Linköping Studies in Science and technologyDissertation No. 1179

Molecular Aspects of

Transthyretin Amyloid Disease

Karin Sörgjerd

BiochemistryDepartment of Physics, Chemistry and Biology

Linköping University, SE- 58183 Linköping, Sweden

Linköping 2008

© 2008 Karin Sörgjerd Cover art: Palmyra Wall Paper, Spring Collection 2006, Osborne & Little ISBN 978-91-7393-906-5 ISSN 0345-7524 Printed in Sweden by LiU Tryck Linköping 2008

Att springa vilse ibland gör att man får

upptäcka mer av världen

Abstract This thesis was made to get a deeper understanding of how chaperones interact with unstable, aggregation prone, misfolded proteins involved in human disease. Over the last two decades, there has been much focus on misfolding diseases within the fields of biochemistry and molecular biotechnology research. It has become obvious that proteins that misfold (as a consequence of a mutation or outer factors), are the cause of many diseases. Molecular chaperones are proteins that have been defined as agents that help other proteins to fold correctly and to prevent aggregation. Their role in the misfolding disease process has been the subject for this thesis. Transthyretin (TTR) is a protein found in human plasma and in cerebrospinal fluid. It works as a transport protein, transporting thyroxin and holo-retinol binding protein. The structure of TTR consists of four identical subunits connected through hydrogen bonds and hydrophobic interactions. Over 100 point mutations in the TTR gene are associated with amyloidosis often involving peripheral neurodegeneration (familial amyloidotic polyneuropathy (FAP)). Amyloidosis represents a group of diseases leading to extra cellular deposition of fibrillar protein known as amyloid. We used human SH-SY5Y neuroblastoma cells as a model for neurodegeneration. Various conformers of TTR were incubated with the cells for different amounts of time. The experiments showed that early prefibrillar oligomers of TTR induced apoptosis when neuroblastoma cells were exposed to these species whereas mature fibrils were not cytotoxic. We also found increased expression of the molecular chaperone BiP in cells challenged with TTR oligomers. Point mutations destabilize TTR and result in monomers that are unstable and prone to aggregate. TTR D18G is naturally occurring and the most destabilized TTR mutant found to date. It leads to central nervous system (CNS) amyloidosis. The CNS phenotype is rare for TTR amyloid disease. Most proteins associated with amyloid disease are secreted proteins and secreted proteins must pass the quality control check within the endoplasmic reticulum (ER). BiP is a Hsp70 molecular chaperone situated in the ER. BiP is one of the most important components of the quality control system in the cell. We have used TTR D18G as a model for understanding how an extremely aggregation prone protein is handled by BiP. We have shown that BiP can selectively capture TTR D18G during co-expression in both E. coli and during over expression in human 293T cells and collects the mutant in oligomeric states. We have also shown that degradation of TTR D18G in human 293T cells occurs slower in presence of BiP, that BiP is present in amyloid deposition in human brain and mitigates cytotoxicity of TTR D18G oligomers.

Included papers Paper I: Detection and characterization of aggregates, prefibrillar amyloidogenic oligomers, and protofibrils using fluorescence spectroscopy., Lindgren M, Sörgjerd K, Hammarström P., Biophys J. 2005 Jun;88(6):4200-12. Paper II: Prefibrillar Amyloid Aggregates and Cold Shocked Tetrameric Wild Type Transthyretin are Cytotoxic. Sörgjerd K, Klingstedt T, Lindgren M, Kågedal K, Hammarström P. In manuscript. Paper III: Retention of misfolded mutant transthyretin by the chaperone BiP/GRP78 mitigates amyloidogenesis., Sörgjerd K, Ghafouri B, Jonsson BH, Kelly JW, Blond SY, Hammarström P., J Mol Biol. 2006 Feb 17;356(2):469-82. Paper IV: BiP can function as a molecular shepherd that alleviates oligomer toxicity and amass amyloid. Sörgjerd K.,Wiseman R.L, Kågedal K., Berg I., Klingstedt T., Budka H, Nilsson K.P.R., Ron D., Hammarström P. In manuscript.

Sammanfattning Denna avhandling handlar om proteiner. Särskilt de som inte fungerar som de ska utan har blivit vad man kallar ”felveckade”. Anledningen till att proteiner veckas fel beror ofta (men inte alltid) på mutationer i arvsmassan. Felveckade proteiner kan leda till sjukdomar hos människor och djur (man brukar tala om amyloidsjukdomar), ofta av neurologisk karaktär. Exempel på amyloidsjukdomar är polyneuropati, där perifera nervsystemet är drabbat, vilket leder till begränsad rörelseförmåga och senare till förlamning; och Alzheimer´s sjukdom, där centrala nervsystemet är drabbat och leder till begränsad tankeförmåga och minnesförluster. Studierna som presenteras i denna avhandling har gått ut på att få en bättre förståelse för hur felveckade proteiner interagerar med det som vi har naturligt i cellerna och som fungerar som skyddande, hjälpande proteiner, så kallade chaperoner. Transtyretin (TTR) är ett protein som cirkulerar i blodet och transporterar tyroxin (som är ett hormon som bland annat har betydelse för ämnesomsättningen) samt retinol-bindande protein (vitamin A). I TTR genen har man funnit över 100 punktmutationer, vilka har kopplats samman med amyloidsjukdomar, bland annat ”Skellefteåsjukan”. Mutationer i TTR genen leder ofta till att proteinet blir instabilt vilket leder till upplösning av TTR tetrameren till monomerer. Dessa monomerer kan därefter sammanfogas på nytt men denna gång på ett sätt som är farligt för organismen. I denna avhandling har fokus legat på en mutation som kallas TTR D18G, vilken har identifierats i olika delar av världen och leder till en dödlig form av amyloidos i centrala nervsystemet. Det chaperon som vi har studerat benämns BiP och är beläget i en cellkomponent som kallas för det endoplasmatiska retiklet (ER). I ER finns cellens kontrollsystem i vilket det ses till att felveckade proteiner inte släpps ut utan istället bryts ned. Denna avhandling har visat att BiP kan fånga upp TTR D18G inuti celler och där samla mutanten i lösliga partiklar som i detta fall är ofarliga för cellen. Avhandligen har också visat att nedbrytningen av TTR D18G sker mycket långsammare när BiP finns i riklig mängd.

Abbreviations ANS 8-anilino-1-naphthalene sulfonic acid Bis-ANS 4-4-bis-1-phenylamino-8- naphthalene sulfonate CNS central nervous system CtD C-terminal domain DCVJ 4-(dicyanovinyl)-julolidine ER endoplasmic reticulum FAP familial amyloidotic polyneuropathy LCP luminescent conjugated polymer MTT 3[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide NBC neuroblastoma cells NtD N-terminal domanin RBP retinol binding protein SDS-PAGE sodium dodecyl sulphate polyacrylamide gel electrophoresis SSA senile systemic amyloidosis ThT thioflavin T T4 thyroxin TEM transmission electron microscopy TTR transthyretin TTR D18G transthyretin with amino acid substitution from aspartic acid to

glycine at position 18 UPR unfolded protein response

Tack Denna bok sammanfattar vad jobbat med och kommit fram till under nästan fem års tid. Jag är glad att den äntligen är skriven, och att den verkligen blev skriven. För det hade inte hänt om det inte varit för vissa personer som finns i mitt liv. Personer som indirekt eller dierkt har stöttat mig under min tid som doktorand (och som människa). Somliga har varit involverade i mitt forskningsprojekt medan andra helt enkelt har bidragit till att göra mig lycklig. Alla bidrag har varit lika viktiga. Om jag inte nämnt någon är det inte för att jag glömt utan för att jag är tankspridd.

Per Hammarström. Du är garanterat den bästa handledare man kan ha! Tack för ditt oändliga tålamod. Tack för alla konferenser jag fått åka på. Tack för att du stöttade mig i viljan och beslutet att åka till och jobba i USA. Jag är imponerad över att du kan så mycket, förstår så mycket och för att du villigt och generöst delar med dig av det. Tack också för all den tid du har gett mig, för att du alltid trott på mig, för ditt enormt trevliga sätt, för ditt engagemang och för ditt stöd. Katarina Kågedal. Det har varit underbart att få samarbeta med dig. Vårt gemensamma projekt har varit fantastiskt roligt och det är sorgligt att det inte kan pågå för alltid. Tack för ditt varma sätt, dina uppmuntrande ord och för din hjälpsamhet. Jag tycker att du är toppen och kommer att sakna dig. Luke Wiseman and David Ron. Thank you for welcoming me in your lab and for letting me spend six months there. Thank you for your time and for your endless support. Thank you for pushing me, for helping me, and for believing in me. The people in the Ron lab. Thank you for contributing to a nice environment at Skirball institute and for making me feeling comfortable. Stefan Klintström. Tack för den positiva energi du sprider omkring dig och för allt det goda du gör för Forum Scientium och alla doktorander som är med där. Tack också alla i Forum Scientium för inspiration, härlig stämning och roliga studieresor. Mikael Lindgren. Tack för trevliga samarbeten. Du är en oändlig källa av kunskap. Tack för att du hjälpte mig att skriva en post doc ansökan. Bijar Ghafouri. Tack för tålmodigt sekvenserande av transtyretin med mig. Tack vare dig föll sista pusselbiten på plats i min första artikel. Bengt-Arne Fredriksson. Tack för hjälp med elekrtonmikroskopi-körningar som resulterade i utmärkta bilder av TTR. Rajesh Mishra. Thank you for being a nice office partner for two years, for helping me with writing my resume and for being a good companion. Sylvie Blond. Thank you for

De jag vill tacka är:

nice cooperation with the BiP work. Linda Hendershot. Thank you for providing me with BiP antibody. Tack alla på Patologen II för att ni låtit mig springa där och för att ni alltid varit trevliga och välkomnande. Min mentor Kajsa Uvdal. Tack för givande samtal som gjort mig inspirerad och glad. Tack för ditt engagemang och för att du övertygade mig om att 30 inte är en ålder när jag hade 30-årskris. Susanne Andersson, tack för hjälp med blanketter, iväg-skickande av paket, svarande på dumma frågor och mycket annat. Du verkar ha koll på det mesta. Louise Gustavsson Rydström, tack för att du är en så bra personalintendent och tack Rita Fantl för att även du har haft koll på administriva saker och alltid varit vänlig och trevlig. Tack alla medarbetare på biokemi-avdelningen, ni som kommit, ni som gått och ni som fortfarande finns kvar. Ni har alla bidragit till en positiv stämning i labbet, fikarummet och i korridorerna, vilket är ovärderligt. Särskilt tack till ”seniorerna” Uno Carlsson, Maria Sunnerhagen, Peter Nilsson, Nalle Jonsson, Magdalena Svensson, Kristina Borén och Martin Karlsson för all kunskap ni besitter och för att ni spridit den vidare. Speciellt tack till Nalle som med stor välvilja hjälpte mig med ultracentrifugkörnigar, även när kvällarna blev sena. Jag vill såklart också nämna självaste Hammarström-gruppen som har jag varit stolt (nästan mallig) över att tillhöra och som jag inte kunnat haft mer tur med. Ni har både varit trevlig sällskap och god hjälp under många år. En liten familj på jobbet helt enkelt, och ni består av (förutom ovan nämnda handledare).... Mina kontorskompisar Ina och Therése. Vad jag kommer att sakna er. Tack Ina för den ljusblå alla-hjärtans-dags filten jag fick när jag som mest behövde den, den har värmt mig varje dag och värmer mig fortfarande. Tack för diskussioner, råd och stöd som både har haft och inte har haft med jobbet att göra men alltid varit lika viktiga. Therése, min första och sista exjobbare, som jag lämnade för en vistelse i USA, jag önskar att jag vore så duktig som du är. Tack för input och för hjälp jag fått i mitt manusförfattande och avhandlingsarbete. Och tack för leverens av alla salta blåbär. De har gett mig oförväntad energi när jag mest behövt det. Karin, mitt bästa konferensresesällskap, vad hade Salt Lake City varit utan dig? Jag vet att du tycker att jag har överdrivit temperaturen på hotellrummen ibland men tur att du stått ut ändå... Ingen annan har heller förstått vikten av onödigt dyrt schampo. Sofie, det var bara du och jag när det började. Sedan blev det fler. Tack för att du varit med mig under hela resan. Du är imponerande och duktig och jag tror starkt på dig. Daniel, tack för att du är så hjälpsam och så snäll, för att du lyfter saker som är tunga och för att du fixar med tekniska prylar och sånt jag inte begriper mig på. För att bibehålla energi, motivation och gott humör har det även varit viktigt med luncher och fikaraster. Och vad hade dessa varit om inte ni runt runda bordet funnits... Tack alla ni som suttit där, men inte bara där.... Speciellt vill jag tacka:

Cissi, för att du provkörde en bil med mig, även fast jag inte köpte den. Tack också för att du alltid ställer upp och har en så avslappnad och skön attityd. Veronica, tack för att du hälsade på i New York, det var upplyftande och roligt. Du verkar ha en bekymmersfri ådra som gör att man blir glad av att umgås med dig. Patricia, du har garanterat bakat de godaste kakorna under åren. Tack för dem. Synd att de inte blev fler... Anngelica, du var saknad redan första dagen du var borta från jobbet. Dessutom är du den bästa hyresvärdinna man kan ha, tack för att du hade bäddat sängen åt mig när jag kom hem från USA och tack för att du lade fram rena handdukar. Satish, it is nice to have you around. You contribute to the international feeling on Biokemi, I like that. Stort tack också till alla exjobbare som varit med under delar av resan, samt övriga doktorander som jag stöter på i korridorer och i fikarum. Till gamla medarbetare som jag vill nämna hör: Carin, vad jag saknade dig när du slutade och flyttade till Stockholm, men vad roligt vi har när vi ses. Tack för trevliga stunder och för ditt uppriktiga engagemang i mitt ibland röriga liv. Amanda, tack för springturer (trots att vi springer vilse ibland), promenader och givande samtal. Tack för att du alltid lyssnar, stöttar och för att du får mig att skratta. Louise, tack för att du tar med mig på konserter, tack för att du tar dig tid att träffa mig när och tack för att du är en sådan bra vän. Laila, du är min favorit-fotograf. Tack för att jag fick vara fotomodell. Tack också för trevligt resesällskap, många resmål har det blivit... Alla studiekompisar och övriga kompisar har varit viktiga för mig. Speciellt tänker jag på: Malin, du var min bästa studiekamrat som hjälpte mig att fixa tentor. Utan dig hade jag kanske inte varit här idag. Synd att vi inte ses oftare men när vi träffas vet vi att vår vänskap varar ändå. Tack för många och långa givande telefonsamtal under åren. Isabelle, tack för att du alltid förstår vad jag säger och tack för att jag får vara din vän. Tack för alla år som jag varit korridorskamrat med dig och för den fantastiska tiden i Provence som var en av de bästa i mitt liv. Tack också för att jag alltid är välkommen till dig och för att jag fick ha min 30-års-fest hos dig. Erika, min barndomsvän och äldsta vän. Vi har känt varandra sedan vi var pyttesmå. Tack för att du alltid funnits vid min sida och för att du aldrig svikit mig, för att du vet allt om mig, och litet till, men tycker om mig ändå. För att du aldrig dömer men alltid uppmuntrar. Jennie, du är den bästa ryssa man kan ha. Tack för att jag kan skratta så hysteriskt när jag är med dig. Det började i franska-gruppen (alla minns oss därifrån, jag lovar...) och vi fnittrar lika mycket än. Det är roligt. Tack också Marcus för att även du accepterar mig som en liten del av ditt liv ;-) Gummorna; Johanna, Sarah och Sofia, tack för fantastiska tjejträffar med energipåfyllnad, struntprat och glädje. Det måste bli många fler. Sarah, jag är glad över att jag har fått uppleva så mycket roligt och tokigt med dig under många år. Sofia, jag är glad över att ha fått lära känna dig och jag hoppas vi kommer att ses oftare när jag kommer till Stockholm. Johanna, tack för det fantastiska stöd du varit för mig många gånger när jag känt mig nere och för den energispruta du är när vi ses. Jag har alltid blivit gladare av att träffa dig, oavsett hur jag känt mig från början. Mari, tack för att du kom till New York och hälsade på mig, för att du alltid säger så bra saker och tack för att du alltid lyssnar. Linda, jag är glad över att du lämnat solen i Spanien och kommit hem till kalla Sverige, då kan vi ju ses

oftare i framtiden... Izabela, tack för alla goda minnen sedan lååångt tillbaka. Hanna och Maria, vad hade Flamman varit utan er... Min New York-tid var rolig och intensiv. Mycket arbete blev det men även mycket annat i den stad som aldrig sover.... Tack Sara, Michelle, Cissi, Thomas, Thomas och Per Anders. Det var ni som gjorde New York för mig. Utan er hade jag inte varit något där. Sara, utan dig hade jag inte haft så många par skor med mig hem till Sverige. Tack för excellent shoppingsällskap. Michelle, synd att vi bara var roomates i en månad men bara efter några dagar med dig kändes det som vi känt varandra i ett helt liv. Cissi och Thomas; det kändes tryggt att lära känna Östgötar i New York, och kul att lära känna just er! Merci Thomas, pour d´être fantastique och tack Per Anders för tillförsel av djup i mitt liv. Polle, min gamle kompis och Ulla, min söta gudmor, tack för att ni alltid funnits i mitt liv. Christian, tack för alla år du var tillsammans mig. Du kommer alltid att ha en speciell plats i mitt hjärta. Min syster Maria, jag är glad över att just du är min syster. Tack för energipåfyllnad jag får när jag kommer till Uppsala och för att jag alltid är välkommen dit. Min svägerska Caroline, du hade inte kunnat vara en bättre svägerska, jag är glad över att du träffade Maria, och för att du finns även på min sida när jag behöver det. Mina älskade syskonbarn Estelle och Lovisa, ni ger mig livsglädje. Mamma och pappa, tack för att ni älskar mig, för att ni tror på mig och för att ni finns.

1

Table of contents

Table of contents _____________________________________________________1

Introduction _________________________________________________________3

Proteins_____________________________________________________________5 Protein production- the background story ___________________________________________6 Protein folding ________________________________________________________________8 Genetic mutations _____________________________________________________________9 Protein misfolding ____________________________________________________________10

Transthyretin (TTR)__________________________________________________13 The TTR D18G mutation_______________________________________________________18

Molecular chaperones ________________________________________________21 BiP ________________________________________________________________________22 Structure and mechanism of BiP _________________________________________________23 A role for BiP during translocation _______________________________________________25 The role of molecular chaperones in misfolding diseases ______________________________25

The ER, cellular stress and cell death____________________________________27 The unfolded protein response (UPR) _____________________________________________28 Apoptosis ___________________________________________________________________31 Caspases____________________________________________________________________31

Methods ___________________________________________________________33 Cloning, mutagenesis__________________________________________________________33 SDS-PAGE and Western blotting ________________________________________________33 Circular Dichroism____________________________________________________________33 Fluorescence spectroscopy______________________________________________________34 Chemical cross-linking ________________________________________________________35 Transmission Electron Microscopy (TEM) _________________________________________35 Affinity chromatography and immunoprecipitation___________________________________36 Analytical ultracentrifugation ___________________________________________________36

Results_____________________________________________________________37 Paper I________________________________________________________________ 38

Cross-linking to probe formation of aggregates______________________________________38 Size and morphology of aggregates and protofibrils __________________________________39 Characterization of TTR conformers using molecular probes ___________________________40 Different kinetics for different probes _____________________________________________41

Paper II _______________________________________________________________ 42 Early oligomeric species of TTR kill human cells____________________________________42 Early oligomeric species of TTR induce ER stress ___________________________________44

Paper III ______________________________________________________________ 45 BiP selectively binds to destabilized variants of TTR _________________________________45 Composition of the BiP- TTR D18G complex_______________________________________46 The BiP- TTR D18G interaction _________________________________________________46 BiP plays a protective role against the toxic effects of TTR D18G _______________________47

2

Paper IV ______________________________________________________________ 49 BiP interacts with TTR D18G in the mammalian ER _________________________________49 The degradation rate of TTR D18G is slowed down in the presence of BiP ________________49 The BiP- TTR D18G complex was present in a wide distribution of molecular weights ______51 BiP can protect cells from TTR D18G cytotoxicity___________________________________52 BiP is found in TTR D18G aggregates in patient tissue _______________________________53

Conclusions ________________________________________________________55

References _________________________________________________________57

3

Introduction This thesis summarizes what I have been working on for the past five years and what

conclusions I have made from my findings. My main characters are two proteins

called BiP; which is a molecular chaperone believed to play a protective role in cells,

and transthyretin (TTR); which is associated with human misfolding disease. It has

been known for a long time that TTR misfolding disease starts with TTR denaturation

and leads to aggregation and fibrillation of TTR, which accumulates in tissues and

organs in patients suffering from the disease. Still, there are no cures for most of these

kinds of diseases and the pathogenesis and mechanisms are not fully understood.

The aims with my studies have been to elucidate what role BiP plays in TTR

misfolding diseases. I have specifically studied a mutant of TTR called TTR D18G,

since that mutant is the most destabilized and most unusual form of TTR. I have also

aimed to follow the mechanisms for TTR misfolding and to study the consequences in

human cells when exposed to misfolded variants of TTR.

I have included four papers in this thesis. In the first, the aggregation process of

transthyretin is described, and the different states in the process are characterized. In

the second paper, the effect of different conformational states of aggregated TTR

variants on human cells has been studied. In the third paper, the interactions of TTR

D18G with BiP are characterized and hypothesizes about what role BiP plays in TTR

misfolding diseases have been made. In the fourth paper, the BiP TTR D18G

interaction is studied from a mammalian point of view and the effects of BiP and TTR

D18G on human cells are elucidated.

4

5

Proteins

Proteins are responsible for most of the reactions occurring in the human body such as

transport of nutrients and oxygen, defense against microorganisms, control of gene

expression, transmission of signals etc. In all organisms and in each cell, they exist

and they work. In the human body, over 30 000 different proteins (and around 30 000

protein coding genes) [1] with almost as many different functions are present,

however, most of them still have unknown functions. The building blocks for proteins

are called amino acids. There are 20 amino acids used for protein synthesis and 12 of

them can be produced by human cells whereas eight need to be supplied with the diet.

Thus, a versatile diet is important for the body to work properly with all its > 30 000

different proteins to be properly synthesized. An average protein consists of hundreds

of amino acids, linked together in different sequences by the peptide bond, and it is

the order of the amino acids that dictate the final shape of the protein. The amino

acids, the building blocks in a polypeptide, have different properties; they can be

polar, non-polar or charged and the hydrophobic ones are usually buried in the interior

of the folded protein. The structures of the proteins, i.e., their conformations, differ

due to different types of secondary structures, called α-helices or β-sheet and how

these structural elements are arranged. It is the conformation that dictates the protein

function. Every single amino acid in the folded protein can contribute to and play a

role for protein function. A substitution of one amino acid to another might in some

cases lead to re-arrangements of the whole protein structure, and thereby induce a new

behavior of the protein (often leading to destabilization and degradation). In other

cases, an amino acid substitution does not influence the conformation at all.

6

Figure 1) The primary structure of a protein showing amino acids as a string of pearls. The side

chains of the amino acids, denoted with R can be polar, non-polar or charged. When the amino acids

are connected to form a poly peptide chain, the COO- group of one amino acid reacts with the NH3+ of

another, and a peptide bond is formed with release of a water molecule.

Protein molecules can interact with each other, and protein-protein interactions are

necessary for many biological functions. Interactions can be prolonged, when a

complex is formed, or transient, i.e. when signals are transferred within or between

cells. Interactions can also be non-preferred, such as protein aggregation.

Protein production- the background story

All proteins begin as linear sequences of amino acids linked together as a string of

pearls (Figure 1). The information about the amino acid sequence of the protein,

leading to their different conformations is encoded in the deoxyribonucleic acid

(DNA), in the specific genes for the proteins of interest, called the genetic code.

When synthesis of a polypeptide begins, the DNA information is transferred to a piece

of messenger RNA (mRNA). Formation of RNA from DNA is a process called

transcription and occurs with help from an enzyme called RNA polymerase. DNA

was identified in 1944 and its double helical conformation was revealed in 1953 [2].

Since DNA is situated in the cell nucleus whereas the protein synthesis occurs in the

cytoplasm, an intermediate needs to be involved in the transcription. In the fifties,

Primary protein structure Amino acid

C COO-

NH3+

R

H

Primary protein structure Amino acid

C COO-

NH3+

R

H C COO-

NH3+

R

H C COO-CC COO-

NH3+

R

H

7

there were discussions about that intermediate and it was proposed that another

nucleic acid, single stranded ribonucleic acid (RNA), would be the intermediate

responsible for transferring information from the cell nucleus to the cytoplasm. Later,

it was formulated that the genetic information in the DNA is transcribed to RNA and

then translated from RNA into a protein. The idea was developed over time and in the

sixties, it was proposed that a gene is transcribed into a specific RNA species, called

mRNA and that a short-lived, non-ribosomal RNA directs the synthesis of proteins. In

1965, Francois Jacob, Jacques Monod and André Lwoff received the Nobel Prize for

their research about mRNA. In the seventies, it became known that mRNA could be

spliced after transcription, resulting in that the primary transcript can generate

different mRNAs and different proteins. In the 1980s, it was found that small RNA

molecules could bind to a complementary sequence in mRNA and inhibit translation

[3]. mRNA is single stranded and its sequence is called sence because it results in a

protein. The complementary sequence is called antisense. When sence mRNA base

pair with anti sence mRNA, translation is blocked. The mechanism has not been fully

understood until Craig C. Mello coined the term RNA interference in his work from

1997, published in Cell [4]. In 2006, Andrew Z. Fire and Craig C. Mello received the

Nobel Prize for uncovering the mechanism of RNA interference. They discovered that

genes could be silenced, i.e. gene activity could be turned off, by double-stranded

RNA [5].

Protein production is intitiated by transcription from DNA to mRNA. The

transcription starts with binding of RNA polymerase to the DNA which, together with

different cofactors, unwinds the DNA. The unwinding helps the RNA polymerase to

bind the single stranded DNA template. But there is also need for different

transcription factors to make the interaction possible. Once RNA polymerase has

bound, the elongation starts, which means that an RNA copy of the DNA template is

made as RNA polymerase is traversing along the template strand. The copy (mRNA)

is transported to the cytoplasm once it is finished. In the cytoplasm, the sequence is

translated into amino acids with help from the ribosome. Ribosomes consist of

different subunits that surrounds mRNA and use its sequence as a template for amino

acid synthesis where the ribosome is constantly fed with amino acids from transport

RNA (tRNA) molecules, each specific for one amino acid (Figure 2). When the amino

8

Figure 2) From DNA to protein. In the nucleus, mRNA is made, which is a copy of the DNA

template containing all the genetic information. Protein synthesis is performed in the cytoplasm by the

ribosomes.

acid sequence is finished, it is released from the ribosome and folds into a three

dimensional structure and is transported to its predestined destination.

Protein folding

The normal protein function does not appear until the polypeptide has developed its

final three dimensional conformation, i.e. the protein has been folded. Protein folding

involves interactions of the amino acids within the polypeptide to form different kinds

of secondary structures; α-helices and β-sheet (Figure 3). The secondary structure

elements can arrange into a tertiary structure mediated by side chain interactions. The

folding process occurs right after the synthesis of the polypeptide and is normally a

relatively fast process, it can even occur on a milli or micro second time scale. The

Translation

Cytoplasm

Mature protein

Folding

Nucleus

DNA Transcription to mRNA Mature mRNA

Transport to cytoplasm

Ribosome

Amino acids

Translation

Cytoplasm

Mature protein

Folding

Nucleus

DNA Transcription to mRNA Mature mRNA

Transport to cytoplasm

Ribosome

Amino acids

Nucleus

DNA Transcription to mRNA Mature mRNA

Nucleus

DNA Transcription to mRNA Mature mRNA

Transport to cytoplasm

Ribosome

Amino acids

9

American biochemist Christian Anfinsen was the first to show that the order of amino

acids in the primary structure is what dictates the final protein conformation [6]. He

also found that if a folded protein was denatured, i.e. the non-covalent native H-

bonds, the charge-charge interactions and the hydrophobic interactions were broken

and the protein became unfolded, it could again find its folded, native conformation,

i.e. refold, under permissive conditions. Anfinsen was awarded the Nobel Prize in

1972. However, for many proteins, the process is not as simple as it was initially

described by Anfinsen. Some proteins can fold, unfold and refold spontaneously in

vitro (usually the smallest ones) in a one-step process [7], some fold through one or

many intermediates and most need assistance to adopt their ultimate conformation.

The cells are provided with molecules whose major function is to help proteins to fold

correctly, these molecules are also proteins, and are called molecular chaperones.

Genetic mutations

Mutations are permanent alterations in the DNA sequence of an organism and are

classified as point mutations (often leading to amino acid substitutions), insertions

(where a nucleotide has been added to the DNA sequence) or deletions (where a

nucleotide has been removed from the DNA sequence). There can also be large scale

mutations in the chromosomal structure leading to loss of genes or massive repetitive

insertions of DNA. Most mutations do not have any significantly effect on the

organism, but some of them are harmful. Mutations occur in two different ways; they

are either inherited, i.e. they are passed from parent to child (and are present in every

cell throughout the person’s life), or they are acquired during the lifetime. Acquired

mutations are usually results of environmental factors e.g. exposure of DNA to UV

light, viral infections, chemicals etc. But they can also be copying errors during

division of cells. Acquired mutations cannot be passed on to the next generation

unless they have occurred in sperm or egg cells.

10

Figure 3. Illustration of secondary and tertiary structures of a protein. a) α-helix b) β-sheet, within

a β-hairpin c) tertiary structure of a folded protein consisting of both α-helices and β-sheets and the

spacial orientation of the secondary structure elements are dictated by side chain interactions

(www.pdb.org).

Protein misfolding

When errors occur in the protein folding machinery, it can result in protein misfolding

and misfolding disease. Misfolding diseases are often associated with amyloidosis [8].

The reasons for protein misfolding could be mutations in the genes that code for the

proteins that are to be misfolded, or outer factors like stress. For some cases, it is

unknown why proteins start to misfold. The consequences could be harmful to the

surrounding cells and to the organism in general. Protein misfolding diseases strike

many people during their lifetime, and it seems like the phenomenon has become

more prevalent during the past years. One early disease of this kind was described in

1906, when the German neurologist Alois Alzheimer found a form of amyloidosis

that affects the brain. The disease was later named Alzheimer´s disease (AD). Most

AD patients get the disease sporadically, i.e., it is usually not inherited. The symptoms

for disease include memory disturbance and loss of other intellectual abilities, the

symptoms are also called dementia. To date, more than 20 million people are believed

to suffer from dementia [9]. In the fifties, the first form of transmissible human

amyloid disesase, named kuru, was found among people practicing cannibalism in

Papua New Guinea. Shortly afterwards, a disease with similar pathology was

discovered in Europe and United States; transmissible spongiform encephalopathy

(TSE), among individuals that had been treated with growth hormones extracted from

human cadavers. The TSEs include Bovine Spongiform Encephalopathy or Mad Cow

11

Disease (BSE), Creutzfeldt Jakob Disease (vCJD), Gerstmann-Straussler-Scheinker

(GSS) disease and fatal familial insomnia (FFI). Certain misfolded proteins, called

prions, are implicated in these diseases. The word "prion" stands for "proteinaceous

infectious particle", referring to its pathogenic variants. Stanley B. Pruisiner was the

first to identify the molecular mechanisms of prions [10-13] and he was awarded the

Nobel Prize in Medicine in 1997 for discovering a new infectious agent- a protein.

All misfolding diseases, whether they are sporadic, inherited or transmissible, are

associated with deposition of proteins in different organs, depending on the disease

(Table 1). The proteins involved are normally soluble but have become insoluble and

aggregated and have developed fibril-like structures. Remarkably, the final fibrils are

strikingly similar regardless of the precursor protein, consisting of cross-β-sheet

structure with twisted morphology. The mechanisms for developing fibrils are also

similar; starting with a conformational change in the protein which becomes a

building block for aggregates or clusters that develop long fibril like structures over

time and accumulate in tissues or organs in the body with consequences like impaired

organ function and cell death [14].

Alzheimer´s, Parkinson´s and Creutzfeldt Jakob disease are examples of notorious

misfolding diseases, but there are also less known diseases like Familiar amyloidotic

polyneuropathy (FAP) with related pathology (Table 1). The precursor protein for

FAP is transthyretin (TTR).

12

Clinical syndrome Precursor protein Alzheimer´s disease Aβ-protein Primary Systemic Amyloidosis Ig Light Chain Secondary Systemic Amyloidosis Serum Amyloid A Senile Systemic Amyloidosis (SSA) Transthyretin Familial Amyloid Polyneuropathy (FAP) Transthyretin Finnish Hereditary Systemic Amyloidosis Gelsolin Type II Diabetes Islet Amyloid Peptide Non-Neuropathic Systemic Amyloidosis Lysozyme Cerebral Amyloid Angiopathy Cystatin C Atrial Amyloidosis Atrial Natriuretic Factor Familial Amyloidosis Type III Apolipoprotein A-1 Hereditary Renal Amyloidosis Fibrinogen Table 1) A selection of diseases coupled to protein misfolding and amyloidosis and their precursor proteins.

13

Transthyretin (TTR) Transthyretin (TTR) was discovered in 1942, it became known as prealbumin and was

by then detected in the cerebrospinal fluid. In the fifties, prealbumin was identified as

a thyroxine (T4) binding protein by Sidney Harold Ingbar, which was published in

1958 in Endocrinology [15]. Kanai et al [16] characterized prealbumin as a retinol

binding protein and published their finding in a paper in Journal of Molecular

Biology, and prealbumin became Retinol Binding Prealbumin (RBPA). The structure

of RBPA was described by Blake and colleagues in 1978 [17] and in 1981, the name

transthyretin was accepted [18].

TTR has a molecular weight of 55 kDa and its structure (studied by X-ray

crystallography) is a homotetramer with four identical, monomeric subunits,

composed of 127 amino acids [19]. Each subunit has a molecular weight of 14 kDa

and contains eight β-strands denoted A-H and a helix between strands E and F. The β-

strands in each monomer form a β-barrel consisting of two antiparallel four stranded

β-sheets containing the DAGH and CBEF strands. Association of two monomers to a

dimer forms a β-sandwich stabilized by hydrogen bonds between the H-H (Figure 4)

and F-F strands. Association of another dimeric β-sandwich produces the tetrameric

conformation. The dimers are connected through hydrophobic interactions between

the A-B loop of one monomer and the H-strand of the opposite dimer.

14

Figure 4) The dimeric form of TTR. The dimer is held together and stabilized by hydrogen bonds

between strands H and H and F and F (not shown) in the TTR structure (www.pdb.org, pdb code

1DVQ).

Tetrameric TTR contains two identical T4- binding sites (Figure 5) located in a

channel at the center of the molecule [20]. If one of the binding sites is occupied with

T4, it becomes harder for the second T4-molecule to bind because of an allosteric

effect (negative cooperativity) that takes place in the molecule upon binding of T4. T4

is a thyroid hormone and it plays a role in the metabolism, but is also important for

neuronal development [21]. Free T4 is metabolically active. Plasma TTR functions as

a transporter of T4 in the blood and transports 15-20 % of serum-T4 and around 80%

of CNS-T4 [22]. Other T4 binding proteins and transporters include albumin and

thyroxine-binding globulin. TTR is also involved in the transportation of retinol

(vitamin A) in complex with retinol binding protein (RBP). The TTR–RBP–retinol

complex is formed in the endoplasmic reticulum (ER) of hepatocytes. In TTR, there

are four binding sites for RBP, however, only two molecules can bind at the same

time because of steric hindrance. In the plasma, most of the TTR does not bind RBP

[23, 24]. The source for plasma TTR is the liver. In human plasma, TTR is present at

D A G H H G A D

D A G H H G A D

15

a concentration of 0.25 g/l [23]. The major sites for TTR production are in the liver,

the choroid plexus of the brain and the retinal pigment epithelium in the eye.

There are over 100 known mutations in the TTR gene that are associated with

amyloid deposition, with varying phenotype depending on the mutation [25] (some

are shown in Figure 8). The most common neurodegenerative disease associated with

TTR mutations is familial amyloidotic polyneuropathy (FAP) [23, 26-28], which first

was described in 1952 by Corino Andrade [29] in Portugal. For more than 20 years,

Andrade and colleagues had observed 74 cases from different families suffering from

a progressive and mortal, by then unknown disease. The genetic defect and cause for

disease has been identified as the V30M TTR mutation [30]. FAP V30M TTR has

also been reported in Japan, Spain, America and Scandinavia [23, 31, 32] and is

caused by misfolding of TTR resulting in amyloid fibril formation where TTR

monomers have associated into cross-β-sheets [33]. The amyloids in FAP affect the

peripheral- and the autonomic nervous system. In FAP patients, amyloidogenic,

mutated TTR is produced and found in the plasma and deposited in tissues. The

frequency of the mutant gene is considered at 1 in 100 000 to 1 million [34], however,

some carriers never develop the disease. In patients who develop disease, the age of

onset is between 17 and 78 (with more than 80% developing symptoms before the age

of 40) for the Portuguese patients and it becomes fatal around 10 years after the initial

symptoms [26, 35]. In Sweden is FAP known as “Skellefteåsjukan”. For Swedish

patients, FAP onset is significally later in life with an average age of 57 years. The

first symptoms are usually loss of sensibility in fingers and toes and walking

disability. Later symptoms include cardiac dysfunction, emaciation and renal failure

[35]. Since the main source of TTR production is the liver, FAP patients can be

treated with liver transplantation whereby the FAP mutant secreting liver is replaced

by a liver that only secretes TTR wt. Liver transplantation usually halts the disease

progression and results in amyloid clearance over time, but often with cardiac

amyloid progression [25].

16

Figure 5) Tetrameric TTR (from rat). The binding sites for T4 are pointed out with the arrows

(www.pdb.org, pdb code 1IE4).

Although TTR amyloid deposit disease is associated with TTR variants, senile

systemic amyloidosis (SSA) is a disease associated with TTR wild type and affects up

to 25% of people over the age of 80 and is characterized by amyloid deposits in the

heart [36, 37]. The primary structure of TTR is therefore not the only explanation for

development of TTR amyloidosis [38]. SSA is usually benign and without symptoms,

and mostly men are severely affected [39]. Analysis of the amyloid fibril deposits in

SSA patients have revealed that the amyloids contain fragments of TTR and those

fragments dominate over full length TTR [40, 41]. The fragmentation has occurred at

certain positions (predominantly at positions 46, 49 and 52) in the molecule which

makes it tempting to believe that the cleavage of TTR is the cause of disease since the

cleavage might expose sequences that are prone to aggregate. However, the cleavage

mechanism is not fully understood.

Thyroxin ThyroxinThyroxin Thyroxin

17

Formation of TTR amyloids starts with dissociation of the tetramer into monomers,

that in turn partly unfold and develop aggregates and amyloid fibrils over time [42-

44] (Figure 6). While the structure and properties of amyloid fibrils have been in the

focus for diagnosing and understanding the pathogenesis for amyloid disease, there is

now increasing evidence that the intermediate states in the amyloid formation

pathway, are the most toxic species [45-48].

Another role for TTR, which recently has been published, is the ability to bind the

Aβ-protein, which plays the major role in the pathology of Alzheimer´s disease. TTR

can act in a chaperone like manner and thereby prevent formation of Aβ amyloid

aggregates and thereby possibly halt progression of Alzheimer disease [49].

18

Figure 6.) The amyloid formation process of TTR. The native tetrameric structure of TTR is

destabilized and form a rearranged structure that dissociates into monomers. The monomeric species

are unstable and aggregation prone and can mature into long, inert fibril structures or unfold. Unfolded

TTR can also rearrange to a molten globule like structure (A-state) that has very similar properties as

the monomeric amyloidogenic intermediates.

The TTR D18G mutation

TTR D18G is a naturally occurring mutation in the TTR gene. The mutation was

originally discovered in a Hungarian family, where four definite and three probable

affected members were identified. It leads to amyloidosis in the central nervous

system (CNS) with disease onset at an average age of 44. The affected family

members had extensive amyloid deposition in the leptomeningeal vessels and in the

subarachnoid membrane. In all patients, the symptoms of the disease included

memory disturbance, psycomotor decelleration, ataxia, hearing loss and a majority

also suffered from migraine-like headache, nausea and tremor [50, 51]. The CNS

phenotype is very rare for TTR disease and the fairly late age of onset is surprising

since TTR D18G was identified as the most destabilized TTR mutant found to date.

Recently, it has been demonstrated that a combination of thermodynamic and kinetic

stability of TTR mutants is strongly correlated to disease progression. Therefore, it

Rearranged tetrameric TTR

Monomeric amyloidogenic TTR

nMature amyloid fibril of TTR

Oligomeric prefibrillar TTR

A- state structure of TTR

pH 2

Unfolded, monomeric TTRRearranged tetrameric TTR

Monomeric amyloidogenic TTR

nMature amyloid fibril of TTR

Oligomeric prefibrillar TTR

A- state structure of TTR

pH 2

Rearranged tetrameric TTR

Monomeric amyloidogenic TTR

nMature amyloid fibril of TTR

Oligomeric prefibrillar TTR

A- state structure of TTR

pH 2

Monomeric amyloidogenic TTR

nMature amyloid fibril of TTR

Oligomeric prefibrillar TTR

Monomeric amyloidogenic TTR

nMature amyloid fibril of TTR

nMature amyloid fibril of TTR

Oligomeric prefibrillar TTR

A- state structure of TTR

pH 2

Unfolded, monomeric TTR

19

was expected that the disease onset for D18G carriers should be lower. For

comparision, the L55P mutation is a very aggressive FAP variant with early disease

onset (patients get their first symptoms in their teens to early twenties). In L55P

patients, the variant protein can be detected in serum at amounts comparable to wt

subunits [52]. Analysis of serum and cerebrospinal fluid (CSF) of a heterozygote

D18G patient revealed that only TTR wt could be detected [53], which is an

indication of degradation or accumulation of D18G within the cell or rapid

degradation post secretion. This could explain why patients do not develop disease

until 44 years of age.

TTR D18G is monomeric (Figure 7) and unable to form tetramers under physiological

conditions. The mutant is aggregation prone and aggregates 1000-fold faster that TTR

wt under physiological conditions [53]. The location of the D18G mutation is at the

end of the A-strand of TTR. The neighboring residues (Figure 7) are known to

stabilize the tetrameric structure. For example, the A25T mutation results in

destabilization of the TTR tetramer and cause CNS amyloidosis [54], the V20I

mutation leads to destabilization of the TTR tetramer and cause cardiac amyloidosis

[55], the F87M/L110M mutations engineer the TTR molecule to be monomeric [56]

and the L111M mutation leads to cardiac amyloidosis [57].

Figure 7) Position of D18 in the TTR monomer and residues believed to influence the tetramer

stability. A) Structure of a TTR monomer. Residues within a radius of B) 5Å C) 7Å

D18D18A) D18D18

D18

A25

V20

L110

L111

D18

A25

V20

L110

L111

B)

C)

20

Figure 8) Primary sequence of TTR with naturally found mutations marked below the wild type

sequence (Hou et al 2007).

The region with the D18 mutation obviously has high impact on TTR tetramerization

and stability. TTR D18G cannot efficiently form hybrid tetramers with TTR wt. T4

binding was found to facilitate tetramerization of D18G in the choroid plexus, where

concentrations of T4 is high, but not in the CSF where the concentrations are lower.

That could explain the TTR D18G prevalence to accumulate in the CNS. Expression

of TTR D18G in E.coli leads to the formation of inclusion bodies [53].

21

Molecular chaperones The term “molecular chaperone” was coined by Ron Laskey in 1978. Laskey

observed that a nuclear protein, called nucleoplasmin, could solve a misassembly

problem during the assembly of histone proteins, termed nucleosomes, in amphibian

eggs. Nucleosomes bind DNA by electrostatic interactions. If the interactions are

broken, e.g. by changes in the physiological conditions, the nucleoplasmin is not able

to rebind the DNA, even if the physiological conditions are readapted, which leads to

aggregation of the protein. This can be prevented by the presence of nucleoplasmin, is

able to bind the nucleosomes and protect them [58].

The molecular chaperones are today known as folding helpers and it is believed that

they are essential for cell survival and for life processes in general. They are present

in the mitochondria, the Golgi, the ER and in the cytoplasm of all cells. The

chaperones can correct mistakes in the folding machinery, unfold and send misfolded

species to be degraded, or hold on to proteins that cannot be folded in a productive

way, thereby preventing escapes of misfolded proteins that could cause damage. The

chaperones direct their substrates into productive folding, transport or degradation

pathways, but they do not become parts of the final structures of the proteins they

interact with [59]. The majority of newly synthesized proteins need assistance to

adopt their final conformation. Molecular chaperones stabilize non-native proteins,

unfold incorrectly folded proteins and send abnormal proteins for degradation. They

do not interact with native proteins, only the unfolded or partially unfolded ones.

They are interacting with the proteins for a finite time and thereafter release their

substrates, often mediated through ATP hydrolysis. Some chaperones interact with a

wide variety of polypeptide chains whereas others are very restrictive and only bind to

specific targets.

22

The heat shock response was first discovered in 1962 in Drosophila flies and the heat

shock proteins (HSP) were identified as a set of proteins whose expression was

induced when the cells were exposed to elevated temperature [60]. Shortly after they

had been discovered, it became evident that their synthesis was not only due to

elevated temperatures in cells but also to other forms of outer stresses, like radiation

(UV or gamma-irradiation), oxidative stress, exposure to heavy metals, amino acid

analogues etc. Protein misfolding and aggregation can lead to acute or chronic stress

and activation of inappropriate signaling pathways. HSPs have strong cytoprotective

effects [61] and are thought to restore the cellular homeostasis when it is disturbed.

Mammalian HSPs are classified according to their molecular weights (in kilodaltons)

and are divided into two main groups, the high molecular weight HSPs and the small

molecular weight HSPs. The first group includes three major families: Hsp60, Hsp70

and Hsp90. The first group (the heavy HSPs) consists of ATP dependent chaperones

whereas the second group (the light HSPs) consists of ATP independent chaperones.

The Hsp70 family is the most studied HSP family, containing proteins from 66 to 78

kDa. Some of the Hsp70 proteins are localized in the cytosol (Hsp70 and Hsp72), one

is found in the mitochondrion (mtHsp70) and one in the ER (BiP).

BiP

BiP was first defined as glucose regulated protein with a molecular weight of 78 kDa

(GRP78) or immunoglobulin heavy chain binding protein. Its function as a molecular

chaperone was established by Munro et al [62] in 1986, who demonstrated that BiP is

an ATP dependent member of the Hsp70 family, located in the ER lumen. BiP

interacts with newly synthesized proteins and chaperones them during transport

through the cell and is believed to be one of the most important components in

facilitating folding in the ER. BiP has a molecular mass of 74 kDa and its 3D

structure is not known but it has been defined by X-ray crystallography for DnaK (an

E.coli Hsp70 and BiP homologue).

23

Structure and mechanism of BiP

All the Hsp70 family members have the same structural organization with a 44 kDa

N-terminal ATPase domain (NtD), a 18 kDa C-terminal substate binding domain

(CtD) (Figure 9) and a third domain, belonging to the C-terminal domain whose

function is unknown. The NtD and the CtD communicate allosterically with each

other. If the NtD is occupied by an ATP molecule, the affinity for the substrate in the

CtD is low but if the NtD is occupied by an ADP molecule, the affinity for substrate is

high. If the CtD binding site is occupied by a substrate, the rate of ATP hydrolysis in

the NtD increases [63, 64]. Thus, an unfolded polypeptide captured by BiP, can

undergo cycles of binding and release, cycles that will proceed until BiP binding

motifs no longer are present in the released and folded polypeptide. BiP recognizes a

wide variety of nascent polypeptides with no obvious sequence similarity. However,

experiments that have been done in order to find sequences that BiP preferentially

binds to, have shown that the binding motifs consist of a high proportion of

hydrophobic residues, normally located in the interior of a folded protein. It has also

been shown that those motifs preferably consist of seven amino acids [63].

Figure 9) The NtD and the CtD of the bacterial BiP homologue DnaK. The NtD binds ATP/ADP

(ADP is marked with black, left figure) whereas the CtD is substrate binding (substrate (NRLLLTG) is

marked with black, right figure). The domains communicate allosterically with each other. When ATP

is bound to the NtD, the CtD releases its substrate (www.pdb.org, pdb code 1S3X for left figure and

1Q5L for right figure).

24

A computer scoring system was used to predict BiP binding sites in natural proteins

[65]. The BiP score program uses an algorithm that scores the amino acid sequences

for every chosen substrate, giving a measure for the binding probability of every

heptapeptide. Sylvie Blond, Chicago, USA, used an updated matrix (7pep3) that

includes results from initial affinity panning experiments as well as in vitro binding

studies of synthetic peptides [66-69] and data provided by Dr. Mary-Jane Gething,

University of Melbourne, Australia. The program calculates an integer for each

heptapeptide in the substrate. A relative high value of the integer, i.e. values higher

than five, means that there is a big chance that BiP would bind to the corresponding

amino acid sequence. Aberrant high values for isolated heptapeptides with a proline

residue at position 6 which are not flanked by a hydrophobic residue such as Phe or

Trp were recalculated using the score value 0 instead of 12 for the proline

contribution. In addition, the software allows prediction of putative binding sites for

the bacterial DnaK based on the matrix developed by Bukau and colleagues [70] and

also calculates the hydropathy index for every overlapping seven residue-long stretch

of protein sequence using the hydropathy scale of Kyte and Doolittle [71]. The BiP

scoring software is protected under a license agreement with the University of Illinois

at Chicago and is available upon request (blond@uic.edu).

BiP can self associate into different oligomeric species and it is the CtD that is

responsible for oligomerization. The more oligomeric BiP is, the less active is it [65].

BiP can also be post-translationally modified by phosphorylation and by ADP

ribosylation. These modifications are believed to play a role in the synthesis and the

polypeptide binding of BiP. Accumulation of unfolded proteins in the ER leads to an

decreased amount of modified BiP whereas unmodified, monomeric BiP increases

[64].

Many chaperones need co-chaperones to be effective. Hsp70 chaperones often need J-

domain containing Hsp40 proteins. The function for the Hsp40 proteins is to stimulate

the ATPase activity which is crucial for Hsp70 chaperone activity. BiP can interact

with different J-domain proteins [72] which are necessary for the chaperone function.

The bacterial homologue for Hsp70 is called DnaK and the bacterial homologue for

Hsp40 is called DnaJ. Interaction of DnaK with DnaJ is mediated by the J-domain of

25

DnaJ, which includes residues 2–75. DnaJ can function as a chaperone either by itself

or in complex with DnaK [73].

A role for BiP during translocation

Another role for BiP besides acting as a chaperone is to block the passage of proteins

into the ER, which is also believed to be a protective mechanism for cells. On the ER

membrane, there are sites called translocons. A translocon functions as a passage for

proteins that are going to pass the ER membrane and contains an aqueous pore,

through which proteins can pass. Both on the outer side of the ER and on the ER

luminal sides, there are seals that prevent small molecules to pass the ER membrane.

The ER luminal seal is closed until a nascent polypeptide reaches a size of ~ 70 amino

acids [74]. There are different possibilities regarding how the sealing mechanism on

the luminal side works, and one possibility is that there are proteins in the ER lumen

that can bind to the end of the pore and close it. Art Johnson and colleagues [75] have

demonstrated that BiP seals the luminal end of the translocon to separate the ER from

the cytosol and prevent small peptides to enter the ER. In other words, BiP has

multiple roles in the ER and it can act as a seal only in presence of ATP/ADP.

The role of molecular chaperones in misfolding diseases

A current opinion is that the chaperones play important roles in the protein misfolding

diseases since they are parts of the control system in the cell [76, 77]. All proteins

associated with the classical amyloid diseases are secreted proteins and will therefore

pass the quality control checks within the ER, where they interact with a number of

proteins facilitating protein folding. In some cases, misfolded proteins are

accumulated in the ER [5]. This accumulation causes “ER-stress”, a condition that

normal cells respond to by increasing the transcription of genes encoding ER-

chaperones, such as BiP, to facilitate protein folding or by suppressing the mRNA

translation to synthesize proteins. These mechanisms are called “the unfolded protein

response” (UPR). Once proteins are aggregated into extracellular amyloid deposits

they are quite resistant to degradation.

26

27

The ER, cellular stress and cell death The ER is a membrane bound cellular organelle, consisting of tubules, vesicles and

cisternae. The environment is oxidizing, which facilitates formation of disulphide

bonds in maturing proteins and thereby stabilizing their structures. ER is involved in

protein translation, folding, post translational modifications and quality control of

proteins that are to be secreted from the cell. The majority of secreted or plasma

membrane proteins enter the ER and fold within it. The vesicles of the ER are

responsible for transport of proteins to be used in the cell membrane or to be secreted

from the cell. Molecular chaperones and folding enzymes assist nascent proteins to

fold inside the ER and correctly folded proteins are transported to the Golgi apparatus.

Proteins that are not able to fold or that are misfolded, are accumulated in the ER

since they cannot be exported. There are different mechanisms responding to

accumulation of unfolded or misfolded proteins inside the ER. One of the

mechanisms is termed ER-associated degradation (ERAD), which recognizes the

misfolded proteins and retrotranslocates them to the cytoplasm and send them for

degradation by the ubiquitin-proteasome degradation machinery [78]. Another

mechanism that responds to accumulation of unfolded proteins in the ER is the

unfolded protein response (UPR). Accumulation of unfolded proteins in the ER may

also lead to cell death (apoptosis), if the condition is prolonged and cannot be solved

(Figure 10). ER chaperones and ER components play a crucial role for recognition of

unfolded proteins and are continuously expressed in the ER. [79].

28

Figure 10) The ER functions. Proteins entering the ER are facing different destinies. The correctly

folded proteins are sent for export, whereas proteins that are not able to fold are sent for degradation.

Accumulation of incorrectly folded proteins leads to ER stress, which in turn can result in apoptosis if

the condition is prolonged. Most ER processes involve several chaperone systems as indicated in the

figure.

The unfolded protein response (UPR)

ER has a certain loading capacity, which varies between different cell types and

during a cell’s life. When unfolded proteins are accumulated in the ER, the cell

becomes stressed and the folding machinery gets perturbed. Unfolded proteins have

hydrophobic residues exposed, which normally are buried in the interior of the folded

protein. These hydrophobic parts tend to form (protein) aggregates that are toxic to

cells. ER stress can also occur as a result of starvation, virus infections or heat, and

other outer factors that influence cells negatively, and the condition is either transient

or permanent. The cells respond to the stress by activating a pathway of signals

leading to transcription of more chaperones, e.g BiP. Simultaneously, the translation

of new proteins and the loading of proteins into ER are reduced, and further

accumulation of unfolded proteins is decreased. The phenomenon is called the

RibosomemRNA

Nascent protein

ER

ER stress

Unfolded proteinFolded protein

APOPTOSIS

ERAD

DEGRADATION

EXPORT

Chaperone

Chaperone

Chaperone Chaperone

RibosomemRNA

Nascent protein

ER

ER stress

Unfolded proteinFolded protein

APOPTOSIS

ERAD

DEGRADATION

EXPORT

RibosomemRNA

Nascent protein

ER

ER stress

Unfolded proteinFolded protein

APOPTOSIS

ERAD

DEGRADATION

EXPORT

RibosomemRNA

Nascent protein

ER

ER stress

Unfolded proteinFolded protein

APOPTOSIS

ERAD

DEGRADATION

EXPORT

Chaperone

Chaperone

Chaperone Chaperone

29

unfolded protein response (UPR) [80, 81]. UPR is a cytoprotective phenomenon, but

if the condition is prolonged, it can lead to activation of caspases and ultimately

apoptosis.

ER stress leads mainly to three sets of responses: first, the amount of unfolded

proteins that enters the ER is reduced (lowered protein synthesis and translocation

into the ER); second, the ER folding capacity is increased (transcriptional activation

of UPR target genes) and third, if the homeostasis has not been re-established, cell

death (the cells commit suicide (apoptosis) to protect the organism). ER stress leads to

activation of different signaling pathways, mediated by trans-membrane proteins, so

called stress transducers, which sense the ER overload and transmit a signal to the

cytosol where the transcription and translation of proteins take place. Three pathways

have been identified (Figure 11), mediated by inositol-requiring protein-1 (IRE1),

activating transcription factor-6 (ATF6) or protein kinase RNA (PKR)-like ER kinase

(PERK) [82].

Ire1 was the first stress transducer to be identified. It is a trans-membrane protein with

an ER luminal domain, which can bind to BiP or unfolded ER proteins, and a

cytoplasmic domain which has a site for protein kinase activity. Ire1 is called Ire1p in

yeast and Ire1α or Ire1β in mammals [83]. When unfolded proteins are accumulated

inside the ER, BiP is released from the luminal domain of Ire1, leading to

oligomerization and trans-autophosphorylation of Ire1. The trans-autophosphorylation

leads to activation of Ire1´s effector function which causes double cleavage of mRNA

that encodes a transcription factor called Hac1 (in yeast) or XBP1 (in mammals)

leading to excision of a small fragment. The remaining parts of the mRNA are ligated

and can thereafter work as an activator of UPR targets [82].

The ATF6 pathway involves ATF6α, which is a membrane spanning protein with a

stress sensing domain in the ER lumen. The stress sensing domain binds to BiP and

when unfolded proteins are accumulated in the ER, BiP is released, which makes it

possible for ATF6α to travel to the Golgi. In the Golgi, ATF6α is cleaved, which

results in a released fragment called ATF6f. After its release, ATF6f moves from

Golgi to the nucleus, binds to DNA and activates gene expression. Both the

Ire1/XBP1 and ATF6α pathways lead to increased folding capacity in the ER [82-84].

30

The third mammalian UPR pathway is via PERK, which is an ER localized

transmembrane protein with homology to Ire1 [85]. Activation of PERK is through

dissociation from BiP, which leads to dimerization and trans-autophosphorylation

(similar to Ire1). PERK has also a cytosolic domain with kinase activity. Activation of

PERK leads to phosphorylation of eIF2α, which in turn inhibits protein synthesis.

Translation of a trans activator of UPR, the activating transcription factor 4 (ATF4), is

on the contrary induced when eIF2α is phosphorylated [83].

Figure 11) The unfolded protein response with a central role for BiP.

ER lumen

BiP

Cytoplasm

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

UPR stimulus

Nucleus

Activation of transcription factors Cleavage Inhibition

of protein synthesisGolgi

Activation of ATF4

ER lumen

BiPBiP

Cytoplasm

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

UPR stimulus

Nucleus

Activation of transcription factors Cleavage Inhibition

of protein synthesisGolgi

Activation of ATF4

Cytoplasm

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

UPR stimulus

Nucleus

Activation of transcription factors Cleavage Inhibition

of protein synthesisGolgi

Activation of ATF4

Cytoplasm

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

UPR stimulus

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

UPR stimulus

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

Ire1

ATF6

αAT

F6α

BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP BiPBiP

PE

RK

PE

RK

PE

RK

PE

RK

BiPBiP

P P

UPR stimulusUPR stimulus

Nucleus

Activation of transcription factors Cleavage Inhibition

of protein synthesisGolgi

Activation of ATF4

31

Apoptosis

Sometimes, cells have to die. They can do it in different ways and for different

reasons. One reason for cell death is tissue damage, which results in a process called

necrosis. During necrosis, damaged cells swell and burst and release their contents to

the surrounding area, which in turn can damage the neighbouring cells and give rise to

an inflammation.

Apoptosis means programmed cell death and was first described in 1972 by Kerr and

colleagues [86]. The phenomenon is distinguished from necrosis since it a genetically

controlled process, mediated by proteins called proteases. Proteases have the ability to

cut other proteins into pieces. And the proteases involved in apoptosis are called

caspases. The role of apoptosis is to eliminate unhealthy or unnecessary cells from an

organism without release of harmful substances to the surrounding cells. In apoptosis,

the cells shrink and form something called apoptotic bodies. These in turn can be

digested by other cells. Apoptosis is necessary for an organism to be healthy, and to

regulate the balance between cell death and cell renewal in mammals by destroying

excess or damaged cells. Failure in the apoptotic cascade may therefore have fatal

consequences. For example, in 50-75 % of all cancers, the apoptotic signal that

normally regulates division of genetically damaged cells is lost, and cancer cells can

continue to accumulate. Apoptosis can be blocked with HSPs, since they have been

shown to interfere with caspases [61]. Depletion of Hsp70 can trigger apoptosis

through the caspase 3 pathway, without other stressful stimulus [87].

Caspases

Caspases is a family of calcium dependent cysteine proteases and they are able to

cleave their substrates after aspartate residues. Robert Horwitz and colleagues

identified a gene (in C.elegans) called Ced-3, which coded for a protein with similar

properties to the, by then, only known caspase (caspase 1) and what they found was

required for cell death [88]. After that discovery, other caspases in different organisms

were soon identified and their roles were surveyed [89]. Caspases contain three

domains; an N-terminal domain (which vary in size between different caspases), a

large domain containing the active site and a small C-terminal domain. Between the

different domains, Asp cleavage sites are situated. When caspases are cleaved at the

32

Asp residues, they also become activated. Once a caspase has been activated, it starts

to cleave its substrates at their Asp residues. Since they both are activated when they

are cleaved at their Asp residues and activate their substrates by cleaving at Asp

residues, they can activate each other. An active caspase is a tetramer composed of

two large domains and two small domains, and has two catalytic sites (Figure 12).

Cleavage of the substrates have effects on the cell morphology, which is the hallmark

of apoptosis [90].

Caspase 3 (also known as CPP32, apopain, or YAMA) has been identified as a key

mediator of apoptosis in mammalian cells. It is the most apoptotic protein and is

synthesized as an inactive proenzyme. Caspase 3 is activated by cleavage at the

Asp28/Ser29 bond and the Asp175/Ser176 bond. The activated caspase 3 cleaves and

activates other caspases (e.g. caspases 6, 7 and 9), and there are also different targets

in the cells that act as substrates for caspase 3 (e.g. poly–adenosine diphosphate ribose

polymerase (PARP) and gelsolin).

Figure 12) The inactive procaspase and the active caspase.

Inactive procaspases

Cleavage sites

Cleavage and activation of caspases

Active caspaseInactive procaspases

Cleavage sites

Cleavage and activation of caspases

Active caspase

33

Methods

Cloning, mutagenesis

To make the TTR mutants used in this thesis, and to generate cloned fragments of the

TTR gene for cloning into specific plasmids required for transgenic cells, we

performed site directed mutagenesis, using the polymerase chain reaction (PCR)

technique. The PCR technique uses a cycling program, alternately heating and cooling

the PCR sample in a defined series of temperature steps. At a specific temperature, the

double stranded DNA helix separates and permits DNA synthesis with help from the

enzyme pfu turbo DNA polymerase. To perform the site directed mutagenesis, we

used the Quikchange site directed mutagenesis kit from Stratagene.

SDS-PAGE and Western blotting

In this work, Sodium dodecyl sulphate polyacrylamide electrophoresis (SDS-PAGE)

was used to detect interactions between BiP and different TTR variants and to

determine the stoichiometric composition of BiP-TTR complexes. Western blotting

(based on antibody recognition of specific proteins) was used after

immunopreciptiation experiments to investigate whether BiP was pulled down with

TTR D18G and vice versa.

Circular Dichroism

All optical spectrophotometric methods are built on the principle that electromagnetic

radiation i.e. light; interacts with electrons in a molecule. A molecule that is radiated

will absorb the light if the energy in the radiation is high enough to excite an electron

from its ground state to an excited state. Circular dichroism (CD) means that a

34

substance differently absorbs left and right vectors of circulary polarized light to

different amounts. The substance that have the ability to do so are chiral

chromophores e.g. the peptide bond within a protein.

Circular Dichroism of proteins can be measured at wavelengths where light

absorption takes place. When the dichroism is measured as a function of wavelength,

a CD spectrum is obtained. CD spectra can be divided into far UV (≤ 240 nm) or near

UV (~ 240 – 300 nm). Near UV CD is used to study tertiary interactions within a

folded protein and provides a fingerprint for the specific protein. In the near UV

region, aromatic side chains (Trp, Tyr, Phe are the reporting chromophores). Far UV

is used to study the secondary structure within a protein.

Fluorescence spectroscopy

Fluorescence occurs when a molecule absorbs light of a given wavelength and

thereafter emits light of another, longer wavelength. The process occurs in certain

molecules, called fluorophores, which often are based on a polyaromatic hydrocarbon

scaffold. When a photon is absorbed by the fluorophore, an electron is excited from

its lowest energy level (its ground state), to a higher energy level (excited state), and

stays there for a finite time. When the electron returns to its ground state, energy is

released as an emitted photon. The emitted photon has lower energy than the absorbed

photon because energy is lost during the excited state lifetime, and is therefore of

longer wavelength. Under optimal conditions the fluorescence signal is proportional

to the concentration of the excited molecule. Fluorophores can either be intrinsic

(naturally present in the molecule; Trp residues are the major intrinsic fluorophores in

proteins) or extrinsic (non natural molecules that can be bound at specific sites in

proteins). In this work, the fluorescence technique has been used to characterize

different conformational states of TTR under the oligomerization process. In the first

paper, four different molecular probes were used; 1 anilinonaphthalene-8-sulphonate

(ANS), 4,4-bis-1-phenylamino-8-naphtalene sulphonate (bis-ANS), 4-(dicyanovinyl)-

julolidine (DCVJ) and thioflavin T (ThT). In the second paper we used site directed

insertion of pyrene. Fluorescence measurements were performed with a Hitatchi F-

4500 spectrofluorometer equipped with a Xe-lamp and thermostatic cell holder. Time

resolved fluorescence was performed using a an IBH (Glasgow, UK) 5000 U

35

fluorescence lifetime spectrometer system with 1 nm resolved excitation and emission

monochromators (5000 M), equipped with a TBX-04D picosecond photon detection

module with IBH laser.

Chemical cross-linking

Chemical cross-linking with glutaraldehyde is a method that can be used to determine

the oligomerization state of a protein under different conditions. Glutaraldehyde is

frequently used for induction of covalent cross links between proteins because of its

ability to react with free amino groups, predominantly lysines. The results from the

cross-linking reactions were analyzed by SDS-PAGE. In paper I and II, chemical

cross-linking was performed to assay the rate of the oligomerization process.

Transmission Electron Microscopy (TEM)

The principle for TEM is the same as for an ordinary light microscope. However, in

TEM, there are electrons instead of photons, the light source is an electron gun instead

of a light bulb and there are magnetic lenses instead of glass lenses. The first electron

microscope was invented and designed in 1933 by Ernst Ruska, who later was

awarded the Nobel Prize in Physics for his finding (in 1986). TEM can be used to

study different properties and structures of both physical and biological materials. A

resolution of 2 Å can be achieved under optimal conditions. The electron gun creates

a high intensity electron beam focused on the sample (condensor lenses collect the

beam and keeps it focused when it hits the sample). To obtain an image, the electron

beam must interact with the sample and provide contrast. Generally, the difference in

mass number provides the contrast since a material transmits a different number of

electrons depending on how electron dense it is (i.e. the heavier a material is, the

fewer electrons are transmitted). Thus, the TEM image is a gray scale pattern where

the darker areas correspond to those areas in the sample that scatter electron more

than the lighter areas. Proteins mostly consist of carbon and are therefore stained with

the much heavier uranyl acetate or phosphotungsten acetic acid during sample

preparation, since the samples are analyzed on carbon coated copper grids. In paper I

and II, TEM was used to study the morphology of different oligomerization states of

TTR.

36

Affinity chromatography and immunoprecipitation

Affinity chromatography is a method used to purify and isolate a protein from a

mixed sample. This is performed by loading a column with a matrix containing a

ligand that the protein of interest will interact with and bind to. Non binding proteins

can thereafter be washed out and the protein of interest can be eluted and collected. In

this work, we used an Ni-NTA-affinity chromatography column which has the ability

to bind to imidazol and hexa histidine tags (6-His), which is described in paper III.

The column was used to purify or isolate 6-His-BiP and ligands bound to it.

Immunoprecipitation was performed using anti-TTR and anti-Flag-BiP antibodies

immobilized on sepharose beads, to pull down either TTR with Flag-BiP or BiP with

TTR.

Analytical ultracentrifugation

Analytical ultracentrifugation was used to determine sizes of protein complexes, and

this technique was used in paper III. The principle of analytical ultracentrifugation is

that molecules sedimentate at different rotation speeds depending on their size due to

the applied gravimetric field. Equilibrium sedimentation experiments were run at

different rotation speeds until equilibrium was reached at each rotation speed, while

monitoring absorbance at 280 nm as a function of sample depth. All sedimentation

equilibrium data were obtained on a temperature-controlled Beckman XL-A/ XL-I

analytical ultracentrifuge (Beckman Coulter, Fullerton, CA) equipped with an An50Ti

rotor and a photoelectric scanner. Evaluation of the data was performed using the

Beckman XL-A/XL-1 data analysis software applying the self-association model.

37

Results The findings will be presented in four papers, which are summarized below. The first

paper is a study about TTR and its misfolding and fibrillation pathway. The

oligomeric intermediates in the process were studied and characterized. In paper II,

the different states in the TTR oligomerization pathway were captured, and added to

neuroblastoma cells to elucidate which species were toxic to cells when applied from

the outside. We could see that early oligomers were toxic to neuroblastoma cells. This

resulted in apoptosis and release of BiP into the cytoplasm. In the third paper, the

misfolding of TTR is studied from a disease point of view and the role of BiP in

misfolding diseases is discussed. The most unstable TTR variant found to date, TTR

D18G, was used as a model for the study and we found that BiP strongly interacted

with this mutant, which was not the case for TTR wt or other mutants. Paper IV is a

study of the role of BiP for TTR D18G misfolding within a eukaryotic cell. Most of

the work done previously was in vitro studies and measurements were performed on

purified proteins, made in E.coli cells. However, the fourth study was done in an in

vivo context, to get a cell biologigal aspect of the work. Human kidney cells were

used to express proteins and the interactions inside the cells were studied. We could

see that the in vivo results in human cells correlated well to what we had seen earlier

from E.coli expressed complexes.

38

Paper I

The purpose with the study in the first paper was to characterize and understand the

aggregation process of TTR. We used different techniques to detect structural changes

in the aggregation process. An in vitro protocol for creating TTR oligomers was used.

Oligomers were studied by using fluorescence spectroscopy, circular dichroism,

chemical cross-linking and transmission electron microscopy.

The process for making amyloidogenic oligomers was initiated by unfolding the

native, tetrameric TTR in 10 mM HCl, pH 2. The aggregation pathway was thereafter

induced by addition of NaCl, giving a final NaCl concentration of 100 mM. Salt

induced aggregation is a simple way for creation of soluble aggregates where addition

of a high concentration of salt to acid unfolded monomeric TTR leads to immediate

shielding of the positive charges in the protein which in turn change the structure of

the protein. The modified structure is called the A-state [91] and is molten globule

like and has a high β-sheet content (higher than in native TTR). As soon as the A-state

is formed, the aggregation process takes place. This is a much cleaner experiment to

detect oligomers than starting with the tetrameric protein exposed to pH 4.4 [91].

Therein the rate determing step; tetramer dissociation, and it also produces

precipitated protein which is difficult to study by spectroscopy.

Cross-linking to probe formation of aggregates

Chemical cross-linking using glutaraldehyde was used to study formation of

aggregates over time. After 1-2 hours, soluble aggregates of molecular weights

around 300- 500 kDa were detected. After 2 hours, very large aggregates had formed

with molecular weights > 1000 kDa (Figure 13).

39

Figure 13) Analysis of cross-linked TTR aggregates. A) The SDS PAGE gel demonstrated that

already after 5 minutes, the aggregation process had gone far and large oligomers had been formed. B)

Quantification of the bands showed that more than 50% of the protein was aggregated after 5 minutes

and after 15 minutes, more or less all TTR molecules were in aggregated forms.

Size and morphology of aggregates and protofibrils

Transmission electron microscopy (TEM) was used to investigate samples taken from

a TTR aggregation reaction. Samples were collected after 30 minutes, 2 hours, 24

hours and 430 hours. Small aggregates were visible after 2 hours (<10 nm length and

<5 nm wide), after 24 hours small fibrillar structures had developed with sizes around

20-80 nm long and 3-8 nm wide, but there were also spherical clusters of protein

aggregates with diameters of 30-50 nm present in the sample. After 430 hours, the

fibrillar structures had increased in length to 100 – 200 nm and the spherical clusters

had shrunk to 20-30 nm in diameter (Figure 14A).

A

B

40

Figure 14) From oligomers to fibrils. A) TEM micrographs demonstrated that in the beginning of the

TTR aggregation process, small spherical structures had formed that later in the process shrunk and

fibrils had grown in length, resulting in long fibril like structures. B) Schematic illustration of the

process.

Characterization of TTR conformers using molecular probes

To kinetically detect different states of the TTR aggregation process, we used four

different molecular probes (Figure 15): 8-anilino-1-naphthalene sulfonic acid (ANS),

4-4-bis-1-phenylamino-8-naphthalene sulfonate (Bis-ANS), 4-(dicyanovinyl)-

julolidine (DCVJ) and thioflavin T (ThT). The four probes all have ability to bind to

misfolded TTR and are useful tools for following the aggregation process of the

protein.

8-anilino-1-naphthalene sulfonic acid (ANS) is a hydrophobic, charged and

fluorescent molecule. Hydrophobic interaction of ANS with proteins is a widely used

B

A

41

method for characterizing partially folded states of proteins. The A-state of TTR binds

to ANS [91]. In the thyroxin binding cavities in native TTR, ANS can also bind (and

also other small molecules). Labeling of TTR with ANS or bis-ANS showed that bis-

ANS did not completely fit into the thyroxin cavity but worked better than ANS when

reacted with an aggregated A-state TTR. DCVJ is a molecular probe belonging to a

class called molecular rotors with a unique sensitivity to torsional rigidity of the

surrounding medium. We have used it as a TTR amyloid detector. ThT is the most

widely used fluorescent detector of amyloids, whose action still not is completely

understood [92].

Figure 15) Four different molecular probes were used to characterize different TTR conformers.

Different kinetics for different probes

The fluorescence response when different probes bound to unfolded monomeric TTR,

native tetrameric TTR and different stages in the oligomerization process varied

depending on the probe. Three probes; ANS, DCVJ, and ThT, showed increased

fluorescence as the aggregation process proceeded, Bis-ANS on the contrary showed

less fluorescence throughout the aggregation process (Figure 16). ANS and Bis-ANS

also bound to native, tetrameric TTR. Both ThT and DCVJ could detect both early

oligomers and mature fibrils with the strongest signals from the mature fibrils.

42

Figure 16) Different probes were used to follow the TTR misfolding reaction. Aliquots of the

aggregation reaction of TTR were withdrawn and assayed at 2μM probe + 2 μM TTR. Symbols: ANS

(circles), Bis-ANS (inverted triangles), DCVJ (triangles) and ThT (squares). The fluorescence intensity

of the different probes in the presence of the unfolded monomer state and the burst amplitude from the

fit is indicated with horizontal lines labeled with the letter U and “burst” in colors corresponding to the

probe.

Paper II

In paper I, we characterized different states in the TTR aggregation pathway. In paper

II, we harvested the different states in the aggregation pathway (during fibrillation)

and challenged with neuroblastoma cells with these species. We exposed the cells to

both early, oligomeric TTR species, long and mature fibrils of TTR and native TTR

wt. We also wanted to investigate if BiP was upregulated in cells as a marker for UPR

activation when exposed to TTR oligomers and used immunostaining for BiP to

detect BiP localization in cells.

Early oligomeric species of TTR kill human cells

Phase contrast images of human SH-SY5Y neuroblastoma cells exposed to early

oligomers or mature fibrils of TTR in 48 hours, revealed that cells that had been

exposed to oligomers were dying, whereas cells exposed to mature fibrils were still

alive and healthy (Figure 17). Results from MTT and caspase 3 analysis correlated

well with the cell morphology and we concluded that the toxicity pathway was

apoptotic. Surprisingly, TTR purified and stored at 4˚C was also highly toxic, whereas

43

Figure 17) Phase contrast pictures of cell viability after exposure to TTR. Cells exposed to early

TTR oligomers, which had aggregated for 5 min, demonstrated apoptotic morphology, such as

decreased cell size and more sparse population compared to non exposed cells (C) or cells exposed to

native TTR wt, 22˚C (wt). Cells exposed to TTR that had aggregated for longer than 1 h were more

dense. Cells exposed to TTR that had aggregated for more than one day (24 h and 1 week (1 w))

showed similar morphology as control cells (C).

TTR purfied at ambient temperature (22˚C) was not. The TTR tetramer is labile under

cold temperatures and our results indicate a possible role for an alternate disordered

tetramer toxicity or that dissociation of the tetramer into monomers render this

toxicity.

CCwtwt24 h24 h 1 w1 w

5 min5 min 30 min30 min 1 h1 h 5 h5 h

CCwtwt24 h24 h 1 w1 w CCwtwt24 h24 h 1 w1 w

5 min5 min 30 min30 min 1 h1 h 5 h5 h5 min5 min 30 min30 min 1 h1 h 5 h5 h

44

Figure 18) BiP is upregulated in cells stressed with TTR oligomers and cold, native TTR wt. A)

Confocal microscopy images of cells immunostained for BiP (green). Top micrograph: cells exposed to

vehicle (C). Middle micrograph: cells exposed to early TTR oligomers (5 min). Bottom micrograph:

cells exposed to cold native TTR wt (wt 4 °C). B) Western blot analysis of BiP expression in cells

following exposure to early oligomers of TTR or cold native TTR wt (4 °C). GAPDH was used as a

protein loading control to quantify the level of BiP expression.

Early oligomeric species of TTR induce ER stress

BiP was used as a marker for ER stress. Elevated expression of BiP and cytoplasmic

localization of BiP were detected in cells challenged to early TTR oligomers or cold

native TTR wt. Immunostaining of BiP showed a diffuse staining of BiP in untreated

cells. In cells exposed to early TTR oligomers, BiP was more granulated and brighter

stained than untreated cells and displayed a cytoplasmic staining pattern (Figure 18A).

These results strengthen the hypothesis that BiP has a cytoprotective role in

misfolding disease. Cells treated with cold tetrameric TTR also displayed increased

immunoreactivity towards BiP but showed less significant cytoplasmic staining than

the oligomer treated cells (Figure 18A). The Western blot verified an increased

A B

C

5 min

BiP

GAPDH

wt (4 °C)

0

0.2

0.4

0.6

0.8

C 5 min Native wt (4 °C)

Rat

io (B

iP/G

APD

H)

A B

C

5 min

BiP

GAPDH

wt (4 °C)

0

0.2

0.4

0.6

0.8

C 5 min Native wt (4 °C)

Rat

io (B

iP/G

APD

H)

C

5 min5 min

BiP

GAPDH

wt (4 °C)

0

0.2

0.4

0.6

0.8

C 5 min Native wt (4 °C)

Rat

io (B

iP/G

APD

H)

45

expression of BiP after exposure to TTR oligomers (Figure 18B) and for cells

exposed to cold tetrameric TTR compared to control cells.

Paper III

To understand how the chaperone BiP could interact with an unstable, aggregation

prone, protein mutant like TTRD18G, plasmids containing genes for His6-BiP or FT2-

TTRD18G were introduced into E.coli cells grown in LB media. The proteins were

thereafter expressed with IPTG, the cells were harvested and the protein containing

lysate was purified on a Ni-NTA-affinity chromatography column for capturing His6-

BiP. As controls, BiP was also expressed with FT2-TTRA25T, FT2-TTRL55P or FT2-

TTRwt and were treated in a similar way. The protein containing cell lysate was

separated on the column and fractions containing column flow through, wash buffer,

and protein eluate were analyzed by SDS-PAGE.

BiP selectively binds to destabilized variants of TTR

It became obvious from the SDS-PAGE analysis that FT2-TTRD18G was captured by

His6-BiP since strong TTR bands were observed in the elution fractions when His6-

BiP was co-expressed with FT2-TTRD18G (Figure 19). On the contrary, when His6-BiP

was co-expressed with FT2-TTRwt or the other TTR mutants analyzed, none or very

little TTR was detected on the gel. The binding capacity of BiP to TTR seemed to be

correlated to the stability of the TTR variant which in turn is correlated to previous

work where relation between secretion effiency and thermodynamic and kinetic

stability was found [43].

46

Figure 19) Comparison of different TTR mutants in their ability to bind to BiP. TTR wt was

compared with the mutants TTR D18G, TTR A25T and TTR L55P. FT= Flow Through, w1= wash 1,

w2= wash 2, el= eluate, MW= Molecular Weight marker.

Composition of the BiP- TTR D18G complex

The purified BiP-TTRD18G complex was further analyzed to get an idea of its

molecular composition. Size exlusion chromatography using a Superdex 75

gelfiltration column and analytical ultra centrifugation revealed that the BiP-TTRD18G

complex was composed of a variety of species with molecular masses ranging from

336 kDa to 75,5 kDa. By calculating the ratio between the two proteins against the

molecular mass of complexes, it became clear that the ratio between FT2-TTRD18G and

His6-BiP increased with increasing molecular weight, showing that a mixture formed

of 1:1 complexes and complexes dominated by mutant TTR over BiP.

The BiP- TTR D18G interaction

To identify which part of TTR that BiP binds, we used acid unfolded TTR wt as a

template, which we cleaved with trypsin and analyzed with MALDI-TOF mass

spectrometry. 97% of the TTR sequence is covered with trypsin digestion and three of

the best predicted binding sites, according to the BiP score software, were separated.

His6-BiP was incubated with the TTR fragments and any captured fragment was

purified with Ni-NTA chromatography (binding His6-BiP). One fragment was bound

47

Figure 20) The binding site for BiP in the TTR molecule. The F-strand in the TTR molecule was

found to be its binding site for BiP.

to BiP, identified with MALDI-TOF. The sequence of the fragment was identified

using electron spray ionization (ESI) tandem mass spectrometry (MS/MS). The

yielded tag was searched against NCBI database and was identified as the TTR

sequence 88-103 which corresponds to the C-terminal part of the 81-103 fragment

from trypsinated TTR. Interestingly, the 88-103 sequence includes the complete

sequence of β-strand F in the folded, monomeric TTR (Figure 20). Since the F strand

in one monomer connects to another F-strand in another monomer to form a dimer,

BiP binding to F-strands would therefore compete with F-strands in monomeric TTR

and compromise formation of dimers and also tetramers. The F-strand is also involved

in packing of growing amyloid fibers [93]. Formation of fibrils might therefore be

interrupted when BiP binds to the amyloid formation interface and kept it in a soluble

form.

BiP plays a protective role against the toxic effects of TTR D18G

To investigate what would happen with TTR D18G bound to BiP when released from

BiP, we purified the complex and then added ATP to the sample. Since BiP is an ATP

dependent chaperone, it releases its substrate when treated with ATP. We followed

48

A

B

Figure 21) BiP protects from TTR D18G cytotoxicity by keeping it in a soluble form. A)

Micrographs of TTR D18G complex before (upper left) and after (upper right) addition of ATP. ATP

releases substrates from BiP, and release of TTR D18G results in aggregation. B) ThT fluoresecence of

BiP/TTR D18G (filled bars) and BiP (open bars) before and after addition of ATP or addition of the

competitive peptide 88-103 TTR. Without incubation (w/o inc), without ATP, 37˚C incubation for 18 h

(37˚C no add), with ATP, 37˚C, 18h (37˚C ATP), with 88-103 TTR peptide, 37˚C, 18h (37˚C 88-103).

the process with transmission microscopy, ThT binding and SDS-PAGE before and

after treatment with ATP. Release of TTR D18G from BiP resulted in formation of

visible, insoluble aggregates that not could be seen when it was in complex with BiP.

We also treated the BiP/TTR D18G complex with the peptide TTR 88-103 to see if it

could compete with TTR D18G for BiP binding sites. The competing peptide resulted

in a small amount of released TTR D18G which was detected with ThT fluorescence.

We received about 30% increase of ThT fluorescence compared to the sample that

was not incubated with peptide (Figure 21B). We assume that BiP plays a protective

role against the toxic effects of TTR D18G by maintaining TTR in a soluble, here

benign, oligomeric form.

49

Paper IV

In this paper, the idea was to obtain a cell biological aspect of BiP/TTR D18G binding

and to investigate if the E.coli derived complexes could be confirmed in a human

cellular system. We wanted to study complex formation between BiP and TTR D18G

in vivo and used human 293T kidney cells to overexpress the proteins. We also

wanted to investigate if BiP would influence the degradation rate of TTR D18G.

We could clearly see that the complex was formed in vivo in human cells, which

confirmed our previous results. It is also known from before that BiP has the ability to

oligomerize. We found that BiP did not prevent aggregation of TTR D18G, but rather

oligomerized with it, both in soluble and insoluble aggregates. Surprisingly,

aggregates seemed to accumulate inside the ER. We also found that the degradation of

TTR D18G was altered in presence of a high amount of BiP.

BiP interacts with TTR D18G in the mammalian ER

In paper IV, we transfected human kidney cells with TTR D18G or BiP and thereafter

analyzed the cell lysate using immunoprecipitation and western blotting (Figure 22).

Both soluble and insoluble proteins were analyzed and both BiP and TTR D18G were

present in both fractions. BiP seemed to increase the amount of soluble TTR D18G

when they were co-expressed. BiP appeared to co-fractionate with aggregated TTR

D18G, this is also in accordance of what we found in paper III.

The degradation rate of TTR D18G is slowed down in the presence of BiP

Since TTR D18G is extremely aggregation prone and appeared to accumulate in the

ER, we were interested in investigating its degradation rate in live cells. Pulse chase

(S35) analysis of transfected 293T cells with TTR D18G or TTR D18G and BiP were

performed and showed that overexpression of BiP reduced the degradation rate

(Figure 23).

50

Figure 22) Selective binding of mutant TTR. Cells expressing Flag-BiP and/or TTRwt or TTR D18G

were lysed with triton. The triton supernatants (containing the soluble protein fractions) and the triton

pellets (containing the insoluble protein fractions) were collected. Immunoprecipitation with anti-TTR

antibody shows that Flag-BiP is pulled down with TTR D18G and to a small extent with TTR wt. The

amount of soluble TTR D18G seemed to increase when BiP was overexpressed.

Figure 23) The TTR D18G degradation rate is slowed down in presence of BiP. Pulse chase

analysis results showed that the degradation of TTR D18G alone (circles) occurred faster than when

BiP was overexpressed in the cells (squares).

0

0,2

0,4

0,6

0,8

1

0 2 4 6 8 10

TTR D18GBiP: TTR D18G

Tota

l fra

ctio

n

Pulse

Chase (h) 0 3 6 9 0 3 6 9

Supernatant

Pellet

BiP D18GD18G

Chase (h) 0 3 6 9 0 3 6 9

Supernatant

Pellet

BiP D18GD18G

Triton PelletTriton Supernatant

Flag-BiP

TTR

eIF2alpha

+ + + +

+ + + +

+ + + + + +

75

37

15

TTRwt

TTR D18G

Flag BiP

Triton PelletTriton Supernatant

Flag-BiP

TTR

eIF2alpha

+ + + +

+ + + +

+ + + + + +

75

37

15

Flag-BiP

TTR

eIF2alpha

+ + + +

+ + + +

+ + + + + +

+ + + +

+ + + +

+ + + + + +

75

37

15

TTRwt

TTR D18G

Flag BiP

51

The BiP- TTR D18G complex was present in a wide distribution of molecular weights

In paper III, we showed that soluble complexes of BiP and TTR D18G, purified from

E.coli bacteria, existed as a broad distribution of molecular weights. In paper IV, we

analyzed both the soluble protein fractions, and the insoluble protein fractions from

eukaryotic cells overexpressing TTR D18G and BiP. A glycerol gradient was

performed to determine the size distribution of oligomers in the soluble fractions, and

the influence of BiP on oligomer size (the insoluble fractions were treated with SDS

and DTT in order to denature them; the results were investigated using SDS-PAGE

and Western blotting with TTR antibody). We found that both BiP and TTR D18G

were abundant the soluble fractions and that for the TTR D18G fractions expressed

alone, the large oligomers dominated and also a significant amount of small oligomers

were present. For the TTR D18G protein co-expressed with BiP, the distribution of

molecular weights was evenly spread throughout the glycerol gradient. Hence, BiP

appear to shift large oligomers towards intermediate size oligomers.

Figure 24) Distribution of oligomers in cells expressing TTR D18G alone or TTR D18G and BiP.

Input

TTR D18G

Molecular weightlow high

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4TTR-D18G BiP TTR-D18G

Frac

tion

Input

TTR D18G

Molecular weightlow high

0

0,05

0,1

0,15

0,2

0,25

0,3

0,35

0,4TTR-D18G BiP TTR-D18G

Frac

tion

52

BiP can protect cells from TTR D18G cytotoxicity

In paper II, we worked with neuroblastoma cells to investigate the cells viability after

exposure to oligomeric species of TTR. We found that early oligomers were toxic to

the cells, which died in an apoptotic like manner. In paper IV, we wanted to study the

influence of BiP on toxic oligomeric TTR species and used the used the same

principles as in paper II but exposed the cells to TTR D18G instead of TTR wt

oligomers and mixed cold TTR D18G with cell medium and challenged NBC cells.

To compare with, we used the cold BiP:TTR D18G complex and cold TTR wt. Phase

contrast pictures of human NBC showed that cells exposed to TTR D18G were dead

after 48 hours, whereas cells exposed to TTR D18G in complex with BiP were more

viable. Analysis of DNA fragmentation inside the cells nuclei which is a hallmark of

apoptosis revealed that TTR D18G induced apoptosis which was not the case for TTR

D18G in complex with BiP. From this we concluded that BiP could protect the cells

from TTR D18G cytotoxicity.

Figure 25) BiP can rescue cells from dying in apoptosis. Phase contrast pictures of human NBC

exposed to cold TTR wt (wt), TTR D18G (D18G) or TTR D18G in complex with BiP (D18G + BiP)

showed that cells treated with TTR D18G were dead after 48 hours (decreased cell size, modified

morphology and sparse population) whereas TTR D18G in complex with BiP were more viable (more

dense and normal morphology).

53

BiP is found in TTR D18G aggregates in patient tissue

Tissue sections from the cerebellum of a TTR D18G patient was analyzed using an

amyloidotropic luminescent conjugated polymer (LCP) which emits light of a specific

signature when bound to amyloid fibrils [94]. BiP antibody was used to detect the

localization of BiP in the tissue sections and TTR antibodies were used to detect

where the TTR was localized. Triple staining using LCP, anti-BiP and anti-TTR

antibodies showed that BiP co-localized with the TTR containing amyloid in the brain

(Figure 26).

Figure 26) BiP co-localizes with TTR containing amyloid in the brain. Amyloid staining (LCP),

TTR, and BiP are all co-localized, however there are patters where the individual signals dominates,

indicating that the amyloid composition is layered within the deposits (indicated with arrows, right

panel).

TTR

BiPLCP

mergeTTR

BiPLCP

merge

54

55

Conclusions From the studies in this thesis, it could be concluded that:

Fluorescence spectroscopy used intelligently (various molecular probes and time

resolves techniques) is a very powerful tool to assay formation of amyloid and

prefibrillar oligomers.

Formation of TTR oligomers during acidic conditions from the A-state occurs very

fast (within minutes) after the process has been initiated. The formation of amyloid

like fibrils occurs via oligomeric intermediates.

Oligomeric, intermediate species in the TTR aggregation pathway are toxic to

neuroblastoma cells and cause apoptosis. Mature fibrils are less toxic. Cold stored

native, tetrameric TTR is also cytotoxic suggesting an additional pathway for a labile

tetramer or monomer to be toxic.

The ER chaperone BiP selectively captures the pathogenic, misfolding prone mutant

of TTR; TTR D18G.

The binding site for BiP in TTR is the part in TTR that is involved in formation of the

tetrameric structure and is possibly an elongation site in fibrils, comprising residues

88-103. Hence, BiP maintains TTR D18G in a soluble oligomeric form which should

be a protection mechanism against oligomer toxicity.

56

BiP co-aggregates with TTR D18G. The larger the TTR containing aggregates are, the

fewer BiP are in the complex. We ascribe this collection role for BiP as a molecular

shepherd.

The degradation process of TTR D18G is slowed down in the presence of BiP.

BiP can escape the ER in complexes with TTR D18G and accumulate as extracellular

amyloid in human brain.

57

References

1. Venter, J.C., et al., The sequence of the human genome. Science, 2001. 291(5507): p. 1304-51.

2. Watson, J.D. and F.H. Crick, Molecular structure of nucleic acids; a structure for deoxyribose nucleic acid. Nature, 1953. 171(4356): p. 737-8.

3. Mizuno, T., M.Y. Chou, and M. Inouye, A unique mechanism regulating gene expression: translational inhibition by a complementary RNA transcript (micRNA). Proc Natl Acad Sci U S A, 1984. 81(7): p. 1966-70.

4. Rocheleau, C.E., et al., Wnt signaling and an APC-related gene specify endoderm in early C. elegans embryos. Cell, 1997. 90(4): p. 707-16.

5. Fire, A., et al., Potent and specific genetic interference by double-stranded RNA in Caenorhabditis elegans. Nature, 1998. 391(6669): p. 806-11.

6. Anfinsen, C.B., Principles that govern the folding of protein chains. Science, 1973. 181(96): p. 223-30.

7. Jackson, S.E. and A.R. Fersht, Folding of chymotrypsin inhibitor 2. 1. Evidence for a two-state transition. Biochemistry, 1991. 30(43): p. 10428-35.

8. Dobson, C.M., Getting out of shape. Nature, 2002. 418(6899): p. 729-30. 9. Ferri, C.P., et al., Global prevalence of dementia: a Delphi consensus study.

Lancet, 2005. 366(9503): p. 2112-7. 10. Prusiner, S.B., Molecular biology and genetics of prion diseases. Philos Trans

R Soc Lond B Biol Sci, 1994. 343(1306): p. 447-63. 11. Prusiner, S.B., The prion diseases. Brain Pathol, 1998. 8(3): p. 499-513. 12. Estebe, J.P., [Anesthesia and non-conventional transmissible agents (or prion

diseases)]. Ann Fr Anesth Reanim, 1997. 16(8): p. 955-63. 13. Horwich, A.L. and J.S. Weissman, Deadly conformations--protein misfolding

in prion disease. Cell, 1997. 89(4): p. 499-510. 14. Buxbaum, J., The genetics of the amyloidoses: interactions with immunity and

inflammation. Genes Immun, 2006. 7(6): p. 439-49. 15. Ingbar, S.H., Pre-albumin: a thyroxinebinding protein of human plasma.

Endocrinology, 1958. 63(2): p. 256-9. 16. Kanai, M., A. Raz, and D.S. Goodman, Retinol-binding protein: the transport

protein for vitamin A in human plasma. J Clin Invest, 1968. 47(9): p. 2025-44. 17. Blake, C.C., et al., Structure of prealbumin: secondary, tertiary and

quaternary interactions determined by Fourier refinement at 1.8 A. J Mol Biol, 1978. 121(3): p. 339-56.

18. Nomenclature Committee of IUB (NC-IUB) IUB-IUPAC Joint Commission on Biochemical Nomenclature (JCBN). Newsletter 1981. J Biol Chem, 1981. 256(1): p. 12-4.

19. Hamilton, J.A. and M.D. Benson, Transthyretin: a review from a structural perspective. Cell Mol Life Sci, 2001. 58(10): p. 1491-521.

20. Blake, C.C., et al., Strjcture of human plasma prealbumin at 2-5 A resolution. A preliminary report on the polypeptide chain conformation, quaternary structure and thyroxine binding. J Mol Biol, 1974. 88(1): p. 1-12.

21. Smith, J.W., et al., Thyroid hormones, brain function and cognition: a brief review. Neurosci Biobehav Rev, 2002. 26(1): p. 45-60.

22. Hagen, G.A. and W.J. Elliott, Transport of thyroid hormones in serum and cerebrospinal fluid. J Clin Endocrinol Metab, 1973. 37(3): p. 415-22.

58

23. Hou, X., M.I. Aguilar, and D.H. Small, Transthyretin and familial amyloidotic polyneuropathy. Recent progress in understanding the molecular mechanism of neurodegeneration. Febs J, 2007. 274(7): p. 1637-50.

24. Monaco, H.L., M. Rizzi, and A. Coda, Structure of a Complex of Two Plasma Proteins: Transthyretin and Retinol-Binding Protein Science, 1995. 268(5213): p. 1039-1041.

25. Benson, M.D. and J.C. Kincaid, The molecular biology and clinical features of amyloid neuropathy. Muscle Nerve, 2007. 36(4): p. 411-23.

26. Sousa, M.M., et al., Deposition of transthyretin in early stages of familial amyloidotic polyneuropathy: evidence for toxicity of nonfibrillar aggregates. Am J Pathol, 2001. 159(6): p. 1993-2000.

27. Sousa, M.M., et al., Evidence for early cytotoxic aggregates in transgenic mice for human transthyretin Leu55Pro. Am J Pathol, 2002. 161(5): p. 1935-48.

28. Costa, P.P., A.S. Figueira, and F.R. Bravo, Amyloid fibril protein related to prealbumin in familial amyloidotic polyneuropathy. Proc Natl Acad Sci U S A, 1978. 75(9): p. 4499-503.

29. Andrade, C., A peculiar form of peripheral neuropathy; familiar atypical generalized amyloidosis with special involvement of the peripheral nerves. Brain, 1952. 75(3): p. 408-27.

30. Saraiva, M.J., et al., Amyloid fibril protein in familial amyloidotic polyneuropathy, Portuguese type. Definition of molecular abnormality in transthyretin (prealbumin). J Clin Invest, 1984. 74(1): p. 104-19.

31. Joao Saraiva, M., et al., Familial amyloidotic polyneuropathy: protein aggregation in the peripheral nervous system. J Mol Neurosci, 2004. 23(1-2): p. 35-40.

32. Suhr, O.B., et al., Hereditary transthyretin amyloidosis from a Scandinavian perspective. J Intern Med, 2003. 254(3): p. 225-35.

33. Thomas, P.J., B.H. Qu, and P.L. Pedersen, Defective protein folding as a basis of human disease. Trends Biochem Sci, 1995. 20(11): p. 456-9.

34. Ii, S., et al., Two-tiered DNA-based diagnosis of transthyretin amyloidosis reveals two novel point mutations. Neurology, 1991. 41(6): p. 893-8.

35. Gertz, M.A., R.A. Kyle, and S.N. Thibodeau, Familial amyloidosis: a study of 52 North American-born patients examined during a 30-year period. Mayo Clin Proc, 1992. 67(5): p. 428-40.

36. Gustavsson, A., et al., Amyloid fibril composition and transthyretin gene structure in senile systemic amyloidosis. Lab Invest, 1995. 73(5): p. 703-8.

37. Cornwell, G.G., 3rd, et al., Frequency and distribution of senile cardiovascular amyloid. A clinicopathologic correlation. Am J Med, 1983. 75(4): p. 618-23.

38. Westermark, P., et al., Fibril in senile systemic amyloidosis is derived from normal transthyretin. Proc Natl Acad Sci U S A, 1990. 87(7): p. 2843-5.

39. Pitkanen, P., P. Westermark, and G.G. Cornwell, 3rd, Senile systemic amyloidosis. Am J Pathol, 1984. 117(3): p. 391-9.

40. Cornwell, G.G., 3rd, et al., Evidence that the amyloid fibril protein in senile systemic amyloidosis is derived from normal prealbumin. Biochem Biophys Res Commun, 1988. 154(2): p. 648-53.

41. Felding, P., et al., Prealbumin in Swedish patients with senile systemic amyloidosis and familial amyloidotic polyneuropathy. Scand J Immunol, 1985. 21(2): p. 133-40.

59

42. Reixach, N., et al., Tissue damage in the amyloidoses: Transthyretin monomers and nonnative oligomers are the major cytotoxic species in tissue culture. Proc Natl Acad Sci U S A, 2004. 101(9): p. 2817-22.

43. Sekijima, Y., et al., The biological and chemical basis for tissue-selective amyloid disease. Cell, 2005. 121(1): p. 73-85.

44. Foss, T.R., R.L. Wiseman, and J.W. Kelly, The pathway by which the tetrameric protein transthyretin dissociates. Biochemistry, 2005. 44(47): p. 15525-33.

45. Andersson, K., et al., Only amyloidogenic intermediates of transthyretin induce apoptosis. Biochem Biophys Res Commun, 2002. 294(2): p. 309-14.

46. Kelly, J.W., Towards an understanding of amyloidogenesis. Nat Struct Biol, 2002. 9(5): p. 323-5.

47. Bucciantini, M., et al., Inherent toxicity of aggregates implies a common mechanism for protein misfolding diseases. Nature, 2002. 416(6880): p. 507-11.

48. Selkoe, D.J., Folding proteins in fatal ways. Nature, 2003. 426(6968): p. 900-4.

49. Buxbaum, J.N., et al., Transthyretin protects Alzheimer's mice from the behavioral and biochemical effects of Abeta toxicity. Proc Natl Acad Sci U S A, 2008. 105(7): p. 2681-6.

50. Garzuly, F., et al., Familial meningocerebrovascular amyloidosis, Hungarian type, with mutant transthyretin (TTR Asp18Gly). Neurology, 1996. 47(6): p. 1562-7.

51. Vidal, R., et al., Meningocerebrovascular amyloidosis associated with a novel transthyretin mis-sense mutation at codon 18 (TTRD 18G). Am J Pathol, 1996. 148(2): p. 361-6.

52. Yamamoto, K., et al., Familial amyloid polyneuropathy in Taiwan: identification of transthyretin variant (Leu55-->Pro). Muscle Nerve, 1994. 17(6): p. 637-41.

53. Hammarstrom, P., et al., D18G transthyretin is monomeric, aggregation prone, and not detectable in plasma and cerebrospinal fluid: a prescription for central nervous system amyloidosis? Biochemistry, 2003. 42(22): p. 6656-63.

54. Sekijima, Y., et al., Energetic characteristics of the new transthyretin variant A25T may explain its atypical central nervous system pathology. Lab Invest, 2003. 83(3): p. 409-17.

55. Jenne, D.E., et al., A new isoleucine substitution of Val-20 in transthyretin tetramers selectively impairs dimer-dimer contacts and causes systemic amyloidosis. Proc Natl Acad Sci U S A, 1996. 93(13): p. 6302-7.

56. Jiang, X., et al., An engineered transthyretin monomer that is nonamyloidogenic, unless it is partially denatured. Biochemistry, 2001. 40(38): p. 11442-52.

57. Ranlov, I., et al., A Danish kindred with familial amyloid cardiomyopathy revisited: identification of a mutant transthyretin-methionine111 variant in serum from patients and carriers. Am J Med, 1992. 93(1): p. 3-8.

58. Laskey, R.A., et al., Nucleosomes are assembled by an acidic protein which binds histones and transfers them to DNA. Nature, 1978. 275(5679): p. 416-20.

59. Ellis, R.J., Do molecular chaperones have to be proteins? Biochem Biophys Res Commun, 1997. 238(3): p. 687-92.

60

60. Tissieres, A., H.K. Mitchell, and U.M. Tracy, Protein synthesis in salivary glands of Drosophila melanogaster: Relation to chromosome puffs. J Mol Biol, 1974. 85(3): p. 389-98.

61. Lanneau, D., et al., Heat shock proteins: essential proteins for apoptosis regulation. J Cell Mol Med, 2008.

62. Munro, S. and H.R. Pelham, An Hsp70-like protein in the ER: identity with the 78 kd glucose-regulated protein and immunoglobulin heavy chain binding protein. Cell, 1986. 46(2): p. 291-300.

63. Chevalier, M., et al., Substrate binding induces depolymerization of the C-terminal peptide binding domain of murine GRP78/BiP. J Biol Chem, 1998. 273(41): p. 26827-35.

64. Gething, M.J., Role and regulation of the ER chaperone BiP. Semin Cell Dev Biol, 1999. 10(5): p. 465-72.

65. Blond-Elguindi, S., et al., Affinity panning of a library of peptides displayed on bacteriophages reveals the binding specificity of BiP. Cell, 1993. 75(4): p. 717-28.

66. Fourie, A.M., et al., HSP70 binding sites in the tumor suppressor protein p53. J Biol Chem, 1997. 272(31): p. 19471-9.

67. Fourie, A.M., J.F. Sambrook, and M.J. Gething, Common and divergent peptide binding specificities of hsp70 molecular chaperones. J Biol Chem, 1994. 269(48): p. 30470-8.

68. Knarr, G., et al., BiP binding sequences in antibodies. J Biol Chem, 1995. 270(46): p. 27589-94.

69. Knarr, G., et al., BiP-binding sequences in HIV gp160. Implications for the binding specificity of bip. J Biol Chem, 1999. 274(42): p. 29850-7.

70. Rudiger, S., et al., Substrate specificity of the DnaK chaperone determined by screening cellulose-bound peptide libraries. Embo J, 1997. 16(7): p. 1501-7.

71. Kyte, J. and R.F. Doolittle, A simple method for displaying the hydropathic character of a protein. J Mol Biol, 1982. 157(1): p. 105-32.

72. Weitzmann, A., et al., The heat shock protein 70 molecular chaperone network in the pancreatic endoplasmic reticulum - a quantitative approach. Febs J, 2007. 274(19): p. 5175-87.

73. Greene, M.K., K. Maskos, and S.J. Landry, Role of the J-domain in the cooperation of Hsp40 with Hsp70. Proc Natl Acad Sci U S A, 1998. 95(11): p. 6108-13.

74. Crowley, K.S., et al., Secretory proteins move through the endoplasmic reticulum membrane via an aqueous, gated pore. Cell, 1994. 78(3): p. 461-71.

75. Hamman, B.D., L.M. Hendershot, and A.E. Johnson, BiP maintains the permeability barrier of the ER membrane by sealing the lumenal end of the translocon pore before and early in translocation. Cell, 1998. 92(6): p. 747-58.

76. Sitia, R. and I. Braakman, Quality control in the endoplasmic reticulum protein factory. Nature, 2003. 426(6968): p. 891-4.

77. Roth, J., et al., Protein quality control: the who's who, the where's and therapeutic escapes. Histochem Cell Biol, 2008. 129(2): p. 163-77.

78. Romisch, K., Endoplasmic reticulum-associated degradation. Annu Rev Cell Dev Biol, 2005. 21: p. 435-56.

79. Yoshida, H., ER stress and diseases. Febs J, 2007. 274(3): p. 630-58. 80. Bernales, S., F.R. Papa, and P. Walter, Intracellular signaling by the unfolded

protein response. Annu Rev Cell Dev Biol, 2006. 22: p. 487-508.

61

81. Harding, H.P., et al., Transcriptional and translational control in the Mammalian unfolded protein response. Annu Rev Cell Dev Biol, 2002. 18: p. 575-99.

82. Ron, D. and P. Walter, Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev Mol Cell Biol, 2007. 8(7): p. 519-29.

83. van Anken, E. and I. Braakman, Endoplasmic reticulum stress and the making of a professional secretory cell. Crit Rev Biochem Mol Biol, 2005. 40(5): p. 269-83.

84. Shen, J., et al., ER stress regulation of ATF6 localization by dissociation of BiP/GRP78 binding and unmasking of Golgi localization signals. Dev Cell, 2002. 3(1): p. 99-111.

85. Harding, H.P., Y. Zhang, and D. Ron, Protein translation and folding are coupled by an endoplasmic-reticulum-resident kinase. Nature, 1999. 397(6716): p. 271-4.

86. Kerr, J.F., A.H. Wyllie, and A.R. Currie, Apoptosis: a basic biological phenomenon with wide-ranging implications in tissue kinetics. Br J Cancer, 1972. 26(4): p. 239-57.

87. Nylandsted, J., et al., Selective depletion of heat shock protein 70 (Hsp70) activates a tumor-specific death program that is independent of caspases and bypasses Bcl-2. Proc Natl Acad Sci U S A, 2000. 97(14): p. 7871-6.

88. Yuan, J., et al., The C. elegans cell death gene ced-3 encodes a protein similar to mammalian interleukin-1 beta-converting enzyme. Cell, 1993. 75(4): p. 641-52.

89. Thornberry, N.A. and Y. Lazebnik, Caspases: enemies within. Science, 1998. 281(5381): p. 1312-6.

90. Kothakota, S., et al., Caspase-3-generated fragment of gelsolin: effector of morphological change in apoptosis. Science, 1997. 278(5336): p. 294-8.

91. Lai, Z., W. Colon, and J.W. Kelly, The acid-mediated denaturation pathway of transthyretin yields a conformational intermediate that can self-assemble into amyloid. Biochemistry, 1996. 35(20): p. 6470-82.

92. Naiki, H., et al., Fluorometric determination of amyloid fibrils in vitro using the fluorescent dye, thioflavin T1. Anal Biochem, 1989. 177(2): p. 244-9.

93. Olofsson, A., et al., Probing solvent accessibility of transthyretin amyloid by solution NMR spectroscopy. J Biol Chem, 2004. 279(7): p. 5699-707.

94. Nilsson, K.P., et al., Conjugated polyelectrolytes--conformation-sensitive optical probes for staining and characterization of amyloid deposits. Chembiochem, 2006. 7(7): p. 1096-104.