Integrity of Membrane Structures in Giant Unilamellar Vesicles As

lized emulsion and fusion of small unilamellar vesicles (SUVs) or large ... 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56...
0 downloads 0 Views 940KB Size
Subscriber access provided by Gothenburg University Library

Article

Integrity of Membrane Structures in Giant Unilamellar Vesicles As Assay for Antioxidants And Prooxidants Xiao-Chen Liu, Hui-Hui Du, Li-Min Fu, Rui-Min Han, Peng Wang, Xi-Cheng Ai, Jian-Ping Zhang, and Leif H. Skibsted Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.7b04383 • Publication Date (Web): 03 Jan 2018 Downloaded from http://pubs.acs.org on January 12, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

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

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

Analytical Chemistry

Xiao-Chen Liu,† Hui-Hui Du,† Li-Min Fu,† Rui-Min Han,† Peng Wang,† Xi-Cheng Ai,† Jian-Ping Zhang,†,* Leif H. Skibsted‡,* †

Department of Chemistry, Renmin University of China, Beijing, 100872, China Department of Food Science, University of Copenhagen, Rolighedsvej 30, DK-1958 Frederiksberg C, Denmark



*J.-P. Zhang: E-mail: [email protected]; Phone: +86-10-62516604 *L. H. Skibsted: E-mail: [email protected]; Phone: +45 3533 3221

ABSTRACT: We have attempted to evaluate, on the basis of optical microscopy for single giant unilamellar vesicle (GUV), the potency of antioxidants in protecting GUV membranes from oxidative destruction. Photosensitized membrane budding of GUVs prepared from soybean phosphatidylcholine with chlorophyll a (Chl a) and β-carotene (β-Car) as photosensitizer and protector, respectively, were followed by microscopic imaging. A dimensionless entropy parameter, E, as derived from the time resolved microscopic images, was employed to describe the evolution of morphological variation of GUVs. As an indication of membrane instability, the budding process showed three successive temporal regimes as a common feature: A lag phase prior to the initiation of budding characterized by LP (in s), a budding phase when E increased with a rate of kE (in s1), and an ending phase with morphology stabilized at a constant Eend (dimensionless). We show that the phase-associated parameters can be objectively obtained by fitting the E-t kinetics curves to a Boltzmann function, and that all of the parameters are rather sensitive to β-Car concentration. As for the efficacy of these parameters in quantifying the protection potency of β-Car, kE is shown to be most sensitive for β-Car in a concentration regime of 95%), N,N-dimethyl-6-propionyl-2-naphthylamine (Prodan, >98%) and 1,6-diphenyl-1,3,5-hexatriene (DPH, 98%) were purchased from Sigma-Aldrich (St. Louis, MO, USA). Extraction and purification of chlorophyll a (Chl a, >95%) from fresh spinach leaves were done as described elsewhere.33 See Scheme S1 for the chemical structures of the aforementioned regents. Sucrose (≥99%) was purchased from Amresco, Inc. (Solon, OH, USA). Chloroform and methanol in analytical reagent grade were purchased from Beijing Chemical Works (Beijing, China). Purified water was prepared using a MingcheTM-D 24UV Water Purification System (Merck Millipore Corp., Shanghai, China). All experiments were performed in a thermostated room (25±2 C). Electroformation of GUVs. GUVs were prepared using a published electroformation method with certain modifications.12,29 Briefly, the cell used for GUV preparation consisted of two indium tin oxide (ITO) plates separated by a Teflon spacer (thickness, 3 mm). A drop of 25 μL PC solution was deposited on the surface of an ITO plate, which was dried in vacuum for 2 h. A volume of 3 mL sucrose solution (100 mM) was injected into the cell. An electrical function generator (DG1022U, Rigol Technologies, Beijing, China) was employed to apply an alternative electrical field via the ITO electrodes to generate GUV suspension in the cell. In GUV preparation, the PC solution was prepared by dissolving PC reagent in chloroform at a concentration of 4 mg/mL. Chl a as the photosensitizer

was dissolved in methanol at a concentration of 8.82×10−7 M, and β-Car as the antioxidant was dissolved in acetone. The final molar ratio of PC reagent to Chl a was 100:1, PC reagent to βCar was 200:1, 500:1, 1000:1, 1500:1 or 2000:1. Optical Microscopy and Digital Image Processing. The experimental setup for fluorescence microscope (TE-2000U, Nikon Corp., JPN) has been described elsewhere.20 Briefly, 200 μL of GUVs suspension was loaded to a Costar 24-well cell culturing cluster (Corning Incorporated, Corning, NY), which was placed on a translational stage. The excitation light from an ultrahigh-pressure mercury lamp (400−440 nm; 4.0 μW) was focused into an area with a diameter of 80 μm (80 mW/cm2). A long-pass filter (cutting wavelength, 475 nm) and a band-pass filter (500−570 nm) were used to remove the excitation scattering and the fluorescence from Chl a. To trace the morphological change, GUV images were recorded every second by a EMCCD detector (Cascade II: 512, Photometrics Inc., Tucson, AZ, USA). The sampling number for each experiment was 8. The heterogeneity of images derived from digital image analysis were used to assess the physical integrity of GUVs. 21,34 Briefly, 100×100 pixels of region of interest (ROI) were picked out from the 16-bit raw images in grayscale format (512×512 pixels). The grayscale interval was divided into 256 subintervals by default. Histogram counts (p) over these subintervals were used to calculate the entropy, E = −∑plog2(p), which was used as a statistical scalar for measuring image heterogeneity.34 The GUV budding process was characterized by the change of entropy, ΔE, with reference to the average entropy at minus delay time, i.e. before exposure to light.20 Membrane Polarity and Fluidity by Fluorescence Spectroscopy. Prodan or DPH as fluorescent probes for polarity and fluidity, respective, were introduced in membranes during GUV preparation. The molar ratios of PC reagent to Prodan and DPH were 200:1. Prodan-containing GUVs doped with Chl a were illuminated in a quartz cuvette by continuous wave (CW) blue laser (405 nm, 10 mW; OBIS 405LX, Coherent Inc., USA). Fluorescence spectra were recorded under an excitation wavelength of 286 nm with a FLS 980 spectrometer (Edinburgh Instruments Ltd., UK) at a specified time. DPH-containing GUVs doped with Chl a were illuminated in a quartz cuvette by the 405-nm laser. Fluorescence polarization (P) was derived from the fluorescence intensity under light excitation at 350 nm, P = (I∥GI⊥)/(I∥+GI⊥), where I∥ and I⊥, respectively, represent fluorescence intensity measured with the emission polarization parallel and vertical to the excitation polarization. The instrumental correction factor, G = I⊥/I∥, was determined from GUVs with Chl a but without DPH.35,36

As seen in Figure 1, when protected against light exposure or exposed to ambient light, the electroformed GUVs with an average diameter of 1020 μm did not change sizes or morphology (cf. Blank), in agreement with previous findings. 30 However, GUVs added Chl a became rather sensitive to light exposure, i.e. weak blue light illumination (400−440 nm, 80 mW/cm2) for 25 s could induce morphological changes (cf. Control).

ACS Paragon Plus Environment

Page 2 of 8

Page 3 of 8

Analytical Chemistry 0s

20 s

30 s

40 s

50 s

80 s

120 s

150 s

Blank

(ΔE0+ΔEend)/2. With these fitting parameters, the middle point slope, i. e. the rate of entropy change of budding phase, can be obtained as k∆E =

PC:β-Car (200:1)

Figure 1. Microscopic GUV images recorded during the process of photosensitized oxidation. Blank: GUVs without Chl a and β-Car. Control: GUVs with Chl a (8.82×10−7 M) but without β-Car. The bottom row is for GUV incorporated with Chl a (8.82×10−7 M) and β-Car (4.41×10−7 M).

Eend

kE

PC:-Car=500:1 PC:-Car=200:1 Blank

0.6

,

(2)

(a) 6

-1

SC = 3.010 M -7 C1/2 = 1.5310 M

60

30

0

0.025

(b) 7

-1

SC = 3.910 M -8 C1/2 = 1.810 M

0.020

-1

Control PC:-Car=2000:1 PC:-Car=1500:1 PC:-Car=1000:1

4Wm

90

kE (s )

In the present experiments, the blue-light sensitive photosensitizer Chl a is efficient in yielding singlet oxygen (1O2*). The morphological changes, appeared as budding of GUV membrane, are ascribed to PC oxidation by 1O2*. The morphological changes were followed with micro-imaging and, for quantification, the digital images were subjected to heterogeneity analysis as described in Materials and Methods. As shown in Figure 2, for GUVs incorporated with Chl a (Control), the change of entropy (ΔE) increases with increasing exposure time (t). Further incorporation of β-Car at various concentration leads to a decrease of ΔE. The ΔE-t curves in Figure 2 share common features in terms of the following time regimes in sequence: A lag phase of little ΔE variation as characterized by a temporal duration LP (in s), a budding phase showing a monotonic increase of ΔE with a rate of kE (in s1), and an ending phase with a constant level of ΔEend.

∆Eend - ∆E0

where 4Wm is the time span of budding phase (in s). As exemplified by the inset of Figure 2, the Control shows a nice agreement of the fitting curve with the experimental data, and the time duration of the lag phase, LPControl, can be determined from the intersecting point of the slope line at the middle point, ΔE(t)−ΔEm = (ΔE)′m(t−tm), with the baseline, ΔE(t) = ΔE0. The GUV morphological changes are found to be dependent on the concentration of β-Car. Figure 3 shows the distinct concentration effects on the LP, kE and ΔEend parameters derived from the ΔE-t curves in Figure 2. The LPs are relative to that of the Control, i.e. LP = LPSample – LPControl (See Table S1 for details). It turns out that, with increasing β-Car concentration, LP prolongs whereas kE and ΔEend decline.

LP (s)

Control

0.015 0.010 0.005

0

20

40

60

80

LPControl

0.6

(c) 6

0.3

0.0

-1

SC = 6.110 M -7 C1/2 = 1.1310 M

0.5

Eend

E

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

0.4 0.3

0

50

t (s)

100

150 0.2

Figure 2. ΔE-t curves of GUVs with a fixed concentration of Chl a (PC: Chl a = 100:1) and varied concentration of β-Car (PC: β-Car = 2000:1, 1500:1, 1000:1, 500:1 or 200:1). Blank and Control are as described for Figure 1. Light exposure: 400−440 nm, 80 mW/cm2. Each curve is an average of 8 independent measurements. Solid lines were obtained by fitting the curves to a Boltzmann model function (Eq. 1). Inset shows the phase-associated parameters.

The parameters LP, kE and ΔEend of different temporal regimes can be simultaneously derived by fitting a ΔE-t curve to a Boltzmann model function, ΔE(𝑡) =

ΔE0 - ΔEend 1+e(t - tm)/Wm

+ ΔEend ,

(1)

where ΔE0 and ΔEend represent the lower and the upper bounds of ΔE, tm stands for the middle point time, ΔEm equals to

0

1

2 3 7 Car (10 M)

4

5

Figure 3. β-Car concentration effects on (a) lag phase parameter LP, (b) budding phase parameter kE, and (c) ending phase parameter ΔEend. The data were fitted to a mono-exponential model function to derive the concentration sensitivity, Sc in M1, and the half maximum concentration, C1/2 in M, as depicted in each panel. Fitting curves are shown in dashed lines.

The budding of GUVs has been correlated with the change in dipole moments of PCs owing to lipooxidation. 20 For Chl a doped GUVs upon light exposure, the unsaturated fatty acid moieties of PCs are easily attacked by 1O2* from Type II photosensitization of Chl a to yield lipid hydroperoxides (PC-OOH) of higher polarity with reference to parent PCs. To confirm this,

ACS Paragon Plus Environment

Analytical Chemistry

IF (normalized)

2.0

(a)

0 min 3 min 30 min 60 min

Peak 2

Peak 1

1.5

selected solvents.40 (c) Membrane polarity (EN T ) determined based on the standard curve in (b) and the peak wavelengths in (a).

The polarity increase is expected to result in a decrease in membrane fluidity, because the oxidative product, PC-OOH, may lead to an augment of hydrogen bonding. Figure 5 shows the result of fluorescence polarization probed with DPH in membrane. It is seen that the polarization parameter P increases with increasing exposure time, indicating an increase of membrane rigidity and, consequently, a retardation of 1O2* diffusion. Figure 6a shows the Chl a fluorescence images of GUVs recorded along the time course of morphological change (see Supporting Information S3 for experimental details). The fluorescence images were further processed to yield the kinetics curves as shown in Figure 6b. It is seen that the fluorescence intensity decay exponentially, indicating a first-order degradation reaction of Chl a (see Table S2 for the kinetics parameters). Clearly, the fluorescence decay becomes slower with increasing β-Car concentration, proving that β-Car to some extent prevents Chl a from photosensitized degradation. 0.85

P

we examined the polarity change of GUV membranes during the photosensitized morphological evolution. There are variety of fluorescent probes for characterizing membrane polarity, fluidity and phase separation.27,32,37 Prodan as a polarity sensitive probe is known to deposit in different regions of lipid membrane. Those locating in the interior and at the surface of membrane exhibit characteristic fluorescence maxima at 435 nm and 465 nm, respectively.38,39 In the case of Control, the pair of fluorescence maxima both shift to longer wavelengths with increasing exposure time (Figure 4a). We took the normalized electronic transfer energy (ENT ) as a metric of the environmental polarity,40 and built an ENT -wavelength curve based on the fluorescence spectra of Prodan in selected solvents (Figure 4b). The calibration curve in Figure 4b together with the fluorescence data in Figure 4a allow us to derive the dynamic change of membrane polarity (Figure 4c). It is seen that light exposure for an hour resulted in an ENT increase of 0.03, corresponding to a dielectric constant increase of 2.54 for the membranes. Thus the oxidation induced polarity increase of membrane is confirmed.

0.80

0.75

1.0 0

0.5 0.0 420

450

480

510

Wavelength (nm) 1.2

50

100

150

t (min)

Figure 5. Time evolution of the fluorescence polarization (P) probed by DPH in GUVs with Chl a (8.82×10−7 M) but without βCar. 1s

(a)

(b) water

20 s

50 s

100 s

200 s

300 s

Control 10 μm

0.8 PC:β-Car (1500:1)

ENT

10 μm

0.4

dimethylformamide

PC:β-Car (200:1) 10 μm

chlorobenzene

0.0

(b)

n-hexane

400

440

480

IF (normanlized)

(c)

Control PC:-Car=1500:1 PC:-Car=200:1

520

Wavelength (nm) 0.55

1.2

0.8

Peak 1 Peak 2

0.4

0.50

ENT

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

Page 4 of 8

0.0 0

0.30

100

200

300

t (s) 0.25 0

20

t (min)

40

60

Figure 4. (a) Fluorescence spectra of Prodan in membrane normalized at the saddle point of 452 nm. The contribution from Prodan free in solution to the longer-wavelength side were deducted based on spectral deconvolution. (b) Standard curve of fluorescence wavelength versus normalized electronic transition energy (EN T ) of

Figure 6. (a) Chl a fluorescence images of GUVs exposed to blue light (400−440 nm, 80 mW/cm2). (b) Time evolution of the fluorescence intensity obtained by integrating the brightness of the images in grey scale. Both Control and samples had a Chl a concentration of 8.82×10−7 M (PC:Chl a = 100:1). The sample GUVs contained β-Car at indicated concentration.

To further investigate the fate of Chl a and β-Car in GUVs, we simultaneously measured Chl a fluorescence and β-Car resonance Raman scattering spectra in the processes of photosensitized oxidation (Supporting Information S4). Figure 7 shows

ACS Paragon Plus Environment

Page 5 of 8

the kinetics traces of Chl a fluorescence intensity (IF) and β-Car Raman scattering intensity (IR), both of which decay monotonically. It is seen that, when IF drops nearly to a noise level, IR remains at a substantial level. These results suggest that Chl a degraded faster than β-Car did, and that β-Car remained after the depletion of Chl a. -Car Raman scattering

8

12 9

4

6

2

3

0

4

4

6

IR(10 )

Chl a fluorescence

IF (10 )

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

Analytical Chemistry

0 0

50

100

150

200

250

300

t (min) Figure 7. Relative concentration of Chl a and β-Car as indicated by spectral intensity. IR: Raman intensity of β-Car obtained as an average area under the peaks at 1154 cm−1 and 1520 cm−1 (Figure S1a). IF: Fluorescence intensity of Chl a obtained as an area under the band peaking at 678 nm (Figure S1b). Statistics are based on 3 independent measurements.

We have shown that GUVs containing photosensitizer Chl a change morphology upon exposure to blue light, and that the budding of GUVs as an indication of increased membrane instability can be quantified by a dimensionless parameter, ΔE, derived from microscopic imaging. The evolution of morphological change, ΔE-t, exhibits three cascading temporal regimes, which can be characterized by the phase-associated parameters LP, kE, and ΔEend. Hereafter, we look into the molecular mechanisms underlying the morphological evolution of GUVs, and assess the efficacy of these parameters for the characterization of membrane integrity. h O2 1O* 2 oo

oo

oo

(a) β-Car

hν 1 Chl

a*

Chl a

a [Chl PC ]

GUV budding Oxidation products

1 O* 2

3 Chl

β-Car

a* Heat

β-Car

β-3Car*

Heat

O2

β-3Car*

(b)

Scheme 1. (a) Photosensitized oxidative processes destabilizing GUV membrane. Red symbols ‘oo’ indicate the lipid hydroperoxides (PC-OOH) preferentially formed at the unsaturated C=C bonds

of PCs.41,42 Morphological variation is shown as corrugated lipid bilayer. (b) Chl a and β-Car involved photochemical and photophysical reactions in GUV membrane during photosensitized oxidation. Shaded regions indicate the case of Control without β-Car involved.

Mechanisms of Morphological Change in the Case of Control. In the case of Control, the photosensitized lipooxidation results in a three-phase morphological evolution of GUV membrane. (i) Lag phase. As illustrated by the reaction paths of shaded areas in Scheme 1b, upon light irradiation, 1O2* starts to diffuse in lipid bilayer to attack the unsaturated C=C bonds of PC to form PC-OOH (cf. Scheme 1a). This is in accord with the membrane polarity increases owing to PC-OOH accumulation, which is verified by the time dependent spectral shift of Prodan fluorescence (Figure 4). In the lag phase spanning 25 s, Chl a itself can also be oxidized as indicated by the fluorescence decay of the time resolved images (Figure 6). Previous studies suggest that PC-OOH tends to move to the lipid-water interface to be hydrated, which results in an increase of the total membrane area. When the area grows to a turning point breaking the mechanical balance, budding takes place for membrane to find a new equilibrated morphology. 27,42-44 Before the turning point, however, photosensitized oxidation does not lead to any obvious morphological change. The reactions involved in lag phase may be categorized into the followings. Type II photosensitization: Chl a + hν  1Chl a* Chl a*  3Chl a*

1

Chl a* + O2  Chl a + 1O2*. Lipooxidation:

(3)

O2* + PC  PC-OOH. Oxidative degradation of Chl a:

(4)

3

1

O2 + Chl a + hν  {O2∙∙∙3Chl a*}  {1O2*∙∙∙Chl a}  Oxidation product. (5) In Eq. 5, we suggest the involving of {O2∙∙∙3Chl a*} and {1O2*∙∙∙Chl a} complexes as reaction intermediates, which draws support from the first-order reaction behavior of Chl a degradation (Figure 6). From lag phase to budding phase, the diffusion of 1O2* becomes more and more sluggish owing to the increased membrane rigidity as reported by DPH fluorescence polarization (cf. Figure 5). A low 1O2* diffusibility should be prone for the formation of {1O2*∙∙∙Chl a} complexes. (ii) Budding phase. The budding phase accounting for distinct morphological change begins immediately after the lag phase and, as the result of increased amount of PC-OOH, it propagates over a timescale comparable to lag phase (25 s; Figure 2). Chl a molecules are degraded continuously also in this phase as indicated by the time-resolved fluorescence images or by the corresponding kinetics curves (Figure 6). (iii) Ending phase. The photosensitized oxidation ends in this phase, because Chl a photosensitizer is depleted at the end of its previous evolution phase. Therefore, the stabilized membrane morphology characterized by Eend is due to the terminated 1 O2* supply. Mechanisms of GUV Membrane Structure Protected by -Car And the Parametric Quantification. As depicted in Scheme 1b, -Car is involved in antioxidation via the reaction

ACS Paragon Plus Environment

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

paths highlighted in orange. The involved reactions are known as, Quenching of 3Chl a*: Chl a* + Car  Chl a + 3Car* Quenching of 1O2*:

(6)

O2* + Car  O2 + 3Car* Excitation dissipation:

(7)

3

1

Car*  Car + heat (8) Both of the excitation quenching reactions prevent PC and Chl a from being oxidized by 1O2*. The triplet excitation energy transfer reaction of Eq. 6, occurring on a timescale of 10 ns,45,46 is much faster than 1O2* diffusion in membrane with a timescale of 10 s over a length of 0.1 m as estimated from a diffusion constant of (14.75)×10−9 m2s−1.47,48 On the other hand, the scavenging process of Eq. 7 is nearly diffusion limited as reported for Cars in plasma.49 Therefore, -Car incorporation helps to prolong both lag phase and budding phase of the morphological change, i.e. to maintain the membrane integrity of GUVs. Note that ending phase correlates somehow with both lag phase and budding phase, i.e. a smaller ΔEend is accompanied by a longer LP and a lower kE (Figure 3). 3

As shown in Figure 7 by resonance Raman spectroscopy, Car in membrane experiences photodegradation, which is likely via a mechanism similar to the self-destruction of Chl a (Eq. 5). However, unlike Chl a that was depleted, β-Car remained in the ending phase despite a low concentration. These results suggest that Chl a molecules are not fully protected by β-Car, because the prerequisite for reaction Eq. 6, a Chl a in van der Waals contact with a β-Car, is not necessarily always satisfied. In addition, sinceβ-Car survived even after the oxidative depletion of Chl a, we conclude that β-Car plays an antioxidation role over the entire time course of morphological evolution. In this context, the Chl a concentration in our experiments was 8.82×10−7 M. In air-saturated sucrose solutions, the oxygen concentration is known to be on the order of 1×10−3 M, and the oxygen diffusivity is reported to be (1.392.53)×109 m2s−1.50,51 The high diffusivity and the much higher concentration than Chl a imply that oxygen participated in the entire processes of photooxidation. Regarding the parametric quantification of the membrane integrity protected by -Car, the results of concentration effects enable us to assess the efficacy of LP, kE and ΔEend in characterizing the protection potency of β-Car. As shown in Figure 3, the concentration dependence for each parameter can be described by a mono-exponential model function, and the according curve fitting yields a concentration sensitivity, SC (in M1), and a half maximum concentration, C1/2 (in M). It is seen in Figure 3 that SC (C1/2) of kE is 6-fold larger (smaller) than those of LP and ΔEend. Therefore, the efficacy of the parameters for characterizing the protection potency of β-Car can be described below: For β-Car at a low concentration regime, i.e.