The photopolymerization of DC8,9PC in microbubbles

The photopolymerization of DC8,9PC in microbubbles

Accepted Manuscript Title: The photopolymerization of DC8,9 PC in microbubbles Author: Maarten Callens Marco Beltrami Emiliano D’Agostino Helge Pfeiff...

6MB Sizes 0 Downloads 3 Views

Accepted Manuscript Title: The photopolymerization of DC8,9 PC in microbubbles Author: Maarten Callens Marco Beltrami Emiliano D’Agostino Helge Pfeiffer Dirk Verellen Gaio Paradossi Koen Van Den Abeele PII: DOI: Reference:

S0927-7757(18)31030-6 https://doi.org/doi:10.1016/j.colsurfa.2019.01.038 COLSUA 23136

To appear in:

Colloids and Surfaces A: Physicochem. Eng. Aspects

Received date: Revised date: Accepted date:

8 September 2018 17 November 2018 18 January 2019

Please cite this article as: Maarten Callens, Marco Beltrami, Emiliano D’Agostino, Helge Pfeiffer, Dirk Verellen, Gaio Paradossi, Koen Van Den Abeele, The photopolymerization of DC8,9 PC in microbubbles, (2019), https://doi.org/10.1016/j.colsurfa.2019.01.038 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

The photopolymerization of DC8,9 PC in microbubbles. a,∗

f

b,c

Dirk Verellen

d

f

, Gaio Paradossi , Koen Van Den Abeele

a

Propagation and Signal Processing Group, Department of Physics, KU Leuven

b GZA

KULAK, Kortrijk, Belgium Ziekenhuizen, Iridium Kankernetwerk, Antwerp, Belgium

cr

a Wave

e

, Marco Beltrami , Emiliano D'Agostino , Helge Pfeier ,

ip t

Maarten Callens

c Faculty of Medicine and Pharmacy, Vrije Universiteit Brussel, Brussels, Belgium. d Department of Materials Engineering, KU Leuven, Leuven, Belgium. e DoseVue NV, Hasselt, Belgium.

di Scienze e Tecnologie Chimiche, Università di Roma Tor Vergata, Rome,

us

f Dipartimento

an

Italy

Abstract

M

The polymerization of diacetylene lipids is a valuable mechanism that can be exploited for various applications ranging from sensor materials to triggered drug delivery.

Therefore, mixtures of diacetylene lipid and saturated

d

phospholipids have been extensively studied in liposome vesicles and Langmuir monolayers. The present study investigates the polymerization of 1,2-bis(10,12-

sn -glycero-3-phosphocholine

te

tricosadiynoyl)-

(DC8,9 PC) in a new assembly, i.e.

in a lipid microbubble shell. Lipid microbubbles have a high degree of echogenic-

Ac ce p

ity which allows to probe the properties of the mixed lipid monolayer shell using ultrasound spectroscopy. Microbubbles with varying DC8,9 PC content in a 1,2-

distearoyl-

sn -glycero-3-phosphocholine (DSPC) matrix were characterized using

ultrasound transmission spectroscopy, UV-vis absorption spectroscopy and microscopy imaging. The microbubble populations are characterized before and after UV radiation (254 nm), in order to investigate the eect of introducing DC8,9 PC into the microbubble shell, and whether diacetylenes can polymerize in this conguration. Microbubbles were found to be more stable when either DSPC or DC8,9 PC dominates the composition. UV radiation induced the for-

∗ Corresponding author

Email addresses: [email protected] (Maarten Callens), [email protected] (Koen Van Den Abeele)

Preprint submitted to Colloids and Surfaces A

January 24, 2019

Page 1 of 31

R1

R1

C C

C

C C R2

C R2

R1

C

C

C

C C

R2

R1

R2

C

C

C

R1

cr

d

C

C

C

C

C

us

R2

C

ip t

C C C

R1

C

R2

an

Figure 1: The topotactic polymerization of diacetylenes.

mation of a diacetylene polymer in all DC8,9 PC containing compositions that

M

were investigated, and the polymerization was found to impair the stability of the microbubbles.

Keywords:

Ac ce p

1. Introduction

te

agents, lipid monolayers

d

diacetylene polymerization, UV-radiation, microbubbles, ultrasound contrast

Polydiacetylenes (PDA's) have been extensively studied, partially because

of their remarkable optical properties [1], but also because of the substantial modication of the mechanical properties of diacetylene assemblies imparted by

5

the cross linking of diacetylene units [2]. The polymerization of diacetylenes, which is shown in Fig. 1, is a topochem-

ical reaction which can be initiated by heat, UV, or by ionising radiation [3, 4]. The reaction can proceed when the packing of the diacetylene monomers fulls a set of steric constraints, which have largely been deduced from systematic stud-

10

ies on the reactivity of diacetylene crystals [5]. Besides in crystals, the required packing can also be realized in supramolecular self assembled structures composed of amphiphilic diacetylenes [6, 7]. One of the most studied self assembling

sn -

diacetylene amphiphiles is the diacetylene lipid 1,2-bis(10,12-tricosadiynoyl)-

2

Page 2 of 31

O

O O O H

O P O O

N

ip t

DC8,9PC

O

DSPC

O

O O O H

O P O O

C4F10 gas core

N

cr

O

PEG(40)S O

}

}

O

OH

us

40

Figure 2: The chemical structure of DC8,9 PC, DSPC and PEG40S, and a schematic drawing

an

of a diacetylene based polymerizable microbubble.

glycero-3-phosphocholine, DC8,9 PC (see Figure 2 for the chemical structure of this lipid and other compounds that are used in this study).

M

15



When DC8,9 PC is dispersed in excess water at a temperature over 43.8 C, the geluid transition temperature [8], it forms spherical liposome vesicles,

d

but when the temperature is lowered, the vesicles transform to hollow cylinders

20

te

[9, 10]. To attain quasi-spherical polymerizable vesicles of DC8,9 PC at physiological conditions, one introduces DC8,9 PC in a liposome matrix of long chained phospholipids, such as 1,2-dioleoyl-

sn -glycero-3-phosphocholine

(DOPC), 1,2-

sn -glycero-3-phosphocholine (DMPC), 1,2-dipalmitoyl-sn -glycero-3phosphocholine (DPPC), or 1,2-distearoyl-sn -glycero-3-phosphocholine (DSPC)

Ac ce p

dimiristoyl-

[11], alternatively, also formulations of PEGylated lipid and DC8,9 PC where

25

DC8,9 PC dominates the composition results in the formation of nano-sized vesicles [12].

A DC8,9 PCmolecule has two diacetylenic hydrocarbon tails can bind with

up to four neighbouring DC8,9 PC molecules and form an extensive cross linked

polymer network.

30

The formulation of mixed diacetylene/lipid assemblies has

been strongly driven by the physical stabilization of the structure, the increased thermal stability and mechanical strength that is imparted by cross linking [13, 14]. For instance, Moragaki

et al [15] showed that supported bilayer mem-

branes of DC8,9 PC showed high stability after UV exposure, as the polymerized

3

Page 3 of 31

domains became resistant against solubilization and were mechanically robust. 35

Furthermore, Alonso-Romanowski

et al [16] reported that partially polymerized

ip t

liposomes composed of DC8,9 PC and the nonpolymerizable lipid DMPC are more stable against leakage at physiologic conditions than conventional (unpolymer-

cr

ized) liposomes.

On the other hand, the exact opposite of a stabilizing eect has also been observed. For example the UV (254 nm) exposure of DC8,9 PC:DPPC liposomes

us

40

triggered the release of entrapped contents by creating pores in the membrane [11]. The phototriggered pore creation has also been demonstrated in unsup-

an

ported articial DC8,9 PC:DPPC bilayers [17] and provided the basis for phototriggered drug delivery systems [18, 19]. 45

A number of systematic studies of the photopolymerization of DC8,9 PC in a

M

matrix of long chained phospholipids have attempted to determine under which conditions a stabilizing eect can be expected, and when destabilization occurs. These studies provided guidelines for the optimal design of polymerizable assem8,9

PC is polymerizable when

the temperature is below the transition temperature, in this conguration the

te

50

et al [20] showed that DC

d

blies. For instance, Lopez

hydrocarbon tails full the necessary steric requirements. Baek

et al [21] system-

Ac ce p

atically studied the polymerization of DC8,9 PC in matrices of an unsaturated lipid DOPC, and of saturated phospholipids with dierent melting temperature







DMPC (24 ), DPPC (41 ) and DSPC (55 ). They monitored the leakage of a

55

dye from the liposomal vesicles (to detect pore formation), and the lateral surface pressure of Langmuir monolayers at dierent compression states, before and after UV treatment. From the study it was concluded that the polymerization only occurs when the hosting matrix has a liquid condensed (LC) phase in the monolayer, or a gel phase in the bilayer liposome, hence only in closely packed

60

DMPC, DPPC and DSPC matrices. The polymerization process could not be observed in uid matrices, for instance in the liquid expanded (LE) phase of DOPC monolayers or in DOPC liposomes. The eect of the polymerization in these systems consists in the formation of pores in liposome membranes, and a decrease in the lateral surface pressure of Langmuir lms as a result of the area

4

Page 4 of 31

65

contraction of the lm when DC8,9 PC cross links to neighbouring molecules and forms a covalent network [21].

ip t

In this work, we investigate the polymerization of the diacetylene lipid DC8,9 PC in an alternative assembly to liposomal vesicles and Langmuir mono-

70

Microbubbles are micrometer sized gas

cr

layers, namely microbubble systems.

cores, typically a heavy gas with a low solubility in water, stabilized by a lipid,

us

polymer, or protein shell. The most notable application of microbubbles is their use as ultrasound contrast agents (UCA's) in medical ultrasound imaging [22]. UCA's are ecient ultrasound scatterers and appear as bright spots in a sono-

75

an

graphic image due to the high compressibility and low density of the gas core. Moreover, microbubbles oscillate in an ultrasound eld and produce a characteristic echo which can be selectively picked up by the ultrasound scanner

M

[23]. When intravenously injected, UCA's reside in the bloodstream for several minutes and aid the visualization of the vasculature and organ perfusion. The applications of UCA's are not limited to diagnostic imaging. Microbubbles also have an immense potential as targeted vehicles for ultrasound-triggered gene

te

and drug delivery [24].

d

80

Aside from their conventional use as UCA's, lipid-shelled microbubbles also

Ac ce p

serve as a platform to study the eects of the polymerization of DC8,9 PC in a phospholipid matrix because of the unique possibility to ultrasonically interro-

85

gate the mechanical properties of the shell.

For example, the stability of the

microbubbles can be evaluated from the evolution of the attenuation coecient of the bubbly medium in time. Moreover, the elasticity and the viscosity of the shell can be deduced from the frequency dispersion of the acoustic response [25]. We will exploit these unique interrogation methods to investigate the eects

90

of UV-radiation (254nm) on DC8,9 PC:DSPC monolayers at the microbubble surface. Based on the work by Baek

et al [21],

we expect that DC8,9 PC, when

residing in the LC phase of DSPC, can be polymerized in this arrangement. The hypothesized architecture of the diacetylene-based polymerizable microbubble is shown in Figure 2. The formation of a polymer network of covalently 95

bound molecules at the surface of the bubble is expected to alter the mechanical

5

Page 5 of 31

properties of the microbubble as a whole. In this work we investigated (1) if microbubbles containing DC8,9 PC can be

ip t

formed, (2) if DC8,9 PC can be polymerized with UV light in this system, and (3) what the eect of the photopolymerization on the stability of microbubbles is.

To this end, microbubbles were prepared from DC8,9 PC:DSPC:PEG40-stearate

cr

100

lipid solutions at several molar fractions of DC8,9 PC. The presence of polymers

us

after UV-exposure was probed using UV-vis spectroscopy, and the stability of UV-exposed and unexposed microbubbles was analysed using ultrasound transmission spectroscopy and bright eld microscopy imaging. This paper is structured as follows:

section 2 describes the microbubble

an

105

preparation, the characterization methods, and the measurement protocol that was used to assess the stability of UV-exposed and unexposed microbubbles.

M

Under section 3, rst, the results from each experimental technique are reported. Next, the inuence of the DC8,9 PC content on the stability is discussed, and nally we elaborate on the eects of UV exposure.

d

110

te

2. Experimental section

2.1. Microbubble preparation

Ac ce p

Microbubbles were prepared using the probe sonication technique as de-

scribed in the work by Kim

115

et al

[26].

The base composition of the micro-

bubble shell comprises three compounds: DC8,9 PC, DSPC, and polyoxyethylene(40)stearate (PEG40S). DC8,9 PC and DSPC were purchased from Avanti

Polar Lipids Inc. (Alabaster, AL) in powder form, PEG40S was purchased from Sigma-Aldrich (St Louis, MO). Microbubbles with a molar fraction of DC8,9 PC equal to 0, 10, 20, 30, 50, 85 and 95 mol% were studied.

120

The fraction of

PEG40S was xed at 5 mol%, and the remaining fraction is DSPC. The lipids were weighed and mixed in 36 ml phosphate buered saline solution (PBS, pH 7.4) to a total concentration of 1.33 mM. The lipid solution was stirred and



heated on a hot plate with the temperature set at 75 C for 20 min.

6

Page 6 of 31

Next, the lipid solution was sonicated using a Branson 450 Analog Sonier 125

(Branson Ultrasonics Corporation, Danbury, CT) at low power (setting 3 on a

ip t

scale of 1-10) and duty cycle 100% for 30 seconds, with the tip of the sonier disruptor horn (with a diameter of 9.5 mm) immersed in the liquid to a depth

The power delivered to the sample is 21.0 W. Sub-

cr

of approximately 1 cm.

sequently, decauorobutane lled microbubbles were generated by high power sonication at the gas-water interface. Prior to the second sonication, decau-

us

130

orobutane (F2 Chemicals Ltd, Preston, UK) gas was gently streamed into the liquid at a ow rate of approximately 100 ml/min for 30 seconds.

The lipid

an

solution was sonicated with the sonier tip right underneath the gas-liquid interface for 10 seconds with the C4 F10 gas still owing, the power was set at 135

10 and the duty cycle was 100%, the power delivered to the sample is 30.9 W.

M

After sonication, the solution was placed in an ice bath for 4 min. Subsequently, the cooled solution was collected in capped syringes and centrifuged at

300×g

for 4 minutes to separate the buoyant microbubbles from the free lipids and

aside. The infranatant was recuperated and underwent a second cycle of low

te

140

d

aggregates. The microbubble cake that accumulated near the plunger was kept

power sonication, high power sonication, cooling and centrifugation, to gener-

Ac ce p

ate more microbubbles. A number of production cycles were performed and the microbubble cake was collected for each of them. All microbubble cakes were combined to undergo a number of washing cycles.

145

A washing cycle consists

of diluting the microbuble cake with fresh PBS, centrifugation at

300×g

for 4

minutes, extracting the microbubble cake and discarding the infranatant containing free lipids and aggregates. This washing process was repeated until the infranatant is no longer turbid. The nal microbubble cake was diluted with three times its volume in PBS:glycerol (8:2 vol%). The solution was stored in a

150

capped syringe at

4 − 6◦ C.

2.2. Experiment outline A protocol has been established to interrogate the stability of microbubbles with dierent shell compositions, and assess the inuence of UV radiation. The

7

Page 7 of 31

Ultrasound spectroscopy Microscopy imaging UV- vis spectroscopy

ip t

UV exposure

cr

Exposed sample

0

30

60

90 120 Time [min]

150

us

Unexposed reference 180

an

Figure 3: Experiment timeline for investigating the stability of microbubbles as a function of time for the unexposed reference, and as a function of UV dose and time for the UVexposed sample. The experiment starts with splitting a microbubble sample into two identical

M

populations. This is represented by the separation of the black line at time = 0 min. The two populations are referred to as the Exposed sample (red line) and the Unexposed reference (yellow line). On each population, a number of measurements are carried out. The type of measurement is denoted by the black markers (squares, circles and diamonds), which refer to

d

the suite of probe measurements indicated in the legend. The start time of each measurement can be read on the horizontal time axis. The exposed sample is exposed to UV radiation in

te

four fractions (grey rectangles), resulting in a cumulative dose of 23.9 J. Note that a probe measurement is never performed on the population as a whole, but only a small portion of the population is transferred to the respective experimental set-up, this is indicated by the

Ac ce p

curved arrows. In contrast, the UV irradiation is done for the entire population of the exposed sample.

experiment outline is presented in Figure 3.

155

The next paragraphs describe for each technique at what time a measurement

is taken, and what can be deduced from it.

Microscopy imaging.

Microscopy imaging is performed before the populations

are separated (one will be exposed to UV, the other will not) and at

(for the exposed sample), and 160

t = 170

min (for the unexposed reference), after

the separation. Comparison of the images of the unexposed reference at min to images taken at time microbubble population.

< 0

t = 160

t = 170

min allows to assess the evolution of the

On the other hand, images of the exposed sample

8

Page 8 of 31

taken at

t = 160

min can be compared to images taken at time

< 0

min to

observe changes that are a combined eect of the UV exposure and the inherent evolution.

Ultrasound transmission measurements.

ip t

165

Ultrasound transmission measurements

cr

of the exposed sample at the start of the experiment, and after every UV expo-

sure, allows to assess the change of the acoustic properties of the microbubbles

170

us

with increasing UV dose. The unexposed reference is also ultrasonically monitored at regular intervals to survey the evolution and the stability of the micro-

an

bubble population in time.

UV-vis spectroscopy.

UV-vis spectroscopy of the exposed sample and the un-

exposed reference at

t = 180

min and

t = 190

min, respectively, allows to

175

M

monitor radiation-induced changes in the optical absorption properties of the microbubble suspension.

All experiments were performed on microbubble samples that were less than To compare the stability between dierent samples and between

d

a week old.

te

dierent compositions the concentration of the samples must be the same, since this parameter inuences the dissolution dynamics.

experiment as outlined in 3, the concentration of microbubbles was probed by

Ac ce p

180

Prior to performing the

measuring the maximum attenuation. All samples have been diluted such that initial maximum attenuation is 12 dB/cm. The concentration of microbubbles at the start of the experiment is estimated at

1.7 · 108

bubbles/mL based on the

model of the acoustic properties of microbubbles by Stride and Saari [27].

185

2.3. UV irradiation

Microbubble suspensions are exposed to 254 nm UV light from a laboratory

UV lamp (Model ENF-260C, Spectroline Corporation, Westbury, NY) at room temperature.

The distance between the protective window of the lamp and

the surface of the microbubble solution is 5 cm. The UV light ux density at 190

this distance is 1.20

±

2

0.06 mW/cm , as measured with an AccuMAX XRP-

3000 digital meter and a UV-C sensor (Spectroline Corporation). 0.7 ml of the

9

Page 9 of 31

8

microbubble solution (1.7·10

± 0.01 cm.

and cover an area of 4.15 is 4.98

±

± 0.04 cm2 .

The radiant power received by the surface

0.25 mW. The suspension receives a total UV dose of 23.9

split over four fraction of 20 minutes (6.0

±

0.3 J per fraction).

±

0.6 J,

cr

195

The microbubbles oat to the top of the suspension

ip t

diameter of 2.30

bubbles/mL) is poured into a 10 ml beaker with a

2.4. Microscopy imaging and image processing

us

Bright eld microscopy images of microbubbles are acquired with a Zeiss Primo Star microscope (Zeiss, Oberkochen, Germany) with an oil immersion

×100

magnication lens. The size distribution and morphology of the micro-

an

200

bubbles is deduced from a series of 15 to 20 images acquired at random positions on the microscopy slide. An algorithm for the automated detection and sizing

M

of microbubbles was developed using the NI Vision Development Module and NI LabVIEW software (National Instruments Corporation, Austin, TX). The algorithms detects circles in the image and keeps track of their radius and position.

d

205

te

2.5. Ultrasound transmission

The acoustic response of the microbubble medium was measured using a

Ac ce p

dedicated ultrasound through-transmission set-up shown in Figure 4a.

210

Two identical at immersion transducers (Olympus V309-SU; Olympus Cor-

poration, Waltham, MA), with a nominal diameter of 13 mm and a center frequency of 5 MHz, are aligned, facing each other.

The distance between the

wear plates of the transducers is 1.8 cm. The signal generation and acquisition is performed with the FGEN (Function Generator) and the MSO (Mixed Signal

215

Oscilloscope) channels of an NI VirtualBench device (NI VB-8012), respectively, controlled by NI LabVIEW. The microbubble suspension ows through a customized water tight polystyrene ow cell at a ow rate of 32 ml/min. The total volume circulating in the tubing and the ow cell is 5.2 ml.

Microbubbles are introduced in the ow through

10

Page 10 of 31

Oscilloscope Pump microbubble inlet

Transducer

Transducer

Water tank

Unexposed reference

25

25

5 4 3 2 1 0 20

30

40

10 5 0

20

50

60

4

2

0

6

8

10

Frequency [MHz]

Exposed sample

0.0 J / 20 min 6.0 J / 50 min 12.0 J / 80 min 17.9 J / 110 min 23.9 J / 140 min

an

15

A!enuation [dB/ cm]

30 min 60 min 90 min 120 min 150 min

20

15 10

5

M

A!enuation [dB/ cm]

6

Sweep count

c.

0

2

4

6

Frequency [MHz]

8

10

0

2

4

6

8

10

Frequency [MHz]

Panel (a) shows a schematic drawing of the experimental set-up for ultrasound

d

Figure 4:

7

0 10

Flow cell

0

8

cr

A!enuation [dB/ cm]

Function generator

ip t

b.

PC

us

a.

te

transmission spectroscopy measurements of a microbubble medium in ow. Panel (b) shows an example of a measurement of the attenuation as a function of frequency and of the sweep count (a single sweep takes approximately 5 seconds).

Initially, for sweep counts 1-14, the

Ac ce p

attenuation is zero for all frequencies, corresponding to the situation where there is only water in the ow cell. The microbubble are injected into the ow at sweep count 15. At this point, the attenuation promptly increases. The dispersion of the microbubble attenuation shows a broad peak with a maximum around 2 MHz and does not vary during the time of the measurement. The series of ultrasound sweeps is stopped when 5 minutes have passed after the injection of the microbubbles. In the experiment as outlined in Figure 3, ten such attenuation measurements are performed, ve on the exposed sample, and ve on the unexposed reference. In panel (c), the average attenuation dispersion of the unexposed reference (left) and the exposed sample (right) are shown.

For the unexposed reference, the ve attenuation spectra correspond to

the ultrasound scans denoted by the black dots on the yellow line in Figure 3. The legend denotes the time of the measurement. Similarly, the ve attenuation curves for the exposed sample correspond to the ultrasound scans denoted by the black dots on the red line in Figure 3. The legend denotes the time of measurement and the total UV dose that the sample has received at that time.

11

Page 11 of 31

220

the microbubble inlet, using a high precision analytical syringe (SGE eVol XR digital analytical syringe, Trajan Scientic, Australia).

ip t

The microbubble medium is probed by performing discrete frequency sweeps from 0.25 MHz to 12.5 MHz, with a frequency step of 0.25 MHz. A single sweep

225

have passed after the injection of the microbubbles.

µs,

For every frequency, a

and an amplitude of 12 V

us

cosine windowed sine wave with a duration of 5

cr

takes approximately 5 seconds. The measurement is stopped when 5 minutes

is sent from the function generator to the emitting transducer at a repetition rate of 1 kHz. The waveform data are post processed to extract information on

230

particular frequency

f

is calculated as

p¯mb (f ) p¯ref (f )

!

1 d

Vmb (f ) Vref (f )

!

1 d

M

α(f ) = −20 log

d

= −20 log

at a

p¯mb (f )

(1)

is the acoustic pressure amplitude

te

and is expressed in units dB/cm.

α(f )

an

the microbubble medium. In particular, the attenuation coecient

of the wave with frequency

f

that has propagated through the microbubble

Ac ce p

medium. The subscript mb denotes microbubbles. The bar indicates that it concerns the pressure averaged over the receiving transducer surface.

235

pref (f )

is the acoustic pressure amplitude of the signal received when there are no microbubbles in the ow cell. This signal serves as a reference, which is denoted by the subscript ref . The ratio

p¯mb /¯ pref equals Vmb /Vref , where V

amplitude of the received signal. The parameter

d

is the voltage

represents the path length

in the microbubble medium, being equal to 1.0 cm, i.e. the inner dimensions of

240

the ow cell.

Figure 4b shows an example of the measured attenuation as a function of the frequency and as a function of the sweep count.

The rst 14 frequency

sweeps (sweep counts 1 to 14 ) were used to obtain the reference signal

Vref ,

the attenuation for these sweeps is 0 dB/cm, by design. The microbubbles are 245

injected into the ow cell at sweep count 15, causing a prompt increase in the

12

Page 12 of 31

attenuation. The injected volume is 10

µL for the rst attenuation measurement

in the sequence of ve (see black circles in 3), corresponding to a concentration

3.3 · 105

bubbles/mL in the ow cell. For the next four measurements, the

injected volume is gradually increased up to 20

to compensate for the de-

creasing microbubble concentration. A new ow cell is prepared for each micro-

cr

250

µL

ip t

of

bubble injection to avoid cross contamination between dierent microbubble

us

populations.

Figure 4c shows the attenuation spectrum for each of the ve measurements on the unexposed sample (left), and the ve measurements of the UV-exposed sample (right). Each curve corresponds to a black circle in Figure 3. The legend

an

255

labels indicate the time of measurement with respect to the time axis in Figure 3. For the exposed sample, the legend labels also denote the total UV dose that

M

the sample has received.

To quantify the attenuation evolution by a single parameter, we introduce 260

the integrated attenuation measure, dened as the integral over the aver-

to

R fmax fmin

α(f )df ,

in the frequency range from

fmin = 0.25

MHz

te

bubble injection,

d

age attenuation in a one minute interval starting 20 seconds after the micro-

fmax = 9.00 MHz,

and we normalize this value to the injected volume and to

Ac ce p

the integrated attenuation of the rst ultrasound measurement (i.e. obtained 265

on the exposed sample prior to the rst UV exposure).

2.6. UV-vis spectroscopy

The optical absorption properties of the microbubble solutions were probed

with a UV-vis spectrometer (CSS100, Thorlabs Inc., Newton, NJ) in the wavelength range from 350 nm to 700 nm, with a resolution of 0.1 nm, and a signal

270

to noise ratio of

≤ 2000:1.

The light source is a stabilized tungsten-halogen light

source with a wavelength range from 360 nm to 2600 nm (SLS201L, Thorlabs Inc., Newton, NJ). Both the light source and the spectrometer are ber coupled. The ends of the optical bers, coupled to the light source and the spectrometer, are mounted in two aligned ber mounts facing each other. A uorimeter

275

cuvette with a path length of 10 mm is placed in between the ber mounts to

13

Page 13 of 31

acquire the light that is transmitted through a liquid sample. The spectrometer data acquisition was controlled by NI LabVIEW (National Instruments Cor-

ip t

poration, Austin, TX). The optimal integration time of the spectrometer was

280

cr

found to be 1.4 ms. Every spectrum is the average of 40 acquisitions.

3. Results and discussion

us

3.1. Microbubble preparation

An overview of the microbubble formation and photopolymerization of the

an

investigated composition is listed in Table 1. The microbubble solutions with 0, 10, 20, 30 50 and 85 mol% were found to be stable for several months. 285

Microbubbles composed of 95 mol% DC8,9 PC and 5 mol% PEG40S could not These microbubbles were very un-

M

be prepared using the sonication method.

stable and it was not possible to collect enough to perform measurements. All DC8,9 PC-containing formulation were found to polymerize upon UV (254 nm)

d

exposure.

te

Table 1: Concentration dependence of DC8,9 PC towards formation and photopolymerization of DC8,9 PC:DSPC:PEG40S microbubbles. The mol% of PEG40S is xed at 5%, the DSPC

Ac ce p

fraction is the remainder.

290

DC8,9 PC mol%

Microbubble

UV (254 nm)-triggered

formation

polymerization

0

yes

no

10

yes

yes

20

yes

yes

30

yes

yes

50

yes

yes

85

yes

yes

95

no

-

3.2. Microscopy imaging Microscopy imaging was performed to investigate the size distribution and the morphology of the microbubbles.

14

Page 14 of 31

20 mol%

30 mol%

ip t

10 mol%

us

cr

0 mol%

85 mol%

M

an

50 mol%

Figure 5: Size distributions of the unexposed reference samples at the start of the experiment (t

=0

min) and at the end of the experiment (t

= 170

min) for each formulation. The molar

fraction of DC8,9 PC is indicated above each pair of histograms. The legend shows the total

The size distribution of unexposed reference microbubbles

te

Size distribution.

N.

d

number of microbubbles counted

at the start and at the end of the experiment are shown in Figure 5 for the formulations with 0, 10, 20, 30, 50 and 85 mol% DC8,9 PCsamples. The mean

Ac ce p

295

diameter and the standard deviation of the size distributions is reported in Table 2.

A qualitative size distribution of UV exposed, DC8,9 PC-based microbubbles

could not be determined due to the low microbubble content of the UV-exposed

300

samples. After a UV dose of 23.9 J, hardly any microbubbles could be detected.

Morphology.

Microscopy images of unexposed microbubbles are shown in Figure

6. From the microbubble contours, one can observe that the microbubble shape starts to deviate from that of a sphere when the molar concentration of DC8,9 PC exceeds 50 mol%. The curvature of the shell is not constant, and in particular 305

cases attened surface contours can be identied.

15

Page 15 of 31



and the standard deviation

σ

the unexposed reference microbubbles as measured at the start (t

min), and at the end

min) of the experiment.

DC8,9 PC mol%

t = 0 min d¯ [µm] σ [µm]

t = 170 min d¯ [µm] σ [µm]

0

2.5

1.2

3.5

1.7

10

2.1

1.0

3.6

1.4

20

2.1

0.9

4.7

2.3

30

2.5

1.3

4.2

1.9

50

2.3

1.0

2.9

1.6

85

2.9

1.4

2.9

1.6

us

= 170

=0

an

(t

of the size distribution of

ip t

The average diameter

cr

Table 2:

The irregular shapes of the microbubbles when DC8,9 PC is



50 mol% indi-

cates that the packing of the DC8,9 PC lipids at the interface does not conform

M

with a perfect sphere. It was previously found that DC8,9 PC liposomes spontaneously formed water-lled tubules when the temperature of the lipid solution 310

was lowered below the lipid transition temperature [9, 28].

This transforma-

d

tion indicates that the packing behaviour of DC8,9 PC diers from DSPC due

te

to the extended unsaturated hydrocarbon chains. Similarly, in the microbubble sample with 85 mol% DC8,9 PC, multiple stretched microbubbles can be found.

Ac ce p

A change in morphology of the microbubbles after UV exposure was not 315

observed. Microbubble compositions with less than 50 mol% DC8,9 PC remain spherical. Compositions with higher molar concentrations of DC8,9 PC still ap-

peared wrinkled after UV exposure. It must be noted that the few microbubbles that were left in the UV-exposed sample are the ones that have not been affected by the UV treatment. As such, we can not exclude that UV exposure

320

induces an instantaneous deformation of the microbubble shell via the formation of a polymer network of interconnected chains, which makes the microbubble unstable.

3.3. Ultrasound transmission To assess the stability and acoustic properties of UV-exposed and unexposed 325

microbubbles of a particular composition, ten ultrasound transmission measure-

16

Page 16 of 31

ip t cr us an M

Figure 6: Microscopy images of unexposed microbubbles. The scale bar in the top right image

d

applies to all images. The molar fraction of DC8,9 PC is indicated above the images.

ments are performed (as outlined in Figure 3): ve on the unexposed reference,

te

and ve on the exposed sample.

From the sequence of attenuation dispersions of the exposed samples [red

Ac ce p

squares and lines in Figure 4c (right)], one can observe that the attenuation dis-

330

persion drastically decreases with UV dose. However, as 30 min have elapsed in between each attenuation measurement, the microbubbles might have changed due to other environmental factors besides UV light. Therefore, the same measurement sequence was performed on an unexposed reference [yellow circles and lines in Figure 4c (left)].

335

In the evolution of the attenuation dispersion

of the unexposed reference, it is apparent that the acoustic properties of the microbubble medium change in the timespan of an experiment. However, the dispersion evolutions of the exposed sample and the unexposed reference differ strongly, indicating that UV light inuences the attenuation properties of a microbubble medium.

340

The normalized integrated attenuation of exposed samples and unexposed

17

Page 17 of 31

references for dierent microbubble compositions is shown in Figure 7.

This

quantity is proportional to the concentration of microbubbles. An exploration

ip t

of microbubble compositions with varying DC8,9 PC content was conducted to

investigate if the DC8,9 PC content has an eect on the microbubble stability, and if there is an eect of UV radiation. The normalized integrated attenuation

cr

345

decreases in time for all formulations. In DC8,9 PC-containing formulations the

us

normalized integrated attenuation of the exposed sample is found to decrease faster than the unexposed reference.

350

an

3.4. UV-vis spectroscopy

The mechanism by which UV radiation aects the microbubbles is hypothesized to be the polymerization reaction of DC8,9 PC in the microbubble shell.

M

To investigate that polydiacetylene polymers or oligomers have been created in the exposed sample, the optical properties of the unexposed reference and the exposed sample are probed using UV-vis absorption spectroscopy. Figure 8a shows the absorption spectra of a UV-exposed sample with 85

d

355

te

mol% DC8,9 PC for dierent UV dose levels. The progression of the absorbance in the region below 550 nm indicates the presence of polymerized DC8,9 PC [29]. The insert shows the dierence

∆A

between the mean absorbance from 470

Ac ce p

nm to 480 nm and from 600 nm to 650 nm as a function of the UV dose. The

360

subtraction is to translate the absorbance and remove the oset which can result from the turbidity of the dispersions. The parameter

∆A

is a crude indicator

of the presence of diacetylene polymers. A positive value of presence of polymers, whereas

∆A

∆A

suggests the

negative or equal to zero alludes to the fact

that there are no diacetylene polymers in the sample. The data are averaged

365

over three independent preparations of the same composition. The bar graph in Figure 8b shows the dierence

the exposed (red), and at

t = 190

∆A

at

t = 180

min for

min for the unexposed samples (yellow),

for the six compositions that were investigated. From this survey, it is found that polymers are formed in all UV exposed microbubble samples that contain a 370

non-zero amount of DC8,9 PC. No sign of an absorption peak was observed in the

18

Page 18 of 31

a. 0 mol%

b. 10 mol% UV Dose [J]

17.9

23.9

1.2

1.0 0.8 0.6 0.4

Exposed Reference

0.2 0.0

0

30

60 90 Time [min]

c. 20 mol% 17.9

Norm. int. a .

23.9

0.8 0.6 0.4 0.2

1.2

Exposed Reference

0

30

0.6 0.4 0.2 60 90 Time [min]

0.0

6.0

120

17.9

23.9

Exposed Reference

0.8 0.6 0.4 0.2

0

30

6.0

17.9

Ac ce p

1.0

12.0

0.8

1.2

Exposed Reference

0.6 0.4 0.2

0

30

60 90 Time [min]

120

UV Dose [J]

23.9 Norm. int. a .

0.0

60 90 Time [min]

f. 85 mol%

UV Dose [J]

Norm. int. a .

12.0

1.0

0.0

120

d

30

te

0

e. 50 mol%

0.0

60 90 Time [min]

M

0.8

1.2

23.9

UV Dose [J]

12.0

1.0

0.0

17.9

Exposed Reference

an

6.0

Norm. int. a .

1.2

12.0

d. 30 mol% UV Dose [J]

0.0

6.0

1.0

0.0

120

0.0

cr

12.0

us

6.0

Norm. int. a .

Norm. int. a .

0.0

ip t

UV Dose [J]

1.2

6.0

12.0

17.9

23.9

1.0 0.8 0.6 0.4

Exposed Reference

0.2 0.0

120

0.0

0

30

60 90 Time [min]

120

Figure 7: Overview of the normalized integrated attenuation of six microbubble compositions with varying DC8,9 PC content. Each graph shows the normalized integrated attenuation of the unexposed reference and of the UV-exposed sample as a function of time and dose. The dose received by the UV-exposed sample can be read on the top horizontal axis, whereas the time passed since the rst ultrasound measurement can be found on the bottom horizontal axis.

Each curve is the weighed mean of three independent batches, and each batch was

measured twice. The error bars indicate the standard error on the mean.

19

Page 19 of 31

ip t cr

a.

1.0

A

0.6 0.4

∆A

0.8

12.0 J 17.9 J 23.9 J

0.6 0.4 0.2 0.0

0

us

1.0

0.0 J 6.0 J

0.8

5

10

15

20

25

UV dose [J]

0.0 400

450

500

550

600

650

700

Wavelength [nm]

b.

M

0.08 Exposed Reference

0.06 0.04 0.02

d

∆A

an

0.2

−0.02

te

0.00

0%

10 %

20 %

30 %

50 %

85 %

Ac ce p

DC8,9PC content [mol%]

Figure 8:

(a) The progression of the absorption spectrum with increasing UV dose of a

microbubble solution with 85 mol% DC8,9 PC. The insert shows the dierence

∆A

between

the mean absorbance from 470 nm to 480 nm and the mean absorbance from 600 nm to 650 m as a function of the UV dose. A positive value of

∆A

indicates the presence of DC8,9 PC-

derived polymers. The bar diagram in (b) shows the dierence at a UV dose of 23.9 J (and at

t = 180

∆A

for the exposed sample

min), and for the unexposed sample at time

t = 190

min, for microbubbles with dierent molar fractions of DC8,9 PC.

20

Page 20 of 31

unexposed references, nor in the microbubble samples without DC8,9 PC lipid. The bar graph in Figure 8b is the result of measurements on a single batch for

ip t

each composition.

375

cr

3.5. Microbubble stability

The stability of microbubbles as a function of the DC8,9 PC content can examined from: (1) the evolution of the size distribution, and (2) the normalized

us

integrated attenuation from the ultrasound transmission measurements.

This

section will focus on the unexposed reference samples and the inuence of the

380

an

DC8,9 PC content rst. Subsequently, a comparison with the behaviour of UV exposed samples is made to identify the added eects of UV light.

3.5.1. Microbubble stability as a function of DC PC content

M

8,9

The studied microbubble compositions only dier in the molar fraction of the shell compounds.

Hence, the dierent evolution of the size distribution

385

d

and the normalized integrated attenuation can only be attributed to the shell composition.

te

The size distribution displayed in Figure 5 show that the mean size and the width of the distribution increase in time in the 0, 10, 20, 30 and 50 mol% Interestingly, the samples with 85 mol% do not show such a

Ac ce p

compositions.

strong change in size distribution.

390

The increase of the mean radius of the

microbubble population is also reected in the attenuation curves. Because the microbubble resonance frequency decreases with increasing radius, we observe that the frequency at which maximal attenuation occurs decreases in time (for example, see Figure 4c).

It is important to mention that the size distributions at

395

t = 0

min are

obtained from microscopy images of microbubbles sampled from a microbubble population that has been transferred from the storage medium to air-saturated PBS. Upon transfer, the microbubbles experience a transient growth due to the uptake of air. The rapid growth takes only a few seconds and the microscopy images were acquired approximately 5 minutes after this initial growth stage.

21

Page 21 of 31

400

Hence, dierences in size distributions between

t=0

min and

t = 170

min are

not the result of this equilibration process.

ip t

The evolution of the size distribution in compositions with 0, 10, 20, 30 and

50 mol% DC8,9 PC can be attributed to Ostwald ripening and coalescence [30].

405

cr

Both processes require a non-zero surface tension, hence the surface tension in these samples is greater than zero.

us

Dierent properties of the 0 mol% DC8,9 PC samples, relative to the 10, 20, 30 and 50 mol% samples, become apparent when comparing the normalized integrated attenuation (yellow squares in Figure 7). This parameter is propor-

410

an

tional to the concentration of microbubbles in the sample. Clearly, the 0 mol% sample is more stable than the 10, 20, 30 and 50 mol% samples.

The addi-

tion of DC8,9 PC to the DSPC matrix aects the stability of the monolayer. We

M

hypothesize that 10, 20, 30 and 50 mol% samples collapse under the Laplace pressure which is not completely reduced since the surface tension is non-zero. The Laplace pressure increases with decreasing radius.

microbubbles shrink, they reach a critical radius where the rigid monolayer col-

d

415

Consequently, as the

te

lapses. The pressure at which the monolayer collapse occurs is lower in 10, 20, 30 and 50 mol% samples than in a 0 mol% DC8,9 PC sample. This hypothesis is

Ac ce p

supported by a study on Langmuir monolayers with 10 mol% DC8,9 PC and 90 mol% DPPC, which were found to collapse at lower surface pressures than pure

420

DPPC or pure DC8,9 PC monolayers [21, 31].

The observations thus indicate

that mixed monolayers of DSPC and DC8,9 PC are not as resistant against high surface pressures as bubbles with 0 mol% DC8,9 PC. At 85 mol% DC8,9 PC the microbubbles become more stable again, this is

apparent from the moderate decrease in normalized integrated attenuation and

425

the stability of the size distribution. Their stability is comparable to the stability of the 0 mol% DC8,9 PC samples. Unlike the 0 mol% sample, the size distribution of the 85 mol% sample is stable.

We hypothesize the high stability of the

size distribution is due to: (1) the high gas permeation resistance of the lipid monolayer which is expected from the long hydrocarbon chains of DC8,9 PC 430

(23 carbons) [32], and (2) the low surface tension. Both properties slow down

22

Page 22 of 31

Ostwald ripening [30]. Another indication of the low surface tension of these microbubbles can be

ip t

found in the morphology of the microbubbles with 85 mol% DC8,9 PC (Figure 6). Many microbubbles in these samples have irregular contours, where the surface

area is not minimized. As these microbubbles are stable in the timespan of the

cr

435

experiment, they must have a low surface tension. The transition temperature

43◦ C

[10], hence, at room temperature, DC8,9 PC monolayers are

us

of DC8,9 PC is

in a condensed state where the hydrocarbon chains are aligned and packed in an periodic manner [33, 34]. Under these conditions, an eective surface tension of zero can be accomplished.

an

440

The limiting molar fraction at which the macroscopic properties of the bubble are determined by one particular compound can be estimated from perPercolation theory studies the behaviour

M

colation theory considerations [35].

of clusters in a lattice, and can be applied to a 2D lipid monolayer at the 445

gas-water interface. An import quantity is the percolation threshold, which is

Rigidity percolation, an extended version of the percolation theory,

te

lattice.

d

the concentration at which an innite network/cluster appears in an innite

is particularly useful to describe the behaviour of a lipid monolayer, as it re-

Ac ce p

quires that the formed cluster not only spans the lattice but is also rigid with 450

respect to the applied forces. The theoretical value of the percolation threshold for a generic square lattice is 0.69755

±

0.0003 [36].

The concepts of 2D

rigidity percolation were rst mapped to mixed lipid monolayers to interpret the composition-collapse problem by Gopal and Lee [37]. They experimentally observed that the collapse pressure of a DPPC:PA (palmitic acid) monolayer

455

considerably increased when the molar fraction of DPPC exceeded 70 mol%. Their observations are in agreement with the theoretical calculations, when a lattice site occupied by a PA molecule is considered empty, and only a DPPC molecule (with a bulkier headgroup) can provide rigidity. Further, Kwan and Borden observed that the headgroup rigidity percolation threshold also applies

460

to the stability of DPPC:PEG40S microbubbles [38]. Microbubbles with > 70 mol% DPPC are more stable than microbubbles with less DPPC due to the for-

23

Page 23 of 31

mation of an extended condensed cluster. In this perspective, one can assume that at 85 mol% DC8,9 PC, a network of condensed DC8,9 PC molecules extends

465

ip t

over the surface of the microbubble, and governs the macroscopic properties, such as the rigidity and the permeability of the monolayer.

cr

However, microbubbles composed of 95 mol% DC8,9 PC, 5 mol% PEG40s and no DSPC were very unstable and dicult to prepare. This suggests that DSPC,

us

or another vesicle forming component, is a crucial in the formation process, or in the stabilization process, or in both. 470

In summary, compositions with 0 mol% and 85 mol% are more stable than

an

compositions with 10, 20, 30 and 50 mol% DC8,9 PC. The latter compositions have non-zero surface tension as is apparent from the observed Ostwald ripening, and are expected to collapse at lower surface pressures than pure DSPC

475

Microbubbles with 85 mol% DC8,9 PC are more sta-

M

or DC8,9 PC monolayers.

ble against Ostwald ripening due to the high gas permeation resistance of the

d

monolayer and the low surface tension owing to the condensed state of the lipid.

te

3.5.2. The eect of UV-exposure on the microbubble stability A survey of the UV-sensitivity of the studied compositions is shown in Figure 7, where the red lines and the red squares denote the normalized integrated attenuation of the UV-exposed samples. A common trend in all compositions,

Ac ce p 480

is that the normalized integrated attenuation of the exposed sample decreases faster than that of the unexposed reference. This indicates that UV light degrades the stability of microbubbles. The dierence between the exposed sample and the unexposed reference is most pronounced in the formulation with

485

85 mol% DC8,9 PC. The hypothesis that the impaired stability results from the polymerization reaction of the DC8,9 PC monomers in the microbubble shell is

experimentally supported by: (1) the presence of polymerized DC8,9 PC in the

UV exposed samples, as noticeable from the increased absorbance in the UVvis absorption spectrum (Fig. 8b), and (2) the observation that the dierence 490

in normalized integrated attenuation between the unexposed reference and the exposed sample increases with DC8,9 PC content. For instance, the formulation

24

Page 24 of 31

with 0 mol% is not strongly inuenced by UV exposure (i.e. the exposed sample and the unexposed reference follow the same trend in Figure 7a), while for all

495

ip t

DC8,9 PC containing formulation the evolution of the unexposed reference and the exposed sample diers.

cr

The ndings suggest that the polymerization in the monolayer disrupts the

integrity of the microbubble shell and impairs the microbubble stability. The

us

formation of a polymer is accompanied by a structural change in the packing of the lipids, since the intermolecular distance between consecutive diacetylene 500

units is smaller in the polymer than in the monomeric condensed monolayer The reduction in the intermolecular distance results in a contraction of

an

[39].

the total area covered by the cross-linked DC8,9 PC molecules.

Such a UV-

induced contraction has been observed in Langmuir monolayers of single [40],

505

The moment when a condensed

M

and double chained diacetylene lipids [34].

diacetylene Langmuir monolayer is exposed to UV light, a sudden drop in the surface pressure exerted on the barriers of the trough is observed, indicating a

d

contraction of the mean area per molecule. The decrease occurs within seconds

te

after the start of the exposure. Depending on the type of diacetylene, the area contraction ranges from 1 % (acids), to 3 % (lecithin), up to 10 % for single-chain lipids at room temperature [41].

Ac ce p

510

Studies on binary DC8,9 PC:DMPC liposomes indicated that not all DC8,9 PC

molecules are polymerized after UV exposure but cross-linking only occurs in DC8,9 PC-dense domains [8].

It is likely that the contraction of DC8,9 PC-rich

regions in the microbubble shell also results in the formation of pores, and

515

abruptly increases the surface tension since a part of the gas-water interface is no longer covered with amphiphiles for a fraction time. Because lipid molecules in condensed domains are practically immobile [42], it takes tens of microseconds for the unpolymerized molecules to redistribute at the surface. Furthermore, it would be energetically unfavorable for a DSPC molecule to migrate out of the

520

condensed domain into a void space at the interface. The increase in surface tension amplies the Laplace pressure, and as such, advances the shrinkage of the microbubble until the void is no longer there. The presence of the polymer

25

Page 25 of 31

in the membrane acts as a mechanical instability at which the monolayer folds

525

ip t

and collapse takes place [43, 44, 45, 46].

4. Conclusion

cr

The diacetylene lipid DC8,9 PC can be incorporated in a microbubble shell

and can be polymerized in this conguration. The inclusion of DC8,9 PC molecules

us

in a DSPC matrix, aects the stability of the microbubble as is apparent from the decrease in the attenuation coecient of the microbubble medium over time. 530

The stability is strongly impaired when the molar fractions of DC8,9 PC and

an

DSPC are similar. When the shell composition is dominated by DC8,9 PC, for instance when the molar fraction of DC8,9 PC is 85 mol%, the microbubbles are

M

stable and show a high gas permeation resistance due to the long hydrocarbon chains. The UV-sensitivity survey showed that microbubbles with DC8,9 PC 535

incorporated in the shell are sensitive to UV light with a wavelength of 254

d

nm. The concentration of microbubbles was found to drop as a function of UV dose in all compositions that were investigated. The impaired stability can be

te

attributed to the polymerization of DC8,9 PC in the microbubble shell.

Ac ce p

Acknowledgement

540

The research leading to these results has gratefully received funding from

the Research FoundationFlanders (FWO).

References

[1] R. Jelinek, M. Ritenberg, Polydiacetylenesrecent molecular advances and applications, RSC Advances 3 (2013) 2119221201.

545

[2] A. Mueller, D. F. O'Brien, Supramolecular materials via polymerization of mesophases of hydrated amphiphiles, Chem. Rev. 102 (2002) 727757.

26

Page 26 of 31

[3] G. Wegner,

Topochemische reaktionen von monomeren mit konjugierten

dreifachbindungen/tochemical

reactions

of

monomers

with

conjugated

550

[4] H. Bässler,

Photopolymerization of diacetylenes,

in:

Polydiacetylenes,

cr

Springer-Verlag, Berlin Heidelberg, Germany, 1984, pp. 148.

Solid-state synthesis of large polymer single crystals,

J.

us

[5] R. Baughman,

ip t

triple bonds, Zeitschrift für Naturforschung B 24 (1969) 824832.

Polym. Sci. B Polym. Phys. 12 (1974) 15111535.

[6] G. Lieser, B. Tieke, G. Wegner,

merizability of multilayers of some diacetylene monocarboxylic acids, Thin

an

555

Structure, phase transitions and poly-

Solid Films 68 (1980) 7790.

tives in layer structures, 1985, pp. 79151.

in:

Analysis/Reactions/Morphology, Springer,

[8] C. F. Temprana, E. L. Duarte, M. C. Taira, M. T. Lamy, S. del Valle Alonso,

d

560

M

[7] B. Tieke, Polymerization of butadiene and butadiyne (diacetylene) deriva-

te

Structural characterization of photopolymerizable binary liposomes containing diacetylenic and saturated phospholipids,

Ac ce p

1008410092.

Langmuir 26 (2010)

[9] P. Yager, P. E. Schoen, Formation of tubules by a polymerizable surfactant,

565

Mol. Cryst. Liq. Cryst. 106 (1984) 371381.

[10] G. L. Jendrasiak, A. A. Ribeiro, N. Mark A, P. E. Schoen, A temperature study of diacetylenic phosphatidylcholine vesicles, Biochim. Biophys. Acta 1194 (1994) 233238.

[11] A. Yavlovich, A. Singh, S. Tarasov, J. Capala, R. Blumenthal, A. Puri,

570

Design of liposomes containing photopolymerizable phospholipids for triggered release of contents, J. Therm. Anal. Calorim. 98 (2009) 97.

[12] M. Viard, H. Reichard, B. A. Shapiro, F. A. Durrani, A. J. Marko, R. M. Watson, R. K. Pandey, A. Puri,

Design and biological activity of novel

27

Page 27 of 31

stealth polymeric lipid nanoparticles for enhanced delivery of hydrophobic photodynamic therapy drugs, Nanomedicin 14 (2018) 22952305.

[13] S. Sheth, D. Leckband,

ip t

575

Direct force measurements of polymerization-

dependent changes in the properties of diacetylene lms,

cr

(1997) 56525662.

Langmuir 13

580

us

[14] M. A. Reppy, B. A. Pindzola, Biosensing with polydiacetylene materials: structures, optical properties and applications, 43174338.

Chem. Commun. (2007)

an

[15] K. Morigaki, T. Baumgart, U. Jonas, A. Oenhäusser, W. Knoll, Photopolymerization of diacetylene lipid bilayers and its application to the construction of micropatterned biomimetic membranes, Langmuir 18 (2002) 40824089.

M

585

[16] S. Alonso-Romanowski, N. S. Chiaramoni, V. S. Lioy, R. A. Gargini, L. I.

d

Viera, M. C. Taira, Characterization of diacetylenic liposomes as carriers

te

for oral vaccines, Chem. Phys. Lipids 122 (2003) 191203.

[17] S. Punnamaraju, H. You, A. Steckl, Triggered release of molecules across droplet interface bilayer lipid membranes using photopolymerizable lipids,

Ac ce p

590

Langmuir 28 (2012) 76577664.

[18] A. Yavlovich, A. Singh, R. Blumenthal, A. Puri, A novel class of phototriggerable liposomes containing DPPC:DC(8,9)PC as vehicles for delivery of doxorubcin to cells, BBA Biomembranes 1808 (2011) 117126.

595

[19] A. Puri, H. Jang, A. Yavlovich, M. A. Masood, T. D. Veenstra, C. Luna, H. Aranda-Espinoza, R. Nussinov, R. Blumenthal, Material properties of matrix lipids determine conformation and intermolecular reactivity of a diacetylenic phosphatidylcholine in the lipid bilayer, Langmuir 27 (2011) 15120.

28

Page 28 of 31

600

[20] E. Lopez, D. O'Brien, T. Whitesides, Eects of membrane composition and

membranes, Biochim. Biophys. Acta 693 (1982) 437443.

ip t

lipid structure on the photopolymerization of lipid diacetylenes in bilayer

[21] S. Baek, M. D. Phan, J. Lee, K. Shin, Packing eects on polymerization

605

cr

of diacetylene lipids in liposomes and monolayers matrices, Polym. J. 48 (2016) 457463.

us

[22] D. Cosgrove, Ultrasound contrast agents: an overview, Eur. J. Radiol. 60 (2006) 324330.

an

[23] N. de Jong, P. J. Frinking, A. Bouakaz, F. J. Ten Cate, Detection procedures of ultrasound contrast agents, Ultrasonics 38 (2000) 8792.

610

[24] K. Ferrara, R. Pollard, M. Borden,

Ultrasound microbubble contrast

M

agents: fundamentals and application to gene and drug delivery.,

Annu.

Rev. Biomed. Eng. 9 (2007) 41547.

d

[25] S. M. van der Meer, B. Dollet, M. M. Voormolen, C. T. Chin, A. Bouakaz, N. de Jong, M. Versluis, D. Lohse, Microbubble spectroscopy of ultrasound contrast agents., J. Acoust. Soc. Am. 121 (2007) 648656.

te

615

Ac ce p

[26] D. H. Kim, M. J. Costello, P. B. Duncan, D. Needham, Mechanical properties and microstructure of polycrystalline phospholipid monolayer shells: Novel solid microparticles, Langmuir 19 (2003) 84558466.

[27] E. Stride, M. Tang, R. J. Eckersley, Physical phenomena aecting quan-

620

titative imaging of ultrasound contrast agents,

Appl. Acoust. 70 (2009)

13521362.

[28] M. S. Spector, K. R. Easwaran, G. Jyothi, J. V. Selinger, A. Singh, J. M. Schnur, Chiral molecular self-assembly of phospholipid tubules: A circular dichroism study, P. Natl. Acad. Sci. USA 93 (1996) 1294312946.

625

[29] D. S. Johnston, S. Sanghera, M. Pons, D. Chapman, Phospholipid polymer synthesis and spectral characteristics, BBA Biomembranes 602 (1980) 57 69.

29

Page 29 of 31

[30] P. Taylor, Ostwald ripening in emulsions, Adv. Colloid Interfac. 75 (1998)

630

[31] A. Gopal, K. Y. C. Lee,

ip t

107163. Morphology and collapse transitions in binary

cr

phospholipid monolayers, J. Phys. Chem. B 105 (2001) 1034810354.

[32] M. A. Borden, M. L. Longo, Dissolution behavior of lipid monolayer-coated, air-lled microbubbles: Eect of lipid hydrophobic chain length, Langmuir

635

us

18 (2002) 92259233.

[33] D. S. Johnston, L. R. McLean, M. A. Whittam, A. D. Clark, D. Chapman,

an

Spectra and physical properties of liposomes and monolayers of polymerizable phospholipids containing diacetylene groups in one or both acyl chains,

M

Biochemistry 22 (1983) 31943202.

[34] P. Meller, R. Peters, H. Ringsdorf, Microstructure and lateral diusion in 640

monolayers of polymerizable amphiphiles, Colloid Polym. Sci. 267 (1989)

d

97107.

1994.

te

[35] D. Stauer, A. Aharony, Introduction to percolation theory, CRC press,

Ac ce p

[36] D. Jacobs, M. Thorpe,

645

Generic rigidity percolation in two dimensions,

Phys. Rev. E 53 (1996) 3682.

[37] A. Gopal, K. Y. C. Lee, Headgroup percolation and collapse of condensed langmuir monolayers, J. Phys. Chem. B 110 (2006) 2207922087.

[38] J. J. Kwan, M. A. Borden,

Lipid monolayer collapse and microbubble

stability, Adv. Colloid Interf. Sci. 183 (2012) 8299.

650

[39] Y. Lifshitz, Y. Golan, O. Konovalov, A. Berman, Structural transitions in polydiacetylene Langmuir lms., Langmuir 25 (2009) 446977. [40] D. Y. Sasaki, R. W. Carpick, A. R. Burns, High molecular orientation in mono-and trilayer polydiacetylene lms imaged by atomic force microscopy, J. Colloid Interf. Sci. 229 (2000) 490496.

30

Page 30 of 31

655

[41] B. Ostermayer, O. Albrecht, W. Vogt,

Polymerizable lipid analogues of

diacetylenic phosphonic acids. synthesis, spreading behaviour and polymer-

ip t

ization at the gas-water interface, Chem. Phys. Lipids 41 (1986) 265291.

[42] J.-F. Tocanne, L. Dupou-Cézanne, A. Lopez, Lateral Diusion of Lipids in

[43] H. Diamant, T. A. Witten, A. Gopal, K. Y. C. Lee, Unstable topography

us

660

cr

Model and Natural Membranes, Prog. Lipid Res. 33 (1994) 203237.

of biphasic surfactant monolayers, Europhys. Lett. 51 (2000) 171.

[44] H. Diamant, T. Witten, C. Ege, a. Gopal, K. Lee,

Topography and in-

an

stability of monolayers near domain boundaries, Phys. Rev. E 63 (2001) 061602.

[45] L. Pocivavsek, S. L. Frey, K. Krishan, K. Gavrilov, P. Ruchala, A. J. War-

M

665

ing, F. J. Walther, M. Dennin, T. A. Witten, K. Y. C. Lee, Lateral stress relaxation and collapse in lipid monolayers, Soft Matter 4 (2008) 2019.

d

[46] L. Bourdieu, D. Chatenay, J. Daillant, D. Luzet, Polymerization of a diJournal de

Physique II 4 (1994) 3758.

Ac ce p

670

te

acetylenic phospholipid monolayer at the air-water interface,

31

Page 31 of 31