Analyst

PAPER

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article OnlineView Journal | View Issue

aFaculty of Chemistry, Jagiellonian Universi

E-mail: birczyns@chemia.uj.edu.pl; Fax: +4bDepartment of Infectious Diseases, Jag

Sniadeckich 5, 31-501 Krakow, PolandcStudents' Scientic Society, Jagiellonian Un

Krakow, Poland

Cite this: Analyst, 2013, 138, 7157

Received 6th August 2013Accepted 26th September 2013

DOI: 10.1039/c3an01493b

www.rsc.org/analyst

This journal is ª The Royal Society of

Raman micro-spectroscopy tracing human lymphocyteactivation

A. Wesełucha-Birczynska,*a M. Kozicki,a J. Czepielb and M. Birczynskac

The activation of lymphocytes occurs when they are exposed to viruses or other foreign antigens. The aim

of this work was to verify if Raman spectroscopy can be used to screen the activation of lymphocytes during

viral infection. There are distinct peaks that reveal differences between activated and intact cells. The most

important marker of the lymphocyte activation process is the prominent 521 cm�1 disulfide band which

marks the immunoglobulin formation. The up shift of the S–S mode from the broad band centered at

510 cm�1 of human normal immunoglobulin to a single band at 521 cm�1 of human B cells indicates a

selection of the optimal geometry of the disulphide bridges to bind to a foreign antigen. Polarization

data is used to detect the structural alteration between domain fragments. Differences in other band

intensities may be due to different protein compositions in both the investigated forms. B cell activation

causes the change of the intracellular cytoplasm composition due to the secretion of immunoglobulins

during the fighting of the infection.

1. Introduction

The use of Raman spectroscopy to investigate biologicalmolecular structures and the recognition of their particularfunctional groups has been growing rapidly since the 1970s.1,2

In recent years the use of Raman spectroscopy has expandedtowards the cellular level. The advantage of this method is thefact that there is no need to use any labeling or staining whichmakes the sample preparation easier and the spectra free of anyadditional disturbances.3,4 Therefore, while investigatinghuman white blood cells we aimed to decode the features of theinammatory response.

Lymphocytes, which constitute up to 40 percent of all whiteblood cells, are an important part of the human immune systemdetermining the line of defense to infectious microorganisms.5

The activation of lymphocytes occurs when they are exposed toviruses or other foreign antigens.6 In recent years there has beensignicant progress in the understanding of this process.7 Therst assumptions rely on the idea that B cells might mainlyrecognize antigens on the surfaces of other cells, which isbelieved to reect a physiological means of antigen-induced Bcell activation. Batista et al.8 analyzed, using an electronmicroscopy method and articial bilayers, the specicity of Bcell receptors which activate B cells resulting in the gathering ofantigens for subsequent presentation to the T cells. B-cell

ty, Ingardena 3, 30-060 Krakow, Poland.

8 126340515; Tel: +48 126632067

iellonian University, Medical College,

iversity, Medical College, Anny 12, 31-008

Chemistry 2013

interactions with antigens, that are immobilized on the surfaceof target cells, lead to the formation of a synapse and theacquisition, of membrane-integral antigens from the target.8

Antigens are usually encountered in vivo in a membrane-anchored form, bound by Fc or complementary receptors.9 It isaccepted that the immunological synapse formation andantigen acquisition will probably enhance the activation of Bcells, even at low antigen concentration. An immunologicalsynapse is also recognized in helper T-cells, natural killer NK-cells and cytotoxic T-lymphocytes.10–13

As a response to increased lymphocyte activation a signi-cant change in their behavior and shape is detected. Some ofthe stimulated B lymphocyte cells become active in theproduction of antibodies against foreign antigens, transformedinto plasma cells.

Recent advances in microspectroscopy have contributed toa new understanding of the molecular environment forlymphocyte activation.14 It is worth noting the use of three-dimensional confocal microscopy (3D-CM) in the explanationof the immunological synapse function that inuences the Bcell acquisition of membrane-attached antigens.8 An inter-esting strategy of the microuidic system was to design andquantify the biological T cell adhesion to the endothelial cellsby means of cell adhesion molecules that are important tomaintain not only the function, but also the migration of theleukocytes in the immune system.15 Vibrational spectroscopy,infrared and Raman microspectroscopy methods have greatlycontributed to a new perspective on human cells, tissues, andthe opportunity to observe physiological and pathologicalchanges. Diem et al.16 note that the spectra obtained from alymph node, removed during colon cancer surgery, from the

Analyst, 2013, 138, 7157–7163 | 7157

Analyst Paper

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article Online

lymphoid follicles and cortical areas may be differentiated dueto B and T cell content. However, for individual cells, suchdissimilarities within the lymphoid system have not beenreported. Successively, the structural information from theresolved IR microspectroscopy of individual cells, althoughrevealing only small spectral differences between distinctregions of a cell, presents considerable potential as a rapidscreening method.17 Raman spectroscopy, introducing reso-nance methods and using certain chromophores, can bedirected to denite chemical species, and may even beconsidered as a diagnostic method.18

The aim of this work is to verify if Raman spectroscopycan be used to screen lymphocytic activity during viral infection(the u).

Fig. 1 (A) A photomicrograph of the thin blood film with the investigated intacthuman lymphocyte (reflected light; objective 100�); (B) intact B lymphocyte withthe marked region over which the map experiment was performed; (C) Ramanspectrum in the range 1750–150 cm�1; (D) distribution of immunoglobulins inthe lymphocyte (521 cm�1 marker band); excitation laser line 785 nm.

2. Materials and methods2.1. Blood samples

The blood from healthy volunteers (and the authors) wasobtained by venipuncture, placed in tubes containing heparinas an anti-coagulant, and kept at 4 �C. Then it was deposited asa blood lm so it could be measured. The cells were allowed tosettle for about ten minutes prior to measurement. The blood ofthe same donors was also collected during the initial phase ofthe viral infection (u).

A human normal immunoglobulin solution for the infusion(100 mg ml�1) and dried powder, of purity of at least 98% IgG,were used as a reference (Baxter AG, Austria).

The studies were conducted in accordance with the guide-lines for good clinical practice (GCP) according to the EthicalPrinciples for Medical Research Involving Human Subjects(Declaration of Helsinki). The study was approved by the localethics committee.

Fig. 2 (A) A photomicrograph of the thin blood film with the activated humanlymphocyte in the center (reflected light; objective 100�); (B) activated Blymphocyte with the marked region over which the map experiment was per-formed; (C) Raman spectrum in the range 1750–150 cm�1; (D) distribution ofimmunoglobulins in the lymphocyte (map created from 521 cm�1 marker band);excitation laser line 785 nm.

2.2. Raman microspectroscopy

A Renishaw inVia spectrometer, working in confocal mode,connected to a Leica microscope, was used for the measure-ments of the Raman spectra. The beam from a 785 nm HPNIR (high power near IR) diode laser was focused by 100�magnication, a high numerical aperture (NA ¼ 0.9) top-classLeica objective for standard applications. Raman light wasdispersed by a diffraction grating with 1200 grooves per mm.The polarization spectra were obtained for samples excitedwith the 514.5 nm line of the Ar+ laser, using a diffractiongrating with 2400 grooves per mm. The laser power was keptlow, c.a. 1–3 mW at the sample, to ensure a minimuminvasion of cells. The laser power was measured at the tubeoutput. However, the desired power was set using WiREspectrometer soware by introducing appropriate lters intothe laser beam.

During the point mapping experiments, the sample wasshied on a motorized stage (Prior Scientic) at xed intervalswithin the dened area Fig. 1B and 2B for the intact lymphocyte(47 spectra, 2-dimensional characterization, in x and y directionsteps equal to 2 and 1 mm, respectively) and plasma cell (69spectra, 2-dimensional characterization, in x and y direction

7158 | Analyst, 2013, 138, 7157–7163

steps equal to 1 mm), respectively. Polarization spectra werecollected within the same lymphocyte (9 spectra, 2-dimensionalcharacterization, in x and y direction steps equal to 2 mm and1.6 mm, respectively). The data acquisition times per one pointwere equal to 20 and 10 minutes for 785 and 514.5 nm,respectively.

For the mapping experiments, the factory supplied sowarewas used (Renishaw, WIRE v. 2.0 and 3.2). The Raman imagesare based on Raman intensities (Fig. 3A and B), integratedRaman intensities (Fig. 1D, 2D, 6A and B) or peak to baselineintensities (Fig. 6C, D and 5).

This journal is ª The Royal Society of Chemistry 2013

Fig. 3 The distribution of the background intensity evaluated at 650 cm�1 for:(A) activated and (B) intact human lymphocyte; excitation laser line 785 nm.

Paper Analyst

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article Online

3. Results and discussion

The aim of our work is to inspect how the confocal Ramanmicrospectroscopy technique can detect lymphocyte cell acti-vation in standard thin blood lm samples. The obtainedRaman spectra have features which can enable the observationof the human lymphocyte activation process. (Fig. 1 and 2).19,20

Lymphocyte activation occurs when it is exposed to a foreignsubstance or microorganism (antigen). Some activated Blymphocytes differentiate into plasma cells that produce anti-body molecules. Plasma cells have a different shape and theirRaman spectra differ from that of intact lymphocytes (fromhealthy donors). Antibodies (immunoglobulins) produced bythem bind to the target antigens and initiate their neutraliza-tion or destruction.

The rst observed characteristic in the Raman spectra of aninfected lymphocyte is the increased intensity of the back-ground shown in Fig. 3A. The background intensity level is aclear indicator of the spread of the activation within the cell.Additionally, there are distinct peaks observed that revealdifferences between the activated and intact cells (Table 1).

The indicative marker of the lymphocyte activation process,is a prominent 521 cm�1 disulde band which marks theimmunoglobulin formation (Fig. 2C). Although immunoglob-ulins can differ structurally, they are all built from the samebasic units. All immunoglobulins have a four chain structure astheir basis. They are composed of two identical light chains (L)

Table 1 Assignments of significant Raman bands observed for a human lymphocy

Raman bands [cm�1] Assignment1–3,6,7,26–29,34

1662 Amide I1554 Amide II1450 Proteins and lipids, CH2

1339 Nucleic acids and protei1252 Amide III1100 PO2

� phosphate backbo1031 Phe, C–H in-plane stretc1003 Phe, ring breathing mod897 Ring breathing mode855/835 Tyr, exposed/buried762 CH2 rocking mode750 Nucleic acids, O–P–O sy621, 665, 695 C–S stretching mode458–567 S–S bridge551 CL domain S–S bridge521 VL domain S–S bridge510 Interchain S–S bridge

This journal is ª The Royal Society of Chemistry 2013

and two identical heavy chains (H). The heavy and light chainsand the two heavy chains are held together by inter-chaindisulde bonds and by non-covalent interactions. Within eachof the polypeptide chains there are also intra-chain disuldebonds.6 High resolution X-ray crystal structures have beendetermined for various S–S bonded protein fragments.21,22 Theimmunoglobulin modular structure may be diversied byaltering the association of the VH and VL domains.23 Predictingthe relative orientations of the variable heavy and light chainsmay be used in crystal structure based drug design.24 Disuldebonds are one of the characteristic types of bond used in Ramanspectroscopy to monitor the conformation of a protein and itsstability.25,26 Raman bands that arise from C–S and S–Sstretching modes are usually well dened and appear in aregion of the Raman spectrum that is relatively free from otherintense bands.1

Raman spectra of human normal immunoglobulin IgG arepresented in Fig. 4A and B in a dried powder form and insolution, respectively. Both of the spectra are very similar,although they differ mainly in the S–S stretching mode region.IgG in solution shows a rather broad band centered at510 cm�1, while for the powder sample the highest peakappeared at 490 cm�1. The broad band centered at 510 cm�1

indicates a disulde linkage characteristic for the hingeregion.27 However, within this relatively broad band envelopethere are also all other S–S modes characteristic for humannormal immunoglobulin IgG contained, depending on theirinternal rotation about the C–S and C–C bonds of C–C–S–S–C–Cin solution, so without constraints. There are two views on theS–S stretching frequency, the rst one is correlated with thedihedral angle, as described by Wart et al.,28 while the secondshows its dependence on the torsional angles in C–C–S–S–C–Caccording to Sugeta et al.29 The intensity ratio of the 510 cm�1

band to the 1450 cm�1, band typical for aliphatic protein sidechain residues, is equal to 1.18 and 0.65 for the solution anddried form, respectively. The ratio of the intensities indicates achange in the local geometry of the disulde bridges, the higher

te

–36 In activated form

Y

deformation modesns, CH deformation

Yne vibration (DNA marker) Yhing Ye

mmetric stretching

[

Analyst, 2013, 138, 7157–7163 | 7159

Fig. 4 Raman spectrum of human normal immunoglobulin in the range 1750–300 cm�1: (A) dried powder; (B) solution for infusion; excitation laser line 785 nm.In the inset there is a curve fitted S–S stretching vibration region to constituentbands presenting different disulphide bridge conformations.

Fig. 5 Raman spectrum of: (A) human normal immunoglobulin solution forinfusion (Raman intensity scale 0–300); human lymphocyte cell (B) acquisition 7from Fig. 7 (Raman intensity scale 0–2000); (C) acquisition 12 from Fig. 7 (Ramanintensity scale 0–2000); (D) acquisition 42 from Fig. 2 (Raman intensity scale 0–3000); excitation laser line 785 nm, range 850–300 cm�1.

Analyst Paper

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article Online

ratio demonstrates a lack of constraints in the solution. Theweak band at 693 cm�1, observed for the solution sample andassigned to the C–S stretching mode, points towards the inter-chain region.27 The intensity ratio of the 510 cm�1 band to the695 cm�1 C–S stretching band in solution is equal to about 4.6,while the 490 cm�1 band to 647 cm�1 in dried form is equal to 4.The strength of the S–S in the Raman spectra and also thepresence of the C–S stretches allows for this assignment to thedisulde bonds. It is worth paying attention to the position ofthe amide I and III bands at 1672 and 1240 cm�1, respectively,which indicate that immunoglobulin is predominantlycomposed of an antiparallel b-sheet conformation.1,2 The amideII band appears at 1554 cm�1. The band at 897 cm�1 isconsidered to be conformationally sensitive. The half-width ofthe 1672 cm�1 band slightly increases for the sample in thesolution as it also includes water H–O–H bending vibrations.Another important band is the 855 cm�1 peak which is due tothe buried tyrosine residues in the proteins which is observedfor the solution. In the powder form tyrosine is partly exposed.These bands with their assignments are collected in Table 1.

The disulde 521 cm�1 band is not present in an intacthuman lymphocyte, as is shown in Fig. 1C. The immunoglob-ulin distribution for intact and activated lymphocytes is pre-sented in Fig. 1D and 2D, respectively. The behavior of this bandimplies activation marked by an enormous increase of the

7160 | Analyst, 2013, 138, 7157–7163

521 cm�1 to 1450 cm�1 intensity ratio from a close to zero valuefor the intact form to almost 1 for the activated lymphocyte(Fig. 2C). The intriguing value close to zero results from the factthat a single lymphocyte cell contains not only the immuno-globulins. Fig. 1D was intentionally scaled to the peak intensityin the activated lymphocyte (Fig. 2D), so it may seem that thegure shows only the noise. The disulde band position at521 cm�1 indicates, according to normal coordination calcula-tions, a variable domain VL intrachain S–S bridge .27,30

Fig. 5 shows the process of increasing the amount of one typeof S–S conformation (in immunoglobulins), which is taken on bythe activated lymphocytes. A Raman band at 521 cm�1 in a singlecell is selected, from a variety of conformations observed in asolution (Fig. 5A), and then enhanced according to theincreasing IgG secretion by the activated lymphocytes(Fig. 5B–D). The intensity ratio of the 510 cm�1 band to 695 cm�1,for a single cell, increases from 1.9 (Fig. 5B), by 2.1 (Fig. 5C) until6.8 (Fig. 5D). While the intensity of a Raman band depends onthe variation of bond polarizability, when the molecule vibrates,the intensity is affected by the environment.31 The S–S bondconformation (in IgG), characterized by the 521 cm�1 band,expresses the fact of immunoglobulins activation and theproduction of monoclonal antibodies in large quantities.32

Fig. 6 shows the polarization data for a recently activatedlymphocyte excited by the 514.5 nm laser line. This laser linewas chosen because of the laser spot size, which is smaller forthis wavelength than for the 785 nm. The depolarization ratio ofsome of the bands was measured in order to resolve them.2 Thebiochemical changes within the lymphocyte towards its activa-tion is marked by the appearance of a strong S–S band in point1, which shows a depolarization ratio of r1 ¼ 0.15. The depo-larization ratio was determined in accordance with the

This journal is ª The Royal Society of Chemistry 2013

Fig. 6 Polarization Raman spectra for a just being activated human lymphocyte, || andt describe the direction of the scattered light as parallel and perpendicular tothe polarization of the incident laser light, respectively; excitation laser line 514.5 nm.

Fig. 7 Polarization Raman spectra (parallel to thepolarizationof the incident laserlight) for an activated human lymphocyte, excitation laser lines 785 and 514.5 nm.

Paper Analyst

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article Online

denition as r ¼ It/I||, where I|| and It are the intensity ofscattered light parallel and perpendicular to the polarization ofthe incident laser light, respectively.31 The spectra in points 1–3are presented in the whole range to indicate the changes in theintensites of the marker bands in relation to the other vibra-tions represented by other cell components. The marker bandintensity in the neighboring point 2 is 10 times smaller incomparison to point 1, while it is almost invisible in point 3(Fig. 4). However, for points 2 and 3, the inverse polarizationeffect is observed (r2 ¼ 2.2 and r3 / N). The phenomenon ofthe inverse polarization was observed by Spiro et al.33 for someprominent bands in the Resonance Raman spectra of hemo-globin and cytochrome c. The position in which the S–S bandis observed in point 3 for the perpendicular polarization is551 cm�1, characteristic for the CL domain.27 It seems that thisdata reects variations of the binding site within a basic,structural motif. This structural variation thus contributes tomodications in the recognition properties of the antibody.

Fig. 7 shows the distribution of the immunoglobulins in theactivated lymphocyte, created from the 521 cm�1 band,measured using both laser lines: 785 and 514.5 nm. Thisexperiment illustrates that the image of the cell activation isanalogous in both cases, different cell components do notdisturb the picture of the immunoglobulin distribution.

Once stimulated by binding to a foreign antigen, such as avirus in our case, a lymphocyte is getting ready to multiply into a

This journal is ª The Royal Society of Chemistry 2013

clone of identical cells.27 To do this the content of the cell ismodied (Table 1). Considering another peak, 1031 cm�1 (Phe;protein C–N str.), it can be seen that it loses its intensity for

Analyst, 2013, 138, 7157–7163 | 7161

Fig. 8 Distribution of the proteins in an: (A) activated and (B) intact humanlymphocyte (integrated Raman intensities, 1662 cm�1 amide I marker band);tyrosine distribution in a: (C) activated and (D) intact human lymphocyte(835 cm�1 marker band due to buried tyrosine residues in proteins); excitationlaser line 785 nm.

Analyst Paper

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article Online

activated lymphocyte.3 The 1100 cm�1 phosphate backbonevibration, which may be regarded as a marker mode for DNAconcentration, disappeared from the activated form.34 Thischange originates from evolving lymphocyte towards theplasma cell, which has a small eccentric nucleus and most ofthe cell is lled by cytoplasm and well developed rough endo-plasmic reticulum (RER) in which the synthesis of the immu-noglobulins occurs. This is also why lipid Raman bands(1450 cm�1) are more common in plasmocytes (because ofRER). On the other hand, amide I (C]O stretching and N–Hwagging vibrations) Raman band at 1662 cm�1 is more inten-sive in the intact form35,36 (Fig. 8A and B). The position of amideI and amide III band at 1662 and 1252 cm�1, respectively,indicates that immunoglobulin is predominantly composed ofa mixture of antiparallel b-sheet structures and random coils.1,2

Differences in band intensities may be due to differentprotein compositions in both the investigated forms. B cellactivation causes the intracellular cytoplasm compositionchange, due to immunoglobulin secretion during ghtinginfections.

Another characteristic band, which occurs as well in intactlymphocytes as in the activated form, is the 850 cm�1 peak dueto the buried tyrosine residues in the proteins. This amino acidis an important element in signal transduction and regulatingcellular activity (by tyrosine kinase).37 In the absence of antigen,mature lymphocytes reside in a quiescent state. The differencebetween the two forms is in the tyrosine distribution whichreects the arrangement of this amino acid in the nucleus andcytoplasm (Fig. 8C and D). Cytoplasmic domains with a func-tionally important region of homology are called immunor-eceptor tyrosines based on the activation motif (ITAM).

4. Conclusions

The observed biochemical structural alterations are caused byviral infection (the u).

The presence and concentration of immunoglobulins makelymphocyte cells spectroscopically “labeled”. The activated formis “stained” by the 521 cm�1 band.

7162 | Analyst, 2013, 138, 7157–7163

The up shi of the S–S mode from the broad band centeredat 510 cm�1 of a human normal immunoglobulin to a singleband at 521 cm�1 of a human B cell indicates a selection of theoptimal geometry of the disulphide bridges to bind to a foreignantigen.

A single sharp band at 521 cm�1 in the Raman spectra ofsingle lymphocytes suggests that the disulphide linkages are insimilar geometries in the activated cell.

It is believed that the formation of immunoglobulins and thefollowing aggregation occurs during B cell activation and thisactually leads to a remarkable increase in the intensity of 521cm�1 marker bands.

Polarization data reects structural alteration betweendomain fragments.

Protein content in intact and activated cells is different.

References

1 A. T. Tu, Raman Spectroscopy in Biology: Principles andApplications, John Wiley & Sons, New York, 1982.

2 V. Fawcett and D. A. Long, Biological Applications of RamanSpectroscopy, in Molecular Spectroscopy, ed. R. F. Barrow,Royal Society of Chemistry, 1976, vol. 4, ch. 4.

3 I. Notingher, Raman Spectroscopy Cell-based Biosensors,Sensors, 2007, 7, 1343–1358.

4 J. Czepiel, M. Birczynska, M. Kozicki, A. Wesełucha-Birczynska, G. Biesiada, T. Mach, A. Garlicki, Abstract ofPapers, 23rd International Conference on Raman Spectroscopy,Bangalore, India, 2012; p. 80.

5 A. Stevens and J. S. Lowe, Human Histology, WydawnictwoLekarskie PZWL & Wydawnictwo Medyczne Słotwinski Verlag,ed. M. Zabel, Warszawa, Poland, 2nd edn, 2000.

6 D. Male, J. Brostoff, D. B. Roth and I. Roitt, Immunology,Elsevier Urban & Partner, Wrocław, 2nd and 7th edn,2008.

7 F. Batista, Facundo Batista: watching B cells spread and grabantigens, J. Exp. Med., 2008, 205, 1718–1719.

8 F. D. Batista, D. Iber and M. S. Neuberger, B cells acquireantigen from target cells aer synapse formation, Nature,2001, 411, 489–494.

9 C. M. Karsten and J. Kohl, The immunoglobulin, IgG Fcreceptor and complement triangle in autoimmunediseases, Immunobiology, 2012, 217, 1067–1079.

10 S. Booth, G. M. Griffiths and W. Dunn, The ImmunologicalSynapse of CTL Contains a Secretory Domain andMembrane Bridges, Immunity, 2001, 15, 751–761.

11 P. Roda-Navarro, M. Mittelbrunn, M. Ortega, D. Howie,C. Terhorst, F. Sanchez-Madrid and E. Fernandez-Ruiz,Dynamic Redistribution of the Activating 2B4/SAP Complexat the Cytotoxic NK Cell Immune Synapse, J. Immunol.,2004, 173, 3640–3646.

12 Y. M. Vyas, K. M. Mehta, M. Morgan, H. Maniar, L. Butros,S. Jung, J. K. Burkhardt and B. Dupont, SpatialOrganization of Signal Transduction Molecules in the NKCell Immune Synapses During MHC Class I-RegulatedNoncytolytic and Cytolytic Interactions, J. Immunol., 2001,167, 4358–4367.

This journal is ª The Royal Society of Chemistry 2013

Paper Analyst

Publ

ishe

d on

27

Sept

embe

r 20

13. D

ownl

oade

d by

Uni

vers

ity o

f C

alif

orni

a -

Irvi

ne o

n 30

/10/

2014

18:

28:4

8.

View Article Online

13 C. Barcia, C. E. Thomas, J. F. Curtin, G. D. King,K. Wawrowsky, M. Candol, W.-D. Xiong, C. Liu,K. Kroeger, O. Boyer, J. Kupiec-Weglinski, D. Klatzmann,M. G. Castro and P. R. Lowenstein, In vivo matureimmunological synapses forming SMACs mediateclearance of virally infected astrocytes from the brain,J. Exp. Med., 2006, 203, 2095–2107.

14 P. Roda-Navarro, Microspectroscopy reveals mechanisms oflymphocyte activation, Integr. Biol., 2013, 5, 300–311.

15 J. Y. Park, H. O. Kim, K.-D. Kim, S. K. Kim, S. K. Lee andH. Jung, Monitoring the status of T-cell activation in amicrouidic system, Analyst, 2011, 136, 2831–2836.

16 M. Diem, M. Romeo, S. Boydston-White, M. Miljkovic andC. Matthaus, A decade of vibrational micro-spectroscopy ofhuman cells and tissue (1994–2004), Analyst, 2004, 129,880–885.

17 P. Lasch, M. Boese, A. Pacico and M. Diem, FT-IRspectroscopic investigations of single cells on thesubcellular level, Vib. Spectrosc., 2002, 28, 147–157.

18 D. I. Ellis and R. Goodacre, Metabolic ngerprinting indisease diagnosis: biomedical applications of infrared andRaman spectroscopy, Analyst, 2006, 131, 875–885.

19 C. A. Owen, I. Notingher, R. Hill, M. Stevens and L. L. Hench,Progress in Raman spectroscopy in the elds oftissueengineering, diagnostics and toxicological testing,J. Mater. Sci.: Mater. Med., 2006, 17, 1019.

20 N. Uzunbajakava, A. Lenferink, Y. Kraan, B. Willekens,G. Vrensen, J. Greve and C. Otto, Nonresonant Ramanimaging of protein distribution in single human cells,Biopolymers, 2003, 72, 1–9.

21 F. A. Saul, L. M. Amzel and R. J. Poljak, Preliminaryrenement and structural analysis of the Fab fragmentfrom human immunoglobulin new at 2.0 A resolution,J. Biol. Chem., 1978, 253, 585–597.

22 J. Novotny, R. Bruccoleri, J. Newell, D. Murphy, E. Haber andM. Karplus, Molecular anatomy of the antibody binding site,J. Biol. Chem., 1983, 258, 14433–14437.

23 E. Vargas-Madrazo and E. Paz-Garcia, An improved model ofassociation for VH–VL immunoglobulin domains:Asymmetries between VH and VL in the packing of someinterface residues, J. Mol. Recognit., 2003, 16, 113–120.

24 D. Kuroda, H. Shirai, M. P. Jacobson and H. Nakamura,Computer-aided antibody design, Protein Eng., Des. Sel.,2012, 25, 507–521.

This journal is ª The Royal Society of Chemistry 2013

25 T. Kitagawa, T. Azuma and K. Hamaguchi, The Ramanspectra of Bence-Jones proteins. Disulde stretchingfrequencies and dependence of Raman intensity oftryptophan residues on their environments, Biopolymers,1979, 18, 451–465.

26 W. Qian, W. Zhao and S. Krimm, Vibrational studies ofdisulde group in proteins. Part IV. SS and CS stretchfrequencies of known peptide and protein disuldebridges, J. Mol. Struct., 1991, 250, 89–102.

27 W. Qian and S. Krimm, Vibrational studies of the disuldegroup in proteins, J. Raman Spectrosc., 1992, 23, 517–521.

28 H. E. Van Wart, A. Lewis, H. A. Scheraga and F. D. Saevat,Disulde Bond Dihedral Angles from Raman Spectroscopy,Proc. Natl. Acad. Sci. U. S. A., 1973, 70, 2619–2623.

29 H. Sugeta, A. Go and T. Miyazawa, S–S and C–S stretchingvibrations and molecular conformations of dialkyldisuldes and cystine, Chem. Lett., 1972, 83–86.

30 L. J. Harris, E. Skaletsky and A. McPherson, Crystallographicstructure of an intact IgG1 monoclonal antibody, J. Mol.Biol., 1998, 275, 861–872.

31 K. Nakamoto, Infrared and Raman spectra of Inorganic andCoordination Compounds, John Wiley & Sons, Hoboken,USA, 6th edn, 2009.

32 Ch. A. Janeway, Jr, P. Travers, M. Walport andM. J. Shlomchik, Immunobiology, Garland Science, NewYork, 5th edn, 2001.

33 T. G. Spiro and T. C. Strekas, Resonance Raman Spectra ofHemoglobin and Cytochrome c: Inverse Polarization andVibronic Scattering, Proc. Natl. Acad. Sci. U. S. A., 1972, 69,2622–2626.

34 J. W. Chan, D. S. Taylor, T. Zwerdling, S. M. Lane, K. Iharaand T. Huser, Micro-Raman spectroscopy detectsindividual neoplastic and normal hematopoietic cells,Biophys. J., 2006, 90, 648–656.

35 Y. Takai, T. Masuko and H. Takeuchi, Lipid structure ofcytiotoxic granules in living human killer T lymphocytesstudied by Raman microspectroscopy, Biochim. Biophys.Acta, Gen. Subj., 1997, 1335, 199–208.

36 V. V. Pully, A. T. M. Lenferink and C. Otto, Time-lapse Ramanimaging of single live lymphocytes, J. Raman Spectrosc., 2011,42, 167–173.

37 A. L. DeFranco, Transmembrane signaling by antigenreceptors of B and T lymphocytes, Curr. Opin. Cell Biol.,1995, 7, 163–175.

Analyst, 2013, 138, 7157–7163 | 7163