Glycosome Biogenesis in Trypanosomes : Identification And

Home , Flagellum, Glycosome, PEX1, PEX14, PEX16, PEX6

Glycosome biogenesis in trypanosomes ‐ Identification and characterization of PEX16 and inhibitors of PEX5‐PEX14 interaction

Dissertation to obtain the degree Doctor Philosophiae (Doctor of Philosophy, PhD) At the Faculty of Biology and Biotechnology Ruhr‐University Bochum

International Graduate School of Biosciences Ruhr‐University Bochum Institute of Biochemistry and Pathobiochemistry Department of Systems Biochemistry Faculty of Medicine

Submitted by Vishal C. Kalel M.Sc. Biochemistry

from Walchandnagar, India

Bochum October 2015

First supervisor: Prof. Dr. Ralf Erdmann Second supervisor: PD Dr. Mathias Lübben

Biogenese von Glykosomen in Trypanosomen ‐ Identifizierung und Charakterisierung von PEX16 und Inhibitoren der PEX5‐PEX14‐Interaktion

Dissertation zur Erlangung des Grades eines Doktors der Naturwissenschaften der Fakultät für Biologie und Biotechnologie

an der Internationalen Graduiertenschule Biowissenschaften der Ruhr‐Universität Bochum

angefertigt im Institut für Biochemie und Pathobiochemie Abteilung für Systembiochemie der Fakultät für Medizin

vorgelegt von Vishal C. Kalel M.Sc. Biochemie

aus Walchandnagar, Indien

Bochum Oktober 2015

Referent: Prof. Dr. Ralf Erdmann Korreferent: PD Dr. Mathias Lübben

ERKLÄRUNG

Hiermit erkläre ich, dass ich die Arbeit selbstständig verfasst und bei keiner anderen Fakultät eingereicht und dass ich keine anderen als die angegeben Hilfsmittel verwendet habe. Es handelt sich bei der heute von mir eingereichten Dissertation um sechs in Wort und Bild völlig übereinstimmende Exemplare. Weiterhin erkläre ich, dass digitale Abbildungen nur die originalen Daten enthalten und in keinem Fall inhaltsverändernde Bildbearbeitung vorgenommen wurde.

Bochum, den

(Unterschrift)

III

Dedicated to...

The Parasitologists and The Organziations

who are helping to fight the Neglected Diseases.

With special mention of following scientists after whom the Trypanosomatid parasite species are named

David Bruce William Leishman Charles Donovan Ostwaldo Cruz

Table of contents

I INTRODUCTION ...... 1 1. Trypanosomatid parasites and the diseases ...... 1 2. Glycosomes ...... 3 3. Peroxisome and glycosome biogenesis ...... 5 3.1 Matrix protein import ...... 5 3.1.1 PEX5‐PEX14 interaction ...... 9 3.2 Membrane protein import ...... 10 3.3 Inheritance and degradation of peroxisomes ...... 14 3.4 Mechanisms of peroxisome biogenesis ...... 15 3.5 Comparison of PEX proteins between human and trypanosomes ...... 17 4. Aims and objectives ...... 19

II RESULTS ...... 21

Chapter I: Identification and functional characterization of Trypanosoma brucei peroxin 16 ...... 22

a) Publication ...... 23 b) Supplementary (unpublished) ...... 37 1. Supplementary materials and methods ...... 38 1.1 Cells and growth conditions...... 38 1.2 Plasmids construction ...... 38 1.3 Microscopy ...... 39 1.4 Yeast two‐hybrid ...... 40 2. Supplementary results ...... 41 2.1 Trypanosomatid PEX16 localization in human cells ...... 41 2.2 Parasite PEX16 proteins do not rescue human PEX16 defects ...... 42 2.3 Trypanosomatid PEX16 localization in yeast ...... 44 2.4 Identification of Trypanosoma PEX16 binding partners ...... 45 2.5 Overexpression of GFP‐PEX16 leads to glycosome aggregation ..... 49

Chapter II: Identification and characterization of inhibitors of PEX5‐PEX14 interaction ...... 51 1. Materials and methods ...... 52 1.1 Anti‐trypanosomal assay ...... 52 1.2 Human cell cytotoxicity assay ...... 52 1.3 Glucose dependence of inhibitor toxicity ...... 53 1.4 ATP Assay and microscopy ...... 53 2. Results ...... 54 2.1 in vitro and bioactivity assays of PEX5‐PEX14 inhibitors (schemes) ..... 54 2.2 Anti‐trypanosomal activity and human cell cytotoxicity of inhibitors .. 57 2.3 Correlation between in vitro and in vivo activities ...... 60 2.4 Inhibitors are also active on procyclic form trypanosomes ...... 61 2.5 Glucose is toxic in presence of inhibitors ...... 62 2.6 Cellular ATP levels are depleted by the inhibitor ...... 64 2.7 Inhibitor disrupts the glycosomal protein import ...... 65 2.8 Inhibitor reduces parasite load in mouse model of trypanosomiasis ... 66

III DISCUSSION ...... 68

IV SUMMARY ...... 89 IV ZUSAMMENFASSUNG ...... 91

V REFERENCES ...... 93

VI APPENDIX ...... 103 1. Abbreviation used in the study ...... 103 2. Publications ...... 104 2.1 Complete list of publications ...... 104 2.2 Conference contributions ...... 105 2. Curriculum vitae ...... 106 3. Acknowledgments ...... 107 4. Publication (Review) ‐ Nagotu and Kalel et al., BBA Mol Bas. Dis. (2012) ...... 109

INTRODUCTION

I INTRODUCTION

1. Trypanosomatid parasites and the diseases

Trypanosomatidae family parasites Trypanosoma and Leishmania are the causative agents of infectious diseases termed Trypanosomiasis and Leishmaniasis respectively. These along with 15 other infectious diseases are regarded as Neglected Tropical Diseases (NTDs) which collectively affect 1 billion people worldwide, mostly in poor and developing countries (WHO NTD report 2015). The protozoan parasites Trypanosoma and Leishmania not only infect humans but also affect livestock animals like cattle leading to economic loss of over 1 billion US$ annually (Kristjanson et al., 1999). Trypanosoma brucei rhodesiense (east African form) and T. b. gambiense (west African form) are transmitted by the bite of Tsetse fly (Glossina species) causing Human African Trypanosomiasis (HAT). During the first stage of the disease, trypanosomes reside and proliferate in bloodstream of human host. During second stage, the parasites invade central nervous system causing disruption of normal sleep cycle and hence the disease is commonly known as Sleeping Sickness (Mogk et al., 2014). Mainly in South America, T. cruzi is the infectious agent of Chagas’ disease (also referred to as Human American Trypanosomiasis). Bloodsucking triatomine insects also known as kissing bugs transmit T. cruzi parasites to humans through their feces laid at bite wound. Although the symptoms in acute stage are mild and non‐specific, chronic stage Chagas’ cardiomyopathy is life‐threatening (Bern, 2015). Leishmania genus comprises the most diverse pathogenic species causing wide spectrum of clinical manifestations collectively termed as Leishmaniasis (Pace, 2014). Female sandflies from the Phleobotomus and Lutzomyia transmit these parasites. In South Asian sub‐continent, L. donovani is the causative agent of visceral Leishmaniasis commonly known as Kala‐Azar which is lethal if untreated. Cutaneous and muco‐cutaneous Leishmaniasis mainly occurs in South America.

Among Trypanosomatids, T. brucei is the most extensively studied model organism. Trypanosoma brucei alternates between two life cycle stages, procyclic form (PCF) in insect stage and bloodstream form (BSF) in mammalian host stage. Intermediate transient stages adapt the parasites for switching of the host (Fig. I1). During the bite of Tsetse fly, the infective metacyclic trypomastigotes in the insect saliva are transmitted into the human

INTRODUCTION

host. The metacyclic trypomastigotes transform into bloodstream trypomastigote which reside extracellularly and proliferate in host bloodstream trypomastigotes.

Fig. I1 ‐ Life cycle of T. brucei. T. brucei alternates between two life cycles stages, procyclic form in insect (Tsetse fly) stage, and bloodstream form in human stage. Metacyclic trypomastigotes are transferred to human host by Tsetse fly bite (1), transformed to bloodstream form trypomastigotes (2) and proliferate in blood (3). Infective form of bloodstream form trypmastigotes (4), short stumpy form is taken up by Tsetse fly during blood meal (5), transform into procyclic stage in insect midgut (6). Epimastigotes which leave midgut (7) transform into metacyclic stage in salivary gland (8) to complete a round of life cycle. (From Blum et al. 2008)

A metabolic signal triggers differentiation of BSF trypanosomes differentiate into short‐ stumpy form (Mony et al., 2014), which is taken up by the Tsetse flies during bite. In the insect midgut, the parasites differentiate into procyclic trypomastigote (PCF). Further they develop into epimastigotes which leave the insect midgut. The terminal differentiation into infective form metacyclic trypomastigote occurs in the salivary gland of insects, thus completing the life cycle (Blum et al., 2008).

INTRODUCTION

There are no vaccines possible against T. brucei due to its unique feature of antigenic variation which involves switching of coat proteins. Currently used treatments suffer severe drawbacks such as low efficacy against different strains, difficulty to administer the drugs and emergence of drug resistance. There is little interest of pharmaceutical industry since the disease is mainly affecting poor and developing countries. Suramin, pentamidine and Melarsoprol were discovered before 1950s, but are still used to treat HAT despite their severe limitations as mentioned above. Therefore there is an urgent need of new drugs and identification of drug targets.

2. Glycosomes

Fig. I2 ‐ Microbodies: Peroxisomes, glyoxysomes and glycosomes. Hansenula polymorpha cells grown in glucose containing medium have very few peroxisomes (P) (left top) while in methanol containing medium, peroxisome proliferation is induced where they occupy more than 80% of cell volume (left bottom) (Erdmann and Schliebs, 2005). Glycosomes (G) in Leishmania (top right) (Souza 2008). Glyoxysome in germinating plant seed (bottom right) (Lehninger Principles of biochemistry, 2nd edition 1993).

Microbodies were first observed by Rhodin in 1965, as single membrane bound organelles in mouse kidney cells (Fig. I2). Cell fractionation studies pioneered by Christian de Duve demonstrated that the microbodies contain enzymes for peroxide metabolism like catalase and oxidase. Therefore these organelles were named as peroxisomes. Peroxisomes are

INTRODUCTION

typically involved in functions such as fatty acid metabolism and peroxide detoxification (Reviewed in Wanders, 2014). Glyoxysomes in plants perform glyoxylate cycle which converts fats into carbohydrates during seed development. The peculiar feature of Trypanosomatid parasites is that the first 7 enzymes of glycolysis are compartmentalized inside peroxisome, and therefore termed glycosome. Glycosomes are also involved in other essential metabolic pathways such as pentose phosphate pathway, gluconeogenesis, purine salvage, pyrimidine biosynthesis (Michels et al., 2005; Michels et al., 2006; Szoor et al., 2014).

Fig. I3 ‐ Glycosome ‐ Compartmentalized glycolysis. First seven enzymes of glycolysis are compartmentalized inside glycosomes. For every glucose molecule entering glycosome, 2 ATPs consumed by hexokinase (HK) and phosphofructokinase (PFK), are replenished by 2 ATP generated by phosphoglycerate kinase (PGK). Net ATP synthesis occurs in the cytosol by enolase (Eno) and pyruvate kinase (PK). End product of glycolysis pyruvate is not further metabolized instead secreted outside the cell. (Modified from Coley et al., 2011)

During mammalian stage of Trypanosomiasis, mitochondria are highly repressed while the glycolysis is the sole source of ATP for the parasite. Though glycosomes compartmentalize first seven enzymes of the glycolytic pathway inside glycosomes, there is no net ATP

INTRODUCTION

generated inside glycosomes (Fig. I3). The last two enzymes of the glycolytic pathway localize to cytosol which provide net 2 ATP molecules per molecule of glucose metabolized. The end product pyruvate is secreted outside. Trypanosomatid hexokinase and phosphofructokinase lack feedback inhibition therefore the unique glycosomal compartmentation of these enzymes exerts spatial control on glycolysis (Kessler and Parsons, 2005; Hanstra et al., 2008). Disruption of glycosome biogenesis is deleterious for the parasite since the enzymes mislocalised to the cytosol, deplete cellular ATP levels by unregulated phosphorylation of glucose and intermediates. At the same time there is accumulation of toxic metabolites leading to parasite cell death. Defects in glycosome biogenesis thus make glucose toxic to trypanosomes (Furuya et al., 2002; Bakker et al. 2002). Therefore disruption of glycosome biogenesis is an attractive drug target. Although the mechanism and the proteins involved in glycosome and peroxisome biogenesis are conserved (also see Table 1), there are unique feature of glycosome biogenesis which will be introduced in next section in the light of peroxisome biogenesis in other organisms.

3. Peroxisome and glycosome biogenesis

3.1 Matrix protein import Peroxisomes do not contain DNA, and hence peroxisomal matrix proteins encoded by nuclear genome have to be imported post‐translationally from the cytosol. Two types of peroxisome targeting signals (PTS) PTS1 or PTS2 are involved in correct targeting of the peroxisomal matrix proteins (Fig. I4). Unique feature of peroxisomal protein import is that peroxisomes can import folded even oligomeric proteins.

PTS1 is C‐terminal tripeptide signal with consensus sequence SKL and variants thereof (Gould et al., 1989; Gould et al., 1990). On the other hand, PTS2 is localized towards N‐

terminus with degenerate sequence RLX5HL (de hoop and Ab, 1992, Faber et al., 1995). Glycolysis occurs in the cytosol in all other organisms except in parasites of family Trypanosomatidae which has evolved to acquire peroxisome targeting signals in the first 7 glycolytic enzymes (Fig. I4 B). Very few proteins use PTS2 signal in yeast and metazoans, but in plants, PTS2 signal is found in one third of peroxisomal proteins (Reumann, 2004). Similar

INTRODUCTION

to plants, trypanosomes appear to use PTS2 signal more frequently for e.g. glycolytic enzymes hexokinase and fructose 1‐6 bisphosphate aldolase (Opperdoes and Szikora, 2006).

Fig. I4 ‐ Peroxisomal Targeting Signals (PTS) in matrix proteins. A. C‐terminal PTS1 or PTS2 near N‐terminus target proteins to peroxisomal matrix (1‐letter Amino acid code) (Subramani, 1998). B. PTS signals in glycolytic enzymes of T. brucei (HK ‐ Hexokinase, PGI ‐ Glucose‐6‐phosphate isomerase, PFK ‐ Phosphofructokinase, ALD ‐ Fructose‐1, 6‐ bisphosphate aldolase, TPI ‐ Triose phosphate isomerase, GAPDH ‐ Glyceraldehyde‐3‐ phosphate dehydrogenase, G3PDH ‐ Glycerol‐3‐phosphate dehydrogenase).

During the peroxisomal import, peroxisomal protease Tysnd1 (Trypsin domain containing 1) in cooperation with peroxisomal Lon protease (PsLon) is involved in removal of PTS2 from thiolase and processing of PTS1 containing proteins involved in peroxisomal β‐oxidation (Kurochkin et al., 2007, Okumoto et al., 2011). Whether PTS2 or PTS1 signals in glycosomal enzymes are processed is currently unknown.

Proteins involved in peroxisomal proteins import are termed peroxins abbreviated as PEX. Until now 33 peroxin proteins have been identified (Fig. I5). (Current convention of nomenclature for peroxin proteins will be used in this thesis as follows ‐ ‘Pex’ for yeast and ‘PEX’ for human, plant and trypanosomes) Proteins which harbor PTS1 are recognized by the soluble import receptor PEX5 in the cytosol (Van der Leji et al., 1993). PTS2 proteins are recognized by WD‐repeat containing protein PEX7 (Braverman et al., 1997) which also requires additional co‐receptors, Pex18 and Pex21 in Saccharomyces cerevisiae (Purdue et al., 1998) and Pex20 in Pichia pastoris (Titrorenko et al., 1998). In plants and mammals PEX5 exists in two isoforms generated by alternative splicing, shorter isoform PEX5S recognizes PTS1 proteins, while longer isoform PEX5L contains PEX7 binding region and hence function

INTRODUCTION

as co‐receptor for PEX7 (Braverman et al., 1998; Lee et al. 2006). Apart from PTS1 and PTS2, proteins like TPI (triose phosphate isomerase, Fig. I4 B) in trypanosomes have internal signals (I‐PTS) which are recognized by the import receptor and targeted to the glycosome (Galland et al., 2010). Some proteins like Cu/Zn superoxide dismutase 1 (SOD1) in mammals (Islinger et al., 2009) and Pnc1 in yeast (Effelsberg et al., 2015) which lack PTS signal do not bind to import receptors directly and instead are imported into peroxisomes via piggy‐ backing onto other PTS containing protein.

Fig. I5 ‐ Composite model of peroxisomal matrix protein import cycle. The import machinery components present in yeast and humans are depicted in different font colors of the PEX number, for e.g. ‘1’ indicates PEX1. (black ‐ components present in both yeast and human, red ‐ components specific to humans, blue ‐ components specific to yeast). PTS ‐ peroxisome targeting signal, Ub ‐ Ubiquitin, UbcH5 and Ubc4 ‐ Ubiquitin conjugating enzymes, USP9X and Ubp15 ‐ Ubiquitin specific proteases. For detailed description see text. (Review published Nagotu and Kalel et al. 2012, Review copy in Appendix)

After recognition of the peroxisomal cargo protein by import receptors in the cytosol, the receptor–cargo complex is directed to the peroxisomal membrane. The docking complex consists of peroxisomal membrane proteins PEX14 and PEX13 (Stein et al., 2002). So far, Trypanosomatid parasites are the only known organisms which encode two very different PEX13 isoforms, PEX13.1 (Verplaetse et al., 2009) and PEX13.2 (Brennand et al., 2012) (Table 1). Additional striking feature of Trypanosomatid PEX13.1 is the presence of

INTRODUCTION

conserved C‐terminal PTS1 signal tripeptide TKL which is exposed to the cytosol (Verplaetse et al., 2012). PEX13 including Trypanosome PEX13.1 typically contain SH3‐domain. On the other hand, such SH3 domain is completely absent in TbPEX13.2 (Brennand et al., 2012). Despite the differences in domain architecture, both TbPEX13.1 and TbPEX13.2 are part of the docking complex (Verplaetse et al., 2012). Another constituent of the docking complex in yeast is Pex17 which is absent in human, plants and trypanosomes (Huhse et al., 1998).

Assembly of the cargo bound Pex5 with the docking complex results into the formation of a multimeric transient pore (Meinecke et al., 2010). The highly dynamic pore mainly consists of Pex5 and Pex14, where Pex5 behaves like an integral membrane protein. The detailed architecture and components of this pore still remains elusive. Dissociation of the cargo‐ receptor complex and translocation of the cargo into the peroxisomal lumen occurs in an unknown manner. The process likely involves Pex8 in yeast which is peripheral membrane protein containing both PTS1 and PTS2 (Agne et al., 2003).

After the cargo translocation, receptors PEX5 or PEX7 (along with co‐receptors) are exported back into the cytosol for another round of protein import or degradation. The RING (Really Interesting New Gene)‐complex comprises of ubiquitin ligases PEX2, PEX10 and PEX12 (El Magraoui et al., 2012). Ubiquitin‐conjugating enzyme PEX4 is anchored to the peroxisomal membrane by the anchor protein Pex22 in yeast (Williams et al., 2012). Pex4 or mammalian Pex4‐like isoforms UbcH5 a, b, c are responsible for Pex5 mono‐ubiquitination which serves as a signal for the ATP‐dependent dislocation of Pex5 from the peroxisomal membrane back to the cytosol. The mono‐ubiquitination occurs on the conserved cysteine residue at the N‐terminus of PEX5. Although Trypanosomatid PEX5 also harbors this conserved cysteine, the mono‐ubiquitinated species is resistant to DTT, suggesting that ubiquitination of lysine residue is more likely (Gualdron‐Lopez and Michels, 2013). The dislocation step is mediated by the AAA (ATPase Associated with various cellular Activities) peroxins PEX1 and PEX6 (Platta et al., 2005). PEX1 and PEX6 are anchored to the peroxisomal membrane by Pex15 in yeast (Birchmann et al., 2015), PEX26 (Matsumoto et al., 2004) or APEM9 in plants (Goto et al., 2011). Identity or existence of PEX1‐PEX6 complex anchor protein in Trypanosomatid parasites remains unknown (Table 1). Ubiquitin‐ conjugating enzyme Ubc4 in yeast together with the redundant proteins Ubc5 and Ubc1

INTRODUCTION

lead to the polyubiquitination of Pex5 which primes Pex5 for degradation with 26S proteasome. Thus mono‐ubiquitination of Pex5 serves as signal for receptor recycling step while polyubiquitination exerts quality control for damaged Pex5. During receptor recycling or export step, The AAA peroxin Pex6 interacts Ubiquitin hydrolase Ubp15 in yeast which deubiquitinates mono‐ubiquitinated Pex5 (Debelyy et al., 2011). In human, PEX6 interacts with adaptor protein AWP1 (Miyata et al., 2012) and the deubiquitination of PEX5 is performed by USP9X in the cytosol (Grou et al., 2012).

3.1.1 PEX5‐PEX14 interaction

Fig. I6 ‐ Schematic representation of PEX5 and PEX14. Domains and binding motifs in PEX5 and PEX14 of human (A) and trypanosomes (B) are shown. C‐terminal tetratricopeptide repeat (TPR) domain in PEX5 recognizes the PTS1 cargo proteins (green). Di‐aromatic pentapeptide motifs (WxxxF/Y) in the N‐terminal region of PEX5 interact with PEX14 N‐terminus (red) with high affinity during docking step. 2nd di‐aromatic motif (white bar) in TbPEX5 does not bind TbPEX14‐NT. 3rd TPR motif in TbPEX5 (Light green box) adopts unusual extended conformation.

Knockdown of PEX proteins is lethal to trypanosomes; therefore disruption of glycosome biogenesis by inhibiting PEX protein cascade is an attractive drug target. There are several studies going on for development of inhibitors of Trypanosomal glycolytic enzymes like hexokinase or phosphofructokinase (Hudock et al., 2006; Sharlow et al., 2010; Walsh et al., 2011). However there are no reports on small‐molecule compounds inhibiting glycosome biogenesis. Docking of import receptor PEX5 at glycosomal membrane PEX14 is the key step

INTRODUCTION

which allows import of both PTS1 and PTS2 containing glycosomal enzymes into the organelle. Therefore in this study, PEX5‐PEX14 interaction was chosen as a target for structure based design of small molecule compounds which inhibit the docking step and will be described later in Section Results ‐ Chapter II.

Functional domains and important interaction motifs in human PEX5 and PEX14 are shown in Fig. I6 A, while those of Trypanosomal counterparts in Fig. I6 B. PEX5 contains seven helix‐rich tetratricopeptide repeats (TPR) in C‐terminal region. They form a domain which binds to the PTS1 signal in its cargo proteins. Third TPR motif in Trypanosomal PEX5 adopts a unique conformation different from any known TPR motif (shown in light green in TbPEX5 in Fig. I6 B) (Kumar et al., 2001). N‐terminal regions of PEX5 contain one or more PEX14‐ binding peptides with consensus sequence WXXX(F/Y) (Saidowsky et al., 2001). Human PEX5 contains 7 di‐aromatic pentapeptide motifs and a recently identified new non‐canonical motif LVAEF (Neuhaus et al., 2014). On the other hand, only three such di‐aromatic motifs occur in Trypanosoma PEX5 (Choe et al., 2003). These di‐aromatic motifs form the high affinity binding sites that interact with the N‐terminal domain of PEX14 (Schliebs et al., 1999). 3D NMR structure of a human PEX5 peptide containing one di‐aromatic motif bound to the N‐terminus of PEX14 has been solved (Neufeld et al., 2009). Trypanosomal PEX5 and PEX14 show low degree of sequence conservation as compared to the human counterparts (~20% identity, ~40% similarity Table 1). Homology modelling of the Trypanosoma PEX5 peptide in complex with N‐terminus of PEX14 (based on human complex Neufeld et al., 2009) revealed several differences in the binding site (Personal communication Prof. Michael Sattler, Munich) which enabled us the successful design of specific inhibitors of the TbPEX5‐TbPEX14 interaction which do not affect human PEX5‐PEX14 interaction (Described in Results Section chapter II).

3.2 Membrane protein import Peroxisomal membrane protein (PMP) import involves three proteins PEX19, PEX3 and PEX16 (Fig. I7). PEX19 is the predominantly cytosolic receptor which acts as chaperone for newly synthesized membrane proteins in the cytosol and targets them to the peroxisomes (Class I PMPs) (Sacksteder et al., 2000; Jones et al., 2004). Currently PEX3 is the only known

INTRODUCTION

PMP which is targeted to peroxisomes independent of PEX19 (Class II PMP) (Jones et al., 2004). Some PMPs are targeted to peroxisome via ER (introduced later in Section 3.4). At its C‐terminus, PEX19 contains a rigid domain that binds to the cargo PMPs which may harbor multiple mPTS (membrane Peroxisome Targeting Signal). Unlike PTS1 or PTS2 in peroxisomal matrix proteins, mPTS in PMPs is less clearly defined at sequence level (Jones 2001). PEX19 binding motifs are conserved in membrane proteins destined for peroxisomes in mammals and glycosomes in trypanosomes (Saveria et al., 2007). Human, plant as well as yeast Pex19 proteins contain a conserved C‐terminal CAAX motif which is the site for farnesylation. Farnesylation is shown to be critical for the function of yeast PEX19 (Rucktaschel et al. 2009). However PEX19 in all Trypanosomatid species lack such CAAX motif (Banerjee et al., 2005).

Fig. I7 ‐ Peroxisomal membrane protein (PMP) import. Pex19, Pex3 and Pex16 are involved in PMP import. Pex19 is the receptor for PMPs synthesized in the cytosol. Pex3 is anchor for Pex19‐PMP complex at peroxisomal membrane. Pex16 acts as anchor for Pex3. ATP requirement for PMP import is not known (Rucktaschel et al., 2011).

Tail anchored (TA) proteins anchor to membranes by single C‐terminal transmembrane span with C‐terminus freely facing the cytosol. Peroxisomal TA protein Pex15 in yeast is inserted into ER by GET (guided entry of TA proteins) pathway (Schuldiner et al., 2008) and routed to

INTRODUCTION

peroxisomes in PEX19‐dependant manner (Lam et al., 2010). Human TA protein PEX26 is however directly targeted to the peroxisomes by PEX19 independently of GET pathway (Yagita et al., 2013). Both yeast Pex15 and human PEX26 contain PEX19 binding sites at their C‐terminus (Halbach et al., 2006). Knowledge about whether glycosomes contain TA proteins and how they are targeted to glycosomes remains obscure.

At peroxisomal membrane, PEX3 acts as receptor for the PEX19‐PMP complex formed in the cytosol (Fang et al., 2004). The N‐terminal 56 amino acids region in human PEX19 is sufficient for binding to PEX3 and docking at the peroxisomal membrane (Fransen et al., 2005; Matsuzono et al., 2006). PEX3 itself is anchored to the peroxisomal membrane by N‐ terminal 33 amino acids (Soukopova et al., 1999) while the C‐terminus is exposed to the cytosol forming a large helical bundle that can bind to PEX19 with nanomolar affinity (Schmidt et al., 2010). How membrane proteins are then inserted in their correct orientation and whether it requires ATP remains to be understood. After the insertion of PMP, PEX19 is supposed to be recycled back into the cytosol for another round of import. Trypanosomal PEX homologs were identified based on sequence similarity to peroxins from other organism. Bioinformatics attempts to identify trypanosome PEX3 have failed and the identity of crucial PEX19 docking factor in trypanosomes remains a mystery.

PEX16 was first identified in Yarrowia lipolytica (Y. lipolytica) as an intra‐peroxisomal peripheral membrane protein which upon overexpression leads to enlargement of peroxisomes (Eitzen et al., 1997). During the maturation of peroxisomes in Y. lipolytica, heteropentameric complex of acyl‐CoA oxidase (Aox) in the matrix relocates to the inner face of the peroxisomal membrane and specifically associates with PEX16 (Guo et al., 2003). This binding event triggers remodeling of the peroxisomal membrane lipids and leads to the assembly of a protein complex on peroxisome membrane initiating division of the mature peroxisome (Guo et al., 2007). Using Y. lipolytica PEX16 sequence, Human PEX16 was identified based on sequence similarity and functional complementation of the mutant patient cell line (Honsho et al., 1998). Human PEX16 is an integral membrane protein in humans with both its termini facing towards the cytosol (Honsho et al., 2002). Human PEX16 acts as the docking site for PEX3 and as a receptor specifically for PEX3‐PEX19 complex at peroxisomal membrane. PEX16 itself is a class‐I PMP which along with PEX19 requires PEX3

INTRODUCTION

at peroxisomal membrane for its insertion (Fig. I7 and I8). Therefore PEX3 and PEX16 are mutually dependent on each other for their direct targeting to peroxisomes (Matsuzaki and Fujiki, 2008).

Fig. I8 ‐ Mutually dependent direct targeting of PEX3 and PEX16 to peroxisomes. PEX3 acts as anchor for PEX19‐PMP complex including PEX16. PEX16 acts as a specific anchor for PEX19‐PEX3 complex (Matsuzaki and Fujiki, 2008).

In plants, PEX16 was originally identified as a shrunken seed 1 (SSE1) mutant that accumulates starch instead of proteins and lipids in the seeds (Lin et al., 1999). These seeds lack recognizable protein bodies (protein storage vacuoles) and contain reduced number of oil bodies which together occupy over 90% of the cell volume. Protein encoded by SSE1 gene shows sequence similarity (26% identity) to Y. lipolytica PEX16 and functionally complements the Y. lipolytica PEX16 mutant. Lack of normal peroxisomes in sse1 embryos validated that indeed SSE1 gene product is a plant PEX16 homolog (Lin et al., 2004).

Deletion or inactivating mutations in PEX19, PEX3 or PEX16 lead to complete loss of peroxisomes. Peroxisomal matrix proteins are mislocalised to the cytosol while some PMPs are mis‐targeted to other membranes or get degraded. Out of these three peroxins which are essential for peroxisome biogenesis, neither PEX3 nor PEX16 is known from trypanosomes and hence there is severe lack of knowledge about glycosomal membrane protein import. In this study, Trypanosomal PEX16 was identified and demonstrated to be

INTRODUCTION

required for glycosome biogenesis and survival of parasites (described in Results section ‐ Chapter I).

3.3 Inheritance and degradation of peroxisomes

During cell division, organelles are segregated between mother and daughter cells in a regulated manner. Components involved and the mechanism of peroxisome inheritance in yeast has been well studied. Myo2 mediates transport of peroxisomes to daughter cells by binding to peroxisomal receptor protein Inp2 (reviewed in Knoblach and Rachubinski, 2015; Fagarasanu et al., 2010). Inp1 is involved in retaining of peroxisomes into mother cells. Inp1 acts as tether between ER‐resident and peroxisome‐resident Pex3 (Knoblach et al., 2013). Peroxisome inheritance proteins Inp1 and Inp2 are conserved but restricted to yeast. Mammalian cells contain high number of peroxisomes; therefore it is believed that during cell division peroxisomes get partitioned into two cells randomly. Inp1 and Inp2 homologs also do not exist in trypanosomes.

Degradation of excess or damaged peroxisomes occurs by selective autophagy termed pexophagy (Oku and Sakai, 2015; Young et al., 2015). Both PEX3 and PEX14 have been reported to act as peroxisomal anchors for autophagy proteins to mark them for pexophagy (Yamashita et al., 2014; Jiang et al., 2015). In trypanosomes, glycosomes turnover occurs during the transition from bloodstream form to procyclic form stage (Herman et al., 2008). During bloodstream stage, glycolytic enzymes contribute to more than 90% of glycosomal matrix protein content. These glycosomes are degraded by autophagy when trypanosomes shift to insect stage where glucose is limited and amino acids are the major nutrient source. Similar autophagic degradation of glycosomes is observed in Leishmania when promastigote stage parasites containing ~20 glycosomes undergo stage transition into intracellular amastigotes with ~10 glycosomes (Cull et al., 2011). Glycosomal anchors for autophagy proteins and mechanism by which selective number of glycosomes is degraded is not known.

INTRODUCTION

3.4 Mechanisms of Peroxisome biogenesis Peroxisomes can multiply by growth and division of preexisting organelles or they could be formed de novo (Fig. I9). The predominance of either of these two pathways differs between organisms and growth conditions. S. cerevisiae cells grown on glucose contain very few peroxisomes whereas shifting the cells to fatty acid rich media such as oleic acid causes proliferation of peroxisomes where peroxisomes multiply by growth and division by fission (Motley et al., 2007).

Fig. I9 ‐ Routes of peroxisome biogenesis. Peroxisomes can multiply by growth and division of preexisting peroxisomes by fission (bottom half) or form de novo from ER (top half). Fission of mature peroxisomes by Pex11 and Dynamin related proteins forms daughter peroxisomes which import matrix and membrane proteins to mature and divide again. Peroxisomes can arise de novo by Pex19 mediated ATP consuming budding of pre‐ peroxisomal vesicles from ER. (Ma et al., 2011)

Fission of peroxisomes involves Pex11, Dynamin related proteins (DLP in yeast) (Schrader et al., 2012), Mff (Gandre‐babbe and van der Bliek, 2008) and GDAP (Huber et al., 2013). Daughter peroxisome import matrix protein and the mature peroxisomes then divide again. Pex11 contains one amphipathic helix at its N‐terminus conserved from yeast to humans, which mediates elongation or tabulation of peroxisomes in first step of peroxisomal fission (Opalinski et al., 2011). Dynamin Like‐proteins (DLPs, Dnm1 in yeast) which are mitochondrial fission factors also mediate terminal steps of peroxisome fission where Pex11 acts as GTPase activating protein (GAP) for Dnm1 (Williams et al., 2015). Trypanosomes

INTRODUCTION

encode three PEX11 family proteins, PEX11, GIM5A and GIM5B. Overexpression of PEX11 leads to elongation and clustering of glycosomes. RNAi knockdown of PEX11 or GIM5 proteins leads to the reduction in glycosome number and is lethal to trypanosomes (Lorenz et al., 1998, Maier et al., 2001, Voncken et al., 2003). Different Dynamin or dynamin family proteins are involved in functions like scission of endocytic vesicle, fission of mitochondria, chloroplast and peroxisomes. Trypanosomes encode single dynamin family proteins (TbDLP) (Morgan et al., 2004). RNAi knockdown of TbDLP blocks mitochondrial fission and endocytosis (Chanez et al., 2006); however there was no effect apparent on glycosome morphology. Bloodstream form trypanosome contains ~65 glycosomes (Tetley and Vickerman, 1991). How this glycosome population is maintained in trypanosomes which divide every ~7hours and whether glycosomes divide by fission of pre‐existing glycosomes remains completely unknown.

Defects in PEX19, PEX3 or PEX16 lead to complete loss of peroxisomes. In Pex3 mutant yeast cells which lack peroxisomes, reintroduction of Pex3 leads to de novo formation of peroxisomes from ER (Hoepfner et al., 2005). However in proliferating conditions like growth in fatty acid rich media, growth and division of preexisting peroxisomes by fission is the dominant pathway (Motley et al., 2007). In humans, peroxisomes are formed predominantly formed from ER by de novo pathway dependent on PEX16 (Kim et al., 2006). PEX16 recruits PEX3 and other PMPs to ER and constantly targets them to peroxisomes (Aranovich et al., 2014), and this mechanism is also conserved in plants (Hua et al., 2015). Plant PEX16 even at steady state levels, coexists in peroxisomes as well as in ER (Karnik and Trelease, 2005). Mammals have two Sec16 isoforms, longer isoform Sec16A and shorter Sec16B. Sec16B. Sec16B performs specialized and non‐redundant functions than the canonical isoform Sec16A (Budnik et al., 2011). Sec16B is required for the exit of PEX16 from ER (Yonekawa et al., 2011). Cell‐free preperoxisomal vesicle‐budding was reconstituted in vitro to show that trafficking of PMPs Pex3, Pex15 and Pex11 from ER to peroxisomes requires Pex19 and ATP (Lam et al., 2010; Agarwal et al., 2011).

Apart from direct targeting of PMPs to the peroxisomes from cytosol, Pex19, 3 and 16 are also required for de novo pathway. As mentioned earlier, only PEX19 is known in trypanosomes while PEX3 and PEX16 homologs were unknown. Unlike in mammals, yeast

INTRODUCTION

and trypanosomes encode single Sec16 isoform. Yeast Sec16 is similar to longer human Sec16A. However trypanosomal Sec16 resembles shorter human isoform Sec16B (Sealey‐ Cardona et al., 2014) which is required for ER‐exit of PEX16 in mammalian cells during peroxisome biogenesis. In this study trypanosomal PEX16 was identified and evidence was obtained suggesting de novo pathway is likely the pathway for glycosome biogenesis.

3.5 Comparison of PEX proteins between human and trypanosomes

Trypanosomes diverged from the eukaryotic lineage very early during the evolution. Although glycosome biogenesis in trypanosomes involves similar mechanisms as in Peroxisome biogenesis in humans, trypanosomal PEX homologs have very low sequence conservation with human counterparts (~13‐30% identity, ~26‐45% similarity in amino acid sequences) (Table 1). There are also some unique features of trypanosomal PEX proteins no seen in human PEX proteins. Since glycosome biogenesis is essential for trypanosomes, PEX proteins and their interactions are attractive drug targets.

Present/identified in % Identity % Similarity Role Human Peroxin Trypanosomatids to human to human Pex1 Yes 22.3 37.1 Pex2 Yes 20.3 39.8 Pex5 Yes 24.4 41.7 Pex6 Yes 29.5 44.9 Matrix Pex7 Yes 28.2 44.9 protein Pex10 Yes 21.1 38 import Pex12 Yes 22.9 36.5 Yes Pex13 13.1 ‐ 15.1 26.1 ‐ 32.3 (PEX13.1, PEX13.2) Pex14 Yes 19 37.4 Pex26 No ‐‐ ‐‐ Pex11 Yes 15.6 ‐ 19.4 34 ‐ 35.9 (Pex11α, Pex11β) (PEX11, GIM5α,β) Pex3 No ‐‐ ‐‐ Membrane Yes protein Pex16 15.5 32.5 (Identified in this study) import Pex19 Yes 19.1 38.5

Table 1 ‐ Percentage identity and similarity between human and T. brucei PEX proteins. Protein sequences were retrieved from Uniprot or Tritrypdb.org. Percent identity and similarity between human and Trypanosomal PEX homolog was calculated with MatGAT 2.02 (Matrix Global Alignment Tool).

INTRODUCTION

All PEX homologs involved in matrix protein import into glycosomes have been identified except PEX26 which remains unknown. On the other hand, out of three peroxins (PEX3, 16 and 19) involved in membrane protein import or de novo biogenesis from ER, only PEX19 was known from trypanosomes. Therefore there is severe lack in knowledge of how glycosomes import membrane proteins and whether they can from de novo.

To address this, PEX16 candidate protein from Trypanosomatid parasites was identified in this study. The first chapter of this thesis was aimed at demonstrating that this candidate protein is PEX16 homolog. Experiments were performed to assess if the candidate PEX16 localizes to glycosomes. PEX16 mutations in other organisms disrupt peroxisome biogenesis. Therefore RNA interference mediated knockdown of PEX16 candidate in T. brucei was performed and its impact on glycosome biogenesis was assessed by microscopy and biochemical techniques. Since glycosomal compartmentation is shown to be essential for ATP homeostasis and survival of bloodstream form trypanosomes. The cellular ATP levels, motility and survival of trypanosomes upon knockdown of PEX16 candidate were analyzed.

There are no safe and effective drugs currently available for treating human African trypanosomiasis and the pipeline of new lead candidates is scarce. Glycosome biogenesis is an attractive drug target but still there are currently no known inhibitors of glycosome biogenesis. To fill this gap, in collaboration with Helmholtz Zentrum München, we initiated a structure based drug design to identify small molecule compounds which can inhibit PEX5‐ PEX14 interaction and disrupt the glycosome biogenesis. In vitro studies were performed by Helmholtz Zentrum München. Second chapter of thesis was aimed at establishing a medium‐throughput in vivo inhibitor screening assay using cultured trypanosomes. Promising anti‐trypanosomal activity in these assays served as activity based guide for the optimization of the compound structure which were retested using our assay. Similar bioactivity was simultaneously performed in this thesis on human cells to eliminate non‐ specific cytotoxic compounds. Multiple rounds of activity guided in vitro and in vivo assays on trypanosomes and human cells were performed till the most potent glycosome biogenesis inhibitor was obtained with low or no toxicity to human cells. Another objective of the study was to validate on‐target action of the PEX5‐PEX14 inhibitors. So in this study it was investigated whether these inhibitors disrupt the glycosomes in trypanosomes by using different approaches which directly and indirectly indicate defects in glycosome biogenesis.

Aim and Objectives

Aim and objectives

The aim of this thesis is (I) to identify and characterize PEX16 in trypanosomes and (II) to characterize the novel PEX5‐PEX14 inhibitors generated by structure‐based design. Regarding these aims, the following objectives were set:

Chapter I

Objectives ‐

1. Identification of PEX16 homolog in Trypanosomatid parasites. Bioinformatic analysis of the parasite genomes will be performed to identify candidate protein which contains PEX16 domain or has similarity to known PEX16 from other organisms.

2. Investigation of the localization and characteristic features of the Trypanosoma PEX16. Subcellular localization will be investigated by using GFP fusion constructs and microscopy. Biochemical experiments will be performed to test whether it is a peripheral or integral membrane protein.

3. Analysis of the essentiality of PEX16 in trypanosomes. The expression of PEX16 will be downregulated using RNA interference (RNAi). Effect of PEX16 RNAi on trypanosome growth rate and survival will be analyzed.

4. Elucidate the role of PEX16 in glycosome biogenesis in trypanosomes For this purpose, influence of PEX16 RNAi on glycosome biogenesis will be analyzed in detail. Various biochemical and microscopic methods will be employed to assess the effects on the glycosomal protein import and the morphology of glycosomes.

5. Analysis of degree of conservation in targeting signal and function of PEX16. Heterologous expression of parasite PEX16 proteins in human and yeast cells will be done to test whether they can target to peroxisomes. Pex16 defective human cell line will be used to test if parasite PEX16 can rescue the dysfunction.

Aim and Objectives

Chapter II

Objectives ‐

1. Establishment of in vivo drug screening assay using cultured cells. 96‐well plate drug screening assay will be established for determining the anti‐trypanosomal activity as well as human cell cytotoxicity of the Trypanosoma PEX5‐PEX14 inhibitors. The assay will be performed with all inhibitors to identify inhibitors that are highly potent against trypanosomes but show low cytotoxicity to human cells (Biological activity guided inhibitor optimization).

2. Validation of on‐target activity and molecular mechanism of action in trypanosomes. Inhibitors showing highest activity against trypanosomes will be used to verify their on‐ target activity by analyzing the effects on the glycosomal protein import. Molecular mechanism of action of the inhibitor will be studied by investigating the glucose toxicity and changes in cellular ATP levels in presence of inhibitor.

RESULTS ‐ Chapter I

II RESULTS

Chapter I : Identification and functional characterization of Trypanosoma brucei peroxin 16.

a) Publication

Kalel, V. C., Schliebs, W., Erdmann, R. (2015) Identification and functional characterization of Trypanosoma brucei peroxin 16. Biochim Biophys Acta. 2015 Oct; 1853(10 Pt A):2326‐37.

Own contribution to the publication I identified the Trypanosomatid PEX16 homolog using bioinformatics approach. I was trained for 1 month for handling and cultivation of trypanosomes in the lab of Prof. Paul Michels at Research Unit for Tropical Diseases (TROP), "de Duve" Institute, University of Louvain (UCL). Afterwards I established the trypanosome culture facility in the lab of Prof. Ralf Erdmann at Ruhr‐University Bochum. All experiments were performed and figures prepared by me. Electron microscopy in Fig. 7B was performed with assistance of technician.

Planning (P): 80%, Experimental procedure: 95%, Writing of manuscript (M): 75% b) Supplementary (Unpublished)

Chapter II :

Identification and characterization of inhibitors of PEX5‐PEX14 interaction.

RESULTS ‐ Chapter I

Chapter I :

Identification and functional characterization of Trypanosoma brucei peroxin 16.

a) Publication

Identification and functional characterization of Trypanosoma brucei peroxin 16.

Kalel, V. C., Schliebs, W., Erdmann, R. (2015)

Biochim Biophys Acta. 2015 Oct;1853(10 Pt A):2326‐37.

Biochimica et Biophysica Acta 1853 (2015) 2326–2337

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbamcr

Identiﬁcation and functional characterization of Trypanosoma brucei peroxin 16

Vishal C. Kalel, Wolfgang Schliebs, Ralf Erdmann ⁎

Department of Systems Biochemistry, Institute of Biochemistry and Pathobiochemistry, Faculty of Medicine, Ruhr-University Bochum, Germany

article info abstract

Article history: Protozoan parasites of the family Trypanosomatidae infect humans as well as livestock causing devastating Received 20 February 2015 diseases like sleeping sickness, Chagas disease, and Leishmaniasis. These parasites compartmentalize glycolytic Received in revised form 22 May 2015 enzymes within unique organelles, the glycosomes. Glycosomes represent a subclass of peroxisomes and they Accepted 24 May 2015 are essential for the parasite survival. Hence, disruption of glycosome biogenesis is an attractive drug target for Available online 27 May 2015 these Neglected Tropical Diseases (NTDs). Peroxin 16 (PEX16) plays an essential role in peroxisomal membrane ﬁ Keywords: protein targeting and de novo biogenesis of peroxisomes from endoplasmic reticulum (ER). We identi ed ﬁ Peroxisome trypanosomal PEX16 based on speci c sequence characteristics and demonstrate that it is an integral glycosomal Peroxin membrane protein of procyclic and bloodstream form trypanosomes. RNAi mediated partial knockdown of PEX16 Trypanosoma brucei PEX16 in bloodstream form trypanosomes led to severe ATP depletion, motility defects Trypanosomatids and cell death. Microscopic and biochemical analysis revealed drastic reduction in glycosome number and Glycosome mislocalization of the glycosomal matrix enzymes to the cytosol. Asymmetry of the localization of the remaining Trypanosoma brucei glycosomes was observed with a severe depletion in the posterior part. The results demonstrate that trypanosomal PEX16 is essential for glycosome biogenesis and thereby, provides a potential drug target for sleeping sickness and related diseases. © 2015 Published by Elsevier B.V.

1. Introduction novel drug targets and development of new affordable drugs against these diseases [6]. Neglected Tropical Diseases (NTDs) comprise seventeen infectious Trypanosomatid parasites harbor glycosomes, unique organelles that parasitic diseases, affecting more than 1 billion people worldwide, mostly compartmentalize glycolytic enzymes and other metabolic pathways, in developing countries [1,2]. Currently, more than 10 million people are which normally occur in the cytosol in other organisms [7]. This unique infected with protozoan parasites of the family Trypanosomatidae [3,4]. compartmentation is essential for the parasites, mislocalization of the African sleeping sickness is caused by Trypanosoma brucei, which is trans- glycosomal enzymes to the cytosol kills the parasite [8–10].Glycosomes mitted by the tsetse fly. Triatomine bugs transmit T. cruzi, the infectious belong to the family of peroxisomes, sharing the same principles of bio- parasite causing the American trypanosomiasis (Chagas disease). Leish- genesis [11]. Proteins required for the biogenesis of these organelles are maniasis is caused by Leishmania species that are transmitted by collectively called peroxins. Until now, thirty-three yeast peroxins and sandflies. Without treatment, African sleeping sickness is fatal, with pro- fifteen human and plant peroxins have been identified [12,13] but only gressive mental deterioration leading to coma, systemic organ failure, 10 trypanosomatid counterparts have been discovered till date [14]. and death [5]. The currently used drugs have several limitations such as Since trypanosomatids diverged very early from other eukaryotes during toxicity, adverse side effects, difficult to administer. NTDs mainly affect evolution [15], the level of sequence similarity is low, making it difficult countries, which only can provide limited resources for the development to identify orthologous peroxins of trypanosomatids by bioinformatic of new therapies. Hence, there is an urgent need for identification of approaches. Most of the few known peroxins of trypanosomatids are involved in glycosomal matrix protein import, which is reasonably well characterized

Abbreviations: PEX, peroxin; PTS, peroxisomal targeting signal; DAPI, 4′,6′-diamidino- [16]. However, our knowledge on glycosome membrane biogenesis is still 2-phenylindole; GAPDH, glycosomal glyceraldehyde-3 phosphate dehydrogenase; PFK, scarce. In other organisms, peroxisome membrane biogenesis requires phosphofructokinase; DIC, differential interference contrast; TbPEX16, Trypanosoma brucei three peroxins, PEX19, PEX3, and PEX16 [17]. So far, only trypanosomatid PEX16 PEX19 has been identiﬁed, while the corresponding PEX3 and PEX16 ho- ⁎ Corresponding author at: Department of Systems Biochemistry, Institute of mologs are still unknown [18]. PEX19 is the cytosolic receptor for newly Biochemistry and Pathobiochemistry, Ruhr-Universität Bochum, Universitätsstr. 150, 44780 Bochum, Germany. Tel.: +49 234 32 24943; fax: +49 234 32 14266. synthesized peroxisomal membrane proteins. The integral membrane E-mail address: [email protected] (R. Erdmann). protein PEX3 acts as anchor for PEX19 [19].PEX16ispresentinmammals

http://dx.doi.org/10.1016/j.bbamcr.2015.05.024 0167-4889/© 2015 Published by Elsevier B.V. V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 2327 and plants and in some yeasts like Yarrowia lipolytica, though in aldolase including the PTS2 signal), containing HindIII and BamHI over- Saccharomyces cerevisiae it seems to be absent. The functional role of hangs were annealed and ligated into HindIII–BamHI site in pGN1. PEX16 still remains elusive. Evidence has been provided suggesting a role of human PEX16 for proper targeting of PEX3 to the peroxisomes 2.4. Trypanosome stable transfection [20]. Mutations in peroxins required for membrane formation lead to complete loss of peroxisomes. In humans, such mutations are responsible All transfections were genomic integrations of NotI linearized plas- for inherited severe metabolic disorders, like the Zellweger syndrome mids and stably transfected clones were selected by limiting dilution. [21]. Since glycosomes are essential for trypanosomatid parasites, the par- Procyclic trypanosome transfection was performed as described in asite peroxins might provide suitable novel drug targets, especially as the [26]. Bloodstream form trypanosomes were transfected using Human degree of conservation between the parasite and human proteins is low. T-Cell nucleofector Kit (Lonza), the Amaxa Nucleofector II device and Here we identified and functionally characterized PEX16 of T. brucei. program X-001. 30 million Log phase cells were used for each transfec- We show that TbPEX16 is a glycosomal membrane protein. PEX16 spe- tion. The transformation mixture was transferred to 30 ml HMI11 medi- cific RNAi, which results in a partial knockdown of PEX16, kills the um (see Section 2.1) and serially diluted 1:10 and 1:100 to a final bloodstream form of the parasites in culture. Reduced TbPEX16 expres- volume of 30 ml. 1 ml aliquots of the cultures were transferred to sion directly affects glycosomes leading to drastic reduction in their three 24 well plates. Antibiotics were added after incubation for 8 h or number and mislocalization of glycosomal matrix proteins to the cyto- overnight (5 μ g/ml blasticidin from InvivoGen for pGC1, pGN1 and sol. The remaining glycosomes were asymmetrically distributed in the pHD1336 constructs; 5 μg/ml hygromycin from Invitrogen for pHD677 cell, most localized in the anterior part. The glycosome biogenesis defect and pHD918 constructs). Transformed cells were monitored 5–6days leads to a mislocalization of glycolytic enzymes, depletion in ATP levels, after transfection. The clones were induced with tetracycline (1 μg/ml) motility defects, and cell death. and positive clones were selected and further cultivated. Glycerol stocks of positive clones were stored at −80 °C in appropriate medium con- 2. Materials and methods taining 12% glycerol.

2.1. Trypanosome strains, growth conditions, and transfection 2.5. RNAi, RT-PCR, and qRT-PCR

Bloodstream and procyclic form cell line 449 (T. brucei 427 strain sta- Double stranded (stem-loop) RNA was inducibly expressed from bly transfected with pHD449 plasmid, thus stably expressing Tet repres- genomically integrated pHD677 construct described in Section 2.3, sor) were used in this study [22]. The bloodstream form was grown in bearing a tetracycline regulated trypanosome-specific promoter. RNAi HMI-11 medium containing 10% fetal bovine serum (FBS, Sigma) at was induced by addition of tetracycline (1 μg/ml) to the cultures of den- 37 °C in humidified incubator with 5% CO2 [23]. The procyclic form sity 2 × 105 cells/ml. Cells were diluted back to 2 × 105 every 24 h and was grown in SDM-79 medium supplemented with 15% FBS at 28 °C fresh tetracycline was added. The growth of uninduced and RNAi- in humidified 5% CO2 incubator [24]. Bloodstream form cells were main- induced cultures was monitored up to 8 days. RNA was isolated from tained in logarithmic phase (below 2 × 106 cells/ml) and procyclic form day-3 and day-7 cultures using RNAeasy Mini Kit (Qiagen), transcribed 5 8 cells were maintained at 5 × 10 –5×10 cells/ml. to cDNA using Oligo(dT)18 primers with RevertAid First Strand cDNA Synthesis Kit (Thermo scientific). Quantitative Realtime PCR was per- 2.2. Bioinformatics formed using MESA GREEN qPCR masterMix Plus on MJ Research DNA Engine Opticon thermal cycler. RT-PCR primers for tubulin are described BLAST searches were performed against trypanosomatid genomes at in [27]. tritrypdb.org. Pfam domain search was done at http://pfam.xfam.org. Multiple sequence alignment was generated using the Clustal Omega 2.6. Carbonate extraction online server and FASTA aligned sequences were visualized using JALVIEW with Clustalx color code. Amino acid percentage identity and Carbonate extraction was essentially performed as described by Lo- similarity matrix were generated using MatGAT2.01 with BLOSUM62 renz et al. [28], with the following modifications. 10 × 107 bloodstream scores. Prediction of transmembrane domains and topology was done form cells of Protein A- or TbPEX16-Protein A-expressing cell line 449, using Phobius prediction software (http://phobius.sbc.su.se). induced with 0.1 μg/ml tetracycline for 24 h, were used. Extracts were directly denatured with Laemmli buffer and subjected to SDS-PAGE. 2.3. Cloning PEX11 (1:5000) and GIM5 (1:5000) antisera were used as integral membrane protein marker, enolase (1:75,000) as cytosolic marker Genomic DNA of T. brucei 449 cell line or Leishmania major was used and aldolase (1:75,000) as glycosomal marker [29]. as template for PCR amplifications of desired genes using peqGOLD Pwo-DNA-Polymerase (Peqlab). 2.7. Microscopy TbPEX16 was C-terminally fused with Protein A tag by cloning the blunt PCR product of RE3135–RE3651 in HpaI digested pHD918 (kindly Trypanosomes were sedimented by centrifugation and resuspended donated by Dr. F. Voncken, The University of Hull). N- or C-terminally in 4% paraformaldehyde in PBS (supplemented with 250 mM sucrose in GFP tagged TbPEX16 was generated by cloning BamHI digested PCR case of RNAi experiments) and incubated on rotatory wheel for 20 min. products of RE3135–RE3130 and RE3135–RE3131 into BamHI site in Fixed trypanosomes were washed two times with PBS and immobilized trypanosome specific expression vectors pGC1 and pGN1, both derived on poly-L-lysine coated wells. Aldolase (1:1500) and GAPDH (1:5000) an- from plasmid pHD1336 [25], respectively. LmPEX16 was C-terminally tibodies were used as glycosomal markers. PEX11 and GIM5 antibodies GFP-tagged by cloning of the HindIII–BglII double digested PCR product (1:200) were used as glycosomal membrane markers. Goat anti-rabbit of RE3144–RE3145 into HindIII–BamHI sites of pGN1. Stem loop con- Alexa 594 was used as secondary antibody. Nuclear and kinetoplast struct for TbPEX16 RNAi was generated by cloning the 550 bp fragment DNA were stained with DAPI. Co-staining in Fig. 4C was done sequentially, (RE3285–RE3286, digested with HindIII–NcoI) and the complementary first with PEX11 and Alexa 488 (green) as secondary antibody, thorough- 500 bp fragment (RE3287–RE3288, digested with NcoI–BamHI) in tan- ly washed with PBS supplemented with 250 mM sucrose (5 min, 6 times) dem but opposite orientation in HindIII–BamHI Site in pHD677 [22].For followed by staining with aldolase and Alexa 594 (red) as secondary an- generation of fluorescent glycosomal marker PTS2-GFP, complementary tibody. Since both PEX11 and aldolase antibodies were raised in rabbit, oligonucleotides RE3474 and RE3475 (coding for first 15 amino acids of cells labeled with only one antibody and respective secondary antibody 2328 V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 were used as controls, which were confirmed not to interfere with the in T. brucei and suggests the topology of both N- and C- termini facing other channel during microscopy. Stained cells were layered with Mowiol towards cytosol, similar to human PEX16 (Fig. 1B). To study the mem- (Sigma) antifade-medium, covered with coverslips and allowed to poly- brane topology of TbPEX16, carbonate extraction of cells expressing merize overnight. Images were captured with a Carl Zeiss Microscope TbPEX16-Protein A was performed (Fig. 1C). Similar to known integral using Axiovision 4.6.3 software. In case of weak fluorescence of membrane proteins TbPEX11 and TbGIM5, TbPEX16-Protein A was TbPEX16-GFP, Z-stacks were acquired, deconvoluted, and merged. For completely resistant to carbonate extraction. The Protein A tag alone is statistical analysis, a minimum of 100 cells were counted in each soluble as it is completely extracted similar to cytosolic marker enolase, experiment. which is predominantly released with low salt buffer. Aldolase is an Motility of uninduced and PEX16 RNAi-induced live cells was abundant glycosomal matrix enzyme, which is strongly attached or as- screen-captured with a handheld mobile device camera. Electron mi- sociated with the glycosomal membrane in Leishmania [29]. The extrac- croscopy was performed as described in [28] and images of cells with tion behavior indicates that TbPEX16 is an integral membrane protein, complete transverse section visible (n = 7) were chosen for quantifica- like human PEX16. tion of glycosomes. 3.2. Trypanosomatid PEX16 localizes to glycosomes 2.8. Digitonin fractionation For the assessment of the subcellular localization of PEX16 by fluores- Digitonin fractionation was performed according to [30] using equal cence microscopy, tetracycline inducible N- and C-terminal GFP fusion amount of cells (corresponding to 30 μg protein) treated with increas- constructs of TbPEX16 were generated and expressed in the procyclic ing amount of digitonin (0 to 2 mg digitonin/mg protein) as indicated and bloodstream form trypanosomes. Several single clones were selected in Fig. 5. 1% Triton-X100-solubilized trypanosomes served as control by limiting dilution of transformants. Induction of GFP with 1 μg/ml tetra- for the starting material, which corresponds to the complete release of cyclineledtotheappearanceofagreenpunctatepattern(Fig. 2AandB). enzymes in cells. The treated cells were centrifuged to yield superna- Glycosomes of formaldehyde-fixed trypanosomes were visualized by im- tants which were subjected to SDS-PAGE and probed with glycosomal munofluorescence microscopy for the glycosomal marker enzyme aldol- matrix markers phosphofructokinase (1:25,000), aldolase (1:75,000), ase. The labeling revealed a punctate pattern, which is typical for a GAPDH (50,000), hexokinase (1:50,000), cytosolic marker enolase glycosomal localization (Fig. 2A and B, top panels). Localization of (1:75,000), and glycosomal membrane markers GIM5 and PEX11 PEX16 is indicated by GFP-fluorescence. The green punctate pattern of (both 1:5000). “Total” corresponds to the cells treated with 1% Triton- both, the N- and C-terminal GFP fusion of TbPEX16 colocalized with aldol- X100, which were directly denatured with Laemmli buffer without the aseinbothprocyclicformtrypanosomes(Fig. 2A) and bloodstream form centrifugation step. trypanosomes (Fig. 2B), indicating that PEX16 is a glycosomal protein. Ag- gregation of glycosomes was observed in procyclic form trypanosomes 2.9. ATP measurements expressing higher levels of GFP-TbPEX16 (Supplementary Fig. S1), similar to aggregation of peroxisomes observed in plant cells expressing N- ATP extracts were prepared according to [31] with modifications. terminal GFP fusion of PEX16 [35]. 5 μl of ATP extract was added to 95 μl ATP assay buffer and 100 μl To investigate whether Leishmania PEX16 is also localized in CellTiter-Glo® Reagent (Promega). After 10 min incubation at room glycosomes, heterologous expression of LmPEX16-GFP was performed temperature, luminescence was monitored with a Synergy H1 plate in bloodstream form trypanosomes (Fig. 2B, lower panel). Similar to reader (BioTek GmbH). TbPEX16, Leishmania PEX16-GFP fusion also colocalized with glycosomal marker aldolase. The glycosomal localization of TbPEX16 is supported by 3. Results previous proteomic findings in which TbPEX16 (Tb09.160.4700) was identified in the glycosomal proteome of bloodstream form trypano- 3.1. Identification of PEX16 in trypanosomatids somes [36]. Recently, a PEX16-homolog was also detected in the mem- braneproteomeofpurified glycosomes from Leishmania tarentolae [29]. To identify putative PEX16 candidates from trypanosomatid para- These findings together with our results verified that Trypanosoma and sites, BLAST searches were performed against the parasite proteins Leishmania PEX16 are authentic glycosomal proteins. (Tritrypdb.org), using known yeast (Yarrowia), human and plant PEX16 protein sequences. The hits were checked manually for the presence of the PEX16 pfam domain (PF08610). The BLAST search using 3.3. RNAi knockdown of PEX16 impairs the growth of bloodstream Arabidopsis PEX16 (SSE1) protein sequence identified a single protein form trypanosomes in T. brucei TREU927 genome currently annotated as hypothetical protein (Tb927.9.6450/Tb09.160.4700 in T. brucei Lister strain 427, hereaf- To study the functions of PEX16 in trypanosomes, tetracycline- ter TbPEX16), which contained the PEX16 domain with E-value 2.8e-22. inducible RNA interference was applied to reduce its expression level. Orthologs of putative TbPEX16 in the available genomes of all species of A stem-loop construct of TbPEX16 was generated and transfected into the Trypanosoma and Leishmania genera also contain the PEX16 pfam bloodstream form trypanosomes. Preliminary analysis of RNAi- domain. induction by addition of tetracycline showed specificgrowthdefectsof Multiple sequence alignment of known and trypanosomatid puta- several clones. The RNAi experiment was initiated with wild-type (con- tive PEX16 candidates revealed a low degree of conservation over the trol), uninduced (RNAi clone without tetracycline), and induced (RNAi entire protein length (Fig. 1A). Trypanosomatid putative PEX16 proteins clone plus tetracycline) cultures at density of 2 × 105 cells/ml. Viable are longer than the known PEX16 proteins due to the presence of addi- (motile) cells were manually counted with a Neubauer chamber and tional internal stretches of amino acids. Trypanosomatid PEX16 candi- the cultures were daily diluted to 2 × 105 cells/ml. Growth defect started dates have ~16% identity and ~30% similarity to known PEX16 appearing specifically in tetracycline-induced RNAi cell line from day 2 proteins (Supplementary Table S2). onwards. ~70% growth reduction compared to uninduced cells was ob- PEX16 is an integral membrane protein in mammals with both N- served on day 3 (Fig. 3A). Cell death was evident in RNAi-induced cul- and C-termini facing cytosol [32], while Yarrowia and Arabidopsis tures by appearance of dead cell debris. Quantitative real time PCR PEX16 are reported as peripheral membrane proteins [33,34].Com- (qRT-PCR) analysis revealed that TbPEX16 mRNA levels were reduced bined transmembrane topology and signal peptide prediction using to 30% on day 3 (Fig. 3B). PEX16-mRNA depletion was also confirmed Phobius server indicates the presence of two transmembrane domains by semi-quantitative analysis of tubulin (control) and PEX16 mRNA V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 2329

Fig. 1. Bioinformatics features and membrane association of PEX16. (A) Multiple sequence alignment of putative trypanosomatid PEX16 and representative PEX16 proteins. UniProt accession numbers — T. brucei (Q38ET6), T. cruzi (Q4D653), L. major (Q4QFA4), L. donovani (E9BCC5), H. sapiens (Q9Y5Y5), A. thaliana (Q8S8S1), Y. lipolytica (P78980). (B) Schematic representation of human and T. brucei PEX16, black box — transmembrane span (phobius prediction), gray box — PEX16 Pfam PF08610 aligned region. (C) TbPEX16 is an integral membrane protein. Wild-type (non-transfected cell line 449), Protein A- and TbPEX16-Protein A-expressing cell lines were sequentially extracted with low salt buffer (lanes 1), high salt buffer (lanes 2) and alkaline sodium carbonate (lanes 3). Lysates were centrifuged and the resulting supernatants (lanes 1–3) and carbonate resistant sediments (lanes 4) were subjected to immunoblot analysis with antibodies against glycosomal integral membrane proteins (PEX11, GIM5), cytosolic enolase and glycosomal aldolase as indicated. *cross-reactivity with Protein A.

levels by routine PCR with cDNA isolated from uninduced, RNAi-induced Our qRT-PCR analysis suggests that the partial growth phenotype is and wild-type 449 cells (Fig. 3C). due to incomplete RNAi knockdown. Therefore, for further analysis of The growth defect was consistently observed from day 3 to day 6. the phenotype, day-3 or day-4 RNAi-induced cells were used. From day 7 onwards, the cell density in RNAi-induced cells increased. Appearance of RNAi-insensitive revertants is commonly observed in 3.4. TbPEX16 RNAi affects import of matrix proteins into glycosomes prolonged growth in culture [37,38]. qRT-PCR analysis showed that indeed PEX16-mRNA levels were increased at later time points To analyze the effects of knockdown of TbPEX16 expression on (~65% day 7, data not shown). In genome-scale knockdown studies in glycosomal protein import, immunoﬂuorescence microscopic localiza- T. brucei brucei TREU927 strain using RNA interference target sequenc- tion of glycosomal matrix and membrane protein markers was ing (RIT-seq), a similar growth phenotype was seen for PEX16 [39]. performed. Aldolase and GAPDH are abundant glycosomal matrix 2330 V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337

marker aldolase revealed that both proteins colocalize completely in uninduced cells (Fig. 4C upper panel). However, in RNAi-induced cells aldolase was mislocalized to the cytosol but the membrane protein PEX11 still localized to the remaining glycosomes which also contained aldolase (Fig. 4C lower panel). There was no apparent mislocalization of PEX11. In another approach, we used a PTS2-GFP expressing PEX16-RNAi cell line, where complete colocalization of PTS2-GFP with PEX11 was observed under uninducing conditions (Fig. 4D, upper panel). Upon RNAi induction, a partial mislocalization of PTS2-GFP to the cytosol was evident, while PEX11 localized to the few present glycosomes (Fig. 4D, lower panel). As a complementary biochemical approach, RNAi uninduced and induced cells were subjected to digitonin fractionation to assess the mislocalization of glycosomal matrix proteins to the cytosol (Fig. 5). Cells were treated with increasing amounts of digitonin to permeabilize the plasma membrane, leading to the release of the cytosolic proteins to the supernatant upon centrifugation. Enolase is a cytosolic enzyme and accordingly it was completely released to the supernatant of both uninduced and RNAi-induced cells, even at low concentration of digitonin (0.05 mg/mg protein). Whereas higher concentrations of digitonin (0.5– 0.75 mg/mg protein) were required to liberate glycosomal proteins into the supernatant from uninduced control cells (Fig. 5, panels labeled “Un”), minor portions of aldolase and hexokinase were released at intermediate digitonin concentrations (0.25–0.5 mg/mg protein). This could be due to extra-glycosomal localization of glycolytic enzymes which was previously reported for hexokinase [41]. However in RNAi-induced cells (Fig. 5, panels labeled “In”), both PTS1-containing (PFK, GAPDH) and PTS2-containing (aldolase, hexokinase) glycosomal enzymes were already well detected in the supernatants at lower digitonin concentrations (0.05–0.5 mg/mg protein). This further demonstrates the mislocalization of glycosomal proteins due to TbPEX16 RNAi.

3.5. TbPEX16 knockdown affects glycosome number and distribution

As described in Section 3.5, the mislocalization of glycosomal matrix proteins was observed in cells, which contained only few import competent glycosomes. This might indicate that the matrix protein mislocalization is an indirect effect of strong reduction in the glycosome number. To estimate the number of remaining normal glycosomes, we used PTS2-GFP expressing PEX16-RNAi cell line and performed an immunofluorescence microscopic analysis with the glycosomal membrane marker PEX11. Combination of these two markers did provide a Fig. 2. PEX16 GFP fusion proteins localize to glycosomes. C- or N-terminally GFP-tagged reliable way to count remaining normal glycosomes. Wild type blood- TbPEX16 as indicated was expressed in (A) procyclic form (PCF) and (B) bloodstream stream form trypanosomes contain ~65 glycosomes per cell s[42].A form trypanosomes (BSF). (B, lower panel) Heterologous expression of Leishmania PEX16 in bloodstream form trypanosomes. GFP-fusions were monitored by GFP-autofluorescence. shown in Fig. 6A (upper panel), PEX11 completely colocalized with The glycosomal marker aldolase was visualized by immunofluorescence microscopy. Top PTS2-GFP in uninduced cells giving numerous distinct and co- panels in A and B are uninduced cells as control for normal glycosome morphology. Scale localizing puncta. In contrast, the number of co-localizing spots was — μ μ bar 10 m(inA);5 m(inB). drastically reduced in RNAi induced cells, for e.g. below 10 in the representative cell shown (Fig. 6A, lower panel). The number of remaining glycosomes was counted in 100 individual cells of both uninduced and RNAi-induced cultures. The statistical distribution analysis is shown in enzymes targeted to glycosomes by PTS2- and PTS1-targeting signals, Fig. 6B. The majority of uninduced cells displayed ~21–35 glycosomes respectively. Both of these enzymes appear as distinct punctate pattern marked with PTS2-GFP and PEX11 (Fig. 6B, blue bars). The number is in wild type (Fig. 2) as well as in the RNAi-cell line without induction of a slight understatement of the actual number as several spots were RNAi (Fig. 4). not individually distinguishable, especially in uninduced cells where However, upon induction of PEX16 RNAi, both aldolase and GAPDH glycosome number is high. However, in RNAi-induced cells, ~25% of gave diffuse cytosolic signals, indicative of a cytosolic mislocalization of cells have less than 10 normal glycosomes, while more than 50% cells the glycosomal matrix proteins (Fig. 4A and B). Fewer import competent have less than 15 glycosomes (Fig. 6B, red bars). glycosomes (less than 10) were still present in these cells. A complete Another striking feature of RNAi induced cells concerns the subcellu- mislocalization of matrix proteins was not observed, which was not ex- lar distribution of glycosomes. Usually, glycosomes are equally distribut- pected as PEX16-mRNA was not completely knocked down. Upon RNAi, ed in both anterior and posterior parts of the slender trypanosome with only newly synthesized proteins will be mislocalized and accumulated nucleus in the middle (Fig. 7A). Upon PEX16 knockdown, the few re- in cytosol to initiate their toxic effect (see discussion), whereas already maining glycosomes were not equally distributed but mainly present imported proteins persist in glycosomes for several generations due to in the anterior part of the trypanosomes. About 5% of cells contained their stability and longer lifetime [40]. Immunofluorescence microscopi- no or just one glycosome in the posterior part (Fig. 7A). The fluorescence cal co-staining of the glycosomal membrane marker PEX11 and matrix microscopy data were corroborated by electron microscopic analysis of V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 2331

Fig. 3. PEX16 RNAi knockdown leads to growth defect in bloodstream form trypanosomes. (A) Cumulative growth curve of wild-type (non-transfected 449 cell line), uninduced and induced TbPEX16 RNAi cell lines. Cultures were inoculated in triplicates at density of 2 × 105 cells/ml and RNAi was induced with 1 μg/ml tetracycline. Cells were daily counted; cultures were diluted to 2 × 105 each day and fresh tetracycline added. The Log10 of cumulative cell counts are plotted using Graphpad Prism (version 6.04). Error bars — SD of triplicate readings. p b 0.0001 (B) qRT-PCR quantiﬁcation of TbPEX16 mRNA levels (normalized with tubulin mRNA levels) in uninduced and induced cultures from day 3. Values are shown in percentage relative to uninduced cells. Error bars — SD of triplicate readings. p b 0.0001 (calculated with Graphpad Prism 6.04 using an unpaired, two-tailed Student's T-test) represented by *** in graph. (C) Semi-quantitative analysis of tubulin (control) and PEX16 mRNA levels by routine PCR with cDNA isolated from wild-type 449 cells, uninduced and RNAi-induced day-3 cultures.

uninduced and RNAi-induced cells (Fig. 7B). Glycosomes appear as single 4. Discussion membrane-bound organelles with a granular matrix. Counting of glycosomes visible in electron micrographs revealed that uninduced Glycosomes are typical and essential peroxisome-like organelles of cells contained roughly equal number of glycosomes in anterior as well protozoan parasites of the family Trypanosomatidae. They harbor first as posterior part of the cell (Fig. 7C). Reduction in total number of seven enzymes of glycolysis, the pathway which is the sole source of glycosomes in PEX16 knockdown cells, especially in the posterior part ATP for the bloodstream form of the parasite. As such, they provide a suit- of the cells was also evident in the PEX16-depleted cells. able target for the development of drugs against devastating African sleeping sickness, Chagas disease, and Leishmaniasis. Here we report on 3.6. Reduced ATP levels and motility defect upon PEX16 knockdown the identification of PEX16 as a novel peroxin of trypanosomatid parasites. TbPEX16 localizes to the glycosomal membrane and knockdown of Glycolysis is the sole source of ATP in bloodstream form trypano- its expression in bloodstream form parasites did result in defects in somes. The first seven glycolytic enzymes are compartmentalized in glycosome biogenesis, especially reduction in the glycosome number the glycosomal matrix where they do not provide net ATP. The last and mislocalization of glycosomal matrix proteins to the cytosol. Due to two enzymes of the glycolytic pathway are located in the cytosol and lack of feedback inhibition of hexokinase and phosphofructokinase, produce 2 mol ATP/mol glucose. This unique distribution of glycolytic their mislocalization results in depletion of cellular ATP levels which is ag- enzymes is crucial for the parasite, since the trypanosomal glycolytic en- gravated by accumulation of toxic metabolites causing death of trypano- zymes like hexokinase and phosphofructokinase lack feedback inhibi- somes [8,10,43,44]. Similar to TbPEX16, the knockdown of peroxins tion. Accordingly, if these enzymes are mislocalized to the cytosol, PEX19 or PEX2 leads to a reduction in glycosome number and matrix pro- they deplete cellular ATP levels by unregulated runaway phosphoryla- tein import defects in procyclic form trypanosomes [18,45].Glycosome tion and accumulate toxic metabolites leading to cell death [8,10,43, number is also reduced upon knockdown of PEX11 or GIM5 expres- 44]. Since the glycosomal enzymes including hexokinase and phospho- sion [28,46]. The glycosome biogenesis defects due to PEX16 knockdown fructokinase were mislocalized to the cytosol upon PEX16 knockdown, observed in this study resemble peroxisome biogenesis defects reported we tested whether this accompanied by a depletion of the cellular ATP for mutations of PEX16 or siRNA silencing of PEX16 expression in other levels. Analysis of ATP levels in equal number of uninduced and RNAi- organisms [33,35,47–49]. PEX16 defects in Zellweger human patient induced cells revealed that the total ATP levels were drastically reduced cells [48,49], Arabidopsis [35,50] and Y. lipolytica [33] lead to complete upon knockdown of PEX16 (Fig. 8A). Up to 70% decrease in the ATP absence of morphologically detectable peroxisomes and exhibit levels was observed. mislocalization of peroxisomal proteins. However, some mutations in Continuous motility of trypanosomes is a major ATP consuming pro- PEX16 which are associated with milder clinical phenotype in humans, cess. As ATP levels were drastically reduced in trypanosomes upon lead to reduced number of enlarged import competent peroxisomes [47]. PEX16 RNAi-induction, we investigated the motility of bloodstream Decrease of the expression rate of TbPEX16 of up to 70% led to re- form parasites. Microscopic observation of these cells disclosed motility duced number of glycosomes and mislocalization of glycosomal matrix defects in ~5% of the cells. The defective motility was not seen in proteins. Glycosomal membrane proteins PEX11 and GIM5 still local- uninduced cells. The motility defective cells move slowly, display com- ized to the glycosomes. In Arabidopsis, knockdown of PEX16 with sim- plete loss of flagellar tip-to-base beating and thus exhibit reversed di- ilar efficiency (as for TbPEX16 in this study) resulted in a decrease in rection of motility in comparison to normal cells (Fig. 8B, Suppl. number but increase in size of peroxisomes [51]. siRNA knockdown of Video). The observed defect might be secondary or indirect. However, PEX16 expression in mammalian cells led to specific mislocalization of as the motility defect is only seen in cells with drastically reduced PEX3, while peroxisomal targeting of PEX11 and PEX14 was not affect- glycosome number it appears not to be an off-target effect. ed. Therefore it was proposed that PEX16 functions as an anchor for 2332 V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337

Fig. 4. RNAi knockdown of PEX16 leads to mislocalization of glycosomal matrix proteins. Uninduced and RNAi-induced cells (day 3) were fixed with formaldehyde and processed for immunofluorescence microscopy with (A) aldolase, (B) GAPDH, and (C) PEX11 and aldolase (sequential co-staining). (D) PTS2-GFP expressing PEX16 RNAi cell line stained with PEX11 antibody. White arrows in lower panels of (A) and (B) indicate affected cells, Cells marked with white box in (A) and (B) are magnified on the right side for comparison. Scale bar — 5 μm.

PEX3 import into peroxisomes [20]. In trypanosomatid parasites, the did not result in promising PEX3 candidates. So far, PEX3 has been iden- PEX3 orthologue is not known. Bioinformatics search for trypanosomal tiﬁed in most species and its prominent role in formation of the perox- proteins with PEX3 Pfam domain as well as proteomic approaches so far isomal membrane and de novo synthesis of the organelles makes it V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 2333

Fig. 5. Effect of PEX16-RNAi on the subcellular distribution of glycosomal proteins analyzed by digitonin-dependent cell fractionation. Intact uninduced (Un) and RNAi-induced (In) cells were incubated with increasing concentrations of digitonin as described in Materials and methods. Cell suspensions were centrifuged to yield supernatants, which were resolved by SDS- PAGE and analyzed by immunoblotting with antibodies against GAPDH, PFK (glycosomal PTS1 proteins), aldolase, hexokinase (glycosomal PTS2 proteins), enolase (cytosolic marker), GIM5, and PEX11 (glycosomal membrane markers) as indicated. 1% Triton indicates supernatants of cells incubated with 1% Triton-X100 to dissolve all membranes. Total corresponds to cell suspension with 1% Triton-X100 without centrifugation.

Fig. 6. Glycosome number is reduced upon knockdown of PEX16 expression. (A) Immunoﬂuorescence microscopy analysis of PEX11 in an uninduced and RNAi-induced PTS2-GFP expressing PEX16-RNAi cell line. Scale bar — 5 μm (B) Statistical analysis of number of glycosomes (marked by PTS2-GFP and PEX11) per cell in uninduced (blue bars) and RNAi-induced (red bars) cells (100 cells were counted in each case). 2334 V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337

Fig. 7. Differential effect of PEX16-depletion on glycosomes distribution. (A) Representative images of uninduced and induced TbPEX16 RNAi cells (DIC, PTS2-GFP and TbPEX11 as glycosomal markers and DAPI) showing disappearance of glycosomes from the posterior part of the cell (scheme — first panel). Scale bar — 5 μM (B) Ultrastructural analysis of uninduced and RNAi-induced cells from day-3, glycosomes are marked with black asterisks. (C) Number of glycosomes (total, posterior part, anterior part) in electron micrographs shown in Fig. 6B were counted. Number of glycosomes from posterior and anterior part and total number (sum of both) are plotted (n = 7). rather unlikely that its function is dispensable for glycosome biogenesis. Binding and release of dynein with microtubule requires ATP [54], TbPEX3 might have escaped its identification due to a low sequence which in the bloodstream form is generated solely by glycolysis. Upon de- conservation or its function in trypanosomatids is performed by an un- pletion of PEX16, around 5% of the cells lost the tip-to-base flagellar beat related protein. Here PEX16 as possible TbPEX3-membrane anchor and directionality of motility was reversed. Such motility defect was ob- might be a good tool for its identification. served previously in procyclic form trypanosomes upon knockdown of Trypanosome usually contains approximately equal number of flagellar proteins Dynein Light Chain 1 (LC1) or outer-arm dynein subunit glycosomes in both the anterior and posterior part of the cell. Remarkably, DNAI1,whichbothleadtolossofouter-armdyneincomplex[55].Inthe depletion of TbPEX16 did not only lead to a reduction in glycosome num- bloodstream form, flagellar protein knockdown is mostly lethal. Point ber, but the remaining glycosomes were predominantly in the anterior mutants of LC1 have been reported, which display motility defects but part, while the posterior part was nearly depleted from glycosomes. A are viable in culture [55]. Same mutants in procyclic form have defective similar effect was also noticed in GIM5knockdownbloodstreamformtry- motility but still move in normal forward direction. Persistent backward panosomes [52], while in GIM5-knockdown cells of the procyclic form, motility was first reported in the wild-type bloodstream form trypano- the number of glycosomes was equally reduced over entire cell body somes under specificgrowthcondition,i.e.physicalconfinement of para- [46]. Long slender bloodstream form trypanosomes are highly polarized sites using for e.g. artificial narrow spaced micropillar arrays, which cells with several single organelles that occupy defined subcellular posi- mimic the physical conditions trypanosome may encounter in host tion [53]. A single lysosome and golgi apparatus are positioned between body [56]. However, in standard growth HMI-medium, persistent back- nucleus and flagellar pocket in the posterior part of the cell. Both blood- ward motility or persistent reversed flagellar beat is not observed [56]. stream and procyclic form trypanosomes are highly motile with a tip- Therefore, the motility defect observed in PEX16 knockdown cells, is to-base flagellar beat imparting the directionality of propulsive motility. first such report for bloodstream-form trypanosomes grown in standard V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 2335

Fig. 8. Depletion of ATP levels and motility defect. (A) Total ATP levels in equal number of uninduced and RNAi-induced cells from days 1–4werequantified with CellTiterGlo reagent and expressed as percentage relative to uninduced cells; Error bars — SD from triplicate readings. (B) Motility defect in PEX16 knockdown cells. Screenshots were extracted from video (see Supplementary Video) using VLC video player. Trypanosome marked with black arrow depicts normal directional motion and black cyclic arrows depict tumbling. Trypanosome marked with white arrow shows reversed motility direction and slow speed. Time in seconds. (See Supplementary video). culture conditions (HMI-medium). ATP depletion may cause overall cel- protein TbDLP (Dynamin-Like Protein), whose knockdown did not affect lular abnormalities including loss of outer-arm-dynein complex, which glycosome biogenesis [63], indicating that fission of the preexisting can explain the backward motility as observed in LC1 [57] and DNAI1 glycosomes might not be the main pathway to maintain glycosome num- [58] knockdowns and PEX16 knockdown in this study. ATP depletion in ber. The de novo peroxisome biogenesis from ER is dependent on PEX16 bloodstream form trypanosomes has been reported in various studies in humans [64], where specifically Sec16B is involved in exit of PEX16 such as RNAi knockdown of different proteins or using chemical inhibitors from ER [65,66]. Trypanosomes harbor only one isoform of Sec16. This [59],low-glucosemediaor2-deoxy-D-glucose [60], but under these con- isoform resembles human Sec16B and forms a single ER exit site in the ditions backward motility defects have not been observed. Thus, ATP de- posterior part of the trypanosome between nucleus and kinetoplast pletion does not per se lead to motility defects as observed for the PEX16 [67]. Investigation of TbSec16 role in PEX16 trafficking and glycosomes knockdown. It should be noted that the growth defect observed in PEX16 biogenesis may shed light on whether also glycosomes are formed de knockdown (Fig. 3) is less severe than observed for PEX5 or PEX7 knock- novo from the ER. The reduced number of glycosomes in the region of down [38]. Accordingly, a motility phenotype might have escaped the at- the ER exit site is indicative of the existence of a de novo biogenesis tention as cells die rapidly. There are some glycosomal enzymes known to pathway. During transition between bloodstream to procyclic form, localize to the flagellum such as hexokinase-2 (HK2) [41], adenylate ki- glycosomes containing glycolytic enzymes are degraded by autophagy nase isoforms [61]. HK2 shares 98% similarity to HK1, both harbor PTS2 [68]; therefore new glycosomes are more likely to be formed by de signal at N-terminus and differ only at the C-terminus [41].Neitherthe novo biogenesis rather than fission in order to adapt to glucose limiting function of HK2 nor the regulation of the bipartite localization of HK is conditions in insect host [68]. There are organism specificdifferencesre- known. In PEX16-knockdown, glycosomal hexokinase was mislocalized garding the contribution of fission of preexisting peroxisomes or de novo to the cytosol, which could interfere the flagellar targeting or function of biogenesis from ER in maintaining the number peroxisomes in normal flagella localized HK2. Further studies with PEX16 RNAi cells may reveal cells. In yeast like S. cerevisiae, there is ample evidence that fission is the whether there is a direct link between glycosomes and regulation of major pathway, while de novo synthesis of peroxisomes is only occurring motility. under certain conditions [69]. In mammalian cells, the de novo pathway is Peroxisomes number is maintained by two pathways; by fission of supposed to play a more prominent role [64]. According to our data, de preexisting peroxisomes, or de novo biogenesis from ER. PEX16 plays an novo biogenesis might be the predominant pathway for glycosome bio- important role in both pathways. Y. lipolytica PEX16 is involved in perox- genesis in trypanosomes even in non-differentiating conditions. isome fission as it contributes to the recruitment of dynamin-like protein Our knowledge on the biogenesis of glycosomes is scarce. The newly Vps1 to peroxisomes [62]. Trypanosomes encode a single dynamin family identified PEX16 and PEX19 might be instrumental for the identification 2336 V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 of the long-searched trypanosomal PEX3, which will contribute to our [25] C. Yernaux, et al., Trypanosoma brucei glycosomal ABC transporters: identification and membrane targeting, Mol. Membr. Biol. 23 (2) (2006) 157–172. understanding of the topogenesis of glycosomal membrane proteins [26] H. Krazy, P.A. Michels, Identification and characterization of three peroxins—PEX6, and de novo synthesis of the organelles. Furthermore, owing to their PEX10 and PEX12—involved in glycosome biogenesis in Trypanosoma brucei, low similarity to human counterparts, PEX16 and associated proteins Biochim. Biophys. Acta 1763 (1) (2006) 6–17. [27] A. Alloatti, et al., Genetic and chemical evaluation of Trypanosoma brucei oleate could be attractive targets for developments of drugs against desaturase as a candidate drug target, PLoS One 5 (12) (2010) e14239. trypanosomatid parasites. [28] P. Lorenz, et al., Elongation and clustering of glycosomes in Trypanosoma brucei Supplementary data to this article can be found online at http://dx. overexpressing the glycosomal PEX11p, EMBO J. 17 (13) (1998) 3542–3555. doi.org/10.1016/j.bbamcr.2015.05.024. [29] C. Colasante, et al., Proteins and lipids of glycosomal membranes from Leishmania tarentolae and Trypanosoma brucei 2 (2013) 27 F1000Res. [30] T. Hausler, et al., Import of a DHFR hybrid protein into glycosomes in vivo is not inhibited by the folate-analogue aminopterin, J. Cell Biol. 132 (3) (1996) 311–324. Acknowledgements [31] M. Delgado, et al., Neuropeptides kill African trypanosomes by targeting intracellular compartments and inducing autophagic-like cell death, Cell Death Differ. 16 (3) We are grateful to Prof. Paul Michels for providing strains, plasmids, (2009) 406–416. and antibodies. We are thankful to Dr. Frank Voncken (The University of [32] M. Honsho, T. Hiroshige, Y. Fujiki, The membrane biogenesis peroxin PEX16p. Topogenesis and functional roles in peroxisomal membrane assembly, J. Biol. Hull) for providing pHD918 plasmid and PEX11, GIM5 antisera. We thank Chem. 277 (46) (2002) 44513–44524. Monica Bürger and Beate Menzel for technical assistance. This work was [33] G.A. Eitzen, R.K. Szilard, R.A. Rachubinski, Enlarged peroxisomes are present in oleic supported by grants from the Deutsche Forschungsgemeinschaft (Er178/ acid-grown Yarrowia lipolytica overexpressing the PEX16 gene encoding an intraperoxisomal peripheral membrane peroxin, J. Cell Biol. 137 (6) (1997) 1265–1278. 4-1). [34] S.K. Karnik, R.N. Trelease, Arabidopsis peroxin 16 coexists at steady state in peroxisomes and endoplasmic reticulum, Plant Physiol. 138 (4) (2005) 1967–1981. [35] Y. Lin, J.E. Cluette-Brown, H.M. Goodman, The peroxisome deficient Arabidopsis mu- References tant sse1 exhibits impaired fatty acid synthesis, Plant Physiol. 135 (2) (2004) 814–827. [36] D. Vertommen, et al., Differential expression of glycosomal and mitochondrial pro- [1] Working to Overcome the Global Impact of Neglected Diseases — First WHO Report teins in the two major life-cycle stages of Trypanosoma brucei,Mol.Biochem. on Neglected Tropical Diseases, 2010. Parasitol. 158 (2) (2008) 189–201. [2] Working to Overcome the Global Impact of Neglected Tropical Diseases: First WHO [37] Y. Chen, et al., Development of RNA interference revertants in Trypanosoma brucei Report on Neglected Tropical Diseases — Update, 2011. cell lines generated with a double stranded RNA expression construct driven by [3] Drugs for neglected diseases initiative (DNDi) — websitecited; Available from: two opposing promoters, Mol. Biochem. Parasitol. 126 (2) (2003) 275–279. http://www.dndi.org/index.php/diseases.html. [38] N. Galland, et al., Characterization of the role of the receptors PEX5 and PEX7 in the [4] F. Chappuis, et al., Human African trypanosomiasis in areas without surveillance, import of proteins into glycosomes of Trypanosoma brucei, Biochim. Biophys. Acta Emerg. Infect. Dis. J. 16 (2) (2010) 354. 1773 (4) (2007) 521–535. [5] K. Stuart, et al., Kinetoplastids: related protozoan pathogens, different diseases, [39] S. Alsford, et al., High-throughput phenotyping using parallel sequencing of RNA in- J. Clin. Invest. 118 (4) (2008) 1301–1310. terference targets in the African trypanosome, Genome Res. 21 (6) (2011) 915–924. [6] R. Pink, et al., Opportunities and challenges in antiparasitic drug discovery, Nat. Rev. [40] C.E. Clayton, Most proteins, including fructose bisphosphate aldolase, are stable in Drug Discov. 4 (9) (2005) 727–740. the procyclic trypomastigote form of Trypanosoma brucei, Mol. Biochem. Parasitol. [7] F.R. Opperdoes, Compartmentation of carbohydrate metabolism in trypanosomes, 28 (1) (1988) 43–46. Annu. Rev. Microbiol. 41 (1987) 127–151. [41] A.C. Joice, et al., Extra-glycosomal localisation of Trypanosoma brucei hexokinase 2, [8] T. Furuya, et al., Glucose is toxic to glycosome — deficient trypanosomes, Proc. Natl. Int. J. Parasitol. 42 (4) (2012) 401–409. Acad. Sci. U. S. A. 99 (22) (2002) 14177–14182. [42] L. Tetley, K. Vickerman, The glycosomes of trypanosomes: number and distribution [9] R. Kumar, et al., Expression of a PTS2-truncated hexokinase produces glucose toxic- as revealed by electron spectroscopic imaging and 3-D reconstruction, J. Microsc. ity in Leishmania donovani, Mol. Biochem. Parasitol. 170 (1) (2010) 41–44. 162 (Pt 1) (1991) 83–90. [10] B.M. Bakker, et al., Compartmentation protects trypanosomes from the dangerous [43] J.R. Haanstra, et al., Compartmentation prevents a lethal turbo-explosion of glycoly- design of glycolysis, Proc. Natl. Acad. Sci. U. S. A. 97 (5) (2000) 2087–2092. sis in trypanosomes, Proc. Natl. Acad. Sci. U. S. A. 105 (46) (2008) 17718–17723. [11] P.A. Michels, et al., Peroxisomes, glyoxysomes and glycosomes (review), Mol. [44] P.S. Kessler, M. Parsons, Probing the role of compartmentation of glycolysis in Membr. Biol. 22 (1-2) (2005) 133–145. procyclic form Trypanosoma brucei: RNA interference studies of PEX14, hexokinase, [12] S. Nagotu, et al., Molecular basis of peroxisomal biogenesis disorders caused by de- and phosphofructokinase, J. Biol. Chem. 280 (10) (2005) 9030–9036. fects in peroxisomal matrix protein import, Biochim. Biophys. Acta 1822 (9) (2012) [45] C. Guerra-Giraldez, L. Quijada, C.E. Clayton, Compartmentation of enzymes in a 1326–1336. microbody, the glycosome, is essential in Trypanosoma brucei, J. Cell Sci. 115 (Pt [13] J. Hu, et al., Plant peroxisomes: biogenesis and function, Plant Cell. 24 (6) (2012) 13) (2002) 2651–2658. 2279–2303. [46] F. Voncken, et al., Depletion of GIM5 causes cellular fragility, a decreased glycosome [14] M. Gualdron-Lopez, et al., Ubiquitination of the glycosomal matrix protein receptor number, and reduced levels of ether-linked phospholipids in trypanosomes, J. Biol. PEX5 in Trypanosoma brucei by PEX4 displays novel features, Biochim. Biophys. Acta Chem. 278 (37) (2003) 35299–35310. 1833 (12) (2013) 3076–3092. [47] M.S. Ebberink, et al., Identification of an unusual variant peroxisome biogenesis dis- [15] M. Gualdron-Lopez, et al., When, how and why glycolysis became compartmentalised order caused by mutations in the PEX16 gene, J. Med. Genet. 47 (9) (2010) 608–615. in the Kinetoplastea. A new look at an ancient organelle, Int. J. Parasitol. 42 (1) (2012) [48] M. Honsho, et al., Mutation in PEX16 is causal in the peroxisome-deficient Zellweger 1–20. syndrome of complementation group D, Am. J. Hum. Genet. 63 (6) (1998) 1622 –1630. [16] M. Gualdron-Lopez, et al., Translocation of solutes and proteins across the [49] S.T. South, S.J. Gould, Peroxisome synthesis in the absence of preexisting peroxi- glycosomal membrane of trypanosomes; possibilities and limitations for targeting somes, J. Cell Biol. 144 (2) (1999) 255–266. with trypanocidal drugs, Parasitology 140 (1) (2013) 1–20. [50] Y. Lin, et al., The Pex16p homolog SSE1 and storage organelle formation in [17] L. Dimitrov, S.K. Lam, R. Schekman, The role of the endoplasmic reticulum in perox- Arabidopsis seeds, Science 284 (5412) (1999) 328–330. isome biogenesis, Cold Spring Harb. Perspect. Biol. 5 (5) (2013) a013243. [51] K. Nito, et al., Functional classification of Arabidopsis peroxisome biogenesis factors [18] S.K. Banerjee, et al., Identification of trypanosomatid PEX19: functional characteriza- proposed from analyses of knockdown mutants, Plant Cell Physiol. 48 (6) (2007) tion reveals impact on cell growth and glycosome size and number, Mol. Biochem. 763–774. Parasitol. 142 (1) (2005) 47–55. [52] A. Maier, et al., An essential dimeric membrane protein of trypanosome glycosomes, [19] F.L. Theodoulou, et al., Peroxisome membrane proteins: multiple traffi cking routes Mol. Microbiol. 39 (6) (2001) 1443–1451. and multiple functions? Biochem J. 451 (3) (2013) 345–352. [53] K.R. Matthews, The developmental cell biology of Trypanosoma brucei,J.CellSci.118 [20] T. Matsuzaki, Y. Fujiki, The peroxisomal membrane protein import receptor PEX3p is (Pt 2) (2005) 283–290. directly transported to peroxisomes by a novel PEX19p- and PEX16p-dependent [54] K.L. Hill, Biology and mechanism of trypanosome cell motility, Eukaryot. Cell 2 (2) pathway, J. Cell Biol. 183 (7) (2008) 1275–1286. (2003) 200–208. [21] Y. Fujiki, Y. Yagita, T. Matsuzaki, Peroxisome biogenesis disorders: molecular basis [55] K.S. Ralston, N.K. Kisalu, K.L. Hill, Structure-function analysis of dynein light chain 1 for impaired peroxisomal membrane assembly: in metabolic functions and biogen- identifies viable motility mutants in bloodstream-form Trypanosoma brucei, esis of peroxisomes in health and disease, Biochim. Biophys. Acta 1822 (9) (2012) Eukaryot. Cell 10 (7) (2011) 884–894. 1337–1342. [56] N. Heddergott, et al., Trypanosome motion represents an adaptation to the crowded [22] S. Biebinger, et al., Vectors for inducible expression of toxic gene products in blood- environment of the vertebrate bloodstream, PLoS Pathog. 8 (11) (2012) e1003023. stream and procyclic Trypanosoma brucei, Mol. Biochem. Parasitol. 85 (1) (1997) [57] D.M. Baron, Z.P. Kabututu, K.L. Hill, Stuck in reverse: loss of LC1 in Trypanosoma 99–112. brucei disrupts outer dynein arms and leads to reverse flagellar beat and backward [23] H. Hirumi, K. Hirumi, Continuous cultivation of Trypanosoma brucei blood stream movement, J. Cell Sci. 120 (Pt 9) (2007) 1513–1520. forms in a medium containing a low concentration of serum protein without feeder [58] C. Branche, et al., Conserved and specific functions of axoneme components in try- cell layers, J. Parasitol. 75 (6) (1989) 985–989. panosome motility, J. Cell Sci. 119 (Pt 16) (2006) 3443–3455. [24] R. Brun, Schonenberger, Cultivation and in vitro cloning or procyclic culture forms of [59] C. Worthen, B.C. Jensen, M. Parsons, Diverse effects on mitochondrial and nuclear Trypanosoma brucei in a semi-defined medium. Short communication, Acta Trop. 36 functions elicited by drugs and genetic knockdowns in bloodstream stage (3) (1979) 289–292. Trypanosoma brucei, PLoS Negl. Trop. Dis. 4 (5) (2010) e678. V.C. Kalel et al. / Biochimica et Biophysica Acta 1853 (2015) 2326–2337 2337

[60] S.K. Natesan, et al., Evidence that low endocytic activity is not directly responsible [65] K. Tani, et al., Dual function of Sec16B: Endoplasmic reticulum-derived protein se- for human serum resistance in the insect form of African trypanosomes, BMC Res. cretion and peroxisome biogenesis in mammalian cells, Cell Logist. 1 (4) (2011) Notes 3 (2010) 63. 164–167. [61] M.L. Ginger, et al., Intracellular positioning of isoforms explains an unusually large [66] S. Yonekawa, et al., Sec16B is involved in the endoplasmic reticulum export of the adenylate kinase gene family in the parasite Trypanosoma brucei,J.Biol.Chem. peroxisomal membrane biogenesis factor peroxin 16 (PEX16) in mammalian cells, 280 (12) (2005) 11781–11789. Proc. Natl. Acad. Sci. U. S. A. 108 (31) (2011) 12746–12751. [62] T. Guo, et al., A signal from inside the peroxisome initiates its division by promoting [67] M. Sealey-Cardona, et al., Sec16 determines the size and functioning of the Golgi in the remodeling of the peroxisomal membrane, J. Cell Biol. 177 (2) (2007) 289–303. the protist parasite, Trypanosoma brucei. Trafﬁc. 15 (6) (2014) 613–629. [63] G.W. Morgan, D. Goulding, M.C. Field, The single dynamin-like protein of Trypanosoma [68] M. Herman, et al., Turnover of glycosomes during life-cycle differentiation of brucei regulates mitochondrial division and is not required for endocytosis, J. Biol. Trypanosoma brucei, Autophagy 4 (3) (2008) 294–308. Chem. 279 (11) (2004) 10692–10701. [69] A.M. Motley, E.H. Hettema, Yeast peroxisomes multiply by growth and division, [64] P.K. Kim, et al., The origin and maintenance of mammalian peroxisomes involves a de J. Cell Biol. 178 (3) (2007) 399–410. novo PEX16-dependent pathway from the ER, J. Cell Biol. 173 (4) (2006) 521–532. RESULTS ‐ Chapter I

Supplementary table (from the publication)

T. brucei T. cruzi L. major L. donovani H. sapiens A. thaliana Y. lipolytica T. brucei 35.2 28.8 28.8 15.5 17.8 15.3 T. cruzi 52.7 32.6 32.4 18.1 17.3 17.2 L. major 46.4 49.9 95.2 16.1 14.9 16.9 Identity (%) L. donovani 45.8 49.7 97.3 16.6 14.9 17.4 H. sapiens 32.5 29.6 26.5 26.3 23.9 22 A. thaliana 32.2 31.9 28.4 28.2 44.7 21 Y. lipolytica 34.9 31.9 28.8 29.1 41.4 38.6 similarity (%)

Percentage Identity and Similarity Matrix of PEX16 protein. Amino acid sequences of PEX16 proteins shown in Fig. 1A were compared using BLOSUM62 scoring method in MatGAT2.01 software.

Supplementary Figure (from the publication)

Aggregation of glycosomes in procyclic form trypanosomes overexpressing GFP‐TbPEX16. Procyclic form trypanosomes expressing GFP‐TbPEX16 were induced with 1 μg/ml tetracycline. Immunofluorescence microscopy co‐localization was performed with glycosomal marker aldolase. Scale bar — 10 μm.

RESULTS ‐ Chapter I

List of Primers used in the publication

Primer Sequence (Restriction site, stop codon) RE3135 GTAGGATCCATGGTGTCCCGGGTATTCAG RE3651 AACCTCTCCGATAGAGTACATGAAC RE3131 CAGGATCCTCTCCGATAGAGTACATG RE3130 CAGGATCCTCACTCTCCGATAGAGTAC RE3144 ACTAAGCTTCGATGCAATCAATGTCTCGTG RE3145 ACTAGATCTCGCTGCCGCAGGGTGTACAG RE3285 GATCAAGCTTTGCGACGCCAG RE3286 GATACCATGGAAGGGTCCAGATACTCC RE3287 GATCCCATGGTATCAACGACTTCACAG RE3288 GATAGGATCCTGCGACGCCAGCTCC AGCTTATGTCCAAGCGTGTTGAAGTTCTGCTTACCCAACTCCCTGC RE3474 GTACGCG GATCCGCGTACGCAGGGAGTTGGGTAAGCAGAACTTCAACACGCT RE3475 TGGACATA RE3443 AGCGGTTTACGCACATGCTC RE3444 CTTCAGGAGCTGCAGAATCA

RESULTS ‐ Chapter I

Chapter I: Identification and functional characterization of Trypanosoma brucei peroxin 16.

b) Supplementary (unpublished)

1. Supplementary materials and methods 2. Supplementary results

RESULTS ‐ Chapter I

b) Supplementary (Unpublished)

1. Supplementary materials and methods

1.1 Cells growth conditions and transformations GM5756 fibroblasts (Wild type or pex16 defective Zellweger patient cell line) immortalized with SV40 antigen were grown in DMEM medium supplemented with 10% FCS (Sigma) in 0 humidified 37 C incubator with 8% CO2. S. cerevisiae cells of BY4742 strain (WT, ∆pex19, ∆pex3) were grown in 0.3% glucose containing medium (Uracil Histidine double dropout to maintain GFP and DsRed constructs) at 300C. Peroxisome proliferation was induced by growing cells in 0.3% oleate containing medium. For yeast two‐hybrid studies, different PCY2 strains (WT, ∆Pex19, ∆Pex3) were used. Human cells were transfected by Amaxa Nucleofector II device using Nucleofector kit (Lonza) or by using liposomal reagent X‐ tremeGENE (Roche). Yeast cells were transformed by routine Lithium‐acetate method.

1.2 Plasmids construction Trypanosoma, Leishmania or human PEX16 proteins were fused to GFP at N‐ or C‐terminus for expression in human cells (Fig. S1 and S2) or in yeast (Fig. S3). Primer pairs and the respective restriction sites and the cloning vector used are listed in Table 2. Primer sequences are listed in Table 3.

Table 2 ‐ Cloning strategies Expression Restriction Construct Primer pair Cloned in vector in sites pEGFP‐N1 TbPEX16‐GFP RE3129 ‐ RE3131 EcoRI ‐ BamHI (Clonetech) pEGFP‐C1 GFP‐TbPEX16 RE3129 ‐ RE3130 EcoRI ‐ BamHI (Clonetech) Human cells LmPEX16_GFP_F pEGFP‐N1 LmPEX16‐GFP BglII ‐ HindIII ‐ RE3022 (Clonetech) pEGFP‐N1 HsPEX16‐GFP RE3146 ‐ RE3148 XhoI ‐ BglII (Clonetech) pAG‐416‐Gal‐ccdB‐ TbPEX16‐GFP RE3119 ‐ RE3121 AttB1 ‐ AttB2 EGFP (Addgene) Yeast pUG35 (AG LmPEX16‐GFP RE3020 ‐ RE3022 SpeI ‐ HindIII Erdmann)

RESULTS ‐ Chapter I

Table 3 ‐ Primer sequences Primer name Sequence (Restriction site, Start codon, Stop codon, AttB sites) LmPEX16_GFP_F ACTAGATCTCTGCCGCAGGGTGTACAGGA RE3020 ACAACTAGTATGCAATCAATGTCTCGTGAG RE3022 ATCAAGCTTCTGCCGCAGGGTGTACAG GGGGACAAGTTTGTACAAAAAAGCAGGCTTCACCATGGTGTCCCGGGTATT RE3119 CAGTGCG GGGGACCACTTTGTACAAGAAAGCTGGGTCCTCTCCGATAGAGTACATGAA RE3121 CGAG RE3129 ACGAATTCTATGGTGTCCCGGGTATTC RE3130 CAGGATCCTCACTCTCCGATAGAGTAC RE3131 CAGGATCCTCTCCGATAGAGTACATG RE3146 ACACTCGAGCAACCATGGAGAAGCTGCGGCTC RE3148 ACTAGATCTCGGCCCCAACTGTAGAAGTAG

For restriction based cloning, standard molecular biology techniques and high quality enzymes (NEB, Peqlab) were used. For expression of TbPEX16‐GFP in yeast “Gateway technology” was used (Invitrogen). TbPEX16 gene was amplified using primers RE3119‐ RE3121 which contain AttB1 and AttB2 sites respectively. The PCR product was cloned in Gateway Donor vector pDONR221 by BP‐clonase mediated recombination. The donor vector was further recombined with yeast Gateway destination vector pAG‐426GAL‐ccdB‐EGFP by using LR‐clonase to get Gal inducible TbPEX16‐EGFP construct. GFP fusion constructs o T. brucei PEX16 for expression in trypanosomes are described in the publication (Kalel VC et al. Chapter I publication)

1.3 Microscopy Human cells expressing various fluorescent proteins were grown on coverslips, fixed with 4% paraformaldehyde in Dulbecco’s phosphate buffered saline (DPBS from Invitrogen) for 20 minutes at room temperature. Fixed cells were washed two times with DPBS. For immunofluorescence microscopy, α‐HsPEX14 antibody (1:200 dilution) was used as peroxisomal marker. α‐HsTRAP1 (1:250, ABR) was used as mitochondrial marker and α‐ HsCalreticulin (1:200, ABR) as ER marker. Goat anti‐rabbit or anti‐mouse Alexa 594 was used as secondary antibody. Coverslips were mounted with Mowiol (Sigma) on glass slide. DsRed‐ SKL (plasmid pUG34‐DsRed‐SKL, AG Erdmann) was used as peroxisomal marker in yeast

RESULTS ‐ Chapter I

fluorescence microscopy studies. Yeast cells expressing various fluorescent proteins were directly visualized on microscope without fixation. Microscopy was performed with Carl Zeiss Microscope using Axiovision 4.6.3 software.

1.4 Yeast two‐hybrid Yeast two‐hybrid was performed in cooperation with Udaya Bhandari (Master’s thesis). Full length or various truncations of Trypanosoma PEX16 were cloned in pPC97 vector containing Gal‐Binding Domain (BD). Full length yeast PEX19, PEX13; Trypanosoma PEX19, PEX13.1; human PEX19 in pPC86 vector containing Gal‐Activation domain (AD) were from plasmid collection (AG Ralf Erdmann). Yeast PEX17169‐178 in pPC97 and PEX14 in pPC86 were provided by Anna Chan. Co‐transformation of various BD and constructs were transformed in yeast two‐hybrid strain PCY2 by routine Lithium acetate method. The clones were selected using double dropout (‐Tryptophan ‐Leucine) synthetic dextrose plates. pPC vectors and filter based β‐galactosidase assay was done as described previously (Chevray and Nathans 1992).

RESULTS ‐ Chapter I

2. Supplementary results

2.1 Trypanosomatid PEX16 localization in human cells

Fig. S1 ‐ Localization of heterologously expressed Trypanosoma and Leishmania PEX16 in human cells. A. N‐ or C‐terminally GFP tagged Trypanosoma brucei PEX16 or GFP alone (control) were expressed in human GM5756 fibroblast cells. Colocalization of GFP tagged TbPEX16 with peroxisomal marker PEX14 (red) by immunofluorescence microscopy is seen in merge (yellow spots) in middle and bottom panels. B. C‐terminally GFP‐tagged Leishmania major PEX16 colocalized partially with peroxisomal marker PEX14 (top) and partially with mitochondrial marker TRAP (bottom). Scale bars ‐ 10µm.

To assess whether human peroxisomal protein import machinery can recognize trypanosomal PEX16 proteins, heterologous expression of Trypanosoma and Leishmania PEX16 proteins was performed in human fibroblast cell line GM5756. PEX16 proteins were fused to GFP on either N‐ or C‐terminus and were transfected into human cells. After 24

RESULTS ‐ Chapter I

hours of incubation, cells were processed for immunofluorescence microscopy (see suppl. Methods, Section 1.3). Both Human PEX16 GFP fusion constructs HsPEX16‐GFP or GFP‐ HsPEX16 localize to peroxisomes (Kim et al. 2006, Matsuzaki and Fujiki 2008), confirming that the GFP fusion on either C‐ or N‐terminus does not interfere with peroxisomal targeting of human PEX16. As seen in Fig. S1 A, Both TbPEX16‐GFP and GFP‐TbPEX16 gave punctate staining characteristic of peroxisomes and colocalized completely with peroxisomal marker HsPEX14 (Fig. S1 A, middle and bottom panels respectively). GFP alone was used as control, which localizes to cytosol appearing as diffuse pattern (Fig. S1 A top panel).

However in case of C‐terminal GFP fusion of Leishmania PEX16 (LmPEX16‐GFP), dual localization was observed, punctate and network like staining (Fig. S1 B). Punctate pattern of LmPEX16‐GFP colocalized with the peroxisomal marker HsPEX14 and therefore corresponds to the peroxisomal localization of LmPEX16 (Fig. S1 B, Top panel). The network staining turned out to be mitochondrial localization since the pattern colocalized with the mitochondrial marker HsTRAP (Fig. S1 B, Lower panel). Same peroxisomal and mitochondrial dual localization was also observed for GFP‐LmPEX16.

2.2 Parasite PEX16 proteins do not rescue human PEX16 defects

Trypanosoma PEX16 completely localized to peroxisomes whereas Leishmania PEX16 dually localized to peroxisomes and mitochondria when expressed in normal human cells. To study whether the parasite PEX16 proteins can complement the deficiency of human PEX16 function, GFP tagged Trypanosoma or LeishmaniaPEX16 were expressed in Zellweger patient fibroblast cells which have function inactivating mutation in PEX16. Due to lack of functional PEX16, the cells completely lack peroxisomes and peroxisomal matrix proteins are mislocalised to cytosol. Expression of peroxisomal marker DsRed‐SKL in these cells resulted in the diffuse cytosolic labelling (Fig. S2 A). Upon reintroduction of functional full length Human PEX16 such as HsPEX16‐GFP, new peroxisomes are formed in the cells in which DsRed‐SKL is imported into peroxisomes and colocalized with HsPEX16‐GFP indicating that the function of PEX16 is restored (Fig. S2 A ‐ lower panel).

RESULTS ‐ Chapter I

Fig. S2 ‐ Trypanosomatid PEX16 proteins do not rescue dysfunction of PEX16 in human cells. A. Peroxisomal marker DsRed‐SKL appears as punctate pattern in normal human fibroblast (top left), while in PEX16 defective Zellweger patient cells which lack peroxisomes, diffuse cytosolic staining is seen (top right). Expression of C‐terminal GFP fusion of Human PEX16 (HsPEX16‐GFP) leads to the formation of new peroxisomes which import DsRed‐SKL and colocalization is seen as congruent punctate pattern (bottom panel). B. C‐terminally GFP tagged Trypanosoma (TbPEX16‐GFP) or Leishmania PEX16 (LmPEX16‐GFP) both fail to rescue human PEX16 dysfunction since DsRed‐SKL is still mislocalised to the cytosol (top and bottom panels respectively). C. TbPEX16‐GFP is mislocalised to the ER as seen by colocalization with ER marker calreticulin (top panel). LmPEX16 is mislocalised to the mitochondria evident from colocalization with mitochondrial marker TRAP (bottom panel).

Unlike HsPEX16‐GFP, expression of neither Trypanosoma nor Leishmania PEX16 GFP fusion proteins (TbPEX16‐GFP or LmPEX16‐GFP) could restore the function of defective human PEX16 (Fig. S2 B). The peroxisomal reporter protein DsRed‐SKL was still mislocalised to the cytosol giving diffuse staining. This demonstrated that though Trypanosoma and Leishmania PEX16 can localize to peroxisomes in normal human cells, both parasite PEX16 proteins fails to rescue the dysfunction of human PEX16. TbPEX16‐GFP was mislocalised to ER in PEX16

RESULTS ‐ Chapter I

defective human cells as seen by colocalization with ER marker calreticulin (Fig. S2 C, top panel). On the other hand, LmPEX16 was mislocalised to mitochondria in these cells as evident from colocalization with mitochondrial marker TRAP (Fig. S2 C, bottom panel).

2.3 Trypanosomatid PEX16 localization in yeast

Fig. S3 ‐ Localization of heterologously expressed Trypanosoma and Leishmania Pex16 in yeast. A. C‐terminally GFP tagged Trypanosoma or Leishmania were co‐expressed with peroxisomal marker DsRed‐SKL in wild type S. cerevisiae. TbPEX16‐GFP partly co‐localizes with peroxisomes (Middle panel) while LmPEX16 completely co‐localizes with peroxisomes (Bottom panel). B. Tb and LmPEX16‐GFP (Middle and bottom panels respectively) completely mislocalised to ER in Peroxisome deficient ∆pex19 S. cerevisiae cells where DsRed‐SKL is mislocalised to the cytosol. Scale bars ‐ 5µm in top and bottom panels, 2µm in middle panel.

Similar to heterologous expression in human cells (Section 2.1 and 2.2), the localization of Tb and LmPEX16 was investigated in yeast cells (S. cerevisiae). Fig. S3 panel A, shows the localization of C‐terminally GFP fused TbPEX16 and LmPEX16 in wild type BY4742 cells grown in oleate containing media to induce peroxisome proliferation. GFP alone was used as control which localizes to cytosol while peroxisomal marker DsRed‐SKL is imported into peroxisomes leading to appearance of a punctate pattern (Fig. S3, top panel). In wild type yeast cells, TbPEX16‐GFP always yielded punctate and reticular dual localization (Fig. S3, middle panel). Peroxisomes labelled with DsRed‐SKL colocalized with TbPEX16‐GFP puncta indicating that TbPEX16‐GFP can only partially localize to peroxisomes in yeast. However major portion of TbPEX16‐GFP mislocalised to ER giving characteristic reticular pattern and

RESULTS ‐ Chapter I

moreover, all peroxisomes appear to be attached to the ER. In human cells we previously showed that TbPEX16 completely localizes to peroxisomes in normal cells (Fig. S1 A) while in pex16 mutant fibroblast, it completely mislocalised to ER (Fig. S2 B). This indicates that human and yeast peroxisome biogenesis machinery recognize parasite PEX16 differently.

Conversely opposite effect was seen for localization of LmPEX16 in human or yeast cells. In wild‐type yeast cells, LmPEX16 was efficiently targeted to peroxisomes as seen by complete colocalization of LmPEx16‐GFP with peroxisomal marker DsRed‐SKL (Fig. S3 A, lower panel). In wild‐type human cells LmPex16‐GFP was only partially localized to peroxisome while major portion mislocalised to mitochondria (Fig. S1 B).

Both Trypanosoma and Leishmania PEX16 could not complement the human PEX16 function in PEX16 defective patient cell line. Similar complementation studies were not possible in budding yeast (S. cerevisiae) since Pex16 homolog or a protein with redundant function is not yet identified. Yeast pex19 and pex3 deletion strains completely lack peroxisomes. Therefore we used these two strains to investigate where parasite PEX16 proteins would localize in absence of peroxisomes. In ∆pex19 yeast cells devoid of peroxisomes, peroxisomal marker DsRed‐SKL completely mislocalised to the cytosol and colocalized with GFP alone (control) (Fig. S3 B, top panel). In these cells, both Trypanosoma and Leishmania PEX16‐GFP proteins were completely mislocalised to ER and gave characteristic reticular pattern, while DsRed‐SKL still mislocalised to the cytosol (Fig. S3 B, middle and bottom panel respectively). Same ER localization was observed for N‐terminal GFP tagged Lm and TbPEX16 (not shown).

2.4 Identification of Trypanosoma PEX16 binding partners

Though PEX16 is essential for peroxisome biogenesis in different organisms, very little is known about the binding partners of PEX16. Currently PEX19 is the only known binding partner of PEX16 based on yeast and bacterial two‐hybrid assays (Fransen et al. 2002). Therefore we investigated the interaction between Trypanosoma PEX16 and PEX19 by using yeast two‐hybrid method. Full length TbPEX16 (1‐453 amino acids) was fused to GAL‐Binding domain (BD), while full length PEX19 protein from different organisms (Trypanosoma,

RESULTS ‐ Chapter I

Human and S. cerevisiae) were fused to GAL‐Activation domain (AD). Since PEX19 GAL‐BD fusion proteins are auto‐activating, the yeast two‐hybrid in reverse order was not possible to perform. We could not detect interaction between full length TbPEX16 with any of the PEX19 proteins tested (Fig. S4 Left panel). This was not unexpected since the most common limitation of yeast two‐hybrid is that membrane proteins might be unstable or unable to translocate to nucleus to activate Gal promoter. Since TbPEX16 is an integral membrane protein, we chose to use soluble fragment for interaction studies.

Human PEX16 harbors PEX19 binding site towards its N‐terminus. Sequence analysis of TbPEX16 indicated that a fragment consisting of first 100 amino acid region to be soluble.

Therefore we tested the interaction between first 100 amino acids of TbPEX16(1‐100) with PEX19 from different organisms. First 100 amino acids of Trypanosoma PEX16 were sufficient to interact with Trypanosoma PEX19 and gave blue color in filter based β‐ galactosidase assay (Fig. S4 A, left panel). All constructs were also tested for auto‐activation and no auto‐activation was seen for constructs used in the experiment (Fig. S4 A, right

panel). Positive control for yeast two‐hybrid interaction was BD‐ScPEX17(169‐178) and AD‐ ScPEX14 (Fig. S4 A light bottom panel) (plasmids provided by Anna Chan, AG Erdmann).

Although TbPEX16(1‐100) interacts with TbPEX19, there was no interaction observed with human as well as yeast PEX19 (Fig. S4 A, right panel). The Trypanosoma PEX16‐PEX19 was also seen in PCY2 yeast two‐hybrid strains in which pex19 or pex3 gene is deleted (not shown). This demonstrated that the interaction between TbPEX19 and TbPEX19 is very specific and could serve as drug target for design of specific inhibitors to block glycosome biogenesis.

RESULTS ‐ Chapter I

Fig. S4 ‐ TbPEX16 interacts specifically with TbPEX19 and TbPEX13.1 at different sites A. Full length (1‐453) or first 100 amino acids of Trypanosoma PEX16 were tested in yeast two‐hybrid for interaction with full length PEX19 from different organisms. TbPEX16(1‐100) interacts specifically with Trypanosoma PEX19 as seen with blue color in X‐gal assay (left). Respective negative controls (right) and a positive control ScPex17‐ScPex14 is shown (right bottom). B. Different fragments of TbPEX16 were tested for interaction with full length Trypanosoma PEX13.1 or yeast Pex13. Full length TbPEX16 (1‐453) and (1‐300) specifically interact with TbPEX13.1 in wild type (WT) PCY2 strain (right), while interaction between only full length TbPEX16 with TbPEX13.1 was observed in ∆pex3 or ∆pex19 PCY2 strains (right). Experiment performed in collaboration with Udaya Bhandari (Master’s thesis).

RESULTS ‐ Chapter I

To identify more binding partners, we tested the interaction of TbPEX16 with various other trypanosomal as well as yeast peroxisomal membrane proteins. Trypanosoma PEX14 and all yeast PMPs tested did not interact with TbPEX16 (not shown). However surprisingly we detected interaction between Trypanosoma PEX16 and TbPEX13.1 (Fig. S4 B, top left) in wild‐type PCY2 strain. This is a novel finding since there are no reports of similar interaction in any organism. We therefore used different truncation constructs of TbPEX16 for interaction with full length TbPEX13.1. Positive interaction with TbPEX13.1 was seen with full length TbPEX16(1‐453) as well as TbPEX16(1‐300) in which C‐terminal part of TbPEX16 is removed (Fig. S4 B) in wild type PCY2 strain. However, no interaction was seen between

shorter fragments of TbPEX16 i.e. TbPEX16(1‐200) and TbPEX16(1‐100) and full length TbPEX13.1. Though TbPEX19 interacts with first 100 amino acids of TbPEX16, no interaction was seen with longer constructs (not shown) which suggests that TbPEX16‐TbPEX19 interaction might not be sufficient to stabilize the full length or longer fragments of TbPEX16 which contain transmembrane domains. Positive and specific interaction of longer fragments of TbPEX16 with TbPEX13.1 suggests that TbPEX13.1 likely contributes to stability of TbPEX16. Role of PEX13 in matrix protein import is well established. In yeast, Pex13 is required for peroxisomal targeting of Pex14. Our results suggest a potentially novel role of PEX13 in PMP targeting.

The interaction between TbPEX16 and TbPEX13.1 is very specific since neither full length or fragments of TbPEX16 interact with S. cerevisiae Pex13 (Fig. S4 B, bottom left) or Neurospora crassa PEX13 (not shown). Therefore TbPEX16‐TbPEX13.1 interaction could be another novel drug target for disrupting glycosome biogenesis in Trypanosomes.

We found that full length TbPEX16(1‐453) also interacts with TbPEX13.1 in ∆pex19 and ∆pex3 deleted PCY2 strains but unexpectedly unlike in wild‐type PCY2 strain, the shorter fragment

TbPEX16(1‐300) fails to interact with TbPEX13.1. One possible explanation is that there are potentially two binding sites in TbPEX16 for TbPEX13.1, where binding site in shorter

fragment TbPEX16(1‐300) is potentially bridged by yeast protein such as Pex19 and Pex3.

RESULTS ‐ Chapter I

Based on yeast two‐hybrid results, it can be concluded that Trypanosoma PEX16 interacts specifically with Trypanosoma PEX19 and PEX13.1 at two different sites which individually could be novel drug targets.

2.5 Overexpression of GFP‐PEX16 leads to glycosome aggregation

GFP DIC Merge

TbPEX16-GFP

GFP-TbPEX16

Fig. S5 ‐ Overexpression of GFP‐TbPEX16 leads to aggregation of glycosomes. C‐terminally GFP tagged TbPEX16 (TbPEX16‐GFP) localizes to glycosomes in procyclic form trypanosomes giving punctate pattern over entire cell body (top panel). However overexpression of N‐ terminally GFP tagged TbPEX16 leads to aggregation of glycosomes and accumulation of the aggregate specifically in the posterior part of the cell (bottom panel). DIC ‐ Differential interference contrast.

RNAi knockdown of PEX16 in Trypanosomes leads to drastic reduction in glycosome number where glycosomes in posterior part disappeared prominently (Fig. 7 in the publication in Chapter I). Expression of C‐ or N‐terminal GFP fusion of TbPEX16 in procyclic form trypanosomes exhibited drastically distinct effects on glycosome morphology. C‐terminal GFP tagged TbPEX16 (TbPEX16‐GFP) localized to glycosomes giving a normal punctate pattern over entire cell body (Fig. S5, top panel). However, in case of N‐terminal GFP fusion of TbPEX16 (GFP‐TbPEX16) upon overexpression by adding higher amounts of inducer of expression (tetracycline) or by incubating cells over longer time points, we observed that

RESULTS ‐ Chapter I

glycosome aggregated and accumulated at posterior end of cell (Fig. S5, bottom panel). This distinct change in glycosome distribution is first report in literature. Aggregation of peroxisomes in plant cells overexpressing GFP‐PEX16 is reported. Results of yeast two‐ hybrid studies provide a possible explanation for this phenotype. N‐terminal GFP fusion likely masks the binding sites at N‐terminus of TbPEX16 and C‐terminus is available in excess for interaction for e.g. with TbPEX13.1 as seen in yeast two‐hybrid studies, which could aggregate glycosomes (section 2.4). Plausible explanation of why this aggregate accumulates predominantly in posterior part of the cell will be discussed in Discussion section.

RESULTS ‐ Chapter II

Chapter II

Identification and characterization of inhibitors of PEX5‐PEX14 interaction

1. Materials and methods 2. Results

This chapter reports the invention of novel glycosome biogenesis inhibitors as potential therapy against Human African Trypanosomiasis. The work described in this section was done in collaboration with Helmholtz Zentrum München. Compounds design, synthesis and in vitro screening of PEX5‐PEX14 interaction inhibition (AlphaScreen) was performed by collaborators at Helmholtz Zentrum München. As part of this thesis, I tested all compounds (~200) for their biological activities on Trypanosomes and key compounds were tested on human cells for cytotoxicity (Fig. 1‐4). The results of my studies enabled Structure Activity relationship (SAR) guided optimization of compounds into highly potent anti‐trypanosomal inhibitors. I also performed different experiments on Trypanosomes to validate that the compounds act on target in vivo by providing evidence for inhibition glycosome biogenesis (Fig. 5‐8). Animal studies (Fig. 9) were performed by Swiss Tropical and Public Health Institute (Swiss TPH), Basel. The invention has been covered with Patent (Priority Art October 2014) and international patent application (Patent Cooperation Treaty ‐ PCT) has been filed in September 2015. The results described in this chapter will be published in near future (Manuscript in preparation).

RESULTS ‐ Chapter II

1. Materials and methods

1.1 Anti‐trypanosomal assay

Anti‐trypanosomal activities were tested on T. brucei brucei bloodstream form parasites 0 (strain ‐ Lister 427, VATMITat 1.2) grown in HMI‐11 medium in 37 C incubator with 5% CO2 (Hirumi and Hirumi, 1989). Alternatively, procyclic form trypanosomes (29‐13) were used that were grown in SDM79 medium (Brun and Schonenberger, 1979) without NaHCO3 at 0 27 C in absence of CO2. Stock solutions of compounds (50mM) were prepared in DMSO. Two fold serial dilutions of each compound (10 wells in each row) were prepared in 96 well plates in HMI‐11 or SDM79 medium (100μl/well, quadruplicates). One well without compound, and one well with media alone were included in each row. Bloodstream or procyclic form trypanosomes were inoculated in all wells (100 μl of 4x104/ml BSF or 1x106/ml PCF cells), except in the well with "media alone". Final concentration of parasites was 2x104/ml for BSF and 5x105/ml for PCF. The plates were incubated for 66 hours (BSF) or 70 hours (PCF). 25μl resazurin (Sigma, 0.1 mg/ml in Hanks Balanced Salt Solution) was added to all wells and the plates were further incubated till 72 hour time‐point. Change in color of resazurin from blue to fluorescent pink by living cells was measured Synergy H1 96 well plate reader (excitation at 530 nm, emission at 585 nm, filter cut‐off: 570 nm). Percent inhibition values were calculated by setting the well without compound to "0% inhibition". Non‐linear regression graphs were plotted in Graphpad Prism to yield sigmoidal dose‐

response curves and IC50 values were determined (compound concentration giving 50% inhibition).

1.2 Human cell cytotoxicity assay

Human Embryonic Kidney (HEK) or GM5756 fibroblast cells grown in DMEM supplemented with 10% FCS (Sigma) were seeded in 96 well plate (Plate 1) at density of 5x104 cells/ml (100µl per well) in wells 2‐12. 200µl DMEM alone was added in well 1 as “media alone” 0 control. Cells were grown overnight by incubating in 37 C incubator with 8% CO2. In a separate 96 well plate (Plate 2), two fold serial dilutions of drug in DMEM were made (10 concentrations). 100µl of drug dilutions from plate 2 were added to wells 3‐12 in plate 1. 100ul DMEM was added in well 2 which served as “cells alone” control i.e. without drugs. After 70hrs incubation, 50µl of 2.5mg/ml MTT (Sigma) in PBS added to each well. Plate was

RESULTS ‐ Chapter II

further incubated at 370C for 2 hours. Media was removed from all wells with multichannel pipette without disturbing the adherent cells. 200µl DMSO was added to all wells and the MTT formazan crystals were dissolved by pipetting up and down. Absorbance was read on Synergy H1 96 well plate reader at 550nm with reference filter of 620nm. Absorbance values were processed in Graphpad Prism as in section 1.1 above.

1.3 Glucose dependence of inhibitor toxicity

For experiment described in Section 2.5 and Fig. 6, procyclic form trypanosomes grown in two versions of SDM79 medium supplemented with 10% FCS (Sigma) were used. Complete SDM79 medium which contains 10mM glucose corresponds to ‘Glucose containing’ or ‘+ Glucose’ medium. For Glucose‐free ‘‐ Glucose’ medium, glucose was omitted during the preparation of SDM79 medium and also supplemented with 50mM N‐Acetyl Glucosamine

(Sigma) to prevent uptake of residual glucose from FCS. IC50 of inhibitors in SDM79 with or without glucose was estimated by resazurin based assay as described in Chapter II Section 1.1 above.

1.4 ATP Assay and microscopy

Bloodstream form trypanosomes were treated with inhibitor at IC50 concentration or equivalent amount of DMSO. Live cells were manually counted using Neubauer chamber at different time points. Equal numbers of control or drug treated cells were harvested. Extraction and quantification of ATP using CellTiter‐Glo reagent was performed as described in Chapter I publication ‐ Material and methods (Kalel et al., 2015)

For microscopy, labelling of glycosomes in procyclic form trypanosomes with FITC‐PTS1 (Fig. 8) was performed as described in (Lin et al., 2013).

RESULTS ‐ Chapter II

2. Results

2.1 in vitro and bioactivity assays of PEX5‐PEX14 inhibitors (schemes)

Structural basis for interaction of PEX5 with N‐terminus of PEX14 in humans has been determined by solving the 3D NMR structure of the complex of PEX5 derived peptide containing one of the di‐aromatic pentapeptide motif and PEX14 N‐terminal fragment (Neufeld et al., 2009). Homology modeling of Trypanosomal PEX5 peptide and PEX14‐NT complex revealed the differences between PEX5 binding sites in TbPEX14 (personal communication, Prof. Michael Sattler). This model was used to identify scaffolds which could bind to TbPEX14 at the PEX5 binding site. Scaffolds that were commercially available were chosen for initiating in vitro studies (Helmholtz Zetrum Munich).

Interaction between PEX5 peptide and Pex14‐NT has been studied using various techniques such as Surface Plasmin Resonance, NMR titration, peptide scan, fluorescence titration etc. (Schliebs et al., 1999; Saidowsky et al., 2001; Neufeld et al., 2009; Neuhaus et al., 2014). Binding affinities of TbPEX14‐NT to different Trypanosoma or human PEX5 peptides were tested in vitro. Human PEX5 peptide containing a Wxxx(F/Y) motif exhibited highest affinity to TbPEX14‐NT, therefore it was chosen as model binding partner of TbPEX14 for the identification of novel inhibitors which could disrupt TbPEX5‐TbPEX14 interaction with very high affinity. For high throughput discovery of small molecules disrupting TbPEX14‐TbPEX5 interaction, AlphaScreen technology was used (Bielefeld‐Sevigny, 2009). AlphaScreen is “Amplified Luminescent Proximity Homogeneous Assay” system which can be used to screen inhibitors with high sensitivity. The assay comprised of two types of beads, donor beads and acceptor beads (Fig. 1A left). Donor beads comprised of streptavidin coated beads bound to Biotin labelled PEX5 derived peptide containing a di‐aromatic motif. Acceptor beads were coated with Anti‐His antibody which were bound to His6 tagged TbPEX14 N‐terminus. Donor beads can be excited at 680nm wavelength which leads to formation of oxygen‐species which is short lived. When the donor beads are mixed together, PEX5‐PEX14 interaction brings these two beads in very close proximity, where the oxygen‐species is accepted by acceptor bead leading to emission at 615 nm. This enables accurately quantification of the PEX5‐PEX14 interaction by spectrophotometry. Addition of

RESULTS ‐ Chapter II

small molecules which can bind to TbPEX14 at the same site where PEX5 binds, it disrupts the Donor‐Acceptor bead proximity leading to reduction or absence of emission at 615 nm.

Fig. 1 in vitro and bioactivity assays of PEX5‐PEX14 inhibitors A. AlphaScreen assay scheme (Left panel) ‐ Streptavidin coated donor beads bound to Biotin‐PEX5 peptide and Anti‐His acceptor beads bound to His‐tagged PEX14‐NT are mixed. Donor beads are excited at 680nm which emit Oxygen species that are only accepted by donor beads in close vicinity, leading to emission at 615nm. Small molecule mediated disruption of PEX5‐PEX14 interaction lowers or abolishes the emission at 615nm. Right panel shows an example of AlphaScreen assay. AlphaScreen signal (ALS) is plotted against varying concentrations (µM, Logarithmic scale) of test compound giving a dose‐response curve. B. Anti‐trypanosomal assay (scheme) ‐ Bloodstream form T. brucei brucei cells are treated with varying concentrations of the compounds (quadruplicates) in 96‐well plate for 66 hours. Resazurin dye is added and after 6 hours the change in color is measured with 96‐ well plate reader. C. Human cell cytotoxicity assay (scheme) ‐ HEK cells are treated with

RESULTS ‐ Chapter II

compounds for 70 hours and MTT dye is added. After 2 hours, MTT reduced to purple formazan by live cells is quantified on 96 well plate reader (for details see Methods section) Fig.1A Right panel provided by Grzegorz Popowicz (Helmholtz Zentrum München)

Representative AlphaScreen assay result is shown in Fig.1A right panel. AlphaScreen Signal (ALS) is plotted against increasing concentration of different compounds (Log scale, micromolar) which were synthesized at Helmholtz Zentrum München. Inhibition of PEX5‐ PEX14 interaction is seen for several compounds as the AlphaScreen signal is highly reduced

at highest concentration testes (100µM). From such graphs it is possible to calculate EC50 i.e. concentration of compound which reduces the signal by 50%. Over 100 compounds were synthesized by structure‐based design and EC50 concentrations were determined using AlphaScreen.

In this thesis, compounds showing promising activities in AlphaScreen were tested for anti‐ trypanosomal activity against bloodstream form trypanosomes and also tested for cytotoxicity against human cell lines. For anti‐trypanosomal assays, we used 96 well plate resazurin dye based method (Fig. 1B). Resazurin is a blue colored dye which is metabolized by live cells to fluorescent pink colored product which can be easily measured on 96‐well plate reader. Different concentrations of a compound (2 fold dilution series, 10 concentrations, in quadruplicates) were incubated with trypanosomes for 66 hours after which resazurin solution was added to wells and further incubated for 6 hours to develop the pink color by live cells. As shown in the (Fig. 1B right), first two wells in each lane are control wells. First well contains no trypanosomes and remains blue after incubation which is also expected at high drug concentration which kills parasites completely (last wells; correspond to 100% growth inhibition). In second well which contains trypanosomes but no drug, resazurin is completely turned pink and corresponds to 0% growth inhibition. The fluorescence is read on 96‐well plate reader and the fluorescence values are transformed into percent growth inhibition.

Since both human and Trypanosoma PEX5‐PEX14 interaction is based on similar binding site, compounds were also tested on cultured human cells to eliminate compounds which might also affect human peroxisomes or are toxic to human cells in non‐specific manner. Similar to anti‐trypanosomal assay, cytotoxicity against human cell lines was estimated using 96‐well

RESULTS ‐ Chapter II

plate MTT dye based assay (Fig. 1C). MTT is yellow colored tetrazolium dye (3‐[4,5‐ dimethylthiazol‐2‐yl]‐2,5‐diphenyltetrazolium bromide) which is reduced by living cells into purple colored insoluble formazan. Formazan crystals are then dissolved in DMSO and absorbance measured with 96‐well plate reader. Results of anti‐trypanosomal and human cell cytotoxicity assays are shown in Section 2.2 for four key compounds.

2.2 Anti‐trypanosomal activity and human cell cytotoxicity of inhibitors

During the stages of the inhibitors development, compound 5M was the earliest molecule found to active in AlphaScreen with in vitro EC50 ~100µM (structures of the compounds are shown in Fig. 3). Resazurin based anti‐trypanosomal assay with bloodstream form stage of T. brucei brucei was performed with compound 5M to obtain a dose‐response curve. Percent growth inhibition is plotted against increasing drug concentration (in Log scale). As seen in Fig. 2A, compound 5M caused 100% growth inhibition of the trypanosomes at 100µM (highest concentration tested). Since resazurin is an indicator of cell viability whose transformation to pink color also depends on growth rate of cells, the Y‐axis is denoted as percent growth inhibition instead of percent killing. However, complete killing of trypanosomes at 100µM was confirmed by visual inspection of plates on microscope before

adding resazurin dye. From the graph, IC50 values (concentration causing 50% growth

inhibition) were calculated. 5M had an IC50 of 21µM against trypanosomes which is lower than the EC50 obtained in AlphaScreen. This is not surprising, since even minor inhibition of glycosomes is toxic to trypanosomes due to mislocalisation of enzymes to the cytosol which consume ATP in absence of feedback inhibition. Similarly, cytotoxicity of compound 5M towards human cells was determined using MTT

assay. Fig. 2B shows that 5M was also toxic to human cells with IC50 88µM. Selectivity Index

(SI) is an indicator of drug safety. SI is defined as ratio between IC50 values for human cells and trypanosomes. For a compound to be good lead against neglected diseases, Selectivity index of minimum 10 is recommended (Ioset et al., 2009). Higher Selectivity Index corresponds to safer drug. Unfortunately compound 5M had Selectivity Index of only 4.1. Therefore 5M was modified with different side‐chains, and a series of 5M derivatives was synthesized. All derivatives were tested in AlphaScreen for PEX5‐PEX14 inhibition (Grzegorz Popowicz, Munich) and their anti‐trypanosomal activity was determined in this thesis

RESULTS ‐ Chapter II

Fig. 2 Anti‐trypanosomal and human cell cytotoxicity assays. A. Bloodstream form trypanosomes were treated with different compounds at increasing concentrations. Resazurin based assay results are plotted as percent growth inhibition against drug concentration (nanomolar, Logarithmic scale). B. HEK cells are treated with different compounds at increasing concentrations. MTT‐based assay results are plotted as percent growth inhibition against drug concentration (nanomolar, Logarithmic scale). IC50 is the concentration leading to 50% growth inhibition (shown as dotted line in A and B)

In one of the 5M derivatives, single aromatic ring in one of the sidechains of 5M were replaced with double aromatic ring (Fig. 3, Blue). By this change in the structure, toxicity of compound 5MA was improved 10 where dose‐response curve for 5MA (blue line) shifted

RESULTS ‐ Chapter II

towards left as compared to 5M (black line) (Fig. 2A). IC50 values decreased from 88uM (5M) to 2.45uM (5MA), indicating 5MA is more toxic to trypanosomes even at 10 times lower concentration than 5M. Unfortunately cytotoxicity assay on human cells showed that 5MA is also 10 times more toxic to human cells (Fig. 2B, blue line), with IC50 8.55uM. Therefore the selectivity index still remained poor (~3.5).

Fig. 3 Structure‐Activity Relationship (SAR) of inhibitors. Chemical evolution of the initial compound 5M into nanomolar active anti‐trypanosomal compound 5M‐III‐MAB‐NH2 with improved Selectivity Index (SI). Potency of compounds was determined against both bloodstream form Trypanosomes and human cells. SI (Selectivity

Index) is the ratio of IC50 against human cells and Trypanosomes.

An extensive fragment optimization of 5MA was performed to synthesize more derivatives. Majority of 5MA derivatives were active in low micromolar range but still exhibited high cytotoxicity to human cells (not shown). Replacement of two sidechains in 5MA to yield 5M‐ III‐MAB (Fig. 3 purple) turned out to be very successful. 5M‐III‐MAB retained anti‐

RESULTS ‐ Chapter II

trypanosomal activity in low micromolar range (IC50 5MA ‐ 2.45uM, 5M‐III‐MAB ‐ 3.4uM) (Fig. 2A purple line), while it was completely non‐toxic to human cells even at highest concentration tested (100µM). Selectivity index was highly improved from ~4 to more than 30 (Fig. 3 purple).

To become an attractive lead compound for HAT treatment, our goal was to obtain highly potent PEX5‐PEX14 inhibitor with anti‐trypanosomal activity in low nanomolar range. One of the major hurdles was that majority of compounds had very limited solubility. The effort to further improve compound 5M‐III‐MAB was tedious since majority of derivatives either lost the activity or were active still in low micromolar range. However a major landmark success was obtained with one derivative where hydroxyl group in sidechain of 5M‐III‐MAB (Fig. 3 purple) was replaced with primary amine to obtain 5M‐III‐MAB‐NH2 (Fig. 3 red). Amino compound was highly active on trypanosomes where the dose‐response curve shifted towards far left (Fig. 2A, red line). The amino compound displayed toxicity to human cells

(Fig. 2B, red line) as with previous parent compound 5MA. 5M‐III‐MAB‐NH2 has IC50 100nM for trypanosomes while 4.45uM on human cells (Fig. 3 red). Since the compound is highly active on trypanosomes than on human cells, the selectivity index is further improved to 45.

2.3 Correlation between in vitro and in vivo activities

It is important to validate that compounds developed to block PEX5‐PEX14 interaction in

vitro are also acting on the same target in trypanosomes. Correlation between in vitro EC50 and trypanosomal IC50 is one such indicator. In Fig. 4, IC50 values against trypanosomes are

plotted against in vitro EC50 values obtained in AlphaScreen assay. IC50 or EC50 values of each of the 23 compounds from 5M series are shown where both axes have scale of Log10 µM. r‐ square (r2) is square of the linear regression correlation coefficient which is 1 when then is tight correlation. For our inhibitors, we observed strong correlation of in vitro (AlphaScreen assay) and in vivo (cell based growth inhibition assay) activities with r2 of 0.87. This reflects that compounds which are active in vitro are also active in vivo, and if a derivative loses in vitro activity the in vivo activity is lost.

RESULTS ‐ Chapter II

Fig. 4 Correlation between AlphaScreen and Anti‐trypanosomal activity

IC50 concentrations (µM) obtained in resazurin based anti‐trypanosomal assays are plotted

against IC50 values (µM) from AlphaScreen based in vitro PEX5‐PEX14 inhibition assays. Both axes are in Logarithmic scale. 23 compounds of the 5M family are shown. r2 (square of the linear regression correlation coefficient) is 0.87. Experiment performed in cooperation with Grzegorz Popowicz (Helmholtz Zentrum München).

2.4 Inhibitors are also active on procyclic form trypanosomes

Both bloodstream (BSF) and procyclic form (PCF) trypanosomes harbor glycosomes. Glucose is the rich source for BSF trypanosomes in the host bloodstream. PCF trypanosomes in insect midgut however use amino acids as nutrient source due to scarcity of glucose. However in culture, PCF trypanosomes utilize glucose in addition to amino acids. Therefore we tested whether PEX5‐PEX14 inhibitors are also active on PCF trypanosomes in glucose containing culture medium. As expected, the inhibitors were also active on PCF parasites

(Fig. 5). 5M‐III‐MAB had an IC50 of ~19µM while 5M‐III‐MAB‐NH2 which was highly potent against BSF trypanosomes, was also toxic to PCF cells in nanomolar concentration (IC50 ~400nM). Activities of both compounds were 4‐5 less than the activity against BSF form.

RESULTS ‐ Chapter II

Fig. 5 PEX5‐PEX14 inhibitors are also active on procyclic form trypanosomes Procyclic form (PCF) trypanosomes were treated with increasing concentration of compounds for 72hours in 96 well plate resazurin based assay. Percent growth inhibition is plotted against the drug concentration (nanomolar, Logarithmic scale). IC50 is the concentration leading to 50% growth inhibition (shown as dotted line).

This was expected since PCF trypanosomes can also utilize amino acids as ATP source where mitochondrial TCA cycle is active so can survive better than BSF trypanosomes in which glycolysis is the sole ATP source and mitochondria are inactive. Procyclic from trypanosomes can be cultivated in different culture media where nutrient source can be easily manipulated. Since the amino compound was active in nanomolar range, we used to study toxicity of this compound in media containing or lacking glucose (Section 2.5).

2.5 Glucose is toxic in presence of inhibitors Inhibition of glycosome biogenesis is toxic to trypanosomes in presence of glucose due to accumulation of toxic glucose metabolites in the cytosol by unregulated phosphorylation of glucose by Hexokinase and phosphofructokinase which lack feedback inhibition. Trypanosomes with defective glycosomes can survive in absence of glucose provided that they have alternate source of energy. Therefore we used two kinds of culture media or PCF trypanosomes, glucose containing (+ Glucose) or glucose free (‐ Glucose). In glucose free

RESULTS ‐ Chapter II

medium, amino acids act as energy source for parasites. 96‐well plate resazurin assay was performed with selected inhibitors which were active in nanomolar range (Fig. 6). Blasticidin

is an antibiotic which prevents protein synthesis and was used as control. Blasticidin has IC50 9.9uM in glucose containing medium. While in glucose free medium, the dose‐response curve shifted towards left indicating that blasticidin is more toxic in the absence of glucose

(fig. 6 top left) and the IC50 reduced 2 fold to 4.7uM.

Fig. 6 Glucose dependent toxicity of PEX5‐PEX14 inhibitors Procyclic form trypanosomes growing in presence and absence of glucose were treated with different compounds at increasing concentrations. The results of resazurin based assay were plotted as percent growth inhibition against the drug concentration (Logarithmic scale). Blasticidin (control) is more toxic to trypanosomes in media lacking glucose (Top left). The PEX5‐PEX14 inhibitors (Top right and bottom panels) are more toxic to trypanosomes in presence of glucose as evident from the lower IC50 values in presence glucose. In media lacking glucose, the toxicity of inhibitors is reduced (green arrow).

Conversely, in case of PEX5‐PEX14 inhibitors the dose response curve shifted in opposite direction towards right in glucose free medium. The IC50 of inhibitors was 2 fold higher in

RESULTS ‐ Chapter II

glucose free medium. In glucose containing medium, the inhibitors are about 2 times more toxic to trypanosomes. This is in agreement with the expected phenotype where disruption of glycosome biogenesis makes glucose toxic to trypanosomes.

2.6 Cellular ATP levels are depleted by the inhibitor

Disruption of glycosome biogenesis is toxic to trypanosomes since mislocalised glycosomal enzymes like hexokinase and phosphofructokinase deplete cellular ATP levels by unregulated glucose phosphorylation. Therefore we analyzed the changes in ATP levels in bloodstream form trypanosomes treated with our PEX5‐PEX14 inhibitor.

Fig. 7 ATP levels are disrupted by the PEX5‐PEX14 inhibitor Bloodstream form (BSF) trypanosomes were treated with compound 5M‐III‐MAB or equivalent amount of DMSO for 25 hours. ATP content of equal number of cells at different time points was estimated using CellTiter‐Glo assay. The luminescence values of inhibitor treated cells (red) were normalized against DMSO Control (blue). At initial time points, the inhibitor treated cells have a higher ATP content (see discussion); while subsequent ATP depletion leads to cell death.

Bloodstream form trypanosomes were treated with 5M‐III‐MAB at IC50 concentration and equivalent amount of DMSO was added to control cultures. After different time intervals, live cells were counted manually using Neubauer chamber. ATP was extracted from Equal number of control and inhibitor treated cells and quantified using CellTiter‐Glo assay. As

RESULTS ‐ Chapter II

shown in Fig. 7, we found that initially the ATP levels in inhibitor treated cells (red line) were higher than control cells (blue line). The explanation for this unusual observation will be discussed later in Discussion section. However at later time points the ATP levels dropped in inhibitor treated cells which led to death of trypanosomes. This indirectly indicates that the PEX5‐PEX14 inhibitor kills trypanosomes by disrupting glycosome biogenesis.

2.7 Inhibitor disrupts the glycosomal protein import

We could show that the PEX5‐PEX14 inhibitors deplete cellular ATP levels (Section 2.6) and are toxic to trypanosomes in glucose dependent manner (Section 2.5). However to obtain a direct evidence that inhibitors disrupt glycosomal protein import of matrix proteins, it was necessary to demonstrate the mislocalisation of glycosomal protein upon treatment with the inhibitors. Digitonin fractionation of inhibitor treated Bloodstream form trypanosomes failed to demonstrate the mislocalisation of glycosomal proteins. The reasons for this negative result will be discussed later in Discussion section. Alternatively we used a FITC labelled PTS1 peptide which is reported as a glycosomal marker (Lin et al., 2013) which gives punctate glycosomal staining in procyclic form trypanosomes.

Fig. 8 PEX5‐PEX14 inhibitors disrupt glycosomal protein import Procyclic form trypanosomes were incubated with 80µM FITC‐labelled PTS1 peptide as glycosomal marker for 1 hour. In absence of inhibitor the peptide gives a punctate pattern characteristic of glycosomes. In presence of inhibitor 5M‐III‐MAB‐NH2, FITC‐PTS1 remains in the cytosol and appears as diffuse pattern, indicating disruption of glycosomal protein import by the inhibitor. Scale bar ‐ 5 µm, (DIC ‐ Differential Interference Contrast).

RESULTS ‐ Chapter II

As reported before, FITC‐PTS1 gave punctate labelling in PCF trypanosomes in absence of inhibitor (control) indicating that glycosomal import is normal (Fig. 8 left). However when trypanosomes were incubated with FITC‐PTS1 and inhibitor 5M‐III‐MAB‐NH2 together, the FITC‐PTS1 peptide accumulated in the cytosol and no punctate pattern similar to control was seen (Fig. 8 middle and right panel). Besides the cytosolic accumulation, the single spot of FITC‐PTS1 seen in the inhibitor treated cells likely represent the fraction of peptide trapped in endocytic vesicle. These observations indicated that the PEX5‐PEX14 inhibitors disrupt glycosomal protein import.

2.8 Inhibitor reduces parasite load in mouse model of trypanosomiasis

5M‐III‐MAB‐NH2 was highly potent against bloodstream form T. brucei brucei in culture (Section 2.2) which is a model laboratory strain that infects cattle. As next step in drug development, the activity of this compound was tested against clinically relevant species of trypanosomes T. brucei rhodesiense by our collaborators at Swiss TPH, Basel. We found that the inhibitor 5M‐III‐MAB‐NH2 was actually four times more potent against T. b. rhodesiense

(IC50 23nM) than T. b. brucei (IC50 100nM). Melarsoprol in the same assay has IC50 of 5nM. Melarsoprol is highly active against trypanosomes but due to its severe side effects, it is only used in rare cases of stage 2 cases of Human African Trypanosomiasis.

Encouraged by this positive result, we further determined the pharmacokinetic properties of 5M‐III‐MAB‐NH2 compound (Bienta, Ukraine) before proceeding to animal studies. Toxicological studies indicated that the inhibitor showed no signs of severe toxicity at doses upto 100mg/kg in mice. Further pharmacokinetic studies in mice at 5mg/kg dose showed that the inhibitor is orally active since ~21‐27% of the compound was orally bioavailable. In vivo half‐life in mice varied depending on the route administration (Intravenous ‐ 156 min., intraperitoneal ‐ 228 minutes, oral ‐ 281 minutes). Peak plasma concentration reached in

mice was 57nM which is well above the IC50 against T. b. rhodesiense (23nM). However the major limitation was that the Vd (volume of distribution) was very high 22 liters/kg, indicating that the compound was rapidly distributed into tissues. Despite this limitation, the compound was tested in mouse model of Human African Trypanosomiasis by our collaborators at TPH, Basel. Mice infected with T. b. rhodesiense were treated with 15 or

RESULTS ‐ Chapter II

30mg/kg doses of the inhibitor administered intraperitoneally (2 doses/day, 4 day treatment). Parasitaemia in mice blood on 2nd day after the treatment was semi‐quantified by counting trypanosomes in blood smears. The values are represented in Fig.9 on arbitrary scale. It is evident that at 30mg/kg dose, the parasite load in animal blood was significantly reduced as compares to control. This shows that the PEX5‐PEX14 inhibitor can partially cure the parasite load despite its unfavorable pharmacokinetic property.

Fig. 9 Partial in vivo efficacy of PEX5‐PEX14 inhibitor in trypanosomiasis mouse model Mice infected with T. brucei rhodesiense were treated with 5M‐III‐MAB‐NH2 (two doses of 15mg/kg or 30mg/kg per day) for 4 days. Parasitaemia in blood on 6th day was visually semi‐ quantified and plotted as arbitrary units. Partial reduction of parasite load was observed at 30mg/kg dose despite the suboptimal pharmacokinetic (PK) properties of the compound. Experiment performed in cooperation with Pascal Mäser (Swiss Tropical and Public Health Institute, Basel) and Grzegorz Popowicz (Helmholtz Zentrum München).

DISCUSSION

Discussion

Vector‐borne infections of Trypanosoma and Leishmania occur in most vulnerable population of low and middle income countries of the world. Human African trypanosomiasis (Sleeping sickness) and visceral leishmaniasis (kala‐azar) are lethal if left untreated. Apart from mortality, disability‐associated loss of human productivity and economic loss of livestock animals results in the endless cycle of poverty.

Human African trypanosomiasis is endemic in 37 countries. Due to sustained disease control efforts, the reported number of cases of sleeping sickness has dropped below annual 10000 cases. However 70 million people still live at the risk infection. Furthermore, lack of complete surveillance in some areas and sporadic outbreak of the infections observed in the past indicate that the actual number of cases is likely higher (Simarro et al., 2011). Vaccines against sleeping sickness are not possible due to unique property of the antigenic variation by trypanosomes. Currently used drugs have limitations like severe toxicity in case of Melarsoprol or lack of activity of Suramin in second stage of the disease. All current therapies of sleeping sickness also require course of repeated injections over weeks which is difficult to realize in remote areas lacking professional healthcare and also leads to poor compliance to the therapy by patients. Glycosomes are the unique feature of trypanosomatids which partly compartmentalize glycolytic cycle inside peroxisomes. This peculiar glycosomal compartmentation is essential for the parasite due to lack of feedback inhibition in trypanosomal glycolytic enzymes (Bakker et al., 2000; Furuya et al., 2002). During human infection, trypanosomes rely on glycolysis as the sole source of ATP in glucose rich bloodstream of host. Disruption of glycosome biogenesis is toxic to trypanosomes since knockdown of proteins involved in glycosome biogenesis is proven to be lethal in bloodstream form stage. Therefore the aim of this thesis was to identify currently unknown components involved in glycosome biogenesis which can serve as new drug targets. Furthermore, the work was also aimed at designing small molecule inhibitors of glycosome biogenesis using the current knowledge of structures of proteins mediating glycosomal protein import.

DISCUSSION

Chapter I

Glycosome membrane biogenesis in trypanosomes ‐ bridging the gaps In yeast, 32 peroxin proteins (PEX) involved in peroxisome biogenesis have been identified so far, most of which are restricted to yeast species and some are redundant in function. However in humans and plants, 14 PEX proteins have been identified out of which 10 mediate matrix protein import and remaining 4 are involved in membrane protein import, de novo formation or elongation (Nagotu and Kalel et al., 2012). Based on sequence similarity to known PEX proteins, 11 PEX homologs have been identified in trypanosomes most of which are involved in matrix protein import. On the other hand, of three proteins PEX3, PEX16 and PEX19 which are required for peroxisomal membrane protein import and de novo formation, only PEX19 homolog has been so far identified in trypanosomes. Therefore it is poorly understood how glycosomes import membrane proteins and in addition the mechanism by which new glycosomes form remain unknown. Aim of the thesis was to identify the potential PEX3 or PEX16 homologs in trypanosomes. Trypanosomes diverged very early from the eukaryotic lineage and therefore the trypanosomal PEX proteins show low degree of sequence similarity to human proteins (Gualdron‐Lopez et al., 2012). This makes it difficult to identify the currently unknown peroxins in trypanosomes. But it is also an opportunity since the proteins are not very conserved between human and trypanosomes; trypanosomal peroxins provide an attractive drug target to disrupt glycosome biogenesis for developing new therapies against the parasite.

Bioinformatic analysis using BLAST search of trypanosomal proteins revealed similarity to known PEX16 proteins. No reliable PEX3 homolog could be identified in trypanosomes using this approach. A single trypanosomal protein sequence was found to contain PEX16 domain (Chapter 1 Fig. 1A and 1B). All species of Trypanosomatid parasites possess single orthologous protein containing a PEX16 domain and still annotated as hypothetical protein in the sequence database. Pfam is a database of protein families based on multiple sequence alignments and hidden Markov models (HMMs). Therefore identified candidate proteins need to be experimentally validated. In this thesis, the candidate trypanosomal PEX16 protein was demonstrated to be an integral membrane protein which localized to the glycosomes (Chapter I

DISCUSSION

Fig. 1C and Fig. 2). When the expression of the candidate PEX16 protein was knocked down by RNA interference, the glycosome biogenesis was severely affected (Chapter I Fig. 3 to 7). These observations validated that the identified candidate protein is the trypanosomal PEX16. Similar to other peroxins involved in glycosome biogenesis, PEX16 knockdown was toxic to the bloodstream form trypanosomes and hence it provides a new drug target to disrupt glycosome biogenesis. PEX16 in mammals acts as an anchor for PEX3 at the peroxisomal membrane (Matsuzaki and Fujiki, 2008). Although trypanosomal PEX3 identification using bioinformatics approach failed, the trypanosome PEX16 identified in this study can be used as an anchor protein to identify potential PEX3 in trypanosomes. This will bridge the gap in understanding of glycosome membrane biogenesis in trypanosomes and also serve as new drug targets.

Disruption of glycosome biogenesis by knockdown of PEX16 expression. Function inactivating mutations in either of PEX3, PEX16 or PEX19 in human cells lead to complete loss of peroxisomes (Fujiki et al., 2012). Peroxisomal matrix proteins are completely mislocalised to the cytosol while membrane proteins are either mislocalised to other membranes or become unstable and get degraded. In this study, it was found that PEX16 expression knockdown in bloodstream form parasites resulted into a drastic reduction in glycosome number from ~65 to less than 10 in a cell (Chapter I Fig. 6). The efficiency of knockdown was 70% which is commonly observed in trypanosomes (Chen et al., 2003, Galland et al., 2007). It can be envisaged that complete knockdown of PEX16 would also result into complete loss of glycosomes. Similar reduction in glycosome number is reported for RNAi knockdown of PEX2, PEX11, GIM5 or PEX19 in trypanosomes (Guerra‐Giraldez et al., 2002; Lorenz et al., 1998; Voncken et al., 2003; Banerjee et al., 2005). Downregulation of plant PEX16 expression also results into reduction in number of peroxisomes (Nito et al., 2007).

In PEX16 knockdown trypanosomes containing very few glycosomes, endogenous glycosomal enzymes as well as the reporter proteins like GFP fused to PTS signal were mislocalised to the cytosol (Chapter I Fig. 4A and 4B). This is likely due to saturation of the import capacity of the fewer remaining glycosomes rather than any direct role of PEX16 in matrix protein import. On the other hand, mislocalisation of glycosomal membrane proteins was not observed. Both

DISCUSSION endogenous membrane proteins PEX11 and GIM5 still localized to the few remaining glycosomes (Chapter I Fig. 4C and 4D). It could be possible that a portion of glycosomal membrane proteins was mislocalised but below detection range in immunofluorescence microscopy. Use of GFP fusion to test the potential mislocalisation of glycosomal membrane proteins was avoided since their overexpression is known to affect glycosome morphology (Lorenz et al., 1998).

In mammalian cells, PEX16 knockdown did not affect targeting of membrane proteins PEX11 and PEX14 to peroxisomes but specifically led to mislocalisation of PEX3 (Matsuzaki and Fujiki 2008). This is expected since PEX16 acts as anchor specifically for PEX3‐PEX19 complex at peroxisomal membrane in mammalian cells. As mentioned before, PEX3 in trypanosomes has not been identified so far. Alternatively using heterologous expression, it can be tested whether human PEX3 can localize to glycosomes in trypanosomes, and whether it is mislocalised upon TbPEX16 knockdown. Such observations would support the existence of potential PEX3 in trypanosomes.

Conservation of targeting signal but divergence of PEX16 function. Trypanosoma as well as Leishmania PEX16 localized to glycosomes when expressed in trypanosomes. When the parasite PEX16 proteins were heterologously expressed in normal human or yeast cells, they could also localize to the peroxisomes either completely or at least partially (Chapter I Fig. S1 and Fig. S3). This suggests that the peroxisome targeting signal in parasite PEX16 can be recognized by the peroxisomal import machinery in humans and yeast, albeit with different efficiencies. Previously it was shown that the PEX19‐binding motifs required for targeting of membrane proteins to mammalian peroxisomes and trypanosomal glycosomes are conserved (Saveria et al., 2007). However both parasite PEX16 proteins could not rescue the function of defective human PEX16 in Zellweger patient cells (Chapter I Fig. S2). Human PEX16 is integral membrane protein with both N‐ and C‐terminus facing towards the cytosol. Peroxisome targeting signal in human PEX16 consists of positively charged amino acids in the N‐terminal region and the first downstream transmembrane domain (Honsho 2002). T. brucei PEX6 is also an integral membrane protein whose predicted topology is similar to human

DISCUSSION

PEX16. Yeast two‐hybrid studies in cooperation with Udaya Bhandari (Master’s thesis), showed that TbPEX16 N‐terminal region binds to TbPEX19 (Chapter I Fig. S4 A). Therefore similar to human PEX16, the glycosome or peroxisome targeting signal in TbPEX16 is also most likely in its N‐terminal region. The only known function inactivating mutation in human PEX16 which causes Zellweger syndrome occurs near C‐terminus of PEX16 (Honsho et al., 1998). Authors also showed that overexpression of only C‐terminal fragment of human PEX16 also interferes with the formation of new peroxisomes. Therefore it was proposed that the C‐terminal region of PEX16 is implicated during early stages of peroxisome formation i.e. de novo biogenesis. Failure of parasite PEX16 proteins in rescuing the function of mutant human PEX16, indicates that the C‐terminal regions of both proteins differ from each other functionally. Although very crucial, how PEX16‐CTD is involved in peroxisome biogenesis is still not known. Using yeast two‐hybrid studies, TbPEX13.1 was identified as first known direct binding partner of PEX16 during this work (Chapter I Fig. S4 B). Further work with PEX16‐CTD of humans and trypanosomes will be worth to get better insight into peroxisome biogenesis in humans and as drug target to block glycosome biogenesis in trypanosomes.

Evidences supporting de novo glycosome biogenesis Peroxisomes proliferate by growth and division via fission of pre‐existing peroxisomes or they can arise de novo from the Endoplasmic Reticulum. Each trypanosomal cell contains ~65 glycosomes and the parasite divides every ~6‐7 hours (Tetley and Vickerman 1991). How the glycosomes are formed and whether fission or de novo pathway is involved remains completely unknown. During the transition of bloodstream form trypanosomes to procyclic stage, the glycosomes are degraded via autophagy. In BSF stage, more than 90% of glycosomal matrix proteins are glycolytic enzymes (Aman et al. 1985). PCF trypanosomes in glucose limited environment of insect midgut degrade these “glycolytic” glycosomes by autophagy (Herman 2008). Therefore at least during BSF to PCF differentiation, de novo glycosome biogenesis likely exists. Fission appears to be not critical for maintaining glycosome number even in non‐ differentiating trypanosomes, since knockdown of single Dynamin like protein of trypanosomes

DISCUSSION

(TbDLP1) affected mitochondria and endocytic vesicles but not the glycosomes (Morgan et al. 2004, Chanez et al. 2006).

In this study, it was found that upon downregulation of PEX16, glycosome number was greatly reduced and the remaining few glycosomes were predominantly in anterior part of cell. Posterior half part of the cell was almost completely depleted of glycosomes with often a single glycosome visible in this region upon RNAi (Chapter I Fig. 7).

Fig. D1 ‐ Glycosomes likely from de novo. RNAi knockdown of PEX16 caused reduction in glycosome number and mislocalisation of matrix proteins to cytosol. Remaining glycosomes were mainly found in anterior part of the cell while posterior part was nearly devoid of glycosomes (left). Trypanosomal Sec16 is similar to human Sec16B which is required for ER‐exit of PEX16 (right top). TbSec16 localizes to a single ER‐exit site in trypanosomes in posterior part. Glycosomes may form de novo at ERES. Block in de novo formation could explain the PEX16 RNAi phenotype. The newly formed glycosomes may require transport to both parts of cells (right bottom). (The Sec16 domains scheme in top‐right panel is modified from Sealey‐Cardona et al., 2014).

Fluorescent pulse‐chase studies showed that delivery of PEX16 from ER to peroxisome requires Sec16B. There are two isoforms of Sec16 in humans, the longer Sec16A and shorter Sec16B isoform. A recent study reported that Sec16B but not Sec16A is implicated in exit of PEX16 from ER in mammalian cells (Yonekawa et al. 2011). Trypanosomes harbor single Sec16 isoform (TbSec16) which is similar to human Sec16B. TbSec16 localizes to a single ER exit site in posterior part of the cell near nucleus (Sealey‐Cardona et al., 2014). The faster depletion of

DISCUSSION glycosomes in the posterior part of the cell upon PEX16 RNAi can be explained by a model in which new glycosomes are formed de novo in a Pex16 dependent manner at the single ER‐exit site in posterior part of the cell (Fig. D1). The model needs to be validated by investigating the glycosome morphology and localization of PEX16 upon RNAi knockdown of TbSec16.

Although de novo pathway is predominant in mammalian cells, only after overexpression PEX16 can be detected in ER in normal cells. The authors argue that this could be due to efficient ER‐ exit of mammalian PEX16 in normal conditions (Kim et al., 2006). However in peroxisome deficient mammalian cells such as PEX19 or PEX3 defective cells, PEX16 accumulates in the ER. Another evidence which can support de novo pathway for glycosome biogenesis is to demonstrate the ER localization of Trypanosoma PEX16 in certain conditions. In wild‐type trypanosomes, ER localization of GFP tagged PEX16 was not observed. This could be due to efficient ER exit of PEX16 in normal cells similar to mammalian cells. Investigating whether PEX16 accumulates in ER upon knockdown of PEX19 needs to be tested. Nevertheless, ER localization of Trypanosoma PEX16 was seen when expressed in peroxisome deficient mammalian cells (PEX19 or PEX3 defective cells). While in yeast, GFP tagged TbPEX16 localized partially to ER even in wild‐type cells. A peculiar change in peroxisome morphology was seen in these cells in which all peroxisomes appear to be attached to the ER. This phenotype resembles overexpression of Pex30‐GFP in yeast where de novo biogenesis appears to be upregulated. These indirect evidences of Trypanosoma PEX16 being able to localize to the ER, the peculiar effect of PEX16 RNAi on glycosome distribution and dispensable nature of fission protein TbDLP1 for glycosomes, all are in favor of de novo glycosome biogenesis model.

If indeed de novo pathway is the only mechanism in trypanosomes for glycosome formation, then the glycosomes which are newly formed at the ER‐exit site need to be transported to the anterior and posterior part of the cell. This transport of glycosomes must be regulated to prevent accumulation of glycosomes in one part of cell. In next section, the evidences supporting existence of such mechanism are provided.

DISCUSSION

Evidence for regulated glycosome distribution Long range and short range movement of peroxisomes along the microtubule cytoskeleton is known to occur in human cells (Bharti et al., 2011). In yeast, transport of peroxisomes on actin cytoskeleton is required to transfer peroxisomes from mother to daughter cells (Fagarasanu et al., 2006). A single trypanosome contains more than 60 glycosomes uniformly spread allover but densly packed in the long‐slender cell. Although it appears that transport of glycosomes in a trypanosome cell might not be required as in human and yeast cells, the results obtained in this study suggest that glycoosme transport occurs within a trypanosome cell (Fig. D2).

Fig. D2 ‐ Evidence supporting regulated glycosome distribution. Glycosomes are uniformly distributed in the slender trypanosome (middle). Glycosome number is reduced upon PEX16 RNAi in which remaining glycosomes are mainly in anterior part of the trypanosome (left). Overexpression of GFP‐PEX16 leads to aggregation of glycosomes which accumulates in posterior part of cell. (+ or ‐ ends indicate polarity of cell body microtubule cytoskeleton)

The knockdown of PEX16 led to a reduction in glycosome number and remaining few glycosomes were seen in anterior part of the cell (Chapter I Fig. 7). Conversely, overexpression of N‐terminally GFP tagged TbPEX16 led to aggregation of glycosomes and the aggregate accumulated in the posterior part of the cell (Chapter I Fig. S5). The differential effects of

DISCUSSION downregulation and overexpression of PEX16 on glycosome distribution suggest that in normal cells a mechanism is needed to distribute glycosomes evenly. PEX16 could be directly or indirectly involved in this mechanism. In yeast, Inp1 mediates retention of peroxisomes in mother cell while Inp2 moves them to daughter cells (Knoblach and Rachubinski, 2015). Both Inp1 and Inp2 homologs appear to be absent in trypansomes.

Trypanosome is a highly polarized cell where cytoskeleton is highly organized along the extended cell body. The pellicular microtubules are organized minus to plus end from anterior to posterior part of the cell body. In previous section, evidences supporting de novo pathway for glycosome biogenesis were provided. If the de novo glycosome biogenesis occura at ER‐exit site in posterior part, glycosomes have to be transported to both ends of the cell body. This will require association of glycosomes with both dyneins and kinesins which mediate minus and plus end directed movement along microtubules.

Unique motility defect and likely consequences PEX16 RNAi led to mislocalisation of glycosomal enzymes to the cytosol including hexokinase and phosphofructokinase. Both these enzymes lack the property of feedback regulation which is common in glycolytic enzymes in other organisms. As a consequence these enzymes perform unregulated glucose phosphorylation in cytosol of trypanosomes by consuming the cellular ATP pool and accumulate the phosphorylated metabolites to toxic levels which lead to death of trypanosomes (Bakker et al., 2000; Furuya et al., 2002). Accordingly, drastic reduction in the cellular ATP levels in trypanosomes was observed upon PEX16 RNAi (Chapter I Fig. 8A). Interestingly a fraction (~5%) of cells exhibited abnormal flagellar motility. The direction of motility was opposite to that of normal cells (Chapter I Fig. 8B). Normally the trypanosome flagellum beat starts at the tip and extends along the cell body ending at flagellum base in the flagellar pocket (Fig. D3). This provides the forward propelling motility to trypanosomes (Hill 2003). However a population PEX16 RNAi cells showed loss of tip to base flagellar beating. This caused cells to move slowly but continuously in opposite direction. This study is the first report of such motility in bloodstream form trypanosomes in normal culture medium.

DISCUSSION

Forward motility of trypanosomes enables hydrodynamic sorting of host antibodies bound to trypanosome cell surface towards the flagellar pocket where they are internalized in and degraded (Engstler et al., 2007). This antibody clearance is a crucial mechanism by which trypanosomes evade destruction by host immune system. The cells which move in opposite‐

Fig. D3 ‐ Motility defect in PEX16 RNAi cells. In normal trypanosome, the flagellar beat starts at tip and ends at base in the flagellar pocket, giving forward motility to trypanosome in the direction of flagellar tip (left). This motility is required for antibody clearance by the parasite to evade host immune system. In PEX16 RNAi, 5% of cells exhibited loss of forward motility in which the tip‐to‐base flagellar beat was lacking. These cells move slowly but persistently in opposite direction due to abnormal base‐to‐tip beat (right).

‐direction as observed in this study will be defective in antibody clearance in human host. Hence the cells will be easily recognized by host immune system and cleared away. The unique abnormal motility observed in PEX16 knockdown gives first indication of a possible link between glycosomes and the directional motility. Hexokinase is predominantly localized to glycosomes but portion also localizes to flagella (Joice et al., 2012). The functional role of extra‐ glycosomal localization of hexokinase to the flagellum and the mechanism by which a PTS2 containing hexokinase is targeted to flagellum remains poorly understood. In this study it was shown that knockdown of PEX16 leads to mislocalisation of glycosomal enzymes including

DISCUSSION hexokinase to the cytosol. This mislocalised hexokinase may in turn disrupt the flagellar targeting of hexokinase or disrupt its function in flagella.

Aggregation of glycosomes ‐ possible explanation The aggregation of glycosomes in a trypanosome was seen upon overexpression of GFP‐ TbPEX16 but not with TbPEX16‐GFP (Chapter I Fig. S5). This indicates that the aggregation is caused when overexpressed PEX16 has C‐terminus free while N‐terminus is blocked by the bulky tag.

Fig. D4 ‐ Aggregation of glycosomes ‐ possible mechanism. Glycosomes aggregate in procyclic form trypanosomes overexpressing GFP‐PEX16 (left top) similar to peroxisome aggregation in plant cells overexpressing plant PEX16 (left bottom) (Lin et al. 2004). Yeast two‐hybrid showed that C‐terminal domain of Trypanosoma PEX16 interacts with PEX13.1. Inter‐glycosomal interactions between GFP‐PEX16 and PEX13.1 can provide explanation for aggregation of glycosomes.

Also in plant cells which overexpress plant PEX16 with N‐terminal GFP fusion (GFP‐PEX16), similar aggregation of peroxisomes is observed. In yeast two‐hybrid studies, glycosomal membrane protein TbPEX13.1 specifically interacted with C‐terminal domain of PEX16. Based on this interaction, a model is proposed which can explain the mechanism of glycosome aggregation (Fig. D4). In this model, PEX16 with bulky tag at its N‐terminus in the glycosome

DISCUSSION membrane exposes its C‐terminal domain (CTD) to the cytosol. The excess of free PEX16‐CTD at the membrane of one glycosome has access to interact with PEX13.1 in the membrane of other glycosome. Such inter‐organellar contacts between GFP‐PEX16 and PEX13.1 can explain the aggregation of glycosomes. As similar peroxisome aggregation is also observed in plant cells overexpressing GFP‐PEX16, whether such inter‐peroxisomal PEX16 and PEX13 contacts occur in plants needs to be investigated. In plants, PEX16 and PEX13 are shown to independently interact with another plant peroxisomal membrane protein DAU/APEM9 (Li et al., 2014) but direct interaction of plant PEX16 and PEX13 has not been studied so far.

The glycosome aggregate was predominantly seen in the posterior part of the trypanosomes. The results from this study provided evidence supporting de novo glycosome biogenesis which may occur in the posterior part of the cell where single ER‐exit site is located. Therefore it is most likely that the glycosomes which are newly formed in the posterior part of trypanosomes tend to aggregate in the same region and the transport of glycosomes to the other part of the cell is inhibited.

Chapter II

Choice of Target for inhibitor development Glycolysis is the sole source of ATP for trypanosomes in the bloodstream of infected mammalian host. Several studies are going on for the development of inhibitors of trypanosomal glycolytic enzymes such as hexokinase and phosphofructokinase (Hudock et al., 2006; Sharlow et al., 2010; Walsh et al., 2011). The unique compartmentation of first seven enzymes of glycolysis pathway in glycosomes including hexokinase and phosphofructokinase is essential to trypanosomes since these enzymes lack feedback regulation and cause toxicity if localized to the cytosol. Knockdown of PEX proteins causes the mislocalisation of glycosomal enzymes to the cytosol. Therefore disruption of glycosome biogenesis is thought to be an attractive drug target. Based on the knowledge of peroxisome biogenesis and glycosome biogenesis, PEX5‐PEX14 interaction was chosen as a target for development of inhibitors.

DISCUSSION

Docking of cargo loaded PEX5 at the peroxisome or glycosome membrane requires interaction of diaromatic motifs in PEX5 and N‐terminal domain of glycosomal membrane protein PEX14. This interaction is well characterized in humans as well as trypanosomes (Saidowsky et al., 2001; Choe et al., 2003). The major limitation of inhibitors against glycolytic enzymes is that the inhibitor must cross the glycosomal membrane to bind and inhibit the glycolytic enzymes. Furthermore glycolytic enzymes constitute more than 90% content of glycosomal matrix. Therefore the permeability barrier of glycosomal membrane and the abundance of glycolytic enzymes in bloodstream form trypanosomes are major obstacles for glycolytic enzyme inhibitors (Aman et al. 1985). On the other hand, the interaction of PEX5‐PEX14 chosen in this study occurs at the cytosolic face of glycosome membrane. Therefore the inhibitors of this interaction do not have additional permeability barrier like in case of glycolytic enzyme inhibitors. PEX5 is a cycling receptor for import of proteins inside peroxisomes or glycosomes. Therefore inhibition of PEX5 binding to its glycosomal receptor PEX14 with small molecules which competitively bind PEX14 at the same site will lead to blockade of protein import. Since even minor mislocalisation of glycosomal enzymes is deleterious to the parasite (Krazy and Michels, 2006), inhibition of even few PEX14 molecules would be enough to initiate the toxic effect on trypanosomes.

Structure‐based drug design Two different approaches can be used for discovery of potential drug candidates ‐ phenotypic screening or target‐based screening (Swinney and Anthony, 2011). Both approaches have certain advantages or disadvantages over each other. In phenotypic screening, large numbers of compounds are tested for their defined biological activity in cell‐based assays or using animal models. Such approach has been used in drug discovery against Trypanosomatid parasites. About 2 million compounds were tested in a high‐throughput screen to identify compounds that kill trypanosomes in culture (Pena et al. 2015, Jones and Avery 2013). Although a large number of compounds identified in such assays, the major limitation of phenotypic screening is that the molecular mechanism of action (MMOA) is not known. This hampers the optimization of the molecular properties of the compounds. On the other hand in target‐based approach, a library of compounds is screened against a defined target such as an enzyme. The candidates

DISCUSSION identified in target‐based screening may have excellent biochemical activity in vitro, but their biological activity in vivo is not known. The compound might fail to reach the target within cell or the target may not be relevant in the disease pathogenesis. In this study, PEX5‐PEX14 interaction was target for development of inhibitors to disrupt glycosome biogenesis. Glycosomal metabolism as well as glycosome biogenesis are well validated drug targets. The molecular mechanism of action (MMOA) of toxicity due to glycosome biogenesis inhibition is also characterized in detail in different studies (Bakker et al., 2000; Furuya et al., 2002; Kessler and Parsons, 2005; Haanstra et al., 2008). Glycolytic enzymes upon cytosolic mislocalisation make glucose toxic to trypanosomes by consuming cellular ATP pool and accumulating glucose metabolites to toxic levels. Interaction between PEX5 and PEX14 is well characterized in human, yeast and trypanosomes. Trypanosome PEX5 contains only two functional diaromatic motifs while in human PEX14 there are 8 such motifs. At sequence level, there are clear differences between human and Trypanosoma PEX14.

Fig. D5 ‐ Inhibitor binds to the PEX5 binding site in PEX14 A. Modelled structure of TbPEX14 – TbPEX5 pentapeptide complex based on the NMR structure of human complex. B. X‐ray crystal structure of TbPEX14‐NTD in complex with inhibitor 5‐MA. Figure provided by Grzegorz Popowicz.

Within target‐based screening approach, more rational approach ‘structure‐based design’ was used. Here the NMR structure of human PEX5 peptide in complex with PEX14‐NTD was used for homology modeling of Trypanosoma PEX5‐PEX14 which provided the starting point for designing molecules which can fit in PEX5 binding pocket PEX14. The competitive inhibitors showing activity in the in vitro PEX5‐PEX14 inhibition assays were also tested in biological

DISCUSSION activity assays on trypanosomes and human cells. High‐resolution crystal structures of the inhibitors in complex with Trypanosoma PEX14 were solved (Helmholtz Zentrum München) which confirmed that the inhibitor indeed binds to PEX14 at PEX5 binding site by competing it out (Fig. D5).

Activity guided inhibitor optimization The most common drawback of phenotypic screening is the limited ability to optimize the molecular properties of identified compound due to lack of knowledge about their target and mechanism of action.

Fig. D6 ‐ Biological activity guided inhibitor optimization. Structure‐based design, synthesis and in vitro assays generate several compounds. The in vivo screening (using cultured cells) step involves estimation of anti‐trypanosomal activity and human cell cytotoxicity. This step acts as a filter for eliminating non‐specific compounds and selects only those potent against trypanosomes. The selected compounds are again optimized with a cycle in this cascade. Inhibitor with highest potency and specificity is used for validation of the on‐target action in vivo (in trypanosomes) and provided for further pharmacokinetic and animal model studies.

DISCUSSION

The cascade used for identifying glycosome biogenesis inhibitors (Fig. D6) involved structure‐ based design of molecules which can fit in PEX5 binding pocket of PEX14, synthesis of these compounds and testing of competitive inhibitor activity in vitro. The in vivo screening performed in this study served as a benchmark for prioritizing small number of compounds which pass the criteria of promising trypanocidal activity and low human cell cytotoxicity (Chapter II Fig. 1 to 3). Compounds which showed undesirable activities such as toxicity to human cells were eliminated at this stage. The selected compounds were optimized by repeated rounds of compound modification, in vitro and in vivo screening.

The biological activity guided prioritization of selected compounds by eliminating the poor inhibitors in several rounds of the workflow led to rapid optimization of the inhibitors. Initial compound identified by in vitro assays had very poor anti‐trypanosomal activities and high human cell cytotoxicity. Within few rounds of activity‐guided optimization, a series of PEX5‐ PEX14 inhibitors was identified which was about 200 times potent against trypanosomes and 45 times less toxic to human cells (Chapter II Fig. 3). These high affinity specific PEX5‐PEX14 inhibitors also allowed validation of the target in vivo in this study. It was demonstrated that the inhibitor leads to disruption of glycosomal protein import, glucose toxicity to trypanosomes and ATP depletion (Chapter II Fig. 6 to 8). This validates that the compounds act on‐target in trypanosomes and the molecular mechanism of action is same as observed in knockdown of PEX proteins. The activity‐based selection of compounds at in vivo screening stage also aided in prioritizing very few compounds for the pharmacokinetic studies in mice and the animal model of trypanosomiasis. This reduced the unnecessary costs that would have incurred with poor inhibitors.

Disruption of glycosome biogenesis ‐ inhibitors versus PEX knockdown Knockdown of PEX proteins in trypanosomes including PEX5 and PEX14 lead to the mislocalisation of glycosomal enzymes to the cytosol. Similarly upon PEX16 knockdown glycosomal enzymes were mislocalised to the cytosol as a consequence of severe reduction in glycosome number. The mislocalisation of endogenous glycosomal proteins was demonstrated

DISCUSSION by microscopic as well as biochemical methods such as immunofluorescence microscopy and digitonin fractionation of trypanosomes upon PEX16 RNAi (Chapter I). In similar experiments using PEX5‐PEX14 inhibitors, significant mislocalisation of endogenous glycosomal proteins was not detected as it is seen in PEX14 knockdown. Both endogenous and GFP‐PTS1 reporter were punctate in the inhibitor treated cells similar to the control DMSO treated cells. But using FITC coupled PTS1 which is reported as glycosomal marker (Lin et al., 2013); it was shown that glycosomal import is disturbed by inhibitor treatment (Chapter II Fig. 8). There are key differences between disruption of glycosome biogenesis by knockdown of PEX14 (Furuya et al., 2002) or by inhibiting the PEX14‐PEX5 interaction with small molecule inhibitors. In case of knockdown, it takes several days before the phenotype is visible after induction of PEX14 RNAi. The PEX14 protein levels are reduced and therefore the docking protein PEX14 is absent or in low amounts at the glycosomal membrane. With inhibitor treatment, the compounds act faster within few hours unlike with PEX14 RNAi in which it takes several days before the mislocalisation is visible. Therefore it is possible that the trypanosome cells in which glycosomal protein import is blocked by the inhibitor, the cell death occurs very fast and thus mislocalisation of endogenous proteins could not be detected in significant amounts. The glycosomal reporter FITC‐PTS1 takes only one hour treatment to label the glycosomes. Therefore use of FITC‐PTS1 to demonstrate disruption of glycosome biogenesis by inhibitors appears to have worked better.

Moreover unlike in RNAi knockdown of PEX14 expression in which PEX14 protein is absent or in low amounts at glycosome membrane, inhibitors only block the PEX5‐binding site PEX14. Oligomerization of mammalian PEX14 has been shown before (Itoh and Fujiki, 2006). The peroxisomal importomer characterized in yeast is a high molecular weight complex containing PEX5 and PEX14 (Meinecke et al., 2010). Therefore it could be possible that the PEX5 along with the glycosomal proteins as cargo can still dock in the inhibitor treated cells, but the translocation inside glycosomes is blocked. Such scenario can provide explanation why endogenous proteins as well as GFP‐PTS1 reporter protein were not detected in significant amounts in the cytosol upon inhibitor treatment. can be investigated using techniques such as

DISCUSSION sensitivity to protease in digitonin treated trypanosomes. Biotin coupled inhibitors were tested to show that they act on‐target by specifically binding to TbPEX14 in trypanosomes. However due to insolubility of endogenous Trypanosoma PEX14 even in the harsh detergent conditions (upto 1% Triton‐X100 or NP‐40), the biotin‐inhibitor pulldown were unsuccessful. As an alternative to demonstrate on‐target action on PEX14, overexpression of PEX14‐NTD in trypanosomes can be tested to investigate whether the inhibitor toxicity is reduced. Attempts to use FITC coupled inhibitors to demonstrate that they reach PEX14 at glycosomes were also unsuccessful since such fluorescent conjugates of inhibitors accumulated in the flagellar pocket of trypanosomes. This indicates that coupling of small molecule inhibitors to fluorescent tags may interfere with their uptake into cells.

Blocking of PEX5 binding site in PEX14 is relevant not only for import of glycosomal matrix proteins. In humans, diaromatic motif in PEX19 similar to those in PEX5 also interacts with the same binding site in PEX14 (Neufeld et al. 2009). Moreover, also tubulin‐PEX14 interaction is mediated by diaromatic motifs in tubulin in human cells which is required for long‐range movement of peroxisomes (Bharti et al., 2011). The results obtained with PEX16 (Chapter I) suggested that glycosome transport is required to ensure proper distribution of glycosomes in a trypanosome. Therefore apart from glycosomal protein mislocalisation, it will be worth investigating if trypanosomal PEX19‐PEX14 interaction or tubulin‐PEX14 interaction is inhibited by the inhibitors which were developed to inhibit PEX5‐PEX14 interaction. For example PEX5‐ PEX14 inhibitor treatment of trypanosomes overexpression of GFP‐PEX16 should prevent their aggregation by accumulation.

Changes in ATP Levels upon inhibitor treatment Significant reduction in cellular ATP levels was seen upon knockdown of PEX16 expression (Chapter I Fig. 8A). However after inhibitor treatment, cellular ATP levels actually increased during first few hours. The ATP levels reduced at later time points which coincide with the trypanosome cell death observed (Chapter II Fig. 7). The increase in ATP levels at initial time points is not surprising since mislocalisation of minor amounts of glycosomal proteins to the

DISCUSSION cytosol is not toxic to trypanosomes till they produce phosphorylated glucose metabolites to toxic concentrations. The major fraction of glycolytic enzymes still inside glycosomes can support the normal ATP production with additional ATP produced by the minor amounts of enzymes mislocalised to the cytosol. But over the time of inhibitor treatment, as higher amounts of glycosomal enzymes are mislocalised to cytosol and the balance is disturbed. The mislocalised enzymes start consuming ATP and accumulating glucose metabolites reaching toxic concentrations. And hence the drop in ATP levels at later time points also coincides with the death of trypanosomes observed which was verified by manual counting of trypanosomes after the drug treatment.

Minimizing the effects on human peroxisomal protein import Since interaction of PEX5 to PEX14 involves same basic principles in human and yeast, several factors were considered to avoid the disruption of human PEX5‐PEX14 interaction by the inhibitors developed to disrupt the trypanosomal proteins. Human and Trypanosoma PEX14 differ in protein sequence including the differences in the PEX5 binding site. This factor was already considered by modeling the structure of Trypanosoma PEX5‐PEX14 complex to identify key differences in the PEX5 binding site and also validated by solving the crystal structures of Trypanosoma PEX14 bound to the inhibitors. During in vitro experiments, the inhibitors were also tested for binding to human PEX14 and the compounds which also bound human PEX14 were eliminated at in vitro screening stage itself. Further during this thesis, the compounds were tested on human cells to determine their cytotoxicity and effect on peroxisomal protein import. The inhibitor identified during the study with highest potency against trypanosomes is 45 times less toxic to human cells. Selectivity index of minimum 10 is recommended by DNDi for development of new lead compounds against the Trypanosomatid parasites (Ioset et al. 2009). The selectivity index of most active Trypanosoma PEX5‐PEX14 inhibitor is 45 (Chapter II Fig. 3), which indicates that the inhibitor will be safe for the therapy against trypanosomiasis. It was additionally verified that the peroxisomal import in human cells was not affected by the inhibitors. Human PEX5 contains 7 diaromatic motifs and an additional similar motif which interacts with PEX14 (Saidowsky et al., 2001; Neuhaus et al., 2014). On the other hand there are

DISCUSSION only two diaromatic motifs in Trypanosoma PEX5 which interact with PEX14 (Choe et al., 2003). Therefore inhibition of Trypanosoma PEX5‐PEX14 is much more sensitive to disruption than human counterparts. Peroxisomes are essential for neurological development during neonatal stage in humans since defects in PEX proteins are responsible for genetic disorders (Nagotu and Kalel et al., 2012, Review copy in appendix) where the children with peroxisome biogenesis disorder die within short period after birth. Therefore limited doses of Trypanosoma PEX5‐ PEX14 inhibitor to cure trypanosomiasis in adult patients are highly unlikely to have any undesirable effect. This is further supported by the fact that even long term peroxisomal dysfunction caused by the genetic defects can be cured with gene therapy (Cartier et al., 2009).

Future prospects The most promising PEX5‐PEX14 inhibitor identified during this study was also highly active against clinically relevant human infective strain T. brucei rhodesiense which causes African

Trypanosomiasis (tested by collaborators at Swiss TPH) (Chapter II Fig. 9). The IC50 of the inhibitor on this strain was ~23nM which comes close to the activity of currently used drug

Melarsoprol (IC50 ‐ 5nM) in the same assay. Melarsoprol is only effective in combination with other inhibitor Nifurtimox which is current combination therapy prescribed by WHO. Pharmacokinetic studies in mice show that our inhibitor 5M‐III‐MAB‐NH2 is also orally bioavailable which is preferred criteria for development of new drugs against trypanosomiasis. The half‐life of this inhibitor is 281 minutes when administered orally and 227 minutes upon intraperitoneal administration. The highest concentration reached in mice blood was more than

50nM which is well above the IC50 concentration required to kill trypanosomes. However the inhibitor has large volume of distribution in mice (Vd ~22 liters), which suggests that the inhibitor is rapidly distributed in tissues of mice. This explains why there was partial reduction of parasite load in blood of mice infected with trypanosomes upon treatment with the inhibitor. Further optimization of the inhibitor is needed to alter its pharmacokinetic properties while retaining its anti‐trypanosomal activity. Apart from novel drugs against trypanosomiasis, the PEX5‐PEX14 inhibitors will be of scientific importance to gain more insights into the mechanism of glycosomal protein import. Some inhibitors obtained during intermediate stages

DISCUSSION of optimization also bind to human PEX14 and even these non‐specific inhibitors offer us a valuable tool to study peroxisomal protein import in humans.

SUMMARY

Summary The protozoan parasites Trypanosoma and Leishmania infect humans as well as the livestock animals and against these infections no safe and effective drugs are currently available. Glycosomes are peroxisome‐like organelles that are essential for the parasites. Therefore disruption of the glycosome biogenesis is an attractive drug target. In this study, a protein essential for glycosome biogenesis was identified and structure‐based small molecule inhibitors of the glycosome biogenesis were designed and characterized.

Identification and functional characterization of Trypanosomatid PEX16 PEX3, PEX16 and PEX19 are essential proteins in different organisms for the peroxisomal membrane protein import and de novo peroxisome formation from the endoplasmic reticulum (ER). Of these three proteins, only PEX19 is known in trypanosomes. Using bioinformatics approach, we identified a candidate PEX16 homolog in Trypanosomatid parasites. T. brucei PEX16 is an integral membrane protein localized to the glycosomes. RNA interference (RNAi) knockdown of PEX16 expression led to disruption of glycosome biogenesis. The glycosome number was drastically reduced and the glycosomal enzymes were mislocalised to the cytosol. Consequently, a strong reduction in the cellular ATP levels and defects of the normal flagella‐driven motility was observed. The few remaining glycosomes were distributed unevenly in a trypanosome. Opposite effects on the glycosome distribution in a trypanosome were seen upon downregulation or overexpression of PEX16. These observations suggest that de novo pathway plays major role in the glycosome biogenesis and regulated transport of glycosomes likely exists in trypanosome. Upon heterologous expression, parasite PEX16 proteins localized to the peroxisomes in normal yeast and human cells, but failed to rescue the function of defective human PEX16 in patient cells. Due to the low sequence similarity and the functional differences with human protein, Trypanosomatid PEX16 provides a novel drug target for the specific disruption of the glycosome biogenesis.

Identification and characterization of the inhibitors of PEX5‐PEX14 interaction (Generated by structure‐based design) Di‐aromatic pentapeptide motifs in PEX5 interact with the well‐conserved N‐terminal domain of PEX14. Based on the known 3D structure of human complex, homology modeling of the Trypanosoma PEX5‐PEX14 complex revealed differences in the binding site. Structure‐ based design of small molecules, synthesis and in vitro inhibition studies (by collaborators in Helmholtz Zentrum München) provided an early stage competitive inhibitor of Trypanosoma PEX5‐PEX14 binding. During this thesis, a medium‐throughput assay was established for screening the activity of these inhibitors against trypanosomes and human

SUMMARY

cells. Using this assay, compounds with promising anti‐trypanosomal activity and low human cell cytotoxicity were identified and were further subjected to the chemical structure optimization. Multiple cycles of bio‐activity assays followed by modification of compound and in vitro inhibition tests were performed. More than 200 compounds were screened during activity‐guided optimization which rapidly yielded highly potent inhibitors with trypanocidal activity in nanomolar range and very low toxicity to human cells. Despite of some suboptimal pharmacokinetic properties, the treatment of trypanosome‐infected mice with the inhibitor led to the partial cure of parasite load in the blood. A direct correlation between in vitro PEX5‐PEX14 inhibition and anti‐trypanosomal activity was observed, which is an indicator of the compounds acting on target in vivo. The most active compounds allowed the validation of the action of the inhibitors on biogenesis of glycosomes (on‐ target). Successful inhibition of the glycosomal protein import was shown and corroborated by the inhibitor‐induced glucose toxicity and ATP depletion in trypanosomes.

ZUSAMMENFASSUNG

Zusammenfassung Die den Protozoen zugeordneten Parasiten Trypanosoma und Leishmania, gegen die bislang keine sichere und effektive Medikation verfügbar sind, infizieren sowohl den Menschen wie auch Tiere. Glykosomen sind peroxisomenartige Organellen, welche für den Parasiten essentiell sind. Daher ist die Störung der glykosomalen Biogenese ein attraktives Ziel für Medikamente. In dieser Arbeit wurde ein glykosomales Protein identifiziert, welches eine wichtige Funktion in der glykosomalen Biogenese der Parasiten besitzt. Außerdem wurden niedermolekularen Inhibitoren der Glykosomen‐Biogenese auf der Basis von Proteinstrukturdaten generiert und charakterisiert.

Identifikation und funktionelle Charakterisierung von trypanosomalem PEX16 PEX3, PEX16 und PEX19 sind in einer Vielzahl von Organismen für den Import von peroxisomalen Proteinen und die de novo Entstehung von Peroxisomen aus dem endoplasmatischen retikulum (ER) essentielle Proteine. Von diesen drei Proteinen ist bislang lediglich PEX19 in Trypanosomen bekannt. Durch Anwendung von bioinformatischen Methoden konnte ein Kandidat für das Homolog von PEX16 in Trypanosomen identifiziert werden. PEX16 in T. brucei ist ein integrales Membranprotein mit glykosomaler Lokalisation. Die Erniedrigung der Expressionsrate von PEX16 mittels RNA‐Interferenz (RNAi) führte zu einer gestörten Biogenese der Glykosomen. Ihre Anzahl war stark reduziert und glykosomale Proteine wiesen eine Fehllokalisation im Cytosol auf. Infolgedessen wurden eine deutliche Abnahme der zellulären ATP‐Menge sowie Defekte der normalen Flagellum‐getriebenen Bewegung beobachtet. Die wenigen verbleibenden Glykosomen waren nicht mehr gleichmäßig in der trypanosomalen Zelle verteilt. Gegensätzliche Effekte wurden bei der Herunterregulierung bzw. der Überexpression von PEX16 beobachtet. Diese Beobachtungen lassen vermuten, dass die de novo Biogenese von Glykosomen eine wichtige Stellung in Trypanosomen einnimmt und für Glykosomen ein regulierter Transportmechanismus existiert. Heterolog in humanen sowie Hefezellen exprimiertes trypanosomales PEX16 zeigte eine peroxisomale Lokalisierung, jedoch konnte ein PEX16‐Defekt in humanen Patientenzellen nicht komplementiert werden. Aufgrund der geringen Sequenzähnlichkeit und funktionalen Unterschieden zwischen der humanen und trypanosomalen Variante stellt PEX16 ein neues Ziel für Wirkstoffe dar, welche die glykosomale Biogenese hemmen.

Identifizierung und Charakterisierung von Inhibitoren der PEX5‐PEX14‐Interaktion (Strukturbasierte Generierung) Di‐aromatische Pentapeptid‐Motive aus PEX5 interagieren mit der stark konservierten N‐ terminalen Domäne von PEX14. Basierend auf der bekannten 3D Struktur des humanen Komplexes zeigte die Homologie‐Modellierung von Trypanosoma PEX5‐PEX14 Unterschiede

ZUSAMMENFASSUNG in den Bindebereichen. Strukturbasiertes Design, Synthese und in vitro Studien (durchgeführt von Kooperationspartnern des Helmholtz Zentrums in München) lieferten erste Inhibitoren der Trypanosoma PEX5‐PEX14 Interaktion. In dieser Arbeit wurde ein Mitteldurchsatz‐Selektionsverfahren für die Analyse der Wirkung der Inhibitoren gegen trypanosomale und humane Zellen etabliert. Hierdurch konnten vielversprechende Stoffe mit anti‐trypanosomaler Wirkung und geringer Cytotoxizität für humane Zellen identifiziert und im Weiteren für die chemisch‐strukturelle Optimierung verwendet werden. Auf diese Art und Weise wurden Bioaktivitätstests mit nachfolgenden Modifizierungen der Stoffe sowie in vitro Inhibierungstests in mehreren Durchgängen durchgeführt. Die so vollzogene Durchmusterung von über 200 Stoffen lieferte schnell sehr potente Inhibitoren mit trypanozider Aktivität im nanomolaren Bereich bei gleichzeitig sehr geringer Toxizität für humane Zellen. Trotz suboptimaler pharmakokinetischer Eigenschaften führte die Behandlung von Trypanosomen‐ infizierten Mäusen zu einem deutlichen Rückgang der Parasitenzahl im Blut. Es konnte eine direkte Korrelation zwischen der Inhibierung der PEX5‐ PEX14‐Interaktion in vitro und der anti‐trypanosomalen Wirkung der Stoffe festgestellt werden. Die Inhibitoren mit der höchsten Aktivität konnten verwendet werden, um den Einfluss dieser Wirkstoffe auf die Biogenese der Glykosomen („on‐target“) zu validieren. Die erfolgreiche Hemmung des glykosomalen Proteinimports konnte durch die Inhibitor‐ induzierte Toxizität von Glukose und den Verlust von ATP in Trypanosomen bestätigt werden.

REFERENCES

(2011). "Working to overcome the global impact of neglected tropical diseases: first WHO report on neglected tropical diseases ‐ Update."

(2015). Investing to overcome the global impact of neglected tropical diseases ‐ Third WHO report on neglected tropical diseases, WHO/Department of control of neglected tropical diseases.

Agrawal, G, et al. (2011). "Cell‐free sorting of peroxisomal membrane proteins from the endoplasmic reticulum." Proc Natl Acad Sci U S A 108(22): 9113‐8.

Alvar, J, et al. (2012). "Leishmaniasis worldwide and global estimates of its incidence." PLoS One 7(5): e35671.

Aman, RA, et al. (1985). "Cross‐linking of the enzymes in the glycosome of Trypanosoma brucei." J Biol Chem 260(11): 6966‐73.

Aranovich, A, et al. (2014). "PEX16 contributes to peroxisome maintenance by constantly trafficking PEX3 via the ER." J Cell Sci 127(Pt 17): 3675‐86.

Bakker, BM, et al. (2000). "Compartmentation protects trypanosomes from the dangerous design of glycolysis." Proceedings of the National Academy of Sciences of the United States of America 97(5): 2087‐2092.

Banerjee, SK, et al. (2005). "Identification of trypanosomatid PEX19: functional characterization reveals impact on cell growth and glycosome size and number." Mol Biochem Parasitol 142(1): 47‐55.

Bern, C (2015). "Chagas' Disease." N Engl J Med 373(5): 456‐66.

Bharti, P, et al. (2011). "PEX14 is required for microtubule‐based peroxisome motility in human cells." J Cell Sci 124(Pt 10): 1759‐68.

Bielefeld‐Sevigny, M (2009). "AlphaLISA immunoassay platform‐ the "no‐wash" high‐throughput alternative to ELISA." Assay Drug Dev Technol 7(1): 90‐2.

Birschmann, I, et al. (2003). "Pex15p of Saccharomyces cerevisiae provides a molecular basis for recruitment of the AAA peroxin Pex6p to peroxisomal membranes." Mol Biol Cell 14(6): 2226‐36.

Blum, JA, et al. (2008). "Cardiac involvement in African and American trypanosomiasis." Lancet Infect Dis 8(10): 631‐41.

Braverman, N, et al. (1997). "Human PEX7 encodes the peroxisomal PTS2 receptor and is responsible for rhizomelic chondrodysplasia punctata." Nature Genetics 15(4): 369‐376.

Braverman, N, et al. (1998). "An isoform of pex5p, the human PTS1 receptor, is required for the import of PTS2 proteins into peroxisomes." Hum Mol Genet 7(8): 1195‐205.

Brennand, A, et al. (2012). "Trypanosomes contain two highly different isoforms of peroxin

REFERENCES

PEX13 involved in glycosome biogenesis." FEBS Lett 586(13): 1765‐71.

Brun, R and Schonenberger (1979). "Cultivation and in vitro cloning or procyclic culture forms of Trypanosoma brucei in a semi‐defined medium. Short communication." Acta Trop 36(3): 289‐92.

Cartier, N, et al. (2009). "Hematopoietic stem cell gene therapy with a lentiviral vector in X‐linked adrenoleukodystrophy." Science 326(5954): 818‐23.

Chanez, AL, et al. (2006). "Ablation of the single dynamin of T. brucei blocks mitochondrial fission and endocytosis and leads to a precise cytokinesis arrest." J Cell Sci 119(Pt 14): 2968‐74.

Chen, Y, et al. (2003). "Development of RNA interference revertants in Trypanosoma brucei cell lines generated with a double stranded RNA expression construct driven by two opposing promoters." Mol Biochem Parasitol 126(2): 275‐9.

Chevray, PM and Nathans, D (1992). "Protein interaction cloning in yeast: identification of mammalian proteins that react with the leucine zipper of Jun." Proc Natl Acad Sci U S A 89(13): 5789‐93.

Choe, J, et al. (2003). "Analysis of the sequence motifs responsible for the interactions of peroxins 14 and 5, which are involved in glycosome biogenesis in Trypanosoma brucei." Biochemistry 42(37): 10915‐22.

Coley, AF, et al. (2011). "Glycolysis in the african trypanosome: targeting enzymes and their subcellular compartments for therapeutic development." Mol Biol Int 2011: 123702.

Cull, B, et al. (2011). "Glycosome turnover in Leishmania major is mediated by autophagy." Autophagy 10(12): 2143‐57.

de Hoop, MJ and Ab, G (1992). "Import of proteins into peroxisomes and other microbodies." Biochem J 286 (Pt 3): 657‐69.

Effelsberg, D, et al. (2015). "Role of Pex21p for piggyback import of Gpd1p and Pnc1p into peroxisomes of Saccharomyces cerevisiae." J Biol Chem.

Eitzen, GA, et al. (1997). "Enlarged peroxisomes are present in oleic acid‐grown Yarrowia lipolytica overexpressing the PEX16 gene encoding an intraperoxisomal peripheral membrane peroxin." J Cell Biol 137(6): 1265‐78.

El Magraoui, F, et al. (2012). "The RING‐type ubiquitin ligases Pex2p, Pex10p and Pex12p form a heteromeric complex that displays enhanced activity in an ubiquitin conjugating enzyme‐selective manner." Febs J 279(11): 2060‐70.

Engstler, M, et al. (2007). "Hydrodynamic flow‐mediated protein sorting on the cell surface of trypanosomes." Cell 131(3): 505‐15.

Faber, KN, et al. (1995). "The N‐terminus of amine oxidase of Hansenula polymorpha contains a peroxisomal targeting signal." FEBS Lett 357(2): 115‐20.

REFERENCES

Fagarasanu, A, et al. (2006). "The peroxisomal membrane protein Inp2p is the peroxisome‐specific receptor for the myosin V motor Myo2p of Saccharomyces cerevisiae." Dev Cell 10(5): 587‐600.

Fagarasanu, A, et al. (2010). "Molecular mechanisms of organelle inheritance: lessons from peroxisomes in yeast." Nat Rev Mol Cell Biol 11(9): 644‐54.

Fang, Y, et al. (2004). "PEX3 functions as a PEX19 docking factor in the import of class I peroxisomal membrane proteins." J Cell Biol 164(6): 863‐75.

Fransen, M, et al. (2002). "Analysis of mammalian peroxin interactions using a non‐transcription‐based bacterial two‐hybrid assay." Mol Cell Proteomics 1(3): 243‐52.

Fransen, M, et al. (2005). "Analysis of human Pex19p's domain structure by pentapeptide scanning mutagenesis." J Mol Biol 346(5): 1275‐86.

Fujiki, Y, et al. (2012). "Peroxisome biogenesis disorders: molecular basis for impaired peroxisomal membrane assembly: in metabolic functions and biogenesis of peroxisomes in health and disease." Biochim Biophys Acta 1822(9): 1337‐42.

Furuya, T, et al. (2002). "Glucose is toxic to glycosome‐deficient trypanosomes." Proc Natl Acad Sci U S A 99(22): 14177‐82.

Galland, N, et al. (2007). "Characterization of the role of the receptors PEX5 and PEX7 in the import of proteins into glycosomes of Trypanosoma brucei." Biochim Biophys Acta 1773(4): 521‐35.

Galland, N, et al. (2010). "An internal sequence targets Trypanosoma brucei triosephosphate isomerase to glycosomes." Mol Biochem Parasitol 171(1): 45‐9.

Gandre‐Babbe, S and van der Bliek, AM (2008). "The novel tail‐anchored membrane protein Mff controls mitochondrial and peroxisomal fission in mammalian cells." Mol Biol Cell 19(6): 2402‐12.

Goto, S, et al. (2011). "Arabidopsis ABERRANT PEROXISOME MORPHOLOGY9 is a peroxin that recruits the PEX1‐PEX6 complex to peroxisomes." Plant Cell 23(4): 1573‐87.

Gould, SJ, et al. (1989). "A conserved tripeptide sorts proteins to peroxisomes." J Cell Biol 108(5): 1657‐64.

Gould, SJ, et al. (1990). "Peroxisomal protein import is conserved between yeast, plants, insects and mammals." Embo J 9(1): 85‐90.

Grou, CP, et al. (2012). "Identification of ubiquitin‐specific protease 9X (USP9X) as a deubiquitinase acting on ubiquitin‐peroxin 5 (PEX5) thioester conjugate." J Biol Chem 287(16): 12815‐27.

Gualdron‐Lopez, M, et al. (2012). "When, how and why glycolysis became compartmentalised in the Kinetoplastea. A new look at an ancient organelle." Int J

REFERENCES

Parasitol 42(1): 1‐20.

Gualdron‐Lopez, M, et al. (2013). "Ubiquitination of the glycosomal matrix protein receptor PEX5 in Trypanosoma brucei by PEX4 displays novel features." Biochim Biophys Acta 1833(12): 3076‐92.

Gualdron‐Lopez, M and Michels, PA (2013). "Processing of the glycosomal matrix‐protein import receptor PEX5 of Trypanosoma brucei." Biochem Biophys Res Commun 431(1): 98‐103.

Guerra‐Giraldez, C, et al. (2002). "Compartmentation of enzymes in a microbody, the glycosome, is essential in Trypanosoma brucei." J Cell Sci 115(Pt 13): 2651‐8.

Guo, T, et al. (2003). "Peroxisome division in the yeast Yarrowia lipolytica is regulated by a signal from inside the peroxisome." J Cell Biol 162(7): 1255‐66.

Guo, T, et al. (2007). "A signal from inside the peroxisome initiates its division by promoting the remodeling of the peroxisomal membrane." J Cell Biol 177(2): 289‐303.

Haanstra, JR, et al. (2008). "Compartmentation prevents a lethal turbo‐explosion of glycolysis in trypanosomes." Proc Natl Acad Sci U S A 105(46): 17718‐23.

Halbach, A, et al. (2006). "Targeting of the tail‐anchored peroxisomal membrane proteins PEX26 and PEX15 occurs through C‐terminal PEX19‐binding sites." J Cell Sci 119(Pt 12): 2508‐17.

Herman, M, et al. (2008). "Turnover of glycosomes during life‐cycle differentiation of Trypanosoma brucei." Autophagy 4(3): 294‐308.

Hill, KL (2003). "Biology and mechanism of trypanosome cell motility." Eukaryot Cell 2(2): 200‐8.

Hirumi, H and Hirumi, K (1989). "Continuous cultivation of Trypanosoma brucei blood stream forms in a medium containing a low concentration of serum protein without feeder cell layers." J Parasitol 75(6): 985‐9.

Hoepfner, D, et al. (2005). "Contribution of the endoplasmic reticulum to peroxisome formation." Cell 122(1): 85‐95.

Honsho, M, et al. (1998). "Mutation in PEX16 is causal in the peroxisome‐deficient Zellweger syndrome of complementation group D." Am J Hum Genet 63(6): 1622‐30.

Honsho, M, et al. (2002). "The membrane biogenesis peroxin Pex16p. Topogenesis and functional roles in peroxisomal membrane assembly." J Biol Chem 277(46): 44513‐24.

Hua, R, et al. (2015). "Multiple Domains in PEX16 Mediate Its Trafficking and Recruitment of Peroxisomal Proteins to the ER." Traffic 16(8): 832‐52.

Hudock, MP, et al. (2006). "Inhibition of Trypanosoma cruzi hexokinase by

REFERENCES

bisphosphonates." J Med Chem 49(1): 215‐23.

Huhse, B, et al. (1998). "Pex17p of Saccharomyces cerevisiae is a novel peroxin and component of the peroxisomal protein translocation machinery." J Cell Biol 140(1): 49‐60.

Ioset JR, et al. (2009). Drug screening for kinetoplastids diseases: a training manual for screening in neglected diseases: 1–74.

Islinger, M, et al. (2009). "Hitchhiking of Cu/Zn superoxide dismutase to peroxisomes‐‐evidence for a natural piggyback import mechanism in mammals." Traffic 10(11): 1711‐21.

Itoh, R and Fujiki, Y (2006). "Functional domains and dynamic assembly of the peroxin Pex14p, the entry site of matrix proteins." J Biol Chem 281(15): 10196‐205.

Jensen, RE and Englund, PT (2012). "Network news: the replication of kinetoplast DNA." Annu Rev Microbiol 66: 473‐91.

Jiang, L, et al. (2015). "Peroxin Pex14p is the key component for coordinated autophagic degradation of mammalian peroxisomes by direct binding to LC3‐II." Genes Cells 20(1): 36‐49.

Joice, AC, et al. (2012). "Extra‐glycosomal localisation of Trypanosoma brucei hexokinase 2." Int J Parasitol 42(4): 401‐9.

Jones, AJ and Avery, VM (2013). "Whole‐organism high‐throughput screening against Trypanosoma brucei brucei." Expert Opin Drug Discov 8(5): 495‐507.

Jones, JM, et al. (2001). "Multiple distinct targeting signals in integral peroxisomal membrane proteins." J Cell Biol 153(6): 1141‐50.

Jones, JM, et al. (2004). "PEX19 is a predominantly cytosolic chaperone and import receptor for class 1 peroxisomal membrane proteins." J Cell Biol 164(1): 57‐67.

Kalel, VC, et al. (2015). "Identification and functional characterization of Trypanosoma brucei peroxin 16." Biochim Biophys Acta 1853(10 Pt A): 2326‐37.

Karnik, SK and Trelease, RN (2005). "Arabidopsis peroxin 16 coexists at steady state in peroxisomes and endoplasmic reticulum." Plant Physiol 138(4): 1967‐81.

Kessler, PS and Parsons, M (2005). "Probing the role of compartmentation of glycolysis in procyclic form Trypanosoma brucei: RNA interference studies of PEX14, hexokinase, and phosphofructokinase." J Biol Chem 280(10): 9030‐6.

Kim, PK, et al. (2006). "The origin and maintenance of mammalian peroxisomes involves a de novo PEX16‐dependent pathway from the ER." J Cell Biol 173(4): 521‐32.

Knoblach, B, et al. (2013). "An ER‐peroxisome tether exerts peroxisome population control in yeast." Embo J 32(18): 2439‐53.

REFERENCES

Knoblach, B and Rachubinski, RA (2015). "Sharing the cell's bounty ‐ organelle inheritance in yeast." J Cell Sci 128(4): 621‐30.

Knoops, K, et al. (2014). "Preperoxisomal vesicles can form in the absence of Pex3." J Cell Biol 204(5): 659‐68.

Krazy, H and Michels, PA (2006). "Identification and characterization of three peroxins‐‐PEX6, PEX10 and PEX12‐‐involved in glycosome biogenesis in Trypanosoma brucei." Biochim Biophys Acta 1763(1): 6‐17.

Kristjanson, P, et al. (1999). "Measuring the costs of African animal trypanosomosis, the potential benefits of control and returns to research." Agricultural Systems 59(1): 79‐98.

Kumar, A, et al. (2001). "An unexpected extended conformation for the third TPR motif of the peroxin PEX5 from Trypanosoma brucei." J Mol Biol 307(1): 271‐82.

Kurochkin, IV, et al. (2007). "Novel peroxisomal protease Tysnd1 processes PTS1‐ and PTS2‐containing enzymes involved in beta‐oxidation of fatty acids." Embo J 26(3): 835‐45.

Lam, SK, et al. (2010). "A vesicle carrier that mediates peroxisome protein traffic from the endoplasmic reticulum." Proc Natl Acad Sci U S A 107(50): 21523‐8.

Lee, JR, et al. (2006). "Cloning of two splice variants of the rice PTS1 receptor, OsPex5pL and OsPex5pS, and their functional characterization using pex5‐deficient yeast and Arabidopsis." Plant J 47(3): 457‐66.

Li, XR, et al. (2014). "Arabidopsis DAYU/ABERRANT PEROXISOME MORPHOLOGY9 is a key regulator of peroxisome biogenesis and plays critical roles during pollen maturation and germination in planta." Plant Cell 26(2): 619‐35.

Lin, S, et al. (2013). "Peptide‐targeted delivery of a pH sensor for quantitative measurements of intraglycosomal pH in live Trypanosoma brucei." Biochemistry 52(21): 3629‐37.

Lin, Y, et al. (1999). "The Pex16p homolog SSE1 and storage organelle formation in Arabidopsis seeds." Science 284(5412): 328‐30.

Lin, Y, et al. (2004). "The peroxisome deficient Arabidopsis mutant sse1 exhibits impaired fatty acid synthesis." Plant Physiol 135(2): 814‐27.

Lorenz, P, et al. (1998). "Elongation and clustering of glycosomes in Trypanosoma brucei overexpressing the glycosomal Pex11p." Embo J 17(13): 3542‐55.

Ma, C, et al. (2011). "Peroxisome assembly: matrix and membrane protein biogenesis." J Cell Biol 193(1): 7‐16.

Matsumoto, N, et al. (2003). "The pathogenic peroxin Pex26p recruits the Pex1p‐Pex6p AAA ATPase complexes to peroxisomes." Nat Cell Biol 5(5): 454‐60.

REFERENCES

Matsuzaki, T and Fujiki, Y (2008). "The peroxisomal membrane protein import receptor Pex3p is directly transported to peroxisomes by a novel Pex19p‐ and Pex16p‐dependent pathway." J Cell Biol 183(7): 1275‐86.

Matsuzono, Y, et al. (2006). "Functional domain mapping of peroxin Pex19p: interaction with Pex3p is essential for function and translocation." J Cell Sci 119(Pt 17): 3539‐50.

Meinecke, M, et al. (2010). "The peroxisomal importomer constitutes a large and highly dynamic pore." Nat Cell Biol 12(3): 273‐7.

Michels, PA, et al. (2005). "Peroxisomes, glyoxysomes and glycosomes (review)." Mol Membr Biol 22(1‐2): 133‐45.

Michels, PA, et al. (2006). "Metabolic functions of glycosomes in trypanosomatids." Biochim Biophys Acta 1763(12): 1463‐77.

Miyata, N, et al. (2012). "AWP1/ZFAND6 functions in Pex5 export by interacting with cys‐monoubiquitinated Pex5 and Pex6 AAA ATPase." Traffic 13(1): 168‐83.

Mogk, S, et al. (2014). "The lane to the brain: how African trypanosomes invade the CNS." Trends Parasitol 30(10): 470‐7.

Mony, BM, et al. (2014). "Genome‐wide dissection of the quorum sensing signalling pathway in Trypanosoma brucei." Nature 505(7485): 681‐5.

Morgan, GW, et al. (2004). "The single dynamin‐like protein of Trypanosoma brucei regulates mitochondrial division and is not required for endocytosis." J Biol Chem 279(11): 10692‐701.

Motley, AM and Hettema, EH (2007). "Yeast peroxisomes multiply by growth and division." J Cell Biol 178(3): 399‐410.

Nagotu, S, et al. (2012). "Molecular basis of peroxisomal biogenesis disorders caused by defects in peroxisomal matrix protein import." Biochim Biophys Acta 1822(9): 1326‐36.

Neufeld, C, et al. (2009). "Structural basis for competitive interactions of Pex14 with the import receptors Pex5 and Pex19." Embo J 28(6): 745‐54.

Neuhaus, A, et al. (2014). "A novel Pex14 protein‐interacting site of human Pex5 is critical for matrix protein import into peroxisomes." J Biol Chem 289(1): 437‐48.

Nito, K, et al. (2007). "Functional classification of Arabidopsis peroxisome biogenesis factors proposed from analyses of knockdown mutants." Plant Cell Physiol 48(6): 763‐74.

Oku, M and Sakai, Y (2015). "Pexophagy in yeasts." Biochim Biophys Acta.

Okumoto, K, et al. "Two proteases, trypsin domain‐containing 1 (Tysnd1) and peroxisomal lon protease (PsLon), cooperatively regulate fatty acid beta‐oxidation in peroxisomal matrix." J Biol Chem 286(52): 44367‐79.

REFERENCES

Opalinski, L, et al. (2011). "Membrane curvature during peroxisome fission requires Pex11." Embo J 30(1): 5‐16.

Opperdoes, FR and Szikora, JP (2006). "In silico prediction of the glycosomal enzymes of Leishmania major and trypanosomes." Mol Biochem Parasitol 147(2): 193‐206.

Pace, D (2014). "Leishmaniasis." J Infect 69 Suppl 1: S10‐8.

Pena, I, et al. (2015). "New compound sets identified from high throughput phenotypic screening against three kinetoplastid parasites: an open resource." Sci Rep 5: 8771.

Platta, HW, et al. (2005). "Functional role of the AAA peroxins in dislocation of the cycling PTS1 receptor back to the cytosol." Nat Cell Biol 7(8): 817‐22.

Purdue, PE, et al. (1998). "Pex18p and Pex21p, a novel pair of related peroxins essential for peroxisomal targeting by the PTS2 pathway." J Cell Biol 143(7): 1859‐69.

Reumann, S (2004). "Specification of the peroxisome targeting signals type 1 and type 2 of plant peroxisomes by bioinformatics analyses." Plant Physiol 135(2): 783‐800.

Rucktaschel, R, et al. (2009). "Farnesylation of pex19p is required for its structural integrity and function in peroxisome biogenesis." J Biol Chem 284(31): 20885‐96.

Rucktaschel, R, et al. (2011). "Protein import machineries of peroxisomes." Biochim Biophys Acta 1808(3): 892‐900.

Sacksteder, KA, et al. (2000). "PEX19 binds multiple peroxisomal membrane proteins, is predominantly cytoplasmic, and is required for peroxisome membrane synthesis." J Cell Biol 148(5): 931‐44.

Saidowsky, J, et al. (2001). "The di‐aromatic pentapeptide repeats of the human peroxisome import receptor PEX5 are separate high affinity binding sites for the peroxisomal membrane protein PEX14." J Biol Chem 276(37): 34524‐9.

Saveria, T, et al. (2007). "Conservation of PEX19‐binding motifs required for protein targeting to mammalian peroxisomal and trypanosome glycosomal membranes." Eukaryot Cell 6(8): 1439‐49.

Schliebs, W, et al. (1999). "Recombinant human peroxisomal targeting signal receptor PEX5. Structural basis for interaction of PEX5 with PEX14." J Biol Chem 274(9): 5666‐73.

Schmidt, F, et al. (2010). "The role of conserved PEX3 regions in PEX19‐binding and peroxisome biogenesis." Traffic 13(9): 1244‐60.

Schrader, M, et al. (2012). "Fission and proliferation of peroxisomes." Biochim Biophys Acta 1822(9): 1343‐57.

Schuldiner, M, et al. (2008). "The GET complex mediates insertion of tail‐anchored proteins into the ER membrane." Cell 134(4): 634‐45.

100

REFERENCES

Sealey‐Cardona, M, et al. (2014). "Sec16 determines the size and functioning of the Golgi in the protist parasite, Trypanosoma brucei." Traffic 15(6): 613‐29.

Sharlow, ER, et al. (2010). "A target‐based high throughput screen yields Trypanosoma brucei hexokinase small molecule inhibitors with antiparasitic activity." PLoS Negl Trop Dis 4(4): e659.

Simarro, PP, et al. (2011). "The human African trypanosomiasis control and surveillance programme of the World Health Organization 2000‐2009: the way forward." PLoS Negl Trop Dis 5(2): e1007.

Soukupova, M, et al. (1999). "Identification and characterization of the human peroxin PEX3." Eur J Cell Biol 78(6): 357‐74.

Souza, W (2008). "Electron microscopy of trypanosomes‐‐a historical view." Mem Inst Oswaldo Cruz 103(4): 313‐25.

Stein, K, et al. (2002). "Interactions of Pex7p and Pex18p/Pex21p with the peroxisomal docking machinery: implications for the first steps in PTS2 protein import." Mol Cell Biol 22(17): 6056‐69.

Subramani, S (1998). "Components involved in peroxisome import, biogenesis, proliferation, turnover, and movement." Physiol Rev 78(1): 171‐88.

Swinney, DC and Anthony, J (2011). "How were new medicines discovered?" Nat Rev Drug Discov 10(7): 507‐19.

Szoor, B, et al. (2014). "Evolution, dynamics and specialized functions of glycosomes in metabolism and development of trypanosomatids." Curr Opin Microbiol 22: 79‐87.

Tani, K, et al. (2011). "Dual function of Sec16B: Endoplasmic reticulum‐derived protein secretion and peroxisome biogenesis in mammalian cells." Cell Logist 1(4): 164‐167.

Tetley, L and Vickerman, K (1991). "The glycosomes of trypanosomes: number and distribution as revealed by electron spectroscopic imaging and 3‐D reconstruction." J Microsc 162(Pt 1): 83‐90.

Titorenko, VI, et al. (1998). "Pex20p of the yeast Yarrowia lipolytica is required for the oligomerization of thiolase in the cytosol and for its targeting to the peroxisome." J Cell Biol 142(2): 403‐20.

Van der Leij, I, et al. (1993). "PAS10 is a tetratricopeptide‐repeat protein that is essential for the import of most matrix proteins into peroxisomes of Saccharomyces cerevisiae." Proc Natl Acad Sci U S A 90(24): 11782‐6.

Verplaetse, E, et al. (2009). "Identification, characterization and essentiality of the unusual peroxin 13 from Trypanosoma brucei." Biochim Biophys Acta 1793(3): 516‐27.

Verplaetse, E, et al. (2012). "Studies on the organization of the docking complex involved in matrix protein import into glycosomes of Trypanosoma brucei." Biochem Biophys

101

REFERENCES

Res Commun 424(4): 781‐5.

Voncken, F, et al. (2003). "Depletion of GIM5 causes cellular fragility, a decreased glycosome number, and reduced levels of ether‐linked phospholipids in trypanosomes." J Biol Chem 278(37): 35299‐310.

Walsh, MJ, et al. (2011). "Identification of Selective Inhibitors of Phosphofructokinase as Lead Compounds Against Trypanosomiasis."

Wanders, RJ (2014). "Metabolic functions of peroxisomes in health and disease." Biochimie 98: 36‐44.

Williams, C, et al. (2012). "Insights into ubiquitin‐conjugating enzyme/ co‐activator interactions from the structure of the Pex4p:Pex22p complex." Embo J 31(2): 391‐402.

Williams, C, et al. (2015). "The membrane remodeling protein Pex11p activates the GTPase Dnm1p during peroxisomal fission." Proc Natl Acad Sci U S A 112(20): 6377‐82.

Yagita, Y, et al. (2013). "Tail‐anchored PEX26 targets peroxisomes via a PEX19‐dependent and TRC40‐independent class I pathway." J Cell Biol 200(5): 651‐66.

Yamashita, S, et al. (2014). "The membrane peroxin PEX3 induces peroxisome‐ubiquitination‐linked pexophagy." Autophagy 10(9): 1549‐64.

Yonekawa, S, et al. (2011). "Sec16B is involved in the endoplasmic reticulum export of the peroxisomal membrane biogenesis factor peroxin 16 (Pex16) in mammalian cells." Proc Natl Acad Sci U S A 108(31): 12746‐51.

Young, PG and Bartel, B (2015). "Pexophagy and peroxisomal protein turnover in plants." Biochim Biophys Acta.

102

APPENDIX

1. Abbreviations

AAA ATPase Associated with various cellular Activities BSF bloodstream form CTD C‐terminal domain DAPI 4′,6′‐diamidino‐2‐phenylindole DIC differential interference contrast EC50 effective concentration giving 50% response Eno enolase ER endoplasmic reticulum ERES ER‐exit site FITC Fluorescein isothiocyanate GAPDH glycosomal glyceraldehyde‐3 phosphate dehydrogenase GFP green fluorescent protein HK Hexokinase HsTRAP TNF receptor‐associated protein 1 (Human) IC50 50% Inhibitory concentration Lm Leishmania major 3‐(4,5‐dimethylthiazol‐2‐yl)‐2,5‐diphenyltetrazolium MTT bromide NMR nuclear magnetic resonance NTD N‐terminal domain NTDs neglected tropical diseases PCF Procyclic form PEX peroxins PFK phosphofructokinase PMP peroxisomal membrane protein PTS peroxisomal targeting signal RNAi RNA interference SAR Structure‐Activity relationship SI Selectivity Index Tb Trypanosoma brucei TbDLP1 T. brucei Dynamin like protein 1 TPR tetratricopeptide repeat

103

APPENDIX

2. Publications

2.1 List of publications

Paper : Identification and functional characterization of Trypanosoma brucei peroxin 16. Vishal C. Kalel, Wolfgang Schliebs, Ralf Erdmann Biochim Biophys Acta. 2015 Oct;1853(10 Pt A):2326‐37.

Review : Molecular basis of peroxisomal biogenesis disorders caused by defects in peroxisomal matrix protein import. Shirisha Nagotu*, Vishal C. Kalel*, Ralf Erdmann, , Harald W. Platta Biochim Biophys Acta. 2012 Sep;1822(9):1326‐36. * Joint first authors

Patent : PYRAZOLOPYRIDINE DERIVATIVES AND THEIR USE IN THERAPY Priority Date ‐ 8. September 2014 Grzegorz Popowicz, Michael Sattler, Maciej Dawidowski, Leonidas Emmanouilidis (Helmholtz Zentrum, München) Vishal Kalel, Wolfgang Schliebs, Ralf Erdmann (Ruhr‐universität, Bochum)

Manuscripts under preparation (Tentative titles)

Structure based design of novel molecules targeting parasite glycosomal (peroxisomal) protein import Grzegorz Popowicz, Maciej Dawidowski, Vishal Kalel, Leonidas Emmanouilidis, Wolfgang Schliebs, Ralf Erdmann, Michael Sattler

Identification and validation of novel druggable targets in glycosomal membrane protein import Vishal C. Kalel, Udaya Bhandari, Wolfgang Schliebs, Ralf Erdmann

104

APPENDIX

2.2 Conference contributions

Talks

Title ‐ Identification and characterization of TbPEX16 43rd Microsymposium on glycosomes, trypanosomatid metabolism and drug discovery 27‐28 October 2011, Brussels (Belgium)

Title ‐ Towards the role of trypanosomal Pex16 in glycosome biogenesis 44th Microsymposium Parasite/Trypanosomatid Metabolism, Drug Design and Glycosomes 25‐26 October 2012, Amsterdam (Netherlands)

Title ‐ Development of novel anti‐trypanosomal compounds selectively interfering with the glycosomal protein import 46th Microsymposium on Parasite/Trypanosomatid metabolism, drug design and glycosomes 23‐24 October 2014, Amsterdam (Netherlands)

Posters

Open European Peroxisome Meeting 2012 (OEPM 2012) 05‐06 July2012, Dijon (France)

Open European Peroxisome Meeting 2012 (OEPM 2014) 09‐11 September 2014, Neuss (Germany)

German Society for Cell Biology (DGZ) meeting 2015 24‐27 March 2015, Cologne (Germany)

105

APPENDIX

3. Curriculum vitae

Personal information

Name: Vishal C. Kalel Date of Birth: 03.06.1984 Place of Birth: Walchandnagar Nationality: India

Higher Education 2010 ‐ till present Ph.D. student at the Ruhr-University Bochum Institute of Biochemistry and Pathobiochemistry Department of Systems Biochemistry Faculty of Medicine Advisor: Prof. Dr. Ralf Erdmann Title: “Glycosome biogenesis in trypanosomes ‐ Identification of PEX16 and inhibitors of PEX5‐PEX14 interaction”

06.2009 ‐ 12.2009 DAAD Visiting Research Fellowship (Deutscher Akademischer Austausch Dienst, Deutschland) Host Professor‐ Prof. Dr. Ralf Erdmann Ruhr‐Universität Bochum Topic ‐ “Investigating the Leishmania PTS2 and Pex7 interactions using yeast as model organism”

2006 ‐ 2009 Research Fellowship Council for Scientific and Industrial Research (CSIR), India At National Centre for Cell Science (NCCS), Pune, India Topic ‐ “Molecular characterization of glycosomal proteins in Leishmania”.

2004 ‐ 2006 M.Sc. Biochemistry Biochemistry Division, Department of Chemistry, University of Pune, Pune, India Grade‐ “Outstanding”. First rank in University with Gold Medal

2001 ‐ 2004 B.Sc. Chemistry (Physics, Botany, zoology as subsidiary subjects) University of Pune, Maharashtra, India Grade‐ First Class with Distinction

106

APPENDIX

4. Acknowledgments

I would like to express my sincere gratitude to Prof. Dr. Ralf Erdmann for providing me the opportunity to pursue research work in his laboratory towards achieving my PhD degree. Further I wish to extend my gratefulness to him for hosting me as a DAAD visiting researcher during which I came to know about peroxisomes and got fascinated by it. His excellent advices not only related to the lab‐bench work but also regarding career as scientist will help me immensely in finding my future career path.

Heartfelt thanks to Prof. Dr. Wolfgang Schliebs for being an excellent advisor for the scientific landscape of my research work. Discussing every experiment with you including those which worked and mostly the ones which failed pushed me positively in the proper direction. Also thanks to Prof. Dr. Wolfgang Girzalsky for additional advice and support during my research work. I am also grateful to Prof. Dr. Wolf‐H. Kunau for brief but motivating discussions about peroxisome research

I thank PD Dr. Mathias Lübben for agreeing to become my second supervisor for the PhD thesis. I express my gratefulness to Prof. Paul Michels for hosting me in his lab for learning to work with trypanosomes. Melisa, Ana, Nathalie ‐ Paul and your help during this training and also ample assistance afterwards was pivotal for establishing and using the trypanosomes for glycosome research. Without your help, the work would have been impossible. I am also thankful to Prof. Michael Sattler and team at the Helmholtz Zentrum München for the collaborative work on glycosome biogenesis inhibitors and also for useful Telephonic discussions.

I am thankful to Prof. Dr. Ralf Erdmann and Prof. Dr. Michael Ehrmann for including me in the Nobel laureate mentoring program. I am indebted to Nobel laureate Prof. Dr. Erwin Neher for the highly motivating mentoring sessions which inspire me to keep working on research topics relevant to developing and poor countries.

107

APPENDIX

I would like to thank Thomas Schröter for the translation of Summary of my thesis and the peak time technical assistance. I appreciate Patrick Basitta and Udaya Bhandari for your patience and fruitful discussions during your Master’s thesis work in cooperation with me.

I am very indebted to Ms. Marion Witt‐Reinhardt, Ms. Brigitte Scharf and Mr. Burkhard Koppitz whose assistance was crucial to help me acclimatize to RUB and Bochum during my first arrival here. I also thanks to Ms. Meike Jade (Möller), Ms. Christina Pintado and Ms. Britta Stickel for the administrative assistance during my PhD thesis.

Thanks to all my labmates for the cordial help during research work and in between. Also the help of lab technicians is highly acknowledged which supported my research work. I am also thankful to Satya, Shyam and Ding Dong for pleasant company and calorific support. Thanks to Rhishi because of whom I realized my passion for research. Thank you Shashi for being there with me as a close friend in thick and thin. I apologize to all whom I could not mention here.

Finally, I dedicate this thesis to my parents. I express my love and appreciation to my father who gave me ultimate freedom to pursue whatever I WANT to do. You made every possible (and impossible) material available for me to READ. Mom (Aai), I can never thank you enough for your unconditional love to not only for me but absolutely for everyone around. Both of you gave me The unparalleled support to my choices, freedom and literature, without which I would not have been here doing what I LIKE to do.

108

Biochimica et Biophysica Acta 1822 (2012) 1326–1336

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta

journal homepage: www.elsevier.com/locate/bbadis

Review Molecular basis of peroxisomal biogenesis disorders caused by defects in peroxisomal matrix protein import☆

Shirisha Nagotu 1, Vishal C. Kalel 1, Ralf Erdmann ⁎, Harald W. Platta ⁎

Abteilung für Systembiochemie, Medizinische Fakultät der Ruhr-Universität Bochum, D-44780 Bochum, Germany article info abstract

Article history: Peroxisomal biogenesis disorders (PBDs) represent a spectrum of autosomal recessive metabolic disorders Received 19 December 2011 that are collectively characterized by abnormal peroxisome assembly and impaired peroxisomal function. Received in Revised form 26 March 2012 The importance of this ubiquitous organelle for human health is highlighted by the fact that PBDs are mul- Accepted 9 May 2012 tisystemic disorders that often cause death in early infancy. Peroxisomes contribute to central metabolic Available online 19 May 2012 pathways. Most enzymes in the peroxisomal matrix are linked to lipid metabolism and detoxification of reactive oxygen species. Proper assembly of peroxisomes and thus also import of their enzymes relies on Keywords: fi Peroxisome biogenesis disorders speci c peroxisomal biogenesis factors, so called peroxins with PEX being the gene acronym. To date, 13 Zellweger syndrome spectrum PEX genes are known to cause PBDs when mutated. Studies of the cellular and molecular defects in cells PEX derived from PBD patients have significantly contributed to the understanding of the functional role of the Peroxin corresponding peroxins in peroxisome assembly. In this review, we discuss recent data derived from both Ubiquitination human cell culture as well as model organisms like yeasts and present an overview on the molecular mechanism underlying peroxisomal biogenesis disorders with emphasis on disorders caused by defects in the peroxisomal matrix protein import machinery. This article is part of a Special Issue entitled: Metabolic Functions and Biogenesis of Peroxisomes in Health and Disease. © 2012 Elsevier B.V. All rights reserved.

1. Peroxisomes — general introduction Unlike in yeast and plants, β-oxidation of fatty acids occurs in both peroxisomes and mitochondria of mammalian cells. Very long chain fatty acids, long chain dicarboxylic acids, some unsaturated fatty acids, Peroxisomes are single membrane bound, dynamic organelles of pristanoic acids, di and tri-hydroxycholestanoic acids are metabolized eukaryotic cells. They were first identified in the electron microscopy in peroxisomes. In humans, peroxisomes are also involved in synthesis images of mouse kidney cells [1]. They are mostly spherical, 0.1 to of cholesterol, bile acids and ether lipids such as plasmalogens, which ac- 1 μm in diameter and surrounded by a single lipid bilayer membrane count for the major portion of the ethanolamine glycerophospholipids in [2]. They do not contain DNA or a protein translation machinery and the adult human brain, notably 80–90% of these lipids in the white mat- hence import nuclear coded proteins synthesized in the cytosol post- ter of the brain [5–10]. Apart from the above mentioned functions, re- translationally [3]. The number, size and function of peroxisomes de- cently a role for peroxisomes in the antiviral innate immunity has been pend on the cell type or organism and the environmental conditions. described. In mouse embryonic fibroblasts and human hepatocytes, the This is also reflected by their unique variability in enzyme content antiviral signaling protein MAVS (mitochondrial antiviral-signaling pro- and thus metabolic functions, which marks them as “multi-purpose or- tein) has been localized to both peroxisomes and mitochondria. The ganelles” that adjust their metabolic capabilities according to the cellu- data demonstrate that peroxisomes provide a significant site of antiviral lar needs [4]. Until now 50 different enzymes have been identified in signal transduction and that they promote a rapid response to viral infec- the peroxisomal matrix, which are linked to different biochemical path- tion [11]. ways. However, the central function of these organelles in all instances After their initial description in 1954 [1], peroxisomes were first is β-oxidation of fatty acids and hydrogen peroxide detoxification. associated with human disease in 1973, when it was discovered that kidney and liver tissue from Zellweger syndrome patients were devoid of peroxisomes [12]. In 1984, a specific biomarker was identified [13] ☆ This article is part of a Special Issue entitled: Metabolic Functions and Biogenesis of Peroxisomes in Health and Disease. that could be used for the screening of patients and the first gene defect ⁎ Corresponding authors at: Institut für Physiologische Chemie, Ruhr-Universität associated with a peroxisomal biogenesis defect was identified in 1992 Bochum, Universitätsstr, 150, D-44780 Bochum, Germany. Tel.: +49 234 322 4943; [14]. fax: +49 234 321 4266. The finding that peroxisomes constitute the sole site for breakdown E-mail addresses: [email protected] (R. Erdmann), [email protected] (H.W. Platta). of fatty acids in fungi and the synthesis of plasmalogens in mammalian 1 Joint first authors. cells, has been used for the screening of yeast [15] and Chinese hamster

0925-4439/$ – see front matter © 2012 Elsevier B.V. All rights reserved. doi:10.1016/j.bbadis.2012.05.010 S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 1327 ovary (CHO)-cell mutants [16] affected in peroxisome function. These PTS2). Notable exceptions are C. elegans and diatoms, which have lost mutants were instrumental for the identification of the corresponding the genes for targeting PTS2 proteins during evolution and import all genes by functional complementation which led to the discovery of matrix proteins via PTS1 signals [36,37]. the first proteins essential for peroxisome formation [17,18].Mostof PTS1 was initially discovered as the carboxyterminal tripeptide the currently known peroxisome biogenesis factors, collectively called SKL in the firefly luciferase protein [38]. Sequence comparisons be- peroxins (PEX), were identified by these genetic approaches, some tween species led to a consensus sequence (S/A/C)-(K/R/H)-(L/M) [39]. later on by proteomic approaches [19] or as in the case of the recently In recent years, it became clear that additional amino acid residues with- discovered PEX34, by their interaction with known peroxins [20]. in the cargo protein are of relevance for the interaction with the PTS1- Apart from the study of their metabolic functions another widely receptor, which led to a refinement of the definition of the PTS1 [40]. studied research area is peroxisome biogenesis. Essential aspects of ThePTS1asatargetingsignalisusedbymajorityoftheperoxisomal peroxisome biogenesis concern the cellular origin of the organelle proteins. Mammalian catalase and alanine-glyoxylate aminotransferase and the matrix and membrane protein import, which are completely are examples of two proteins that need interactions in addition to the independent processes. A role for the endoplasmic reticulum (ER) in binding of the PTS1 for proper targeting to the peroxisomes. A KANL the origin of peroxisomal membrane was hypothesized initially sequence at the C-terminus is required for the targeting of mammalian based on the close proximity of ER and peroxisomes in electron mi- catalase [41] whereas two signals are required for the targeting of crographs [21]. However the post-translational import of peroxisomal alanine-glyoxylate aminotransferase, a C-terminal KKKL and an internal proteins developed the idea of peroxisomes as an autonomous organ- eight amino acid sequence [42]. The PTS2 is located near the N-terminus elle which multiply by growth and division [22]. This concept of per- of the proteins and conforms to the motif R-(L/V/I/Q)-xx-(L/V/I/H)-(L/S/ oxisomes being autonomous organelles was challenged with the G/A)-x-(H/Q)-(L/A) [43]. Very few proteins are targeted to the peroxi- discovery that peroxisomes can be formed from the endoplasmic re- somes by PTS2 in mammals. However, in plants approximately one ticulum upon the reintroduction of Pex3 in peroxisome-deficient third of the peroxisomal proteins are targeted via the PTS2 pathway cells. [23–25]. While, the possibility of de novo formation of peroxi- [44].InSaccharomyces cerevisiae, thiolase (Fox3) and glycerol phosphate somes is now well accepted, the significance of this contribution to dehydrogenase (Gpd1) are the only proteins known to use the PTS2 the peroxisome content is still under debate. While one view is that [45,46]. all peroxisomal membrane proteins traffic via the ER [26], other Proteins that harbor a peroxisomal targeting signal are recognized work demonstrates that under wild type conditions peroxisomes by their respective receptors in the cytosol — which is Pex5 for the are generally formed by growth and division and only under circum- PTS1 [47] and Pex7 for the PTS2 import [48]. The C-terminal domain stances when there are no peroxisomes in a cell, they can be formed of Pex5 binds to the PTS1 sequence of the cargo and is characterized from the ER after reintroduction of the corresponding gene [27]. Also by seven tetratricopeptide repeats. Pex7 is a WD40 protein which recog- in higher eukaryotes, it is shown that peroxisomal membrane pro- nizes the PTS2 signal. A major difference between Pex5 and Pex7 is that teins are directly targeted to peroxisomes [28]. Several recent data, Pex5 functions independently whereas Pex7 requires co-receptors such however, point towards a more semi-autonomous nature of peroxi- as Pex18 and Pex21 in S. cerevisiae or Pex20 in Yarrowia lipolytica, Pichia somes, where certain peroxisomal membrane proteins might traffic pastoris, Hansenula polymorpha and Neurospora [49–51]. Humans and to peroxisomes via the ER [29–31]. plants lack these co-receptors but express a longer splice variant of In contrast to peroxisomal membrane protein targeting, the basic Pex5 which contains a Pex7 binding site [52–54]. Two isoforms of principles of matrix protein import are widely accepted. The ability Pex5 are present in mammalian cells, namely a short (Pex5S) and a lon- to import folded, co-factor bound and oligomeric proteins distinguishes ger species (Pex5L). Both of these isoforms function as PTS1 receptors peroxisomes from other organelles like mitochondria [32–34]. whereas only Pex5L is also required for PTS2 import due to its additional The importance of peroxisomes for human health is highlighted Pex7-binding domain [53]. Thus, while PTS1 and PTS2 import pathways by severe inborn peroxisomal diseases. These can be caused by defects function independently in yeasts and fungi, they converge in higher in peroxisome biogenesis or can be due to peroxisomal single enzyme eukaryotes such as mammals and plants already at the level of cargo deficiencies. Peroxisome biogenesis disorders are caused by defects recognition. in PEX genes. At present, 33 different PEX genes are known and 13 While most peroxisomal proteins use either the PTS1 or PTS2 sig- orthologous human PEX genes have been described (Table 1). nal for targeting, there are some proteins which do not contain such In this review, we focus on the import of peroxisomal matrix pro- a signal [55]. Interestingly, the import of these non-PTS proteins still teins and the importance of this process for humans. We highlight the depends on the presence of Pex5, mostly via an interaction to the increased understanding of the molecular mechanism of this trans- N‐terminal portion of Pex5. Some examples include acyl-CoA oxidase port process and explain the manifestation of the severe peroxisomal in S. cerevisiae and Y. lipolytica, alcohol oxidase in H. polymorpha [56,57]. biogenesis disorders (PBD), which occur when this import process is Moreover, some proteins that lack a PTS might hijack the PTS-pathway defective. Recent developments of various animal models and thera- by piggy-backing onto proteins that do contain a PTS. Enoyl-CoA peutic strategies for mild PBDs associated with a defective matrix im- isomerases Eci1 and Dci1 in S. cerevisiae and the five acyl-CoA oxi- port will also be discussed here. dase isoforms in Y. lipolytica are examples for this mechanism [58,59]. Also in mammals, the copper–zinc superoxide dismutase 1 (SOD1), a 2. Matrix protein import non-PTS protein, is targeted to peroxisomes with the aid of its interacting partner, the copper chaperone of SOD1 [60].

A remarkable feature of peroxisomes is the import of fully folded 2.2. Docking of the receptor/cargo complex at the membrane even oligomeric and co-factor bound proteins [35]. This import process can be divided into ﬁve steps as (i) cargo recognition, (ii) receptor Subsequent to the formation of the receptor–cargo complex in the docking, (iii) cargo translocation, (iv) cargo release and (v) receptor cytosol, the receptor ferries the cargo protein to the peroxisomal release and recycling (Fig. 1). membrane. The membrane-bound peroxins Pex13 and Pex14 are required for the initial binding step. Lack of one of these docking com- 2.1. Cargo recognition by the import receptors in the cytosol plex peroxins signiﬁcantly affects the import of both PTS1 and PTS2 targeted proteins into the peroxisome [61–66]. Pex13 is an integral Proteins that are to be imported into the peroxisomal matrix typ- membrane protein and binds to Pex14 by its SH3 domain which con- ically contain one of the two peroxisome-targeting signals (PTS1 and tains a proline rich SH3 ligand motif. However, a second binding site 1328 S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336

Table 1 Peroxisomal biogenesis factors and associated biogenesis disorders. The table lists the known human and S. cerevisiae proteins required for peroxisomal biogenesis. Abbreviations used are IRD (infantile Refsum disease), NALD (neonatal adrenoleukodystrophy), RCDP1 (Rhizomelic Chondrodysplasia Punctata type 1) and ZS (Zellweger syndrome). The asterisk * points out that the protein is absent in S. cerevisiae and other yeast species with an exception of Y. lipolytica. The ✓ marks the corresponding matrix protein import defect. The (✓) indicates a stress-related matrix protein import defect. For detailed information we refer to the text and for cross-references to [20,35,102,103,113,116–118,127,129,133].

Role H. sapiens S. cerevisiae Human disease Gene deletion phenotype in S. cerevisiae — mislocalization of matrix proteins

PTS1 PTS2

Matrix protein import PTS1 receptor Pex5 Pex5 ZS, NALD ✓ PTS2 receptor Pex7 Pex7 RCDP1, Adult Refsum disease ✓ PTS2 co-receptor Pex5L Pex18 ✓ Pex21 ✓ Docking complex Pex13 Pex13 ZS, NALD ✓✓ Pex14 Pex14 ZS ✓✓ Pex17 ✓✓ Importomer assembly Pex8 ✓✓ RING ﬁnger ligase complex Pex2 Pex2 ZS, IRD ✓✓ Pex10 Pex10 ZS, NALD ✓✓ Pex12 Pex12 ZS, NALD, IRD ✓✓ Receptor ubiquitin conjugation UbcH5a/b/c Pex4 Unknown ✓✓ (E2D1/2/3) Pex22 ✓✓ Receptor deubiquitination Ubp15 (✓) Unknown USP9X Unknown AAA export complex Pex1 Pex1 ZS, NALD, IRD ✓✓ Pex6 Pex6 ZS, NALD, IRD ✓✓ Pex26 Pex15 ZS, NALD, IRD ✓✓ AWP1 Unknown Membrane biogenesis and regulatory peroxins Gene deletion phenotype in S. cerevisiae Membrane biogenesis Pex3 Pex3 ZS Absence of peroxisomes Pex19 Pex19 ZS Absence of peroxisomes Pex16 (Pex16)* ZS Peroxisome proliferation Pex11α,β,γ Pex11,25,27 Unknown Reduced number of peroxisomes Regulation of size, number and distribution Pex28–32, Pex34 Altered peroxisome number and/or morphology

in Pex13 for Pex14 was also identified and it has also been shown that deficiency of Pex17 affects import of PTS1 as well as PTS2 proteins, how- the interactions of Pex13 with Pex14 are mediated by several ever, the functional role of the protein remains enigmatic [75,76]. direct or indirect interactions [67]. The N-terminus of Pex13 binds to Pex7 and the SH3 domain interacts with the WXXXF/Y motif of the PTS1 receptor in yeast [63,65,66,68]. However, the Chinese hamster 2.3. Cargo translocation Pex5 does not interact with the SH3 domain of Pex13 but interacts with its N-terminus which also contains the Pex7 binding region The import of folded and in some cases even cofactor bound or [69]. oligomerized proteins into the matrix distinguishes peroxisomes sig- Pex14 is an integral peroxisomal membrane protein in most spe- nificantly from mitochondria and chloroplasts. However, it remained a cies but in some species it is also described to be peripherally associ- matter of speculation for a long time how the cargo proteins traverse ated with the peroxisomal membrane [61,70–72]. Several studies the membrane without affecting the permeability barrier. One hypoth- suggest that Pex14 is the initial docking protein. NMR and crystal esis was that the import receptors themselves might temporally be- structures have revealed a structural basis for the Pex14 and Pex5 income part of the dynamic import pore [79]. This was mainly based on teractions [73,74]. The interaction between Pex14 and Pex5 is shown to the observations that the PTS1-receptor Pex5 can bind lipids and be mediated by the Pex5 WXXXF/Y motifs. The N-terminus of Pex14 changes its topology at the peroxisomal membrane. At the membrane, comprises two hydrophobic cavities which recognize the WXXXF/Y it is partially carbonate resistant, adapts a partial protease-protected motif of Pex5. These hydrophobic cavities are separated by two aromat- state and thus behaves like an integral membrane protein [80–82].In ic residues and are flanked by several basic amino acids leading to a line with this view, Pex5 together with Pex14 turned out to represent positively charged protein surface. The Pex5 peptide adopts an amphi- the minimal unit for the import of the intraperoxisomal protein Pex8 pathic α-helical conformation in the Pex5–Pex14 complex structure in P. pastoris [83]. Furthermore, by the use of the planar lipid bilayer and binds diagonally across helices α1andα2 in Pex14 [73]. technique, it has been shown that the import receptor Pex5 and its Yeast Pex17 is a peripheral membrane protein of unknown function, docking protein Pex14 in the presence of a cargo protein form a pore which associates to peroxisomes via Pex14 [75,76]. A homolog of Pex17 with features expected for a protein-conducting channel [84].More- in higher eukaryotes has not yet been identified. In filamentous fungi, over, the dynamic behavior and physical properties of this channel a chimeric protein that consists of Pex14-like N-terminal domain and with a diameter of up to 9 nm appear to fulfill the criteria for the pas- a Pex17-like C-terminal domain has been described [77,78]. In yeast, sage of folded proteins. However, the exact composition of this pore S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 1329 as well as the driving force and the detailed mechanism of cargo trans- 2.5. Receptor ubiquitination and recycling location is still to be disclosed. Subsequent to cargo liberation, the PTS-receptors return to the cytosol for further rounds of import. Early studies discovered that the 2.4. Cargo release into the matrix peroxisomal matrix protein import is an energy-dependent process requiring the hydrolysis of ATP [94]. The idea of import receptors The release of the cargo into the peroxisomal lumen after transloca- shuttling between the peroxisomal membrane and the cytosol was tion is still not well characterized. A role for the less conserved yeast first described by Marzioch et al. (1994) [95], based on the predominant Pex8 in this process has been suggested. Pex8 is an intra-peroxisomal cytosolic localization of the PTS2-receptor. Investigations in perme- peripheral membrane protein, which contains both PTS1 and PTS2 sig- abilized cell systems of human fibroblasts provided the first evidence nals for its targeting to peroxisomes [85,86]. There is some evidence for for cycling of the PTS1 receptor as the protein accumulated reversibly the role of S. cerevisiae Pex8 in the dissociation of the PTS1-receptor at the peroxisomal membrane under conditions when protein transport cargo complex [87]. Another function assigned to S. cerevisiae Pex8 was blocked [96]. Furthermore, in vitro studies revealed that the binding is that it is required to physically connect the docking complex to and translocation of Pex5 does not require ATP while the export of Pex5 the RING finger complex of the export machinery (see below) [88]. back to the cytosol is the ATP-dependent step [97]. The corresponding However, this structural function seems to be taken over in P. pastoris ATPase was identified as the peroxisomal AAA (ATPases associated by Pex3 [89]. As Pex8-like proteins are not yet identified in humans, with diverse cellular activities)-complex, consisting of Pex1 and Pex6, the general mechanism of cargo release remains elusive. both in human [98] and yeast cells [99]. The function of Pex1 and Pex6 A role for Pex14 in the release of catalase from the translocation is not redundant [86,87] and depends on the presence of their mem- machinery into the matrix has been suggested recently [90]. The au- brane anchor, Pex26 in mammalian cells and its ortholog Pex15 in thors suggest that the release of the cargo is facilitated by the interac- yeast [100,101]. The binding and consumption of ATP by the AAA pro- tion of Pex5 with the N-terminal domain of Pex14 molecules at the teins are believed to induce conformational changes that generate translocation machinery. the driving force to pull the receptor out of the membrane [102,103]. After the release into the matrix, a subset of proteins is processed The exact mechanism of substrate recognition and extraction from in peroxisomes of mammals and plants [91,92]. In mammalian cells, the membrane is not known. However, it has become increasingly the intraperoxisomal protease Tysnd1 is responsible both for the re- clear that ubiquitination of the PTS-receptor plays a major role in the removal of the leader peptide from PTS2 proteins and for the specific ceptor release. Ubiquitination is a highly conserved post-translational processing of PTS1 proteins. Tysnd1 turned out to be a key regulator modification that is catalyzed by a three-step enzyme cascade of the peroxisomal β-oxidation pathway. The proteolytic activity of (E1, E2, E3) and results in the covalent attachment of the 76 amino oligomeric Tysnd1 is controlled by self-cleavage and the degradation acid comprising ubiquitin to a substrate [104]. Pex5 exhibits two differ- products are removed by the peroxisomal Lon protease [91–93]. ent modes of ubiquitination, mono and polyubiquitination. For

Fig. 1. Composite model of peroxisomal matrix protein import. Proteins harboring a peroxisomal targeting signal type 1 (PTS1) are recognized by the soluble import receptor Pex5 in the cytosol and proteins with the PTS2 sequence are recognized by Pex7 and the cofactors Pex18 and Pex21 in S. cerevisiae, Pex5L in plants and mammals. After this step, the receptor–cargo complex is directed to the peroxisomal membrane and associates with the docking complex consisting of Pex14 and Pex13 as well as Pex17 in yeast. Assembly of the cargo-loaded Pex5 with the docking complex results in the formation of a transient pore, which mainly consists of Pex5 and Pex14. The exact architecture and components of the pore yet have to be identified. The cargo is translocated into the peroxisomal lumen in an unknown manner. The receptor–cargo complex then dissociates and the cargo is released into the peroxisomal lumen, a process which possibly involves Pex8 in yeast. The RING-complex comprises Pex2, Pex10 and Pex12 which all are ubiquitin ligases, which together with ubiquitin-conjugating enzymes Pex4 and its membrane anchor Pex22 in yeast or the Pex4-like isoforms UbcH5 a, b, c in mammals are responsible for receptor monoubiquitination. This modification serves as a signal for the ATP-dependent dislocation of Pex5 from the peroxisomal membrane back to the cytosol by the AAA peroxins Pex1 and Pex6. Pex1 and Pex6 are anchored to the peroxisomal membrane via Pex15 in yeast and Pex26 in mammals. In yeast the ubiquitin-conjugating enzyme Ubc4 together with the redundant proteins Ubc5 and Ubc1 polyubiquitinate Pex5 which is then degraded by the 26S proteasome. Thus, while polyubiquitination directs Pex5 to a quality control pathway, the modification by monoubiquitination primes Pex5 for new round of import. The AAA peroxin Pex6 in yeast interacts with Ubp15, a deubiquitinating enzyme acting on Pex5, while human Pex6 interacts with AWP1, an adaptor protein for ubiquitinated Pex5 in humans. In mammals, Pex5 is deubiquitinated by USP9X in the cytosol. The import machinery components present in yeast and humans are depicted in different font colors. Black — components present in both yeast and human, red — components specific to humans, blue — components specific to yeast. 1330 S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 monoubiquitination of Pex5, a single ubiquitin protein is attached via a degradation whereas in the case of the peroxisomal import the peroxi- thioester bond to a conserved cysteine residue of the receptor whereas somal receptors are modified. Thus, the receptors in peroxisomal protein attachment of ubiquitin to conserved lysine residues and subsequent import behave like the cargos in ERAD. These considerations led to the ubiquitination of the attached ubiquitin results in polyubiquitination. export-driven import model proposing that the ERAD-like removal of The polyubiquitination cascade acting on Pex5 has been unraveled the peroxisomal import receptor is linked to protein import [124]. in S. cerevisiae. Here, the ubiquitin-conjugating enzyme (E2) Ubc4 In support of this model, the presence of a functional receptor–export and the partially redundant Ubc5 and Ubc1 are required for poly- complex is a pre-requisite for the import of matrix proteins into peroxi- ubiquitination of Pex5 [105,106]. Both Pex2 and Pex10 have been somes. This mode of protein translocation is dependent on ATP and the suggested to function as the corresponding ubiquitin-ligases (E3) AAA-dependent energy driven extraction of the receptor from the im- [107,108]. Polyubiquitinated Pex5 is degraded in the 26S proteasome port pore is supposed to be coupled to the movement and translocation and thus is considered as a quality control system for aberrant PTS1- of the cargo across the membrane [124]. Recently, work on the receptor molecules [105,106]. ubiquitination of the PTS2-co-receptor Pex18 in S. cerevisiae provided The monoubiquitination of Pex5 on a conserved cysteine is a prereq- direct evidence for this model as the cysteine-dependent mono- uisite for the export of the PTS1-receptor back to the cytosol [109–111] ubiquitination of Pex18 which is required for receptor export was and thus is essential for peroxisomal biogenesis. The E2 protein Pex4 found to be a prerequisite for translocation of cargo-loaded Pex7 across together with its membrane anchor Pex22 is required for this modifica- the peroxisomal membrane [119]. tion in yeast [111,112] while in mammals the Pex4-like isoforms UbcH5a, UbcH5b; and UbcH5c (also known as UBE2D1, UBE2D2, 3. Peroxisomal matrix protein import defects and human disorders UBE2D3) catalyze the cysteine-dependent ubiquitination [113]. Fur- thermore, the E3 ligase Pex12 is also required for this process [107]. A wide spectrum of disorders is associated with defects in peroxi- The ubiquitin moiety needs to be removed from the mon- some biogenesis affecting specific steps like matrix protein import, oubiquitinated Pex5 before it can enter a new round of import. This peroxisome membrane biogenesis or organelle division [127,128]. cleavage of ubiquitin from the substrate is generally carried out by Peroxisome biogenesis disorders (PBD) can be generally divided into ubiquitin hydrolases also called deubiquitinating enzymes [114].Re- two subsections namely Zellweger syndrome spectrum (ZSS) and clin- cent in vitro data obtained from rat suggests that the monoubiquitin ically distinct rhizomelic chondrodysplasia punctata type 1 (RCDP1). moiety of Pex5 can be cleaved in two ways, a non enzymatic release The ZSS is a genetically heterogeneous group of disorders with over- of the thioester bond between Pex5 and mono-Ub by a nucleophilic lapping clinical phenotypes, which includes the most severe Zellweger attack of glutathione or, as the major pathway, enzyme catalyzed by syndrome (ZS), less severe neonatal adrenoleukodystrophy (NALD) ubiquitin hydrolases [115]. Ubp15 was identified as such an ubiquitin and relatively milder infantile Refsum disease (IRD). Neurological ab- hydrolase in S. cerevisiae. The protein is a novel interaction partner of normalities and developmental defects are common in PBD patients Pex6 and functions as a deubiquitinating enzyme acting on Pex5 [116]. which appear early after birth. Abnormalities in fatty acid metabolism Recently, its putative mammalian ortholog, the ubiquitin-specificprote- lead to changes in the levels of various metabolites like very long ase 9X (USP9X) has been identified as a deubiquitinase acting on the chain fatty acids (VLCFA) or plasmalogens [129]. Several diseases are ubiquitin-Pex5 thioester conjugate [117]. also attributed to single peroxisomal enzyme deficiencies [130] or de- While it seems clear that the purpose of monoubiquitination is to fects in fatty acid transport across the peroxisomal membrane [131]. prime Pex5 for export mediated by the AAA peroxins as part of the This review focuses mainly on disorders caused by matrix protein im- recycling pathway, the mechanistic role of the modification remains port defects as well as on recent developments in animal models to to be investigated. In this respect, the recent discovery of a novel study such disorders and novel therapeutic strategies. adaptor protein of human Pex6, AWP1 could provide a mechanistic Experimental studies on the fibroblasts isolated from patients link [118]. AWP1, which has been described before to function as a suffering from PBDs led to the observation of distinct types of perox- ubiquitin-binding NF-kappaB modulator, is able to interact with both isomal protein import defects at a subcellular level. Various strategies Pex6 and monoubiquitinated Pex5 and thus might function as a selec- were employed to dissect the degree of these import defects. Slawecki tive linker, which enables the AAA proteins to transfer their suggested et al. [132] conducted immunofluorescence microscopy studies using pulling force to the receptor molecule intended for export. antibodies against the PTS1 (−SKL) or the PTS2 protein thiolase while Not only the PTS1-receptor Pex5, but also components of the PTS2 more recently Ebberink et al. [133] used fluorescent marker proteins pathway are ubiquitinated. The co-receptors Pex18 in S. cerevisiae GFP–PTS1 or PTS2–GFPtoelucidatetheimportdefectsinover600PBD [119,120] and Pex20 in P. pastoris [121,122] are ubiquitinated at the patient cell lines. Three types of matrix protein import defects were peroxisome membrane and this modification is crucial for PTS2-co- observed: (i) defects in PTS1 protein import, (ii) defects in PTS2 protein receptor recycling. import and (iii) defects in PTS1 as well as PTS2 protein import (Table 1).

2.6. Possible link between receptor export and cargo release: the export 3.1. Defects in PTS1 protein import driven import model As described in Section 2.1, Pex5 is the receptor for peroxisomal The components of the peroxisomal import receptor ubiquitination targeting of PTS1-containing proteins. Human Pex5 was identified and export machinery are evolutionarily related to the proteins of by Dodt et al. based on sequence homology with P. pastoris Pex5 the Endoplasmic Reticulum Associated Degradation (ERAD) machinery and it was shown that mutations in Pex5 lead to PBDs (118). [123–125]. ERAD represents a mechanism by which misfolded and Unlike human cells, yeast cells that completely lack Pex5 still polyubiquitinated proteins are extracted from the ER for their subse- exhibit normal PTS2 import. Yeast Pex7 binds to the co-receptors quent proteasomal degradation [126]. The structural and functional Pex18 and Pex21 (S. cerevisiae) or Pex20 (Y. lipolytica, H. polymorpha) similarities of ERAD and the peroxisomal receptor cycle might provide and the peroxisomal targeting of PTS2-proteins in yeast occurs inde- a clue to how the energy-requirement of matrix protein import might pendent of the Pex5 import pathway [52]. In humans, PTS2 protein be connected to the translocation of the cargo through the import pore. import is dependent on Pex5 since Pex7 requires the longer splice The similarity between both machineries is that they use ubiquitination isoform as a co-receptor [52]. Ebberink et al. [133] performed an anal- to mark proteins for ATP-dependent release from the membrane. How- ysis of mutations in patient cell lines and found 11 different muta- ever, the target for ubiquitination is different in both cases — in ERAD tions in Pex5, out of which 4 mutations lead to a specificdefectinthe the cargo is ubiquitinated, released and directed for proteasomal import of PTS1 containing proteins. These four mutations lie in regions S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 1331 which are not involved in PTS2 protein import like the Pex7 binding box ZSS patients exhibit severe congenital neurological abnormalities or regions required for interaction with Pex13 or Pex14. Consequently, like disturbances in neuronal migration and differentiation which PTS2 import is normal in these cell lines. In contrast, the remaining typically leads to death within the first year after birth [141]. Hepatic seven Pex5 mutations lead to the loss of both PTS1 and PTS2 protein and renal dysfunctions are also frequently observed and therefore ZSS import, resulting in the severe clinical Zellweger phenotype (see is also referred to as cerebrohepatorenal syndrome. Patients with Section 3.3) [52,132,133]. milder PBDs like NALD or IRD survive relatively longer than ZS patients, Only PTS1 protein import defects are associated with the disorder but they also display various progressive developmental defects like NALD. Elevated levels of VLCFA and pipecolic acid are accompanied by loss of vision and hearing and other symptoms like hypotonia and adeficiency in plasmalogens. Clinical symptoms include neurological seizures. Biochemical features of these PBDs include elevated plasma abnormalities like de-myelination, hypotonia, seizures, sensorineural levels of very long chain fatty acids (VLCFAs), branched chain fatty hearing loss and psychomotor retardation [134]. acids, di- and trihydroxycholestanoic acid (DHCA/THCA) and L-pipecolic acid. On the other hand, deficiency of plasmalogens and docosahexanoic 3.2. Defects in PTS2 protein import acid (DHA) is observed in the plasma and erythrocytes [142,143]. Peroxisomal disorders were originally described based on clinical Although the PTS2 signal is less commonly utilized for peroxisomal phenotypes without the knowledge of their molecular cause. Several targeting in humans [44], crucial metabolic enzymes like 3-ketoacyl peroxins involved in peroxisome biogenesis were first identified in Co-A thiolase (VLCFA metabolism), alkylglycerone phosphate synthase various yeast species or mammalian cells like CHO [15,16].Usingthis (AGPS) (plasmalogen synthesis) and phytanic acid co-A hydrolase knowledge of peroxins from different species, the orthologous human (PHYH) (phytanic acid catabolism) are targeted by this pathway in counterparts were identified based on their sequence similarity. Subse- humans. Specific defects in the PTS2 import, potentially caused by quently, mutations of the hereby identified human PEX genes were mutations in Pex7 or the Pex7-binding region in Pex5L, lead to the demonstrated to be responsible for PBDs. Thus, the genetic basis of all mislocalization of PTS2 proteins while PTS1 import is not affected. peroxisome biogenesis disorders is known. Metabolic abnormalities caused by specific PTS2 import defects are Mutational analysis of over 600 cell lines from PBD patients revealed associated with RCDP1 disorder which differs genetically as well as clin- that mutations in Pex1 are the most common (58%) followed by Pex6 ically from PBD, ZSS [135,136]. RCDP1 patients have plasmalogen defi- (16%) [133] with the Gly843Asp substitution being the most common ciency and elevated levels of phytanic acid due to the mislocalization mutation found in Pex1. This mutation reduces the Pex1–Pex6 interac- of AGPS and PHYH respectively [136]. However, normal levels of tion [144] and is associated with temperature sensitivity observed in VLCFA are observed despite of the absence of peroxisomal 3-ketoacyl cell lines that show improved PTS1 import when cells are grown at Co-A thiolase. This probably is due to the thiolase activity of Sterol 30 °C. The repertoire of known disease causing mutations in all peroxins carrier protein X which contains a PTS1 [137]. is reported in Ebberink et al. 2011 [133] or is available at www.dbpex. S. cerevisiae Pex7 mutants are characterized by PTS2 import de- org. fects. Motley et al. [138] cloned Kluyveromyces lactis Pex7 by functional complementation in S. cerevisiae lacking the endogenous Pex7 and 4. Disorders associated with peroxisomal protein import identified the conserved residues in Pex7. Multiple sequence alignments of the human ORFs with yeast Pex7 led to the identification 4.1. Unique case of acquired peroxisomal targeting of the human PTS2 receptor Pex7 and it was confirmed that mutations in human Pex7 are responsible for RCDP1 disorder [135]. Peroxisomal disorders are associated with genetic defects in per- Adult Refsum disease is mainly (90% cases) attributed to the peroxisome biogenesis genes or single enzyme deficiencies. However, oxisomal single enzyme deficiency disorder caused by mutations in Shepard et al. [145] described a striking link between mutations in a phytanoyl-CoA hydroxylase (PHYH). The patients are affected in the gene totally unrelated to peroxisomes and a probable role of normal alpha-oxidation pathway [139]. About 10% of the cases of adult peroxisomal import in the disease pathology. Mutations in myocilin Refsum disease are caused by mutations in Pex7 without any defects (MYOC) gene are the genetic cause of primary open angle glaucoma in PHYH [140]. It should be noted that infantile Refsum disease is a (POAG), a disease that belongs to the genetically heterogeneous group distinct disorder from adult Refsum disease with a different genetic of optic neuropathies. Myocilin is a secreted glycoprotein present in cause as well as different clinical phenotypes (See Section 3.3). the trabecular extracellular matrix tissue of the eye and is responsible for the maintenance of intraocular pressure (IOP) [146,147]. Myocilin 3.3. Defects in PTS1 and PTS2 protein import bears a cryptic PTS1 signal at its C-terminus (−SKM). This signal is not accessible for Pex5 recognition in normal cells since the protein is Both PTS1 and PTS2 pathways converge at the peroxisomal mem- secreted via the ER–Golgi pathway [145]. However, mutant myocilin brane where they share the same translocon composed of the docking is retained in the ER leading to ER stress which may trigger ERAD and peroxins (Pex14, Pex13), RING-finger peroxins (Pex2, Pex10, Pex12) dislocate the mutant protein to cytosol for degradation via the ubiquitin and AAA-type ATPase complex peroxins (Pex1, Pex6, Pex26) [102]. proteasome system [148]. The presence of mutant myocilin protein in Hence a mutation in any one of these components affects the import the cytosol or the exposure of its cryptic PTS1 due to misfolding, allows of both PTS1 and PTS2 proteins. Mutations in Pex1, Pex2, Pex5L, Pex6, the PTS1 receptor Pex5 to access the PTS1 signal and to target the Pex10, Pex12, Pex13, Pex14, or Pex26 result in general matrix protein mutant myocilin to peroxisomes. The severity and early onset of IOP import defects. As a consequence, peroxisomal metabolic pathways elevation is correlated with the mutations that cause a stronger interac- are lost leading to the metabolic abnormalities and form the basis tion with Pex5. This indicates that the gain-of-function mutations in for the lethal genetic disorders collectively termed ZSS [129]. MYOC increases its peroxisomal accumulation, facilitating the progres- Cells from ZSS patients suffering from defects in peroxisomal ma- sion of IOP and pathogenesis of POAG [145]. trix protein import however can still correctly target peroxisomal membrane proteins and hence they contain remnant peroxisomal mem- 4.2. Exploitation of peroxisomal matrix protein import by pathogens brane structures, called ghosts. In contrast, mutations in the peroxins Pex3, Pex16 and Pex19 required for the topogenesis of peroxisomal Several pathogens have been shown to hijack or exploit normal membrane proteins (PMP) cause mislocalization of PTS1 and PTS2 peroxisome functions at least in part of their life cycle [149]. It has proteins due to the complete loss of peroxisomes (reviewed by Fujiki been described that rotaviruses which cause infantile gastroenteritis, et al. in this issue). encode the spike protein VP4 that bears a highly conserved PTS1 signal 1332 S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 at its C-terminus [150]. All 153 Rotavirus VP4 sequences in genebank disease allele PEX1–Gly843Asp and treatment with a chemical library have PTS1 signal variants (SKL, CKL, GKL, CRL, and CRI) of which CRL of small molecule drugs. Recovery of peroxisomal import indicated by is the most common variant (62%). redistribution of cytosolic GFP–PTS1 reporter to peroxisomes after Rotavirus strain SA11 that encodes VP4 protein with a C-terminal drug treatment was assessed by high-content imaging. Upon screening CRL sequence was used to infect a permissive simian (monkey) cell of 2080 bioactive compounds, four molecules were found to recover line MA104. Peroxisomal localization of VP4 was demonstrated in GFP–PTS1 import into peroxisomes. Further validation of peroxisomal virus-infected cells by immunofluorescence microscopy [151]. import and metabolism recovery was done by three independent as- Transient expression of N-terminally GFP tagged VP4 in COS-7L says; import of native peroxisomal proteins, plasmalogen synthesis cells shows punctate peroxisomal localization while truncation of and processing of PTS2. This robust and high-throughput screening C-terminal PTS1 leads to diffuse cytosolic fluorescence. This confirms protocol opens up new frontiers in exploring novel drug therapy for that rotaviruses indeed can utilize the normal PTS1 import pathway for PBD treatment. the transport of at least one of its proteins to the host organelle [151]. The functional significance or requirement of peroxisomal VP4 for the virus remains to be elucidated. 5.3. Stimulating peroxisome proliferation Similarly, the presence of functional PTS1 is also found in Poa semilatent virus cysteine-rich γbprotein[152]. Other notable examples of ex- Sodium 4-phenylbutyrate induces peroxisome proliferation and ploitation of human peroxisomes are association of HIV's Nef protein improves biochemical function in fibroblast cell lines from patients with peroxisomal thioesterase and interaction of influenza virus NS1 with milder PBD phenotypes [155]. Recently, docosahexaenoic acid protein with peroxisomal 17b-HSD4/MFP-2. However, in both the cases (DHA, C22:6n−3) was identified as an inducer of peroxisome divi- the physiological relevance of the peroxisome association of these pro- sion [158]. Treatment of fibroblasts from patients that carry defects teins is not known [153,154]. in peroxisomal fatty acid β-oxidation with DHA induced the proliferation of peroxisomes to the level seen in normal fibroblasts [158]. 5. Development of therapies and new animal models for PBDs Earlier studies demonstrated that also patients with milder PBD phenotypes could benefit from DHA-treatment [159,160], indicating that Early onset of disease and lethality in case of the severe forms the pharmacological induction of peroxisomes in PBD patients might of PBDs has limited scope for the treatment. But patients suffering be another worth pursuing approach to improve overall peroxisomal from milder forms of PBD and those with longer survival rates have biochemical function. some scope for correcting or at least slowing down the progressive developmental defects. Several lines of research are ongoing [155] including (a) dietary therapies — limiting the intake of metabolites such 5.4. New model systems for studying PBDs as VLCFA which tend to accumulate due to peroxisomal defects or supplementing the diet with metabolites like docosahexanoic acid Our knowledge of peroxisome biogenesis is mostly obtained from (DHA), ether lipids and bile acids which are deficient in PBD patients, studies on yeast or mammalian cells which were instrumental in identi- (b) stimulating peroxisomal proliferation or peroxisome related gene fying the corresponding human peroxins and the molecular basis under- expression or (c) screening small molecule compound libraries for new lying peroxisomal disorders. However, to gain insight into how these drugs. Since PBDs are associated with abnormalities in peroxisomal mutations correlate with the pathogenesis of the disease phenotype protein import, recovering or improving import may have beneficial ef- and the developmental defects, there is a need for complex multicellular fects. Here we discuss latest efforts towards the development of such animal model systems. Mouse models of PBDs were generated by therapies. targeted deletion of Pex2, Pex5, Pex7 and Pex13 as well as conditional knockouts of Pex5 and Pex13 which allow tissue-specific inactivation 5.1. Nonsense suppressor therapies of these peroxins [161]. An interesting mouse model concerns PEX11β, a peroxisomal Nonsense mutations (point mutation leading to premature stop membrane protein which plays an important role in the peroxisome codon) are observed in ∼15% of PEX gene alleles in PBD patients. proliferation [19,162,163]. Accordingly, in this mouse model, the loss Dranchak et al., 2011 [156] evaluated whether nonsense suppressor of PEX11β resulted in a reduction of peroxisome abundance and in- therapies which promote the translation read-through can be utilized creased peroxisome clustering and elongation [162]. The Pex11β knock- to restore expression of an active peroxin [156]. Pex2- and Pex12- out mice exhibited a similar clinical phenotype as observed in human defective patient fibroblasts responded well to nonsense suppressor PBDs. However, no defects in peroxisomal protein import or changes therapy, which led to an improvement of peroxisomal VLCFA metab- in the levels of VLCFA or plasmalogens in tissues were observed, thus olism and plasmalogen biosynthesis. Immunofluorescence microscopy challenging the view of involvement of import defects or VLCFA toxicity analysis of treated cells showed recovery of peroxisomal assembly. in the disease pathogenesis [162]. On the other hand, RCDP1 related Pex7 nonsense mutations did not C. elegans as a model system to probe human PBDs is described in respond to PTC124 (ataluren) mediated read-through. Nonetheless, [164,165]. Knockdown of five peroxin homologs of Pex5, Pex6, Pex12, the positive results observed for the Pex2- and Pex12-defective cells Pex13 and Pex19 by RNA-mediated interference (RNAi) led to a devel- provide a proof of concept for using nonsense suppressor therapies for opmental arrest in early larval stage. A drawback of the C. elegans system PBDs which certainly needs further attention. is that the PTS2 pathway does not exist in this organism, hence this model is not applicable for the study of RCDP1 which displays PTS2 im- 5.2. Small molecule screening for rescuing the peroxisomal import port defects [37]. More recently, Drosophila models of PBDs have been described. Mast Peroxin mutations may lead to misfolding or instability of the pro- et al. [166] utilized cultured Drosophila cells to characterize effects of tein or reduced interaction with its binding partners. Small molecule RNAi mediated knockdown of fourteen predicted Drosophila PEX gene compounds may alleviate these defects by mechanisms such as assisting homologs. Drosophila larvae harboring inherited PEX1 mutations most proper folding. A high-content screening assay was developed by Zhang frequently found in PBD patients showed developmental defects analo- et al., 2010 for screening small molecule compounds which can recover gous to Zellweger syndrome patients. Studies on Drosophila PEX mu- peroxisomal import functions [157]. The study was based on the expres- tants also demonstrated their requirement for germ cell development sion of GFP–PTS1 reporter in patient fibroblasts carrying the common and VLCFA metabolism [167,168]. S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 1333

6. Concluding remarks [11] E. Dixit, S. Boulant, Y. Zhang, A.S. Lee, C. Odendall, B. Shum, N. Hacohen, Z.J. Chen, S.P. Whelan, M. Fransen, M.L. Nibert, G. Superti-Furga, J.C. Kagan, Peroxisomes are signaling platforms for antiviral innate immunity, Cell 141 (2010) 668–681. The combination of data derived from human patient cells as well [12] S. Goldfischer, C.L. Moor, A.B. Johnson, A.J. Spiro, M.P. Valsamis, H.K. Wisniewski, as model organisms like yeast merges into a more detailed picture of R.H. Ritch, W.T. Norton, I. Rapin, L.M. Gerner, Peroxisomal and mitochondrial defects in cerebrohepatorenal syndrome, Science 182 (1973) 62–64. the peroxisomal matrix protein import. Even though there are some [13] A.E. Moser, I. Singh, F.R.r. Brown, G.I. Solish, R.I. Kelley, P.J. Benke, H.W. Moser, species-specific differences, the basic principles of peroxisomal protein The cerebrohepatorenal (Zellweger) syndrome. Increased levels and impaired import are well conserved among species. Nevertheless, an important degradation of very-long-chain fatty acids and their use in prenatal diagnosis, – question is whether all major components of the human peroxisomal N. Engl. J. Med. 310 (1984) 1141 1146. [14] N. Shimozawa, T. Tsukamoto, Y. Suzuki, T. Orii, Y. Shirayoshi, T. Mori, Y. Fujiki, protein import machinery have been identified. This is especially evident A human gene responsible for Zellweger syndrome that affects peroxisome when comparing S. cerevisiae, which utilizes twenty-seven peroxins assembly, Science 255 (1992) 1132–1134. from which sixteen are dedicated to matrix protein import, with [15] R. Erdmann, M. Veenhuis, D. Mertens, W.-H. Kunau, Isolation of peroxisome-deficient mutants of Saccharomyces cerevisiae, Proc. Natl. Acad. humans, where seventeen proteins required for biogenesis have been Sci. U. S. A. 86 (1989) 5419–5423. described, from which nine are linked to matrix protein import. So far, [16] T. Tsukamoto, S. Yokota, Y. Fujiki, Isolation and characterization of Chinese the identification of human peroxins has been connected to the study hamster ovary cell mutants defective in assembly of peroxisomes, J. Cell Biol. 110 (1990) 651–660. of patient complementation groups. It has been noted earlier that [17] R. Erdmann, F.F. Wiebel, A. Flessau, J. Rytka, A. Beyer, K.U. Fröhlich, W.-H. Kunau, mutants affected in peroxisome biogenesis that might cause only a PAS1, a yeast gene required for peroxisome biogenesis, encodes a member of a mild phenotype could have been missed. Most interestingly, recent novel family of putative ATPases, Cell 64 (1991) 499–510. [18] T. Tsukamoto, S. Miura, Y. Fujiki, Restoration by a 35K membrane protein of work demonstrates that some answers might come from unexpected peroxisome assembly in a peroxisome-deficient mammalian cell mutant, Nature directions. For instance, the central function of yeast Pex4 in mono- 350 (1991) 77–81. ubiquitination of Pex5 seems to be transferred in evolution to three [19] R. Erdmann, G. Blobel, Giant peroxisomes in oleic acid-induced Saccharomyces cerevisiae lacking the peroxisomal membrane protein Pmp27p, J. Cell Biol. 128 redundant E2 enzymes, UbcH5a, b and c, probably explaining why (1995) 509–523. there is no monogenetic defect of these E2-enzymes related with perox- [20] R.J. Tower, A. Fagarasanu, J.D. Aitchison, R.A. Rachubinski, The peroxin Pex34p isomal disease. Moreover, these E2 enzymes are not peroxisome-specific functions with the Pex11 family of peroxisomal divisional proteins to regulate – but are also involved in other pathways. Similarly, the recent discovery the peroxisome population in yeast, Mol. Biol. Cell 22 (2011) 1727 1738. [21] A.B. Novikoff, W.-Y. Shin, The endoplasmic reticulum in the Golgi zone and its of AWP1, the Pex6 adaptor that binds monoubiqitinated Pex5, has relations to microbodies, Golgi apparatus and autophagic vacuoles in rat liver been described to be essential for matrix protein import even though cells, J. Microsc. 3 (1964) 187–206. it also has additional functions in other signaling pathways. [22] Y. Fujiki, R.A. Rachubinski, P.B. Lazarow, Synthesis of a major integral membrane fi polypeptide of rat liver peroxisomes on free polysomes, Proc. Natl. Acad. Sci. U. S. A. Another major leap in the eld of peroxisome import came recently 81 (1984) 7127–7131. by the identification of the peroxisomal import pore in S. cerevisiae. [23] D. Hoepfner, D. Schildknegt, I. Braakman, P. Philippsen, H.F. Tabak, Contribution However there are still several open questions regarding the mechanis- of the endoplasmic reticulum to peroxisome formation, Cell 122 (2005) 85–95. — [24] P.K. Kim, R.T. Mullen, U. Schumann, J. Lippincott-Schwartz, The origin and main- tic aspects of this process like how is the pore assembled, what is tenance of mammalian peroxisomes involves a de novo PEX16-dependent path- the driving force for the translocation? How is the cargo released? way from the ER, J. Cell Biol. 173 (2006) 521–532. How is the opening and closure of the pore regulated? The mechanistic [25] A. Kragt, T. Voorn-Brouwer, M. van den Berg, B. Distel, Endoplasmic reticulum- directed Pex3p routes to peroxisomes and restores peroxisome formation in a function of the AAA peroxins in the receptor release and how the recep- Saccharomyces cerevisiae pex3Delta strain, J. Biol. Chem. 280 (2005) 34350–34357. tor release relates to import also needs to be unveiled. [26] A. van der Zand, I. Braakman, H.F. Tabak, Peroxisomal membrane proteins insert Knowledge obtained from various organisms with respect to the into the endoplasmic reticulum, Mol. Biol. Cell 21 (2010) 2057–2065. [27] A.M. Motley, E.H. Hettema, Yeast peroxisomes multiply by growth and division, above and many more important questions can be utilized to get better J. Cell Biol. 178 (2007) 399–410. insight of various peroxisomal disorders caused by peroxisomal matrix [28] T. Matsuzaki, Y. Fujiki, The peroxisomal membrane protein import receptor protein import defects. Pex3p is directly transported to peroxisomes by a novel Pex19p- and Pex16p-dependent pathway, J. Cell Biol. 183 (2008) 1275–1286. [29] G. Agrawal, S. Joshi, S. Subramani, Cell-free sorting of peroxisomal membrane Acknowledgements proteins from the endoplasmic reticulum, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 9113–9118. We apologize to all the scientists whose work could not be cited [30] S.K. Lam, N. Yoda, R. Schekman, A vesicle carrier that mediates peroxisome protein traffic from the endoplasmic reticulum, Proc. Natl. Acad. Sci. U. S. A. 107 due to space limitations. This work was supported by grants from the (2010) 21523–21528. Deutsche Forschungsgemeinschaft (SFB642, RE178/2-4) to RE and [31] S. Yonekawa, A. Furuno, T. Baba, Y. Fujiki, Y. Ogasawara, A. Yamamoto, M. HWP. SN was supported by a Fellowship of the Alexander von Tagaya, K. Tani, Sec16B is involved in the endoplasmic reticulum export of the peroxisomal membrane biogenesis factor peroxin 16 (Pex16) in mammalian Humboldt foundation. cells, Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 12746–12751. [32] J.R. Glover, D.W. Andrews, R.A. Rachubinski, Saccharomyces cerevisiae peroxisomal References thiolase is imported as a dimer, Proc. Natl. Acad. Sci. U. S. A. 91 (1994) 10541–10545. [33] J.A. McNew, J.M. Goodman, An oligomeric protein is imported into peroxisomes [1] J. Rhodin, Correlation of ultrastructural organization and function in normal in vivo, J. Cell Biol. 127 (1994) 1245–1257. and experimentally changed peroxisomal convoluted tubule cells of the mouse [34] P.A. Walton, P.E. Hill, S. Subramani, Import of stably folded proteins into peroxi- kidney, Dissertation, Aktiebolaget Godvil, Stockholm (1954). somes, Mol. Biol. Cell 6 (1995) 675–683. [2] H. van den Bosch, R.B. Schutgens, R.J. Wanders, J.M. Tager, Biochemistry of [35] S. Leon, J.M. Goodman, S. Subramani, Uniqueness of the mechanism of protein peroxisomes, Annu. Rev. Biochem. 61 (1992) 157–197. import into the peroxisome matrix: transport of folded, co-factor-bound and [3] P.B. Lazarow, Y. Fujiki, Biogenesis of peroxisomes, Annu. Rev. Cell Biol. 1 (1985) oligomeric proteins by shuttling receptors, Biochim. Biophys. Acta 1763 (2006) 489–530. 1552–1564. [4] F.R. Opperdoes, Glycosomes may provide clues to the import of peroxisomal [36] N.H. Gonzalez, G. Felsner, F.D. Schramm, A. Klingl, U.G. Maier, K. Bolte, A single proteins, Trends Biochem. Sci. 13 (1988) 255–260. peroxisomal targeting signal mediates matrix protein import in diatoms, PLoS [5] N. Nagan, R.A. Zoeller, Plasmalogens: biosynthesis and functions, Prog. Lipid Res. One 6 (2011) e25316. 40 (2001) 199–229. [37] A.M. Motley, E.H. Hettema, R. Ketting, R. Plasterk, H.F. Tabak, Caenorhabditis [6] L. Biardi, S.K. Krisans, Compartmentalization of cholesterol biosynthesis. Con- elegans has a single pathway to target matrix proteins to peroxisomes, EMBO version of mevalonate to farnesyl diphosphate occurs in the peroxisomes, Rep. 1 (2000) 40–46. J. Biol. Chem. 271 (1996) 1784–1788. [38] S.J. Gould, G.A. Keller, S. Subramani, Identification of a peroxisomal targeting [7] A.K. Hajara, J.E. Bishop, Glycerolipid biosynthesis in peroxisomes via the signal at the carboxy terminus of firefly luciferase, J. Cell Biol. 105 (1987) acyldihydroxacetone pathway, Ann. N.Y. Acad. Sci. 386 (1982) 170–182. 2923–2931. [8] S.K. Krisans, The role of peroxisomes in cholesterol metabolism, Am. J. Respir. [39] G. Lametschwandtner, C. Brocard, M. Fransen, P. Van Veldhoven, J. Berger, A. Hartig, Cell Mol. Biol. 7 (1992) 358–364. The difference in recognition of terminal tripeptides as peroxisomal targeting sig- [9] R.J. Wanders, H.R. Waterham, Biochemistry of mammalian peroxisomes nal 1 between yeast and human is due to different affinities of their receptor revisited, Annu. Rev. Biochem. 75 (2006) 295–332. Pex5p to the cognate signal and to residues adjacent to it, J. Biol. Chem. 273 [10] R.L. Wyckle, Brain Phospholipids, Plenum Press, New York, 1977. (1998) 33635–33643. 1334 S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336

[40] C. Brocard, A. Hartig, Peroxisome targeting signal 1: is it really a simple tripeptide? [68] Y. Elgersma, L. Kwast, A. Klein, T. Voorn-Brouwer, M. van den Berg, B. Metzig, T. Biochim. Biophys. Acta 1763 (2006) 1565–1573. America, H.F. Tabak, B. Distel, The SH3 domain of the Saccharomyces cerevisiae [41] P.E. Purdue, P.B. Lazarow, Targeting of human catalase to peroxisomes is depen- peroxisomal membrane protein Pex13p functions as a docking site for Pex5p, dent upon a novel COOH-terminal peroxisomal targeting sequence, J. Cell Biol. a mobile receptor for the import of PTS1 containing proteins, J. Cell Biol. 135 134 (1996) 849–862. (1996) 97–109. [42] P.A. Huber, G.M. Birdsey, M.J. Lumb, D.T. Prowse, T.J. Perkins, D.R. Knight, C.J. [69] H. Otera, K. Setoguchi, M. Hamasaki, T. Kumashiro, N. Shimizu, Y. Fujiki, Peroxi- Danpure, Peroxisomal import of human alanine:glyoxylate aminotransferase somal targeting signal receptor Pex5p interacts with cargoes and import ma- requires ancillary targeting information remote from its C terminus, J. Biol. chinery components in a spatiotemporally differentiated manner: conserved Chem. 280 (2005) 27111–27120. Pex5p WXXXF/Y motifs are critical for matrix protein import, Mol. Cell. Biol. [43] R.A. Rachubinski, S. Subramani, How proteins penetrate peroxisomes, Cell 83 22 (2002) 1639–1655. (1995) 525–528. [70] M. Hayashi, K. Nito, K. Toriyama-Kato, M. Kondo, T. Yamaya, M. Nishimura, [44] S. Reumann, S. Quan, K. Aung, P. Yang, K. Manandhar-Shrestha, D. Holbrook, N. AtPex14p maintains peroxisomal functions by determining protein targeting Linka, R. Switzenberg, C.G. Wilkerson, A.P. Weber, L.J. Olsen, J. Hu, In-depth to three kinds of plant peroxisomes, EMBO J. 19 (2000) 5701–5710. proteome analysis of Arabidopsis leaf peroxisomes combined with in vivo sub- [71] M.E. Oliveira, C. Reguenga, A.M. Gouveia, C.P. Guimaraes, W. Schliebs, W.H. cellular targeting verification indicates novel metabolic and regulatory functions Kunau, M.T. Silva, C. Sa-Miranda, J.E. Azevedo, Mammalian Pex14p: membrane of peroxisomes, Plant Physiol. 150 (2009) 125–143. topology and characterisation of the Pex14p–Pex14p interaction, Biochim. [45] J.R. Glover, D.W. Andrews, S. Subramani, R.A. Rachubinski, Mutagenesis of Biophys. Acta 1567 (2002) 13–22. the amino targeting signal of Saccharomyces cerevisiae 3-ketoacyl-CoA thiolase [72] G.K. Will, M. Soukupova, X. Hong, K.S. Erdmann, J.A. Kiel, G. Dodt, W.-H. Kunau, reveals conserved amino acids required for import into peroxisomes in vivo, R. Erdmann, Identification and characterization of the human orthologue of J. Biol. Chem. 269 (1994) 7558–7563. yeast Pex14p, Mol. Cell. Biol. 19 (1999) 2265–2277. [46] S. Jung, M. Marelli, R.A. Rachubinski, D.R. Goodlett, J.D. Aitchison, Dynamic [73] C. Neufeld, F.V. Filipp, B. Simon, A. Neuhaus, N. Schüller, C. David, H. Kooshapur, changes in the subcellular distribution of Gpd1p in response to cell stress, T. Madl, R. Erdmann, W. Schliebs, M. Wilmanns, M. Sattler, Structural basis for J. Biol. Chem. 285 (2010) 6739–6749. competitive interactions of Pex14 with the import receptors Pex5 and Pex19, [47] S.R. Terlecky, W.M. Nuttley, D. McCollum, E. Sock, S. Subramani, The Pichia EMBO J. 28 (2009) 745–754. pastoris peroxisomal protein Pas8p is the receptor for the C-terminal tripeptide [74] J.R. Su, K. Takeda, S. Tamura, Y. Fujiki, K. Miki, Crystal structure of the conserved peroxisomal targeting signal, EMBO J. 14 (1995) 3627–3634. N-terminal domain of the peroxisomal matrix protein import receptor, Pex14p, [48] P. Rehling, M. Marzioch, F. Niesen, E. Wittke, M. Veenhuis, W.-H. Kunau, The im- Proc. Natl. Acad. Sci. U. S. A. 106 (2009) 417–421. port receptor for the peroxisomal targeting signal 2 (PTS2) in Saccharomyces [75] B. Huhse, P. Rehling, M. Albertini, L. Blank, K. Meller, W.-H. Kunau, Pex17p of cerevisiae is encoded by the PAS7 gene, EMBO J. 15 (1996) 2901–2913. Saccharomyces cerevisiae is a novel peroxin and component of the peroxisomal [49] M. Otzen, D. Wang, M.G. Lunenborg, I.J. van der Klei, Hansenula polymorpha protein translocation machinery, J. Cell Biol. 140 (1998) 49–60. Pex20p is an oligomer that binds the peroxisomal targeting signal 2 (PTS2), [76] J.J. Smith, R.K. Szilard, M. Marelli, R.A. Rachubinski, The peroxin Pex17p of the J. Cell Sci. 118 (2005) 3409–3418. yeast Yarrowia lipolytica is associated peripherally with the peroxisomal mem- [50] M. Sichting, A. Schell-Steven, H. Prokisch, R. Erdmann, H. Rottensteiner, Pex7p brane and is required for the import of a subset of matrix proteins, Mol. Cell. and Pex20p of Neurospora crassa function together in PTS2-dependent protein Biol. 17 (1997) 2511–2520. import into peroxisomes, Mol. Biol. Cell 14 (2003) 810–821. [77] D. Managadze, C. Würtz, S. Wiese, M. Schneider, W. Girzalsky, H.E. Meyer, R. [51] V.I. Titorenko, J.J. Smith, R.K. Szilard, R.A. Rachubinski, Pex20p of the yeast Erdmann, B. Warscheid, H. Rottensteiner, Identification of PEX33, a novel com- Yarrowia lipolytica is required for the oligomerization of thiolase in the cytosol ponent of the peroxisomal docking complex in the filamentous fungus Neurospora and for its targeting to the peroxisome, J. Cell Biol. 142 (1998) 403–420. crassa, Eur. J. Cell Biol. 89 (2010) 955–964. [52] G. Dodt, D. Warren, E. Becker, P. Rehling, S.J. Gould, Domain mapping of human [78] L. Opaliński, J.A. Kiel, T.G. Homan, M. Veenhuis, I.J. van der Klei, Penicillium PEX5 reveals functional and structural similarities to Saccharomyces cerevisiae chrysogenum Pex14/17p—a novel component of the peroxisomal membrane Pex18p and Pex21p, J. Biol. Chem. 276 (2001) 41769–41781. that is important for penicillin production, FEBS J. 277 (2010) 3203–3218. [53] H. Otera, T. Harano, M. Honsho, K. Ghaedi, S. Mukai, A. Tanaka, A. Kawai, N. [79] R. Erdmann, W. Schliebs, Peroxisomal matrix protein import: the transient pore Shimizu, Y. Fujiki, The mammalian peroxin Pex5pL, the longer isoform of the model, Nat. Rev. Mol. Cell Biol. 6 (2005) 738–742. mobile peroxisome targeting signal (PTS) type 1 transporter, translocates the [80] A.M. Gouveia, C.P. Guimaraes, M.E. Oliveira, C. Sa-Miranda, J.E. Azevedo, Inser- Pex7p-PTS2 protein complex into peroxisomes via its initial docking site, Pex14p, tion of Pex5p into the peroxisomal membrane is cargo protein-dependent, J. Biol. Chem. 275 (2000) 21703–21714. J. Biol. Chem. 278 (2002) 4389–4392. [54] A.W. Woodward, B. Bartel, The Arabidopsis peroxisomal targeting signal type 2 [81] A.M. Gouveia, C. Reguenga, M.E. Oliveira, C. Sa-Miranda, J.E. Azevedo, Character- receptor PEX7 is necessary for peroxisome function and dependent on PEX5, ization of peroxisomal Pex5p from rat liver: Pex5p in the Pex5p–Pex14p mem- Mol. Biol. Cell 16 (2005) 573–583. brane complex is a transmembrane protein, J. Biol. Chem. 275 (2000) [55] I.J. van der Klei, M. Veenhuis, PTS1-independent sorting of peroxisomal matrix 32444–32451. proteins by Pex5p, Biochim. Biophys. Acta 1763 (2006) 1794–1800. [82] D. Kerssen, E. Hambruch, W. Klaas, H.W. Platta, B. de Kruijff, R. Erdmann, W.H. [56] A.T. Klein, M. van Den Berg, G. Bottger, H.F. Tabak, B. Distel, Saccharomyces Kunau, W. Schliebs, Membrane association of the cycling peroxisome import cerevisiae acyl-CoA oxidase follows a novel, non-PTS1, import pathway into per- receptor Pex5p, J. Biol. Chem. 281 (2006) 27003–27015. oxisomes that is dependent on Pex5p, J. Biol. Chem. 277 (2002) (25011-25009). [83] C. Ma, U. Schumann, N. Rayapuram, S. Subramani, The peroxisomal matrix im- [57] P. Ozimek, P. Kötter, M. Veenhuis, I.J. van der Klei, Hansenula polymorpha and port of Pex8p requires only PTS receptors and Pex14p, Mol. Biol. Cell 20 (2009) Saccharomyces cerevisiae Pex5p's recognize different, independent peroxisomal 3680–3689. targeting signals in alcohol oxidase, FEBS Lett. 580 (2006) 46–50. [84] M. Meinecke, C. Cizmowski, W. Schliebs, V. Kruger, S. Beck, R. Wagner, R. [58] V.I. Titorenko, J.M. Nicaud, H. Wang, H. Chan, R.A. Rachubinski, Acyl-CoA oxidase Erdmann, The peroxisomal importomer constitutes a large and highly dynamic is imported as a heteropentameric, cofactor-containing complex into peroxi- pore, Nat. Cell Biol. 12 (2010) 273–277. somes of Yarrowia lipolytica, J. Cell Biol. 156 (2002) 481–494. [85] P. Rehling, A. Skaletz-Rorowski, W. Girzalsky, T. Voorn-Brouwer, M.M. Franse, B. [59] X. Yang, P.E. Purdue, P.B. Lazarow, Eci1p uses a PTS1 to enter peroxisomes: Distel, M. Veenhuis, W.-H. Kunau, R. Erdmann, Pex8p, an intraperoxisomal either its own or that of a partner, Dci1p, Eur. J. Cell Biol. 80 (2001) 126–138. peroxin of Saccharomyces cerevisiae required for protein transport into peroxi- [60] M. Islinger, K.W. Li, J. Seitz, A. Völkl, G.H. Lüers, Hitchhiking of Cu/Zn superoxide somes binds the PTS1 receptor pex5p, J. Biol. Chem. 275 (2000) 3593–3602. dismutase to peroxisomes—evidence for a natural piggyback import mechanism [86] L. Zhang, S. Léon, S. Subramani, Two independent pathways trafficthe in mammals, Traffic 10 (2009) 1711–1721. intraperoxisomal peroxin PpPex8p into peroxisomes: mechanism and evolu- [61] M. Albertini, P. Rehling, R. Erdmann, W. Girzalsky, J.A.K.W. Kiel, M. Veenhuis, tionary implications, Mol. Biol. Cell 17 (2006) 690–699. W.-H. Kunau, Pex14p, a peroxisomal membrane protein binding both receptors [87] D. Wang, N.V. Visser, M. Veenhuis, I.J. Van Der Klei, Physical interactions of the of the two PTS-dependent import pathways, Cell 89 (1997) 83–92. peroxisomal targeting signal 1-receptor, Pex5p, studied by fluorescence correla- [62] C. Brocard, G. Lametschwandtner, R. Koudelka, A. Hartig, Pex14p is a member of tion spectroscopy, J. Biol. Chem. 278 (2003) 43340–43345. the protein linkage map of Pex5p, EMBO J. 16 (1997) 5491–5500. [88] B. Agne, N.M. Meindl, K. Niederhoff, H. Einwächter, P. Rehling, A. Sickmann, H.E. [63] R. Erdmann, G. Blobel, Identification of Pex13p a peroxisomal membrane recep- Meyer, W. Girzalsky, W.H. Kunau, Pex8p. An intraperoxisomal organizer of the tor for the PTS1 recognition factor, J. Cell Biol. 135 (1996) 111–121. peroxisomal import machinery, Mol. Cell 11 (2003) 635–646. [64] M. Fransen, S.R. Terlecky, S. Subramani, Identification of a human PTS1 receptor [89] P.P. Hazra, I. Suriapranata, W.B. Snyder, S. Subramani, Peroxisome remnants docking protein directly required for peroxisomal protein import, Proc. Natl. in pex3Delta cells and the requirement of Pex3p for interactions between Acad. Sci. U. S. A. 95 (1998) 8087–8092. the peroxisomal docking and translocation subcomplexes, Traffic3(2002) [65] W. Girzalsky, P. Rehling, K. Stein, J. Kipper, L. Blank, W.-H. Kunau, R. Erdmann, 560–574. Involvement of Pex13p in Pex14p localization and peroxisomal targeting signal [90] M.O. Freitas, T. Francisco, T.A. Rodrigues, I.S. Alencastre, M.P. Pinto, C.P. Grou, A.F. 2dependent protein import into peroxisomes, J. Cell Biol. 144 (1999) Carvalho, M. Fransen, C. Sá-Miranda, J.E. Azevedo, PEX5 protein binds mono- 1151–1162. meric catalase blocking its tetramerization and releases it upon binding the [66] S.J. Gould, J.E. Kalish, J.C. Morrell, J. Bjorkman, A.J. Urquhart, D.I. Crane, Pex13p N-terminal domain of PEX14, J. Biol. Chem. 286 (2011) 40509–40519. is an SH3 protein of the peroxisome membrane and a docking factor for the pre- [91] I.V. Kurochkin, Y. Mizuno, A. Konagaya, Y. Sakaki, C. Schönbach, Y. Okazaki, dominantly cytoplasmic PTS1 receptor, J. Cell Biol. 135 (1996) 85–95. Novel peroxisomal protease Tysnd1 processes PTS1- and PTS2-containing [67] A. Schell-Steven, K. Stein, M. Amoros, C. Landgraf, R. Volkmer-Engert, H. enzymes involved in beta-oxidation of fatty acids, EMBO J. 26 (2007) 835–845. Rottensteiner, R. Erdmann, Identification of a novel, intraperoxisomal pex14- [92] H. Schuhmann, P.F. Huesgen, C. Gietl, I. Adamska, The DEG15 serine protease binding site in pex13: association of pex13 with the docking complex is essential cleaves peroxisomal targeting signal 2-containing proteins in Arabidopsis, Plant for peroxisomal matrix protein import, Mol. Cell. Biol. 25 (2005) 3007–3018. Physiol. 148 (2008) 1847–1856. S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336 1335

[93] K. Okumoto, Y. Kametani, Y. Fujiki, Two proteases, Tysnd1 and PsLon, coopera- [120] P.E. Purdue, P.B. Lazarow, Pex18p is constitutively degraded during peroxisome tively regulate fatty-acid {beta}-oxidation in the peroxisomal matrix, J. Biol. biogenesis, J. Biol. Chem. 276 (2001) 47684–47689. Chem. (2011) (Epub ahead of print). [121] S. Leon, S. Subramani, A conserved cysteine residue of Pichia pastoris Pex20p [94] T. Imanaka, G.M. Small, P.B. Lazarow, Translocation of acyl-CoA oxidase into per- is essential for its recycling from the peroxisome to the cytosol, J. Biol. Chem. oxisomes requires ATP hydrolysis but not a membrane potential, J. Cell Biol. 105 282 (2007) 7424–7430. (1987) 2915–2922. [122] S. Leon, L. Zhang, W.H. McDonald, J. Yates III, J.M. Cregg, S. Subramani, Dynamics [95] M. Marzioch, R. Erdmann, M. Veenhuis, W.-H. Kunau, PAS7 encodes a novel yeast of the peroxisomal import cycle of PpPex20p: ubiquitin-dependent localization member of the WD-40 protein family essential for import of 3-oxoacyl-CoA thiolase, and regulation, J. Cell Biol. 172 (2006) 67–78. a PTS2-containing protein, into peroxisomes, EMBO J. 13 (1994) 4908–4918. [123] T. Gabaldon, B. Snel, F. van Zimmeren, W. Hemrika, H. Tabak, M.A. Huynen, [96] G. Dodt, S.J. Gould, Multiple PEX genes are required for proper subcellular dis- Origin and evolution of the peroxisomal proteome, Biol. Direct 1 (2006) 8. tribution and stability of Pex5p, the PTS1 receptor: evidence that PTS1 [124] W. Schliebs, W. Girzalsky, R. Erdmann, Peroxisomal protein import and ERAD: var- protein import is mediated by a cycling receptor, J. Cell Biol. 135 (1996) iations on a common theme, Nat. Rev. Mol. Cell Biol. 11 (2010) 885–890. 1763–1774. [125] A. Schlüter, S. Fourcade, R. Ripp, J.L. Mandel, O. Poch, A. Pujol, The evolutionary origin [97] M.E. Oliveira, A.M. Gouveia, R.A. Pinto, C. Sa-Miranda, J.E. Azevedo, The energet- of peroxisomes: an ER–peroxisome connection, Mol. Biol. Evol. 23 (2006) 838–845. ics of Pex5p-mediated peroxisomal protein import, J. Biol. Chem. 278 (2003) [126] C. Hirsch, R. Gauss, S.C. Horn, O. Neuber, T. Sommer, The ubiquitylation machinery 39483–39488. of the endoplasmic reticulum, Nature 458 (2009) 453–460. [98] N. Miyata, Y. Fujiki, Shuttling mechanism of peroxisome targeting signal type 1 [127] M. Fidaleo, Peroxisomes and peroxisomal disorders: the main facts, Exp. Toxicol. receptor Pex5: ATP-independent import and ATP-dependent export, Mol. Cell. Pathol. 62 (2010) 615–625. Biol. 25 (2005) 10822–10832. [128] H.R. Waterham, J. Koster, C.W. van Roermund, P.A. Mooyer, R.J. Wanders, J.V. [99] H.W. Platta, S. Grunau, K. Rosenkranz, W. Girzalsky, R. Erdmann, Functional role Leonard, A lethal defect of mitochondrial and peroxisomal fission, N. Engl. of the AAA peroxins in dislocation of the cycling PTS1 receptor back to the cyto- J. Med. 356 (2007) 1736–1741. sol, Nat. Cell Biol. 7 (2005) 817–822. [129] S.J. Steinberg, G. Dodt, G.V. Raymond, N.E. Braverman, A.B. Moser, H.W. Moser, Perox- [100] I. Birschmann, A.K. Stroobants, M. Van Den Berg, A. Schäfer, K. Rosenkranz, W.H. isome biogenesis disorders, Biochim. Biophys. Acta 1763 (2006) 1733–1748. Kunau, H.F. Tabak, Pex15p of Saccharomyces cerevisiae provides a molecular [130] R.J. Wanders, H.R. Waterham, Peroxisomal disorders: the single peroxisomal basis for recruitment of the AAA peroxin Pex6p to peroxisomal membranes, enzyme deficiencies, Biochim. Biophys. Acta 1763 (2006) 1707–1720. Mol. Biol. Cell 14 (2003) 2226–2236. [131] H.W. Moser, Adrenoleukodystrophy: phenotype, genetics, pathogenesis and [101] N. Matsumoto, S. Tamura, S. Furuki, N. Miyata, A. Moser, N. Shimozawa, H.W. therapy, Brain 120 (Pt 8) (1997) 1485–1508. Moser, Y. Suzuki, N. Kondo, Y. Fujiki, Mutations in novel peroxin gene PEX26 [132] M.L. Slawecki, G. Dodt, S. Steinberg, A.B. Moser, H.W. Moser, S.J. Gould, Identifi- that cause peroxisome-biogenesis disorders of complementation group 8 pro- cation of three distinct peroxisomal protein import defects in patients with per- vide a genotype–phenotype correlation, Am. J. Hum. Genet. 73 (2003) 233–246. oxisome biogenesis disorders, J. Cell Sci. 108 (1995) 1817–1829. [102] Y. Fujiki, N. Miyata, N. Matsumoto, S. Tamura, Dynamic and functional assembly [133] M.S. Ebberink, P.A. Mooijer, J. Gootjes, J. Koster, R.J. Wanders, H.R. Waterham, of the AAA peroxins, Pex1p and Pex6p, and their membrane receptor Pex26p Genetic classification and mutational spectrum of more than 600 patients with involved in shuttling of the PTS1 receptor Pex5p in peroxisome biogenesis, a Zellweger syndrome spectrum disorder, Hum. Mutat. 32 (2011) 59–69. Biochem. Soc. Trans. 36 (2008) 109–113. [134] R.I. Kelley, N.S. Datta, W.B. Dobyns, A.K. Hajra, A.B. Moser, M.J. Noetzel, E.H. [103] H.W. Platta, M.O. Debelyy, F. El Magraoui, R. Erdmann, The AAA peroxins Pex1p Zackai, H.W. Moser, Neonatal adrenoleukodystrophy: new cases, biochemical and Pex6p function as dislocases for the ubiquitinated peroxisomal import studies, and differentiation from Zellweger and related peroxisomal poly- receptor Pex5p, Biochem. Soc. Trans. 36 (2008) 99–104. dystrophy syndromes, Am. J. Med. Genet. 23 (1986) 869–901. [104] O. Kerscher, R. Felberbaum, M. Hochstrasser, Modification of proteins by ubiquitin [135] N. Braverman, G. Steel, C. Obie, A. Moser, H. Moser, S.J. Gould, D. Valle, Human and ubiquitin-like proteins, Annu. Rev. Cell Dev. Biol. 22 (2006) 159–180. PEX7 encodes the peroxisomal PTS2 receptor and is responsible for rhizomelic [105] J.A. Kiel, K. Emmrich, H.E. Meyer, W.H. Kunau, Ubiquitination of the peroxisomal chondrodysplasia punctata, Nat. Genet. 15 (1997) 369–376. targeting signal type 1 receptor, Pex5p, suggests the presence of a quality con- [136] N.E. Braverman, A.B. Moser, S.J. Steinberg, Rhizomelic Chondrodysplasia Punctata trol mechanism during peroxisomal matrix protein import, J. Biol. Chem. 280 Type 1, in: R.A. Pagon, T.D. Bird, C.R. Dolan, K. Stephens (Eds.), GeneReviews (2005) 1921–1930. [Internet], University of Washington, Seattle, Seattle (WA), 2001. [106] H.W. Platta, W. Girzalsky, R. Erdmann, Ubiquitination of the peroxisomal import [137] R.J. Wanders, P. Vreken, S. Ferdinandusse, G.A. Jansen, H.R. Waterham, C.W. van receptor Pex5p, Biochem. J. 384 (2004) 37–45. Roermund, E.G. Van Grunsven, Peroxisomal fatty acid alpha- and beta-oxidation [107] H.W. Platta, F. El Magraoui, B.E. Baumer, D. Schlee, W. Girzalsky, R. Erdmann, in humans: enzymology, peroxisomal metabolite transporters and peroxisomal Pex2 and Pex12 function as protein–ubiquitin ligases in peroxisomal protein im- diseases, Biochem. Soc. Trans. 29 (2001) 250–267. port, Mol. Cell. Biol. 29 (2009) 5505–5516. [138] A.M. Motley, E.H. Hettema, E.M. Hogenhout, P. Brites, A.L.M.A. ten Asbroek, F.A. [108] C. Williams, M. van den Berg, E. Geers, B. Distel, Pex10p functions as an E3 ligase for Wijburg, F. Baas, H.S. Heijmans, H.F. Tabak, R.J.A. Wanders, B. Distel, Rhizomelic the Ubc4p-dependent ubiquitination of Pex5p, Biochem. Biophys. Res. Commun. chondrodysplasia punctata is a peroxisomal protein targeting disease caused by 374 (2008) 620–624. a non-functional PTS2 receptor, Nat. Genet. 15 (1997) 377–380. [109] A.F. Carvalho, M.P. Pinto, C.P. Grou, I.S. Alencastre, M. Fransen, C. Sa-Miranda, J.E. [139] G.A. Jansen, R.J. Wanders, Alpha-oxidation, Biochim. Biophys. Acta 1763 (2006) Azevedo, Ubiquitination of mammalian Pex5p, the peroxisomal import receptor, 1403–1412. J. Biol. Chem. 282 (2007) 31267–31272. [140] D.M. van den Brink, P. Brites, J. Haasjes, A.S. Wierzbicki, J. Mitchell, M. [110] K. Okumoto, S. Misono, N. Miyata, Y. Matsumoto, S. Mukai, Y. Fujiki, Cysteine Lambert-Hamill, J. de Belleroche, G.A. Jansen, H.R. Waterham, R.J. Wanders, ubiquitination of PTS1 receptor Pex5p regulates Pex5p recycling, Traffic 12 (2011) Identification of PEX7 as the second gene involved in Refsum disease, Am. J. 1067–1083. Hum. Genet. 72 (2003) 471–477. [111] H.W. Platta, F. El Magraoui, D. Schlee, S. Grunau, W. Girzalsky, R. Erdmann, [141] J.M. Powers, H.W. Moser, Peroxisomal disorders: genotype, phenotype, major Ubiquitination of the peroxisomal import receptor Pex5p is required for its neuropathologic lesions, and pathogenesis, Brain Pathol. 8 (1998) 101–120. recycling, J. Cell Biol. 177 (2007) 197–204. [142] R.J. Wanders, Peroxisomal disorders: clinical, biochemical, and molecular as- [112] C. Williams, M. van den Berg, R.R. Sprenger, B. Distel, A conserved cysteine is pects, Neurochem. Res. 24 (1999) 565–590. essential for Pex4p-dependent ubiquitination of the peroxisomal import recep- [143] R.J. Wanders, Metabolic and molecular basis of peroxisomal disorders: a review, tor Pex5p, J. Biol. Chem. 282 (2007) 22534–22543. Am. J. Med. Genet. 126A (2004) 355–375. [113] C.P. Grou, A.F. Carvalho, M.P. Pinto, S. Wiese, H. Piechura, H.E. Meyer, B. [144] B.E. Reuber, C.S. Collins, B.V. Geisbrecht, S.J. Gould, Disruption of a PEX1-PEX6 Warscheid, C. Sa-Miranda, J.E. Azevedo, Members of the E2D (UbcH5) family interaction is the most common cause of the neurologic disorders Zellweger mediate the ubiquitination of the conserved cysteine of Pex5p, the peroxisomal syndrome, neonatal adrenoleukodystrophy, and infantile Refsum disease, Proc. import receptor, J. Biol. Chem. 283 (2008) 14190–14197. Natl. Acad. Sci. U. S. A. 95 (1998) 8630–8635. [114] A.Y. Amerik, M. Hochstrasser, Mechanism and function of deubiquitinating [145] A.R. Shepard, N. Jacobson, J.C. Millar, I.H. Pang, H.T. Steely, C.C. Searby, V.C. enzymes, Biochim. Biophys. Acta 1695 (2004) 189–207. Sheffield, E.M. Stone, A.F. Clark, Glaucoma-causing myocilin mutants require [115] C.P. Grou, A.F. Carvalho, M.P. Pinto, S.J. Huybrechts, C. Sa-Miranda, M. Fransen, the peroxisomal targeting signal-1 receptor (PTS1R) to elevate intraocular pres- J.E. Azevedo, Properties of the ubiquitin-Pex5p thiol ester conjugate, J. Biol. sure, Hum. Mol. Genet. 16 (2007) 609–617. Chem. 284 (2009) 10504–10513. [146] Y.H. Kwon, J.H. Fingert, M.H. Kuehn, W.L. Alward, Primary open-angle glaucoma, [116] M.O. Debelyy, H.W. Platta, D. Saffian, A. Hensel, S. Thoms, H.E. Meyer, B. Warscheid, N. Engl. J. Med. 360 (2009) 1113–1124. W. Girzalsky, R. Erdmann, Ubp15p, a ubiquitin hydrolase associated with the [147] Z.T. Resch, M.P. Fautsch, Glaucoma-associated myocilin: a better understanding peroxisomal export machinery, J. Biol. Chem. 286 (2011) 28223–28234. but much more to learn, Exp. Eye Res. 88 (2009) 704–712. [117] C.P. Grou, T. Francisco, T.A. Rodrigues, M.O. Freitas, M.P. Pinto, A.F. Carvalho, P. [148] G.S. Zode, M.H. Kuehn, D.Y. Nishimura, C.C. Searby, K. Mohan, S.D. Grozdanic, K. Domingues, S.A. Wood, J.E. Rodríguez-Borges, C. Sá-Miranda, M. Fransen, J.E. Bugge, M.G. Anderson, A.F. Clark, E.M. Stone, V.C. Sheffield, Reduction of Azevedo, Identification of ubiquitin-specific protease 9X (USP9X) as a ER stress via a chemical chaperone prevents disease phenotypes in a mouse deubiquitinase acting on the ubiquitin–peroxin 5 (PEX5) thioester conjugate, model of primary open angle glaucoma, J. Clin. Invest. 121 (2011) 3542–3553. J. Biol. Chem. 287 (2012) 12815–12827. [149] P.B. Lazarow, Viruses exploiting peroxisomes, Curr. Opin. Microbiol. 14 (2011) [118] N. Miyata, K. Okumoto, S. Mukai, M. Noguchi, Y. Fujiki, AWP1/ZFAND6 Functions 458–469. in Pex5 Export by Interacting with Cys-monoubiquitinated Pex5 and Pex6 AAA [150] K.V. Mohan, C.D. Atreya, Novel organelle-targeting signals in viral proteins, ATPase, Traffic 13 (2012) 168–183. Bioinformatics 19 (2003) 10–13. [119] A. Hensel, S. Beck, F. El Magraoui, H.W. Platta, W. Girzalsky, R. Erdmann, [151] K.V. Mohan, I. Som, C.D. Atreya, Identification of a type 1 peroxisomal targeting Cysteine-dependent ubiquitination of Pex18p is linked to cargo translocation signal in a viral protein and demonstration of its targeting to the organelle, across the peroxisomal membrane, J. Biol. Chem. 286 (2011) 43495–43505. J. Virol. 76 (2002) 2543–2547. 1336 S. Nagotu et al. / Biochimica et Biophysica Acta 1822 (2012) 1326–1336

[152] N.E. Yelina, T.N. Erokhina, N.I. Lukhovitskaya, E.A. Minina, M.V. [160] M.T. Noguer, M. Martinez, Visual follow-up in peroxisomal-disorder patients Schepetilnikov, D.E. Lesemann, J. Schiemann, A.G. Solovyev, S.Y. Morozov, treated with docosahexaenoic acid ethyl ester, Invest. Ophthalmol. Vis. Sci. 51 Localization of Poa semilatent virus cysteine-rich protein in peroxisomes is (2010) 2277–2285. dispensable for its ability to suppress RNA silencing, J. Gen. Virol. 86 [161] M. Baes, P.P. Van Veldhoven, Generalised and conditional inactivation of Pex (2005) 479–489. genes in mice, Biochim. Biophys. Acta 1763 (2006) 1785–1793. [153] L.X. Liu, F. Margottin, S. Le Gall, O. Schwartz, L. Selig, R. Benarous, S. Benichou, [162] X. Li, E. Baumgart, J.C. Morrell, G. Jimenez-Sanchez, D. Valle, S.J. Gould, Binding of HIV-1 Nef to a novel thioesterase enzyme correlates with PEX11beta deficiency is lethal and impairs neuronal migration but does not Nef-mediated CD4 down-regulation, J. Biol. Chem. 272 (1997) 13779–13785. abrogate peroxisome function, Mol. Cell. Biol. 22 (2002) 4358–4365. [154] T. Wolff, R.E. O'Neill, P. Palese, Interaction cloning of NS1-I, a human protein that [163] P.A. Marshall, Y.I. Krimkevich, R.H. Lark, J.M. Deyer, M. Veenhuis, J.M. Goodman, binds to the nonstructural NS1 proteins of influenza A and B viruses, J. Virol. 70 PMP27 promotes peroxisomal proliferation, J. Cell Biol. 129 (1995) 345–355. (1996) 5363–5372. [164] O.I. Petriv, D.B. Pilgrim, R.A. Rachubinski, V.I. Titorenko, RNA interference of [155] M.C. McGuinness, H. Wei, K.D. Smith, Therapeutic developments in peroxisome peroxisome-related genes in C. elegans: a new model for human peroxisomal biogenesis disorders, Expert Opin. Invest. Drugs 9 (2000) 1985–1992. disorders, Physiol. Genomics 10 (2002) 79–91. [156] P.K. Dranchak, E. Di Pietro, A. Snowden, N. Oesch, N.E. Braverman, S.J. Steinberg, [165] H. Thieringer, B. Moellers, G. Dodt, W.H. Kunau, M. Driscoll, Modeling human J.G. Hacia, Nonsense suppressor therapies rescue peroxisome lipid metabolism peroxisome biogenesis disorders in the nematode Caenorhabditis elegans,J.Cell and assembly in cells from patients with specific PEX gene mutations, J. Cell. Sci. 116 (2003) 1797–1804. Biochem. 112 (2011) 1250–1258. [166] F.D. Mast, J. Li, M.K. Virk, S.C. Hughes, A.J. Simmonds, R.A. Rachubinski, A Dro- [157] R. Zhang, L. Chen, S. Jiralerspong, A. Snowden, S. Steinberg, N. Braverman, Recovery sophila model for the Zellweger spectrum of peroxisome biogenesis disorders, of PEX1-Gly843Asp peroxisome dysfunction by small-molecule compounds, Proc. Dis. Model Mech. 4 (2011) 659–672. Natl. Acad. Sci. U. S. A. 107 (2010) 5569–5574. [167] H. Chen, Z. Liu, X. Huang, Drosophila models of peroxisomal biogenesis disorder: [158] A. Itoyama, M. Honsho, Y. Abe, A. Moser, Y. Yoshida, Y. Fujiki, Docosahexaenoic peroxins are required for spermatogenesis and very-long-chain fatty acid acid mediates peroxisomal elongation, a prerequisite for peroxisome division, metabolism, Hum. Mol. Genet. 19 (2010) 494–505. J. Cell Sci. 125 (2012) 589–602. [168] M. Nakayama, H. Sato, T. Okuda, N. Fujisawa, N. Kono, H. Arai, E. Suzuki, M. [159] S.A. Abdel Maksoud, H.T. El-Bassyouni, H.H. Afifi, A.G. Khalifa, The role of DHA in Umeda, H.O. Ishikawa, K. Matsuno, Drosophila carrying Pex3 or Pex16 muta- amelioration of symptoms of peroxisome biogenesis disorders, Res. J. Med. Med. tions are models of Zellweger syndrome that reflect its symptoms associated Sci. 2 (2007) 5–13. with the absence of peroxisomes, PLoS One 6 (2011).