Disponible à / Available at permalink :

(1)

Dépôt Institutionnel de l’Université libre de Bruxelles / Université libre de Bruxelles Institutional Repository

Thèse de doctorat/ PhD Thesis Citation APA:

Wald, N. (2015). FTIR imaging as a new histopathological technique to characterize melanomas and their immune microenvironment (Unpublished doctoral dissertation). Université libre de Bruxelles, Faculté des Sciences – Ecole Interfacultaire des Bioingénieurs, Bruxelles.

Disponible à / Available at permalink : https://dipot.ulb.ac.be/dspace/bitstream/2013/209046/4/124657e1-cf02-483a-8042-66dd147a0739.txt

(English version below)

Cette thèse de doctorat a été numérisée par l’Université libre de Bruxelles. L’auteur qui s’opposerait à sa mise en ligne dans DI-fusion est invité à prendre contact avec l’Université ([email protected]).

Dans le cas où une version électronique native de la thèse existe, l’Université ne peut garantir que la présente version numérisée soit identique à la version électronique native, ni qu’elle soit la version officielle définitive de la thèse.

DI-fusion, le Dépôt Institutionnel de l’Université libre de Bruxelles, recueille la production scientifique de l’Université, mise à disposition en libre accès autant que possible. Les œuvres accessibles dans DI-fusion sont protégées par la législation belge relative aux droits d'auteur et aux droits voisins. Toute personne peut, sans avoir à demander l’autorisation de l’auteur ou de l’ayant-droit, à des fins d’usage privé ou à des fins d’illustration de l’enseignement ou de recherche scientifique, dans la mesure justifiée par le but non lucratif poursuivi, lire, télécharger ou reproduire sur papier ou sur tout autre support, les articles ou des fragments d’autres œuvres, disponibles dans DI-fusion, pour autant que :

Le nom des auteurs, le titre et la référence bibliographique complète soient cités; L’identifiant unique attribué aux métadonnées dans DI-fusion (permalink) soit indiqué; Le contenu ne soit pas modifié.

L’œuvre ne peut être stockée dans une autre base de données dans le but d’y donner accès ; l’identifiant unique (permalink) indiqué ci-dessus doit toujours être utilisé pour donner accès à l’œuvre. Toute autre utilisation non mentionnée ci-dessus nécessite l’autorisation de l’auteur de l’œuvre ou de l’ayant droit.

--- English Version ---

This Ph.D. thesis has been digitized by Université libre de Bruxelles. The author who would disagree on its online availability in DI-fusion is invited to contact the University ([email protected]).

If a native electronic version of the thesis exists, the University can guarantee neither that the present digitized version is identical to the native electronic version, nor that it is the definitive official version of the thesis.

DI-fusion is the Institutional Repository of Université libre de Bruxelles; it collects the research output of the University, available on open access as much as possible. The works included in DI-fusion are protected by the Belgian legislation relating to authors’ rights and neighbouring rights. Any user may, without prior permission from the authors or copyright owners, for private usage or for educational or scientific research purposes, to the extent justified by the non-profit activity, read, download or reproduce on paper or on any other media, the articles or fragments of other works, available in DI-fusion, provided:

The authors, title and full bibliographic details are credited in any copy;

The unique identifier (permalink) for the original metadata page in DI-fusion is indicated; The content is not changed in any way.

It is not permitted to store the work in another database in order to provide access to it; the unique identifier (permalink) indicated above must always be used to provide access to the work. Any other use not mentioned above requires the authors’ or copyright owners’ permission.

(2)

D 04081

Ecole Interfacultaire de Bioingénieurs

FTIR imaging as a new histopathological

technique to characterize melanomas and

their immune microenvironment

Noémie Wald

Thèse présentée en vue de l'obtention du grade de docteur en sciences

agronomiques et ingénierie biologique.

(3)

ULB

UNIVERSITÉ LIBRE DE BRUXELLES

Ecole Interfacultaire de Bioingénieurs

FTIR imaging as a new histopathological

technique to characterize melanomas and

their immune microenvironment

Noémie Wald

Thèse présentée en vue de l'obtention du grade de docteur en sciences

agronomiques et ingénierie biologique.

Service de Structure et Fonction des Membranes Bioiogiques - Centre de Biologie Structurale

Promoteur : Pr. Erik Goormaghtigh

(4)

«What is research but a blind date with knowledge? » - William J. Henry

(5)

REMERCIEMENTS

En tout premier lieu, je voudrais remercier Erik Goormaghtigh, sans qui tout cela n'aurait pas été

possible. Merci Erik de m'avoir accueillie au sein du SFMB et de m'avoir aiguillée tout au long de ce

parcours. Ton soutien et tes encouragements ont été essentiels lors de chaque étape de ce long

cheminement. Depuis la défense PRIA Jusqu'à la rédaction du manuscrit tu as pu être présent tout en

me laissant beaucoup de liberté.

Then, I would like to sincerely thank ail the jury members; Pr. Philip Heraud, Dr. Véronique Mathieu,

Pr. Vincent Raussens, Pr. Daniel E. Speiser, Pr. Michel Vandenbranden and Pr. Bayden Wood, for

agreeing to be part of my thesis jury. Thank you for your time and for sharing your expertise.

Je remercie le Fonds de la Recherche Scientifique (FNRS) de m'avoir octroyé une bourse PRIA,

l'élément indispensable pour pouvoir entreprendre un projet de recherche.

Je voudrais également remercier les personnes avec qui j'ai eu la chance de collaborer durant ce

travail de recherche. Parmi les collaborateurs de Lausanne, je remercie Daniel Speiser pour ses

nombreux conseils scientifiques mais aussi pour sa passion, sa conscience et son regard très juste.

Ensuite je voudrais remercier Natacha pour sa très grande disponibilité et son enthousiasme ! Merci

aussi à Amandine qui m'a permis de démarrer cette collaboration. Ce travail de thèse a également

mené à une deuxième collaboration avec l'Université d'Angers. Merci aux Docteurs Yannick Le Corre

et Ludovic Martin. Je remercie également Véronique Mathieu qui m'a suivie depuis mon mémoire

jusqu'à la fin de cette thèse. Je la remercie pour ses conseils très judicieux, ses idées et pour m'avoir

ouvert de nouvelles perspectives de recherche.

Mais bien entendu, cette thèse n'aurait pas été la même sans toutes ces merveilleuses personnes qui

constituent le SFMB. Premièrement, merci à toutes les personnes de notre petit groupe infrarouge

avec qui j'ai partagé plein de très agréables moments autant dans le cadre du travail que d'un point

de vue plus personnel. Merci à Allison, Alix, Magali, Marie, Margarita, Mouna et Joëlle. Je garderai

des souvenirs mémorables de nos congrès aux quatre coins du monde (Tha lande, Etats-Unis,

Pologne...). Merci aussi pour votre soutien et vos conseils au quotidien car oui, une thèse n'est pas un

long fleuve tranquille. Merci aussi pour les « papotes » qui permettent d'oublier (ou de résoudre) les

problèmes de la thèse au jour le jour.

(6)

Merci à Rabia, Marie, AIdo, Malvina, Adelin, Sabrina et Véronique pour le partage du bureau. Merci

pour vos conseils et votre présence au quotidien !

Merci à Chloé, Anastassia, Marie, Nancy, Alix, Mag, Allison pour les midis au soleil, pour les petits

verres au Waff et autres sorties ciné mais aussi, bien entendu, pour vos conseils scientifiques.

Un merci spécial à Audrey qui nous a ouvert la voie pour l'utilisation parfois chaotique de notre cher

microscope infrarouge (que je ne remercie pas d'ailleurs !)

Merci à Jean-Marie pour ses blagues et ses commentaires qui essayent constamment de remettre en

question notre travail pour mieux nous « challenger ».

Merci aussi à Caroline, Rosie, Maude, Deborah, Aurélie, Tina, François, Vincent, Gaëlle, Stéphanie,

Mathieu, Fabien, Emilie K et Benjamin pour tout ce que vous avez apporté et ce que vous apportez

auSFMB.

Merci à Michel, Vincent, Fabrice, Cédric, Guy d'avoir été là pour m'apporter votre soutien et vos

conseils si précieux lors de cette thèse, et plus particulièrement lors des séminaires.

D'un point de vue scientifique, je voudrais remercier les personnes qui m'ont aidée ponctuellement

dans ce travail. Merci notamment à Soizic.

Merci aux personnes qui m'ont aidée dans les corrections du manuscrit. Ils se reconnaîtront©.

Merci aussi aux personnes avec qui j'ai organisé deux éditions de la Bioengineer Research Day : Fred,

Thomas, Emilie, Marie, J-F, Allison, Magali, Lauranne et Khadija. C'était vraiment une expérience

enrichissante.

(7)

Et puis un merci spécial à Jess, mon amie d'enfance, toujours présente et c'est vraiment génial. A

mon avis on est en bonne voie pour se retrouver à l'Heureux Séjour ensemble !

Merci aussi à ma famille qui m'a beaucoup encouragée : mes parents qui m'ont toujours soutenue

pour que je donne le meilleur de moi-même et à mes trois frérots : Micha, Jérémie et David qui

comptent beaucoup pour moi. Merci aussi à Virginie et Caroline, mes deux belles-sœurs.

(8)

TABLE OF CONTENT

ABBREVIATIONS LIST...i

ABSTRACT... üi

Chapter I :

GENERAL INTRODUCTION... 1

1. MELANOMA... 1 1.1. Epidemiology and risk factors _____________________________________________________________1 1.2. Progression and subtypes of melanoma___________________________________________________ 2 1.3. Melanoma staging, classification and prognosis____________________________________________ 3 1.4. Diagnosis and treatments_________________________________________________________________ 6 1.5. Anatomopathology of melanoma _________________________________________________________ 8 2. HALLMARKOF CANCER... 12 2.1. Acquired capabilities____________________________________________________________________ 12 2.2. Enabling characteristics________________________________________________________ ^16 2.3. Tumor microenvironment______________________________________________________________ ^17 3. TUMOR-IMMUNITY: INTEGRATING IMMUNITY'S IN CANCER PROGRESSION... 18 3.1. Melanoma and the immune System _____________________________________________________ .18 3.2. Prognosis implication of immune infiltration _____________________________________________ 21 3.3. Immunothérapies to target melanoma___________________________________________________ 22 4. INFRARED SPECTROSCOPY...23 4.1. Principles of IR spectroscopy_____________________________________________________________23 4.2. FourierTransform Infrared technology___________________________________________________ 24 4.3. FTIR micro-spectroscopy ________________________________________________________________ 25 4.4. Applications to biological molécules______________________________________________________27 4.5. Applications to prokaryotic and eukaryotic cells___________________________________________ 28 4.6. Applications to biological tissues________________________________________________________ 30

Chapter II:

AIM OF THE THESIS... 33

(9)

3.4. Data analysis____________________________________________________________________________ 40

4. RESULTS... 42

4.1. Does the FFPE treatment induce similar spectral changes for different cell lines?____________42 4.2. Are spectral proximities and distances between closely related cell lines conserved after FFPE Processing?____________________________________________________________________________________ 44 4.3. Is supervised discrimination power between cell lines conserved after FFPE Processing?_____46 5. DISCUSSION... 48

Chapter IV :

INFRARED IMAGING OF PRIMARY MELANOMAS REVEALS HINTS OF REGIONAL

AND DISTANT METASTASES... 51

1. ABSTRACT... 52

2. INTRODUCTION...53

3. MATERIALS AND METHODS...55

3.1. TMA description________________________________________________________________________ 55 3.2. FTIR measurements_____________________________________________________________________56 3.3. Data analysis___________________________________________________________________________ 56 4. RESULTS...59

4.1. Major cell type identification____________________________________________________________ 59 4.2. Comparison of melanoma cells in the primary tumor and in métastasés____________________ 63 4.3. Corrélation between Ki67 expression rate and infrared signature___________________________ 64 4.4. Melanoma spectral sub-classification accordingto cancer stages___________________________ 64 5. DISCUSSION... 71

6. SUPPLEMENTARYMATERIAL...74

7. NOTE ADDED IN PROOF... 76

Chapter V:

INFRARED SPECTRA OF PRIMARY MELANOMAS CAN PREDICT RESPONSE TO

CHEMOTHERAPY: THE EXAMPLE OF DACARBAZINE...77

1. ABSTRACT...78

2. INTRODUCTION... 79

3. RESULTS... 81

3.1. Spectra from responding patients compared to non-responding patients___________________ 82 3.2. A model of prédiction and cross-validation________________________________________________84 4. DISCUSSION...87

5. MATERIALS AND METHODS... 90

(10)

6.2. Materials and methods_________________________________________________________________ 93

ChapterVI:

AN INFRARED SPECTRAL SIGNATURE OF HUMAN LYMPHOCYTE

SUBPOPULATIONS FROM PERIPHERAL BLOOD...95

1. ABSTRACT...96

3.1. CD4* and CD8"^ T cell isolation from peripheral blood samples______________________________ 99 3.2. Assessment of cell purity by FACS______________________________________________________ 100 3.3. FTIR measurements___________________________________________________________________ 100 3.4. Data analysis__________________________________________________________________________ 101 4. RESULTS... 103

4.1. Identification of CD4* and CD8* T cells___________________________________________________ 103 4.2. Quantification of regulatory T cells_____________________________________________________ 107 5. DISCUSSION...110

6. SUPPLEMENTARY MATERIAL... 114

Chapter VII :

IDENTIFICATION OF MELANOMA CELLS AND LYMPHOCYTE SUBPOPULATIONS IN

LYMPH NODE METASTASES BY FTIR IMAGING HISTOPATHOLOGY... 115

1. ABSTRACT... 116

3.1. Patient samples________________________________________________________________________ 119 3.2. Tissue handling and immunohistochemistry______________________________________________119 3.3. FTIR measurements___________________________________________________________________ 120 3.4. Data analysis__________________________________________________________________________ 120 4. RESULTS... 123

(11)

PUBLICATIONS

173 LIST OF TABLES... 175

(12)

ABBREVIATIONS LIST

AJCC; American Joint Committee on Cancer ALM: Acral Lentiginous Melanoma

Arg-1: Arginase-1

ATR: Attenuated Total Reflection

CGH: Comparative Genomic Hybridization CT; Computerized Tomography

CTLA-4: Cytotoxic T-Lymphocyte-Associated protein 4 DOPC; Dioleoylphosphatidylcholine

FACS: Fluorescence Activated Cell Sorting FDA: Food and Drug Administration FFPE: Formalin-Fixation Paraffin-Embedding FISH: Fluorescence In Situ Hybridization Foxp3: Forkhead box P3

FPA: Focal plane array

FTIR: FourierTransform Infrared HMB45: Human Melanoma Black 45 HCA: Hierarchical Cluster Analysis IDO: Indoleamine 2,3-dioxygenase IFN: Interferon

IL: Interleukin

INOS: Inducible Nitric Oxide Synthase IR: Infrared

(13)

ABBREVIATIONS LIST

PC: Principal Component

PCA: Principal Component Analysis PD-1: Programmée! death 1 protein PD-Ll: Programmed death Ligand 1 PET: Positron Emission Tomography

PLS-DA: Partial Least Square Discriminant Analysis PLSR: Partial Least Square Régression

ROC: Receiver Operating Characteristic S/N: Signal to Noise

SSM: Superficiel Spreading Melanoma TGF-p: Transforming Growth Factor Beta TIL: Tumor Infiltrating Lymphocytes TMA: Tissue Microarrays

TNF: Tumor Necrosis Factor Treg: Regulatory T cells UV: Ultraviolet

(14)

ABSTRACT

An early diagnosis of melanoma is essential to reduce mortality of patients. The diagnosis is aiso

fundamental to predict the outcome of patients and to select the most adapted treatment. Current

diagnostic assessments are obtained after Visual inspection of the histological section of the primary

tumor. The pathologist has first to détermine the malignant nature of the lésion and then to assess

the potentiel of the lésion to form métastasés. Depending on several characteristics of the primary

tumor (mainly tumor thickness, ulcération and mitotic rate), the sentinel node is surgically removed

and the détection of tumor cells is based on its histopathological examination. These assessments

are time consuming, to some degree subjective and are particularly challenging. Among melanoma

patients that are not subject to sentinel node surgery, 6.5 % will develop métastasés while 80 % of

patients that undergo sentinel node surgery will effectively présent métastasés. The search for

biomarkers that can identify malignant cells, evaluate potential of invasion or help selecting a

treatment is still on.

In this thesis we used a new and promising technique of imaging based on infrared spectroscopy to

study melanoma primary tumors and metastatic lymph nodes. Infrared spectroscopy brings

information on the biochemical composition of the main components of the cells. When combined

with a microscope and with multivariate statistical analyses, images that are generated allow the

identification of melanoma cells and stromal cells in the biopsy. We aIso focused on the immune

infiltration as it was shown to carry an important prognosis value for melanoma patients.

The first part of the thesis was a prerequisite for the rest of the study. It addresses the effects of the

process of fixation that tissues obtained by surgical resection undergo for their long term

préservation. In chapter III, we showed that Formalin-Fixation and Paraffin-Embedding (FFPE)

procedure induces small but significant modifications in the infrared spectra of cells but these are

very similar for different cell Unes. In turn, it préserves the potential to identify closely-related cell

Unes by infrared spectroscopy.

We thus pursued our study on primary melanomas. In chapter IV, we first developed an automatic

tool capable of identifying melanoma cells and the main cells of the tumor microenvironment in

tissue sections. Importantly, we built a second model that brings information on the presence of

métastasés on the basis of the spectral signature of the primary tumor.

(15)

CHAPTER I : GENERAL INTRODUCTION

1. MELANOMA

Melanoma is a cancer that develops from mélanocytes, the specialized pigmented cells mainly

présent in the skin but aiso in hair follicles, eyes and mucosa. In the skin, mélanocytes are mainly

located in the basal layer of the epidermis and produce melanin, the brown pigment that has a

protective rôle, in particular for the cell DNA. Thereby, mélanocytes hâve a key rôle in protecting the

skin from the damaging effects of UV radiations and in preventing skin cancer [1]. Mélanocytes are

aIso présent in eyes and mucosa but most melanomas arise from mélanocytes of the skin (more than

90%) [2,3],

1.1. Epidemiology and risk factors

Melanoma represents a global heaith issue as incidence and mortality of this cancer are steadily

increasing in western population. The majority of melanomas is probably related to a single risk

factor, which is sun exposure [4]. The number of new cases increased by twofold in the last 20 years

and is still on the rise [1].

Worldwide incidence of melanoma was evaluated at 232130 new cases per year and mortality at

55488 deaths in 2012. They accounted respectively for 1.7% of ail newiy diagnosed cancer and 0.7%

of cancer deaths. As represented in Figure 1-1, melanoma incidence and mortality are not equally

distributed worldwide and the developed countries are the most impacted by the disease [5]. Almost

85 % of cases occur in developed countries [6]. Statistics obtained from the World Health

Organization, report that in developed régions of the world, melanoma of the skin is the eight most

common cancer and incidence and mortality reach respectively 3.2% of ail newiy diagnosed cancer

and 1.3 % of ail cancer deaths. The most affected countries and régions are New Zealand, Australie,

Northern Europe and the USA [5]. Most of these areas are largely populated by people originating

from Europe, suggesting a direct implication of sun exposure on comparatively low pigmented skin

[7].

(16)

CHAPTER I - GENERAL INTRODUCTION

gene but can aiso be related to phenotypical characteristics [4,8]. There is today sufficient evidence

showing that sun exposure causes melanoma development and that solar UV radiations is the most

important environmental factor [4].

3* Incidence ASR I I <048 No DM

b.

Mortality ASR i—\ <0 21 No Dm

Figure 1-1 : Epidemiologie data for melanoma ofthe skin. Worldwide répartition of incidence (a) and mortality (b) per 100 000 people per yearfor 2012 [5],

1.2. Progression and subtypes of melonomo

(17)

The first anomaly of the pigmentary System is the benign nevus and consists in a high number of

normal mélanocytes. The dysplastic nevus can either develops from a preexisting benign nevus or

from another location and implicates abnormal growth of atypical mélanocytes. Dysplastic nevi are

the precursor lésions of melanoma even if most of these lésions will not resuit in melanoma. The first

phase of malignant progression is a radial growth inside the epidermis and this phase is followed by a

vertical-growth phase through the dermis. This latter growth-phase involves pénétration of the

basement membrane that may lead to metastasis formation. This model proposed by Clark and

colleagues is based on clinical and histopathological features but this sequence of progression is not

observed for every melanoma [1,8-10].

stage Benign Nevus Epidermis Basement membrane Dysplastic Radia|.Growth Nevus Phase Dermis Vertical-Growth Phase Metastatic Melanoma

Figure 1-2 : Schematic ofthe step involved in the progression of melanoma [8].

Primary melanomas can présent different growth patterns and incidence of these growth patterns

varies according to the sex, the âge and the race. The four major growth patterns (subtypes of

melanoma) are nodular melanoma (NM), acral lentiginous melanoma (ALM), lentigo maligna

melanoma (LMM) and superficial spreading melanoma (SSM). While this classification has no

independent prognostic value , it is still considered to be usefui and is based on distinct risk factors,

sites of prédilection and therapeutic implications [11]. ALM are mainly found on palm of hands, sole

of feet and nail bed and are therefore not linked to sun exposure whereas LMM are related to

chronic sun exposure (older people) and SSM to épisodes of severe sunburns mainly in the chiidhood

[

1

].

1.3. Melanoma staging, classification and prognosis

(18)

treatment planning and stratification and assessment of clinical trials [12], The American Joint

Committee on Cancer (AJCC) staging System proposed in 2009 a new version (Seventh édition) of the

AICC Cancer Staging Manual for melanoma [13]. This staging is based on well-established

independent prognostic factors involving clinical and anatomopathological criteria. In the first part of

this section, we will présent an overviev\/ of the current factors included in the AJCC staging System

and in the second part, we will présent the new prognostic factors. Patients are staged from stage I

to IV according to the progression of the disease.

The current AJCC staging System is based on the TNM classification. The TNM System is an expression

of the anatomical extent of the disease and is based on the assessment of three components: the

extent of the primary tumor (T), the absence or presence and the extent of régional lymph node

métastasés (N) and the absence or presence of distant métastasés [14]. Table 1-1 summarizes how

these different criteria are defined and included in the TNM classification.

1.3.1. Primary tumor (T)

T is defined by a number ranging from 1 to 4 that dépends on the tumor thickness and by the letter a

or b related to ulcération and mitotic rate. Ail are assessed by microscopy.

Primary tumor thickness (Breslow index): The first criterion that defines the extent of the primary

tumor is the tumor thickness in mm, aiso expressed as the Breslow index. This is aiso the most

important prognostic factor for stage I patients.

Ulcération: The second most important criterion is ulcération defined as an absence of an intact

epidermis overlaying the primary melanoma. Ulcération is associated with a more aggressive form of

the disease [14].

Mitotic rate: It represents the prolifération of the primary tumor and is assessed as the number of

mitoses per square millimeter by microscopie investigation. The threshold is defined at 1/ mm^.

1.3.2. Lymph node involvement (N)

(19)

viscéral) and lactate dehydrogenase (LDH) level in sérum. LDH level is an independent and highiy

significant predictor for stage IV patients.

Table 1-1 : Melanoma TNM classification [13].

T«bl« 1. TNM Staging Categories for Cutaneous Melanoma Classification Thickness (mm) Ulcération Status/Mitoses

T

Tis NA NA

Tl 2= 1.00 a: Without ulcération and milosis < l/mm* b: With ulcération or mitoses a l/mm* T2 1.01-2.00 a; Without ulcération b: With ulcération T3 2.01-4.00 a; Without ulcération b: With ulcération T4 >4.00 a: Without ulcération b: With ulcération N No. of Metastatic Modes Nodal Metastatic Burden

NO 0 NA NI 1 a; Micrometastasis* b; Macrometastasist N2 2-3 a; Micrometastasis* b; Macrometastasist c: In transit metastases/satellites without metastatic nodes N3 4+ metastatic nodes. or

matted nodes, or in transit

metastases/satellites with metastatic nodes

M Site Sérum LDH

MO No distant métastasés NA Mia Distant skin. subcutaneous. Normal

or nodal métastasés

M1b Long métastasés Normal MIC Ail other viscéral Normal

métastasés

Any distant metastasis Elevated

Abbreviâtions: NA. not applicable: LDH. lactate dehydrogenase. 'Micronnetastases are diagnosed after sentinel lymph node biopsy. tMacrometastases are defined as clinically détectable nodal métastasés confirmed pathologically.

(20)

Figure 1-3 : Survival rates for patients with melanoma, stage I through IV (adapted from [15]/

Among others prognostic variables that are not yet included in the AJCC are lymphocytes infiltration,

tumor vascularity and anatomical location of the primary tumor. While lymphocytes infiltration is

positively correlated to survival, tumor vascularity correlates with a poor prognosis [12]. These two

criteria are discussed later in the thesis.

1.4. Diagnosis and treatments

(21)

The second step is the excision of the lésion (with 1 to 2 mm margin of adjacent normal skin) and its

anatomopathological analysis. The pathologie assessment is the gold standard for the diagnosis of

melanoma and will détermine whether the lésion is benign (nevus) or malignant. Since melanoma

présents a wide range of architectural and cytological characteristics and features, the pathological

diagnosis of melanocytic tumors can be one of the most difficult areas of diagnostic histopathology.

If a malignant melanoma is diagnosed, then the pathologist needs to assess the metastatic potential

of the tumor through observation of several parameters [16]. The anatomopathological analysis

defines the type of melanoma, the thickness, the ulcération and the mitotic rate. According to the

microstaging of the primary lésion, the resection of the sentinel lymph node (first node draining

lymph from the primary lésion) will be required to detect potential métastasés. Primary lésion with

thickness above 1 mm or under 1 mm but ulcerated or with a mitotic rate higher than 1/mm^ are

subject to sentinel lymph node biopsy [17]. Lymph nodes are the most common first site of

metastatic involvement and the détection of métastasés in the sentinel nodes is the most important

prognostic factor for early melanomas [18]. The localization of the sentinel node is usually

determined by use of radiocolloid and/or blue dye [19]. Around 20% of patients with a thickness of

the primary tumor above 1 mm will develop lymph node métastasés [16]. Additionally, some patients

will undergo therapeutic lymphadenectomy (surgical removal of lymph nodes) when lymph nodes

will be detected positive following clinical assessment (palpable nodal enlargement or abnormal

features on ultrasound imaging, with or without guided fine-needle aspiration) [16]. For patients who

présent a positive sentinel lymph node, radiologie tests (CT scans, PET/CT or MRI...) are

recommended to define the précisé staging (e.g. Ml a, b, c) and for the évaluation of spécifie signs or

symptoms. However, these radiologie tests rarely reveal métastasés in the absence of symptoms or

signs that are aiready determined by physical examination [17,20].

(22)

The first drugs approved by the Food and Drug Administration (FDA) for treatment of metastatic

melanomas were chemotherapies: first hydroxyurea and then dacarbazine. Two immunothérapies

were approved later; interferon alpha and interleukin-2. Since recently, advances in immunotherapy

and molecular targeting in melanoma hâve yielded new treatments that improve overall survival of

advanced melanoma patients. Among these newiy FDA approved molécules, there are targeted

thérapies (BRAF or MEK inhibitors) and immune checkpoint blockades (anti-CTLA4, anti-PD-1)

[17,21,23,24]. Today, the therapeutic landscape for advanced melanoma is evolving rapidiy. The

treatments that are available to treat stage III and stage IV patients who présent unresectable

métastasés are multiple. Yet there are no ruies to figure out how to make the best use of these

treatments either in sequence or in combination. Nevertheless, some data exist from randomized

trials that can guide clinicians in this management. Approximatively 45% of patients harbor an

activating mutation in the intracellular signaling kinase, BRAF and can be treated with BRAF or MEK

inhibitors. Some studies indicate that immunothérapies may be considered before targeted thérapies

as immunothérapies can be associated with more durable response than targeted ones and may no

longer be effective after treatment with targeted thérapies [24]. Moreover, each agent is associated

with unique side effects and response patterns. This raised new questions and new challenges that

need to be resolved to progress further in the treatment of advanced melanoma [17].

1.5. Anatomopathology of melanoma

This chapter section provides some basic knowledge about histology of melanoma to better

apprehend the following manuscript. As the field of anatomopathology is very complex, this

description is non exhaustive.

(23)

Mélanocytes in H&E stained sections présent a pale and large cytoplasm with a dark nucléus and are

located at the basal membrane of the epidermis [26]. The ratio mélanocytes/ kératinocytes in term

of number of cells is around 1 for 10 and is relatively constant between individuals. The différences

observed between tanning are due to activity of mélanocytes [27]. An important feature of the skin is

the basal membrane as it linked together the epidermis and the dermis. The dermis is a supportive

tissue for epidermis and is mainly composed of connective tissue synthetized by fibroblasts. It aiso

contains blood and lymphatic vessels, nerves and epidermal appendages (hair follicles and sweat

glands) [26].

Figure 1-5 : H&E stained section of normal skin.

1.5.2. Primary melanoma

(24)

CHAPTER I - GENERAL INTRODUaiON

Figure 1-6 : H&E stained section ofa primary cutaneous melanoma.

1.5.3. Metastatic lymph nodes

(25)

GERMINAL LYMPHOCYTES MELANOMA CAPSULE ERYTHROCYTES

CENTER CELLS

Figure 1-7 : H&E stained section ofa metastatic iymph node.

(26)

2 . HALLMARK OF CANCER

Most cancerous cells share common characteristics that hâve been rationalized and described as 8

hallmarks of cancer in the landmark reviews of Hanahan and Weinberg that give a logical framework

to this extremely compiex disease [30,31]. Cancer formation is a multistep process. The cancerous

celis acquired progressively these hallmarks through the acquisition of genetic alterations that confer

them one or several advantages. Figure 1-9 represents these essentiel acquired capabilities. Together

with the 8 hallmarks, 2 enabling characteristics crucial for capabilities acquisition are aiso described.

These are genomic instability and tumor-promoting inflammation. Tumor has been recognized as a

compiex tissue constituted with multiple distinct cell types, both normal or malignant, that

participate to the tumor progression.

Figure 1-9 : Représentation of the eight acquired capabilities and the two enabling characteristics involved in cancer development and progression (adapted from [30]^.

Sustaining Evading proliférative growth

signaling suppressors

Inducing Activating angiogenesis invasion &

metastasis

(27)

through production of growth factors combined with an increase in the level of receptors at the cell

surface. The second way is a constitutive activation of signaling pathwrays, normally triggered by

growth factors, downstream of the receptors. For example, around 50 % of melanomas contain

activating mutations in the gene coding for the BRAF protein involved in the MAP-Kinase pathway.

The MAPK pathway is a major pathway leading to activation of the cell cycle and prolifération of

cells. Alternatively, a defect in the négative loops that normally operate to dampen prolifération can

enhance proliférative signais [30]. A loss-of-function mutation in the PTEN protein occurs in more

than 25 % of melanomas, leading to enhanced prolifération [8]. Yet, an excessive increase in growth

may induce senescence. This nonproliferative but viable State is prévalent in some cases of

melanoma even though some cells can adapt to high level of oncogenic signais by disabling their

senescence or apoptotic pathways [30].

2.1.2. Evading growth suppressors

Cancer cells must aiso circumvent the powerfui programs that negatively regulate cell prolifération

through suppressor tumor genes. Indeed, prolifération of normal cell is negatively regulated by

several proteins that are gatekeeper of cell cycle progression. The genes of these proteins are called

tumor suppressor genes and are frequently mutated in cancer. The most frequently mutated genes

in cancer are RB and Tumor P53 [30]. They both hâve a central rôle in the decision either to progress

through the cell cycle or alternatively to activate senescence and apoptotic programs. In melanoma,

the most common mutated tumor suppressor gene is CDKN2A. This gene encodes two important

proteins that are gatekeepers of the cell cycle. One of them directiy interacts with P53 [8]. Another

important négative régulation of prolifération in normal tissues is the cell-to-cell contact inhibition.

However, cancerous cells develop several mechanismsto circumvent this régulation.

2.1.3. Resisting cell death

(28)

mélanocytes is not only due to characteristics of the mélanocytes themselves but is aiso related to

the stimulation by kératinocytes and fibroblasts [32]. Moreover, defects in other types of cellular

death (necrosis, autophagy, senescence...) hâve aIso been associated with cancer progression [30].

2.1.4. Enabling réplicative immortality

The three hallmarks described above allow growth of the tumor that is not regulated by

environmental signaling but these capabilities are not sufficient to explain how tumor can generate

macroscopie masses. Normal cells are able to pass through a limited number of successive division

cycles. This is mainly due to the loss of around 100 base pairs of telomeric DNA at the end of every

chromosome following each division. After a certain number of divisions, normal cells enter in

senescence and/or in crisis (cell death). The ongoing telomere maintenance is a key phenotype of

cancer cells that enable them to proliferate indefinitely. The telomerase, the specialized DNA

polymerase that extent telomeres, is almost absent in non-immortalized cells but is widely expressed

by cancer cells [30].

2.1.5. Inducing angiogenesis

Like normal cells, cancer cells need nutrients and oxygen as well as a \way to evacuate metabolic

wastes and carbon dioxide. This need is addressed by the process of new vessel formation that is

called angiogenesis. In normal conditions, this process is only transiently activated but in cancer,

angiogenesis is chronically induced following the angiogenic switch to meet the high needs of rapidiy

growing cells. The vessels that are produced within a tumor by chronic activated angiogenesis are

typically aberrant. The neovasculature is marked by convoluted and excessive vessel branching,

distorted and enlarged vessels, erratic blood flow and microhemorrhaging or leakiness [30].

Angiogenesis is regulated by proangiogenic factors such as VEGF (vascular endothélial growth factor)

or FGF (fibroblast growth factor) and by antiangiogenic factors such as thrombospondin [30,33]. The

induction of angiogenesis by production of VEGF is especially regulated by hypoxia which is defined

by a weak level of oxygen. In melanoma, highiy vascularized primary tumor was shown to correlate

with poor prognosis [12].

(29)

The first step of the process of metastasis formation involves loss of adhérence with the others

épithélial cells and the extracellular matrix. The main proteins involved in cell-to-cell adhesion and

implicated in cancer progression are the cadherins. In a normal epidermis, mélanocytes mainly

express E-cadherins, these proteins allow a cell-to-cell adhesion \with kératinocytes that aiso express

E-cadherins forming adherens junctions. In metastatic melanoma, a decrease expression of E-

cadherin and an increase of N-cadherins are observed. N-cadherins are mainly expressed by

fibroblasts of the dermis and endothélial cells of the vessels. These new cell-to-cell contacts enhance

invasion. This loss of E-cadherins is moreover involved in the cell survival as cadherins are important

proteins that regulate survival pathways [8]. Integrin is a family of proteins that médiate contact with

the extracellular matrix (collagen, fibronectin and laminin). The transition from radial to vertical

growth-phase is associated with the expression of a spécifie integrin and induces the dégradation of

the collagen of the basement membrane.

The transition of cancer cell toward an invasive phenotype was aIso described as involving the

epithelial-mesenchymal transition (EMT). This biological transition normally plays a critical rôle in

embryogenesis and was shown to be implicated in cancer invasiveness. This transition could play a

major rôle in the means that cells acquire the ability to invade, resist to apoptosis and disseminate.

However, EMT transition represents only one type of invasiveness acquisition among others [30,34].

2.1.7. Reprogramming energy metabolism

The adjustment of cell metabolism by cancer cells is considered as an emerging hallmark of cancer.

Normal cells, in aérobic condition, transform glucose into pyruvate through the glycolysis in the

cytosol and then the pyruvate into carbon dioxide in the mitochondria. In contrast, under anaérobie

condition, glycolysis is favored and only few pyruvate molécules are transferred to the oxygen-

consuming mitochondria. The cancer cells présent a reprogrammed metabolism as they limit their

energy metabolism to glycolysis, even in aérobic condition. This is known as the Warburg effect. This

seems counterintuitive, as cancer cells proliferate fast and need high rates of energy but, to

compensate, they overexpress glucose importers to increase the amount of glucose in the cytoplasm.

It was shown that the interest to favor glycolysis rather than aérobic respiration is to increase the

level of intermediates necessary to biosynthetic pathways. This generates amino acids and

nucleosides that are needed for assembling new cells [30].

2.1.8. Evading immune surveillance

(30)

transformed cells. The causes of this mechanisms can vary but they are known to be related either to

cell-autonomous modifications (e.g. a reduced expression of antigens) or to the destruction or

modification in immune System (e.g. by libération of factors that induce an immunosuppressive

microenvironment (e.g. TGF-P)) and are detailed in the next section [30,35].

2.2. Enabling characteristics

2.2.1. Genome instability

The acquisition of the hallmarks described above highiy dépends on the apparition of successive

alterations in the genome of tumor cells. Some mutations or genetic alterations confer a sélective

advantage to one clone over the others which lead to its prolifération. Alterations can be mutations

or epigenetic changes that are associated with the régulation of gene expression. Cancer cells often

increase their mutation rate. This can be achieved by increasing cell sensitivity to mutagenic agents

orthrough alterations of the genomic maintenance machinery. Melanoma was shown to be a cancer

with particularly high rate of mutations when compared to other cancers (Figure 1-10). While

genomic instability is favorable to cancer progression as it accelerates the rate at which fast growing

génotypes appear, it is aiso unfavorable as it increases the rate of appearance of mutated antigens

that immune System can recognize.

(31)

2.2.2. Tumor-promoting inflammation

Tumors are infiltrated by both innate and adaptive arms of the immune System. While, this

infiltration was thought to reflect an attempt of the immune System to eradicate tumors, we now

know that immune cells hâve the paradoxical effect of enhancing tumorigenesis and tumor

progression. Inflammation, which is particularly mediated by innate immune cells, participâtes to the

suppiy of bioactive molécules to the microenvironment and those contribute to multiple hallmarks

capabilities. These bioactive molécules include growth factors that sustain proliférative signais,

survival factors that limit cell death, pro-angiogenic factors and extracellular matrix-modifying

enzymes that facilitate angiogenesis, invasion and metastasis. Furthermore, these inflammatory cells

can release Chemicals, for example reactive oxygen species, that are mutagenic for cancer cells.

These inflammatory cells are macrophage subtypes, mast cells, neutrophils, some type of T and B

cells and myeloid derived suppressor cells (MDSCs). In physiological conditions (normal wound

healing or fighting infections), inflammation appears only transiently. On the contrary, in cancer we

observed a persistence of chronic inflammation [30].

2.3. Tumor microenvironment

(32)

3 . TUMOR-IMMUNITY: INTEGRATING

IMMUNITY'S IN CANCER PROGRESSION

Besides the numerous intrinsic tumor-suppressor mechanisms that exist to prevent development of

malignant cells, the immune System plays an important rôle in the prévention against cancer. An

evidence of this mechanism is that immunosuppressed patients hâve an increased risk of developing

a cancer. The immune System can specifically identify and eliminate tumor cells on the basis of their

tumor-specific antigens expression. This process was originally referred as immunosurveillance.

However, the rôle of immune system has been reviewed recently, as the immune System not only

provides protection against cancer but paradoxically can aiso promote cancer progression especially

by inducing an inflammatory microenvironment [35]. Melanoma is among the most immunogenic

cancer and has been the prototype for the discovery of tumor antigens and for the development of

immune-based thérapies [39].

3.1. Melanoma and the immune system

3.1.1. Immunoediting hypothesis

Nowadays, the immunosurveillance hypothesis has been revisited as immunoediting, taking into

considération the dual rôle of immune system in cancer progression. The immune system acts for the

protection of the host but aIso edits the tumor immunogenicity. This was discovered following

experiments on immunodeficient mice. It was shown that tumors formed in the absence of an intact

immune system were more immunogenic than tumors that arose in immunocompétent hosts

[35,40].

(33)

Elimination:

Different mechanisms aient the immune System to the presence of a developing tumor. Among them

are the classical danger signais such as Type 1 IFNs that are produced early in the development of

tumors and that activate dendritic cells. Other signais are the damage-associated molecular pattern

(DAMP) molécules that are released from dying tumor cells or from damaged tissues. These factors

particularly activate the innate immune System but, in most of the case, cancer immunosurveillance

response requires aiso the expression of tumor-antigens capable of propagating the expansion of

effector CD4"^ and CD8"^ T cells. To protect the host against tumor development both innate and

adaptive arms of the immune System are needed [40]. The different types of antigens in melanoma

cells that can be recognized by T cells are cancer testis antigens (e.g. Melanoma antigen-encoding

(MAGE)), mutated antigens, overexpressed antigens and différentiation antigens that are aIso

expressed by mélanocytes (e.g. Melan-A/MART-1, gpl00/HMB45) [41,42].

Healthytissue Intrinsic tumor suppression (repair, senescencc, and/or apoptosis) Carcinogens Radidtion Viral infections Chronic inflammation Inherited genetic mutations

NKG2D ligands

i

1

Elimination (cancer immunosurveillance) Equilibrium (cancer persistence/dormancy) Escape (cancer progression)

I

(g)©®

rLA^ Innate and adaptive immunity Normal ceil Highiy immur>ogenic Protection

(i.e„ extrinsk tumor suppression)

transformed ceil CTLA-4

■3''

Chronic inflammation Poorly immunogenic transformed celt lmmurK>suppFessive transformed celi Cancer immunoediting

(34)

In melanoma, a spontaneous response against the tumor exists as tumor-specific lymphocytes are

observed in blood samples of patients and spontaneous régression of the lésions may be observed in

some patients [35,42].

Equilibrium:

Rare tumor cell variants survive to the élimination phase and enter in an equilibrium phase which is

the longest phase of the immunoediting. During this phase, the adaptive immune System prevents

the net outgrowth of the tumor. The immune cells involved in this sustainable State of dormancy are

mainly CD4^ and CD8"^ T cells and is mediated via IL-12 and IFN-y but does not involve the innate

immune System anymore. However, these cells aiso provide the sélective pressure that is needed to

promote the outgrowth of less immunogenic variants, leading to the escape phase and to a clinically

apparent tumor.

Escape:

Progression from equilibrium to immune escape happens as a resuit of a change in the cancer cell

population in response to the immune System editing function and/or as a change in the immune

System of the host caused by an increased immunosuppression induced by the cancer cells or a

détérioration of the immune System itself. The tumor can escape through several mechanisms. First,

at the level of the tumor, cells can reduce their immune récognition through the loss of antigens, a

decreased expression of MHC-I proteins or through the loss of antigen processing fonctions.

Additionally, cancer cells can become résistant to cytotoxic effect of immunity in increasing for

example their level of anti-apoptotic factors [40].

Immune escape can aIso arise from the establishment of an immunosuppressive microenvironment

that is promoted by the cancer cells through the sécrétion of immunosuppressive cytokines (VEGF,

TGF-P, IDO, galectin...) and/or by the recruitment of regulatory immune cells.

(35)

they deplete amino acids that are essential for T cells functions (arginine, cysteine and tryptophan)

by producing several enzymes (Arg-1, iNOS, IDO) [35,40]. Additionally, many tumors attract

macrophages. M2 macrophages can inhibit anti-tumor immunity by secreting TGPP or IL-10.

3.1.2. Stumbling blockfor effective anticancer T cells

In addition to the mechanisms of escape described above, a weak immune response against cancer

cells can aiso be due to déficits in maturation, activation, différentiation and fonction of

lymphocytes, preventing therefore an effective destruction of the tumor cells by cytotoxic T cells

(CD8"^) and Thl CD4^ T cells. First of ail, tumor spécifie lymphocytes are subjected to central tolérance

as tumor-antigens are often self-antigens or closely related. Thus, a high number of tumor-specific

lymphocytes are eliminated or converted in regulatory phenotype in the thymus during the normal

process of central tolérance that strongly limits autoimmune response [35,43]. Secondly, tumor-

specific lymphocytes are insufficiently activated. The priming of na'we T cells is mainly mediated

through antigen-presenting cells. In cancer, innate stimulators are relatively rare, leading to weakiy

activated T lymphocytes and resulting in few effector and memory T cells spécifie of tumor-antigens

[43]. And finally, a peripheral hyporesponsiveness of anti-cancer T cells is observed. Two different

States of hyporesponsiveness hâve been described: anergy and exhaustion. T cell anergy is

characterized by a loss of expression of IL-2 under stimulation while T cells exhaustion is

characterized by a progressive loss of prolifération and cytokine production. T cells exhaustion has

been observed in melanoma métastasés and is probably due to a chronic antigen exposure of T cells

[41,44,45].

3.2. Prognosis implication of immune infiltration

There is a growing body of evidence showing that quantity, quality and spatial distribution of tumor-

infiltrating lymphocytes (TILs) correlates with patient's survival [40]. This distribution, density,

localization of TILs has been termed as immunoscore and has been suggested to be added in current

staging classification [46,47].

(36)

reported that both density and phenotype are important for prognostic of survival. A higher density

of both T cells, B cells correlated with increased survival [51]. Moreover, infiltration of TILs was aiso

associated with an improved response to some immunothérapies (e.g. the monoclonal antibody anti-

CTLA-4, ipilimumab) [52].

3.3. Immunothérapies to target melanoma

(37)

4 . INFRARED SPECTROSCOPY

4.1. Principles of IR spectroscopy

The IR région extends from the visible to the microwaves région of the electromagnetic spectrum. In

this interval, three régions of decreasing wavenumbers are defined: the near-IR région, the mid-IR

région and the far-IR région (Figure 1-12). While wavelength (X) is the usual unit of reference to

describe régions of the electromagnetic spectrum, the unit preferentially used to characterize the IR

radiation is the wavenumber (v). It represents the number of waves per centimeter and is the

reciprocal of the wavelength. The whole IR radiation ranges from 10 to about 10 000 cm ^ The mid-IR

région (4000-40 cm‘^) is the most used région to characterize biological samples. It is related to the

vibrational and rotational frequencies of the organic molécules [57,58].

Wavelength

X

Wavenumber

V

Frequency

V

Spectral range

“1— lO-* —1—T 10^ 1 ~T— 1 --- 1---10'2 r-iT" 10^ II ---1 1 1 1^ 10"®

1

10"® cm 1m _{1 mm} 1 Mm 1 nm

T—I—I—I—T

1 10 100 1000 10000 „-1 1 1 1 10® 10^° 1 1 1 10^2 I I 10’® 10’® s-' ^ Visible Radio waves Sh o rt waves Mic ro waves a: IL. IR or. s a: Z UV X-rays y-rays

Figure 1-12 : The electromagnetic spectrum from low frequencies radio waves to high frequencies y-rays. The radiations are characterized by their wavelength, wavenumber and frequency [2],

IR spectroscopy is a technique based on the vibrations of the atoms of a molécule [59]. The atoms

constituting a molécule can oscillate around an equilibrium position by changing the length or the

angle of the bonds when they are excited by quantum of energy. The frequencies of these motions

are comprised in the IR région of the electromagnetic spectrum. Elongation of bonds can be either

symmetric or asymmetric and are called stretching whereas deformations in the bond angle are

called bending. The energy ranges needed for a molécule to Jump from one vibrational level to an

excited level is found in the IR région of the electromagnetic spectrum. They dépend on the strength

of the bond, the relative masses of the atoms as well as the geometry of the atoms involved in the

bond [57].

(38)

energy will be absorbed [59]. An infrared spectrum of a sample is obtained by measuring the

intensity of an IR radiation before (7/j) and after (Ts) the passage of the beam through the sample. In

most cases, an infrared spectrum is presented as the absorbance as a fonction of the wavenumber.

The absorbance is calculated as follows A = — iogTs/Tn [58]. An infrared spectrum is commonly

represented as values of the absorbance as a fonction of the wavenumber. The IR spectrum contains

several peaks and bands which appear at spécifie wavenumbers corresponding to the energy

absorbed in rotational/vibrational transitions. Only molécules which présent a change in their

electrical dipole moment in the course of the transition will absorb energy from the IR radiation

[57,59].

4.2. Fourier Transform Infrared technology

Originally, the spectrometers were based on the use of monochromators to scan the sample one

wavenumber after the other. An enormous progress was achieved with the advent of Fourier

Transform Infrared spectrometer. This technology, combining the Michelson interferometer and the

mathematical Fourier Transform, collects simultaneousiy the data for ail the frequencies [58]. Most

of the modem FTIR spectrometers are nowadays composed of the following éléments: an internai IR

light source, a Michelson interferometer and a single element detector [57]. The schematic

représentation of a Michelson interferometer appears in Figure 1-13. It is composed of a beam

splitter and two perpendicular mirrors (one fixed and one moving). The principle of the

interferometer is to split the IR beam from the source into two beams (a half toward the fixed mirror

and half toward the moving mirror) and to reconstruct them at the sample position leading to

constructive and destructive interférences according to the position of the moving mirror.

(39)

The detector will measure intensity of the beam transmitted as a function of the position of the

moving mirror. It corresponds to the interferogram. The infrared spectrum can be calculated via the

application of FourierTransform to the interferogram [58]. The peak intensifies reflect the amount of

molecular bonds absorbing at different wavenumbers of the IR spectrum and is related to the

relative abundance of different sample constituents [57].

This new development in IR spectroscopy has greatly decreased the acquisition time (from hours to

few seconds) and improved the quality of the spectra [59]. As ail the wavenumbers are recorded

simultaneousiy in one single measurement, the recording time was considerably reduced and

therefore it was possible to record several scans from the same sample leading to an increase of the

Signal to Noise ratio (S/N). Another advantage of the interferometer is that ail the energy of the

source reaches the detector and there are no longer slits that limit the amount of energy. The

spectral resolution is solely related to the maximum distance reached by the moving mirror. The

position of the mobile mirror is followed by an internai laser and the spectral resolution can reach

0.01 cm'^ [58,59].

4.3. FTIR micro-spectroscopy

The combination of the infrared spectroscopy with microscopy has led to a new technology termed

infrared micro-spectroscopy. It combines the Chemical identification of the sample components by IR

spectroscopy and the spatial resolution of the microscopy [60].

Two different configurations are possible depending on the detector selected;

• Mapping: this technique uses a single element detector and the restriction for spatial

resolution appears at the incoming radiation. Apertures restrict the illumination to a single

point at the sample plane and the measurement is achieved point by point by moving a XY

stage [60,61].

• Imaging: this second technique is based on the use of multichannel detectors and is now the

standard for microscopy. The multichannel detectors are referred to as Focal Plane Array

(FPA) detectors and are composed of thousands of single éléments. The entire field of view is

illuminated and each individual detector (pixel) records the transmitted radiation from a

spécifie sample région [61,62].

(40)

On the contrary, the imaging configuration considerably decreases the acquisition time and reaches a

good signal to noise ratio level. The best spatial resolution achieved is the pixel size but the actual

spatial resolution highiy dépends on the diffraction limit which ranges from 5.5 to 10 pm for IR light

[62]. The FPA used in the work is composed of 64x64 pixels of around 2.8x2.8 pm^ generating 4096

entire IR spectra from an area of 180x180 pm^.

The standard micro-spectrometers are equipped with both configurations as represented in Figure

1-14.

Figure 1-14 : Schematic représentation ofa modem imaging micro-spectrometer. The main components are the IR source, the Michelson interferometer, the microscope (objective and condenser), the single-point detector

(mapping configuration) and the FPA detector (imaging configuration) [7],

Different sampling modes exist: transmission, transfiection and attenuated total reflection (ATR). The

transmission mode is the most frequently used and is the one used throughout this work. The

infrared beam passes through the sample and the transmitted beam is detected (Figure 1-15). The

support to deposit the sample needs therefore to be transparent to IR radiation (e. g. barium fluoride

or calcium fluoride). The signal to noise ratio is good even though the samples are limited to a few

(41)

Transmission Transfiection

Figure 1-15 : The three main sampling modes for FTiR spectroscopy: transmission, transfiection and attenuated total reflection [7],

4.4. Applications to biological molécules

For a number of décades, infrared spectroscopy bas been extensively used to study and characterize

isolated biological molécules such as proteins, lipids and nucleic acids. This technique is either

applicable to solid, liquid or powder [59].

In the particular case of the study of proteins, IR spectroscopy was shown to bring valuable

information on the secondary structure [63-65]. The amide vibrations are the largest bands in the IR

spectra of proteins. Nine bands are characteristic of the amide groups but only three of them are

usually used for the investigation of the protein secondary structure. These bands are mainly based

on the vibrations of the amide bonds of proteins. The main contribution to the amide I band is the

stretching of C=0 groups while the main contribution to amide II band originates from N-H bending.

The exact wavelength corresponding to the absorption of the amide I band (from 1600 to 1700 cm'^)

and the amide II band (from 1600 to 1500 cm'^) dépends on the geometry of the polypeptide chain

and the strength of the hydrogen bonds involving C=0 and N-H groups. Depending on the secondary

structure of the protein studied, the frequencies at which the amide I and II bands absorb will change

[63,64].

(42)

Among the numerous studies carry out on nucleic acids, the ones of Malins and colleagues on DNA

extracted from various cancers show a significant différence between DNA of malignant and normal

cells [69-71]. The main bands in the spectrum of nucleic acids can be divided into the contribution of

its main constituents; sugars and phosphate groups [59]. Several other research were dedicated to

the study of hydration or déhydration of biological molécules (proteins, DNA and phospholipid films)

[72-74]. These studies evidenced that several bands in the IR spectrum are sensitive to

hydration/dehydration. These results are important as samples used for IR analyses are often

dehydrated to avoid the contribution of the water in the IR spectrum.

4.5. Applications to prokaryotic and eukaryotic cells

As a rapid and sensitive technique, IR spectroscopy has rapidiy proved to be a valuable tool to study

more complex biological organisms such as bacteria and eukaryotic cells. Given the complex nature

of such organisms, the spectrum of a cell is a superimposition of its component spectra (proteins,

nucleic acids, lipids, carbohydrates...) as represented in Figure 1-16.

Figure 1-16 : Infrared spectra ofthe main biological macromolecules constituting cells (DNA, RNA, lipids and proteins) and a eukaryotic cell spectrum. Protein 1 is rich in 6 sheet and protein 2 is rich in a-helix.

(43)

• The area between 3100-2800 cnr^ is dominated by bands arising from C-H vibrations and is

usually attributed to the acyl chains of lipids in biological samples.

• In the 2800-1800 cm'^ région, there is almost no absorption of biological molécules. A double

peak appears near 2300 cm'^ due to absorption of atmospheric CO

2

.

• Between 1700 and 1500 cm ^ the IR spectrum is dominated by bands related to proteins (i.e.

amide I and amide II bands described above).

• The 1400-900 cm'^ spectral région contains superimposed bands from the nucleic acids,

carbohydrates and phospholipids (head groups). In this région, absorbing groups are the P02'

and the C-O-O-C as represented in Figure 1-17.

Figure 1-17 : Typical spectrum ofa cell showing biomolecular bands assignments. The stretching vibrations are represented by the Symbol v and the bending by 6 (either symmetric or asymmetric). [75].

(44)