Cutaneous T-cell lymphoma pathogenesis : extracellular vesicles, syncytin-1, and metabolites

43 ISBN 978-951-51-7464-2 (PRINT)

ISBN 978-951-51-7465-9 (ONLINE) ISSN 2342-3161 (PRINT) ISSN 2342-317X (ONLINE)

http://ethesis.helsinki.fi HELSINKI 2021

AUKKANEN CUTANEOUS T-CELL LYMPHOMA PATHOGENESIS: EXTRACELLULAR VESICLES, SYNCYTIN-1, AND METABOLITES

dissertationesscholaedoctoralisadsanitatem investigandam universitatishelsinkiensis

DEPARTMENT OF DERMATOLOGY AND ALLERGOLOGY FACULTY OF MEDICINE

DOCTORAL PROGRAMME IN INTEGRATIVE LIFE SCIENCE UNIVERSITY OF HELSINKI

CUTANEOUS T-CELL LYMPHOMA PATHOGENESIS:

EXTRACELLULAR VESICLES, SYNCYTIN-1, AND METABOLITES

KIRSI LAUKKANEN

Doctoral Programme in Integrative Life Science Faculty of Medicine

University of Helsinki Finland

Cutaneous T-cell lymphoma

pathogenesis: extracellular vesicles, syncytin-1, and metabolites

Kirsi Laukkanen

DOCTORAL DISSERTATION

To be presented for public examination, with the permission of the Faculty of Medicine, University of Helsinki, in the Lecture hall 2, Biomedicum Helsinki,

Haartmaninkatu 8,

on 8th October 2021, at 12 noon.

Helsinki 2021

Supervised by

Professor Annamari Ranki

Department of Dermatology and Allergology Clinicum, University of Helsinki

Helsinki, Finland

Reviewed by

Professor Jyrki Heino

BioCity Turku, University of Turku Turku, Finland

Docent Kirsi Rilla

University of Eastern Finland Kuopio, Finland

Opponent

Professor Veli-Matti Kähäri Department of Dermatology University of Turku

Turku, Finland

The Faculty of Medicine uses the Urkund system (plagiarism recognition) to examine all doctoral dissertations.

ISBN 978-951-51-7464-2 (print) ISBN 978-951-51-7465-9 (PDF) ISSN 2342-3161 (print)

ISSN 2342-317X (online) http://ethesis.helsinki.fi

Cover: Artwork Elli Nieminen, design Anita Tienhaara.

Unigrafia

To my dearest daughters Heini and Suvi,

TIIVISTELMÄ

Ihon T-solulymfoomat ovat kroonisia ihosairauksia, joiden ilmaantuvuus on kasvanut maailmanlaajuisesti. Ne jaetaan yhdeksään eri alaryhmään. Sairaudelle on ominaista pahanlaatuisiksi muuntuneiden imusolujen, T-lymfosyyttien kertyminen ensisijaisesti ihoon. Oireet vaihtelevat ihossa esiintyvistä ekseemaa tai psoriaasia muistuttavista läiskistä aggressiivisiin ihokasvaimiin. Pahanlaatuiset T-solut voivat taudin edetessä levitä imusolmukkeisiin sekä muualle elimistöön. Ihon T-solulymfoomien varhainen diagnosointi ja erottaminen muista ihosairauksista on vaikeaa. Tämän vuoksi taudin toteaminen usein viivästyy jopa vuosia ensimmäisten oireiden ilmaantumisesta. Ihon T-solulymfoomia on tutkittu jo pitkään, mutta edelleenkin niiden syntymekanismi on tuntematon.

Tämän tutkimuksen tarkoituksena oli selvittää kuinka ihon T-solulymfooman mikroympäristön tekijät vaikuttavat taudin kehittymiseen. Aiempien tutkimusten mukaan elimistön välttämättömät aminohapot synnyttävät immuniteettiä hillitseviä molekyylejä. Solut, joissa on aminohappoja hajottavia entsyymejä, muodostavat suojaverkon kasvainsolujen ympärille eikä elimistön oma immuunivaste pääse tuhoamaan niitä. Tryptofaanin kataboliareitin indoliamiini 2,3-dioksigenaasi 1 (IDO1) on tällainen entsyymi ja se toimii hajottamalla tryptofaanin kynureniiniksi.

IDO1 on immuunijärjestelmän säätelijä, joka estää puolustusjärjestelmää tuhoamasta syöpäkasvainta. Vähemmän tutkittu tryptofaani 2,3-dioksigenaasi (TDO) -entsyymi vaikuttaa samaan metaboliareittiin. Tulokset osoittivat, että molemmat entsyymit ilmentyvät ihon T-solulymfoomista eristetyissä solulinjoissa, syöpäsoluissa sekä kasvainta ympäröivissä tulehduksellisissa soluissa. Lisäksi eri iholymfooma-alatyypeillä oli oma erityinen ilmentymisprofiilinsa. Tutkimalla potilaiden plasmanäytteitä selvitimme, että ihon T-solulymfoomien yleisimmän alatyypin (mycosis fungoides) potilailla kynureniini/tryptofaanisuhde oli korkeampi verrattuna terveisiin kontrolleihin, mikä oli myös merkkinä pitkälle edenneestä sairaudesta. Kartoitimme tutkimuksessamme myös muita tryptofaanin kataboliareitin metabolitteja. Löydöksemme mukaan neljä metabolittia olivat yli-ilmentyneitä ja kaksi ali-ilmentyneitä mycosis fungoides-potilailla verrattuna terveisiin verrokkeihin.

Tutkimuksessa selvitettiin myös syöpäsolujen erittämien solunulkoisten vesikkelien merkitystä syövän mikroympäristössä. Solut käyttävät vesikkeleitä viestintään ja voivat siirtää niihin pakattuja proteiineja, lipidejä, nukleiinihappoja sekä metaboliitteja verenkierron välityksellä kauaskin alkuperäisestä solusta.

Vesikkeleillä on tärkeä rooli syövän kehityksessä. Niiden avulla syöpäsolu voi kuljettaa syövälle ominaisia molekyylejä terveisiin soluihin sekä muodostaa

helposti siirtymään kudoksesta toiseen jo ennen kuin syöpää on välttämättä edes havaittu. Tämän vuoksi ne ovat hyviä merkkiaineita syövän varhaiseen toteamiseen.

Ihmisen genomiin on muinaisten retrovirusinfektoiden jäljiltä jäänyt endogeenisiä retroviruksen geenejä. Eräs tällainen retrovirusgeeniä ilmentävä proteiini on synsytiini-1, jolla tiedetään olevan keskeinen tehtävä istukkakudoksen synnyssä. Tutkimuksessamme synsytiini-1 oli yli-ilmentynyt ihon T-solulymfoomista eristetyissä solulinjoissa sekä solunulkoisissa vesikkeleissä. Synsytiini-1 on osallisena solujen välisissä fuusioissa. Solufuusiot syövissä aiheuttavat genomista epätasapainoa, aneuploidiaa, kasvainsolujen heterogeenisuutta, lääkeaineresistenssiä sekä etäpesäkkeiden muodostumista. Toiminnallinen tutkimuksemme osoitti, että vesikkeleihin sitoutunut synsytiini-1 lisäsi vastaanottajasoluissa suuria fuusioituneita soluja. Lisäksi tutkimuksemme kartoitti ihon T-solulymfoomien, eturauhassyövän ja paksusuolen syövän solunulkoisten vesikkelien metabolomiikkaa.

Osoitimme, että näiden syöpien aineenvaihdunnassa tapahtui muutoksia verrattuna vastaavien kudosten hyvänlaatuisiin soluihin. Aiemmin syövälle tyypillisiä, eri tavoin ilmentyviä metaboliitteja on löydetty syöpäsoluista, mutta solunulkoisten vesikkelien roolia on tutkittu vähemmän. Tulokset osoittivat, että kaikilla tutkituilla syöpätyypeillä oli yhteisiä vesikkeleihin sitoutuneita metaboliitteja.

Proliini ja sukkinaatti olivat merkitsevästi yli-ilmentyneet ihon T-solulymfoomien, eturauhassyövän ja paksusuolen syövän solulinjojen vesikkeleissä. Lisäksi foolihappo ja kreatiniini olivat yli-ilmentyneet ihon T-solulymfooman ja eturauhassyövän solujen vesikkeleissä.

Tämä tutkimus tarjoaa uutta tietoa syövän mikroympäristöstä ihon T-solu lymfoomassa, eturauhassyövässä sekä paksusuolen syövässä. Erityisesti tutkimus selvitti solunulkoisten vesikkelien ja tryptofaanin hajotukseen osallistuvien immunosuppressivisten metaboliittien osuutta ihon T-solulymfoomien synnyssä.

Tuloksia voidaan hyödyntää ihon T-solulymfoomien varhaisemmassa ja tarkemmassa diagnosoinnissa sekä uusien lääkeaineiden kehityksessä. Lisäksi tutkimuksen uutena löydöksenä oli aiemmin syövän kehitykseen liitettyjen metaboliittien yli-ilmentymisen toteaminen syöpäsolujen solunulkoisissa vesikkeleissä. Näitä metaboliitteja on mahdollista tulevaisuudessa hyödyntää reaaliaikaisena syövän merkkiaineena. Tutkimustuloksemme lisäävät myös yleisemminkin tietoa syövän leviämiseen ja syntyyn vaikuttavista mekanismeista.

ABSTRACT

Cutaneous T-cell lymphomas (CTCLs) constitute an incurable, chronic, heterogeneous group of non-Hodgkin lymphomas, characterized by a malignant population of mature T-lymphocytes that infiltrate the skin. The incidence of CTCLs has increased worldwide. Symptoms vary from indolent skin patches to aggressive skin tumors. In addition to the skin, lymph nodes, blood, and other body organs are affected depending on the stage and subtype of the disease.

Difficulties in differentiating early symptoms from other skin conditions and complex diagnostic criteria make CTCLs challenging to diagnose, often delaying the diagnosis for years. Despite intensive research, the pathogenesis of CTCLs remains mostly unknown.

This study emphasizes the impact of the tumor microenvironment (TME) and the tumor cell-released extracellular vesicles (EVs) on the pathogenesis of CTCLs.

Particularly, this study examines the role of the tryptophan pathway-associated enzymes indoleamine 2,3-deoxygenase 1 (IDO1) and tryptophan 2,3-dioxygenase (TDO). IDO induces immune tolerance in the TME by inflammation and T-cell activation. While TDO catabolizes the same metabolic pathway as IDO1, its role in tumors is poorly characterized. We showed that both IDO1 and TDO are expressed in CTCL tissues and cell lines. Interestingly, different CTCL subtypes show unique patterns of expression. In addition, elevated serum kynurenine/tryptophan ratios observable in plasma samples of mycosis fungoides (the most common subtype of CTCL) patients correlated with advanced disease.

Furthermore, four tryptophan catabolic route metabolites were upregulated, and two were downregulated. These observations could be exploited in clinical tests, and therapeutic potential for CTCLs may result from blocking IDO activity.

EVs are small membranous vesicles released into the extracellular milieu by most cell types – also by cancer cells. The EVs released from cancer cells contain various proteins, lipids, amino acids, and metabolites and have proven to be an essential form of intercellular communication. Malignant cells can release EVs into TME and the circulation to reprogram target cells or prepare a pre-metastatic niche. Such transfer could occur at a very early stage in tumor progression because the small size of EVs allows them to cross barriers that cells cannot. Consequently, the molecular cargo of EVs reflects the cell of origin, which is also detectable in the circulation by the non-invasive liquid biopsy method.

This study shows that human endogenous retrovirus, type W (HERV-W)-coded syncytin-1 in CTCL cells and CTCL cell-derived EVs is upregulated. Functionally, upregulated syncytin-1 can promote cell-to-cell fusion. Cell-to-cell fusion in cancers can induce genomic instability and aneuploidy and contribute to tumor

of EVs are not fully characterized, syncytin-1 could be involved in binding EVs to target cells to facilitate the progressing fusion. Our functional study suggested that EV-harbored syncytin-1 promoted giant plasma membrane-surrounding fusion cells in the recipient T-cell leukemia cells. In addition, we studied alterations of EV-derived metabolite cargo in CTCL, prostate carcinoma, and colon carcinoma cell lines and compared these with the cargo of their healthy counterparts. In cancers, the expression of metabolites is commonly reregulated due to oncogenes or tumor suppressor mutations. The enhanced metabolism supports cancer cell proliferation and survival. Previously, numerous cancer-associated metabolites have been detected, but little is known of their role in cancer-derived EVs. This study shows that despite the differences among the studied cancer types, all cell line-derived EVs shared a common metabolomic feature: upregulation of proline and succinate. In addition, in CTCL- and prostate cancer-derived EVs, folate and creatinine were upregulated.

This doctoral thesis explores the role of EVs and the kynurenine pathway in the pathogenesis of CTCLs. These observations could impact diagnosis and have therapeutic value for currently incurable, complex CTCLs. In addition, our results from three different cancer cell line-derived EVs show that metabolomic reprogramming of cancer cells influences the EV metabolome. In particular, this finding suggests that proline, succinate, folate, and creatinine could constitute a metabolomic fingerprint of cancer, which could serve as a peripherally detectable cancer marker. Thus, exploring the function of EVs in cancer progression will prove valuable in next-generation cancer diagnosis and treatment.

Contents

TIIVISTELMÄ � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �4 ABSTRACT � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �6 LIST OF ORIGINAL PUBLICATIONS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 11 ABBREVIATIONS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 12 1 INTRODUCTION � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 14 2 REVIEW OF THE LITERATURE � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 15 2 . 1 CUTANEOUS T-CELL LYMPHOMA � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 15 2�1�1 Occurrence, classification, and treatment � � � � � � � � � � � � � � � � � � � � � 15 2�1�2 Pathogenesis � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 18 2 . 2 PROSTATE CANCER AND COLON CANCER � � � � � � � � � � � � � � � � � � � � � � � � � � � 20 2�2�1 Prostate cancer � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 20 2�2�2 Colon cancer � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 21 2 . 3 TRYPTOPHAN CATABOLISM � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 22 2 . 4 EXTRACELLULAR VESICLES � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 24 2�4�1 Mechanism of extracellular vesicle uptake � � � � � � � � � � � � � � � � � � � � � 26 2�4�2 Extracellular vesicles: function in cancers � � � � � � � � � � � � � � � � � � � � � 27 2�4�3 Extracellular vesicles as a cancer biomarker � � � � � � � � � � � � � � � � � � � 30 2 . 5 SYNCYTIN-1 AND ITS ROLE IN HEALTH AND DISEASE � � � � � � � � � � � � � � � � � � � � 31 2�5�1 Syncytin-1 in extracellular vesicles � � � � � � � � � � � � � � � � � � � � � � � � � � 34 2 . 6 CANCER METABOLISM � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 34 2�6�1 Aerobic glycolysis in cancers � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 35 2�6�2 Oncometabolites � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 36 2�6�3 Metabolomics and EVs � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 36 3 AIMS OF THE STUDY � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �40 4 MATERIALS AND METHODS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 41

4 . 1 CELL LINES AND CELLS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 43 4 . 2 PATIENTS AND SPECIMENS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 44 4 . 3 RELATIVE GENE EXPRESSION � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ⁴⁵ 4 . 4 IMMUNOASSAYS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 45

4�4�1 Immunohistochemistry, immunofluorescence,

and immunocytochemistry � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 45 4�4�2 Western blot � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 47 4�4�3 Immunoelectron microscopy � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 48 4 . 5 TRYPTOPHAN PATHWAY ASSAYS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 49 4�5�1 ELISA assay to measure kynurenine concentrations � � � � � � � � � � � � � � 49 4�5�2 In vitro inhibition assay of IDO1 � � � � � � � � � � � � � � � � � � � � � � � � � � 50 4�5�3 Mass spectrometry analysis of metabolites

from the tryptophan pathway � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 50 4 . 6 EV ISOLATION AND CHARACTERIZATION � � � � � � � � � � � � � � � � � � � � � � � � � � � � 51 4 . 7 FUNCTIONAL ASSAYS OF SYNCYTIN-1 � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 51 4�7�1 Syncytin-1 knock-down assay � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 51 4�7�2 Cell fusion assay � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 52 4 . 8 METABOLITE ANALYSIS OF EVs WITH LIQUID CHROMATOGRAPHY-MASS

SPECTROMETRY � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 52 4 . 9 STATISTICAL ANALYSIS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � ⁵⁴ 4 . 10 ETHICAL CONSIDERATIONS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 54 5 RESULTS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 55 5 . 1 CTCL SUBTYPES SHOW A UNIQUE EXPRESSION PATTERN OF IDO1 AND TDO (I) � � � 55 5 . 2 SEVERAL CELL TYPES EXPRESS IDO1 IN CTCL TUMOR MICROENVIRONMENT (I) � � � 57 5 . 3 INCREASED PLASMA Kyn/Trp RATIO REFLECTS AN ADVANCED

DISEASE STAGE OF MF (I) � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 58 5 . 4 EPACADOSTANT INHIBITS MALIGNANT MYCOSIS FUNGOIDES CELLS (I) � � � � � � � � 58

5 . 5 METABOLITES FROM THE TRYPTOPHAN PATHWAY ARE DIFFERENTIALLY

EXPRESSED IN PATIENT SERUM (I) � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 59 5 . 6 CTCL CELL LINES EXPRESS SYNCYTIN-1 (II) � � � � � � � � � � � � � � � � � � � � � � � � � � 61 5 . 7 CHARACTERIZATION OF EVs (II, III) � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 63 5 . 8 CTCL CELL LINES AND CTCL CELL LINE-DERIVED EVs EXPRESS CD30 (II) � � � � � � � 64 5 . 9 SYNCYTIN-1 AND ALTERED METABOLITE PROFILE

IN CELL LINE-DERIVED EVs (II, III) � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 66 5�9�1 CTCL cell-derived EVs harbor fusogenic syncytin-1 � � � � � � � � � � � � � � 66 5�9�2 EVs derived from CTCL cells carry altered metabolite cargo � � � � � � � � 67 5 . 10 CANCER CELL-DERIVED EVs HARBOR ALTERED METABOLOMICS (III) � � � � � � � � � �71 6 DISCUSSION � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 74 6 . 1 TRYPTOPHAN CATABOLISM IS ALTERED IN CTCL � � � � � � � � � � � � � � � � � � � � � � � 74 6�1�1 Kyn / Trp ratio is a possible biomarker of MF � � � � � � � � � � � � � � � � � � � 75 6�1�2 Combined IDO1

/

EXTRACELLULAR VESICLES IS ALTERED � � � � � � � � � � � � � � � � � � � � � � � � � � � � � 84 6�3�1 Metabolic fingerprint of extracellular vesicles of CTCL, � � � � � � � � � � 84 prostate carcinoma, and colon carcinoma cells � � � � � � � � � � � � � � � � � � � � � � 84 6�3�2 Extracellular vesicles in tryptophan catabolism � � � � � � � � � � � � � � � � � 86 7 CONCLUSION � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �88 8 ACKNOWLEDGMENTS � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �89 9 REFERENCES � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � � �92

LIST OF ORIGINAL PUBLICATIONS

This thesis is based on the following publications:

I

II

III

Maliniemi P, Laukkanen K, Väkeva L, Dettmer K, Lipsanen T, Jeskanen L, Bessede A, Oefner P, Kadin M, Ranki AM. Biological and clinical significance of tryptophan-catabolizing enzymes in cutaneous T-cell lymphomas.

Oncoimmunology 2017;Feb 10;6(3):e1273310.

Laukkanen K, Saarinen M, Mallet F, Aatonen M, Hau A, Ranki AM. Cutaneous T-Cell Lymphoma (CTCL) Cell Line-Derived Extracellular Vesicles Contain HERV-W-Encoded Fusogenic Syncytin-1. Journal of Investigative Dermatology 2020;140,1466-1469.e4.

Palviainen M*, Laukkanen K*, Tavukcuoglu Z, Velagapudi V, Kärkkäinen O, Hanhineva K, Ranki AM, Siljander P. Cancer Alters the Metabolic Fingerprint of Extracellular Vesicles. Cancers (Basel) 2020;12(11):3292.

*) Authors contributed equally to this paper.

Publication I, in its original submitted version, has been part of the PhD thesis of Pilvi Maliniemi (Novel factors in the pathogenesis of cutaneous T cell lymphoma, Helsinki, 2015), at the University of Helsinki, 2015.

These publications are referred in the text by their Roman numerals and have been reprinted with permission from their copyright holders.

ABBREVIATIONS

ADCC antibody-dependent cellular cytotoxity

ALS amyotrophic lateral sclerosis ASCT amino acid transporter BMDC bone marrow-derived

dendrite cell

CAF cancer-associated fibroblast CDO1 cysteine dioxygenase 1 CRC colorectal cancer CTC circulating tumor cell CTCL cutaneous T-cell lymphoma CTLA-4 cytotoxic T-lymphocyte-

associated protein 4 DMSO dimethyl sulfoxide DSS disease-specific survival ECM extracellular matrix

EGFR epidermal growth

factor receptor ELISA enzyme-linked

immunosorbent assay

EM electron microscopy

EMT epithelial-to-mesenchymal transition

ERV endogenous retrovirus EV extracellular vesicle FasL cell death ligand

FBS fetal bovine serum

FC fold-change

FFPE formalin-fixed

paraffin-embedded

FR folate receptor

GAPDH glyceraldehyde-3-

phosphate dehydrogenase

GPC1 glypican 1

HELLP hemolysis elevated liver enzyme low platelet count

HERV human endogenous

retrovirus

HMDB human metabolome

database

IA immune-affinity technique

ICC immunocytochemistry

IDO1 indoleamine

2,3-deoxygenase 1

IDO2 indoleamine

2,3-deoxygenase 2

IEM immunoelectron microscopy

IF immunofluorescence

IHC immunohistochemistry

IFN-γ interferon gamma IPMC intracellular plasma

membrane-connected compartments

IUGR intrauterine growth restriction

KAT kynurenine aminotransferase

KMO kynurenine-3-

monooxygenase

KRAS Kirsten rat sarcoma viral oncogene homolog Kyn/Trp kynurenine and

tryptophan ratio

KYNU kynureninase

LAG-3 lymphocytic- activation gene 3 LC-ESI-MS/MS liquid chromatography-

electrospray ionization- tandem mass spectrometry LC-MS liquid chromatography-

mass spectrometry LC-qTOF-MS quadrupole time-of-flight

mass spectrometry LPR lichen ruber planus

LPS lipopolysaccharide

LyP lymphomatoid papulosis MDSC myeloid-derived

immunosuppressive cell MetPA metabolomics

pathway analysis

MF mycosis fungoides

MHC major histocompatibility complex

MISEV Minimal Information for Studies of

Extracellular Vesicles M-MDSC monocyte-related myeloid

-derived suppressor cell MRD minimal residual disease

MS mass spectrometry

MSC mesenchymal stem cell MSEA metabolite set

enrichment analysis MSI microsatellite instability MVB multivesicular body NAD+ nicotinamide adenine

dinucleotide

NK natural killer

NTA nanoparticle

tracking analysis

PBMC peripheral blood

mononuclear cell pcALCL primary cutaneous

anaplastic large cell lymphoma

PD-1 programmed death 1

PD-L1 programmed death 1 ligand

PFA paraformaldehyde

PMN-MDSC polymorphonuclear myeloid -derived suppressor cell PSA prostate-specific antigen ROS reactive oxygen species

SDMA symmetric dimethylarginine SDS sodium dodecyl sulfate

SEC size exclusion

chromatography scRNA-seq single-cell RNA

sequencing analysis SPTCL subcutaneous panniculitis-

like T-cell lymphoma STAT3 signal transducer and

activator of transcription 3

SU surface subunit

SV40 Simian vacuolating virus 40

TAM tumor-associated

macrophage TCA tricarboxylic acid

TDO tryptophan 2,3-dioxygenase TEM transmission electron

microscope

T_CM central memory T-cell T_EM effector memory T-cell TGF-β transforming

growth factor β

Th-1 T-helper 1

Th-2 T-helper 2

TM transmembrane subunit

TME tumor microenvironment

TNF tumor necrosis factor

TRAIL tumor necrosis factor-related apoptosis-inducing ligand Treg regulatory T-cell

UC ultracentrifugation

WB Western blot

WHO-EORTC World Health Organization -European Organization for Research and Treatment of Cancer

1. INTRODUCTION

Cancer is listed as the first or second cause of death before the age of 70 years in many countries, accounting for 19.3 million new cases of cancer and almost 10.0 million cancer-related deaths in 2020.¹ In the body, cancer comprises of a heterogeneous group of proliferating cells and the cancer microenvironment, consisting of stromal cells, extracellular matrix, infiltrating inflammatory cells and secreted factors. Crosstalk between malignant cells and their microenvironment is essential for angiogenesis, invasion, and metastasis, all necessary for tumor progression.^{2, 3}

Immune cells are recruited to the tumor site to eliminate cancer cells, but their anti-tumor function is largely restricted to response to tumor-derived signaling.⁴ Cancer cells can remodel their microenvironment by the secretion of soluble factors, including cytokines, chemokines, and growth factors. Consequently, they reprogram the surrounding cells to facilitate tumor survival and progression.⁵ A variety of biomolecules are packed into small EVs enclosed or exposed on their surface, including DNA, RNAs, lipids, metabolites, and proteins to facilitate communication between cancer and microenvironment cells.⁶

EVs were first been observed as platelet-derived particles in normal plasma, referred to as “platelet dust,” and in 1975, lipid bilayer vesicles were discovered to be released from rectal adenocarcinoma.⁷ At present, EVs are known as key mediators in cell-to-cell communication by delivering messages from parent cells to the recipient cells even long-distance via the circulation. Through their parent- cell derived molecules, EVs can promote tumor growth, epithelial-mesenchymal transition, vascular leakiness, and drug resistance. They also prepare the pre- metastatic niche, modulate the extracellular matrix, and impact the immune system.⁸ Consequently, EV-based therapy is anticipated to emerge as a novel treatment option for cancers. Numerous therapeutic strategies, including eradication of cancer-associated circulating EVs, inhibition of EV secretion, and ablation of EV internalization, have been proposed.⁹ Several preclinical and clinical studies are ongoing, including the use of EVs as biologically active drug delivery entities or immune system activators.¹⁰ Interest in the use of EVs for disease monitoring has also increased in the past few years. Tumor-derived EVs circulate in different body fluids and are readily detectable with non-invasive liquid biopsies. Their potential as a cancer biomarker is based on the reflection of cells of origin for longitudinal prognosis monitoring and early relapse detection.¹¹

2. REVIEW OF THE LITERATURE

2 . 1 CUTANEOUS T-CELL LYMPHOMA

Primary cutaneous lymphomas are a heterogeneous group of extranodal non- Hodgkin T- and B-cell lymphomas. In the Western world, cutaneous T-cell lymphomas (CTCLs) constitute 75–80% of all primary cutaneous lymphomas.¹² In CTCL, skin-homing or skin-resident T-cells undergo malignant proliferation and induce skin symptoms ranging from indolent patches to aggressive skin tumors. In the early disease stages, the symptoms may resemble inflammatory skin conditions such as psoriasis or chronic eczema. Commonly, the disease progress is indolent with a favorable prognosis. However, in the advanced stage, the malignant T-cells accumulate, expand, and form tumors. In addition, malignant T-cells may spread to the lymphatic system, blood, or internal organs throughout the body, resulting in a fatal outcome. Despite intensive research for years, the etiology of CTCL remains largely unknown, rendering CTCL a challenging diagnosis.^13,14

2�1�1 Occurrence, classification, and treatment

An increasing incidence of CTCLs has been documented since the 1970s. In the United States, the annual incidence has increased from 6.4 (1973–2002) to 7.7 per million individuals (2001–200w5).¹⁵ Likewise, a recent study from the Finnish population shows that the prevalence of two common CTCL subtypes has increased between the years 1998 and 2016.¹⁶ The prevalence of mycosis fungoides (MF) has increased from 2.04 to 5.38 per 100 000 individuals. The Sézary syndrome has also increased from 0.16 to 0.36 per 100 000 inhabitants.

CTCLs are classified into nine different subtypes (Table 1). MF is the most common form. According to the WHO-EORTC (World Health Organization- European Organization for Research and Treatment of Cancer) classification, the MF variants comprise a distinct group from the classical MF, based on the clinicopathologic features, clinical behavior, and prognosis.¹² Of MF patients, approximately 75% are diagnosed after 50 years of age. In the North American and European populations, the occurrence of juvenile MF varies between 0%

and 5%. However, juvenile MF in Asia is more common, with a prevalence of approximately 25%.¹⁷ MF is understood to arise from skin-resident effector memory T-cells (T_EM), producing inflammatory cytokines and presenting as skin-

*) Frequency (%) is calculated from all primary cutaneous lymphomas. Data from Dutch and Austrian cutaneous lymphoma registers between 2002 and 2017. NK = natural killer, DSS = disease-specific survival. Based on the data of the WHO-EORTC 2018 classification. Modified

Classification Frequency * (%) 5-year * DSS

Mycosis fungoides (MF) Mycosis fungoides variants

• Folliculotropic MF

• Pagetoid reticulosis

• Granulomatous slack skin

39 5< 1

< 1

88 75100 100

Sézary syndrome 2 36

Adult T-cell leukemia/lymphoma < 1 Data

unavailable Primary cutaneous CD30+ lymphoproliferative disorders

• Primary cutaneous anaplastic large cell lymphoma

• Lymphomatoid papulosis

8 12

95 99 Subcutaneous panniculitis-like T-cell lymphoma 1 87 Extranodal NK/T-cell lymphoma, nasal type < 1 16

Chronic active EBV infection < 1 Data

unavailable Primary cutaneous peripheral T-cell lymphoma, rare

subtypes < 1–6 11–100

Primary cutaneous peripheral T-cell lymphoma, not

specified 2 15

limited patches, plaques, or tumors, commonly in sun-protected areas¹⁴ (Figure 1A and B). Typically, malignant T-cells cluster within the epidermis of vesicle- like structures with surrounding haloes, so-called Pautrier’s microabscesses.^18, 19 The skin lesions often remain stable for years.¹⁴ Among MF patients, 71.5%

are diagnosed in the early stage (IA–IIA) and 28.5% in the advanced stage (IIB–IVB).²⁰ A small subset of these patients develops more advanced disease with the involvement of blood, lymph nodes, or other body organs.¹⁴

Table 1. Classification, frequency, and five-year survival rate of CTCLs.

Sézary syndrome is an aggressive, leukemic type of CTCL thought to arise from the central memory T-cells (T_CM). T_CM are highly proliferative cells circulating between peripheral blood, lymph nodes, and skin.^{12, 14} Representative symptoms of Sézary syndrome are generalized erythroderma, which can affect more than 80% of the body surface, lymphadenopathy, and the presence of malignant cells in blood circulation. Sézary syndrome also shows faster progression than MF.²¹ Criteria for distinguishing Sézary syndrome from erythrodermic MF are blood involvement and count of Sézary cells, atypically enlarged lymphocytes with convoluted nuclei.¹⁹ The high prevalence of pruritus (87–100% of patients) is also a characteristic symptom.²²

The spectrum of primary cutaneous CD30-positive lymphoproliferative disorders includes primary cutaneous anaplastic large cell lymphoma (pcALCL) and lymphomatoid papulosis (LyP). CD30 is a membrane protein that belongs to the tumor necrosis factor (TNF) receptor superfamily, expressed by activated T- and B-cells.²³ In tumors, CD30 is primarily expressed in lymphoid malignancies (Hodgkin and non-Hodgkin lymphomas).²⁴ In the lymphoproliferative disorder subtype pcALCL, symptoms present as separate, grouped, or multifocal nodules and papules in the skin. Lesions usually affect the upper body and can measure several centimetres. Typically, the age of onset is 50–70 years. Of patients, 10–15% represent extracutaneous dissemination in regional lymph nodes or

visceral metastasis.^{12, 25, 26} Of note, systemic ALCL is a distinct lymphoma that may secondarily involve the skin,²⁷ whereas LyP is described as a recurrent chronic papulonodular dermatitis. The incidence of LyP is 1.2–1.9 per million individuals.²⁸ Onset of LyP may be at any age, although LyP is unusual in childhood.²⁹ Histologically, LyP represents large variability, thus resembling a variety of CTCL subtypes. The papulonodular eruption may range from a single lesion to hundreds of lesions over the body (Figure 1C). Commonly, skin lesions are self-healing, but 10–40% of patients develop a second lymphoproliferative disease such as pcALCL, MF, or Hodgkin lymphoma.^{12, 28}

In contrast to other CTCL subtypes, subcutaneous panniculitis-like T-cell lymphoma (SPTCL) affects younger individuals. The median age of onset is 36 Figure 1. Clinical presentation of CTCL skin lesions. A) Early-stage patches of MF patient. B) Fungus-like skin tumor of MF patient. C) Nodular skin lesions of LyP patient.

Adapted from the Department of Dermatology archives, University of Helsinki, Finland.

years (range 7–79 years). Symptoms are nodular skin lesions or deep plaques with a diameter of 1-20 cm.³⁰ SPTCL infiltrates preferentially into subcutaneous adipose tissue, resembling benign panniculitis, cellulitis, eczema, or skin infections.

Consequently, the diagnosis is challenging. SPTCL nodules are typically self- healing. However, in advanced disease, patients can suffer from serosal effusion, hemophagocytosis syndrome, and pancytopenia.³¹

Treatment options for CTCL depend on the disease subtype and the clinical staging. Skin-directed therapies, such as topical steroids and radiation, are used for early-stage patients.¹⁴ Also, ultraviolet phototherapy has a longstanding history in the treatment of MF (8-methoxypsoralen combined with UVA or UVB).³² Systemic therapy, such as gemcitabine and doxorubicin, is recommended for patients with advanced disease. Entering a multicenter clinical trial is also a possibility. In addition, numerous other systemic treatments are available, but the long-term remission rate is usually poor.^{14, 33} However, several new targeted drugs, such as mogamulizumab and brentuximab vedotin, have become available. Mogamulizumab is an antibody targeting the chemokine receptor,³⁴ and brentuximab vedotin selectively targets cells expressing the CD30 antigen.³⁵ The PD-1/PD-L1 (programmed death 1 and its ligand) checkpoint system normally

controls the local inflammatory responses and maintains self-tolerance, but it also prevents the immune system from destroying cancer cells. Immune checkpoint inhibitor anti-PD1 targets PD-1 expressing cells such as T-cells, natural killer cells, and B-cells.³⁶ For example, in Sézary syndrome, the PD-1 expression is upregulated in CD4-positive T-cells and skin specimens.^{37, 38} In a phase II trial of PD-1 inhibitor pembrolizumab, the total response rate was 38% of advanced MF and SS patients.³⁹

Also, microbiological agents are proposed as treatment options. Aggressive antibiotic treatment against Staphylococcus aureus in advanced stages of CTCL resulted in significant improvement of clinical symptoms. Of eight patients, six experienced decreases in malignant skin lesions.⁴⁰ However, despite the moderate responses to available treatments of CTCL, the treatments mainly minimize the symptomatic morbidity. Allogenic hematopoietic stem cell transplantation for a fraction of patients with advanced disease is currently the only curative treatment.^41,42

2�1�2 Pathogenesis

Most lymphocytes are found in the lymphoid tissues; only a small number of these are recycled in the circulation.⁴³ When circulating in the peripheral blood, some of them migrate into the skin and are involved in inflammation and injury responses or homeostatic skin surveillance. Typically, healthy skin-homing

Occasionally, skin-homing T-cells proliferate into malignant T-cells and undergo clonal expansion.¹⁵ In CTCLs, each subtype comprises distinct clinical symptoms, but all are characterized by expanding malignant T-cells.¹⁹ However, malignant T-cells differ between each subtype and even between patients with the same CTCL subtype.¹⁴ Furthermore, they may change during disease progression.⁴⁴ A recent study challenged the idea that MF is derived only from skin-resident

memory T-cells. In contrast, it showed that malignant T-cells represented multiple malignant T-cell clones in the skin and blood. The study hypothesized that skin lesions are formed by seeding malignant T-cell clones from circulation. T-cells undergo expansion and additional mutational evolution in the skin, producing genetically different subclones. Some of these subclones may re-enter the circulation and seed other skin lesions.^{45, 46}

The inflammatory environment is a hallmark of CTCL progression from indolent to progressive disease. In the early stage, reactive T-helper1 (Th-1) and cytotoxic CD8 T-cells accumulate in the skin lesions. These cells release interferon-γ (INF- γ) and cytotoxic molecules, having anti-tumor properties.

In advanced disease, Th-1 shifts towards a Th-2 cytokine profile. Consequently, levels of transcription factor GATA-3 and cytokines IL-4, IL-5, and IL-13 are increased.¹³ The Th-2 dominant microenvironment promotes tumor growth and progressive immunosuppression.⁴⁷ Interestingly, the cytokine profile changes may also modulate chemokine expression of benign T-cells, fibroblasts, and keratinocytes in the CTCL microenvironment, further shifting towards a Th-2 inflammatory environment.¹³

Malignant T-cells also change the epidermal architecture. T-cells secrete factors that disorganize the keratinocytes’ stratification, leading to increased epithelial permeability.⁴⁸ Weakened skin barriers expose susceptibility to bacterial infections such as Staphylococcus aureus. Particularly, staphylococcal enterotoxin A activates STAT3 (signal transducer and activator of transcription 3) and IL-17 expression.

STAT3 regulates the oncogenic pathway by a different mechanism and promotes tumor growth. For example, STAT3 further upregulates proto-oncogenes, such as Bcl-2 and survivin, or downregulates tumor-suppressor, such as miR-22.⁴⁹

Besides establishing the role of inflammation in CTCL progression, the genomics of CTCL has been studied intensively. Despite findings of genetic alterations, CTCL is not caused by any specific mutations, copy number variants, or familial accumulation of the gene alteration. However, somatic alterations may occur in the genes involved in cellular processes such as epigenetic regulation, cell cycle control, or signaling pathways^{13, 50}. A recent study challenges the view that CTCL is a monoclonal disease originating from a single T-cell clone.⁴⁶ The study documents that T-cell clones of MF are intratumorally heterogeneous with a branched phylogenetic relationship pattern. Also, stage progression was shown to correlate with increased clonotypic heterogeneity. The driver mutations of CTCLs are unknown. However, the study of single-cell RNA sequencing analysis (scRNA-seq) has highlighted the clusters of unique transcriptomes when

compared CTCL patients with healthy controls. Remarkably, the expression of TOX represents a marker of malignant lymphocytes, and the cluster of 17 genes expression identifies highly proliferating lymphocytes of CTCL patients, including PCNA, ATP5C1, and NUSAP1.⁵¹ In addition, the scRNA-seq analysis of Sésary syndrome patients of early versus late disease stage showed a transition from FOXP3+T cells to GATA3+or IKZF2+tumor cells. The FOXP3+was the most influential factor to predict the disease stage, along with 19 other genes. This transcriptomic diversity was able to predict the disease stage at 80% accuracy.⁵²

Environmental factors have been suggested to contribute to CTCL pathogenesis.

Some evidence shows the clustering of CTCL patients, for instance, in several communities of the state of Texas. CTCL incidence was 3- to 20-fold higher than the US national average. The study proposed that environmental factors, such as pollution, may trigger malignancy. Contrary to other skin cancers, sun exposure is reported to protect against the development of CTCL, suggested to contribute to the low vitamin D levels of MF and SS patients.⁵³ However, these studies are limited, and no clear consensus of the role of environmental triggers yet exists.

2 . 2 PROSTATE CANCER AND COLON CANCER

2�2�1 Prostate cancer

Th e incidence of prostate cancer is the second highest worldwide and the fifth leading cause of cancer-related deaths in men.⁵⁴ The median age of onset is 68 years.⁵⁵ Of all prostate cancers, 75% can be divided into seven subtypes characterized by different molecular mechanisms, including cancer-driving genetic mutations or gene fusions.⁵⁶

Multiple risk factors contribute to the incidence of prostate cancer, including age, family history, ethnicity, and genetic susceptibility.^{56, 57} Malignant transformation of prostate cancer is a multistep process. In the first step, the prostatic intraepithelial neoplastic precursor of prostate carcinoma forms a localized tumor. Second, the tumor evolves into an advanced adenocarcinoma and finally the metastasis spreads.⁵⁸ Commonly, prostate cancers remain latent for a long period, and it is potentially curable at an early stage with a high long-term survival rate.⁵⁹ However, approximately 25% of men sustain either regional or distant metastasis.

Lymph nodes are often the first metastatic site, followed by the liver, lungs, and bones.^{58, 60} The majority of prostate carcinomas are clinically localized at the time of diagnosis with a preferable outcome. Nevertheless, approximately 5% of

the five-year survival for men with metastatic prostate cancer is only 28%.⁶⁰ The lethality and lack of an effective treatment for the advanced stage reflect the

extreme heterogeneity of prostate cancer at genetic and biological levels.⁵⁸ The function of prostate cells is mainly production, storage, and secretion of prostatic fluids, including citrate, polyamines (spermine, myoinositol), and prostate-specific antigen (PSA).^{61, 62} Normal prostate metabolism is highly specialized to produce elevated levels of citrate and an accumulation of zinc (important regulatory cofactor for enzymes and transcription factors). Likewise, the metabolic reprogramming of prostate cancer is unique compared with other tumors, favoring oxidative phosphorylation and lipogenesis but limited glycolysis.

Androgen receptor-driving metabolism has an essential role in prostate cancer.⁶³ Compared with healthy prostate tissue, prostate cancer is characterized by low levels of citrate and polyamines but high levels of lactate, choline, and creatine.

Consequently, the altered metabolism of prostate cancer is a promising target for biomarker discovery.^{61, 62} A wide variety of body fluids can be employed, including urine, serum, plasma, and prostate fluids. However, cultured cells comprise some advantages in biomarker discovery, since invariability, such as in the age of patients, smoking habits, and diet, can be eliminated.⁵⁹

2�2�2 Colon cancer

Colorectal cancer (CRC) is the third most common cancer and the third leading cause of cancer-related deaths worldwide.⁵⁴ The median age of onset is 67 years for men and 71 for women.⁶⁴ In many high-income countries, the mortality rate of CRC has gradually declined. However, mortality has increased in less developed countries, a consequence of limited resources for early detection with a screening program.⁶⁵ The 5-year survival rate depends on the cancer stage and metastatic spread, accounting for 90% of early-stage, 70% of intermediate, and 10% of advanced disease with metastasis.⁶⁶ Epidemiologically, colon cancer and rectal cancer are synonymously termed CRCs. However, according to the American Cancer Society in 2015, 72% of CRCs were diagnosed in the colon and 28%

in the rectum.⁶⁷ CRCs are heterogeneous diseases that exhibit various genetic and epigenetic alterations.⁶⁸ Despite hereditary predisposition, CRC is mostly a sporadic disease with a slow progression over years.⁶⁹ The term CRC is used for rectal cancer, right-sided and left-sided colon cancers.⁷⁰

Colon cancer originates from the epithelial tissue and may develop in the left or right side of the colon. Right- and left-sided cancers comprise different histology and molecular characteristics. The differences originate from anatomical site, developmental origin, or district carcinogenic loading during progression, including the bacterial population and exposure to different nutrients or bile acids.⁷¹ The prognosis of right-sided tumors is worse than left-sided tumors with

distinct genetic alterations.⁶⁶ Microsatellite instability (MSI) is a hypermutable phenotype that is caused by a loss of DNA mismatch repair activity. It is associated with right-sided colon cancer and is detected in approximately 15% of CRCs, including hereditary Lynch syndrome.⁷² The MSI-positive CRCs have unique features such as hereditary predisposition and location in a highly activated lymphocytic microenvironment composed of infiltrated cytotoxic T-lymphocytes and activated Th-1 cells. MSI-positive CRCs can escape the immune system by multiple mechanisms, including upregulating of immunosuppressive molecules, such as PD-1, PD-L1, CTLA-4 (cytotoxic T-lymphocyte-associated protein 4), LAG-3 (lymphocytic-activation gene 3), and IDO1,⁷³ suggesting that MSI-positive CRC patients benefit from targeted immune checkpoint therapy. Recently, the phase III clinical trial of pembrolizumab (PD-1 blockade) led to a significantly longer progression-free survival and fewer treatment-related adverse events than traditional chemotherapy.⁷⁴ However, the therapy response differs from left-sided CRCs, benefiting more from adjuvant chemotherapies and targeted therapies such as anti-epidermal growth factor receptor (EGFR) therapy. Chromosomal instability pathway-related alterations, such as activation of KRAS (Kirsten rat sarcoma viral oncogene homolog) and inactivation of p53 tumor suppressor gene, are features of left-sided CRCs.⁶⁶ Left-sided CRCs predominantly follow the traditional chromosomal instability pathway, proposed first in 1990 by Fearon and Vogelstein, who stated that tumorigenesis of CRC is a multistep process with a continuum of genetic alterations.^{66, 75}

In conclusion, CRC is a complex group of diseases with many risk factors such as diet, age, body weight, social behaviors, history of inflammatory bowel disease, and family history of CRC.⁶⁹

2 . 3 TRYPTOPHAN CATABOLISM

Tryptophan is an essential amino acid for humans, supplied by dietary uptake.

Tryptophan and its downstream metabolites from the catabolic pathway serve as crucial building blocks for protein synthesis and precursors for physiologically active compounds.⁷⁶ However, the tryptophan catabolism imbalance has revealed an essential regulator for inflammation and cancer progression. IDO1, IDO2 or TDO are the main enzymes that catalyze the conversion of tryptophan into kynurenine⁷⁷ (Figure 2).

Figure 2. Tryptophan catabolism pathway. IDO 1, IDO2, and TDO catalyze tryptophan degradation into kynurenine. Kynurenine is catabolized along with three different branches.

KMO = kynurenine-3-monooxygenase, KYNU = kynureninase, KAT = kynurenine aminotransferase, NAD+ = nicotinamide adenine dinucleotide. Modified from Fuertig et al.⁷⁸ and Maliniemi et al.⁷⁹. Created with BioRender.com.

Catabolizing routes of kynurenine depend on the cell type and the enzymes available. In the first branch, kynurenine is converted to 3-OH-kynurenine, which is further degraded to 3-OH-anthranilic acid. When upregulated, these two inter-metabolites can release free radicals and promote tissue damage. In the second branch, kynurenine is transformed to another redox-active intermediate, anthranilic acid, which may hydroxylate to 3-OH-anthranilic acid, followed by metabolizing into quinolinic acid. Finally, the end-product of these two branches is nicotinamide adenine dinucleotide (NAD+). NAD+is a criticalenzyme co-factor involved in various biological processes such as oxidative phosphorylation and electron transfer during glycolysis. In the third branch, kynurenine is produced to kynurenic acid.^{78, 80} Elevated kynurenic acid concentrations are detected from the serum of colon adenocarcinoma and invasive lung cancer patients. However, the contrast results show that the kynurenic acid serum levels of cervical cancer and glioblastoma patients were decreased.⁸¹

The immunosuppressive role of tryptophan catabolism was first reported in the placenta, where IDO1 was shown to prevent rejection against the fetus.⁸²

Similarly, dysregulated tryptophan catabolism promotes tumor progression by suppressing immune responses. All tryptophan-degrading enzymes, IDO1/2, and TDO are expressed in cancers. However, IDO1 is a principal enzyme in this pathway, expressed in approximately 58% of human tumors such as melanoma, colon cancer, and hematological malignancies.⁷⁷ IDO1 is also upregulated in the cells of TME such as regulatory T-cells (Tregs), dendritic cells, antigen-presenting cells, macrophages, and endothelial cells.⁸³ Especially, IDO1 expression is associates with tumor-infiltrating FOXP3+ Tregs to stimulate their generation. In addition, IDO1 expression is detected in circulating peripheral blood mononuclear cells (PBMC) from cancer patients.⁸⁴ Upregulation of IDO1 suppresses T-cell activity and inhibits immune cell function by multiple mechanisms.⁸⁰ Local tryptophan depletion and kynurenine accumulation are suggested to be a crucial immunosuppressive mechanism. The stress-response kinase serves as a molecular sensor of tryptophan deprivation. Depletion of tryptophan by IDO1 activates the stress-response kinase pathway, thus inhibiting T-cell proliferation, and induces immune tolerance.^{80, 85} However, the relevance in in vivo studies is less clear.⁸⁶ Aryl hydrocarbon receptor is a ligand-activated transcription factor, initially

known to control cellular responses against environmental toxins.⁸⁷ Tryptophan catabolism metabolites, such as kynurenine, are agonists for the aryl hydrocarbon receptor. IDO1-producing metabolites activate the aryl hydrocarbon receptor.

Thereafter, the aryl hydrocarbon receptor promotes T-cell differentiation to immunosuppressive Tregs, commonly upregulated in cancers.⁸⁸ Furthermore, activated dendritic cells produce IDO and inhibit T-cell proliferation.⁸⁹

IDO1 and TDO share a similar function of degradation of tryptophan. In contrast, the role of IDO2 in human tryptophan metabolism is proposed to be minimal.⁹⁰ Pro-inflammatory mediators, such as IFN-γ and lipopolysaccharide (LPS), up-regulate IDO1.^{91, 92} In comparison, TDO expression is regulated by

systemic levels of tryptophan and corticosteroids.^{93, 94} The tryptophan degradation pathway and the role of IDO1/2 and TDO in cancers constitute complex, cell- specific, and context-dependent mechanisms. Although research has proceeded intensively over the past two decades, the exact molecular mechanism remains unknown.

2 . 4 EXTRACELLULAR VESICLES

EVs have emerged as an essential mediator of intercellular communications in almost all cell types. EVs comprise a heterogeneous group of vesicles enclosed by a lipid bilayer. EVs can release their contents to the recipient cells, affecting the

between 50 and 500 nm, but they can be even 1–10 µm in diameter.⁹⁵ The EV population is commonly divided into microvesicles (50–500 nm) and exosomes (30–100 nm). These two groups differ in their biogenesis but overlap in size and composition.⁹⁶ Furthermore, many EV isolation methods are supposed to isolate exosomes or microvesicles separately, co-isolating heterogeneous populations of EVs.⁹⁷ Consequently, the exact difference between exosomes and microvesicles cannot be verified. However, vesicles formed by direct budding from the plasma membrane are referred to as microvesicles (Figure 3). In comparison, exosomes form by vesicle budding into endosomes, thereafter maturing into multivesicular bodies (MVBs). The MVBs fuse with the plasma membrane and release exosomes to the extracellular space or undergo degradation pathways in the lysosome.⁹⁶ Interestingly, some studies suggest that also exosomes may bud directly from the plasma membrane, as illustrated by the atomic force microscopy method.⁹⁸ Exosomes may even sprout from the intracellular plasma membrane-connected compartments (IPMCs).⁹⁹ Initially, the IPMCs are demonstrated to provide the compartments of HIV particles reservoir during macrophage infection;¹⁰⁰ however, their role as the EVs reservoir needs further validation.

Figure 3. Releasing of EVs from the cell. 1) Exosomes are released from the cell by the classical MBV-mediated pathway. 2) Exosomes and microvesicles bud by immediate release from the plasma membrane. 3) Exosomes may also use the IPMC-mediated pathway. MBV

= multivesicular body, IPMC = intracellular plasma membrane-connected compartments.

Created with BioRender.com.

EVs carry their cargo enclosed in vesicles or on the surface of EVs. The cargo repertoire is comprehensive and varies between cell types. EVs contain a high range of bioactive molecules such as protein, lipids, nucleic acids (DNA, mRNA, non-coding RNAs), and metabolites. Typically, all EVs share some common proteins used as EV markers. According to the International Society of Extracellular Vesicles, these markers and other guidelines are used to establish validated EV research (Minimal Information for Studies of Extracellular Vesicles, MISEV guidelines).¹⁰¹

2�4�1 Mechanism of extracellular vesicle uptake

Multiple pathways are involved in the ability of EVs to release their cargo or signal molecules to the recipient cells. First, the membrane of EVs can fuse directly to the plasma membrane of recipient cells. Second, cells can uptake EVs and their cargo with endocytosis. Third, EVs can use ligand- or receptor-mediated approaches97, 102, 103 (Figure 4). However, communication between recipient cells and EVs is a complex process, requiring multiple proteins. Furthermore, the dynamic internalization of EVs depends on the cell type and is temporally and spatially regulated.^{103, 104} The uptake of EVs represents a critical process between cell-to-cell communication. Consequently, it has garnered interest in the research field, especially in EV-mediated drug delivery research, but it is only partially understood.

Figure 4. Cells comprise multiple mechanisms of EV uptake.

1) Macropinocytosis-mediated EV uptake. Actin cytoskeleton in the cytoplasm mediates plasma membrane extension into lamellipodia. Folded plasma membrane traps EVs inside the cell.^{103, 105} 2) Direct fusion to the cell membrane. EVs fuse with the plasma membrane to release their cargo directly into extracellular space.¹⁰⁶ 3) Internalization of EVs with the clathrin-dependent pathway. Other mechanisms comprise caveolin-dependent or clathrin- and caveolin-independent pathways.¹⁰⁷ 4) EVs may also be endocytosed via distinct lipid rafts along the plasma membrane.¹⁰⁸ 5) Lysis of EVs in the extracellular space can release free ligands from the EVs, stimulating receptors on the cell surface.¹⁰² 6) EVs can also bind directly to the cell receptor, thus releasing signal molecules to the cytoplasm.¹⁰⁹ 7) Internalization of EVs by phagocytosis.

Phagocytosis is known to ingest large particles, such as pathogens, also employed for EV internalization.^{103, 110} Created with BioRender.com.

2�4�2 Extracellular vesicles: function in cancers

Communication between cancer cells and surrounding cells is historically thought to occur via direct cell-to-cell contact, facilitated by gap junctions and tunneling nanotubes, or by the secretion of soluble factors such as cytokines and growth factors.^{97, 111} Nowadays, straightforward evidence confirms that EVs mediate a crucial function in cell-to-cell communication. In cancers, EVs are known to modulate almost all steps of cancer progression⁹⁷ (Figure 5).

Figure 5. Role of EVs in cancer progression. EVs are small round-shaped vesicles enclosed by a lipid bilayer. The cargo of EVs consists of multiple proteins, DNAs, RNAs, lipids, and metabolites. Created with BioRender.com.