Effects of Glycon and Temperature on Self-Assembly Behaviors of α-Galactosyl Ceramide in Water

α ‑Galactosyl Ceramide in Water

Article

Eﬀects of Glycon and Temperature on Self-Assembly Behaviors of

Ryo Miyazaki, Mahmuda Nargis, Abu Bin Ihsan, Noriyuki Nakajima, Masahiro Hamada,

and Yasuhito Koyama*

Cite This: Langmuir 2021, 37, 7936 −7944

Read Online

ACCESS

Metrics & More

Article Recommendations

sı Supporting Information

ABSTRACT: α-Galactosyl ceramide (GalCer) is an anticancer glycolipid consisting of D-galactose and phytosphingosine-based

ceramide. Although the amphiphilic structure of GalCer is expected to form self-associates in water, the self-assembly behaviors of

GalCer and its derivatives have not been systematically investigated at this moment in spite of its great importance. The evaluation

of morphologies and properties of the associates should open new insights into glycolipid chemistry such as the application of

GalCer derivatives to a nanocarrier and the elucidation of the detailed pharmacological mechanism of GalCer. Herein, we show the

synthesis of the aglycon fragment (Aglycon) of GalCer and the self-assembly behaviors of both GalCer and Aglycon in water. The

critical aggregation concentrations of Aglycon and GalCer were determined using UV−vis spectral measurements at various

concentrations. The transmission electron microscopy observations of the aqueous sample solutions indicated that the solution of

GalCer includes vesicles, while that of Aglycon comprises giant micelles in the absence of vesicles. The vesicle formation in the

solution of GalCer was also conﬁrmed by Triton X-100-triggered dye-release experiments. To reveal the eﬀects of glycon on the self-

assembly behaviors in detail, we performed the measurements of dynamic light scattering, temperature-dependence of turbidity,

diﬀerential scanning calorimetry, and wide-angle X-ray diﬀraction. The results clarify that the glycon moiety of GalCer has a

signiﬁcant role in the formation inhibition of second associates and the plasticization of the hydrophobe. This work will shed light on

the other natural glycosides to evaluate the self-assembly behaviors for supramolecular and pharmacological applications in the near

future.

INTRODUCTION

immuno-stimulatory activity but also Th2-type cytokines with

immuno-suppressive activity. The release of Th1 cytokines is

related to antitumor and antimicrobial activities, while that of

Th2 cytokines leads to alleviation of autoimmune diseases.

Therefore, GalCer has been long used as a standard reference

for positive control in NKT cell studies. On the other hand, in

order to control the Th1/Th2 balance, organic chemists have

■

α-Galactosyl ceramide (GalCer) is a synthetic glycolipid,

which was ﬁrst developed by the researchers at Kirin Brewery

Co. as a candidate for an anticancer drug through the

structure−activity relationship study using an anticancer

natural product, agelasphin-9b (Figure 1). They found that

GalCer exhibits a lymphocyte proliferation stimulatory activity

created a variety of GalCer derivatives. In spite of the

on both human umbilical cord blood and human peripheral

blood at a concentration above 1.0 μM in vitro. After the

Received: February 24, 2021

Revised: May 27, 2021

Published: June 23, 2021

report, the bioactivity and its mechanism of GalCer have been

−7

considerably investigated to date.

It has been revealed that

GalCer forms a complex with CD1d as a non-polymorphic

glycoprotein on the surface of antigen-presenting cells.

Invariant natural killer (NK) T cells recognize the complex

to produce not only helper-T (Th)1-type cytokines with

2021 American Chemical Society

936

Langmuir 2021, 37, 7936−7944

Article

Figure 1. (a) Structures of GalCer as a synthetic glycolipid and agelasphin-9b as a natural glycolipid and (b) synthesis of Aglycon via Schotten−

Baumann reaction.

extensive pharmaceutical studies of GalCer as a lead

compound, the self-assembly behaviors of GalCer in water

have not been suﬃciently investigated. Amphiphilic GalCer

could hardly exist as a single molecule in aqueous media above

the critical aggregation concentration (CAC), which could

form self-associates such as micelles, vesicles, and nano-

structures of GalCer formed at room temperature. To know

the reason for the distinguishable self-assembly behaviors of

Aglycon and GalCer, we performed dynamic light scattering

(DLS), turbidity measurements at various temperatures,

diﬀerential scanning calorimetry (DSC) analyses, and wide-

angle X-ray diﬀraction (XRD) measurements. The results

reveal the special role of the galactosyl moiety in GalCer

skeleton on the self-assembly and thermal transition behaviors

of GalCer.

0−13

tubes.

These associates in water have potential usefulness

as the nanocarrier of organic molecules. In addition, the

bioactivities of GalCer can be inﬂuenced by the physical

properties of its associates in water because the complexation

of GalCer with a host molecule in a living body should include

not only enthalpy gain on the complexation but also enthalpy

loss on the dissociation of self-associates comprising GalCer.

The morphology and curvature of associates are also expected

to control the biological properties of GalCer as an analogy of

the related studies concerning the impact of membrane shape

EXPERIMENTAL SECTION

■

Materials. Dehydrated dichloromethane (Kanto Chemical Co.,

Inc., Tokyo, Japan), cerotic acid (Tokyo Chemical Industry Co., Ltd.,

Tokyo, Japan), oxalyl chloride (Tokyo Chemical Industry Co., Ltd.,

Tokyo, Japan), phytosphingosine (Tokyo Chemical Industry Co.,

Ltd., Tokyo, Japan), potassium carbonate (Fujiﬁlm Wako Pure

Chemical Corporation, Osaka, Japan), GalCer (Tokyo Chemical

Industry Co., Ltd., Tokyo, Japan), 5-carboxyﬂuorescein (5CF, Tokyo

Chemical Industry Co., Ltd., Tokyo, Japan), polyethylene glycol

mono 4-octylphenyl ether (Triton X-100, critical micelle concen-

tration (CMC): 0.2−0.9 mM at 20−25 °C, Tokyo Chemical Industry

Co., Ltd., Tokyo, Japan), and pyrene (Tokyo Chemical Industry Co.,

Ltd., Tokyo, Japan) were used as received. The other chemicals were

used without further puriﬁcation.

14−18

on molecular recognition.

The study on the self-assembly

behaviors of GalCer would ﬁll the gap of knowledge in the

synthetic, supramolecular, and pharmacological work. In this

context, we have recently reported the thermo-responsivity of

micelles comprising quercetin glycosides as a natural ﬂavonoid

9−21

glycoside,

in which we found that giant ribbon-like

micelles comprising quercetin polyglycoside thermally trans-

form to small-sized spherical micelles.

Measurements. H NMR (400 MHz) spectrum was recorded on

a JEOL ECZ400R spectrometer (JEOL Co. Ltd., Tokyo, Japan) using

In this work, we synthesized the aglycon fragment

Aglycon) of GalCer and investigated the self-assembly and

CDCl as the solvent, calibrated using a residual undeuterated solvent

(

signal. Fourier transform infrared spectrum via an attenuated total

reﬂection (ATR) method was measured using an IRSpirit

spectrometer (Shimadzu Co. Ltd., Kyoto, Japan). Matrix-assisted

laser desorption ionization time-of-ﬂight mass spectrometry (MALDI-

TOF MS) spectrum was recorded on a Shimadzu AXIMA-CFR plus

mass spectrometer (matrix: CHCα). UV−vis spectra were recorded

on a JASCO V-550 spectrophotometer (JASCO Co. Ltd., Tokyo,

Japan) using a 1.0 cm path quartz cell with a temperature controller

thermal transition behaviors of both GalCer and Aglycon. The

UV−vis spectra of Aglycon and GalCer aqueous solutions at

various concentrations were measured to determine the CAC

values. The morphologies of the associates were observed by

transmission electron microscopy (TEM). The TEM images

indicated that GalCer forms vesicles, while Aglycon forms

giant cylindrical micelles in the absence of vesicles. The dye-

release and encapsulation experiments using the solution of

GalCer were performed in order to evaluate the associate

(

ETC-505T, JASCO Co. Ltd., Tokyo, Japan). Fluorescent spectra

were recorded on an RF-6000 spectro ﬂuorophotometer using a 1.0

cm path quartz cell at room temperature (Shimadzu Co. Ltd., Kyoto,

937

Langmuir 2021, 37, 7936−7944

Article

Figure 2. Turbidity of (a) Aglycon and (b) GalCer in water at 25 °C as a function of the logarithm of concentration (μM).

Japan). DSC was performed using a DSC-60 plus instrument

Shimadzu Co. Ltd., Kyoto, Japan) under a N atmosphere (ﬂow

rate: 50 mL/min). XRD proﬁles were recorded on a benchtop powder

XRD instrument (MiniFlex600; Rigaku Co. Ltd., Tokyo, Japan)

equipped with a BTS500 thermo-controller (Anton Paar GmbH,

Graz, Austria).

For the determination of CAC, stock solutions of the samples were

prepared by dissolving in water. The aqueous solutions of various

concentrations were prepared using the stock solution by proper

dilution. The UV−vis spectra of the sample solutions in water were

collected at various concentrations (Figure S4 ). The obtained

absorbance values at 354 nm were plotted as a function of the

logarithm of concentration. From the intersection point in the plots,

we determined the CAC values in water.

The morphologies of sample associates in water with the

concentrations above the CAC were observed by cryo-TEM using a

JEM-3100FEF instrument (JEOL Co. Ltd., Tokyo, Japan) with a

carbon ﬁlm-coated TEM grid (200 mesh, JEOL Co. Ltd., Tokyo,

Japan). Aglycon (0.96 mg) was dissolved in water (100 mL) to aﬀord

the sample solution (13.8 μM). The sample solution of GalCer (50

μM) was also prepared by dissolution of GalCer (0.87 mg) in water

intensity and UV−vis absorption of the mixture were monitored for

an additional few hours. The obtained ﬂuorescence values at 512 nm

and the absorbance values at 492 nm were plotted as a function of

time.

(

The pyrene-encapsulation experiments were performed using a

stock solution of pyrene (300 mg/L) in methanol. The pyrene stock

solution (1.0 μL) was added into water (1.0 mL) or GalCer-

containing aqueous solution (50.0 μM, 1.0 mL). The resulting

mixture was shaken well to completely dissolve/disperse pyrene,

which was used for UV−vis spectral measurements (Figure S7 ).

The hydrodynamic diameters of the associates of the sample

solutions were determined by the DLS method using a Zetasizer nano

ZSP instrument (Malvern Co. Ltd., Worcestershire, UK). The time-

dependent correlation function of the scattered light intensity was

measured at a scattering angle of 173°. The size distribution data were

evaluated using the software provided with the instrument. The

measurements at various temperatures were conducted by using the

sample solutions with the concentrations above CAC (1.0 μM for

Aglycon and 30 μM for GalCer). At ﬁrst, the DLS proﬁles were

collected at 25 °C. Then, the solutions were heated at 70 °C. The

further proﬁles at various temperatures were collected upon cooling

from 70 to 0 °C. The time interval for each temperature was 5 min.

The concentration dependency on the DLS proﬁles of GalCer

solutions was investigated by using sample solutions with diﬀerent

concentrations (10, 20, 30, 50, 70, and 100 μM).

(20.0 mL). The obtained sample solution (1.0 μL) was applied on the

TEM grid. The remaining solution on the grid was removed by

contacting with a ﬁlter paper. The obtained grid was then immersed

into liquid ethane by using EM-CPC (Leica Microsystems Co. Ltd.,

Tokyo, Japan), which was immediately used for the cryo-TEM

measurements. The micelle morphology of Aglycon was further

observed by TEM in the presence of phosphotungstic acid as a

negative stain. The pH of 2.0 wt % phosphotungstic acid aqueous

solution was adjusted to be 7.0 by the addition of a small amount of

The UV−vis spectra of the sample aqueous solutions above the

CAC (1.0 μM for Aglycon and 30 μM for GalCer) were collected

with 5 °C temperature intervals upon cooling from 70 to 10 °C. The

waiting time for each temperature was 10 min. The obtained

absorbance values at 340 nm were plotted as a function of

temperature.

.0 M NaOH aqueous solution to the stain solution. Aglycon (1.23

mg) was dissolved in water (25.0 mL) to aﬀord the sample solution

70.0 μM). The sample solution (2.0 μL) was applied on the TEM

Typical Procedure for the Aglycon. Oxalyl chloride (330 μL,

3.84 mmol) was added to a suspension of cerotic acid (381 mg, 0.960

mmol) in CH Cl (5.0 mL) at room temperature. After stirring for 1

(

grid and blotted for 1 min. The remaining solution on the grid was

removed by contacting with a ﬁlter paper. The negative stain solution

d, additional oxalyl chloride (330 μL, 3.84 mmol) was added to the

mixture. The mixture was further stirred for 1 d and concentrated in

vacuo to give cerotic chloride in a quantitative yield, which was used

for the next reaction without further puriﬁcation.

(2.0 μL) was applied to the sample part on the grid and blotted for 1

min. A ﬁlter paper was used for the removal of the remaining solution

on the grid. The obtained grid was dried at room temperature and

used for TEM measurement.

The dye release experiments were carried out using 5CF as a

ﬂuorescent probe. To an aqueous solution of 5CF (0.27 μM, 4.0 mL)

was added GalCer (0.25 mg, 0.29 μmol) at room temperature. The

resulting associate solution was dialyzed by a centrifugal ﬁlter

To a mixture of phytosphingosine (100 mg, 0.315 mmol) in

CH Cl (1.5 mL) was added a solution of K CO (91.2 mg, 0.660

mmol) in H O (1.0 mL) and the mixture of cerotic chloride (138 mg,

0.332 mmol) in CH Cl (1.5 mL) at room temperature. The reaction

mixture was stirred for 7.5 h at the same temperature and

(

Amicon Ultra-4, ULTRAcel-10K, Merck Millipore Ltd.) for the

concentrated in vacuo. The crude material was puriﬁed by silica gel

removal of free 5CF in the medium. The residual part was diluted to

column chromatography (eluent: CHCl /MeOH = 10:1 → 5:1) to

.0 mL with water and used for ﬂuorescent and UV−vis spectral

give Aglycon (30 mg, 0.043 mmol) as a white solid in 14% yield

measurements (Figure S6 ). The time-course of ﬂuorescence intensity

was monitored on an RF-6000 spectro ﬂuorophotometer by exciting

the solution at 492 nm and measuring the ﬂuorescence intensity at

(Figures S1 −S3); mp 78 °C; H NMR (400 MHz, CDCl , 298 K): δ

6.31 (d, J = 7.2 Hz, 1H, NH), 4.1−3.5 (m, 5H), 2.4−2.2 (m, 2H),

1.8−1.0 (m, 70H), 0.86 (t, J = 6.8 Hz, 6H) ppm; IR (ATR)ν: 3315

−

12 nm. The UV−vis spectra of the mixture were also collected after

(NH), 3248 (OH), and 1641 (CO) cm ; MALDI-TOF MS

each ﬂuorescence measurement. After 44 h, Triton X-100 (0.46 μL)

was added to the 5CF-trapped associate solution. The ﬂuorescence

(matrix: CHCα): calcd for C H NNaO [M + Na ], 718.7; found,

718.3.

938

Langmuir 2021, 37, 7936−7944

Article

Figure 3. Cryo-TEM observations of (a) giant cylindrical micelles comprising Aglycon and (b) small unilamellar vesicles comprising GalCer. The

TEM specimens were prepared by using the sample aqueous solutions with high concentrations above CAC (13.8 μM for Aglycon and 50 μM for

GalCer). The scale bars indicate 100 nm.

RESULTS AND DISCUSSION

water with a high concentration above CAC. The sample

solution was applied on a carbon ﬁlm-coated TEM grid at

room temperature and blotted for a few minutes. The

remaining solution on the grid was removed by contacting

with a ﬁlter paper. The obtained grid was then immersed into

super-cooled liquid ethane, which was immediately used for

the cryo-TEM measurements. The TEM images show that

Aglycon aqueous solution forms giant associates (Figures 3a

and S8a). To obtain a clearer image, we also performed TEM

measurements of Aglycon solution in the presence of

phosphotungstic acid as a negative stain. The TEM image

suggests that Aglycon forms giant cylindrical micelles in water

■

GalCer constitutes D-galactose as a glycon and phytosphingo-

sine-based ceramide as an aglycon (Figure 1), which is

commercially available. On the other hand, we synthesized

Aglycon via Schotten−Baumann reaction (Figure 1b). Cerotic

acid was treated with an excess amount of oxalyl chloride at

room temperature. The reaction mixture was concentrated in

vacuo to aﬀord cerotic chloride. The obtained cerotic chloride

was mixed with phytosphingosine in CH Cl in the presence of

K CO aqueous solution, which chemo-selectively proceeded

the amide formation in a quantitative conversion. After careful

puriﬁcation using silica gel column chromatography, we

successfully obtained Aglycon. The product structure was

(

Figure S8b). On the other hand, in the TEM images of

GalCer, we observed small unilamellar vesicles as the spherical

associates with a highlighted rim (Figure 3b). Because we did

not observe giant micelles in any TEM images of GalCer, the

bulky and hydrophilic galactosyl moiety of GalCer may

kinetically suppress further aggregation of the associates.

The vesicle formation of GalCer was evidenced by the

con ﬁ rmed by H NMR, IR, and mass spectrometry (Figures

S1 −S3). In the IR spectrum (Figure S2), the signal at around

−

640 cm strongly indicated the formation of amide linkage

between cerotic acid and phytosphingosine skeletons. The

spectroscopic data of Aglycon was found to be in good

2,23

agreement with the reported values.

entrapment experiments of a water-soluble anionic ﬂuorescent

Having Aglycon and GalCer in hand, we investigated the

self-assembly behaviors by the turbidity change of sample

aqueous solutions upon the change of concentration for the

CAC determination, that is, onset concentration to form

30−32

probe, 5CF, according to the literature.

To an aqueous

solution of 5CF was added GalCer above the CAC at room

temperature. The resulting associate solution was dialyzed by

ultraﬁltration for the removal of free 5CF in the medium. The

residual part was diluted with water and used for ﬂuorescent

and UV−vis spectral measurements. The time-course of

ﬂuorescence intensity was monitored by exciting the solution

at 492 nm and measuring the ﬂuorescence intensity at 512 nm

24−27

associates in water.

Aglycon and GalCer at 25 °C at various concentrations

Figure S4 ). Since the molecules have no visible absorption,

We measured the UV−vis spectra of

the concentration-dependent absorption change comes from

the scattering of incident light by the generated associates. The

absorbance values at 354 nm were plotted against the

logarithm of sample concentration (Figure 2).

(

Figure S6a ). The UV−vis spectra of the mixture were also

collected after each ﬂuorescence measurement (Figure S6b ).

After 44 h, Triton X-100 above the CMC was added to the

The intersection point on the plots of the absorbance against

the logarithm of sample concentration suggests the CAC

values. The appearance of clear intersection points in the plots

means that Aglycon and GalCer form associates in water

above the deﬁned concentration at 25 °C. The CAC values

were estimated to be 0.44 μM for Aglycon and 8.5 μM for

GalCer. It was found that the CAC value of GalCer is higher

than that of Aglycon. Because it has been reported that several

ceramides exhibit very low CAC in nanomolar order due to

CF-trapped associate solution. The ﬂuorescence values at 512

nm and the absorbance values at 492 nm were plotted as a

function of time (Figure S6d,e ). On the basis of ﬂ uorescent

intensity of the ﬁltrate after the ultra ﬁ ltration (Figure S6c ), we

created the release proﬁle of 5CF (Figure S6f). In Figure S6,f,

the remarkable change in the ﬂuorescence intensity was

observed after the addition of Triton X-100. The plots clearly

indicate that Triton X-100 triggers a quick release of 5CF as a

result of the disruption of vesicles comprising GalCer. The

tendency of the plots is in good accordance with those of the

previous reports concerning Triton X-100-triggered disruption

8,29

hydrophobicity,

the higher CAC value of GalCer could be

attributed to the presence of the hydrophilic galactosyl

skeleton.

30−32

of vesicles.

We observed the hyperchromic eﬀect based on

The morphologies of associates were observed by cryo-TEM

measurements with a carbon ﬁlm-coated TEM grid. For the

TEM sample preparations, we used the sample solutions in

the addition of Triton X-100 in the UV−vis spectra (Figure

S6b,d ). The hyperchromism of 5CF suggests the movement of

5CF in the associates comprising GalCer to Triton X-100-

939

Langmuir 2021, 37, 7936−7944

Article

Figure 4. DLS proﬁles of Aglycon (1.0 μM) and GalCer (30 μM) in water above CAC at various temperatures. The DLS proﬁles of freshly

prepared samples were ﬁrst collected at 25 °C (a,f). Then, the sample solutions were heated to 70 °C. Further proﬁles at various temperatures were

collected at 70 (b,g), 50 (c,h), 25 (d,i), and 10 °C (e,j) upon cooling.

based micelles with a more hydrophobic environment,

implying that the GalCer-based associates would have a

hydrophilic inner space, which should be vesicles.

suggest that pyrene moved into a hydrophobic environment

such as the inner space of micelles, supporting the presence of

micelles comprising GalCer in the solution. The encapsula-

tions of both hydrophilic 5CF and hydrophobic pyrene in the

associates comprising GalCer indicate the coexistence of

vesicles and micelles in the aqueous solution of GalCer at

room temperature.

To know the cooling eﬀects of the associate structures in

detail, we performed the DLS measurements of Aglycon and

GalCer in water above CAC at various temperatures (Figure 4,

Table S1). At ﬁrst, the DLS proﬁles of sample aqueous

solutions were collected at 25 °C. Then, the sample solutions

Because we deeply considered the possibility of cooling

eﬀects on the structure of GalCer-based associates during the

TEM observations using super-cooled liquid ethane, we also

performed the encapsulation experiments using pyrene as a

hydrophobic dye into the associates comprising GalCer in

order to conﬁrm the presence of micelles at room temperature

according to the reported method. Surprisingly, it turned out

that the intensity of signals characteristic to pyrene clearly

increases in the presence of GalCer (Figure S7). The results

940

Langmuir 2021, 37, 7936−7944

Article

Figure 5. Absorbance of (a) Aglycon (1.0 μM) and (b) GalCer (30 μM) in water at 340 nm as a function of temperature upon cooling. Lines are

guides for the eyes.

Figure 6. DSC cooling curves for (a) Aglycon and (b) GalCer.

9,20

were heated to 70 °C and further proﬁles at various

temperatures were collected upon cooling from 70 to 10 °C.

The DLS proﬁles of Aglycon at all temperature regions appear

as a unimodal pattern (Figure 4a−d) except for the proﬁle at

of micelles.

In addition, it has also been reported that the

formation of vesicles should lead to an increase in scattering

intensity of incident light and likewise to an increase in

4−37

turbidity.

Figure 5 shows the turbidity of Aglycon and

0 °C in Figure 4e with minor signals. It is indicated that the

GalCer as a function of temperature. We collected the UV−vis

spectra of Aglycon and GalCer with 5 °C temperature

intervals upon cooling from 70 to 10 °C (Figure S5). The

waiting time for each temperature was 10 min. The absorbance

values at 340 nm were plotted against the temperature. In the

plots of Aglycon (Figure 5a), roughly speaking, the turbidity of

the solution linearly decreased upon cooling. Considering the

observations in the TEM images and the DLS proﬁles, the

turbidity change implies the gradual formation of giant micelles

as a thermodynamic product. On the other hand, the plots for

GalCer were found to be not monotonous. The trend of the

plots includes two inﬂection points at around 25 and 45 °C

(Figure 5b). Upon cooling, the solution of GalCer became

more opaque. Interestingly, further cooling led to a decrease in

turbidity. Because we did not observe giant micelles in the

TEM images of GalCer, such results suggest that temperature

can inﬂuence micelle−vesicle transitions. Because it has been

reported that several associates comprising ionic surfactants

show a simple transition as a temperature-induced micelle−

hydrodynamic diameters of associates comprising Aglycon

tend to increase upon cooling probably due to the aggregation

of small micelles to form giant micelles as an enthalpically

stable product upon cooling. On the other hand, the DLS

proﬁles of GalCer appear as bimodal patterns at low-

temperature regions, where the contribution of small associates

on the DLS proﬁle becomes larger upon cooling. The bimodal

pattern may indicate the formations of two kinds of associates

such as vesicles and micelles, although accurate peak

assignments of DLS signals are diﬃcult. Moreover, we

evaluated the dependency of concentration on the DLS

proﬁles of GalCer -based associates. The results are summar-

ized in Figure S9 and Table S2. As the concentration decreases,

the hydrodynamic diameter of GalCer associates becomes

smaller, suggesting that the size of associates is sensitive to the

concentration. Since the size of GalCer-based associates in the

TEM image appears to be smaller than those in the DLS

proﬁles, there may be a possibility that the sample

concentration has decreased during the TEM sample

preparation.

34−37

vesicle transition,

might indicate the coexistence of micelles and vesicles in an

the complicated transition of GalCer

We next planned to investigate the distinguishable thermal

transitions of associates comprising Aglycon and GalCer in

detail. The process of thermal transitions should be observable

using appropriate methods. The simplest way is to analyze the

turbidity of the sample as a function of temperature. We have

previously reported that the temperature-dependence of

turbidity was diagnostic to judge the structural transformation

equilibrium.

The results of DLS and turbidity measurements at various

temperatures imply that the thermal structural transition of

GalCer-based associates would involve complicated processes,

while the transition of Aglycon is from small micelles to giant

micelles. We now have a question about the reason why the

micelles comprising Aglycon do not form the vesicles, despite

941

Langmuir 2021, 37, 7936−7944

Langmuir

Article

the ceramide structure seeming to favorably form the bilayer

associates. Thus, we performed the DSC measurements of

Aglycon and GalCer (Figures 6, S10, and S11). Figure 6

shows the cooling curves of DSC proﬁles. We found that the

DSC curve of Aglycon includes two exothermic signals at 62

and 27 °C upon cooling (T and T in Figure 6a), which

than 0.76. The numerical value of P implies that the GalCer-

based associates would tend to form not a spherical micelle but

a spherical vesicle with a bilayer structure or a cylindrical

micelle, which seems to match with our observations.

CONCLUSIONS

■

should indicate the presence of a liquid crystalline (LC)

In conclusion, this paper describes the eﬀects of the galactose

skeleton on the self-assembly and thermal transition behaviors

of GalCer for the very ﬁrst time, which are discussed through

the comparison with those of synthesized Aglycon. It should

be mentioned that the synthesized Aglycon plays a crucial role

to elucidate the thermo-dependent self-assembly behaviors of

GalCer and Aglycon by CAC, TEM, DLS, temperature-

dependent turbidity, DSC, and XRD measurements. The TEM

images and dye-release experiments of the associates

comprising GalCer clearly support the formation of vesicles

in the solution. The DSC and XRD analyses indicate that the

glycon of GalCer would make the conformation of two

alkylene chains of GalCer temperature-responsive, enabling

the structural transitions. Nonlinear correlations between the

biological activities and the concentration of GalCer have been

6,38,39

phase.

On the other hand, the DSC curve of GalCer

ⁱⁿ^c^l^u^d^e^s^aⁿ^e^x^o^t^h^e^r^mⁱ^c^sⁱ^gⁿ^a^l^a^t⁶⁹^°^C^a^s^a^f^r^e^e^zⁱⁿ^g^p^oⁱⁿ^t⁽^Tf⁾

upon cooling (Figure 6b). In addition, two changes in heat

^c^a^p^a^cⁱ^t^y^w^e^r^e^o^b^s^e^r^v^e^d^a^t⁴²^°^C^aⁿ^d²³^°^C⁽^Tg2 ^aⁿ^d^Tg1 in

Figure 6b), which can be assigned to the glass transition

temperatures (T ) of GalCer. The T s of GalCer at 42 and 23

C match well with the inﬂection points observed in the

temperature-dependency of turbidity in Figure 5b, indicating

the possible correlation between the T and the structural

40−42

transition temperature.

Below each T of GalCer, the

ﬂexible random-coiled conformation of two alkylene chains of

GalCer could translate stepwise to the rigid conformation even

in a solution. Along with the conformational change of alkylene

chains upon cooling, the hydrodynamic volume of the

hydrophobe can be altered to change the hydrophilic−

lipophilic balance of GalCer, leading to the formation of

vesicles.

To conﬁrm the presence of the LC phase of Aglycon, we

performed XRD measurements of Aglycon at various temper-

atures (Figure S12 ). At ﬁrst, we measured the XRD proﬁle of

Aglycon at room temperature (Figure S12a ), which shows the

plural signals attributed to the crystalline phase. When the

sample was heated at 120 °C, all signals disappeared, indicating

the melt of Aglycon to become the isotropic liquid phase

reported, where the concentration-dependent biological

activities have an inﬂection point at around the CAC of

GalCer, although the physiological environment could

certainly alter the self-assembly behaviors. Taking the

importance of the biological activity of GalCer into account,

we believe that this work will certainly motivate chemists and

biologists to design novel nanocarriers, evaluate the dynamic

behaviors, and investigate the detailed biological mechanism of

natural glycolipids in future.

The Supporting Information is available free of charge at

Figure S12b ). Upon cooling, we observed the diﬀraction

ASSOCIATED CONTENT

sı Supporting Information

https://pubs.acs.org/doi/10.1021/acs.langmuir.1c00545.

■

signals at 50 °C (Figure S12c ). Although the intensity of the

signals is not so strong, the pattern is clearly diﬀerent from

those in Figure S12a , suggesting the formation of the LC phase

in Aglycon. The results are in good accordance with the

observation in the DSC cooling curve.

The micelles comprising Aglycon should form such LC

phase as a thermally stable phase at the corresponding

temperature region. The LC phase would suppress not only

the dynamic nature of amphiphiles but also the vesicular

formation. The incorporation of bulky galactose skeleton on

the ceramide would disturb the packing of alkylene chains to

prevent the formation of LC phase, which could endow the

special character for thermal structural transition with the

associates comprising GalCer.

¹H NMR, IR, and MALDI-TOF MS spectra of Aglycon,

UV−vis and ﬂuorescent spectra of the sample solutions,

TEM observations, and DSC proﬁles (PDF)

AUTHOR INFORMATION

Corresponding Author

■

Yasuhito Koyama − Department of Pharmaceutical

Engineering, Faculty of Engineering and Biotechnology

Research Center, Toyama Prefectural University, Imizu,

Toyama 939-0398, Japan; orcid.org/0000-0002-6985-

Finally, we considered the correlation of self-assembly

behaviors of GalCer to theoretical packing parameter (P),

which was estimated by the following equation

499 ; Phone: +81-766-56-7500 (ext. 2913);

Email: ykoyama@pu-toyama.ac.jp; Fax: +81-766-56-6182

Authors

P = V /(al )

(1)

Ryo Miyazaki − Department of Pharmaceutical Engineering,

Faculty of Engineering, Toyama Prefectural University, Imizu,

Toyama 939-0398, Japan

Mahmuda Nargis − Department of Pharmaceutical

Engineering, Faculty of Engineering, Toyama Prefectural

University, Imizu, Toyama 939-0398, Japan; orcid.org/

0000-0001-6292-2631

Abu Bin Ihsan − Department of Pharmaceutical Engineering,

Faculty of Engineering, Toyama Prefectural University, Imizu,

Toyama 939-0398, Japan; orcid.org/0000-0001-8369-

5060

where V and l are the volume and the length of the surfactant

tail and a is the surface area of the hydrophobic core per

molecule at the aggregate interface. For common surfactants,

the ratio V /l is a constant independent of tail length, equal to

2 Å for the double tail. Although the a value of GalCer has

not been estimated to date, the value should be higher than the

cross-sectional area of ceramide-free galactose. Thus, we

determined to use the a value of galactose for the estimation

of P of GalCer-based associates as the reference. By using the

reported hydrodynamic radius (R ) of galactose, 4.2 Å, we

calculated the a value of galactose to be 55.4 Å . As a result, we

Noriyuki Nakajima − Department of Pharmaceutical

Engineering, Faculty of Engineering and Biotechnology

estimated P for the associates comprising GalCer to be less

942

Langmuir 2021, 37, 7936−7944

Langmuir

Hatanaka, A.; Vestergaard, M. d. C.; Takagi, M. Size-dependent

Article

Research Center, Toyama Prefectural University, Imizu,

Toyama 939-0398, Japan

Masahiro Hamada − Department of Pharmaceutical

Engineering, Faculty of Engineering and Biotechnology

Research Center, Toyama Prefectural University, Imizu,

Toyama 939-0398, Japan

(15) Hamada, T.; Morita, M.; Miyakawa, M.; Sugimoto, R.;

partitioning of nano/microparticles mediated by membrane lateral

heterogeneity. J. Am. Chem. Soc. 2012, 134, 13990−13996.

(16) Itoh, T.; Erdmann, K. S.; Roux, A.; Habermann, B.; Werner, H.;

De Camilli, P. Dynamin and the actin cytoskeleton cooperatively

regulate plasma membrane invagination by BAR and F-BAR proteins.

Dev. Cell 2005, 9, 791−804.

Complete contact information is available at:

https://pubs.acs.org/10.1021/acs.langmuir.1c00545

(17) Carlton, J.; Bujny, M.; Peter, B. J.; Oorschot, V. M. J.;

Rutherford, A.; Mellor, H.; Klumperman, J.; McMahon, H. T.; Cullen,

P. J. Sorting nexin-1 mediates tubular endosome-to-TGN transport

through coincidence sensing of high-curvature membranes and 3-

phosphoinositides. Curr. Biol. 2004 , 14 , 1791 −1800.

Notes

The authors declare no competing ﬁnancial interest.

(18) Habermann, B. The BAR-domain family of proteins: a case of

ACKNOWLEDGMENTS

■

bending and binding? The membrane bending and GTPase-binding

functions of proteins from the BAR-domain family. EMBO Rep. 2004,

We thank Masahiro Fujihara, Tomoko Oono, and Sakiko

Fujita for TEM measurements at the Nara Institute of Science

and Technology (NAIST) on request, supported by Nano-

technology Platform Program (Molecule and Material Syn-

thesis) of the Ministry of Education, Culture, Sports, Science

and Technology (MEXT), Japan. This work was ﬁnancially

supported by Ogasawara Toshiaki Memorial Foundation, JSPS

KAKENHI (grant number JP17H03070), and the Grant-in-

Aid for JSPS Fellow relating to JSPS Postdoctoral Fellowships

for Foreign Researchers (no. P19038 for A.B.I.) from JSPS.

, 250−255.

(19) Nargis, M.; Ihsan, A. B.; Koyama, Y. Thermoresponsive

structure and dye encapsulation of micelles comprising bolaamphi-

philic quercetin polyglycoside. Langmuir 2020 , 36 , 10764 −10771.

20) Nargis, M.; Ihsan, A. B.; Koyama, Y. Effects of sugar chain

length of quercetin-3-O-glycosides on micellization in aqueous media.

Chem. Lett. 2020, 49, 896−899.

(21) Nargis, M.; Ihsan, A. B.; Koyama, Y. Bolaamphiphilic properties

and pH-dependent micellization of quercetin polyglycoside. RSC Adv.

019, 9, 33674−33677.

22) Cheng, J. M. H.; Chee, S. H.; Dölen, Y.; Verdoes, M.; Timmer,

(

REFERENCES

■

M. S. M.; Stocker, B. L. An efficient synthesis of a 6 ”-BODIPY-α -

galactosylceramide probe for monitoring α -galactosylceramide uptake

by cells. Carbohydr. Res. 2019, 486, 107840.

(23) Yen, Y.-F.; Sawant, R. C.; Luo, S.-Y. Concise syntheses of α -

galactosyl ceramide, D-ribo-phytosphingosine, and ceramide. Synthesis

(

1) Morita, M.; Motoki, K.; Akimoto, K.; Natori, T.; Sakai, T.; Sawa,

E.; Yamaji, K.; Koezuka, Y.; Kobayashi, E.; Fukushima, H. Structure-

activity relationship of α -galactosylceramides against B16-bearing

mice. J. Med. Chem. 1995, 38, 2176−2187.

2) Anderson, B. L.; Teyton, L.; Bendelac, A.; Savage, P. B.

24) Cai, L.; Gochin, M.; Liu, K. A facile surfactant critical micelle

013, 45, 511−517.

concentration determination. Chem. Commun. 2011, 47, 5527−5529.

25) Mukerjee, P. Critical Micelle Concentrations of Aqueous

Stimulation of natural killer T cells by Glycolipids. Molecules 2013, 18,

5662−15688.

3) Wu, D.; Fujio, M.; Wong, C.-H. Glycolipids as immunostimulat-

ing agents. Bioorg. Med. Chem. 2008, 16, 1073 −1083.

4) Savage, P. B.; Teyton, L.; Bendelac, A. Glycolipids for natural

killer T cells. Chem. Soc. Rev. 2006 , 35 , 771 −779.

5) Brutkiewicz, R. R. CD1d ligands: the good, the bad, and the ugly.

J. Immunol. 2006, 177, 769−775.

6) Berkers, C. R.; Ovaa, H. Immunotherapeutic potential for

ceramide-based activators of iNKT cells. Trends Pharm. Sci. 2005, 26,

52−257.

7) Taniguchi, M.; Harada, M.; Kojo, S.; Nakayama, T.; Wakao, H.

Surfactant Systems; National Standard Reference Data System,

National Bureau of Standards, U.S. Department of Commerce, U.S.

Government Printing Oﬃce: Washington, D.C., U.S.A., 1971; pp 1−

26) Chang, C. Y.; Wang, S. J.; Liu, I. J.; Chiu, Y. C. A simple

22.

method for determining the critical micellar concentration of a

surfactant. J. Chin. Chem. Soc. 1987, 34, 243 −246.

(27) Ihsan, A. B.; Nargis, M.; Koyama, Y. Effects of the hydrophilic-

lipophilic balance of alternating peptides on self-assembly and

thermo-responsive behaviors. Int. J. Mol. Sci. 2019, 20, 4604.

The regulatory role of V α 14NKT cells in innate and acquired immune

response. Annu. Rev. Immunol. 2003 , 21 , 483 −513.

8) Tashiro, T.; Mori, K. Fifteen years since the development of

(28) Zheng, W.; Kollmeyer, J.; Symolon, H.; Momin, A.; Munter, E.;

Sullards, M. C.; Cabot, M.; Merrill, A. H., Jr. Ceramide and other

Wang, E.; Kelly, S.; Allegood, J. C.; Liu, Y.; Peng, Q.; Ramaraju, H.;

bioactive sphingolipid backbones in health and disease: lipidomic

analysis, metabolism and roles in membrane structure, dynamics,

signaling and autophagy. Biochim. Biophys. Acta, Biomembr. 2006,

KRN7000 −structure-activity relationship studies on novel glyco-

sphingolipids which stimulate natural killer T cells. Trends Glycosci.

Glycotechnol. 2010, 22, 280−295.

10) Shimizu, T.; Masuda, M.; Minamikawa, H. Supramolecular

9) Franck, R. W.; Tsuji, M. α -C-Galactosylceramides: synthesis and

immunology. Acc. Chem. Res. 2006, 39, 692−701.

nanotube architectures based on amphiphilic molecules. Chem. Rev.

005, 105, 1401−1444.

11) Kitamoto, D.; Morita, T.; Fukuoka, T.; Konishi, M.-a.; Imura,

T. Self-assembling properties of glycolipid biosurfactants and their

potential applications. Curr. Opin. Colloid Interface Sci. 2009, 14, 315−

758, 1864−1884.

29) Goni, F. M.; Alonso, A. Biophysics of sphingolipids I.

Membrane properties of sphingosine, ceramides and other simple

sphingolipids. Biochim. Biophys. Acta, Biomembr. 2006, 1758, 1902−

921.

30) Mondal, J. H.; Ghosh, T.; Ahmed, S.; Das, D. Dual self-sorting

by cucurbit[8]uril to transform a mixed micelle to vesicle. Langmuir

014, 30, 11528−11534.

31) Jeon, Y. J.; Bharadwaj, P. K.; Choi, S.; Lee, J. W.; Kim, K.

328.

Supramolecular amphiphiles: spontaneous formation of vesicles

12) Hato, M.; Minamikawa, H.; Tamada, K.; Baba, T.; Tanabe, Y.

Self-assembly of synthetic glycolipid/water systems. Adv. Colloid

Interface Sci. 1999, 80, 233−270.

triggered by formation of a charge-transfer complex in a host.

Angew. Chem., Int. Ed. 2002, 41 , 4474 −4476.

(32) Majhi, P. R.; Blume, A. Temperature-induced micelle-vesicle

transitions in DMPC-SDS and DMPC-DTAB mixtures studied by

calorimetry and dynamic light scattering. J. Phys. Chem. B 2002, 106,

10753−10763.

13) Corti, M.; Cantu, L.; Brocca, P.; Del Favero, E. Self-assembly in

glycolipids. Curr. Opin. Colloid Interface Sci. 2007 , 12 , 148 −154.

14) Sugiura, Y.; Ikeda, K.; Nakano, M. High membrane curvature

enhances binding, conformational changes, and fibrillation of

amyloid-β on lipid bilayer surfaces. Langmuir 2015, 31, 11549−

1557.

943

Langmuir 2021, 37, 7936−7944

Langmuir

(33) Fu, J.; Cai, Z.; Gong, Y.; O ’Reilly, S. E.; Hao, X.; Zhao, D. A

Article

new technique for determining critical micelle concentrations of

surfactants and oil dispersants via UV absorbance of pyrene. Colloids

Surf., A 2015, 484, 1−8.

34) Yin, H.; Zhou, Z.; Huang, J.; Zheng, R.; Zhang, Y.

Temperature-induced micelle to vesicle transition in the sodium

dodecylsulfate/dodecyltriethylammonium bromide system. Angew.

Chem., Int. Ed. 2003, 42, 2188−2191.

(35) Yin, H.; Huang, J.; Lin, Y.; Zhang, Y.; Qiu, S.; Ye, J. Heating-

induced micelle to vesicle transition in the cationic-anionic surfactant

systems: comprehensive study and understanding. J. Phys. Chem. B

36) Méndez, S.; Martí, M.; Barba, C.; Parra, J. L.; Coderch, L.

005, 109, 4104−4110.

Thermotropic behavior of ceramides and their isolation from wool.

Langmuir 2007, 23, 1359−1364.

(37) Liu, X.; Wang, J.; Cui, Z.; Yao, H.; Ge, X.; Chen, W.; Sun, F.

Temperature-induced reversible micelle-vesicle transition in aqueous

solution of a pseudogemini surfactant without any additive. RSC Adv.

(

017, 7, 26440−26445.

38) Park, B. D.; Youm, J. K.; Jeong, S. K.; Choi, E. H.; Ahn, S. K.;

Lee, S. H. The characterization of molecular organization of

multilamellar emulsions containing pseudoceramide and type III

synthetic ceramide. J. Invest. Dermatol. 2003, 121, 794−801.

(39) Rahim, N.; Ahmat, N.; Norrizah, J. S.; Fadhilah Kamalul Aripin,

N.; Hamid, H. A. A.; Saleh, A. A review on phase behaviour of

glycolipids derived from plants. Mater. Today: Proc. 2018, 5, S180−

S185.

(

40) Marsh, D. Thermodynamics of phospholipid self-assembly.

Biophys. J. 2012, 102, 1079−1087.

41) Singh, J.; Ranganathan, R. Mixing of oxidized and bilayer

phospholipids. Biochim. Biophys. Acta 2015, 1848, 1472−1480.

42) Kastantin, M.; Ananthanarayanan, B.; Karmali, P.; Ruoslahti, E.;

Tirrell, M. Effect of the lipid chain melting transition on the stability

of DSPE-PEG (2000) micelles. Langmuir 2009 , 25 , 7279 −7286.

(43) Nagarajan, R. Molecular packing parameter and surfactant self-

assembly: The neglected role of the surfactant tail. Langmuir 2002, 18,

1−38.

44) Schultz, S. G.; Solomon, A. K. Determination of the effective

hydrodynamic radii of small molecules by viscometry. J. Gen. Physiol.

961, 44, 1189−1199.

45) Veerapen, N.; Brigl, M.; Garg, S.; Cerundolo, V.; Cox, L. R.;

Brenner, M. B.; Besra, G. S. Synthesis and biological activity of α -

galactosyl ceramide KRN7000 and galactosyl (α 1 →2) galactosyl

ceramide. Bioorg. Med. Chem. Lett. 2009, 19, 4288−4291.

944