DISSERTATIONS | VIKTÓRIA PRANTNER | NUCLEAR RECEPTORS AS MOLECULAR MACHINES | No 469
uef.fi
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND Dissertations in Health Sciences
ISBN 978-952-61-2823-8 ISSN 1798-5706
Dissertations in Health Sciences
PUBLICATIONS OF
THE UNIVERSITY OF EASTERN FINLAND
VIKTÓRIA PRANTNER
NUCLEAR RECEPTORS AS MOLECULAR MACHINES
Mechanistic investigations on DNA-, ligand- and coactivator-nuclear receptor interactions Nuclear receptors are omnipotent proteins,
regulating healthy development of cells within the body. Irregular nuclear receptor-protein
crosstalk may result in the formation of serious metabolic diseases. In this thesis we
acquired better understanding on protein interactions applicable for the understanding
of the metabolic regulation. Some of the findings provide valuable insights that may be
used to design more potent drugs for disease treatment based on selective protein-
protein recruitment.
VIKTÓRIA PRANTNER
30805599_UEF_Vaitoskirja_NO_469_Viktoria_Prantner_Terveystiede_kansi_18_06_05.indd 1 5.6.2018 9.45.22
Nuclear receptors as molecular machines
Mechanistic investigations on DNA-‐‑, ligand-‐‑ and
coactivator-‐‑nuclear receptor interactions
VIKTÓRIA PRANTNER
Nuclear receptors as molecular machines
Mechanistic investigations on DNA-‐‑, ligand-‐‑ and coactivator-‐‑nuclear receptor interactions
To be presented by permission of the Faculty of Health Sciences, University of Eastern Finland for public examination in MD100, Kuopio, on Thursday, June 28th 2018, at 12 noon
Publications of the University of Eastern Finland Dissertations in Health Sciences
Number 469
School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland Kuopio
2018
GRANO Jyväskylä, 2018 Series Editors:
Professor Veli-Matti Kosma, M.D., Ph.D.
Institute of Clinical Medicine, Pathology Faculty of Health Sciences Associate Professor Tarja Kvist Ph.D.
Department of Nursing Science Faculty of Health Sciences Professor Olli Gröhn, Ph.D.
A.I. Virtanen Institute for Molecular Sciences Faculty of Health Sciences
Professor Kai Kaarniranta, M.D., Ph.D.
Institute of Clinical Medicine, Ophthalmology Faculty of Health Sciences
Lecturer Veli-Pekka Ranta, Ph.D. (pharmacy) School of Pharmacy
Faculty of Health Sciences Distributor:
University of Eastern Finland Kuopio Campus Library
P.O.Box 1627 FI-70211 Kuopio, Finland http://www.uef.fi/kirjasto ISBN (print): 978-952-61-2823-8
ISBN (pdf): 978-952-61-2824-5 ISSN (print): 1798-5706
ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
!III
Author’s address: School of Pharmacy
University of Eastern Finland KUOPIO
FINLAND
Supervisors: Dipl. Ing. Ferdinand Molnár, Ph.D.
School of Pharmacy
University of Eastern Finland KUOPIO
FINLAND
Tuomo Laitinen, Ph.D.
School of Pharmacy
University of Eastern Finland KUOPIO
FINLAND
Professor Perttu Permi, Ph.D.
Nanoscience Center University of Jyväskylä JYVÄSKYLÄ
FINLAND
Reviewers: Professor Enikő Kállay, Ph.D.
Department Pathophysiology and Allergy Research Medical University of Vienna
VIENNA AUSTRIA
Professor Yuuki Imai, Ph.D.
Proteo-Science Center Ehime University EHIME
JAPAN
Opponent: Professor Jukka Hakkola, Ph.D.
Director of the Biomedical Research Unit Faculty of Medicine
University of Oulu OULU
FINLAND
!V Prantner, Viktória
Nuclear receptors as molecular machines, Mechanistic investigations on DNA-, ligand- and coactivator- nuclear receptor interactions
University of Eastern Finland, Faculty of Health Sciences
Publications of the University of Eastern Finland. Dissertations in Health Sciences 469. 2018. 86 p.
ISBN (print): 978-952-61-2823-8 ISBN (pdf): 978-952-61-2824-5 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
ABSTRACT
Nuclear receptors are vibrant and vigorous transcriptional factors participating in various physiological processes such as endogenous drug metabolism, cell proliferation and differentiation and energy metabolism. During the last decade, three xeno- and endo-biotic metabolising NRs, constitutive androstane, pregnane X and vitamin D receptors, have been linked to energy metabolism which opened up new challenging possibilities for the development of new drugs for the treatment of metabolic disorders. These nuclear receptors not only share common metabolic target genes, that participate in metabolic pathways they also recruit similar transcriptional coactivator proteins, such as the peroxisome proliferator-activated receptor gamma coactivator 1-alpha, the master regulator of energy metabolism and steroid receptor coactivator 1. Structurally, these coactivators are very similar, by containing several leucine rich LXXLL motifs, that functions as nuclear receptor interacting boxes. Peroxisome proliferator-activated receptor gamma coactivator 1- alpha consist of serine-arginine rich and a special RNA processing motif in addition to the LXXLL motifs. Apart from several LXXLL motifs, steroid receptor coactivator 1 possesses Per-Arnt-Sim domain. Although it is widely believed that the second LXXLL motif plays a role in providing the physical interaction between the coactivators and most of the nuclear receptors, no information is available on peroxisome proliferator-activated receptor gamma coactivator 1-alpha and steroid receptor coactivator 1 interaction sites regarding pregnane X and vitamin D receptor.
To fill this gaps in knowledge, this thesis attempts to identify the mechanistic and structural determinants of these interactions, to investigate the main interacting motif(s) of each of them and to provide the missing information on these interaction. Various in vitro and in silico methods are available to study protein-protein interactions such as cell based assays or using recombinant protein technologies. To achieve the goal of the determination of the detailed interaction profile, a large number of new, mammalian and recombinant expression constructs had to be designed and created. In addition, the optimisation of the human intrinsically disordered domains had to be done for expressions of recombinant proteins in bacteria. In addition, these wet lab experiments were aided by in silico modelling techniques.
Primarily, it was believed that all known coactivators behave in a similar way and have a scheme they follow when binding to nuclear receptors. The findings of this thesis suggests that there are very specific protein interactions that most likely fine-tuned by evolutionary and specifies specific aspects. All three nuclear receptors have a highly specific way in interacting with either of the studied coactivators. Moreover, different types of ligand, bound to the nuclear receptor further influenced the choice of the interacting motif. The relevance of peroxisome proliferator-activated receptor gamma coactivator 1-alpha’s serine- arginine rich and RNA processing motifs were completely diminished in previous research.
However, our results show that for both the vitamin D and pregnane X receptors, the RNA processing motif is one of the main factor influencing their signalling. Serine-arginine rich motifs play partial role for all three nuclear receptors, and do definitely make a contribution. In the possession of such important information, it is suggested that these
nuclear receptor-coactivator interactions may be further distinguishable in different glucose sensing tissues to help to widen the search for novel drugs to treat metabolic disorders.
National Library of Medicine Classification: QU 55.2, QU 55.97, QU 475, WD 200
Medical Subject Headings: Receptors, Cytoplasmic and Nuclear; Nuclear Receptor Coactivators; Receptors, Calcitriol; Receptors, Steroid; Xenobiotics; Metabolism; Drug Agonism; Drug Inverse Agonism; Ligands; Gene Expression Regulation; Metabolic Diseases/drug therapy; Metabolic Syndrome; Gluconeogenesis;
Lipogenesis; Diabetes Mellitus; Biological Assay; Models, Molecular; Cells, Cultured
!VII Prantner, Viktória
Tumareseptorit molekulaarisena koneistona - Mekanistisia tutkimuksia tumareseptorien vuorovaikutuksista DNA:n, ligandien ja koaktivaattorien kanssa.
Itä-Suomen yliopisto, terveystieteiden tiedekunta
Publications of the University of Eastern Finland. Dissertations in Health Sciences 469. 2018. 86 s.
ISBN (print): 978-952-61-2823-8 ISBN (pdf): 978-952-61-2824-5 ISSN (print): 1798-5706 ISSN (pdf): 1798-5714 ISSN-L: 1798-5706
TIIVISTELMÄ
Tumareseptorit ovat tärkeitä transkriptiotekijöitä, jotka osallistuvat moniin fysiologisiin proses-seihin, kuten lääkeainemetaboliaan, solujen kasvuun ja erilaistumiseen sekä energiametabo-liaan. Viime vuosikymmenen aikana myös konstitutiivinen androstaani-, pregnaani X ja D-vitamiini-reseptori, on yhdistetty energiametabolian säätelyyn. Tämä havainto on avannut uusia lääkekehitysmahdollisuuksia energiametabolian häiriöiden hoitoon. Nämä reseptorit säätelevät osin samojen kohdegeenien ilmentymistä sekä sitovat samoja koaktivaattoreita, esimerkkeinä peroksisomiproliferaattori-aktivoituva reseptori γ:n koaktivaattori 1α, joka on erityisen merkit-tävä energiametabolian säätelyssä, sekä steroidireseptorin koaktivaattori 1. Rakenteellisesti nämä koaktivaattorit ovat hyvin samanlaisia ja sisältävät useita leusiinirikkaita LXXLL-motiiveja, jotka on määritelty tumareseptori-interaktioalueiksi. LXXLL-motiivien lisäksi tärkeitä alueita ovat muun muassa peroksisomiproliferaattori-aktivoituva reseptori γ:n koaktivaattori 1α:n sisältämä seriini-arginiinirikas sekä RNA:n prosessointiin liittyvä alue ja steroidireseptorin koaktivaattori 1:n Per-Arnt-Sim domeeni. Erityisesti toisen LXXLL-motiivin on ajateltu mahdollistavan fyysisen interaktion koaktivaattorien ja useimpien tumareseptorien välillä, mutta peroksisomiproliferaattori-aktivoituva reseptori γ:n koaktivaattori 1α:n ja steroidiresep-torin koaktivaattori 1:n ja pregnaani X ja D-vitamiinireseptorien välisistä inter-aktioista ei ole tarkkaa tutkimustietoa.
Työn tavoitteena oli selvittää interaktioiden mekanistisia ja rakenteellisia tekijöitä sekä selvittää pääasialliset interaktioita välittävät motiivit näille reseptoreille ja koaktivaattoreille. Proteiinien välisten vuorovaikutusten tutkimiseen on käytössä erilaisia in silico ja in vitro, kuten solupohjaisia ja rekombinanttiproteiinien käyttöön perustuvia, menetelmiä. Yksityiskohtaisten interaktioprofiilien määrittämiseksi työssä kehitettiin ja optimoitiin useita koaktivaattori-DNA-konstrukteja. In vitro tuloksia täydennettiin molekyylimallituksen keinoin.
Aiemmin on ajateltu, että koaktivaattorien sitoutumiselle tärkeät alueet tai sitoutumistavat tumareseptoreihin ovat samanlaisia. Saatujen tulosten perusteella interaktiot ovat kuitenkin koaktivaattori- ja tumareseptorispesifisiä ja nämä erot saattavat olla evolutionaarisia.
Kaikilla kolmella tutkitulla tumareseptorilla havaittiin olevan erittäin spesifinen interaktiotapa tutkittujen koaktivaattorien kanssa. Lisäksi ligandin sitoutuminen vaikutti siihen, mikä motiivi mahdollisti interaktion. Peroksisomiproliferaattori-aktivoituva reseptori γ:n koaktivaattori 1α:n RNA:n prosessointiin liittyvän alueen havaittiin aiemmista tutkimuksista poiketen olevan merkittävä interaktiossa pregnaani X ja D- vitamiinireseptorin kanssa. Lisäksi seriini-arginiinirikkaalla alueella on pienempi, mutta selvästi havaittava rooli interaktioissa kaikkien kolmen tutkitun reseptorin kanssa.
Interaktioiden spesifisyys voi olla merkittävää esimerkiksi erilaisissa glukoosia tunnistavissa kudoksissa ja auttaa uusien lääkeaineiden kehittelyssä energiametabolian häiriöiden hoitoon.
Luokitus: QU 55.2, QU 55.97, QU 475, WD 200
Yleinen suomalainen asiasanasto: transkriptiotekijät; reseptorit; tuma; geeniekspressio; ligandit; aineenvaihdunta;
aineenvaihduntahäiriöt; metabolinen oireyhtymä; diabetes; soluviljely
!IX
Drága nagyszüleimnek és nagybátyámnak.
Örökké a szívemben maradtok.
Acknowledgements
The present doctoral thesis study was carried out at the Faculty of Health Sciences, School of Pharmacy at Kuopio campus, University of Eastern Finland between 2012-2016.
The study was conducted as a part of the Doctoral Programme in Drug Research and funded by the Doctoral Program in Drug Research at the University of Eastern Finland, Finnish Cultural Foundation and Saastamoinen foundation.
Most importantly, I want express my gratitude to my principal supervisor Dipl. Ing.
Ferdinand Molnár, Ph.D. for the endless support and guidance during my work. His professionalism and maximalism taught me how to become a confident researcher and his valuable advices have made me grow not only as a scientist but as person. He has always encouraged me to be curious, to dare to dream big! All I know about research and science I have learned from Dr. Molnár and I am eternally grateful for the opportunity to be his student.
I also would like to thank for my two other supervisors Tuomo Laitinen, PhD and Professor Perttu Permi, PhD, for the help they have provided for this thesis.
I would like to thank to Professor Paavo Honkakoski for the support and for the opportunity to be part of research group and working with great people.
I would like to thank to the pre-examiners, Professor Enikő Kállay, PhD and Professor Yuuki Imai, PhD for investing their valuable time and evaluating my thesis.
I thank Thomas Dunlop for the proofreading and language correction.
I will be always grateful to Jenni Küblbeck who has been a great colleague, excellent teacher and a amazing friend. Jenni, you have inspired me is so many ways, showed me that research is a wonderful journey and not only work.
I would like to thank to my great friends, Anna&Atte and Emma&Olli for being always there, whenever I needed my “Hungarian” family! I could share with you anything, anytime, through hardship and laughter, you lifted me up whenever I was down, and you pulled me back when I flew too high!
I am thankful for Melinda Szinger, for her eternal support and faith in me. You said once to me that friends are like stars, you don’t always see them, but they are always there. You are the STAR in my life!
I would like to thank to aunt and uncles and their family Éva, Anna, Bálint, Sanyi, Ildi, Márk, Vera, Jani, Eta, Balázs&Klaudia and Árpi for making my time at home relaxing and happy. There is nothing better than to spend the time together, having good food and talking about good old times, share memories and just have great laughs.
Here, I would like to thank to the most important people in Hungarian:
Drága szüleim, Anya és Apa, örökké hálás leszek a ti soha el nem múló szeretetekért.
Mindig is bátorítottatok, hogy kutató legyek. Minden egyes lépésnél ezen a hosszú úton mellettem voltatok, a kezem soha el nem engedtétek! Az, hogy a szüleim vagytok, Isten egyik legnagyobb ajándeka! Feri, szivemből mondom, jobb nevelőapát az ember nem is kivánhat maganak. Mindig is ösztönöztél, hogy csinálni kell, menni előre, fejet emelve, bátran. Hálám nagyobb, mint gondolnád! Drága testvérem, Joci, te vagy az én örök biztonságom, az én mentsváram, az őszinte tükröm, és a legnagyobb boldogságom!
Mindig, mindenütt ott voltál velem, együtt mentünk át a felnőtté válás rögös útján, egymás kezét fogva, mindig biztos támaszként egymás mellett állva. Minding is a legnagyobb támogatóm voltál, köszönöm neked! Márti, a te kitartásod és pozitív életszemléleted mindig is lenyűgözött, emellett erőt és biztatást adott, hogy menni kell előre, meg kell és lehet csinálni bármit, amit megálmodunk! Hatalmas inspiráció vagy számomra. Drága Mama, te vagy az én legnagyobb példaképem! A világon nincs nálad erősebb, kitartóbb asszony, nincs ki jobban szeretné és támogatná a gyermekeit, unokáit! A szívem túlcsordul szeretettel, ha rád gondolok! Te mindig is büszke voltál rám, és te nem is tudod én milyen
!XI
büszke vagyok rád, és arra, hogy te vagy az én csodálatos nagymamám! Mamikám, végtelenül szeretlek!
My family is my greatest support! Without them, my life would be poor, and meaningless, their endless love and support has brought me to this moment, to become a doctor, a dream I had since I was thirteen.
Last, but definitely not least, I want to thank for my husband and my son, Harun for the pure and never ending love. Nejib, your faith in me have lifted me up so many times, and made me look forward. Your incredible love and trust has brought me through this journey and I am happy that we stand together in front of the door that opens to the future! Harun, Áron, my biggest joy, the love of my life, the smile on your face makes everything else so meaningless, because that little, bright and happy face of yours is mommy’s biggest refuge!
I will love you both forever and beyond!
Kuopio, March 2018 Viktória Prantner
Table of Contents
ABBREVIATIONS...XIV
1 INTRODUCTION ...1
2 REVIEW OF LITERATURE...2
2.1 NUCLEAR RECEPTORS...2
2.1.1 Transcriptional regulation by NRs...2
2.1.2 NR superfamily and its classification...3
2.1.3 Modular and structural organisation of NRs...5
2.1.3.1 The activation function 1 (A/B domain)...5
2.1.3.2 The DNA-binding domain (C domain)...6
2.1.3.3 The Hinge region (D domain)...6
2.1.3.4 The LBD (E domain)...7
2.1.3.5 The dimerisation interface...9
2.1.4 The NR signalling...9
2.2 THE CONSTITUTIVE ANDROSTANE RECEPTOR...10
2.2.1 Molecular structure of CAR LBD...10
2.2.2 CAR’s role in energy metabolism...12
2.3 PREGNANE X RECEPTOR...13
2.3.1 The molecular structure of PXR’s LBD and DBD...15
2.3.2. PXR’s role in energy metabolism...15
2.4 VITAMIN D RECEPTOR...16
2.4.1 Molecular structure of VDR’s LBD and DBD...17
2.4.2 VDR’s role in energy metabolism...17
2.5 TRANSCRIPTIONAL COAS...18
2.1.1 PGC-1α, the master regulator of energy metabolism...18
2.1.1.1 Role of PGC-1α in energy metabolism...20
2.1.1.2 Enigmatic role of PGC-1α in T2D...21
2.1.2 The SRC family of CoAs...22
2.1.2.1 SRC-1 in energy metabolism...24
3 AIMS...27
4 MATERIALS AND METHODS...28
4.1 REAGENTS, BUFFERS...28
4.2 DNA CONSTRUCTS...30
4.2.1 Bacterial and mammalian expression constructs of NRs...30
4.2.1.1 SRC-1 synthetic deletion mutant constructs for bacterial and mammalian expression...30
4.2.2 Protein expression constructs containing CoA motif and domains...31
4.3 RECOMBINANT PROTEIN EXPRESSION VECTOR PROPERTIES...31
4.4 MOLECULAR CLONING, RECOMBINANT PROTEIN EXPRESSION AND PURIFICATION...32
4.4.1 Mini scale protein expression of hCAR-LBD, hPGC-1α LXXLL/RS and/or hSRC-1 LXXLL motif containing constructs...33
4.4.2 Western blotting...33
4.4.3 Co-purification of hCAR-LBD and hPGC-1α︎ LXXLL/RS domains ...34
4.5 MOLECULAR CLONING OF SYNTHETIC PGC-1Α LXXLL/RS DOMAINS INTO PET28C BACTERIAL EXPRESSION VECTOR...35
4.6 CLONING OF CAR-LBD INTO BACTERIAL EXPRESSION VECTOR...35
4.7 DELETION MUTATIONS OF HPGC-1Α MAMMALIAN EXPRESSION CONSTRUCTS...35
4.8 SRC-1 CONSTRUCTS...36
4.9 TRANSCRIPTIONAL ACTIVATION ASSAY (TAA)...36
4.9.1 Optimisation of TAA conditions and choice of cell line...36
4.9.2 TAA conditions for testing the response of NRs on different REs...37
4.9.3 TAA to study the interaction sites between NRs and CoAs...38
4.9.4 TAA to determine the different ligand-binding profiles of VDR in the presence of different CoAs...38
4.10 BACTERIAL PROTEIN EXPRESSION AND PURIFICATION OF PET28C/PET15B EXPRESSION VECTOR CONSTRUCTS CONTAINING CAR-LBD OR COA LXXLL/RS MOTIFS...38
4.10.1. Mini-scale protein expression of CoA LXXLL/RS motif containing constructs...38
!XIII
4.10.2 Mini-scale protein expression and purification of pET15b-hCAR LBD...39
4.10.4 Large-scale expression and purification of the pET15b-hCAR LBD...40
4.10.5 Expression and purification of hSRC-1 LXXLL motif containing mutants...40
4.11 PULL-DOWN EXPERIMENTS FOR STUDYING HCAR AND HSRC-1 INTERACTIONS...41
4.12 MARCONI ASSAY...42
4.13 IN SILLICO METHODS - MOLECULAR DYNAMIC SIMULATIONS OF HCAR-LBD WITH SRC-1 LXXLL MOTIF CONTAINING PEPTIDES...43
4.14 ANALYSIS OF THE BASAL EXPRESSION LEVELS OF NRS IN SW480-ADH AND C3A CELL LINES....43
5 RESULTS...45
5.1 MINI-SCALE BACTERIAL EXPRESSION...45
5.1.1 Cloning and mini-scale expression of the hCAR-LBD...45
5.1.2 Mini-scale expression of PGC1α LXXLLsyn and RSsyn motif containing domains...45
5.2 EXPRESSION AND PURIFICATION OF PCOAT7HF BACTERIAL CONSTRUCTS...46
5.2.3 Western blot of pCoAT7H-3xFLAG containing construct...49
5.3 OPTIMISATION FOR TAA...50
5.4 TRANSCRIPTIONAL ACTIVATION ON DIFFERENT NR PROMOTORS...51
5.4.1 CAR’s transcriptional activation on different REs....52
5.4.2 PXR’s transcriptional activation on different REs....52
5.4.3 VDR’s transcriptional activation on different RE containing promotors....53
5.5 IDENTIFICATION OF POSSIBLE PGC-1Α INTERACTION SITES WITH TAA...54
5.5.1 The interaction of CAR with PGC-1α...54
5.5.2 The interaction of PXR with PGC-1α...55
5.5.3 The interaction of VDR with PGC-1α...55
5.6 IDENTIFICATION OF POSSIBLE SRC-1 INTERACTION SITES WITH TAA...56
5.6.1 Activation of CAR with different SRC-1 LXXLL motifs in the absence or presence of ligands...56
5.6.1 Activation of PXR with different SRC-1 LXXLL motifs in the absence or presence of ligands...57
5.6.1 Activation of VDR with different SRC-1 LXXLL motifs in the absence or presence of ligands...57
5.6 LIGAND ACTIVATION PROFILE OF CAR IN THE MARCONI COA INTERACTION ASSAY...59
5.7 PULL-DOWN EXPERIMENTS FOR STUDYING CAR-SRC-1 INTERACTIONS...59
5.8 LIGAND-DEPENDENT SPECIFICITY OF THE VDR-COA RECRUITMENT...60
5.9 MOLECULAR MODELLING OF SRC-1 INTERACTION WITH CAR ...61
5.10 BASAL EXPRESSION OF NRS IN C3A AND SW480-ADH CELL LINES...64
5.11 MORPHOLOGICAL CHANGES OF C3A AND SW480-ADH CELLS UPON TREATMENT WITH VDR AGONISTS ...64
6 DISCUSSION...67
6.1 EXPRESSION, PURIFICATION AND VERIFICATION OF RECOMBINANT PROTEINS...67
6.1.1 Protein expression and purification of pCoAT7HF expression constructs...67
6.1.2 Western blot analysis...68
6.2 MOLECULAR BASIS OF LIGAND-DEPENDENT OR -INDEPENDENT RECRUITMENT OF DIFFERENT COA LXXLL MOTIFS BY CAR...69
6.2.1 The ligand selective recruiting pattern of PGC-1α motifs to CAR...69
6.2.2 Investigation of SRC-1 motif recruitment to CAR in the presence or absence of specific ligands....70
6.2.3 In vitro MARCoNi coregulator assay using SRC-1 LXXLL motifs...71
6.2.4 Molecular modelling of SRC-1 interaction with CAR...71
6.2.5 Investigation of SRC1 - CAR protein interaction using pull-down experiments...72
6.3 MOLECULAR BASIS OF LIGAND-INDEPENDENT AND -DEPENDENT RECRUITMENT OF DIFFERENT COA LXXLL MOTIFS TO PXR...73
6.3.1 PGC-1α motif recruiting profile of PXR in the presence of its ligands...73
6.3.2 The SRC-1 motif recruitment by PXR...73
6.4 MOLECULAR BASIS OF LIGAND-INDEPENDENT AND -DEPENDENT RECRUITMENT OF DIFFERENT COA LXXLL MOTIFS BY VDR...74
6.4.1 Unique ligand-independent association of PGC-1α to VDR...74
6.4.2 The SRC-1 motif recruitment by VDR...75
6.4.3 Ligand-induced selective CoA recruitment by VDR...75
6.5 TRANSCRIPTIONAL ACTIVATION OF NRS ON DIFFERENT RES...76
6.6 GENE EXPRESSION STUDY...77
7 CONCLUSION...79
8 REFERENCES...80
Abbreviations
ΔID deleted insertion domain 3KLCA 3-keto lithocholic acid
3xFLAG three copies of a synthetic FLAG peptide
6xHis six histidine tag
A adenine
aa amino acid
AF activation function
AMPK AMP-activated protein kinase APS ammonium persulfate AR androgen receptor ATP adenosine triphosphate BAT brown adipose tissue
C cytosine
C3A human hepatocarcinoma cell line CAI codon adaptation index
cAMP cyclic adenosine monophosphate CAR constitutive androstane receptor CC carbenicillin
cDNA complementary DNA
CEBP CCAAT/enhancer-binding protein CL crude lysate
CLT clotrimazole
Co2+ bivalent cobalt metal ions CoA coactivator
COOH C-terminal end CoR corepressor
COUP-TF chicken ovalbumin upstream promoter-transcription factor CPEC circular polymerase extension
cloning
CREB cyclic adenosine monophosphate response element-binding protein DAX-1 Dosage-sensitive sex reversal,
adrenal hypoplasia critical region, on chromosome X, gene 1 DBD DNA-binding domain DNA deoxyribonucleic acid
DRIP vitamin D receptor – interacting protein
DRs direct repeat
EAR2 V-erbA-related protein 2 ER estrogen receptor ERR estrogen-related receptor ERs everted repeat
EtOH ethanol
f.c. final concentration FL full length
FOXO1 forkhead box protein O1
FXR farnesoid X receptor
G guanine
G6Pase glucose 6-phosphatase GCNF germ cell nuclear factor GFP green fluorescent protein GR glucocorticoid receptor HAT histone acetyltransferase Hat histidine affinity tag HC% helical content
HEK293T human epithelial embryonic kidney cell line
HNF4 hepatocyte nuclear factor 4 HSP heat-shock protein
ID insertion domain
IMAC immobilised metal ion affinity chromatography
Insig1 insulin-induced gene-1 Insig2 insulin-induced gene-2 INSR insulin receptor gene IPTG isopropyl β-D-1-
thiogalactopyranoside IRs inverted repeat
IRS insulin response sequence
KAN kanamycin
LB Lennox L Broth Base LB agar Lennox L Broth agar LBD ligand-binding domain LBP ligand-binding pocket LCA lithocholic acid
LRH-1 liver receptor homolog 1 LXR liver X receptor
MAPK mitogen-activated protein kinase MCF7 mammalian breast cancer cell line MD molecular dynamic
MEF-2 myocyte enhancer factor-2 mQH2O molecular biology grade water
(milliQ)
MR mineralocorticoid receptor
n any nucleotide
N-CoR nuclear receptor corepressor NBFI-B neuron-derived orphan receptor 1 NGFI-B nerve growth factor IB
NH2 N-terminal end
NLS nuclear localisation sequence NR nuclear receptor
NURR1 nuclear receptor related 1 protein O/N overnight
OD600 optical density at 600 nm
PAGE polyacrylamide gel electrophoresis
PAS Per/ANT/Sim
!XV
PB phenobarbital
PBREM phenobarbital (PB) responsive element module
PBS phosphate buffer saline PCR polymerase chain reaction PDPK proline-directed protein kinase PEI Polyethylenimine
PGC-1α peroxisome proliferator-activated receptor gamma coactivator 1- alpha
PKA cyclic adenosine monophosphate- dependent protein kinase
PNR photoreceptor cell-specific nuclear receptor
PPAR peroxisome proliferator-activated receptor
PR progesteron receptor
PRC peroxisome proliferator-activated receptor gamma coactivator 1- related coativator
ps picosecond
PTM post-translational modification PXR pregnane X receptor
qRT-PCR quantitative real time PCR RA retinoic acid
RAR retinoic acid receptor RE response element Rev-erb reverse c-erbA RIF rifampicin RLU relative light unit
RMSD root mean square deviation RNA ribonucleic acid
RNA pol ribonucleic acid RNA polymerase II
ROR retinoic acid receptor-related orphan receptor
RRM RNA processing motif RXR retinoic X receptor s β-sheet
S soluble fraction in the supernatant
SB super broth
SDS sodium dodecyl sulfate SF1 steroidogenic factor 1 SHP small heterodimer partner
SMRT silencing mediator of retinoid and thyroid hormone receptor
SOB super optimal broth
SOC super optimal broth with catabolite repression
Sox9 sex determining region Y-box 9 SRC-1 steroid receptor coactivator 1 SREBP1 sterol regulatory element-binding
transcription factor 1 SUMO small ubiquitin-like modifier
SW480-ADH human colorectal adenocarcinoma cell line
T thymine
T1D type 1 diabetes T2D type 2 diabetes T3 triiodothyronine
TAA transcriptional activation assays TEV tobacco etch virus
TF transcription factor
TLX homologue of the Drosophila tailless gene
TR thyroid hormone receptor TR2/4 Testicular receptor 2/4 TRAP thyroid hormone receptor–
associated protein TSS transcription start site
UB unbound
UCP uncoupling protein VDR vitamin D receptor
VDRE vitamin D receptor response element
WAT white adipose tissue
X any amino acid
XREM xenobiotic response enhancer module
1 Introduction
Basic molecular biological processes in the human body had been investigated at the molecular level since the 1930s, yet the pioneers in the fields of biology, physics or biochemistry started to emerge in the early 19th century onwards. Even though scientific research in biology has a relatively long history, the basic mechanisms of the processes in the human body at the molecular level are still not been entirely understood. Proteins represent a largest group of macromolecules that have been widely studied. Their dysfunction, for example, based upon structural disturbances, leads to various afflictions such as neurodegenerative disorders, cardiovascular diseases, diabetes and cancer.
The steroid hormone nuclear receptor family represents the largest group of ligand activated eukaryotic transcription factors that play role in one of the most important biological process called transcription, the process transcribing genetic information from DNA substrate info information carrying RNA molecules.
Almost all nuclear receptors recognise specific DNA-sequences which are termed as response elements within the promoters of genes they regulate. Hence, after recognising and binding to response elements, nuclear receptors have a direct control over the transcriptional regulation. Based on nuclear receptor’s role, their structure is distinguishable from other proteins such as they have a DNA-binding domain, which task is to recognise response element sequences and a ligand-binding domain, which enables specific ligand-binding and the recruitment of transcriptional helper proteins, the coregulators. Thus, the finally assembled large multi-protein transcriptional complex specifically controls the fate of different target gene expression. Coordinated co-expression of various target genes in a defined biochemical pathway, but the same NR, leads to expression of proteins (which often are the downstream 3D printout of initial transcription) in that pathway. Therefore, individual NRs have a powerful role in dictating activity of biochemical processes.
Most of the nuclear receptors’ way of action is manifested through the binding of small lipophilic compounds within the ligand-binding domain to the ligand-binding pocket.
Modulation of a nuclear receptor can happen through an agonist or activating ligand, or though an antagonist/inverse agonist or repressing ligand that decreases their activity thereafter repressing the target gene expression. Many of the ligands show high selectivity to their target receptors and have a distinct mode of binding. Upon ligand-binding NR recruits cofactor protein (coactivator or corepressor) to further promote the recruitment of either activation or repression complex that determines the faith of the transcriptional regulation. A lot of the endogenous ligands are products of metabolic pathways. Therefore, it is no surprise that these NRs are potent regulators of those pathways.
This thesis aims to extend the research on understanding the molecular mechanism how nuclear receptors work as molecular machines, how their functional properties such as ligand-, protein- and DNA-interactions are regulated and what are their consequences on gene regulatory level.
2 Review of Literature
2.1 Nuclear receptors
2.1.1 Transcriptional regulation by NRs
To maintain the prefect orchestration of gene expression transcription of DNA encoded genetic information RNA it has to be tightly regulated, complex process. Any mistake during during its execution may bring serious consequences to the cellular function that can ultimately be reflected on the whole body. Transcriptional regulation not only controls the number of genes that are expressed but but also supervises the DNA reading RNA polymerase itself when a gene is transcribed. Even so, this short description does not fully describe how complex this process really is.
Since NRs belong to the large group of transcription factors (TF), they have an extremely important role in the regulation of gene expression. Ideally, each NR has its own target gene set though in reality there is a huge crosstalk that creates complex regulatory networks with overlaps between different NRs such as two or more different NRs contribute to the regulation of the gene at the same time. When mentioning the transcriptional regulatory networks it is apparent that NR family interacts with other TFs classes as well.
NRs can exercise their role through mechanisms other than classical type described above.
One of the non-canonical instruments that is used by NRs to regulated activity of NRs is post-translational modifications (PTMs). PTMs are needed to aid the fine-tuning of regulation complexity in the flow of the information from the genome to proteome, which is the description of whole set of proteins or peptides that are produced by the genome, cell or the whole organism at a given time (Jensen 2004, Gaston et al. 2003, Chavez et al. 2013, Beltrao et al. 2013, Wani et al. 2015, Smutny et al. 2013).
Compared to the non-canonical mechanism of controlling the activity of NRs the typical way is the presence of small lipophilic compounds termed ligands that are recognised by NRs, bind to them and modulate their activity in a positive (activation) or negative (repression) fashion. These ligands can range from macronutrients such as fatty acids or cholesterol and their metabolites to micronutrients such as vitamin A, D, K; or from xenobiotics (food additives, environmental pollutants, medications) to steroid and thyroid hormones (Katzenellenbogen and Katzenellenbogen 1996, Yang et al. 2014, Fernandez 2017). It is striking that many of these ligands are endogenous metabolic products, therefore it should not be surprising that NRs are deeply involved in maintaining metabolic homeostasis. Furthermore, the fact that the activity of NR can be modulated via bound ligands successfully predetermined them for design of drugs to treat various diseases and are considered as one of the most studied drug targets and to this day they remain the mostly highly targeted proteins for successful, approved medications for anti-inflammatory effects (GR), contraception (ER), metabolic disorders such as obesity & diabetes (PPAR𝛾) and cancer (AR and ER for prostate and breast tumours, respectively).
Another layer of functional complexity by which NRs exert the control over gene expression is that they bind to DNA sequence specific motifs within the promoter region of their target genes. The overall gene-regulatory outcome depends on whether an agonist or antagonist/inverse agonist is bound to NR and which coregulator protein complex is recruited. Subsequently there will be either down- or up-regulation of the target gene.
It was widely accepted that the presence of an agonistic ligand recruits a coactivator (CoA) protein, where as an antagonist/inverse agonist will attract a corepressor (CoR). However, new data has surfaced which indicates that agonist-binding may recruit also a CoR instead of a CoA protein (Pascual et al. 2005, Gurevich and Aneskievich 2009, Murai-Takeda et al.
3
2010). For example, for the androgen receptor (AR) is has been found that silencing mediator of retinoid and thyroid hormone receptor (SMRT) and NR CoR (N-CoR) play a role in the transcriptional regulation not only when antagonist is present but also when agonist is interacting with it (Yoon and Wong 2006). Moreover, a recent resolution of the retinoic acid receptor (RAR)-related orphan receptor gamma (ROR𝛾) crystal structure revealed that in the inverse agonist MRL-871 bound ROR𝛾 at helix H12, which usually plays role in the ligand-dependent transactivation of the receptor, through interacting with a CoA protein’s LXXLL motif (L=leucine X=any amino acid) may be modified or even extensively blocked disabling the receptor to recruit any coregulator (Scheepstra et al. 2015).
2.1.2 NR superfamily and its classification
NRs have a crucial role in embryonic development, cell proliferation and differentiation, synthesis and catabolism of hormones, and metabolism of cholesterol, bile acids, glucose and xenobiotics (Huang et al. 2010). Their importance is strengthened by the fact that their misregulation contributes to different pathophysiological conditions including various cancers, type 2 diabetes (T2D), cardiovascular diseases, dyslipidemias, and hormone resistance syndromes (Evans 1988, Aranda and Pascual 2001). All the 48 human members of NR superfamily have been identified upon completion of the sequencing of the human genome. Most of them have at least one identified natural ligand or group of compounds that modulate their function. However there are few for which, to date, no known ligand have been found thus have been called “orphans” (Robinson-Rechavi et al. 2001). NRs appeared evolutionary in metazoans before the separation of vertebrates from the invertebrates (Escriva et al. 1997). In different species the number of the NR members largely varies such as mouse has only one extra NR, compared to the human repertoire, in the family farnesoid X receptor beta (FXRβ) (Robinson-Rechavi et al. 2001), where as, in the case of the common fruit fly, Drosophila melanogaster, the same family consists of only 21 members (Adams et al. 2000). Remarkably, upon resolution of the whole genome sequence, 270 NR family member genes were identified in the roundworm Caenorhabditis elegans (Sluder et al. 1999).
For the classification of NR family members there are several different ways for instance a classification based on phylogenetics, initial subcellular localisation or nature of the ligands (Table 1)(Germain et al. 2006). Furthermore the classification can be established whether it acts as monomer, homodimer or heterodimer.
NR family members can be classified by several different criteria. For instance, NRs can be arranged based on molecular phylogenetics (DNA or protein), initial subcellular localisation or the nature of (or absence) the ligands (Table 1)(Germain et al. 2006).
Furthermore, NRs can also be catalogued by whether they can function as monomers, homodimers, heterodimers or even combinations of these.
Table 1. Human NR classification - nomenclature {Germain et al., 2006, #93397}
# Abbreviation NRNC* Symbol Name Prototypical ligand
1 TRα NR1A1 Thyroid hormone receptor alpha Thyroid hormones
2 TRβ NR1A2 Thyroid hormone receptor beta Thyroid hormones
3 RARα NR1B1 Retinoic acid receptor alpha Retinoic acids
4 RARβ NR1B2 Retinoic acid receptor beta Retinoic acids
5 RAR𝛾 NR1B3 Retinoic acid receptor gamma Retinoic acids
6
PPARα NR1C1 peroxisome proliferator-activated
receptor alpha Fatty acids, leukotriene B4, fibrates
7
PPARβ (PPARδ) NR1C2 peroxisome proliferator-activated
receptor beta/delta Fatty acids
8
PPAR𝛾 NR1C3 peroxisome proliferator-activated
receptor gamma Fatty acids, prostaglandin J2, thiazolidinediones
9 Rev-erb α NR1D1 reverse c-erbA alpha (TRα) Orphan
10 Rev-erb β NR1D2 reverse c-erbA beta Orphan
11
RORα NR1F1
RAR-related orphan receptor alpha Cholesterol, cholesteryl sulfate
12
RORβ NR1F2
RAR-related orphan receptor beta Retinoic acids
13
ROR𝛾 NR1F3 R A R- r e l a t e d o r p h a n r e c e p t o r
gamma Orphan
14
LXRα NR1H3 Liver X receptor alpha Oxysterols, T0901317,
GW3965
15
LXRβ NR1H2 Liver X receptor beta Oxysterols, T0901317,
GW3965
16
FXR NR1H4 Farnesoid X receptor Bile acids, fexaramine,
gugglesteron, SR12813, GW4046
17 VDR NR1I1 Vitamin D receptor Vitamin D, bile acids
18
PXR NR1I2 Pregnane X receptor X e n o b i o t i c s , 1 6α- cyanopregnenolone
19
CAR NR1I3 Constitutive androstane receptor Xenobiotics, endobiotics, environmental pollutants
20
HNF4α NR2A1 Hepatocyte nuclear factor 4 alpha Orphan
21
HNF𝛾 NR2A2 Hepatocyte nuclear factor 4 gamma Orphan
22 RXRα NR2B1 Retinoic X receptor alpha Retinoic acids
23 RXRβ NR2B2 Retinoic X receptor beta Retinoic acids
24 RXR𝛾 NR2B3 Retinoic X receptor gamma Retinoic acids
25 TR2 NR2C1 Testicular receptor 2 Orphan
26 TR4 NR2C2 Testicular receptor 4 Orphan
27
TLX NR2E2 Homologue of the Drosophila
tailless gene Orphan
28 PNR NR2E3 Photoreceptor cell-specific nuclear
receptor Orphan
29
COUP-TFI NR2F1 Chicken ovalbumin upstream
promoter-transcription factor 1 Orphan
30
COUP-TFII NR2F2 Chicken ovalbumin upstream promoter-transcription factor 2 Orphan
31 EAR2 NR2F6 V-erbA-related protein 2 Orphan
32 ERα NR3A1 Estrogen receptor alpha 17ß-estradiol, tamoxifen, raloxifene
33
ERβ NR3A2 Estrogen receptor beta 1 7 ß - e s t ra d i o l , va r i o u s synthetic compounds
34 ERRα NR3B1 Estrogen-related receptor alpha Orphan
35 ERRβ NR3B2 Estrogen-related receptor beta DES, 4-OH, tamoxifen
36 ERR𝛾 NR3B3 Estrogen-related receptor gamma DES, 4-OH, tamoxifen
37 GR NR3C1 Glucocorticoid receptor Cortisol, dexamethasone,
RU486
38 MR NR3C2 Mineralocorticoid receptor Aldosterone, spirolactone
39
PR NR3C3 Progesteron receptor P r o g e s t e r o n e ,
medroxyprogesterone acetate, RU486
40 AR NR3C4 Androgen receptor Testosterone, flutamide
41 NGFI-B (Nurr77) NR4A1 Nerve growth factor IB Orphan
42 NURR1 NR4A2 Nuclear receptor related 1 protein Orphan
43 NBFI-B (NOR1) NR4A3 Neuron-derived orphan receptor 1 Orphan
44 SF1 NR5A1 Steroidogenic factor 1 Orphan
45 LRH-1 NR5A2 Liver receptor homolog 1 Orphan
46 GCNF NR6A1 Germ cell nuclear factor Orphan
47
DAX-1 NR0B1 Dosage-sensitive sex reversal, adrenal hypoplasia critical region, on chromosome X, gene 1
Orphan
48 SHP NR0B2 Small heterodimer partner Orphan
* NRs Nomenclature Committee. 1999 A unified nomenclature system for the NR superfamily Cell. 97: 161–3.
5
The table above summaries all human NRs based on their primary DNA sequence classification. By this ordering, human NRs are divided into seven main groups based on their DNA- (DBD) and ligand-binding domains (LBD) cDNA sequence homology (Germain et al. 2006). The very first column of the table contains the abbreviation of the NR names, the second collects the unified nomenclature classification system, the third gives the common name given to that particular NR and the last column depicts prototypical natural and/or synthetic ligands. Receptors that are classified as orphans, to date have no identified high affinity ligands (Baek and Kim 2014).
2.1.3 Modular and structural organisation of NRs
In order to fulfil their important duties, NR proteins developed a very specific 3 dimensional structure, allowing it to execute their roles in gene expression. This structure can be subdivided into regions handle DNA- and ligand-binding, dimer formation with other NRs and binding to cofactor proteins. Structurally NR contain two main large domains the DBD and LBD (Figure 1). The two different domains are connected by the hinge region that provides a bridge between the two functional parts of the protein. A typical NR’s modular structure comprises of the N-terminal end (NH2), called A/B region which is followed by the DBD (C domain). The previously mentioned hinge region may be also called a linker region (domain D) and at the C-terminal end (COOH) can be found the LBD or domain E. At the very end of the C-terminus some NRs have a so called F region for which to date no clear function is known. Interestingly two NRs, DAX-1 and SHP lack a DBD but the LBD of these NRs are highly conserved (Aranda and Pascual 2001, Gronemeyer et al. 2004, Huang et al. 2010). Some NRs such as Reverb family members lack the whole F domain as well as the part of the C-terminus of the LBD, thus they do not have a helix H12 (Pawlak et al. 2012).
2.1.3.1 The activation function 1 (A/B domain)
In most of the NRs, the A/B region hosts the activation function (AF) -1 domain (Figure 1), a modulatory region that differs among NRs, both in sequence and size. Not much functional data is known about this region. This probably due to its highly flexible 3D structure as no X-ray diffraction based crystal structure has been solved. What is known, from biochemical experiments, is that it is subjected to PTMs that may result in the modulation of the whole NR activity. AR, GR, ER, RARs and PPARs are a few examples of NRs that are phosphorylated at AF-1, there are many more. RARs, for example, have several consensus phosphorylation sites for the proline-directed protein kinase (PDPK). In the case of PPARα, the same type of phosphorylation by a specific PDPK mitogen-activated protein kinases (MAPK) results in the enhancement of its transcriptional activity, where as, in case of another PPARs the same PTMs may have negative effect. Thus, PPAR𝛾 will show decreased transcriptional activity since the ligand-binding will be limited due allosteric changes in the LBD. PTMs in one NR can effect the function of its dimerising partner. For instance, phosphorylation, in case of RXRα as heterodimeric partner, may alter the biological function for the partner protein. As a summary, PTMs can have both positive and negative effect on the function of different NRs (Germain et al. 2006, Bain et al. 2007).
2.1.3.2 The DNA-binding domain (C domain)
DBD, also called domain C, is the most conserved NR domain (Figure 1). The main function of the DBD is to recognise and bind to response elements (REs), which are specific short sequences within the regulatory regions of the target genes (Figure 1). Almost all NRs contain a DBD. However, there are two exceptions, SHP (Seol et al. 1996) and DAX1 (Zazopoulos et al. 1997) do not possess DBD. Nuclear magnetic resonance and X-ray crystallographic studies have revealed that DBDs contain a well conserved 66-residue stem that is constituted of two cysteine-rich zinc finger motifs.
Moreover, this region contains two α-helices and a carboxyl acid extension. The NR's RE- binding specificity is provided by different amino acid sequence elements called A, P, D and T boxes. These boxes not only decide to which RE the DBD will bind, but also contribute to
the dimerisation surface of the DBD, depending on the RE spatial organisation. The amino acids in these boxes are able to interact with the DNA backbone and residues from both strands of the DNA core recognition sequence (Helsen et al. 2012). The first zinc finger between the last two cysteine residues is part of the P box and gives the sequence specificity of the NR-DNA-binding RE. In the cases of ER and GR, only one amino acid change in the P box may change the sensitivity towards the one or another NRs RE (Klinge 2001). The mode of RE recognition by these two NRs (Table 2) depends upon heterodimerisation with RXR#, as well. The heterodimeric RXR# complexes are interacting with a pair of hexameric core binding motifs or half-sites within the promoter region of their target genes. Such hexameric motifs can be organised as inverted repeats (IRs), containing two binding motifs that are facing towards each other (orientated symbolically as arrows, to define polarity, !
") , direct repeats (DRs) with hexameric sequences following each other in the same orientation(! !) e.g. DR3 in case of VDR (Figure 1), or everted repeats (ERs), half sites facing away from each other (" !). In the majority of cases the heterodimers prefer DRs but certain NR dimers are prone to bind to ERs. The DR group is further classified based on number of spacing nucleotides that are located between the half sites, for instance DR1-DR5 (Pawlak et al. 2012).
DNA
Figure 1. Modular and structural organisation of NRs. All the domains are highlighted on the example of heterodimer of retinoid X receptor (blue) and vitamin D receptor (green). The A/B region hosts the activation function 1, a ligand independent activation region of NRs. C domain is the DNA-binding domain, D is the hinge bridging the E domain, which is the ligand-binding domain with C domain and the F domain is a very end of the C-terminal region. In most of the cases the A/B and F regions due to their flexibility are not visible in the crystal structures. The orange cylinder represents the short CoA peptide that binds in the close proximity of the helix 12 et the end of the C-terminus. The DNA is represented in white surface and the orientation of the half hexameric sites are shown with black arrows.
7 2.1.3.3 The Hinge region (D domain)
The role of the poorly conserved hinge region (domain D) is to connect the N-terminal DBD with the C-terminal LBD (Figure 1). Domain D, except for providing the bridge between these functional domains, also allows their rotation relative to each other, thus granting orientational flexibility between the 2 domains. As a result of this flexible structure, the whole protein is permitted to take different conformations, without creating any steric hindrance problems. Studies have also shown that this region accommodates motifs that are responsible for effective sub-cellular translocation, such as a very conserved nuclear localisation sequence (NLS) that targets NRs to nucleus. Examples of these exist for the ER (Burns et al. 2011) and the VDR (Shaffer et al. 2005). For certain NRs such as GR, the hinge also provides a coregulator binding interface (Hong et al. 2008). Mutations in the hinge region may affect coregulator-binding, as shown for PPARα, where the binding to N-CoR will be impaired (Liu et al. 2008). However, in the natural T3R variant, which is phosphorylated on its DBD, in the presence of bound agonist, its steroid receptor CoA 1 (SRC-1) interaction shows an impairment (Pawlak et al. 2012). Despite the fact that hinge region was considered in the past only as a bridge between the functional domains, more evidence indicates that its dysfunction may lead to reduced functionality in both coregulator interactions and dimerisation with partner NRs. A so-called “hinge swap mutation” introduced in the ERα hinge region disrupts the synergy between AF-1 and -2 domains (Zwart et al. 2010). When four amino acids are inserted in the hinge region of FXR, its DBD’s binding affinity is greatly reduced (Zhang et al. 2003). Interestingly, the hinge region may be subjected to different PTMs such as acetylation, methylation, SUMOylation
& phosphorylation. Such modifications may affect NRs transcriptional ability or even sensitivity towards binding different compounds and both negative and positive effects have been observed for various PTMs (Wang et al. 2001). A mutation in ERα that misses a consensus SUMO-1 attachment motif, and consequently impaired SUMOylation, may result in a transcriptionally blunted receptor (Sentis et al. 2005). On the other hand, a SUMOylation in PPARα with an increased recruitment of N-CoR shows increased repression profile (Pourcet et al. 2010).
2.1.3.4 The LBD (E domain)
LBD is a highly conserved domain, though at a level lower than that observed for the DBD.
The LBD is responsible for ligand-binding, It also provides the site for cofactor-binding as well as a surface area for dimerisation with the LBDs of other NR molecules in order to create homo- or hetero-dimers. All the LBD structural data starting with the very first determined X-ray LBD structure, the apo-RXRα (Bourguet et al. 1995) led to the recognition that most LBDs consist of 11-13 α-helices. At tertiary level these helices form a three layer sandwich-like structure. In addition to the α-helices there are also β-sheets that are located in the N-terminal part of the LBDs usually between helices H2 and H3. Functionally the LBD has four different parts i) dimerisation surface, ii) ligand-binding pocket (LBP), iii) cofactor-binding site and iv) AF-2, which contains the helix H12 that is crucial for the activation of the NR in a ligand-dependent manner. This last particular helix undergoes dramatic conformational changes in order to allow CoA-binding in transcriptional activation (Moras and Gronemeyer 1998, Aranda and Pascual 2001, Bain et al. 2007, Jin and Li 2010). Some NRs are able to interact with heat-shock proteins through their LBDs, whilst localised in the cytoplasm (Pratt and Toft 1997). The role of LBP, due to its plasticity and dynamic nature, is to recognise and interact with differently sized ligands, or compounds to achieve the optimal modulation of a NR. Compared to other NR structural regions, the LBP is one the least conserved. Depending on the NR, the size of LBP differs and correlates with its function to bind specific compounds (Jin and Li 2010). In case of CAR, full activation can be achieved by binding of smaller lipophilic compounds, ligands. Although upon binding of various ligands the shape of the LBP changes the volume and size does not show big variation (Molnár et al. 2013). On contrary, the PXR is able to bind both relatively smaller ligands, such as a cholesterol-level lowering drug SR12813 as well as the antibiotic Rifampicin (RIF), which is considered to date the bulkiest ligand for NRs (Table 4). Here,
the smaller volume guarantees perfect hydrogen bonding network between the small compound and the hydrophobic LBP that provides the polar side chains. Upon ligand- binding the first contacts are made with mostly apolar amino acids located in the helices H3, H4, H6, H7, H11 and H12. To ensure the selectivity of the LBP towards the PXR ligand only few hydrophilic residues, close to the β-turn, are contributing to the binding. The biggest contrast can be seen when steroid and adopted orphan receptors’ pocket sizes are compared, the latter have relatively large pocket volume that can shelter differently sized, shaped and structured compounds, whereas the steroid receptors’ pocket is only able to harbour specifically limited number of small, lipophilic ligands (Gronemeyer et al. 2004, Germain et al. 2006, Jin and Li 2010).
The very last C-terminal helix H12 of the LBD may serve as the entrance to the LBP since it acts as a flexible lid above it and has several distinguished/specialised residues that are important for the perfect functioning of AF-2. Helix H12 goes through the greatest conformational changes during ligand-binding. Both, apo-, and holo-NR crystal structures are pointing to similar mechanisms by which the AF-2 exerts its transcriptional activity, which is to allow the interaction with either CoA or proteins (Figure 1, orange). Upon ligand-binding, number of formed interactions will push the helix H11 to reposition as the continuance of helix H10 and accompanying helix H12. The induction of AF-2 has a stabilising effects on helix H3 and H4, loop H3-4 and helix H12 itself creating an interaction surface for CoA-binding. One of the most favourable cofactors for interacting with the NR superfamily is SRC-1 (Bain et al. 2007). SRC-1 and other CoAs that are interacting with NRs contain one or more precise short sequences with leucine rich LXXLL motifs, which bind through hydrophobic interactions to the groove formed by the helices H3, H4-5 and helix H12. The LXXLL motif of CoA has a secondary structure with two turns of α-helix and with
Figure 2. NR as molecular machine. NRs are forming four different molecular interactions. (A) Recognising and binding to a specific DNA sequence through its DBD;(B) forming homo-, or heterodimers to enable the effective binding to their target genes; (C) allowing small lipophilic ligands to bind to LBD;(D) upon ligand-binding the recruitment of transcriptional cofactor proteins.
