Microscopic diffusion in hydrated encysted eggs of brine shrimp

BBA - General Subjects xxx (xxxx) xxx–xxx

Contents lists available at ScienceDirect

BBA - General Subjects journal homepage: www.elsevier.com/locate/bbagen

Microscopic diffusion in hydrated encysted eggs of brine shrimp☆ E. Mamontov Chemical and Engineering Materials Division, Neutron Sciences Directorate, Oak Ridge National Laboratory, Oak Ridge, TN 37831, United States

A R T I C L E I N F O

A B S T R A C T

Keywords: Brine shrimp eggs hydration water aqueous solvents microscopic dynamics

Background: We have studied microscopic diffusion of water in fully hydrated encysted eggs of brine shrimp (Artemia). Methods: We have utilized quasielastic neutron scattering. Results: Dry eggs of brine shrimp were rehydrated using (1) water without additives, (2) eutectic mixture of water and dimethyl sulfoxide, and (3) a concentrated aqueous solution of lithium chloride. Despite the complexity of the hydrated multicellular organism, measurable microscopic diffusivity of water is rather well defined. Pure hydration water in eggs exhibits freezing temperature depression, whereas hydration water in eggs mixed with dimethyl sulfoxide or lithium chloride does not crystallize at all. Conclusions: The characteristic size of the voids occupied by water or aqueous solvents in hydrated brine shrimp eggs is between 2 and 10 nm. Those voids are accessible to co-solvents such as dimethyl sulfoxide and lithium chloride. There is no evidence of intracellular water in the hydrated eggs. General significance: The lack of intracellular water in the fully hydrated (but still under arrested development) state must be linked to the unique resilience against adverse environmental factors documented not only for the anhydrous, but also hydrated encysted eggs of brine shrimp.

1. Introduction Depending on the environmental conditions, females of brine shrimp (genus Artemia) produce either free swimming nauplii, or embryos at gastrula stage. In the latter case, the embryos enclosed in a protective shell are known as cysts. The encysted embryos enter diapause, which is a state of greatly reduced metabolism. Possible role of sugars, lipids, and stress resistant proteins in diapause of Artemia eggs has been recognized [1]. What follows diapause depends on the environmental conditions at the time of the diapause termination. Under favorable conditions embryo development resumes, whereas under environmental stress cryptobiotic dormancy (quiescence) follows. Diapause can be terminated, e.g., by dehydration, but dried Artemia eggs in the state of quiescence can eventually resume their development under favorable conditions of appropriate hydration, temperature, and salinity. Artemia occupies a truly special place among multicellular organisms that may undergo cryptobiosis at some development stage [2,3]. While anhydrobiosis in dried encysted brine shrimp eggs has been widely recognized and is important in commercial aquaculture, it is

anoxybiosis in hydrated eggs that is rather unique. Hydrated eggs could survive years of complete anoxia, apparently slowing down their metabolism to the values too low to be measured experimentally [4,5]. The kinetics of hydration-dehydration in brine shrimp eggs has been studied experimentally [6], but, to our knowledge, the microscopic dynamics of water in Artemia embryos has not been investigated. Such dynamics, primarily of diffusive character, should be indicative of the state of hydration water, which is likely important in molecular transport and metabolic properties at the cellular and intra-cellular level. Here we report a study of microscopic dynamics of aqueous solvents (water with and without co-solvents) in Artemia eggs using quasielastic neutron scattering (QENS). The choice of QENS as a suitable probe of water dynamics in a multicellular organism is far from intuitive. The incoherent neutron scattering cross-section of hydrogen exceeds the scattering cross-sections of most other elements, including deuterium, at least by a factor of 40, making QENS a technique of choice for studies of confined water, but usually in rather simple matrices. Unlike NMR, which can be sensitive to the atomic positions as manifested through the specific chemical shift, QENS measures hydrogen-dominated scattering signal

☆ This manuscript has been authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. The United States Government retains and the publisher, by accepting the article for publication, acknowledges that the United States Government retains a non-exclusive, paid-up, irrevocable, world-wide license to publish or reproduce the published form of this manuscript, or allow others to do so, for United States Government purposes. The Department of Energy will provide public access to these results of federally sponsored research in accordance with the DOE Public Access Plan (http://energy.gov/downloads/doe-public-access-plan). E-mail address: [email protected].

http://dx.doi.org/10.1016/j.bbagen.2017.05.022 Received 6 January 2017; Received in revised form 7 May 2017; Accepted 23 May 2017 0304-4165/ © 2017 Elsevier B.V. All rights reserved.

Please cite this article as: Mamontov, E., BBA - General Subjects (2017), http://dx.doi.org/10.1016/j.bbagen.2017.05.022

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

sample prepared as follows: 0.44 g of eggs was loaded with 0.50 g of D2O, to match the hydration level of the H2O-loaded sample, which had 0.44 g of eggs and 0.56 g of H2O. Prior to hydration with D2O, the eggs were rinsed with D2O and dried three times to remove labile protons, which otherwise could contaminate the D2O solvent in the finally measured sample.

averaged over the entire sample. This lack of selectivity is balanced in part by the sensitivity of QENS not only to the energy, but also momentum transfer, which advantageously allows analysis of time-spatial characteristics of molecular-level diffusion processes. A variation of QENS technique, temperature-dependent elastic neutron scattering with zero energy transfer, gives information similar to that provided by differential scanning calorimetry, but often with enhanced sensitivity to the confined water dynamics and phase state, due to the high scattering cross-section of hydrogen. QENS studies of water in the living cells [7–12] gradually gain acceptance, but feasibility of QENS for probing entire multicellular organisms remains to be explored. Here we attempt such a study with encouraging results that shed a light on the state of water in hydrated Artemia embryos.

2.2. Neutron scattering experiment Quasielastic and elastic neutron scattering measurements were carried out on a backscattering spectrometer BASIS [24], which has an energy resolution of 3.4 μeV (full-width at half-maximum for the Qaveraged resolution value). That is, microscopic dynamics much slower than ca. 0.4 ns would not be measurable on BASIS (the corresponding scattering signal would appear completely elastic). With the chosen incident neutron bandwidth center of 6.15 Å and neutron bandwidth choppers frequency of 30 Hz, the dynamic range of accessible neutron energy transfer of −100 μeV to +500 μeV could be utilized. The incident neutron beam size is 30 mm by 30 mm. The flat plate sample holders were positioned normal to the incident beam direction. QENS data were collected at 300 K. The elastic intensity temperature scan data were collected from the as received eggs and the three samples hydrated with protonated solvents on cooling down with a ramp rate of 1 K/min and were used to calculate the atomic mean-squared displacement (MSD), < u2(T) > , using a Gaussian approximation, Ielastic(Q,T) = Ielastic(Q,T0)exp(−Q2 < u2(T) > /3), where T0 is the lowest measurement temperature, over the range of 0.2 Å− 1 < Q < 1.0 Å− 1. Additional QENS data were collected at the baseline temperature of 10 K at the completion of the measurement in order to obtain the resolution function under conditions when the measurable dynamics has ceased and the scattering signal has become purely elastic. Besides, the reference liquid solvent samples were measured at 300 K in the same sample geometry for comparison with the hydrated Artemia eggs samples.

2. Materials and methods 2.1. Samples Artemia eggs were purchased from Carolina Biological Supply Co. The as received eggs were in quiescent dry state, yet they exhibited further water loss of ca. 8 wt% in a test when placed in a 10− 3 mbar vacuum at ambient temperature for 24 h. The as received eggs were hydrated using (1) deionized water, (2) an aqueous solution of deuterated dimethyl sulfoxide (DMSO) of eutectic composition [13], (H2O)0.67(C2D6OS)0.33, and (3) an aqueous solution of lithium chloride, (H2O)0.86(LiCl)0.14. For brevity, we may refer to these aqueous solvents as water-DMSO and water-LiCl. As described in the next sections, we have found that quasielastic scattering signal from the hydrated eggs was dominated by scattering from the hydration water, more specifically, the incoherent scattering by protons of the water molecules; hence, we probed single-particle water diffusion dynamics in the samples (besides the scattering signal from the eggs themselves, which was predominantly elastic). Dimethyl sulfoxide is a common cryoprotector utilized to prevent freezing and thawing damage to living systems [14,15], which is known for easy penetration through various tissues. Aqueous solutions of lithium chloride are remarkably close to pure water in their glass-forming properties [16], but allow bypassing [17–19] homogenous nucleation temperature of water to enable lowtemperature studies of bulk-like aqueous solutions of biomolecules without resorting to nanoscopic confinement [20–22]. The slightly offeutectic composition of (H2O)6(LiCl) was chosen for its superior glassforming properties and nanoscopic homogeneity [23]. Although hydration from water vapor could be done in a more controlled manner while monitoring the weight uptake, such a vapor hydration protocol would be impossible to replicate with aqueous solutions as solvents. Therefore, Artemia eggs were instead fully hydrated by submerging into the respective aqueous solvent at ambient temperature for 24 h following by drying in open air for 96 h. Exposure to an aqueous solvent not containing sodium chloride is not expected to terminate quiescence. Following air drying, eggs exposed to pure water looked morphologically similar to the as received dry eggs, having appearance of dry powder. They showed water loss of ca. 56 wt% in a test when placed in a 10− 3 mbar vacuum at ambient temperature for 24 h, in agreement with earlier studies [6]. Eggs exposed to water-DMSO and water-LiCl had appearance of a powder (not a paste or a slurry), but retained some minimal residual dampness even at the completion of the air drying, which precluded meaningful water loss measurements. Four powderlike samples (as received and hydrated with H2O, (H2O)0.67(C2D6OS)0.33, and (H2O)0.86(LiCl)0.14 were loaded in flat-plate aluminum sample holders (30 mm wide, 50 mm tall, and 1 mm thick) vacuum-sealed with indium gaskets. The mass of each of the four samples was 1 g. Three reference samples of liquid aqueous solvents, H2O, (H2O)0.67(C2D6OS)0.33, and (H2O)0.86(LiCl)0.14) were loaded in similar indium-sealed flat-plate aluminum sample holders, but only 0.1 mm thick, to minimize multiple neutron scattering effects. Besides, we used a 1 mm thick sample holder to measure a control D2O-loaded

3. Results The temperature dependence of the mean-squared atomic displacement measured in Artemia eggs is presented in Fig. 1. The data is characteristic of confined aqueous solutions [25]. At sufficiently low temperatures, where there is no measurable diffusion mobility, the temperature dependence of the elastic scattering intensity is controlled

Fig. 1. Mean-squared atomic displacement of hydrogen atoms in brine shrimp eggs, dry (as received) and hydrated with protonated aqueous solvents, as measured by elastic neutron scattering.

2

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

hydrated eggs demonstrate increased signal at lower energies, with the broad peaks which are, although not as sharp as their counterparts in the corresponding bulk solvents, sufficiently well-defined and clearly Qdependent. This suggests that, despite considerable complexity of a multicellular organism such as Artemia eggs, the translational diffusivity of its hydration water (or aqueous solvents) is reasonably well defined and can be quantified. For quantitative analysis of water diffusivities, we turn to more traditional fits of the I(Q,E) scattering intensities with a resolution function, R(Q,E), convolved with a superposition of a delta function, δ(E) with a relative spectral weight of x(Q) centered at zero energy transfer, and a model scattering function, S(Q,E), plus a linear background:

by vibrational thermal factors, leading to purely harmonic < u2(T) > ~ T. In glass-forming systems that do not exhibit freezing/ melting, gradual diffusion mobility onset on warming up (or, conversely, diffusion slowing down on cooling down) results in the excess mean-squared displacement above the extrapolation of the low-temperature < u2(T) > ~ T baseline. Such gradually varying with temperature excess MSD, commonly studied in hydrated biomolecules and their hydration water [26–28], is a reflection of general solvent-solute relationship, not specific to either biomolecules or water [29]. In the case of freezing/melting transition, a more abrupt change in the MSD is observed [25]. In the limiting case of bulk unconfined water or bulklike weakly confined water, a sharp melting/freezing step would be evident at or near 273 K. The as received eggs data in Fig. 1 exhibit the temperature dependence characteristic of biomolecular samples that are almost, but not completely, dry, as evidenced by a small intensity upturn above ca. 230 K [28], which is in agreement with 8 wt% weight loss observed when as received eggs were treated under vacuum. The eggs hydrated with H2O show progressive water freezing (intensity drop) on cooling down that is completed at ca. 230 K. On the other hand, the eggs hydrated with water-DMSO and water-LiCl do not show water freezing on cooling down; instead, the water dynamics gradually moves out of the resolution window of the spectrometer. This qualitative difference between the H2O and two aqueous solvents hydrating brine shrimp eggs is evident from the fact that the diffusion mobility giving rise to excess MSD with respect to the low-temperature baseline < u2(T) > ~ T ceases abruptly below 230 K for H2O, but disappears gradually for the two aqueous solvents. In the case of pure water in confinement, a greater suppression of the freezing point is associated with smaller pore sizes. For instance, the following freezing temperatures have been reported in various silica matrices: 252 K and 237 K in 10 nm and 3 nm pores, respectively [30], 260 K in 9 nm pores [31], 232 K in 4.2 nm pores [32], and 255 K in 5 nm pores [33]. In general, greater suppression of the freezing point of confined water is associated with smaller pore sizes. The data in Fig. 1 that show freezing completion at 230 K suggest that characteristic size of voids filled by H2O in the hydrated eggs is above ca. 2 nm (in voids below 2 nm in size, pure water would not exhibit freezing). At the same time, there is no evidence of intra-cellular cytoplasmic water, which is only weakly confined and would freeze not far below 273 K. Adding a co-solvent (DMSO or LiCl) suppresses water freezing in the hydrated eggs, suggesting that co-solvents penetrate into the same voids where water penetrates. The right panels in Fig. 2 show the scattering intensities collected from Artemia eggs converted to dynamic susceptibilities, χ″(E) = I (Q,E)/nb(E), where nb(E) is the thermal Bose factor. The left panels show the corresponding data for bulk aqueous solvents, which look quite similar to the previously reported dynamic susceptibility data for relatively loosely confined hydration water [34], where the lower-energy maximum due to long-range translational diffusion component is separated by a minimum from the higher-energy intensity upturn due to the localized mobility. The former component is Q-dependent, reflecting the fact that, for a reasonably well-defined diffusion process, it takes a certain amount of time (inversely related to E) for a diffusing particle to traverse a certain distance (inversely related to Q). As the Q is increased (corresponding to the shorter probe distance), the separation between the translational and localized dynamics becomes blurred. Qualitatively, the positions of the translational diffusivity maxima shift to the lower energy with addition of DMSO and LiCl, illustrating slowing down of water diffusion in bulk aqueous solutions compared to pure H2O. In the dry eggs (upper right panel), there is no evidence of water translational diffusivity, in agreement with the almost harmonic behavior of the corresponding MSD in Fig. 1 demonstrating very low hydration level. The higher-energy intensity upturn persists, reflecting the localized microscopic dynamics of the eggs constituents. The

I (Q, E ) = [x (Q) δ (E ) + (1 − x (Q)) S (Q, E )] ⊗ R (Q, E ) + (C1 (Q) E + C2 (Q))

(1)

The resolution function is represented by the data collected at 10 K, whereas the information on the water diffusivity is in the model scattering function, which in many cases can be approximated by a Fourier transform of a stretched exponential: ∞

S (Q, E ) =

∫ exp ⎡⎢−⎛ τ (tQ) ⎞ 0

⎜

⎟

⎣ ⎝

⎠

β (Q)

E ⎤ ⎛ ⎞ ⎥ exp ⎝i ℏ t ⎠ dt ⎦

(2)

where 0 < β(Q) < 1. A representative example of the data fits with Eqs. (1)–(2) for hydrated Artemia eggs at 300 K performed over the truncated energy transfer range, − 100 μeV < E < + 100 μeV, to minimize contribution from the localized dynamics, is shown in Fig. 3 for Q = 0.3 Å− 1. As the lowest measured Q point, it is the most closely related to the translational diffusivity. From the data in Fig. 3 one can anticipate the diffusivity values for water to be fairly close among all the aqueous solvents studied. From the fit parameters τ(Q) and β(Q), the average relaxation time can be calculated as < τ(Q) > = (τ(Q)/β(Q))Γ(1/β(Q)), where Γ is the gamma-function, and the average inverse relaxation time, in the energy units, as HWHM(Q) = (ℏ/ < τ(Q) > ). The Q-dependence of the resulting HWHM(Q) then can be fitted with an appropriate diffusion model, e.g., continuous diffusion with infinitely small jump length, HWHM(Q) = ℏDQ2, or jump diffusion with a finite jump length, L, HWHM(Q) = ℏDQ2/(1 + < L2 > Q2/6), to extract the translational diffusion coefficient, D. Such fits of the average inverse relaxation time are presented in Fig. 4, while the fit parameters involved in the calculation of the relaxation times are listed in Table 1. At the Q values higher than those presented in Table 1 and Fig. 4, the QENS broadening due to the translational diffusion is no longer well defined, as evidenced by the merging peaks in the susceptibility data in Fig. 2. This is more of a problem for the relatively less well-defined peaks from the hydrated eggs samples, resulting in the sizable error bars visible for those data in Fig. 4 already for Q = 0.9 Å− 1, unlike for the bulk solvents data. The Qdependence of the data in Fig. 4 followed a jump diffusion model, except for the confined water-LiCl, which had to be fitted with a continuous diffusion model. The applicability of a particular model has limited influence on the obtained D value, which is determined primarily by the lowest Q data as the slope of the HWHM(Q) vs. Q2 plot in the limit of zero Q. The D values reported in the literature, 6.5 ∗ 10− 10 m2/s for waterDMSO sample of the same composition as ours measured at 300 K [35] and 7.5 ∗ 10− 10 m2/s for (H2O)0.88(LiCl)0.12 measured at 290 K [18], demonstrate consistency of the water diffusion coefficients obtained for bulk solvents as shown in Fig. 4 with the previously measured data. While the measured diffusion coefficients values for the solvents hydrating brine shrimp eggs are characteristic of confined solvents, being up to 4 times lower compared to the bulk counterparts and consistent with the elastic intensity scans, their accuracy needs to be discussed. As we demonstrate below, quasielastic scattering signal from 3

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

Fig. 2. Imaginary part of dynamic susceptibility, as measured by quasielastic neutron scattering at 300 K, in brine shrimp eggs, dry (as received) and hydrated with protonated aqueous solvents (right panels). Left panels show the data for the corresponding aqueous solvents in bulk form at 300 K.

sample hydrated to the same hydration level with D2O instead of H2O, the transmission, which is now dominated by scattering from egg constituents, is ca. 88%. Fig. 5 shows comparison between the H2O- and D2O-hydrated samples (the fit parameters for the latter are also included in Table 1). The still large elastic scattering signal in the D2O-hydrated sample (although reduced due to the H-D exchange undertaken prior to hydration with D2O) is obvious (top panel), but the quasielastic scattering now becomes low in intensity and requires the use of logarithmic scale (middle panel, Q = 0.3, 0.5, and 0.7 Å− 1 from left to right) to be visualized. When presented as dynamic susceptibility in the middle panels, the peak intensity in the D2O-hydrated sample is now small compared to the peak intensity in the H2O-hydrated sample, but the peak position at any given Q remains almost unchanged (red vs. black curves). This indicates that the QENS signal in the D2O-hydrated sample predominantly originates from D2O. Thus, quasielastic scattering signal

the hydrated eggs is dominated by scattering from the hydration water, whereas the scattering signal from the eggs themselves is predominantly elastic. Therefore, the scattering contributions can be evaluated using the fitted values of the elastic scattering fraction, x, in Eq. (1), which are 0.39, 0.36, and 0.50 for water, water-LiCl, and waterDMSO, respectively. Thus, 0.44 g of eggs scatter at about 39% level with respect to the 100% scattering from 0.44 g of eggs and 0.56 g of H2O, suggesting that dry eggs constituents scatter at 80% level compared to H2O, on gram per gram basis. In this case, in the as received sample water molecules contribute ca. 10% of the scattering signal, whereas in the H2O-hydrated sample their contribution is ca. 61%. With the sample holders used, in the H2O-hydrated sample the hydration water considered alone would be associated with ca. 82% neutron transmission through the sample, but additional scattering from the eggs constituents decreases the transmission to ca. 73%, making multiple scattering effects not negligible. On the other hand, for the control 4

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

water diffusivity and freezing temperature would be close to those of bulk water due to the limited extent of confinement in the cells, the confinement experienced by water hydrating brine shrimp eggs is much tighter. Comparison between the QENS broadening in Fig. 5 (bottom) for the H2O- and D2O-hydrated samples indicates the magnitude of the multiple scattering effects in the samples hydrated with protonated solvents. The diffusion coefficient calculated from jump-diffusion model is almost the same: (7.2 ± 0.4) ∗ 10− 10 m2/s for the H2O-hydrated eggs vs. (7.7 ± 0.9) ∗ 10− 10 m2/s for the D2O-hydrated eggs, and (7.3 ± 0.5) ∗ 10− 10 m2/s for the difference data obtained by subtraction of the D2O-hydrated data from the H2O-hydrated data. However, it is more instructive to compare directly the QENS broadening, e.g., at the lowest measured Q of 0.3 Å− 1, which is most closely related to model-free diffusivity values. The H2O-hydrated sample exhibits HWHM (Q = 0.3 Å− 1) = 3.4 μeV, whereas for the D2O-hydrated sample HWHM (Q = 0.3 Å− 1) = 2.5 μeV. Thus, the effects of multiple scattering in the samples hydrated with protonated solvents lead to increase of the apparent width of the QENS signal at a level of ca. 33% compared to what would be measured in the canonical samples with neutron transmission close to 90%. The above discussion illustrates the power of comparative dynamic measurements using neutron scattering from hydrogenated and deuterated samples. At the same time, obtaining quantitative information from measurements of the elastic intensities alone may be challenging for multi-component systems. The data presented in Fig. 1 illustrate freezing of lack thereof in the hydrated eggs, but they were calculated with respect to the lowest measurement temperatures of those data sets, as explained in the Materials and Methods section, and, thus, provide only qualitative temperature dependence of MSD. Independently, for more quantitative analysis, we can use the QENS spectra collected at 300 K and the respective resolution data collected at 10 K to obtain the elastic intensities by integrating the spectra over ± 3.4 μeV energy transfer range, similar to the approach used to obtain the data presented in Fig. 1. Using a Gaussian approximation for Q < 1 Å− 1, the apparent MSD is customarily extracted from the slope of the measured 3ln[Ielastic(Q,T)/Ielastic(Q,T0)] vs. Q2. However, for hydrated eggs,

Fig. 3. Example of the data and fits for brine shrimp eggs at Q = 0.3 Å− 1 (300 K).

total total Ielastic (Q, T ) = Ielastic (Q, T0) exp( −Q 2 < u2 (T ) > 3)= eggs 2 Ielastic (Q, T0) exp(−Q 2 < ueggs (T ) > 3) water 2 + Ielastic (QT0) exp(−Q 2 < u water (T ) > 3)

(3)

Therefore, instead of a simple relationship used to extract the apparent MSD from 3ln[Ielastic(Q,T)/Ielastic(Q,T0)] = − < u2(T) > Q2, the < u2 > values for the eggs constituents and hydration water both contribute to the experimentally measured values of 3ln[Ielastic(Q,T)/ Ielastic(Q,T0)] as follows: eggs I total (Q, T ) ⎞ ⎛ Ielastic (Q, T0 ) exp( −Q 2 < u 2 (T ) > 3) 3 ln ⎛⎜ elastic ⎟ = 3 ln ⎜ eggs total total ⎝ Ielastic (Q, T0 ) ⎝ Ielastic (Q, T0 ) ⎠ I water (Q, T0 ) 2 + elastic exp(−Q 2 < u water (T ) > 3) ⎞⎟ total Ielastic (Q, T0 ) ⎠

Fig. 4. Width of the quasielastic broadening (symbols) fitted with translational diffusion models (solid lines), jump diffusion or continuous diffusion, as described in the text. Right panels: brine shrimp eggs, dry (as received) and hydrated with aqueous solvents (300 K). Left panels: the corresponding bulk solvents in bulk form (300 K). The extracted diffusion coefficients are also shown (bulk solvents: top legends, hydrated eggs: bottom legends).

(4)

In general, when the contribution of either term on the right-hand side is not negligible, the interpretation of the Q-dependence of the experimentally measured elastic intensities (on the left-hand side) is not straightforward. Fig. 6 shows, along with the data for the as received eggs, the apparent average MSD at 300 K derived (as slope of the fitted lines) from fitting the Q-dependence of the elastic intensities, which, for the eggs hydrated with H2O, actually include contribution from two terms, as follows:

from the hydrated eggs is dominated by scattering from the hydration water (H2O or D2O), whereas the scattering signal from the eggs themselves is predominantly elastic in the energy range of our analysis. For cells with cytoplasmic water, the quasielastic signal would be a mixture of signals from the cytoplasmic water and cell constituents made mobile by the presence of water. Quasielastic scattering dominated by water, even in the D2O-hydrated sample, thus indicates lack of intra-cellular water in the hydrated eggs, consistent with no freezing near 273 K in the elastic intensity scans and the diffusion coefficient much reduced compared to the bulk water value. While cytoplasmic

eggs − H2 O eggs 2 Ielastic (Q, 300K) = Ielastic (Q, 10K) exp( −Q 2 < ueggs (300K) > 3) H2 O 2 + Ielastic (Q, 10K) exp( −Q 2 < u water (300K) > 3)

5

(5)

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

Table 1 Parameters obtained from the data fits using Eqs. (1)–(2) and the average relaxation times computed as < τ > = (τ/β)Γ(1/β), where Γ is the gamma-function. The standard deviation values are shown in parentheses. Q, Å− 1

0.3 0.5 0.7 0.9 0.3 0.5 0.7 0.9 0.3 0.5 0.7 0.9 0.3 0.5 0.7 0.9

τ(Q), ps

β(Q)

H2O bulk 38.7 (0.3) 0.87 (0.01) 18.9 (0.4) 0.95 (0.02) 11.2 (0.1) 1.0 (0.01) 8.6 (0.2) 1.0 (0.01) (H2O)0.67(C2D6OS)0.33 bulk 148.5 (0.9) 0.87 (0.01) 65.5 (0.3) 0.87 (0.01) 36.0 (0.3) 0.90 (0.01) 25.0 (0.4) 0.91 (0.01) (H2O)0.86(LiCl)0.14 bulk 120.0 (0.6) 0.82 (0.01) 46.9 (0.3) 0.80 (0.01) 23.6 (0.4) 0.82 (0.01) 17.5 (0.6) 0.91 (0.02) D2O in eggs 159.8 (3.4) 0.57 (0.01) 71.1 (0.9) 0.44 (0.01) 23.5 (1.1) 0.35 (0.01) 6.5 (0.8) 0.27 (0.1)

< τ(Q) > , ps

τ(Q), ps

β(Q)

H2O in eggs 124.5 (1.6) 0.59 (0.01) 47.0 (0.7) 0.49 (0.01) 9.4 (0.8) 0.33 (0.01) 1.0 (0.3) 0.23 (0.01) (H2O)0.67(C2D6OS)0.33 in eggs 133.1 (2.4) 0.61 (0.01) 58.8 (0.6) 0.64 (0.01) 25.5 (0.8) 0.53 (0.01) 5.6 (0.7) 0.36 (0.01) (H2O)0.86(LiCl)0.14 in eggs 239.1 (8.8) 0.57 (0.01) 79.8 (0.7) 0.64 (0.01) 37.8 (0.6) 0.65 (0.01) 19.9 (0.7) 0.59 (0.01) Difference data (H2O minus D2O) in eggs 113.3 (2.0) 0.59 (0.01) 40.8 (1.0) 0.51 (0.01) 6.9 (1.0) 0.34 (0.01) 0.3 (0.2) 0.22 (0.02)

41.4 (0.7) 19.2 (0.8) 11.2 (0.1) 8.6 (0.2) 159.2 (1.9) 69.8 (0.6) 38.1 (0.6) 26.1 (0.7) 134.2 (1.5) 53.4 (0.7) 26.3 (0.7) 18.4 (1.0) 259.8 (9.4) 185.1 (4.9) 119.1 (8.0) 99.9 (15.9)

< τ(Q) > , ps

193.0 (4.8) 98.1 (2.7) 58.4 (6.1) 39.4 (13.3) 196.2 (6.7) 81.4 (1.8) 46.4 (2.2) 25.5 (4.1) 386.1 (21.2) 110.6 (2.1) 51.4 (1.4) 30.7 (1.6) 173.4 (6.2) 78.3 (3.5) 38.6 (7.1) 15.6 (12.7)

Fig. 6. Symbols: ratio of the experimentally measured at 300 K and 10 K elastic scattering intensities vs. Q2. The apparent average mean-squared displacements at 300 K are customarily determined from the slope of the measured values of 3ln[Ielastic(Q,300 K)/ Ielastic(Q,10 K)] vs. Q2, as shown by the solid lines for the hydrated samples. Short-dashed lines: the slope at the low Q, attributed to the hydration water. Long-dashed lines: the slope at the high Q (and at all Q values for the as received eggs), attributed to the eggs constituents. eggs − d − D2 O eggs − d 2 Ielastic (Q, 300K) = Ielastic (Q, 10K) exp(−Q 2 < ueggs (300K) > 3) D2 O 2 + Ielastic (Q, 10K) exp(−Q 2 < u water (300K) > 3)

(6)

Fig. 5. Comparison between the H2O- and D2O-hydrated brine shrimp eggs at 300 K. Top panel: data at Q = 0.7 Å− 1 presented over the truncated, −20 μeV < E < + 20 μeV, energy transfer range. Middle panels: imaginary part of dynamics susceptibility, left to right: Q = 0.3, 0.5, and 0.7 Å− 1. Bottom panel: fits of the Q-dependence of the quasielastic broadening with jump diffusion model. The same color scheme as shown in the legend is used for the H2O, D2O, and the difference data in all panels. The extracted diffusion coefficients are also shown.

The difference between the elastic intensities from the H2O- and D2O-hydrated eggs is also plotted, along with the corresponding fitted apparent MSD (slope of the solid blue line). Due to the much larger scattering cross-section of hydrogen compared to deuterium, one would expect: eggs − H2 O eggs − d − D2 O Ielastic (Q, 300K) − Ielastic (Q, 300K)

and, for the eggs hydrated with D2O (following exchange of the labile protons in the eggs), also include contribution from two terms, as follows:

H2 O 2 ≈ Ielastic (Q, 10K) exp(−Q 2 < u water (300K) > 3)

(7)

if not for the difference in scattering between the eggs constituents in the native state and the eggs constituents that underwent H-D exchange of the labile protons. 6

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

hydrating the eggs decreases only by a factor of 2 or less compared to the corresponding bulk values, suggesting that the composition of the confined solvents could differ from the nominal (H2O)0.67(C2D6OS)0.33 and (H2O)0.86(LiCl)0.14. Yet the presence of some DMSO and LiCl in the confined solvents is evident from the elimination of H2O freezing step in Fig. 1. This freezing step would still be present should all DMSO and LiCl have remained on the eggs surface, which is obviously not the case. The morphological complexity of the encysted quiescent Artemia eggs as determined from electron microscopy studies [38] is significant. Beginning from the outermost layer, the layers include outer membrane, cortical layer, alveolar layer, outer cuticular membrane, fibrous layer, and inner cuticular membrane [38]; the latter contains the yolk content with the number of nuclei in the 3000–4000 range [6]. The three outer layers, outer membrane, cortical layer, and alveolar layer are maternally produced and together are termed tertiary envelope [39]. The three inner layers, outer cuticular membrane, fibrous layer, and inner cuticular membrane are embryonically produced. The outer layers are permeable, whereas the outer cuticular membrane is impermeable for non-volatile compounds; this impermeability is essential for normal embryonic development of Artemia [40]. In hydration/dehydration experiments on the intact and decapsulated (with the outer layers, down to the outer cuticular membrane, removed) Artemia eggs, it was determined that the outer layers hosted only a minor fraction of hydration water; a typical water content varied between ca. 60 wt% for the intact eggs and ca. 50 wt% for the decapsulated eggs. This suggests that the main dynamical features observed in the current experiment (completion of freezing at ca. 230 K for pure H2O suppressed by addition of DMSO or LiCl and decrease in the translational diffusion coefficient) predominantly reflect the state of the confined water (or waterDMSO and water-HCl) which permeates both the outer and inner membranes. In view of the considerable morphological complexity of the hydrated host, it is remarkable that the translational diffusion coefficient of the hydration water is sufficiently well defined, despite the fact that freezing of the pure confined H2O does not occur in a single step, suggesting a distribution of the effective confinement pore size over the range of 2 to 10 nm. A remarkable observation made in the present study is the absence in the fully hydrated quiescent Artemia eggs of hydration water with typical characteristics of the intra-cellular cytoplasmic water, such as freezing near 273 K and the translational diffusion coefficient similar to that in bulk water. The fact that hydrated eggs host confined extracellular water rather than intra-cellular water also justifies our choice of the model scattering function in the form of a Fourier transformed stretched exponential. This choice is different from the model scattering functions previously employed in QENS studies of hydrated cells, which may include, besides the elastic line and background, several Lorentzian components: for water molecules in the hydration layer in direct contact with the cell constituents, for the cell constituents whose motions are enabled by water, and for bulk-like intra-cellular water [41]. This is due to, in part, the more limited energy range of our data analysis, − 100 μeV < E < + 100 μeV, but mostly the different character of hydration water observed in the current study. Macromolecular constituents in hydrated cells may be characterized by two diffusion coefficients [42]: for the global diffusivity (on the low 10− 10 m2/s scale) and the localized side chain dynamics (on the low 10− 9 m2/s scale). The latter diffusivity could be measured at about two-thirds of the solvent diffusivity [43]. The absence of intra-cellular water in our samples largely eliminates these dynamic components from consideration and reduces the importance of the dynamic component associated with the hydration water molecules in the layer of direct contact with the eggs constituents. The latter type of water, with diffusivity often measured in biomacromolecules in the form of hydrated powders on a 10− 11 m2/s scale [44], typically contributes only 10–20% of the scattering signal even in QENS measurements of intracellular water, where due to the abundance of intra-cellular macromolecules the interface between the water and the matrix is more

The apparent average MSD values at 300 K (the slope of the solid lines for the hydrated samples and the long-dashed line for the as received sample) are as follows: (0.65 ± 0.04) Å2 for the as received eggs, (1.94 ± 0.04) Å2 for the D2O-hydrated eggs, (2.39 ± 0.05) Å2 for the H2O-hydrated eggs, and (3.23 ± 0.14) Å2 for the difference data between the H2O- and D2O-hydrated eggs. Qualitatively, the average MSD increases along with the increasing relative contribution to the scattering signal from the mobile hydration water (from the as received to D2O-hydrated to H2O-hydrated to the difference data; the latter is the most representative of the water contribution alone). The quantitative trend is more difficult to evaluate. However, we note that only for the as received eggs, where the signal is dominated by the eggs constituents, a single slope fits all the data points well. The other data sets (for the hydrated eggs) appear to exhibit two slopes, for the lowand high-Q data; the latter may resemble the slope of the as received eggs data. Thus, it is possible that the low-Q slope is associated with MSD of the hydration water, whereas the high-Q slope is associated with MSD of the eggs constituents. The latter contribution does not disappear in the H2O-D2O difference data set because of the difference in scattering between the eggs constituents in the native state and the eggs constituents that underwent H-D exchange of the labile protons. Separate fits of the low-Q and high-Q data, as shown by the dashed lines for the hydrated samples in Fig. 6, provide the following MSD values. For the hydration water: (4.15 ± 0.21) Å2 for the H2O-hydrated eggs, (3.76 ± 0.18) Å2 for the D2O-hydrated eggs, and (4.97 ± 0.58) Å2 for the difference data set between the H2O- and D2O-hydrated eggs. For the eggs constituents: (1.61 ± 0.11) Å2 for the H2O-hydrated eggs and (1.31 ± 0.08) Å2 for the D2O-hydrated eggs. These values for the eggs constituents in the hydrated samples are closer to the value of (0.65 ± 0.04) Å2 for the as received eggs than the values for the hydration water. Therefore, quasielastic scattering signal from the hydrated eggs must be dominated by scattering from the hydration water, consistent with the data presented in Fig. 5. 4. Discussion In principle, there could be several possible states of water or water solutions in contact with the eggs, as in the case of any hydrated matrix. Firstly, there could be surface-bound water, which is usually exohedral, as on the surface of various hydrated powders with high surface area and low porosity, but also could be endohedral, as on the surface of inner pores [34]. This type of water is commonly probed in hydrated powders of proteins and other biomolecules [26–28] and is characterized by lack of freezing, behaving as a glass-former instead. Secondly, water confined in pores (filling them) would generally exhibit a freezing step, but it could either be eliminated or become undetectable in elastic neutron scattering intensity scans if the pore size is below ca. 2 nm [36]. This type of water exhibits freezing point depression, depending on the pore size, until it exceeds ca. 10 nm [37]. Finally, in less restrictive confinement environments of larger pores, water properties become bulk-like, exhibiting no freezing point depression and a diffusion coefficient similar to that of bulk water; this applies to intra-cellular cytoplasmic water [7–12]. Some surface-bound water is undoubtedly present in our hydrated eggs samples, but its contribution to the scattering signal is likely rather insignificant because of the negligible surface area of the samples associated with the large individual egg size, > 0.1 mm. Likewise, the contribution from the possible residual bulk-like water in the samples outside the eggs is not substantial, as demonstrated by (1) the lack of pronounced freezing step near 273 K for pure hydration water in Fig. 1 and (2) the diffusion coefficients reduced by a factor of 4 or more for water in the hydrated eggs compared to the bulk value (Figs. 4 and 5). Furthermore, our choice of small-molecule co-solvents capable of penetrating tissues along with water helps illustrate that aqueous solvents hydrating Artemia eggs bear characteristic traits of confined aqueous systems. The diffusion coefficient of water in aqueous solvents 7

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

141–149. [7] M. Jasnin, M. Moulin, M. Haertlein, G. Zaccai, M. Tehei, Down to atomic-scale intracellular water dynamics, EMBO Rep. 9 (2008) 543–547. [8] A.M. Stadler, J.P. Embs, I. Digel, G.M. Artmann, T. Unruh, G. Buldt, G. Zaccai, Cytoplasmic water and hydration layer dynamics in human red blood cells, J. Am. Chem. Soc. 130 (2008) 16853. [9] A. Frolich, F. Gabel, M. Jasnin, U. Lehnert, D. Oesterhelt, A.M. Stadler, M. Tehei, M. Weik, K. Wood, G. Zaccai, From shell to cell: neutron scattering studies of biological water dynamics and coupling to activity, Faraday Discuss. 141 (2009) 117–130. [10] M. Jasnin, A. Stadler, M. Tehei, G. Zaccai, Specific cellular water dynamics observed in vivo by neutron scattering and NMR, Phys. Chem. Chem. Phys. 12 (2010) 10154–10160. [11] F. Natali, C. Dolce, J. Peters, Y. Gerelli, C. Stelletta, G. Leduc, Water dynamics in neural tissue, J. Phys. Soc. Jpn. 82 (2013) SA017. [12] M. Liberton, L.E. Page, W.B. O'Dell, H. O'Neill, E. Mamontov, V.S. Urban, H.B. Pakrasi, Organization and flexibility of cyanobacterial thylakoid membranes examined by neutron scattering, J. Biol. Chem. 288 (2013) 3632–3640. [13] D.H. Rasmussen, A.P. MacKenzie, Phase diagram for the system water-dimethylsulfoxide, Nature 220 (1968) 1315–1317. [14] J.E. Lovelock, M.W.H. Bishop, Prevention of freezing damage to living cells by dimethyl sulfoxide, Nature 183 (1959) 1394–1395. [15] J. Farrant, Mechanism of cell damage during freezing and thawing and its prevention, Nature 205 (1965) (1284- &). [16] C.A. Angell, E.J. Sare, Glass-forming composition regions and glass transition temperatures for aqueous electrolyte solutions, J. Chem. Phys. 52 (1970) (1058- &). [17] E. Mamontov, Diffusion dynamics of water molecules in a LiCl solution: a lowtemperature crossover, J. Phys. Chem. B 113 (2009) 14073–14078. [18] E. Mamontov, A. Faraone, E.W. Hagaman, K.S. Han, E. Fratini, A low-temperature crossover in water dynamics in an aqueous LiCl solution: diffusion probed by neutron spin-echo and nuclear magnetic resonance, J. Phys. Chem. B 114 (2010) 16737–16743. [19] M. Nakanishi, P. Griffin, E. Mamontov, A.S. Sokolov, No fragile-to-strong crossover in LiCl-H2O solution, J. Chem. Phys. 136 (2012) 124512 (9 Pages). [20] X. Chu, E. Mamontov, H. O'Neill, Q. Zhang, Apparent decoupling of the dynamics of a protein from the dynamics of its aqueous solvent, J. Phys. Chem. Lett. 3 (2012) 380–385. [21] E. Mamontov, X. Chu, Water-protein dynamic coupling and new opportunities for probing it at low to physiological temperatures in aqueous solutions, Phys. Chem. Chem. Phys. 14 (2012) 11573–11588. [22] E. Mamontov, H. O'Neill, Q. Zhang, S.M. Chathoth, Temperature dependence of the internal dynamics of a protein in an aqueous solvent: decoupling from the solvent viscosity, Chem. Phys. 424 (2013) 12–19. [23] L.E. Bove, C. Dreyfus, R. Torre, R.M. Pick, Observation of nanophase segregation in LiCl aqueous solutions from transient grating experiments, J. Chem. Phys. 139 (2013) 044501(11 Pages). [24] E. Mamontov, K.W. Herwig, A time-of-flight backscattering spectrometer at the spallation neutron source, BASIS, Rev. Sci. Instrum. 82 (2011) 085109(10 pages). [25] E. Mamontov, D.R. Cole, S. Dai, M.D. Pawel, C.D. Liang, T. Jenkins, G. Gasparovic, E. Kintzel, Dynamics of water in LiCl and CaCl2 aqueous solutions confined in silica matrices: a backscattering neutron spectroscopy study, Chem. Phys. 352 (2008) 117–124. [26] W. Doster, S. Cusack, W. Petry, Dynamical transition of myoglobin revealed by inelastic neutron scattering, Nature (London) 337 (1989) 754–756. [27] W. Doster, The two-step scenario of the protein dynamical transition, J. Non-Cryst. Solids 357 (2011) 622–628. [28] J.H. Roh, R.M. Briber, A. Damjanovic, D. Thirumalai, S.A. Woodson, A.P. Sokolov, Dynamics of tRNA at different levels of hydration, Biophys. J. 96 (2009) 2755–2762. [29] E. Mamontov, V.K. Sharma, J.M. Borreguero, M. Tyagi, Protein-style dynamical transition in a non-biological polymer and a non-aqueous solvent, J. Phys. Chem. B 120 (2016) 3232–3239. [30] T. Takamuku, M. Yamagami, H. Wakita, Y. Masuda, T. Yamaguchi, Thermal property, structure, and dynamics of supercooled water in porous silica by calorimetry, neutron scattering, and NMR relaxation, J. Phys. Chem. B 101 (1997) 5730–5739. [31] D.C. Steytler, J.C. Dore, C.J. Wright, Neutron diffraction study of cubic ice nucleation in a porous silica network, J. Phys. Chem. 87 (1983) 2458–2459. [32] K. Morishige, K. Nobuoka, X-ray diffraction studies of freezing and melting of water confined in a mesoporous adsorbent (MCM-41), J. Chem. Phys. 107 (1997) 6965–6969. [33] M.-C. Bellissent-Funel, J. Lal, L. Bosio, Structural study of water confined in porousglass by neutron-scattering, J. Chem. Phys. 98 (1993) 4246–4252. [34] E. Mamontov, Y. Yue, J. Bahadur, J. Guo, C.I. Contescu, N.C. Gallego, Y.B. Melnichenko, Hydration level dependence of the microscopic dynamics of water adsorbed in ultramicroporous carbon, Carbon 111 (2017) 705–712. [35] J.T. Cabral, A. Luzar, J. Teixeira, M.-C. Bellissent-Funel, Single-particle dynamics in dimethyl-sulfoxide/water eutectic mixture by neutron scattering, J. Chem. Phys. 113 (2000) 8745. [36] L. Liu, A. Faraone, C.-Y. Mou, C.-W. Yen, S.-H. Chen, Slow dynamics of supercooled water confined in nanoporous silica materials, J. Phys. Condens. Matter 16 (2004) S5403–S5436. [37] M. Sliwinska-Bartkowiak, M. Jazdzewska, L.L. Huang, K.E. Gubbins, Melting behavior of water in cylindrical pores: carbon nanotubes and silica glasses, Phys. Chem. Chem. Phys. 10 (2008) 4909–4919. [38] J.E. Morris, B.A. Afzelius, The structure of the shell and outer membranes in

extensive than in the present case of extra-cellular water in brine shrimp eggs. Finally, the localized (oftentimes called rotational) dynamics of water molecules gives rise to the signal at energy transfers above 100 μeV, as evident from our susceptibility data. This leaves us with translational dynamics of water molecules confined in extra-cellular space with characteristic dimensions ranging between ca. 2 and 10 nm, giving rise to “stretched” rather than Lorentzian-type susceptibility peaks. Two-Lorentzian fits of our data, while technically possible, would be difficult to justify on the basis of the susceptibility plots that indicate no visible spit in the Q-dependent (hence, of translational origin) peaks and the localized dynamics prominently visible beyond our data analysis range. 5. Conclusion The current quasielastic neutron scattering study of microscopic diffusion of water in encysted eggs of brine shrimp (Artemia) fully hydrated with (1) water without additives, (2) eutectic mixture of water and dimethyl sulfoxide, and (3) a concentrated aqueous solution of lithium chloride demonstrates that the diffusivity of the hydration water is reasonably well defined, despite the considerable morphological complexity of the hydrated organism. The translational diffusivity of pure water hydrating Artemia eggs decreases by at least a factor of 4 compared to the bulk value. For water-DMSO and water-LiCl hydration mixtures, the decrease in the translational diffusivity compared to the respective bulk values is smaller, yet the penetration inside the eggs of at least some DMSO and LiCl together with H2O is evident from the elimination of the hydration water freezing. Without DMSO and LiCl additives, the hydration water exhibits multi-step freezing on cooling down, completed at ca. 230 K. Together with the at least four-fold reduction in the translational diffusion coefficient, this suggests a distribution of the effective confinement pore size over the range of 2 to 10 nm. Those pores are also accessible to DMSO and LiCl co-solvents. A remarkable absence of the typical characteristics of the intra-cellular cytoplasmic water, such as freezing near 273 K and the translational diffusion coefficient similar to that in bulk water, may be linked to the unique and well documented resilience of Artemia eggs against adverse environmental factors not only in the anhydrous, but also hydrated state. Transparency document The http://dx.doi.org/10.1016/j.bbagen.2017.05.022 associated with this article can be found, in online version. Acknowledgments The neutron scattering experiments at Oak Ridge National Laboratory's (ORNL) Spallation Neutron Source were supported by the Scientific User Facilities Division, Office of Basic Energy Sciences, U.S. Department of Energy (DOE). ORNL is managed by UTBattelle, LLC, for the U.S. DOE under Contract No. DE-AC05-00OR22725. References [1] A.M. King, J. Toxopeus, T.H. MacRae, Artemin, a diapause-specific chaperone, contributes to the stress tolerance of Artemia franciscana cysts and influences their release from females, J. Exp. Biol. 217 (2014) 1719–1724. [2] M. Guppy, P. Withers, Metabolic depression in animals: physiological perspectives and biochemical generalizations, Biol. Rev. 74 (1999) 1–40. [3] K.I. Jonsson, Causes and consequences of excess resistance in cryptobiotic metazoans, Physiol. Biochem. Zool. 76 (2003) 429–435. [4] J.S. Clegg, Embryos of Artemia franciscana survive four years of continuous anoxia: the case for complete metabolic rate depression, J. Exp. Biol. 200 (1997) 467–475. [5] J.S. Clegg, S.A. Jackson, The metabolic status of quiescent and diapause embryos of Artemia franciscana (Kellogg), Arch. Hydrobiol. Spec. Issues Adv. Limnol. 52 (1998) 425–439. [6] S.-W. Wang, S.-C. Sun, R.K. Okazaki, Comparative study on thermotoleranace of Artemia resting eggs from Qinghai-Xizang Plateau, China, Aquaculture 307 (2010)

8

BBA - General Subjects xxx (xxxx) xxx–xxx

E. Mamontov

[42] M. Grimaldo, F. Roosen-Runge, M. Hennig, F. Zanini, F. Zhang, N. Jalarvo, M. Zamponi, F. Schreiber, T. Seydel, Hierarchical molecular dynamics of bovine serym albumine in concentrated aqueous solution below and above thermal dentauration, Phys. Chem. Chem. Phys. 17 (2015) 4645–4655. [43] M. Monkenbusch, A. Stadler, R. Biehl, J. Ollivier, M. Zamponi, D. Richter, Fast internal dynamics in alcohol dehydrogenase, J. Chem. Phys. 143 (2015) 075101. [44] G. Schiro, Y. Fichou, F.-X. Gallat, K. Wood, F. Gabel, M. Moulin, M. Hartlein, M. Heyden, J.-P. Colletier, A. Orecchini, A. Paciaroni, J. Wuttke, D.J. Tobias, M. Weik, Translational diffusion of hydration water correlates with functional motions in folded and intrinsically disordered proteins, Nat. Commun. 6 (2015) 6490.

encysted Artemia salina embryos during cryptobyosis and development, J. Ultrastruct. Res. 20 (1967) 244–259. [39] E. Anderson, J.H. Lochhead, M.S. Lochhead, E. Huebner, The origin and structure of the tertiary envelope in thick-shelled eggs of the brine shrimp, Artemia, J. Ultrastruct. Res. 32 (1970) 497–525. [40] D. De Chaffoy, G. De Maeyer-Criel, M. Kondo, On the permeability and formation of the embryonic cuticle during development in vivo and in vitro of Artemia salina embryos, Differentiation 12 (1978) 99–109. [41] N. Martinez, G. Michoud, A. Cario, J. Ollivier, B. Franzetti, M. Jebbar, P. Oger, J. Peters, High protein flexibility and reduced hydration water dynamics are key pressure adaptive strategies in prokaryotes, Sci. Rep. 6 (2016) 32816.

9