Da). Lane 2, E36(18922Da). Lane 3, E72(36512Da). Lane 4 Halfcharge E35(HC_E35, 31217Da). Lane 5, ELP90(39671Da).Lane 6, K18(10176Da). Lane 7, K36(18888Da). Lane 8, K72(36444Da). Lane 9, Halfcharge K30(HC_K30, 26944Da). Lane10, Zwitterionic KE16(17709 Da). M=Prestained protein ladder. The two bands in lane 4, lane9 and lane 10 could be contributed to the Cysteine residues at C-terminus of HC_E35, HC_K30 and Zwitterionic KE16.
68
Mass spectrometric analysis was performed using a 4800 MALDI-TOF/TOF Analyzer in the linear positive mode. The polypeptide samples were mixed 1:1 v/v with α-cyano-4-hydroxycinnamic acid matrix (SIGMA) (100 mg/mL in 70% ACN and 0.1% TFA). Mass spectra were analyzed with the Data Explorer V4.9 (shown in Figure S5).Values determined by mass spectrometry are in good agreement with the masses that are calculated based on the amino acid sequence (shown in Table S1).
69
Figure S5. MALDI-TOF mass spectra of supercharged polypeptides. a) E18, b)
E36, c) E72, d) K18, e) K36, f) K72,g) Zwitter-ionic KE16, h) Half charge K30, i) Half charge E35, j) ELP90.
Table S1.
Mass Determination of Supercharged Polypeptides.
M calculated *(Da) M ms#(Da)
K18 10176.0 10175.12 +/-50 K36 18888.4 18888.45+/-50 K72 36444.5 36442.53+/-50 E18 10192.95 10189.56+/-50 E36 18922.36 18926.64+/-50 E72 36512.3 36511.53+/-50 Zwitterionic KE16 17709.1 17760.9+/-50 Halfcharge K30 26944.0 26955.71+/-50 Halfcharge E35 31217.0 31238.9+/-50
70
ELP90 39671.0 39674.52+/-50
*average molecular weight calculated with ProtParam tool.
#molecular weight determined by MALDI-TOF mass spectrometry.
Modifying the surfaces with SUPs
First, silicon wafers are rinsed sequentially by sonication in ethanol, acetone and ultrapure water (provided by Milli-Q reference system, 18.2 MΩ cm) (20 minutes for each step). For the preparation of surfaces modified by SUPs, the cleaned silicon wafers are immersed into 3,4-dihydroxyphenylalanine (DOPA) solution (2 mg ml-1) for 24 h at the room temperature, which is obtained by dissolving DOPA in a 10 mmol l-1 bicine buffer solution. Then the DOPA-functionalized surfaces are placed in a 50 ml vial with ultrapure water for 10 minutes, and afterwards the sample surface is blown dry by a flow of high purity argon (this treatment is repeated for three times). The DOPA-functionalized surfaces are immersed into a solution of SUPs (1 mg ml-1, prepared by dissolving SUPs in a phosphate buffered saline solution) for 24 h at the room temperature. Finally, SUPs-modified surfaces are placed in a 50 ml vial with ultrapure water for 10 minutes, and afterwards the sample surface is blown dry by a flow of high purity argon (this process is repeated for three times). The whole substrate preparation process was conducted in a Class II Type A2 biosafety cabinet, in which the number of airborne particles is equivalent to that of a “Class 100” clean room.
Surface characterization and the verification of surface modification with DOPA and SUPs
The static contact angle (CA) and contact angle hysteresis (CAH) on the sample were measured at the room temperature with CA System (DSA100, Kruss Co., Germany). The surface chemistry composition was measured by X-ray photoelectron spectroscopy (XPS, ESCALab220i-XL). The topographical images of the surfaces were acquired by using atomic force microscopy (AFM) (Multimode8, Bruker Nano Inc., Germany) in the ScanAsyst mode.
71
XPS was employed to evaluate the change of surface chemistry composition after the modification (Figure S6). The polydopamine functionalization results in an obvious decrease in the intensity of the Si2p peak that originates from the silicon wafer and the concomitant emergence of a new N1s peak, as well as an increase in the C1s signal, which indicate the successful formation of a polydopamine layer. After the immobilization of SUPs, K36 and E36, onto the DOPA-functionalized surface, XPS spectra show further increased intensities of the N1s and C1s peaks originating from enriched amounts of C and N in SUPs, meanwhile the intensities of Si2p signals greatly decrease. Specifically, the ratios of N to Si (N/Si) are 0.000413, 0.0893, 0.208, 0.261 for the bare silicon wafer, the DOPA-functionalized surface, E36 and K36-modified surfaces, respectively.
Table S2. The values of average surface roughness, Ra, of the surfaces modified by
DOPA and a series of supercharged polypeptides.
samples roughness (Ra/nm)
silicon wafer 0.53 ± 0.14 DOPA-functionalized wafer 1.43 ± 0.24 E18-modified surfaces 1.29 ± 0.20 E36-modified surfaces 1.61 ± 0.29 E72-modified surfaces 1.18 ± 0.25 K18-modified surfaces 1.52 ± 0.39
72 K36-modified surfaces 1.00 ± 0.15 K72-modified surfaces 1.68 ± 0.67 HC-K30-modified surfaces 1.26 ± 0.26 KE16-modified surfaces 1.04 ± 0.16 ELP90-modified surfaces 1.79 ± 0.35 HC-E35-modified surfaces 1.57 ± 0.49
73
Figure S6. The XPS spectra of (a) the bare silicon wafer, (b) DOPA-functionalized
silicon wafer, (c, d) K36 and E36-modified surfaces.
Measurement of ice nucleation temperature of condensed water microdroplets
The substrate to be tested is placed in a sample cell consisting of a rubber O-ring sandwiched between two glass coverslips to ensure that ice nucleation experiments are carried out in a closed environment, which eliminates the effects of environmental conditions such as humidity and possible flow of the surrounding gas on ice nucleation. Similarly, the sample cell preparation was conducted in a biosafety cabinet.
Specifically, for the investigation of ice nucleation temperatures, a macroscopic ultrapure water droplet (by Milli-Q reference system, 18.2 MΩ cm) is first placed at the edge of the sample surface, and then the macroscopic droplet goes through an evaporation-condensation process to have large numbers of condensed microdroplets on the sample surface. Afterwards the nucleation temperature of the condensed water microdroplets is recorded when the sample surface temperature is lowered. This process ensures a constant 100% relative humidity and further minimizes possible impurities. The whole laboratory room is air-conditioned and the temperature is kept at 25 °C, which avoids the possible effect of external temperature.
Measurement of ice nucleation temperature of water macrodroplets
The measurement of the temperatures of ice nucleation of single macroscopic droplets sitting on the SUP-modified surfaces was performed in a home-built cell. The procedure for the deposition of macroscopic droplet and the fabrication of sample cell was conducted in a bio-safety cabinet. Specifically, three to four macroscopic droplets of 0.1 μL ultrapure water were separately placed onto the substrate, and each macroscopic droplet froze independently without triggering/affecting the freezing of the proximal one by keeping enough distance between droplets. Then, the surface was cooled at a rate of 2 °C min-1 to induce ice
74
nucleation and the temperatures of ice nucleation were recorded. The statistical distributions of the freezing temperatures on K36 and E36-modified surfaces were investigated. The numbers of freezing events performed are 153 and 195 for K36 and E36-modified surfaces, respectively. Figure S7 shows that the freezing temperature can be well fitted with a Gaussian distribution and the difference in the freezing temperature on the E36 and K36-modified surfaces is obvious. These measurements of macroscopic water droplets consolidate the effects of SUPs on heterogeneous ice nucleation. One interesting feature to be noted is that the average freezing temperature of macroscopic droplets (-23.6 and -19.6 °C on the surfaces modified by E36 and K36, respectively) is a little bit lower than that of the condensed microdroplets (-22.8 and -19.2 °C on the surfaces modified by E36 and K36, respectively). This can be explained by the high heat capacity of liquid water and the presence of a temperature gradient when macroscopic droplets are cooled down by a cooling stage beneath it. Figure S10 shows the freezing temperatures of macroscopic droplets on the surfaces modified by K36, HC-K30, HC-E35 and E36. The difference in the freezing temperatures clearly demonstrates the reproducible effects of SUPs on heterogeneous ice nucleation.
75
Figure S7. Investigation of the freezing of single macroscopic droplet sitting on
K36 and E36-modified surfaces.The histograms fitted with Gaussians clearly show the different effects of both SUPs on ice nucleation.
Figure S8. The freezing process of water droplet (0.1 µL) on an E36-modified
surface to verify surface-induced nucleation. (a) The red circle shows the location of ice nucleation initiated at liquid-solid interface. (b) The blue arrows show the
76
growth direction of ice nucleation toward the liquid-solid interface. (c, d) The freezing front moves upward along water droplet. The insets in the sequential images show the schematic process of ice nucleation. The cooling rate is 2 °C min-1 and the scale bar is 100 µm.
The effect of the cooling rate on ice nucleation
77
Figure S10. Freezing temperatures of single macroscopic droplets on surfaces
modified by K36, HC-K30, HC-E35 and E36.
Measurement of the delay time
Similarly, the delay time measurement was conducted in a home-built cell to ensure that ice nucleation experiments are carried out in a closed environment to ensure a constant relative humidity of 100%. A drop of 0.1 µl ultrapure water was placed onto the substrate to be tested and the procedure for the deposition of water macrodroplet and the fabrication of sample cell was conducted in a biosafety cabinet. The surface was cooled to a temperature of -19.0 °C and then held constantly. The moment when the surface temperature reaches -19.0 °C is referred to as the starting time, and the difference between the starting time and the time taken by the drop for ice nucleation to occur is defined to be the delay time.
Measurement of quartz crystal microbalance (QCM)
First, the chip of QCM-D was immersed into DOPA solution (2 mg ml-1) for 24 h at the room temperature. After being rinsed by ultrapure water and blown dry with a flow of high purity argon, the DOPA-functionalized surfaces were immersed into a
78
solution of E36 (1 mg ml-1) for 24 h at the room temperature. Finally, E36-modified surfaces were rinsed by ultrapure water and blown dry by a flow of high purity argon.
Quartz crystal microbalance with dissipation monitoring (QCM-D) was employed to investigate the variation in the frequency and dissipation. Figure S11 shows that there is no variation in the frequency and dissipation. The results demonstrate that dissociation of DOPA and SUPs does not occur and that the coatings remain stable upon contact with water.
Figure S11. The change in QCM-D resonant frequency and dissipation as a
function of time for the DOPA-functionalized surface (a) and E36-modified surface (b) immersed in ultrapure water.
Table S3. Advantage and Disadvantage of SUPs for tuning ice nucleation in
comparison to chemistry-based compounds.
Advantage Due to precise control of primary structure, SUPs allow precise tuning of nucleation behavior of water. This property and the proteinaceous nature are promising characteristics to employ SUPs as new generation of cryo-protectants in biomedicine owing to their previously proven biocompatibility and the possibility of
79 fabrication inside cells.
Disadvantage Costs for production of SUPs are high compared to commercially available synthetic polyelectrolytes.
5. REFERENCES
[1] a) B. Vonnegut, J. Appl. Phys. 1947, 18, 593; b) M. Gavish, R. Popovitz-Biro, M. Lahav, L. Leiserowitz, Science 1990, 250, 973; c) K. Shi, H. L. Yu, T.-C. Lee, Q. R. Huang, ACS Appl. Mater. Interfaces 2013, 5, 10456; d) B. Wowk, G. M. Fahy, Cryobiology 2002, 44, 14.
[2] a) B. J. Murray, D. O’Sullivan, J. D. Atkinson, M. E. Webb, Chem. Soc. Rev.
2012, 41, 6519; b) G. J. Morris, E. Acton, Cryobiology 2013, 66, 85.
[3] a) E. Anim-Danso, Y. Zhang, A. Alizadeh, A. Dhinojwala, J. Am. Chem. Soc.
2013, 135, 2734; b) M. H. Smith, R. F. Griffiths, J. Latham, Q. J. R. Meterol. Soc. 1971, 97, 495.
[4] a) I. Braslavsky, S. G. Lipson, Appl. Phys. Lett. 1998, 72, 264; b) W. Sun, X. B. Xu, H. Zhang, C. X. Xu, Cryobiology 2008, 56, 93; c) I. M. Svishchev, P. G. Kusalik, J. Am. Chem. Soc. 1996, 118, 649; d) J. Y. Yan, G. N. Patey, J. Phys.
Chem. A 2012, 116, 7057.
[5] D. Ehre, E. Lavert, M. Lahav, I. Lubomirsky, Science 2010, 327, 672. [6] D. Turnbull, B. Vonnegut, Ind. Eng. Chem. 1952, 44, 1292.
[7] a) A. Kolbe, L. L. del Mercato, A. Z. Abbasi, P. R. Gil, S. J. Gorzini, W. H. C. Huibers, B. Poolman, W. J. Parak, A. Herrmann, Macromol. Rapid Commun. 2011,
32, 186; b) D. Pesce, Y. Z. Wu, A. Kolbe, T. Weil, A. Herrmann, Biomaterials
2013, 34, 4360; c) D. H. Veeregowda, A. Kolbe, H. C. van der Mei, H. J. Busscher,
A. Herrmann, P. K. Sharma, Adv. Mater. 2013, 25, 3426.
[8] D. Chow, M. L. Nunalee, D. W. Lim, A. J. Simnick, A. Chilkoti, Mater. Sci.
80
[9] a) L. N. Zhang, C. S. Tian, G. A. Waychunas, Y. R. Shen, J. Am. Chem. Soc.
2008, 130, 7686; b) D. Lis, E. H. G. Backus, J. Hunger, S. H. Parekh, M. Bonn,
Science 2014, 344, 1138.
[10] M. B. van Eldijk, F. C. Smits, N. Vermue, M. F. Debets, S. Schoffelen, J. C. M. van Hest, Biomacromolecules 2014, 15, 2751.
[11] H. Lee, J. Rho, P. B. Messersmith, Adv. Mater. 2009, 21, 431.
[12] K. Y. Li, S. Xu, W. X. Shi, M. He, H. L. Li, S. Z. Li, X. Zhou, J. J. Wang, Y. L. Song, Langmuir 2012, 28, 10749.
[14] M. Matsumoto, S. Saito, I. Ohmine, Nature 2002, 416, 409. [15] N. Du, X. Y. Liu, C. L. Hew, J. Biol. Chem. 2003, 278, 36000.
[16] a) B. Riechers, F. Wittbracht, A. Hutten, T. Koop, Phys. Chem. Chem. Phys.
2013, 15, 5873; b) B. J. Murray, S. L. Broadley, T. W. Wilson, S. J. Bull, R. H.
Wills, H. K. Christenson, E. J. Murray, Phys. Chem. Chem. Phys. 2010, 12, 10380. [17] a) P. Eberle, M. K. Tiwari, T. Maitra, D. Poulikakos, Nanoscale 2014, 6, 4874; b) T. M. Schutzius, S. Jung, T. Maitra, P. Eberle, C. Antonini, C. Stamatopoulos, D. Poulikakos, Langmuir 2015, 31, 4807; c) S. J. Cox, S. M. Kathmann, B. Slater, A. Michaelides, J. Chem. Phys. 2015, 142, 184705.
[18] N. H. Fletcher, The Chemical Physics of Ice, Cambridge University Press, Cambridge, 1970.
[19] A. J. Rusnak, E. R. Pinnick, C. E. Calderon, F. Wang, J. Chem. Phys. 2012, 137, 034510.
