Transformation between Inverse Bicontinuous Cubic Phases of a Lipid

Subscriber access provided by UNIV LAVAL

Article

Transformation between inverse bicontinuous cubic phases of a lipid from diamond to primitive Toshihiko Oka Langmuir, Just Accepted Manuscript • DOI: 10.1021/la504295v • Publication Date (Web): 26 Feb 2015 Downloaded from http://pubs.acs.org on March 3, 2015

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.

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

Page 1 of 24

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

Langmuir

1

Transformation between inverse bicontinuous cubic

2

phases of a lipid from diamond to primitive

3

Toshihiko Oka*,†,§

4

†Department of Physics, Graduate School of Science, §Nanomaterials Research Division,

5

Research Institute of Electronics, Shizuoka University, Shizuoka 422-8529, Japan

6 7

ACS Paragon Plus Environment

1

Langmuir

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

Page 2 of 24

1

ABSTRACT. I studied the transformation between inverse bicontinuous cubic phases (QII) from

2

diamond (QIID) to primitive (QIIP) in a single-crystal region of monoolein. X-ray diffraction data

3

reveal that the crystallographic orientation of QIIP rotates 55° around the [0 1 1] axis from QIID.

4

This indicates that one direction of the four-branched water channels in the QIID phase is

5

preserved in the six-branched water channels of the QIIP phase. I, therefore, built a transformation

6

model that would keep the direction of the water channels fixed in both phases and cause the

7

water channels along other direction in QIID to shrink and disappear.

8


2

Page 3 of 24

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

1

Langmuir

INTRODUCTION

2

A triply periodic minimal surface (TPMS) has an infinitely extending structure and separates

3

space into two interwoven regions1,2 (Figure 1). Three TPMSs, viz., diamond (D), primitive (P),

4

and gyroid (G), which have cubic symmetry with the space groups Pn3 m , Im3m , Ia3 d ,

5

respectively, are frequently encountered in real systems, such as lyotropic liquid crystals3,4,

6

diblock copolymers5, prolamellar bodies of plants6, and organized smooth endoplasmic

7

reticulum7. Inverse bicontinuous cubic (QII) phases formed by lipid-water systems are typical

8

examples. QII phases having a diamond (QIID), primitive (QIIP), or gyroid (QIIG) structure have

9

been observed3.

10

The three TPMS can be transformed into each other mathematically by Bonnet

11

transformation8, but the transformation requires that parts of the minimal surfaces pass through

12

each other. The transformation also requires tearing the membrane of the physical system, thus

13

the Bonnet transformation between QII phases is physically unrealistic3,9. Another mechanism

14

has been proposed that simplifies the transformations as mergence and separation of junctions of

15

a water network labyrinth 10,11. This mechanism indicated that transformation without tearing is

16

feasible. Moreover, it has been shown that transformations between all three TPMS structure can

17

preserve both topology and zero mean curvature throughout12, and corresponding intermediate

18

structures have been proposed12–14. In the transformation between D and P, an intermediate with

19

rhombohedral symmetry has been proposed. This transformation between the D and P surfaces is

20

obtained by pushing or pulling along one of the directions12. The resulting changes in

21

surface structure have been illustrated13. Thus, it has been predicted that the crystallographic

22

orientation would rotate 60° around one of the axes during transformation in a real

23

system9.


3

Langmuir

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

Page 4 of 24

1

Exploring the orientational (or epitaxial) relationships in transformations between the QII phases

2

requires a technique to control the crystal orientation. Several such techniques have been

3

reported15–18 and employed to study the phase transition between QIIG and QIID 16,19. However,

4

only one direction of the crystal orientation can be aligned through these techniques. Thus, it

5

seems difficult to elucidate orientational relationships in transformations between QII phases. My

6

group, therefore, has developed a method to make a single crystal domain of the QIID phase20,

7

which clearly exhibited a crystal orientation of the QIID phase. Using this method, I studied the

8

QIID-QIIP phase transition in the single crystal region.

9

MATERIALS AND METHODS

10

A single crystal region was prepared according to a procedure reported previously20 except that

11

the additive was replaced by t-butyl alcohol (tBA). A 80:20 v/v solution of water/ tBA (Wako

12

Pure Chemical Industries, Osaka, Japan) was prepared. The solution was mixed with dried

13

monoolein (1-oleoyl-rac-glycerol) (Sigma Chemical Co., St. Louis, MO, USA) in a 60:40 w/w

14

solvent/monoolein ratio. The sample was loaded into a polyimide capillary (PIT-S, Furukawa

15

Electric Co., Tokyo, Japan), whose inner diameter was 0.5 mm. The length of the loaded sample

16

was about 30 mm. One end of the sample was sealed with a tube sealing compound (Chaseal,

17

Chase Scientific Glass, TN, USA). The other end of the sample was soaked in water at 25°C.

18

After about one month, the sample in the capillary separates into two parts: a powder region and

19

a single crystal region of the QIID phase20. The powder region was cut off, and the remaining

20

portion was transferred into a 20 mM L-arginine (Arg) (Wako Pure Chemical Industries, Osaka,

21

Japan) solution in a horseshoe-shaped aluminum frame. The thickness of the frame was 1.0 mm,

22

and its inner width and height were 20 and 35 mm, respectively. Both sides of the frame were


4

Page 5 of 24

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

Langmuir

1

sealed with 20 µm polyimide films. The volume of the 20 mM Arg in the frame was about 500

2

µl.

3

X-ray diffraction measurements were conducted in almost the same manner as in my previous

4

paper20. The rotating crystal method was used. The sample-detector distance was set to 495 mm.

5

The sample capillary in the horseshoe-shaped frame was mounted on a goniometer head. The

6

sample capillary axis was set vertically and adjusted to the rotating axis. Measurements were

7

performed at 25°C. The sample was rotated from 0° to 180° in increment steps of 2°, and X-rays

8

for 10 s in each step. To measure the phase transition at the same sample position, measurements

9

were repeated at intervals of 1 or 2 h. The results of the diffraction data analysis were almost the

10

same as those reported previously20. I determined the peak positions of the diffraction spots with

11

the display software ALBULA (Dectris, Baden, Switzerland). Using these peak positions,

12

rotation matrices of the crystal orientations of QIID (RD) and QIIP (RP) were calculated from the

13

base coordinate. Of the data for the QIID phase, diffraction spots from the {110}, {111}, {200},

14

{211}, {220}, {221}, and {222} planes were used in the calculation. In the case of the QIIP

15

phase, spots from {110}, {200}, {211}, and {222} planes were used in the calculation.

16

Diffraction spots from the {220} planes of the QIIP phase, which were not included in the

17

extinction rules, could not be observed. Some of the spots listed above were not observed

18

because they were located outside the detection area or at a blind angle with respect to the

19

horseshoe-shaped frame. The rotation matrices RD and RP were used to calculate the peak

20

positions of unobservable diffraction spots in the extinction rules, and then the phase of the

21

sample was identified. The rotation matrix from QIID to QIIP, RDP, was calculated via RDP = RD-

22

1

23

RDP is an eigenvector of the matrix. The rotation angle, θ, around the rotation axis was calculated

RP. A rotation matrix is defined by a rotation axis and a rotation angle21. The rotation axis of


5

Langmuir

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

1

from the trace of RDP via 1+2cosθ = TrRDP. The detailed calculations are in the Supporting

2

Information.

3

I constructed the model of a hypothetical intermediate in Figure 5 as follows. I aligned

4

orientations of two four-way junctions of the QIID phase and one six-way junction of the QIIP

5

phase. Next, I calculated the midpoint of the corresponding junctions in the QIID and QIIP phases

6

as the hypothetical intermediate. Then, I calculated the periodical structure of the intermediate

7

using the periodicity of the QIID and QIIP phases.

Page 6 of 24

8 9

RESULTS AND DISCUSSION

10

Figure 2 illustrates the sample preparation process. To obtain a long single crystal region of

11

monoolein in a capillary, I used a method reported previously by my group20, with the

12

modification of using tBA instead of 1,4-butanediol. The length of the single crystal region was

13

about 5 mm. Arg solution is known to convert the QIID phase of monoolein into QIIP 22. The open

14

end of the capillary sample was soaked in 20 mM Arg to change the phase of the single crystal.

15

Thus, the slow penetration of Arg solution led to a gradual QIID-QIIP phase transition from the

16

open end. Figure 3 shows the time evolution of the X-ray diffraction patterns of the sample 3.5

17

mm from the edge (Supporting Information Figure S1). The X-ray diffraction patterns of the

18

capillary sample just after soaking (data not shown) and up to 7.5 h after soaking (Figures 3a and

19

3d) indicated that only one orientation of the crystal existed in the exposure region. Analysis of

20

the data from all angles confirmed that the phase of the lipid was QIID, which agreed with

21

previous results20. After 12 h (Figures 3b and 3e), the spots of the QIID phase were still observed,

22

but new diffraction spots had appeared as well. After 16 h (Figures 3c and 3f), the spots due to


6

Page 7 of 24

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

Langmuir

1

the QIID phase disappeared completely and only the new diffraction spots seen in Figures 3b and

2

3e were observed. Analysis of these new spots indicated that the phase was QIIP. The lattice

3

constants of the QIID and QIIP phases are 10.8 and 13.8 nm, respectively. The ratio of these lattice

4

constant is 1.28, which agrees exactly with the theoretical value of 1.288. Diffraction patterns of

5

the QIID phase at 7.5 h and 12 h show clear spots, and I could pick up the spot positions and

6

easily determine the orientation of the single crystal. The diffraction spots of the QIIP phase at 12

7

h and 16 h have circular tails, but I could still determine the orientation of the single crystal.

8

Some weak spots were observed at positions where no spots appeared from the single crystal.

9

Thus, I concluded there was one large single crystal and some small crystals of the QIIP phase. In

10

Figure 3d, a diffraction spot from the (0 1 1) plane of QIID is observed. In Figure 3e, a new

11

diffraction spot appeared immediately to the left of the (0 1 1) plane of QIID, which was assigned

12

to the (0 1 1) plane of the QIIP phase. In Figure 3b, a diffraction spot from the (111) plane of QIID

13

adjoins one from the (200) plane of QIIP.

14

From the positions of the diffraction spots in the two-dimensional images, I could calculate the

15

positions of the spots in reciprocal space (see Figure 4). Figure 4a is the view along the [0 1 1]

16

direction of the QIID phase, which almost corresponds to the [0 1 1] direction of the QIIP phase.

17

Figure 4b is the view along the [111] direction of the QIID phase, which almost corresponds to

18

the [200] direction of the QIIP phase. I calculated the rotation matrix RDP from QIID to QIIP. A

19

rotation matrix is defined by a rotation axis and a rotation angle21. The normalized rotation axis

20

of the rotation matrix RDP was [0.143 -0.658 0.740], and the calculated rotation angle of the

21

matrix was 54°. Visual inspection in reciprocal space indicated that the [0 1 1] directions of the

22

two phases were nearly the same, as were the [111] direction of QIID and the [200] direction of


7

Langmuir

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

Page 8 of 24

1

QIIP (Figure 4). The rotation matrix satisfying these relationships must have [0 1 1] as its rotation

2

axis and a rotation angle of 55°. The rotation axis calculated from the rotation matrix RDP was

3

close to the [0 1 1] direction, and the angle difference was 9°. The difference between the

4

rotation angles is very small. I, therefore, concluded that the crystallographic orientation rotated

5

55° around the [0 1 1] axis in the phase transition from QIID to QIIP.

6

Using this orientational relation, I propose a model for the phase transition from QIID to QIIP.

7

Figure 5 shows water channels of QIID, a hypothetical intermediate, and QIIP, viewed along the

8

[0 1 1] direction. The lipid surface changes with the water channels, however, to simplify the

9

model, I only treat the transformation of water channels explicitly. The phase transition starts

10

from the QIID phase (Figure 5a). The QIID phase has simplified water channels along four

11

directions: [111], [1 1 1 ] , [ 1 1 1 ] , and [ 1 1 1] . At the start of the transition, water channels along

12

the [1 1 1 ] direction start to shrink. In the hypothetical intermediate, water channels along the

13

[1 1 1 ] direction in QIID shorten (Figure 5b). But one direction of the water channels is fixed,

14

namely, the [111] direction of the water channels in the QIID phase. Then, the [1 1 1 ] direction of

15

water channels in the QIID phase disappear, and the QIIP phase is formed (Figure 5c). The [111]

16

direction of water channels in the QIID phase is still fixed in the QIIP phase, but the Miller indices

17

of the direction become [200] because the crystallographic orientation changes during the

18

transition.

19

The transformation from D to P has been studied theoretically10–14, but the orientational

20

relationships observed in a single crystal during the D-P transformation remains poorly studied.

21

Squire et al. only predicted that one of the axes in the crystallographic orientation of both


8

Page 9 of 24

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

Langmuir

1

phases was fixed during the phase transition according to the rhombohedral pathway model9. I

2

expected that the direction of the water channels of the QIID phase, which disappear in the phase

3

transition, would be fixed. But my data showed that the [0 1 1] axis is fixed as the rotation axis of

4

crystallographic orientation, and one other direction of the water channels is conserved. On the

5

other hand, my model seems to agree well with the theoretical model for the surface

6

transformation. The concept of merging junctions developed by Sadoc and Charvolin10 was used

7

in my transformation model. Hyde and coworkers have reported a rhombohedral pathway model

8

that preserves the topology and zero mean curvature throughout12–14. The hypothetical

9

intermediate in Figure 5b has rhombohedral symmetry. Thus, my transition model seems to

10

agree well with the rhombohedral pathway model. Because the rhombohedral pathway model

11

was developed on the basis of a unit cell, the model tolerate the rotation of the unit cell.

12

Moreover, the transformation model of Squire et al. is in agreement with my model, except for

13

the orientational relationship. Thus, my model and previous theoretical models by other groups

14

differ only in terms of the direction preserved during the transformation.

15

The QIID-QIIP phase transition occurs at the interface between the two phases in the single-crystal

16

region. Thus, the newly formed QIIP phase contacts with the existing QIID phase. This contact

17

presumably affects the crystallographic orientation in the phase transition. Thus, having a

18

membrane forming adequate connection between the QIID and QIIP phases is important for

19

decreasing the transition energy and hence preserving the direction of the water channels. The

20

preservation, therefore, would be energetically favorable in a macroscopic phase transition.

21

Epitaxial relationships involving the type I cubic phase also seem to minimize the movement of

22

the lipid and water during phase transition23. Further experimental and theoretical studies are

23

required in which the effect of the interface between the QIID and QIIP phases are considered.


9

Langmuir

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

Page 10 of 24

1

In the rhombohedral pathway model, the rPD surface was considered as a pathway for the

2

transformation from D to P surfaces12,13. In the transformation, the c axis length of the

3

rhombohedral unit cell of the rPD surface decreases by 36% and the a axis length increases by

4

28%24. This indicates that the length of the [1 1 1 ] direction of D shrinks by 36% in P, while the

5

length of the direction perpendicular to [1 1 1 ] expands. My model in Figure 5 also shows the

6

disappearance of the water channel of the [1 1 1 ] direction in the QIIP phase. The single crystal of

7

the QIID phase is a liquid crystal, and thus, the membrane surface and water channels are

8

movable during the phase transition. But a large length change in one direction would damage

9

the single crystal. Indeed, X-ray diffraction reveals deterioration of the single crystal during the

10

phase transition.

11

I concluded that one direction of the water channels is preserved: the [111] direction of the QIID

12

phase corresponds to [200] of QIIP. But there are eight choices because the set of axes

13

includes eight directions. Thus, there is reason to restrict the crystallographic orientation in the

14

transition. In this experiment, Arg solution was used to induce phase transition in the lipid in the

15

capillary. Thus, the phase transition advanced in one direction along the capillary axis from the

16

open end of the capillary located at the bottom. This would affect the choice of axis. Indeed, my

17

data indicates that the preserved directions of the water channels were almost perpendicular to

18

the capillary axis. The direction in which the phase transition advances might restrict the

19

preserved directions of the water channels.

20

The phase of monoolein is QIID under excess water conditions. In this report, I used Arg to obtain

21

the QIIP phase from the QIID phase22. But the phase transition mechanism is unclear. Yamazaki

22

and coworkers have been reported that the phase of monoolein with charged lipids such as


10

Page 11 of 24

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

Langmuir

1

dioleoylphosphatidyc acid depended on the surface charge, and observed the QIIP phase4,25. Thus,

2

Arg might affect the membrane surface and stabilize the QIIP phase.

3

The X-ray powder diffraction technique has been widely used in the analysis of QII phase

4

transitions9,26,27. However, data obtained by this technique does not include information on

5

sample orientation. By contrast, the single-crystallization method20 and alignment technique15–18

6

of the QII phase allow us to study the orientation of the phases during the transition. These

7

techniques promise to yield high-quality structural information on the QII phases.

8

CONCLUSION

9

By using the single crystal region of the QII phase of monoolein, I revealed orientaional

10

relationships in the QIID-QIIP phase transition. I also proposed a model for the transition in which

11

one direction of the water channels was preserved while the other direction shrank and

12

disappeared. The change of crystallographic orientation in my model did not agree with those

13

reported in previous theoretical works, but in all other respects, my model did agree well with

14

previous models. This result is one example of the benefits produced by the single crystallization

15

technique of the QII phase. Moreover, the techniques used in this paper would help us to obtain

16

acquire further information about the transformation between QII phases.

17


11

Langmuir

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

1

Page 12 of 24

FIGURES

2 3

Figure 1. TPMS structures of (a) D and (b) P. 2×2×2 unit cells are drawn. The TPMS separates

4

space into two interwoven regions. The two different tones of pipes indicate the centers of the

5

two interwoven regions. In the structure of the QII phases of the lipid, the positions of the TPMS

6

are occupied by lipid bilayer membranes, and the two interwoven regions are occupied by water.

7

(c) Adjacent two four-way junctions of interwoven regions in D. (d) One six-way junction of

8

interwoven regions in P. The two four-way junctions of D merge into the one six-way junction of

9

P in the D-P transformation model with rhombohedral symmetry. In this model, the

10

transformations are obtained by pushing along one of the directions in D, and a pipe of

11

the direction in D shrinks and disappears in P.

12


12

Page 13 of 24

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

Langmuir

1

2 3

Figure 2. Schematic of the sample preparation process. (a) Monoolein sample before soaking in

4

water. Concentration of t-butyl alcohol (tBA) was 20%. The phase of monoolein was L3. (b)

5

Sample soaked in water for several days. The tBA in the capillary diffused out gradually from

6

the open edge, and a concentration gradient of tBA was formed in the capillary20. Monoolein at

7

the open edge formed polycrystals of the QIID phase. (c) Sample soaked in water for one month.

8

The concentration gradient of tBA was lower than in (b). A single crystal region of QIID was

9

formed far from the open edge20. (d) The polycrystal region was cut off; the remaining part

10

consisted of only the single crystal region. (e) The single crystal region was soaked in 20mM L-

11

arginine (Arg). Arg diffused slowly into the capillary from the open edge, and the single crystal

12

region of QIIP was formed gradually. (f)(g) X-ray diffraction patterns of the sample in (e) were

13

measured from just after the Arg soak to about 1 day with roughly about 1-2 hour intervals. The

14

capillary axis was adjusted to the rotating axis and perpendicular to the X-ray direction. The

15

crystallographic orientation of the single crystal region was not aligned with the capillary axis.

16

The sample was rotated from 0° to 180° in increment steps of 2°, and exposed to X-rays for 10 s

17

at each step. The phase of the sample in the X-ray exposed area was QIID, as in (f). As time


13

Langmuir

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 14 of 24

1

passed, the QIIP grew, and the boundary between QIID and QIIP passed through the X-ray exposed

2

area. The phase of the sample under X-rays was QIIP, as in (g).

3 4


14

Page 15 of 24

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

Langmuir

1

2 3

Figure 3. X-ray diffraction images of the capillary sample, 3.5 mm from the edge immersed in 20

4

mM Arg solution. The images are cumulative data of rotation angles (a-c) from 30° to 40°, (d-f)

5

from 110° to 120°. Only the right halves of the images are shown because the original images are

6

centrosymmetric. (a), (d) After 7.5 h of soaking in the Argsolution. (b), (e) After 12 h. (c), (f)

7

After 16 h. Diffraction spots of the QIID phase are seen in (a) and (d), and those of the QIIP phase

8

are in (c) and (f). Diffraction spots from both phases can be seen in (b) and (e). Some strong

9

peaks from the QIID and QIIP phases are surrounded by dashed circles of red and blue,

10

respectively. Some Miller indices of the QIID phase are assigned in (a) and (d), and those of the

11

QIIP phase are assigned in (c) and (f).

12 13


15

Langmuir

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 16 of 24

1 2

Figure 4. Observed positions of diffraction spots of the QIID phase (red circles) and the QIIP phase

3

(blue crosses) in reciprocal space. Diffraction spots that satisfy |h|, |k|, |l| ≤ 2 (h, k, l are the Miller

4

indices) are shown. Diffraction spots from {200} planes and the origin are drawn with lines:

5

those of QIID are red solid, and those of QIIP are blue dashed. (a) Viewed along roughly the

6

[0 1 1] direction of both phases, (b) along roughly the [111] direction of the QIID phase, and

7

along roughly the [200] direction of the QIIP phase.

8


16

Page 17 of 24

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

Langmuir

1 2

3 4

Figure 5. Model of the transition pathway from the QIID phase to the QIIP phase. Pipes indicate

5

the centers of the water channels. Different tones of pipes indicate two different water channels.

6

Models are viewed along the [0 1 1] direction of QIID or QIIP. (a) The QIID phase. The [111]

7

direction of the water channels is fixed during the phase transition. Water channels of the [1 1 1 ]

8

direction begin to shrink (red dashed arrows) at the start of the phase transition. (b) Hypothetical

9

intermediate. Water channels of one direction (the [1 1 1 ] direction in the QIID phase) is short. (c)

10

The QIIP phase of the corresponding orientation. The [111] direction of water channels of the

11

QIID phase (green dotted line) is fixed and corresponds to the [200] direction of water channels of

12

the QIIP phase. All directions marked with arrows or lines in the figures are perpendicular to the

13

[0 1 1] viewing direction. The angle between the [200] and [111] (or [1 1 1 ] ) directions is 55°,


17

Langmuir

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 18 of 24

1

which is the rotation angle around the [0 1 1] direction in the transformation. The water channels

2

of the [1 1 1 ] direction in the QIID phase (a) (red dashed line) shrink (b) and disappear (c) during

3

the transformation.

4


18

Page 19 of 24

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

Langmuir

1

ASSOCIATED CONTENT

2

Supporting Information. Sequential X-ray diffraction images for Figures 3 and two movies for

3

Figure 5. This material is available free of charge via the Internet at http://pubs.acs.org.

4

AUTHOR INFORMATION

5

Corresponding Author

6

*[email protected]

7

Notes

8

The authors declare no competing financial interests.

9

ACKNOWLEDGMENT

10

I am grateful to Prof. S. Hyde for helpful discussions. I appreciate the critical reading of the

11

manuscript by Prof. M. Yamazaki.

12 13


19

Langmuir

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

1 2

REFERENCES (1)

3 4

(2)

(3)

Seddon, J. M.; Templer, R. H. In Structure and Dynamics of Membranes; North-Holland, 1995; Volume 1, pp. 97–160.

(4)

9 10

Lord, E. A.; Mackay, A. L. Periodic Minimal Surfaces of Cubic Symmetry. Curr. Sci. 2003, 85, 346–362.

7 8

Hyde, S.; Ninham, B.; Andersson, S.; Larsson, K.; Landh, T.; Blum, Z.; Lidin, S. The Mathematics of Curvature. In The Language of Shape; Elsevier, 1997; pp. 1–42.

5 6

Page 20 of 24

Yamazaki, M. In Advances in Planar Lipid Bilayers and Liposomes; Academic Press, 2009; Volume 9, pp. 163–209.

(5)

Hajduk, D. A.; Harper, P. E.; Gruner, S. M.; Honeker, C. C.; Kim, G.; Thomas, E. L.;

11

Fetters, L. J. The Gyroid: A New Equilibrium Morphology in Weakly Segregated Diblock

12

Copolymers. Macromolecules 1994, 27, 4063–4075.

13

(6)

14 15

Prolamellar Body in Squash Etioplasts. Protoplasma 1985, 128, 147–156. (7)

16 17

Murakami, S.; Yamada, N.; Nagano, M.; Osumi, M. Three-Dimensional Structure of the

Almsherqi, Z. A. Cubic Membranes: A Legend beyond the Flatland* of Cell Membrane Organization. J. Cell Biol. 2006, 173, 839–844.

(8)

Hyde, S. T.; Andersson, S.; Bodil, E.; Kåre, L. A Cubic Structure Consisting of a Lipid

18

Bilayer Forming an Infinite Periodic Minimum Surface of the Gyroid Type in the

19

Glycerolmonooleat-Water System. Z. Kristallogr. - Cryst. Mater. 1984, 168, 213–220.

20

(9)

Squires, A. M.; Templer, R. H.; Seddon, J. M.; Woenkhaus, J.; Winter, R.; Narayanan, T.;

21

Finet, S. Kinetics and Mechanism of the Interconversion of Inverse Bicontinuous Cubic

22

Mesophases. Phys. Rev. E 2005, 72, 011502.


20

Page 21 of 24

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

1 2

Langmuir

(10) Sadoc, J. F.; Charvolin, J. Infinite Periodic Minimal Surfaces and Their Crystallography in the Hyperbolic Plane. Acta Crystallogr. A 1989, 45, 10–20.

3

(11) Benedicto, A. D.; O’Brien, D. F. Bicontinuous Cubic Morphologies in Block Copolymers

4

and Amphiphile/Water Systems: Mathematical Description through the Minimal Surfaces.

5

Macromolecules 1997, 30, 3395–3402.

6 7

(12) Fogden, A.; Hyde, S. T. Continuous Transformations of Cubic Minimal Surfaces. Eur. Phys. J. B. 1999, 7, 91–104.

8

(13) Schröder, G. E.; Ramsden, S. J.; Fogden, A.; Hyde, S. T. A Rhombohedral Family of

9

Minimal Surfaces as a Pathway between the P and D Cubic Mesophases. Phys. A

10

(Amsterdam, Neth.). 2004, 339, 137–144.

11

(14) Schröder-Turk, G. E.; Fogden, A.; Hyde, S. T. Bicontinuous Geometries and Molecular

12

Self-Assembly: Comparison of Local Curvature and Global Packing Variations in Genus-

13

Three Cubic, Tetragonal and Rhombohedral Surfaces. Eur. Phys. J. B. 2006, 54, 509–524.

14

(15) Seddon, A. M.; Lotze, G.; Plivelic, T. S.; Squires, A. M. A Highly Oriented Cubic Phase

15

Formed by Lipids under Shear. J. Am. Chem. Soc. 2011, 133, 13860–13863.

16

(16) Rittman, M.; Amenitsch, H.; Rappolt, M.; Sartori, B.; O’Driscoll, B. M. D.; Squires, A.

17

M. Control and Analysis of Oriented Thin Films of Lipid Inverse Bicontinuous Cubic

18

Phases Using Grazing Incidence Small-Angle X-Ray Scattering. Langmuir 2013, 29,

19

9874–9880.

20 21

(17) Squires, A. M.; Hallett, J. E.; Beddoes, C. M.; Plivelic, T. S.; Seddon, A. M. Preparation of Films of a Highly Aligned Lipid Cubic Phase. Langmuir 2013, 29, 1726–1731.


21

Langmuir

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

1

Page 22 of 24

(18) Rappolt, M.; Gregorio, G. M. D.; Almgren, M.; Amenitsch, H.; Pabst, G.; Laggner, P.;

2

Mariani, P. Non-Equilibrium Formation of the Cubic

Pn

3

3

Phase in a Monoolein/water System. Europhys. Lett. EPL 2006, 75, 267–273.

M

4

(19) Seddon, A. M.; Hallett, J.; Beddoes, C.; Plivelic, T. S.; Squires, A. M. Experimental

5

Confirmation of Transformation Pathways between Inverse Double Diamond and Gyroid

6

Cubic Phases. Langmuir 2014, 30, 5705–5710.

7 8 9 10 11 12 13

(20) Oka, T.; Hojo, H. Single Crystallization of an Inverse Bicontinuous Cubic Phase of a Lipid. Langmuir 2014, 30, 8253–8257. (21) Goldstein, H. Classical Mechanics; Addison-Wesley Pub. Co.: Reading, Mass., 1980. (22) Cherezov, V.; Clogston, J.; Papiz, M. Z.; Caffrey, M. Room to Move: Crystallizing Membrane Proteins in Swollen Lipidic Mesophases. J. Mol. Biol. 2006, 357, 1605–1618. (23) Rancon, Y.; Charvolin, J. Fluctuations and Phase Transformations in a Lyotropic Liquid Crystal. J. Phys. Chem. 1988, 92, 6339–6344.

14

(24) Hyde, S. T. Personal Communication, 2014.

15

(25) Li, S. J.; Yamashita, Y.; Yamazaki, M. Effect of Electrostatic Interactions on Phase

16

Stability of Cubic Phases of Membranes of Monoolein/Dioleoylphosphatidic Acid

17

Mixtures. Biophys. J. 2001, 81, 983–993.

18

(26) Alam, M. M.; Oka, T.; Ohta, N.; Yamazaki, M. Kinetics of Low pH-Induced Lamellar to

19

Bicontinuous Cubic Phase Transition in Dioleoylphosphatidylserine/monoolein. J. Chem.

20

Phys. 2011, 134, 145102.

21

(27) Oka, T.; Tsuboi, T.; Saiki, T.; Takahashi, T.; Alam, J. M.; Yamazaki, M. Initial Step of

22

pH-Jump-Induced

Lamellar

to

Bicontinuous

Cubic

Phase

23

Dioleoylphosphatidylserine/Monoolein. Langmuir 2014, 30, 8131–8140.

Transition

in


22

Page 23 of 24

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

Langmuir

1 2


23

Langmuir

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 24 of 24

1 2

Table of Contents Graphic

3


24

Transformation between Inverse Bicontinuous Cubic Phases of a Lipid

Recommend Documents