Label-Free Infrared Spectroscopy and Imaging of Single Phospholipid

Label-Free Infrared Spectroscopy and Imaging of Single Phospholipid Bilayers with Nanoscale Resolution ... Publication Date (Web): August 3, 2018. Cop...
0 downloads 0 Views 1012KB Size
Subscriber access provided by MT ROYAL COLLEGE

Article

Label-free Infrared Spectroscopy and Imaging of Single Phospholipid Bilayers with Nanoscale Resolution. Adrian Cernescu, Micha# Szuwarzy#ski, Urszula Kwolek, Pawel Wydro, Mariusz Kepczynski, Szczepan Zapotoczny, Maria Nowakowska, and Luca Quaroni Anal. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.analchem.8b00485 • Publication Date (Web): 03 Aug 2018 Downloaded from http://pubs.acs.org on August 4, 2018

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

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

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

Analytical Chemistry

1

Label-free Infrared Spectroscopy and Imaging of Single Phospholipid

2

Bilayers with Nanoscale Resolution.

3 4

Adrian Cernescu1, Michał Szuwarzyński2, Urszula Kwolek3, Paweł Wydro3, Mariusz

5

Kepczynski3, Szczepan Zapotoczny3, Maria Nowakowska3, Luca Quaroni3*

6

1

Neaspec GmbH, Bunsenstrasse 5, 82152 Martinsried, Germany

7

2

AGH University of Science and Technology, Academic Centre for Materials and

8

Nanotechnology, al. A. Mickiewicza 30, 30-059, Kraków, Poland

9

3

Jagiellonian University, Faculty of Chemistry, ul. Gronostajowa 2, 30-387, Kraków, Poland

10

* Corresponding Author: Luca Quaroni, Faculty of Chemistry, Jagiellonian University

11

ul. Gronostajowa 2, 30-387, Kraków, Poland. E-mail: [email protected]

12

ORCID: Paweł Wydro, 0000-0002-9145-1362; Mariusz Kepczynski, 0000-0002-7304-6881;

13

Szczepan Zapotoczny, 0000-0001-6662-7621; Maria Nowakowska, 0000-0001-6456-5463; Luca

14

Quaroni 0000-0002-0225-9775

15

Abstract

16

Mid-infrared absorption spectroscopy has been used extensively to study the molecular

17

properties of cell membranes and model systems. Most of these studies have been carried out on

18

macroscopic samples or on samples a few micrometers in size, due to constraints on sensitivity

19

and spatial resolution with conventional instruments operating with far-field optics. Properties of

20

membranes on the scale of the nanometers, such as in-plane heterogeneity, have to date eluded

21

investigation by this technique. In the present work we demonstrate the capability to study single

22

bilayers of phospholipids with near-field mid-infrared spectroscopy and imaging, and achieve a

1 ACS Paragon Plus Environment

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

Page 2 of 31

1

spatial resolution of at least 40 nm, corresponding to a sample size of about a thousand

2

molecules. The quality of the data and the observed spectral features are consistent with those

3

reported from measurements of macroscopic samples and allow detailed analysis of molecular

4

properties, including orientation and ordering of phospholipids. The work opens the way to the

5

detailed nanoscale characterization of the biological membranes for which phospholipid bilayers

6

serve as a model.

7 8

Main Text

9

Membranes are defining structures of cells. Their role, however, extends beyond basic

10

compartmentalization, and includes overall organization of proteins in space and time, energy

11

storage, control of trafficking and homeostatic regulation, intracellular and intercellular signal

12

transduction, recognition, and adhesion. Despite intense interest, many issues related to the

13

structure and function of membranes are still unaddressed or there is no consensus reached on

14

them among researchers. This is partly due to the challenges intrinsic to the study of individual

15

membranes. A cellular membrane consists of a bilayer of phospholipids and embedded proteins

16

which is a few nanometers in thickness.1 The structure puts severe constraints on spectroscopic

17

studies of individual membranes because of its demands on both the sensitivity and the spatial

18

resolution of the measurement. This is particularly critical for the understanding of in-plane

19

nanoscale organization of the lipid and protein components. The latter has been a fundamental

20

issue in membrane biology since the introduction of the fluid mosaic model by Singer and

21

Nicholson 2 and its subsequent development into the lipid raft model by Simons. 3

22

Over the last few decades, infrared (IR) absorption spectroscopy has been extensively applied to

23

the study of biological systems, thanks to its capability to provide molecular and dynamic 2 ACS Paragon Plus Environment

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

Analytical Chemistry

1

information on a broad variety of samples, including single living cells. 4 Many IR spectroscopy

2

studies have been devoted to understanding membranes and their model systems, such as

3

phospholipid bilayers and multibilayers. The available information includes structure and

4

orientation of headgroups, alkyl chains and embedded proteins, order at the molecular level,

5

phase transitions, lipid-lipid or protein-lipid interactions and membrane dynamics. 5, 6, 7 Most IR

6

absorption measurements have been carried out on samples of macroscopic dimensions, such as

7

vesicle dispersions and multibilayer stacks, using transmission and attenuated total reflection

8

(ATR) configurations. IR microscopy configurations that rely on far-field optical geometries lack

9

the spatial resolution, because of the limits set by diffraction, and the sensitivity to study

10

nanoscale structure in single bilayers. Because of these constraints, the application of IR

11

microscopy to the study of membranes in vivo has so far been possible only in very specific

12

cases, such as the extended stacks formed by the disks of rod outer segments.

13

grazing incidence geometry has allowed the selective study of extended bilayers and single

14

leaflets, 9 either supported on solid substrates or at the air water interface. However, because of

15

the high angle of incidence, these measurements do not provide the spatial resolution necessary

16

to study in-plane membrane organization at the nanoscale. Nanoscale structure and reactivity of

17

membranes and bilayers have, to date, escaped direct investigation by IR absorption

18

spectroscopy. Current studies on this scale rely mostly on atomic force microscopy (AFM), to

19

study topographical and mechanical properties, on fluorescence microscopy, for diffusional

20

dynamic studies, 10 or on super-resolved light microscopy, 11 for localization studies.

21

The introduction of near-field spectroscopic techniques for the IR region has recently provided

22

the sensitivity and spatial resolution to allow the study of soft matter at the nanometer scale.

23

Among these techniques, IR scattering Scanning Near-Field Optical Microscopy (IR sSNOM or

8

The use of a

12

3 ACS Paragon Plus Environment

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

Page 4 of 31

1

simply sSNOM) and nano-FTIR13 have sensitivity and spatial resolution that allow the study of

2

minute amounts of biological material down to the scale of 20 nanometers.14 In both techniques,

3

the near-field IR signal from a sample is provided by the localized electric field in the proximity

4

of an externally illuminated metal coated AFM tip, simultaneously with AFM information.15

5

While sSNOM provides images of the sample, nano-FTIR allows the measurement of extended

6

spectra. Both techniques can be implemented with the same instrument and on the same sample.

7

In the case of sSNOM, an image of the sample is produced by raster scanning under such

8

nanofocused light field at a specific wavelength, thus obtaining the local absorption and

9

reflectivity information. The sSNOM signal is also highly sensitive to crystallinity of the

10

investigated samples such as highly ordered organic materials.16 In the case of nano-FTIR, the

11

scattered light from a broadband source is analyzed by Fourier transform of the light analyzed by

12

an asymmetric Michelson interferometer while the tip is positioned at a specific location, thus

13

providing the local FTIR absorption and reflection spectra of the sample.17

14

images of the sample can be obtained by collecting spectra over a 2D array of positions,18

15

similarly to other IR near-field techniques. 19

16

The capability to provide IR absorption nano-FTIR spectra and IR sSNOM images of single

17

bilayers of the purple membranes from H. salinarum has been demonstrated.20 However, these

18

studies provided the IR absorption of peptide bonds in patches of bacteriorhodopsin proteins.

19

Measurements on portions of the membrane without bacteriorhodopsin molecules were not

20

reported. The measurement of phospholipid headgroups in a bilayer membrane is expected to be

21

a major challenge, because of their lower concentration within the probed volume compared to

22

the peptide bonds of membrane proteins. A previous attempt to use sSNOM to measure

Hyperspectral

4 ACS Paragon Plus Environment

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

Analytical Chemistry

1

supported phospholipid bilayers relied on the measurement of the ester carbonyl band and

2

provided a limiting sensitivity of three bilayers and an in-plane resolution of about 100 nm.21

3

In this work we demonstrate the possibility to measure the nanoscale IR absorption of

4

headgroups and alkyl chains in single phospholipid bilayers by using nano-FTIR and to image

5

their spatial distribution with a resolution better than 40 nm using sSNOM. We use single and

6

multiple bilayers of 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) deposited on a silicon

7

surface as simplified models of biological membranes. Despite the compositional simplicity, the

8

system provides a challenge for the sensitivity and spatial resolution of sSNOM and nano-FTIR

9

that is comparable to the one from real cell membranes.

10 11

Results and Discussion.

12

We deposited single bilayers and multibilayer stacks of DPPC by allowing liposomes to adhere

13

and unfold on the surface of a silicon wafer. The resulting sample, as seen in AFM height maps

14

recorded at room temperature (Figure 1A), displays patches of single and multibilayers separated

15

by areas of exposed silicon surface. Profile analysis shows that each bilayer was about 5.5 nm

16

thick (Figure 1B), as expected for DPPC bilayers below the transition temperature, at 41 °C for a

17

suspension of multilamellar vesicles.22,23 Multibilayers display a terraced structure that exposes

18

steps corresponding to one, two, three or more bilayers. In addition to stacks of bilayers, also

19

particles can be observed on top of the bilayers or on the substrate (Figure 1A). Figure S1

20

compares the AFM height map for a stack of bilayers with the corresponding AFM phase map.

21

The observed difference in phase between the bilayer surface and the surface of the support

5 ACS Paragon Plus Environment

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

Page 6 of 31

1

confirms that the stacks are laying directly on silicon, and not on an extended layer of organic

2

material.

3 4 5

Figure 1. Nano-FTIR spectroscopic measurements of bilayers of DPPC on a silicon substrate.

6

A. AFM height image of a region of the sample. The scale bar corresponds to 200 nm. B. Height

7

profile along the line in panel A. C. Nano-FTIR spectra measured at the positions marked in

8

panel A. Spectra are color coded with position markers. Red, single bilayer; Blue and Green,

9

double bilayers. The nano-FTIR signal plotted in C is the imaginary part of the measured

10

scattering intensity Im [σn(ω)]. The green arrows mark the positions that will be used for on-resonance

11

and off- resonance sSNOM imaging in Figure 4.

6 ACS Paragon Plus Environment

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

Analytical Chemistry

1

We use the AFM height map to survey the sample and select positions for the nano-FTIR

2

spectral measurement, corresponding to different thickness values. Figure 1C shows nano-FTIR

3

spectra measured with laser power emission in the 800 - 1400 cm-1 range. All quantities used in

4

nano-FTIR and sSNOM experiments and their relationship are defined in the Supplementary

5

Information. In contrast to conventional FTIR experiments with a glowbar source, which can

6

simultaneously measure the whole mid-infrared spectra region, nano-FTIR measurements with a

7

DFG source are limited to the measurement of smaller spectral intervals, approximately 600 to

8

800 cm-1 wide, depending on the optical configuration of the source. We selected the 800 - 1400

9

cm-1 spectral region for our experiments because it allows us to measure simultaneously

10

absorption bands from all functional groups of a phospholipid molecule. The spectra were

11

obtained by plotting the nano-FTIR absorbance, defined as the imaginary part of the scattering

12

coefficient, an≡Im [σn(ω)] (Supplementary Information).17 The index n is the demodulation

13

order used to extract the scattering signal. The spectra of Figure 1C were obtained from the 2nd

14

harmonic of the modulation frequency. It has been shown that for weak oscillators, such as

15

phospholipids, the nano-FTIR absorbance an is proportional to the absorbance as measured in a

16

macroscopic transmission experiment.17

17

Measurements were performed in three different positions, corresponding to single and double

18

bilayers, as marked in Figure 1A. Nano-FTIR bands are reproduced in all three different

19

locations and in other sample locations (not shown), indicating that they have little contribution

20

from noise. The signal-to-noise ratio for these measurements is in the range 7 to 10 in the central

21

part of the spectral region, where laser intensity is highest, but poorer at the fringes (between

22

800-850 cm-1 and 1350-1400 cm-1), where laser intensity tapers off. At the very edge of the

23

spectral range, below 850 cm-1 and above 1350 cm-1, decreased reproducibility indicates that 7 ACS Paragon Plus Environment

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

Page 8 of 31

1

noise is overtaking the signal. Comparison of the spectra measured at the three positions in

2

Figure 1A shows that the intensity of the bands from the double bilayer is about 80% to 100%

3

higher than that from the single bilayer. This is expected for bilayers of about 5 nm thickness,

4

which are much thinner than the probing depth of sSNOM, of the order of few tens of

5

nanometers. Figure S2 shows an approach curve on the silicon surface. The curve provides the

6

intensity of the signal (optical amplitude a given Ωn) as a function of the tip-sample distance.

7

Signal intensity decreases exponentially with tip-sample distance. Therefore, additional bilayers

8

provide an increased contribution to the phase signal as long as they are still within the volume

9

probed by the electric field. Because of the decay of the field away from the tip, the contribution

10

of additional bilayers to the total signal decreases progressively, with little or no increase

11

observed after the fifth bilayer (not shown).

12

To compare the nano-FTIR spectra of DPPC in Figure 1 with ATR-FTIR spectra from

13

macroscopic samples, we deposited a suspension of DPPC liposomes on a diamond ATR prism

14

and allowed it to dry in air. Spectra were recorded soon after the disappearance of the absorption

15

bands from bulk water, to allow the sample to retain some hydration. The resulting spectrum is

16

shown in Figure 2A and compared with one of the nano-FTIR spectra for a double bilayer. Peak

17

positions were determined using the plot of the second derivative of absorbance (Figure 2B) and

18

are listed in Table 1. Table 1 also lists the band positions of ATR-FTIR spectra of DPPC

19

reported in the literature.24

20

8 ACS Paragon Plus Environment

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

Analytical Chemistry

1 2

Figure 2. Comparison of nanoFTIR (blue line) and ATR-FTIR (red line) spectra of DPPC. A. ATR-

3

FTIR spectrum of DPPC and nano-FTIR spectrum of a double bilayer of DPPC (from Figure 1). The

4

vertical axis plots absorbance units for the ATR-FTIR spectrum and nano-FTIR absorbance units for the

5

nano-FTIR spectrum. The scale is the same for both spectra except that an offset has been applied to the

6

nano-FTIR spectrum for clarity. B. Second derivative of the ATR-FTIR spectrum and of the nano-FTIR

7

spectrum in panel A. The scale is the same for both spectra except that an offset has been applied to the

8

derivative of the nano-FTIR spectrum for clarity. The dotted lines show the position of peaks in the nano-

9

FTIR spectrum and its derivative. The green arrows mark the positions that will be used for on-resonance

10

and off- resonance sSNOM imaging in Figure 4.

9 ACS Paragon Plus Environment

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

Page 10 of 31

1

Table 1. IR Bands of DPPC Bilayers. Wavenumbers of bands observed in nano-FTIR spectra of

2

DPPC in the 800 – 1500 cm-1 spectral region and comparison with values from ATR

3

measurements of oriented methyl palmitate and DPPC, and corresponding assignments. The

4

band marked with *, corresponding to the antisymmetric stretching mode (νPO2-asym), was not

5

observed in nano-FTIR spectra. sSNOM of DPPC

6

ATR of Methyl Palmitate

ATR of DPPC(§)

ATR of DPPC(¶)

970 970 970 1069 1060-1072 1058 1096 1093 1091 1178 1177 1178 1180 1200 1198 1198 1198 1222 1220 1219 1222 1244 1242 1242 1243 * 1255 1243 1265 1265 1265 1265 1286 1287 1287 1284 1303 1310 1310 1314 1321 1331 (§) Literature data.24 (¶) Values measured in this work.

Assignment

νN+(CH3)3 asym νC-O-PO2νPO2-sym. νCO-O ωCH2 wag ωCH2 wag ωCH2 wag νPO2-asym. ωCH2 wag ωCH2 wag ωCH2 wag ωCH2 wag

7 8

Band intensity and position in nano-FTIR spectra are consistent with the spectra of macroscopic

9

samples of DPPC, both from literature data and from our own measurements. The relative

10

intensity and position of most bands in ATR-FTIR spectra closely match the ones observed in

11

the nano-FTIR spectra. Small differences in band positions reported for the ATR-FTIR spectra of

12

macroscopic DPPC samples in the literature can be ascribed to both differences in hydration of

13

the samples and differences in the order and packing of bilayers as deposited on the ATR crystal.

14

The only major difference is the relative intensity of the bands at 1198 cm-1, 1222 cm-1 and 1243

15

cm-1. This difference is clearer in the 2nd derivative traces of Figure 2B, confirming that it is due 10 ACS Paragon Plus Environment

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

Analytical Chemistry

1

to differences in absorption between the two measurements and not to differences in the baseline.

2

Baseline variations are greatly reduced or cancelled by derivation.

3

Bands in this spectral region are mostly assigned to vibrational modes of phospholipid

4

headgroups, as described in Table 1. In addition, a regular sequence of peaks is observed

5

between 1200 cm-1 and 1300 cm-1 (1200 cm-1, 1222 cm-1, 1244 cm-1, 1265 cm-1, 1286 cm-1, 1303

6

cm-1, 1321 cm-1) that corresponds to the progression of the methylene wagging modes of the

7

alkyl chain. 25 Methylene wagging modes have their transition moment aligned along the axis of

8

the extended alkyl chain (Figure 3). The progression arises from the linear combination of

9

wagging modes of individual methylene groups and is observed for highly oriented chains, while

10

it decreases or disappears as the chains become disordered, such as above the gel to liquid

11

transition temperature of phospholipids.25 In sSNOM and nano-FTIR measurements, the tip-

12

enhanced electric field is oriented along the axis of the tip and normal to the plane of the bilayer

13

(Figure 3). This geometry allows strong absorption by the wagging modes of the alkyl chains,

14

which extend along the normal to the plane. The regularity of the wagging progression in our

15

nano-FTIR measurements is comparable to that observed in the ATR spectra of oriented methyl

16

palmitate and DPPC multibilayers.

17

indicate that the alkyl chains of the phospholipids are extended and approximately perpendicular

18

to the plane of the bilayer.

19

One difference from the spectra of macroscopic samples was observed in the intensity of the

20

wagging mode bands at 1198 cm-1, 1220 cm-1, which show lower absorption than in ATR-FTIR

21

spectra. We suggest that this may be due to the thin oxide layer on the Si substrate.

22

reference measurement was carried out in a region of the substrate without lipids, and Si-O

24

Band position and intensity of the wagging progression

26

The

11 ACS Paragon Plus Environment

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

Page 12 of 31

1

bonds on the surface would add to the reference absorption and provide a negative contribution

2

to the nano-FTIR spectrum, with the apparent suppression of overlapping bands.

3 4

It is notable that in our measurements the absorption of the symmetric stretching mode of the

5

phosphate headgroup (νPO2-sym) is clearly observed at 1095 cm-1, within 2 cm-1 of the value

6

measured by ATR on oriented DPPC films. The absorption of the antisymmetric stretching mode

7

(νPO2-asym) falls at 1255 cm-1 in published ATR-FTIR spectra

8

ATR-FTIR measurements (Figure 2), where its intensity is about half that of νPO2-sym. In nano-

9

FTIR spectra (Figure 1) a band is observed at 1244 cm-1. However, it is sharper and weaker than

10

that observed for νPO2-asym in ATR-FTIR spectra of DPPC. Both intensity and bandwidth match

11

those observed in other systems for the ωCH2 wag band expected in that position. Therefore, the

12

spectral contribution of νPO2-asym is absent or minimal in nano-FTIR spectra.

13

The absence of νPO2-asym can be interpreted considering the orientation of the transition moment

14

for this mode, which has been proposed to lay in the plane of the bilayer.24 In the DPPC

15

molecule, the transition moment of νPO2-sym is aligned along the bisector of the PO2 group, while

16

the transition moment of νPO2-asym is aligned perpendicular to it and in the PO2 plane.

17

In the nano-FTIR experiment, the transition moment is perpendicular to that of the tip-enhanced

18

electric field, making the mode inactive in this experimental geometry (Figure 3). We conclude

19

that the polarization of the νPO2-asym transition together with the high degree of orientation of the

20

sample are the main factors that prevent us from seeing the mode. Therefore, the polarization of

21

the spectra indicates that the PO2 plane lies approximately perpendicular to the silicon surface.

24

and at 1243 cm-1 in our own

12 ACS Paragon Plus Environment

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

Analytical Chemistry

1

Madrid and Horswell 27have shown that, for a bilayer of DPPE in the gel phase supported on an

2

Au (1,1,1) electrode surface, νPO2-asym appears as a strong band on top of the wag progression

3

and overlaps with one of the wagging bands, similarly to what we observe in ATR-FTIR spectra.

4

The intensity of νPO2-asym is a function of electrode potential, and is stronger for positive

5

potentials, while it disappears for negative potentials. The spectrum at negative potential is very

6

similar to the nano-FTIR spectrum that we report in Figure 1. The effect has been interpreted as

7

due to increased organization of the headgroups, which align horizontal to the gold surface at

8

negative potential, what is in agreement with our conclusions. This is also consistent with the

9

structure of the oxidized Si surface that supports the phospholipid bilayer in our sample, which is

10

expected to have a negative interfacial charge distribution due to the electron density distribution

11

in the SiO2 layer.

12

Figure S3 shows the nano-FTIR spectrum recorded in the spectral region 1400 -1900 cm-1. In

13

FTIR and FTIR-ATR spectra of macroscopic phospholipid samples this region is dominated by

14

the two strongest absorption bands, the ester carbonyl stretching mode, νCO, around 1740 cm-1,

15

and the methylene scissoring mode, δCH2sci, around 1470 cm-1. In nano-FTIR both appear as

16

weak bands, about 5 to 6 times weaker than phosphate bands. The weakness of these two bands

17

suggests that the transition moments of both δCH2sci and νCO are perpendicular to the electric

18

field of the incident radiation, or at a high angle to the normal. The absorption from water around

19

1640 cm-1 is very weak in these spectra, showing that water content is very low. Order and

20

headgroup interactions are known to be favored by a low degree of hydration of the sample, as

21

revealed by changes in thermotropic properties,28 supporting our conclusions that the sample

22

under investigation is highly ordered.

23 13 ACS Paragon Plus Environment

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

Page 14 of 31

1

One possible orientation that satisfies the observed spectral properties is shown in Figure 3. The

2

proposed orientation is expected to be energetically favorable in a bilayer with low hydration

3

because it allows head-to-tail alignment of the electric dipoles of the headgroups. Figure 3 shows

4

the alkyl chains of the two leaflets extended in the parallel structure. However, our data cannot

5

discriminate between this structure and the alternative one with crossed chains, in which the

6

alkyl chains of the two leaflets form the same angle with the normal but with opposite sign.

7

DPPC for ATR-FTIR experiments has been prepared by drying vesicle suspensions on the

8

surface of the ATR prism. The resulting samples comprise stacks of bilayers that are fragmented

9

and only partially oriented. In contrast to a nano-FTIR measurement, an ATR-FTIR

10

measurement probes the averaged molecular orientation in the whole sample and is unable to

11

appreciate the local order of individual bilayers. Therefore, in ATR-FTIR measurements all

12

bands are observed, including νPO2-asym, even when spectra are collected with perpendicular

13

polarized light.

14

polarization of the beam may still be present because of multiple reflections by instrument

15

optics. Therefore, the observation of the νPO2-asym band is expected. Furthermore, comparison

16

of nano-FTIR and ATR-FTIR spectra shows that bands are sharper in the former, despite the

17

poorer resolution used in nano-FTIR spectra (10 cm-1 versus 8 cm-1). One possible explanation,

18

consistent with other observations in the text, is the higher degree of ordering that can be

19

appreciated when measuring a single nanoscale bilayer, as opposed to a macroscopic stack of

20

bilayers.

21

The description resulting from the analysis of nano-FTIR spectra agrees with our current

22

understanding of the structure of phospholipid bilayers and multibilayers below the gel-to-liquid

23

phase transition temperature. Under these conditions multilamellar suspensions of DPPC exist in

24

In our ATR-FTIR measurements we did not use a polarizer, although some

14 ACS Paragon Plus Environment

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

Analytical Chemistry

1

either the lyotropic liquid crystalline gel phase (Lβ’) or in a range of pseudocrystalline phases

2

(Lc’).29 The latter are characterized by in-plane ordering of both the headgroups and the alkyl

3

chains in two dimensional lattices in supported multibilayers. 30

4 5 6

Figure 3.Proposed orientation of DPPC molecules in a bilayer leaflet relative to the incident

7

electric field E in the proximity of the tip used for a nano-FTIR or sSNOM measurement. The

8

figure also shows the orientation of the transition moments (black arrows for vectors in the plane

9

of the page and dotted circles for vectors perpendicular to the plane of the page) for νPO2-sym,

10

νPO2-asym, δCH2sci, νCO and ωCH2wag relative to the DPPC molecule and the field E.

11

Dimensions of tip and DPPC molecules are not drawn to scale.

12

In addition, spectra from the single bilayer show small differences in band structure. In

13

particular, the νN+(CH3)3

asym

band at 970 cm-1, associated to the choline headgroup, shows 15 ACS Paragon Plus Environment

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

Page 16 of 31

1

splitting into two components of equal intensity, at 970 cm-1 and 982 cm-1, in the single bilayer.

2

In contrast, one of the two components, at 970 cm-1, dominates the spectra of double bilayers.

3

The splitting may be due to the interaction of the lower leaflet with the silicon substrate, which

4

can affect the frequency of the mode. The contribution of this single leaflet can be appreciated in

5

a single bilayer but becomes less obvious when multiple bilayers are present. This is in

6

agreement with recent work by Wu et al.

7

individual DPPC leaflets in bilayers supported on mica. The differences were ascribed to the

8

selective interaction of the lower bilayer with the substrate.

9

We performed IR sSNOM imaging of multibilayer patches by mapping the intensity of the

10

sSNOM signal over the surface of the sample while keeping the emission of the laser fixed at

11

one wavelength. We measured maps at 1070 cm-1, corresponding to the peak of the strong

12

absorption from νC-O-PO2-, and at 1015 cm-1, where only weak absorption can be observed.

13

Figure 4 shows maps of AFM topography, reflectivity, and sSNOM optical phase at the two

14

wavenumbers.

31

, who reported differences in the properties of

15

16 ACS Paragon Plus Environment

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

Analytical Chemistry

1 2

Figure 4. AFM and sSNOM maps of multibilayers of DPPC with single wavelength excitation.

3

A; AFM topography map recorded while exciting at 1070 cm-1(νC-O-PO2- band). B; Reflectivity

4

map at 1070 cm-1, constructed using the amplitude s(ω) of the scattered light. C; Absorption map

5

at 1070 cm-1, constructed using the phase φ(ω) of the scattered light. D; AFM topography map

6

recorded while exciting at 1015 cm-1, where only weak absorption is observed. E; Reflectivity

7

map at 1015 cm-1. F; Absorption map at 1015 cm-1. The yellow circles indicate the location of

8

flattened convex particles. The orange dotted lines indicate the positions of line scans used for

9

Figure 5. The wavenumbers used for on-resonance and off-resonance imaging are marked by

10

green arrows in Figure 2B.

11

The wavenumber of 1070 cm-1 corresponds to the stretching mode of the C-O-P phosphate ester

12

bond and its spatial distribution tracks that of phospholipid headgroups. Figure 4b shows the

13

reflectivity of the sample at this wavenumber, mapped using the amplitude of the scattered light

14

s(ω). The reflectivity increases with the real part of the refractive index of the material, n(ω),

15

giving rise to larger scattering amplitude from reflection at the Si surface (n(ω) ~ 3.4) and lower 17 ACS Paragon Plus Environment

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

Page 18 of 31

1

amplitude from the DPPC bilayers (n(ω) ~ 1.5). Figure 4c maps absorption at 1070 cm-1 using

2

the phase φ(ω) of the scattered light. Large contrast is observed due to phospholipid bilayers,

3

which give rise to strong absorption at this wavelength. The phase change increases with the

4

number of stacked bilayers in a given position, as expected from the thickness dependence of the

5

spectra (Figure 1). The same region is mapped while exciting at 1015 cm-1, corresponding to a

6

wavenumber where only weak absorption is displayed by DPPC. At this wavenumber, the

7

reflectivity map still displays weaker but still strong contrast. This is expected since the

8

difference in n(ω) between silicon and DPPC is still high at this wavenumber. In contrast, weak

9

absorption contrast is observed at 1015 cm-1 (Figure 4f) throughout most of the sample and

10

images at this wavenumber show very few structures. A residual phase shift, in the range of 15

11

mrad, is still observed on the bare Si surface, due to the fact that images retain the phase

12

differences between the maxima but not the absolute phase of the signal. Single bilayers are

13

barely detectable in the 1015 cm-1 images and only thicker sample locations, such as multiple

14

bilayers and round particles, give appreciable contrast.

15

18 ACS Paragon Plus Environment

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

Analytical Chemistry

1

Figure 5. Line profiles across the edge of a bilayer patch, from the images of Figure 4

2

(Corresponding to the orange dotted lines in Figure 4). Black trace: AFM topography profile.

3

Red trace: profile of the phase of the scattered light φ(ω) at 1070 cm-1.

4 5

Figure 5 compares the AFM topography line profile at the edge of a patch with the profile of the

6

phase φ(ω) of the scattered light at 1070 cm-1, showing that the two track closely. The edge

7

profile can be used to provide an estimate for resolution. To provide statistically significant

8

information, we measured the resolution over twenty one edge profiles on the border of one

9

patch. The 90/10 and 80/20 values for resolution (points at 90% - 10% and 80% - 20% of height

10

in the intensity profile) give an average resolution of 51 ± 26 nm and 39 ± 23 nm respectively,

11

much smaller than is allowed by diffraction and comparable to the size of the tip (approximately

12

20 nm diameter). In conclusion, the imaging capabilities of sSNOM of phospholipid bilayers at a

13

selected absorption wavelength are comparable to the ones provided by AFM.

14

It must be noted that assessing spatial resolution by using a topographical feature, such as an

15

edge, carries the risk that cross-talk between the mechanical and optical measurements may

16

affect the resulting resolution value. In our measurements, any such coupling would be mediated

17

by the movement of the tip and would be obvious at the edges of the patches, where the

18

movement of the tip is affected by the change of topography, as evidenced by mechanical phase

19

shifts. For example, the AFM phase map of Figure S1 shows marked edge effects where the

20

phase changes at the edge of each bilayer. However, no edge effects are seen in the map in

21

Figure 4C (the off-resonance map in Figure 4F is too noisy to allow such analysis). The increase

22

of intensity with the number of bilayers observed in Figure 4C also supports the conclusion that

23

image contrast is a function of the amount of material under the tip, and not of topography. We 19 ACS Paragon Plus Environment

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

Page 20 of 31

1

conclude that cross-coupling between the mechanical and optical signal is negligible in these

2

measurements and the resolution estimated from Figure 5 is a useful metric for this class of

3

samples.

4 5

Conclusions

6

We have reported the first detailed IR spectra of a single phospholipid bilayer using a near-field

7

IR technique. We have investigated the spectral region between 800 cm-1 and 1400 cm-1, which

8

includes bands from both the phospholipid headgroups and their acyl chains. Nano-FTIR spectra

9

provided the same bands as in the IR absorption spectra of macroscopic samples with similar

10

relative intensities. In addition to band position, the relative intensity of the bands is also fully

11

consistent with their polarization properties and their macroscopic ATR-FTIR absorption. The

12

observation validates the hypothesis that for a thin film of weak oscillators the nano-FTIR

13

absorbance, an, is proportional to the absorbance measured in a transmission experiment.17

14

The average spatial resolution of the measurements is better than 40 nm, consistent with the

15

dimensions of the tip. Overall, the quality of the spectra and the anisotropy displayed by some

16

absorption bands from headgroups and alkyl chains indicate that the DPPC bilayers are highly

17

ordered and oriented. Moreover, the measurements of the wagging progression indicates that

18

nano-FTIR measurements can be used to probe nanoscopic changes of order within the bilayers,

19

such as following a phase transition or the interaction with a protein or exogenous molecule. The

20

measurements rely on the intrinsic optical properties of the sample and do not require the

21

introduction of extrinsic chromophores. Therefore, the molecular properties of the system can be

22

studied without any perturbations associated with labeling. 20 ACS Paragon Plus Environment

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

Analytical Chemistry

1

To the best of our knowledge this is the first time that detailed spectra of single phospholipid

2

bilayers have been obtained by sSNOM and nano-FTIR. Work by Amenabar et al.

3

high-quality spectra of the protein component of single membrane layers. However, the intensity

4

of the signal from membrane phospholipids is about one order of magnitude weaker than that

5

from proteins, and it was not observed by these researchers.

6

The portion of the bilayer probed by the sSNOM experiment corresponds to about 500 nm2. The

7

footprint of a DPPC molecule in the bilayer plane is approximately 0.5 nm2. Therefore the signal

8

for a single bilayer arises from about 2000 (1000 per leaflet) DPPC molecules. Despite the small

9

size of the sample, the quality of the spectra allows detailed molecular studies. It is remarkable

10

that such spatially confined measurements provide such an accurate description of molecular

11

order within a monolayer. The high resolution has to be credited for this advantage, because it

12

allows selecting specific nano-sized regions or domains of the sample. In contrast, macroscopic

13

measurements deliver only the average contribution from all regions of the sample. It has already

14

been pointed out that such averaging may be the reason why the polarized ATR-FTIR spectra of

15

macroscopic samples of phospholipids multi-bilayers show such variability in literature reports.8

16

It is expected that the studies of the biological membranes for which phospholipid bilayers are

17

models will also be possible in the near future. The major challenge will be, however, the

18

development of sample holders that will allow performing sSNOM experiments on samples of

19

biological relevance under controlled environmental conditions, such as bilayers in contact with

20

an aqueous phase or living cells.

20

provided

21 ACS Paragon Plus Environment

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

Page 22 of 31

1 2

Methods

3

1,2-Dipalmitoyl-sn-glycero-3-phosphocholine (DPPC) was purchased from SIGMA-Aldrich

4

(Poland); chloroform (99%, stabilized with ethanol) was purchased from Avantor Performance

5

Materials (Poland). Deionized water was produced using a MilliQ deionizer at 18.2 mΩ cm

6

resistivity. Si wafers (P-Type, Boron doped) were purchased from ON Semiconductor (Czech

7

Republic).

8

A modification of an existing procedure

9

dissolved in chloroform to a concentration of 1 mg/ml and brought to dryness under a flow of

10

gas. The DPPC film was suspended in water to a 1 mg/ml concentration by vortex mixing until a

11

cloudy dispersion was obtained. The dispersion was subjected to five freeze-thaw cycles between

12

liquid nitrogen temperature and 40 - 50 °C. Finally, the suspension was sonicated in a bath for ~

13

15 min at a temperature of approximately 45 °C. The liposome suspension was stored at 4 °C.

14

Supported bilayers and multi-bilayers were prepared by depositing a drop of the liposome

15

suspension on a Si wafer. The wafers were previously cleaned by immersion for 30 min in a

16

piranha solution (50/50 v/v concentrated sulfuric acid and 30% hydrogen peroxide solution in

17

water). Piranha solution is hazardous and should be handled with extreme care. Wafers were

18

extensively washed with deionized water and stored under a layer of deionized water. The

19

suspension was allowed to settle for about 1 h and drained by capillary action towards a paper

20

towel touching the side of the wafer.

21

IR s-SNOM and nano-FTIR measurements were performed at room temperature with samples

22

exposed to the atmosphere using a neaSNOM system (neaspec GmbH, Munich) described

32

was used for liposome preparation. DPPC was

22 ACS Paragon Plus Environment

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

Analytical Chemistry

1

elsewhere.17 For nano-FTIR measurements, a PtIr coated AFM tip (20 nm diameter; 42 N/m

2

force constant) was illuminated using the emission of a coherent, broadband beam from a

3

Difference Frequency Generation (DFG) laser source (Toptica) with 100 fs pulses. The power of

4

the beam was set at 400 µW covering the 700 - 1500 cm-1 (6.7 µm – 14 µm) spectral range. The

5

tip was used to probe the sample by operating in tapping mode at the frequency Ω of ca. 250

6

kHz, thus modulating the intensity of the scattered light at Ω and its higher harmonics. The light

7

backscattered from the tip is analyzed with an asymmetric Michelson interferometer, where the

8

sample and tip are positioned in one arm of the interferometer. A liquid nitrogen-cooled mercury

9

cadmium telluride (MCT) detector was used to measure the intensity of the scattered light.

10

Spectral resolution was set at 10 cm-1 with eight levels of zero-filling. Demodulation of the

11

scattered light signal at a higher harmonic nΩ of Ω was used to separate the contribution of the

12

near-field signal from that of the background and to control the probing depth of the light.

13

Fourier Transform of the interferogram from the beam demodulated at the n-th harmonic nΩ

14

provides the complex spectra of the scattered electric field En(ω). Near-field spectra of DPPC

15

were normalized by dividing 〈En(ω) 2〉 by a reference spectrum 〈En,ref(ω) 2〉 recorded and

16

demodulated at the same nΩ value on a clean portion of the Si substrate (See Supporting

17

Information for a definition of the optical quantities). The normalization corrects for the change

18

in intensity of the laser through the spectral region and between experiments (similarly to what is

19

done when calculating classical IR absorption spectra). The normalization ensures that values of

20

nano-FTIR absorbance are comparable throughout the spectral region and between the

21

measurements. Noise levels are quantified by dividing two references measured sequentially in

22

the same position of the Si surface. For this purpose we used the spectral range between 1400

23

and 2000 cm-1, because of the absence of absorption bands from silicon oxide. Noise levels of

17

23 ACS Paragon Plus Environment

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

Page 24 of 31

1

0.007 RMS were measured over the time used to measure the spectra (about 10 min). For the

2

signal intensity value we used the intensity of the strongest bands in Figure 1, corresponding to

3

0.05 and 0.07 au for the single and double bilayer respectively .

4

sSNOM imaging is performed using a pseudoheterodyne interferometric detection scheme 33 and

5

a tunable QCL laser (Mircat, Daylight Solutions, US) while collecting an AFM topography and

6

stiffness information, using the same tip as for nano-FTIR experiments. The phase φn(ω) of the

7

scattered light at the frequency ωn, demodulated at the n-th order, is mapped as a function of tip

8

position on the sample.

9

Information), this provides a map of absorption at the frequency ω superimposed to the

10

topography map. Mapping of the scattered light amplitude sn(ω) at the same wavelength provides

11

a map of sample reflectivity at ω.

12

ATR-FTIR measurements were performed on a Thermo Nicolet iS10 spectrometer using a Smart

13

iTX diamond ATR accessory, without polarizers. The interferometer was setup for

14

measurements in the mid-IR region using a glowbar source, a liquid nitrogen cooled MCT

15

detector and a KBr-supported beamsplitter. Measurements were performed at 8 cm-1 resolution

16

and the Fourier Transform was executed with a Blackmann-Harris 3-term apodization function,

17

two levels of zero filling and a Mertz phase correction algorithm. A suspension of DPPC

18

liposomes, as prepared for sSNOM measurements, was deposited on the diamond ATR crystal

19

and allowed to dry in air. The ATR spectrum was measured soon after the disappearance of

20

absorption bands from bulk water.

21

Spectral data were analyzed using the program OPUS 7.5 (Bruker Optics) and plotted using the

22

program Origin Pro (Origin Lab, US). AFM and sSNOM Images were processed and plotted

23

using the free program Gwyddion 2.50 (http://gwyddion.net/). sSNOM images were corrected

Since the phase shift φn(ω) is related to absorption (Supporting

24 ACS Paragon Plus Environment

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

Analytical Chemistry

1

for horizontal strokes, row alignment, and row level. All images were corrected to bring the

2

minimum value to zero.

3

Supporting Information

4

The Supporting Information is available free of charge on the ACS Publications website. The

5

following files are available free of charge.

6

Definition of Quantities and Relationships: AFM Height and Phase Maps of Sample; nano-FTIR

7

Spectrum of 1400 -1900 cm-1 Region; Approach Curve on Silicon (PDF).

8 9

Acknowledgments and Funding

10

The authors are grateful to Dr. Karol Wolski, Jagiellonian University, for help with FTIR

11

instrumentation, and to Dr. Magdalena Wytrwał-Sarna, AGH University of Science and

12

Technology, for laboratory access.

13

This project has received funding from the European Union’s Horizon 2020 research and

14

innovation program under the Marie Skłodowska-Curie grant agreement No. 665778, managed

15

by the National Science Center Poland under POLONEZ contract 2016/21/P/ST4/01321 (L.Q.)

16 17

Competing financial interests

18

Adrian Cernescu is an employee of neaspec GmbH, which produces the instruments for nano

19

FTIR and IR sSNOM used in this work. The other authors declare no competing financial

20

interests.

25 ACS Paragon Plus Environment

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

Page 26 of 31

1 2

References

3

(1)

Nicolson, G. L. The Fluid - Mosaic Model of Membrane Structure: Still Relevant to

4

Understanding the Structure, Function and Dynamics of Biological Membranes after More

5

than 40 Years. Biochim. Biophys. Acta - Biomembr. 2014, 1838 (6), 1451–1466.

6

(2)

7 8

Membranes. Science. 1972, 175 (4023), 720–731. (3)

9

11411–11416. (4)

12 13

(5)

Dluhy, R. A. Infrared Spectroscopy of Biophysical Monomolecular Films at Interfaces : Theory and Applications. Appl. Spectrosc. Rev. 2000, 35 (4), 315–351.

(6)

16 17

Quaroni, L.; Zlateva, T. Infrared Spectromicroscopy of Biochemistry in Functional Single Cells. Analyst. 2011, 136 (16), 3219.

14 15

Levental, I.; Grzybek, M.; Simons, K. Raft Domains of Variable Properties and Compositions in Plasma Membrane Vesicles. Proc. Natl. Acad. Sci. 2011, 108 (28),

10 11

Singer, S. J. J.; Nicolson, G. L. L. The Fluid Mosaic Model of the Structure of Cell

Tamm, L. K.; Tatulian, S. A. Infrared Spectroscopy of Proteins and Peptides in Lipid Bilayers. Q. Rev. Biophys. 1997, 30 (4), 365–429.

(7)

Tatulian, S. A. Current Topics Attenuated Total Reflection Fourier Transform Infrared

18

Spectroscopy : A Method of Choice for Studying Membrane Proteins and Lipids †.

19

Biochemistry 2003, 42, 11898–11907.

20 21

(8)

Quaroni, L.; Zlateva, T.; Bedolla, D.; Massaro, S.; Torre, V. Measurement of Molecular Orientation in a Subcellular Compartment by Synchrotron Infrared Spectromicroscopy. 26 ACS Paragon Plus Environment

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

Analytical Chemistry

1 2

ChemPhysChem. 2008, 9 (10), 1380–1382. (9)

Dluhy, R. A. Quantitative External Reflection Infrared Spectroscopic Analysis of

3

Insoluble Monolayers Spread at the Air-Water Interface. J. Phys. Chem. 1986, 90 (7),

4

1373–1379.

5

(10)

6 7

Established Technique. Annu. Rev. Biophys. Biomol. Struct. 2007, 36 (1), 151–169. (11)

8 9

Kwiatek, J. M.; Hinde, E.; Gaus, K. Microscopy Approaches to Investigate Protein Dynamics and Lipid Organization. Mol. Membr. Biol. 2014, 31 (5), 141–151.

(12)

10 11

Haustein, E.; Schwille, P. Fluorescence Correlation Spectroscopy: Novel Variations of an

Centrone, A. Infrared Imaging and Spectroscopy Beyond the Diffraction Limit. Annu. Rev. Anal. Chem. 2015, 8 (1), 101–126.

(13)

Keilmann, F.; Hillenbrand, R. Near-Field Nanoscopy by Elastic Light Scattering from a

12

Tip. In Nano-Optics and Near-Field Optical Microscopy; Zayats, A., Richard, D., Eds.;

13

ArtechHouse, Boston, 2009; pp 235–265.

14

(14)

Wiens, R.; Findlay, C. R.; Baldwin, S. G.; Kreplak, L.; Lee, J. M.; Veres, S. P.; Gough, K.

15

M. High Spatial Resolution (1.1 Μm and 20 Nm) FTIR Polarization Contrast Imaging

16

Reveals Pre-Rupture Disorder in Damaged Tendon. Faraday Discuss. 2016, pp 555–573.

17

(15)

18 19

Knoll, B.; Keilmann, F. Near-Field Probing of Vibrational Absorption for Chemical Microscopy. Nature. 1999, 399 (May), 7–10.

(16)

Westermeier, C.; Cernescu, A.; Amarie, S.; Liewald, C.; Keilmann, F.; Nickel, B. Sub-

20

Micron Phase Coexistence in Small-Molecule Organic Thin Films Revealed by Infrared

21

Nano-Imaging. Nat. Commun. 2014, 5 (May), 1–6. 27 ACS Paragon Plus Environment

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

1

(17)

Huth, F.; Govyadinov, A.; Amarie, S.; Nuansing, W.; Keilmann, F.; Hillenbrand, R.

2

Nano-FTIR Absorption Spectroscopy of Molecular Fingerprints at 20 Nm Spatial

3

Resolution. Nano Lett. 2012, 12 (8), 3973–3978.

4

(18)

Amenabar, I.; Poly, S.; Goikoetxea, M.; Nuansing, W.; Lasch, P.; Hillenbrand, R.

5

Hyperspectral Infrared Nanoimaging of Organic Samples Based on Fourier Transform

6

Infrared Nanospectroscopy. Nat. Commun. 2017, 8, 14402.

7

(19)

Donaldson, P. M.; Kelley, C. S.; Frogley, M. D.; Filik, J.; Wehbe, K.; Cinque, G.

8

Broadband near-Field Infrared Spectromicroscopy Using Photothermal Probes and

9

Synchrotron Radiation. Opt. Express. 2016, 24 (3), 1852 - 1864.

10

(20)

Page 28 of 31

Amenabar, I.; Poly, S.; Nuansing, W.; Hubrich, E. H.; Govyadinov, A. A.; Huth, F.;

11

Krutokhvostov, R.; Zhang, L.; Knez, M.; Heberle, J.; et al. Structural Analysis and

12

Mapping of Individual Protein Complexes by Infrared Nanospectroscopy. Nat. Commun.

13

2013, 4, 1–9.

14

(21)

Wollny, G.; Bründermann, E.; Arsov, Z.; Quaroni, L.; Havenith, M. Nanoscale Depth

15

Resolution in Scanning near-Field Infrared Microscopy. Opt. Express. 2008, 16 (10),

16

7453–7459.

17

(22)

18 19

Biltonen, R. L.; Lichtenberg, D. The Use of Differential Scanning Calorimetry as a Tool to Characterize Liposome Preparations. Chem. Phys. Lipids. 1993, 64 (1–3), 129–142.

(23)

Naumann, C.; Brumm, T.; Bayerl, T. M. Phase Transition Behavior of Single

20

Phosphatidylcholine Bilayers on a Solid Spherical Support Studied by DSC, NMR and

21

FT-IR. Biophys. J. 1992, 63 (5), 1314–1319.

28 ACS Paragon Plus Environment

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

1

Analytical Chemistry

(24)

Hubner, W.; Mantsch, H. H. Orientation of Specifically 13C=O Labeled

2

Phosphatidylcholine Multilayers from Polarized Attenuated Total Reflection FT-IR

3

Spectroscopy. Biophys. Jorunal. 1991, 59, 1261–1272.

4

(25)

Senak, L.; Moore, D.; Mendelsobn, R. CH2 Wagging Progressions as IR Probes of

5

Slightly Disordered Phospholipid Acyl Chain States. J. Phys. Chem. 1992, 96 (15), 2749–

6

2754.

7

(26)

8 9

Gokus, T.; Huth, F.; Huber, A. Introducing Nano-FTIR – Imaging and Spectroscopy at 10nm Spatial Resolution. Microsc. Microanal. 2015, 21 (934), 1871–1872.

(27)

Madrid, E.; Horswell, S. L. Effect of Headgroup on the Physicochemical Properties of

10

Phospholipid Bilayers in Electric Fields: Size Matters. Langmuir. 2013, 29 (5) pp:1695-

11

1708.

12

(28)

Lewis, R.; McElhaney, R. The Mesomorphic Phase Behavior of Lipid Bilayers. In The

13

Structure of Biological Membranes, Second Edition; Yeagle, P. L., Ed.; CRC Press, 2004;

14

pp 53–107.

15

(29)

Lewis, R. N. A. H.; McElhaney, R. N. Membrane Lipid Phase Transitions and Phase

16

Organization Studied by Fourier Transform Infrared Spectroscopy. Biochim. Biophys.

17

Acta - Biomembr. 2013, 1828 (10), 2347–2358.

18

(30)

19 20 21

Raghunathan, V. A.; Katsaras, J. Structure of the Lc′ Phase in a Hydrated Lipid Multilamellar System. Phys. Rev. Lett. 1995, 74 (22), 4456–4459.

(31)

Wu, H.-L.; Tong, Y.; Peng, Q.; Li, N.; Ye, S. Phase Transition Behaviors of the Supported DPPC Bilayer Investigated by Sum Frequency Generation (SFG) Vibrational

29 ACS Paragon Plus Environment

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

Page 30 of 31

1

Spectroscopy and Atomic Force Microscopy (AFM). Phys. Chem. Chem. Phys. 2016, 18

2

(3), 1411–1421.

3

(32)

Kepczyński, M.; Nawalany, K.; Jachimska, B.; Romek, M.; Nowakowska, M. Pegylated

4

Tetraarylporphyrin Entrapped in Liposomal Membranes. A Possible Novel Drug-Carrier

5

System for Photodynamic Therapy. Colloids Surfaces B Biointerfaces. 2006, 49 (1), 22–

6

30.

7 8

(33)

Ocelic, N.; Huber, A.; Hillenbrand, R. Pseudoheterodyne Detection for Background-Free near-Field Spectroscopy. Appl. Phys. Lett. 2006, 89 (10), 28–31.

9 10

30 ACS Paragon Plus Environment

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

1

Analytical Chemistry

TOC Figure

2 3

31 ACS Paragon Plus Environment