Biochemical and Biophysical analysis of the GTPase Activating Proteins of the small guanine nucleotide binding protein Rap1 and RheB

Dissertation

zur Erlangung des Grades Doktor der Naturwissenschaften der Fakultät Chemie der Ruhr-Universität Bochum

vorgelegt von Partha Pratim Chakrabarti aus Calcutta / Indien

Bochum 2005

Die vorliegende Arbeit wurde im Zeitraum Juli 2001 bis Januar 2005 in einer Kooperation mit der Arbeitsgruppe von Prof. Dr. Klaus Gerwert (Lehrstuhl für Biophysik, Ruhr-Universität Bochum) am Max-Planck-Institut für molekulare Physiologie in Dortmund unter der Anleitung von Prof. Dr. Alfred Wittinghofer (Abteilung für Strukturelle Biologie) angefertigt.

1. Berichterstatter: Prof. Dr. Alfred Wittinghofer 2. Berichterstatter: Prof. Dr. Klaus Gerwert 3. Berichterstatter: Prof. Dr. Rolf Heumann

Mündliche Prüfung: 27th June, 2005

Index

Index

1 Introduction...... 1

1.1 Guanine nucleotide binding proteins ...... 1 1.1.1 The GTPase cycle...... 1 1.1.2 Overview of the GTPase superfamily...... 2 1.1.3 Biochemical and structural features of small GNBPs...... 3 1.1.4 Fabrications on the minimal G domain...... 6 1.1.5 The Ras signal transduction pathway ...... 7 1.1.6 Biology of Rap ...... 8 1.1.7 The TOR pathway ...... 10 1.1.8 Phosphoryl transfer and the hydrolysis reaction ...... 12 1.2 GTPase Activating Proteins...... 14 1.2.1 RasGAP and Gα: the arginine finger in trans or cis...... 15 1.2.2 RhoGAP and bacterial toxins ...... 16 1.2.3 GAPs without an arginine finger...... 17 1.2.4 GTPase activating proteins of the Rap1GAP family ...... 19 1.2.5 Summary of biochemistry of Rap1GAP and Tsc1/2...... 23 1.3 Objectives of this work...... 24

2 Materials and Methods ...... 25

2.1 Materials...... 25 2.1.1 Chemicals...... 25 2.1.2 18O labeled caged-GTPs ...... 25 2.1.3 ...... 25 2.1.4 Kits...... 25 2.1.5 Columns ...... 26 2.1.6 Microorganisms ...... 26 2.1.7 Mammalian cell lines ...... 26 2.1.8 Media, special reagents and antibiotics ...... 26 2.1.9 Antibodies...... 27 2.1.10 Special Buffers ...... 27 2.2 Molecular biology methods...... 27 2.2.1 Agarose gels...... 27 2.2.2 Isolation of plasmid DNA ...... 28 2.2.3 Polymerase chain reaction (PCR) ...... 28 2.2.4 Ligation ...... 28 2.2.5 Competent cells...... 28 2.2.6 Transformation ...... 28 2.2.7 Bacteria storage ...... 29 2.2.8 Site specific mutagenesis...... 29 2.2.9 DNA sequencing...... 29 I Index

2.2.10 Constructs ...... 29 2.2.11 Point mutants...... 30 2.3 Biochemical methods ...... 30 2.3.1 SDS-PAGE ...... 30 2.3.2 Determination of protein concentration ...... 30 2.3.3 Matrix assisted laser desorption ionisation (MALDI) ...... 30 2.3.4 Test expression of proteins ...... 30 2.3.5 Protein overexpression and preparation of soluble bacteria extract ...... 31 2.3.6 Purification of Rap1GAP75-415 ...... 32 2.3.7 Purification of Rap1B1-167 and mutants...... 32 2.3.8 Purification of RheB1-170...... 32 2.3.9 Rap1-Aedans preparation ...... 33 2.3.10 Nucleotide exchange...... 33 2.3.11 Nucleotide detection using reversed-phase HPLC ...... 34 2.3.12 Fast kinetics using stopped-flow measurement ...... 34 2.3.13 Quenched Flow analysis ...... 35 2.3.14 Electrospray ionization mass spectrometry (ESI-MS)...... 35 2.4 Cell Biological methods ...... 36 2.4.1 Cell culture...... 36 2.4.2 Cryopreservation ...... 36 2.4.3 Transfection...... 36 2.4.4 β-Galactosidase assay ...... 37 2.4.5 Immunoprecipitation ...... 37 2.4.6 Western blot ...... 37 2.4.7 Immunostaining ...... 38 2.5 Time resolved infrared spectroscopy...... 39 2.5.1 Instrumentation...... 39 2.5.2 Basic principles of Infrared spectroscopy...... 41 2.5.3 Sample preparation ...... 41 2.5.4 Initiation of the reaction and data collection ...... 42 2.5.5 Measurement parameters ...... 44 2.5.6 Data analysis ...... 44 2.5.7 Stability of spectral regions...... 45 2.5.8 Reproducibility within the phosphate region...... 46

3 Results...... 48

3.1 FTIR on Rap•RapGAP Reaction ...... 48 3.1.1 Summary of previous FTIR studies on Ras and RasGAP ...... 48 3.1.2 Intrinsic reaction of Rap...... 50 3.1.3 RapGAP catalyzed reaction...... 53 3.1.4 A to B Transition...... 54 3.1.5 B to C Transition...... 58 3.1.6 Reversibility of the GTPase reaction and 18O exchange...... 60 3.1.7 Summary of band assignments...... 62 3.1.8 Kinetic scheme...... 63

II Index

3.1.9 Time Course of The Reaction...... 63 3.1.10 Conclusion...... 65 3.2 analysis of Rap•RapGAP system ...... 68 3.2.1 Effect of T61 and G12 replacement ...... 68 3.2.2 Comparison of hydrolysis rates from HPLC measurement...... 70 3.2.3 Replacement of catalytic aspargine 290 of Rap1GAP ...... 72 3.2.4 Cross activation ...... 73 3.2.5 Transition state mimicking for Rap•RapGAP reaction...... 75

3.2.6 AlFx experiment for Rap and Rap1GAP mutants ...... 77 − 3.2.7 BeF 3 : ground stare mimic or transition state mimic?...... 78 3.2.8 GAP activation of T61Q mutant with FTIR spectroscopy...... 79 3.2.9 Conclusion...... 80 3.3 An extension to RheB-Tsc1/Tsc2 system...... 82 3.3.1 Initial attempts of protein expression from Cos7 cells...... 83 3.3.2 Putative dimerization domain of Hamartin: a bioinformatics approach ...... 84 3.3.3 Fragment designing for recombinant expression and purification ...... 85 3.3.4 Biochemistry of Tsc21538-1729, Tsc21520-1763 and Tsc1645-872 fragments ...... 86 3.3.5 Conclusion...... 87

4 Discussion...... 88 4.1.1 Outcome of time resolved FTIR spectroscopy ...... 88 4.1.2 Assignment of amino acid bands ...... 89 4.1.3 The novel hydrolysis mechanism ...... 90 − 4.1.4 BeF 3 association for Rap mutants ...... 91 4.1.5 Downregulation of Rap: is the strategy unified or diverse?...... 93 4.1.6 Systematic strategy for Tuberin and Hamartin...... 94 4.1.7 A possible Tuberous sclerosis Chimera...... 94

5 Summary ...... 97

6 Zusammenfassung ...... 99

7 References ...... 101

8 Appendix ...... 118

8.1 Abbreviations...... 118

8.2 Mass spectra of Aedans-Rap (A86C) and mutants ...... 119

8.3 Scientist model for global fit analysis...... 122

8.4 Sequence allignment of Tsc1 ...... 123

Danksagung...... 125

III Index

Teilpublikationen dieser Arbeit ...... 126

Erklärung...... 127

Lebenslauf ...... 128

IV Introduction 1 Introduction

1.1 Guanine nucleotide binding proteins

1.1.1 The GTPase cycle

Guanine nucleotide binding proteins (GNBPs) are involved in a variety of cellular processes including signal transduction (Ras related small , Heterotrimeric G proteins (Bourne, 1993)), vesicular transport (/Ypt (Pfeffer, 2001)), cell growth and differentiation (Ras (Reuther and Der, 2000)), receptor mediated endocytosis ( (Hinshaw, 2000)), protein synthesis (elongation and initiation factors (Rodnina and Wintermeyer, 2001)) and protein targeting (signal recognition particle (Doudna and Batey, 2004)). Most GNBPs regulate these processes at the expense of GTP rather than providing mechanochemical work or chemical synthesis. An estimated number of 200 GNBPs exist in eukaryotic cells (Vetter and Wittinghofer, 2001), most of which use a conserved switch mechanism by cycling between a GDP bound ‘OFF’ state and a GTP bound ‘ON’ state as shown in Figure 1.1) (Bourne et al., 1991). The nucleotide state (tri- or diphosphate) determines the function since the downstream effector molecules can only interact with the triphosphate form. The activation (switch ON) of the GNBPs involves the exchange of GDP to GTP and this intrinsically slow process is stimulated by guanine nucleotide exchange factors (GEFs). Depletion of the terminal phosphate of GTP through hydrolysis brings the GNBP to its resting state. This intrinsically slow reaction is catalyzed by GTPase activating proteins (GAPs).

Figure 1.1. Schematic representation of GTPase cycle (after Bourne et al., 1991) shows switching between ‘Off’ and ‘On’ states [This figure was kindly provided by Dennis Fiegen].

In some specific cases (Rho and Rab GTPases), another set of regulators; guanine nucleotide dissociation inhibitors (GDIs) are involved in the cycle. GDIs prevent GDP dissociation and mask the attached isoprenyl group of the GNBP that anchors it to the

1 Introduction membrane, thereby retaining the GNBP in the cytosol (Wu et al., 1996). In order to facilitate membrane attachment and thus activation of GNBPs, another set of regulatory molecules named GDI displacement factors (GDFs) are involved to release bound GDIs (Geyer and Wittinghofer, 1997).

1.1.2 Overview of the GTPase superfamily

Proteins that bind and hydrolyze nucleotide triphosphates are crucial for almost all aspects of life. They belong to several chain folds, most notably, the dinucleotide binding (Rossmann) fold and the related / FtsZ fold, the mononucleotide binding fold (P-loop NTPases), the protein kinase fold, the histidine kinase / HSP90 / TopoII fold, and HSP70 / RNAse H fold (Leipe et al., 2002). The GTPase superclass can be divided into two large classes, each of which has a unique set of sequence and structural signatures. The first class named TRAFAC (after transcription factors) includes enzymes required in translation (initiation, elongation and release factors), signal transduction (in particular the extended Ras like family), cell motility and intracellular transport. The TRAFAC class thus contains the translation factor superfamily, the - and the Ras like superfamily. The second class, designated SIMIBI (after signal recognition particle, MinD and BioD) consists of signal recognition particle (SRP) GTPases, the assemblage of MinD like which are involved in protein localization, chromosome partitioning and membrane transport, and a group of metabolic enzymes with kinase or related phosphate transfer activity. The two classes together contain over 20 distinct families that are further subdivided into 57 subfamilies on the basis of conserved sequence motifs, shared structural features and domain architecture.

The best-known members of the SIMIBI class are the signal recognition particle GTPases (Fth in bacteria) and the alpha subunit of the signal recognition particle receptor (FtsY in bacteria) (Freymann et al., 1997). They exist in all three kingdoms of life and are involved in co-translational cellular targeting of nascent secretary and membrane proteins. The initiation factor IF2 (elF5B in eukaryotes), elF2γ and elongation factors EF-Tu (and the homologous protein SelB) and EF-G are four members of the translation factor superfamily that appear ubiquitously in bacteria and eukaryotes (Leipe et al., 2002). elF2γ forms a ternary complex met with GTP and Met-tRNAi to the ribosome (Sonenberg and Dever, 2003). GTP hydrolysis in elF2γ is triggered in a reaction which requires among other initiation factors, IF2 / elF5B. The eukaryotic cellular motor ATPases kinesin and myosin were grouped in the myosin-kinesin superfamily of TRAFAC GTPases. Liepe et al. argued that these proteins have evolved from an ancestral GTPase at the onset of eukaryotic evolution and have lost their specificity towards GTP. Members of the Dynamin family that are involved in the budding of clathrin- coated vesicles (Urrutia 1997), and GB1 family members including GBPs (Prakash et al., 2000) are also grouped in this superfamily.

2 Introduction

Members of the exist mainly in eukaryotes and have been classified differently (Bourne et al., 1990; Garcia-Ranea and Valencia, 1998; Leipe et al., 2002). The group of monomeric 20-25 kD GNBPs with Ras as the founding member is called the small GNBPs. More than 100 GNBPs have been identified from yeast to human (reviewed in (Takai et al., 2001)) and they were initially classified as a separate superfamily comprising Ras, Rab, Rho, and Arf / Sar1 sub-families (Garcia-Ranea and Valencia, 1998). Within the Ras superfamily, each protein shares high sequence similarity with 50-55% amino acid identity, whereas the Rho and Rab family members share ~30% amino acid identity with the Ras family members. Figure 1.2 shows a dendrogram of small GTP binding proteins.

Figure 1.2. Dendrogram of the small GTP binding (Takai et al., 2001).

However, solely based on sequence, recently it was suggested that the family of heterotrimeric GNBPs should also be included in this superfamily (Leipe et al., 2002).

1.1.3 Biochemical and structural features of small GNBPs

Ras, a small GNBP, is a paradigm of the entire family of small GTP binding proteins, both structurally, and biochemically. It is anchored to the plasma membrane by means of a farnesyl and a palmitoyl group (H and N-Ras), which are required for localization and function in vivo. It consists of 189 amino acids with five conserved sequence elements, which are also found in other GNBPs and are necessary for guanine nucleotide binding and hydrolysis. Residues 1-166 form the minimal G domain, which consists of a mixed six stranded β sheet and five helices, located on both sides (Figure 1.3).

Like all other small GNBPs, Ras binds both GTP and GDP with very high affinity, in the picomolar range. No other standard nucleotide binds Ras with similar affinity, indicating a

3 Introduction high specificity for guanine nucleotides. The β-phosphate is required since guanosine monophosphate (GMP) has a 106 fold reduced affinity compared to GTP / GDP. Again, the binding affinity strongly depends on the presence of a bivalent cation, typically magnesium 2+ ion (Mg ). When magnesium ions are absent, the dissociation rate constant (koff) increases several hundred folds leading to a concomitant increase in the equilibrium dissociation constant (KD) (John et al., 1988).

Figure 1.3. Structure of Ras in the triphosphate analogue (GppNHp) form. PM and G motifs are involved either in binding phosphate and Mg2+ or the guanosine base respectively. (Wittinghofer and Waldmann, 2000)

Structurally, the GNBPs contain four to five conserved sequence elements (Bourne et al., 1990; Valencia et al., 1991; Sprang, 1997; Wittinghofer and Waldmann, 2000; Vetter and Wittinghofer, 2001). The PM1 motif 10GxxxxGKS/T forms a loop, which is involved in binding charged phosphate groups. The main chain nitrogen atoms of residues 13-16 and the side chain of Lys16 point towards the negatively charged phosphate groups thereby creating a polarized environment for these groups. The Ser 17 residue is important for Mg2+ binding for both GTP and GDP. Substitution of this residue by Ala or Asn weakens the affinity to nucleotide and in turn strengthens the affinity to guanine nucleotide exchange factors in a sense that GEFs are more easily sequestered. This dominant negative mutation inhibits signaling through GEFs. The threonine (T35) residue of PM2 is a direct ligand of Mg2+ and a key residue that triggers conformational change after hydrolysis. The 57DxxGQ/H/T motif involves the aspartate group to bind Mg2+ through a water molecule. The Glycine 60 residue, which is also important for conformational change is coordinated to the γ-phosphate via main chain contact. In 116N/TKxD (G2) both Asn and Lys are involved in the interaction to various regions of the base. The aspartate residue forms a double hydrogen bond to the guanine base and is thereby the major contributor for the high specificity for the guanine nucleotides. The G3 motif 145(S/C)A(K/T/L) (Serine S145 in Ras) is located between strand β6 and helix α5. It doesn’t participate in the guanine base recognition, besides a main chain interaction

4 Introduction between alanine 146 and the keto-oxygen of the base. The network of conserved interactions between Ras and the GTP analogue GppNHp is shown in Figure 1.4.

Figure 1.4. Interactions of the Ras-bound triphosphate analogue GppNHp with selected, usually conserved residues in Ras. mc = peptide main chain. (Wittinghofer and Waldmann, 2000)

The switch regions (in blue in Figure 1.3), switch I (amino acids 32-40 in Ras) and switch II (amino acids 61-67 in Ras) are the main determinants for effector binding. While the GDP bound forms of the GNBPs show larger variations in the structural details, the GTP (analogue) bound forms are very similar (Sprang, 1997; Vetter and Wittinghofer, 2001). The most important contributions of the triphosphate form are two additional hydrogen bonds from γ-phosphate oxygens to the main chain NH groups of the invariant Thr and Gly residues (Thr 35 and Gly 60 in Ras) from switch I and switch II respectively. This has been described as a conformational trigger, which is set upon hydrolysis of GTP. Release of the γ phosphate allows the switch regions to relax to a different conformation. The extent of conformational change depends on the protein and involves extra elements for some of them. This is depicted in Figure 1.5.

Figure 1.5. Conformation of GNBPs. The switch regions of the G domains are much more divergent in GDP bound state. Extra elements on the G domain are indicated in different color. (Vetter and Wittinghofer, 2001)

5 Introduction

1.1.4 Fabrications on the minimal G domain

Apart from the small GNBPs, the structures of several α subunits of the heterotrimeric G proteins like Gtα, Giα and Gsα, protein synthesis factors like EF-Tu, EF-G, IF2 and some of the larger GTPases like antiviral human guanylate binding protein GBP1 (hGBP1), have been solved. The straightforward way (Vetter and Wittinghofer, 2001) to compare these structures is to compare the G domains and describe the additional features as ‘tema con variazioni’ as depicted in Figure 1.6.

Figure 1.6. Extra domains of the GNBPs in comparison to Ras and their relative location. The G domain is depicted in yellow worm plot with the switches colored. The additional domains are shown in different colors (Vetter and Wittinghofer, 2001).

The Ras related Rho proteins contain a α-helical insertion of about 13 residues (Figure 1.6) while Arf, Arl and Sar1 proteins contain an amino terminal extension necessary for interaction with the membrane. Ran has an elongated carboxy terminus (Figure 1.5) crucial for its function in nuclear transport. Gα proteins have several extensions and insertions into the G domain, one of which is an independently folding α-helical domain. Human guanylate binding protein (hGBP1) has a 300 residue extended G domain and a carboxy terminal coiled coil domain (Prakash et al., 2000). Obg, a protein involved in regulation of stress response and sporulation in Bacillus subtilis is a new type of guanine nucleotide binding protein. The structure (Buglino et al., 2002; Kukimoto-Niino et al., 2004) reveals an N terminal glycine rich domain with an unique elongated barrel like fold (Obg fold / OCT fold) consisting of an eight stranded β-sheet interfacing the G domain and a six helix bundle, a Ras like G domain, and a C terminal TGS domain, commonly found in several proteins that are involved in bacterial stress response. In the G domain, Obg has an additional β-strand in switch I and a α-helix in the switch II region.

6 Introduction

EF-Tu is a three-domain protein containing an N terminal G-domain (domain 1), a β-barrel middle domain (domain 2) and a C-terminal domain (domain 3) (Kjeldgaard and Nyborg, 1992; Kjeldgaard et al., 1993). EF-Tu forms a ternary complex with aminoacyl-tRNA and controls incorporation of the correct amino acids into the peptide chain (reviewed in (Ogle et al., 2003)). The relative location of the middle and C-terminal domain changes between GTP bound active state or GDP bound inactive state (Berchtold et al., 1993). The highly elongated five-domain protein EF-G catalyzes translocation of tRNA on the ribosome. The origin of other factors involved in protein biosynthesis, e.g. the release factors, differs in the three kingdoms of life (Leipe et al., 2002). The structure of EF-G (Czworkowski et al., 1994; Aevarsson et al., 1994; al Karadaghi et al., 1996) reveals an N-terminal G domain, which is surrounded by the second and third domain (poorly defined) and the C-terminal domain. The fourth domain stands away from the other domains. Although the structure of GTP bound state is known, it has been proposed that the active state of EF-G might mimic the inactive state of EF-Tu-tRNA complex and vice versa (Liljas et al., 1995; Liljas and al Karadaghi, 1997).

The G-domain of SRP (signal recognition particle) and its receptor SR show a divergent topology of the β sheet, in addition to an extension and insertion. Structural studies (reviewed in (Doudna and Batey, 2004)) reveal that SRP has three distinct domains: the N terminal N- domain, the centrally located G-domain, and C terminal M-domain (methionine abundant). The N and G domains are responsible for signal sequence recognition and the M-domain presumably contributes to SRP-RNA binding. Bacterial SRP, FtsY, does not have the M domain. Tubulin and its bacterial homolog FtsZ on the other hand, are structurally unrelated.

1.1.5 The Ras signal transduction pathway

Ras was originally identified as the active principle of sarcoma viruses and was called an oncogene. Later on it became apparent that there are cellular counterparts existing in mammals (reviewed in (Takai et al., 2001; Malumbres and Barbacid, 2003)). In human, these are approximately 21 kD proteins with three different isoforms, H-Ras, K-Ras and N-Ras, mainly differing at the C-terminus. Growth factors like EGF and PDGF bind to the extracellular domain of the specific receptor tyrosine kinases, which lead to receptor dimerization, and subsequent transphosphorylation of the tyrosine residues in the cytoplasmic part of the receptor. Such phosphorylated tyrosines are in turn recognized by proteins having specific phosphotyrosine binding domains, e.g. SH2 domain (src homology domain). This way, the adaptor protein Grb2 binds via its SH2 domain to the phosphorylated growth receptor. SOS (protein from Son of Sevenless gene), a Ras specific guanine nucleotide exchange factor, becomes associated to the membrane through Grb2 and promotes GDP to GTP exchange of Ras, thereby activating it. Activated Ras subsequently activates a downstream effector, such as Raf kinase, which subsequently activates MEK

7 Introduction kinase (Figure 1.7). MEK kinase activates the protein kinase ERK. Finally ERK phosphorylates a class of transcription factors named the ternary complex factors, which associate with a second class of transcription factors, the serum response factors. The resulting protein complex binds to a conserved promoter element and initiates transcription. The pathway is finally terminated by the action of GAPs and by the dissociation of the

Figure 1.7. Simplified scheme of Ras signaling pathway (Takai et al., 2001). For explanation, see text.

SOS-Grb2 complex upon phosphorylation by ERK. Point mutations at residues 12, 13 and 61 render Ras unable to hydrolyze GTP, even in the presence of GAP (Trahey and McCormick, 1987), thus leading to constitutively active Ras. These mutations are also found in human tumors (Almoguera et al., 1988; Bos, 1989) indicating that a constitutively active Ras pathway leads to tumor formation. Now it is estimated that 30% of human tumors carry activated forms of Ras oncogene. Thus, the Ras pathway is an important target for cancer therapy (Wittinghofer and Waldmann, 2000; Downward, 2003). Ras also triggers different pathways, e.g. by activating Type I Phosphoinositol-3 kinase (PI3K), the exchange factor for the small GTPase Ral, RalGEF and Phospholypase Cε (PLCε) (reviewed in (Downward, 2003)).

1.1.6 Biology of Rap

At the sequence level, Rap has the characteristic replacement of Glutamine 61 to a threonine (Figure 1.8). Unlike H and N-Ras, which are modified by a farnesyl and palmitoyl anchors, Rap bears a geranyl-geranyl anchor which attaches this protein to the plasma membrane. Rap was identified by low stringency hybridization of various cDNA libraries with Ras cDNA (Pizon et al., 1988) and soon after as K-rev1 in a screen of cDNAs which can revert the phenotype of K-Ras transformed fibroblast (Kitayama et al., 1989). Rap1 homologues are found in all vertebrates, in Drosophila and in yeast (reviewed in (Bos et al., 2001)). The yeast homologue Bud1 is involved in selecting the new budding site by recruiting and activating 8 Introduction effectors to the selected region in the cell (Park et al., 2002). In Drosophila, maternal and zygotic Rap1 expression is essential for development of the embryo, imaginal disc development and oogenesis (Asha et al., 1999). Drosophila Rap1 was shown to be enriched at adherens junctions, particularly between newly divided sister cells, and to regulate the position of these junctions thereby regulating cell adhesion (Knox and Brown, 2002).

In human, five isoforms of Rap exist, namely Rap1a, Rap1b, Rap2a, Rap2b and Rap2c. Among these, Rap1a and Rap1b (Rap1 from hereon) share more than 90% sequence identity and are similar in biochemical properties. Rap1 in human was shown to be ubiquitously expressed. In fibroblast, it is located in mid-Golgi compartment and early and late endosomes (Beranger et al., 1991; Pizon et al., 1994). However, in platelets and neutrophiles, after activation by different stimuli, it translocates from granules to the plasma membrane (Franke et al., 1997). Such a stimuli can be activation the of tyrosine kinases, heterotrimeric G-protein-coupled receptors and cell adhesion molecules (McLeod et al., 1998; Zwartkruis et al., 1998; Posern et al., 1998). Second messengers like cyclic AMP, Ca2+ and diacylglycerol (DAG) are commonly involved in transducing extracellular signals to Rap1 (Bos et al., 2001) by activating Rap specific GEFs, such as cAMP dependent GEF Epacs (de Rooij et al., 1998) or Ca2+ / DAG dependent CalDAG GEFs (Kawasaki et al., 1998). Two other group of Rap specific GEFs, namely C3G (Gotoh et al., 1995) and PDZ-GEFs (de Rooij et al., 1999) also play important role in Rap1 activation.

From a functional point of view, Rap1 attracted much attention because of the possibility that it regulates Ras-mediated signalling. With greater than 50% sequence identity, especially in the switch regions (Figure 1.8), Rap1 is the closest relative of Ras. Based on the striking similarity of the switch I regions of Ras and Rap, effector interactions have been analyzed ((Nassar et al., 1996), reviewed in (Herrmann, 2003)). All Ras effectors examined can also interact with the GTP bound form of Rap; however, such interaction does not lead to activation of all effectors, as shown for the Raf kinase isoform, Raf1 (Shirouzu et al., 1998). The arrows in Figure 1.8 indicate the amino acid residues of Ras which interact with one of the effectors, phosphatidylinositol 3-kinase γ (PI3K-γ) (Pacold et al., 2000). The striking similarity of the effector interaction between Rap and Ras is the structural view of the Ras antagonism hypothesis (Kitayama et al., 1989) of Rap.

Figure 1.8. Partial sequence alignment of Ras and Rap1 (Bos et al., 2001). Arrows indicate amino acid residues of Ras which interact with PI3K. The asterisk indicates residue 61, which is not a glutamine (Q) in Rap family.

9 Introduction

In line with the suppression of K-Ras action in fibroblasts and the recent finding that Rap1•GTP can inhibit cell proliferation in keratinocytes (Mitra et al., 2003), this suggested initially that Rap1 has an antagonistic role in Ras signaling by sequestering common effectors in an inactive state.

However, many lines of evidence suggest now that Rap1 is also able to activate signal transduction pathways independently of Ras. When microinjected into Swiss3T3 fibroblasts, Rap1 is able to induce DNA synthesis and morphological changes (Altschuler and Ribeiro- Neto, 1998). In fibroblasts, Rap1 activation fails to interfere with Ras-dependent ERK activation (Zwartkruis et al., 1998). In PC12 cells, the Raf-isoform B-Raf is activated by Rap1•GTP and activates ERK, independently of Ras action (Vossler et al., 1997; Kao et al., 2001). B-Raf and likewise Ral-GEF activity are also stimulated in vitro by Rap1•GTP (Ohtsuka et al., 1996). Furthermore, Rap1 was shown to be involved in the process of learning and memory (Morozov et al., 2003), in the development of leukemia (Ishida et al., 2003), in angiogenesis and cerebrovascular diseases (Sahoo et al., 1999).

Apart from Ras antagonism, the best characterized function of Rap is integrin-mediated cell adhesion referred to as inside out signalling (Katagiri et al., 2000; Reedquist et al., 2000; de Bruyn et al., 2002; Sebzda et al., 2002). A new Rap1 effector molecule, RAPL, was identified which is indispensable for the integrin-mediated adhesion and migration of lymphocytes and dendritic cells where the effect is an outcome of linking Rap1 to the integrin LFA-1 (Katagiri et al., 2002; Katagiri et al., 2004).

1.1.7 The TOR pathway

The TOR pathway controls cell size and cellular proliferation (Volarevic et al., 2000; Fingar et al., 2002). Target of Rapamycin (TOR) (Brown et al., 1994; Sabatini et al., 1994; Zheng et al., 1995) is an essential serine/threonine kinase belonging to the phosphoinositide-kinase related kinase family (PIKK) that is conserved in eukaryotes from yeast (Saccharomyces cerevisiae) to plants (Arabidopsis thaliana), worms (Caenorhabditis elegans) and flies (Drosophila melanogaster) (Jacinto and Hall, 2003). TOR directly or indirectly regulates the translation of ribosomal proteins (Fingar and Blenis, 2004), and at least in yeast, ribosome biogenesis (Powers and Walter, 1999). This protein functions as a sensor of mitogens, energy (Dennis et al., 2001) and nutrient levels (Kim et al., 2002; Proud, 2004), acting as a gatekeeper for cell cycle progression from G1 to S phase and is positively regulated by AKT that phosphorylates and inactivates negative regulators such as tuberin (Tsc2) (Potter et al., 2002; Manning et al., 2002; Inoki et al., 2002; Gao et al., 2002b). Pathways upstream of TOR are often activated in cancer (reviewed in (Vivanco and Sawyers, 2002; Harris and Lawrence, Jr., 2003; Bjornsti and Houghton, 2004; Inoki et al., 2005)) which can be due to increased activity of PI3K or AKT, kinases which regulate Tsc2, or through mutations which inactivate Tsc2 (Li et al., 2004). The Tsc1-Tsc2 heterodimer is responsible in the autosomal

10 Introduction dominant disorder Tuberous Sclerosis, which leads to the development of Hamartomas, typically benign tumors that often contain very large cells (Gomez et al., 1999; Ito and Rubin, 1999) causing serious defects in multiple organ systems (reviewed in (Krymskaya, 2003)).

How the Tsc complex regulates TOR signalling was clarified by the discovery that Tsc2 negatively regulates a small G protein – RheB (Ras Homolog enriched in Brain) (Tee et al., 2003; Garami et al., 2003; Zhang et al., 2003; Inoki et al., 2003a). Tsc2 contains partial sequence homology to Rap1GAP (see 1.2.4) in its GAP domain and in singular publications, has been reported to weakly stimulate GTP hydrolysis of Rap1A (Wienecke et al., 1995) and Rab5 (Xiao et al., 1997) in vitro and possibly also Rho in vivo (Astrinidis et al., 2002). Complementary data derived from genetics and biochemistry both in Drosophila and mammalian systems placed RheB between TOR and the Tsc complex. Genetic screens in Drosophila compound eye identified RheB as a gene that, when mutated, decreased and when overexpressed, increased eye and cell size (Saucedo et al., 2003; Stocker et al., 2003). Consistent with the genetic results, RheB overexpression in mammalian cell increases S6K1 (ribosomal S6 kinase 1) and 4EBP1 phosphorylation in a nutrient and growth factor independent manner that is sensitive to rapamycin but insensitive to wortmannin (PI3K inhibitor), indicating that RheB lies upstream of TOR but downstream of PI3K (Castro et al., 2003; Garami et al., 2003).

In summary, Insulin binding activates the insulin receptor (IR) tyrosine kinase, which phosphorylates IRS-1 or IRS-2 (IRS: Insuline receptor substrate) (Figure 1.9). PI3K binds to

Figure 1.9. Schematic representation of the TOR pathway (Harris and Lawrence, Jr., 2003). For explanation, see text. phosphorylated IRS by Src homology domain 2 (SH2) domains in the p85 regulatory subunit. This interaction activates the p110 catalytic subunit which phosphorylates phosphatidylinositol, generating products that recruit the kinase Akt to the membrane, where

11 Introduction it is phosphorylated and activated by phospholipid-dependent kinase 1 (PKD-1). Akt phosphorylates TOR directly and also Tsc2 which decreases the RheB-GAP activity of the Tsc1-Tsc2 complex through an undefined mechanism. Consequent accumulation of RheB- GTP leads to activation of TOR. So far no exchange factor (GEF) has been identified for RheB. Raptor (regulatory associated protein of mTOR) and mLST8 (also called GβL) are mTOR-associated proteins and the downstrem targets are 4EBP1 and S6K1 (ribosomal S6 kinase 1).

It is interesting to note the presence of AMP-activated protein kinase (AMPK) in this pathway, which senses the AMP / ATP ratio in a cell. Activated AMPK can phosphorylate Tsc2, enhancing the ability of the Tsc complex to inhibit mTOR (Inoki et al., 2003b). Thus, energy sufficiency is sensed by AMPK to modulate the function of the Tsc-complex, which in turn regulates TOR function. Hence, TOR may not directly sence ATP levels, as was originally suggested (Dennis et al., 2001), rather may be regulated by the AMP / ATP ratio via AMPK, which in turn is activated by another tumorsuppressor LKB1 (Peutz-Jeghers syndrome kinase) (Woods et al., 2003; Shaw et al., 2004; Lizcano et al., 2004).

The diverse signalling function of Rap1 and RheB are terminated by GTP hydrolysis of the GNBPs, which are stimulated by respective GTPase activating proteins, Rap1GAP and the Tsc1/2 complex. Such hydrolysis reactions are described as “Phosphoryl transfer” and is discussed in the following section 1.1.8. Different GTPase activating proteins and their diversified strategies of catalyzing the otherwise slow intrinsic GTPase reaction are discussed in section 1.2.

1.1.8 Phosphoryl transfer and the hydrolysis reaction

Phosphoryl transfer reactions in biology have been studied for several decades because of the high abundance of phosphate esters and anhydrides in the living world. Phosphoric acid is specially suited for its role in nucleic acids because it can link two nucleotides and still ionize; the resulting negative charges stabilize the diester against hydrolysis. This also holds for phosphates, which are intermediate metabolites, and the ones that serve as energy sources [adenosine triphosphate (ATP), creatine phosphate, phosphoenolpyruvate]. Again phosphate with multiple charges can react in a similar way as that of the monomeric − metaphosphate ion (PO3 ) in case of nucleophilic reaction. No other residue appears to fulfill such a versatile role as that of phosphate in biochemistry (Westheimer, 1987).

The classical description of a phosphoryl transfer reaction is given in terms of dissociative and associative (Figure 1.10, A and B, respectively). A dissociative transition state is dominated by bond cleavage (Maegley et al., 1996); in the extreme case the bond to the outgoing group is fully or nearly broken and the bond to the incoming nucleophile is absent or barely formed. In contrast, in an associative transition state there is a large amount of bond formation to the incoming nucleophile but only a small amount of bond cleavage of the 12 Introduction outgoing leaving group. An enzymatic reaction is more likely to be somewhere in the middle of these two mechanistic extremes and the comparison should always be relative. This is the reason why study of model compounds does not necessarily describe the real situation. A true metaphosphate intermediate probably does not exist in aqueous solution or in active sites (Du et al., 2004). Typically a phosphoryl transfer reaction is concerted with a transition state in between monophosphate and phosphorane extremes (Figure 1.10 C).

Figure 1.10. Mechanism for hydrolysis of phosphoesters and anhydrides. (A) Dissociative and (B) Associative, the mechanistic extremes; (C) the concerted pathway, (D) substrate assisted mechanism (Du et al., 2004).

High resolution crystal structures have attempted to clarify the situation in certain cases (Lahiri et al., 2003), but also have added controversy (Blackburn et al., 2003). Transition states with largely metaphosphate or phosphorane like characteristics have also been described as “loose” or “tight”, respectively (Du et al., 2004).

In small GNBPs, the intrinsic rate of hydrolysis is slow, for Ras, in the range of 0.02 min−1 at 30 °C, and is strictly dependent on the presence of divalent cations (Wittinghofer and Waldmann, 2000). A water molecule close to the γ-phosphate and in hydrogen bond distance to glutamine 61 is considered to be the attacking nucleophile and mutation of this glutamine residue to most other amino acids cause a dramatic reduction of intrinsic GTP hydrolysis rate (Krengel et al., 1990). 31P-NMR studies with various Ras mutants and determination of intrinsic hydrolysis reaction rates at different pH values resulted in a substrate assisted catalysis theory (Figure 1.10 D), where the γ-phosphate itself activates the attacking water molecule by abstracting a proton (Schweins et al., 1995; Schweins et al., 1997). Both the less charged phosphate and the more reactive nucleophile (the hydroxide ion) were suggested to promote catalysis. The role of glutamine 61 is thought to be positioning of the

13 Introduction water molecule in the vicinity of the γ-phosphate. Apart from the Mg2+ ion, the invariable P- loop lysine of both G domain and ATP motor proteins (Lys-16 in Ras), which points toward the β-phosphate, has also been suggested to have a role in the hydrolysis reaction (Allin et al., 2001). Direct experimental evidences for charge shift (from γ to β phosphate) are getting accumulated from spectroscopic methods (Allin et al., 2001) but they are, so far confined to limited systems.

Studies of phosphoryl transfer reaction in different systems reveal some unique conserved as well as some divergent features. The most common feature found in several structures like TMPK ATPase, is the presence of a nucleophilic water molecule (Ostermann et al., 2000). Multiple bivalent cations instead of one are also usual. Such features are seen in RNA polymerase (Cramer et al., 2001), Adenylyl Cyclase (Tesmer et al., 1997) and protein phosphatases (Goldberg et al., 1995; Das et al., 1996). Instead of water, cysteine is the nucleophile in PTP1B (protein tyrosine phosphatase 1B), which attacks the phosphate moiety (Barford et al., 1998). Interestingly in PTP1B, a glutamine residue (Gln 262) serves the purpose of positioning the nucleophilic water similar to the Gln 61 of Ras. Both PTP and GTPases (for GAP catalyzed reaction) use an essential arginine residue that coordinates and stabilizes the pentavalent phosphorous intermediate (Coleman et al., 1994; Strater et al., 1995; Scheffzek et al., 1997). Finally, the classical P-loop lysine, which is considered to be a general feature in GNBPs and several motor proteins, is also not absolutely conserved. In the GMS subfamily of large GTPases (LRG47), the lysine is replaced by a methionine (Boehm et al., 1998). However, these members are not yet biochemically characterized with respect to GTP hydrolysis. Therefore the possibility remains that they do not hydrolyze GTP.

1.2 GTPase Activating Proteins

The intrinsic GTPase reaction of most GNBPs is too slow to be efficient in signal transduction processes. GAPs accelerate the otherwise slow hydrolysis reaction by several orders of magnitude and thus terminate the signal. The slow intrinsic reaction of EF-Tu for example is dramatically stimulated by mRNA charged 70S ribosome (Ogle et al., 2003). GTP hydrolysis in Gα proteins can be simulated by a family of proteins called regulators of G protein signaling (RGS) (Tesmer et al., 1997). For the small GNBPs like Ras, Rho, Ran, Rab and Arf, specific GAPs catalyze GTP hydrolysis, which are all different at the sequence level and employ certain unifying and some divergent features to accelerate GTP hydrolysis of the cognate GNBP. They are briefly discussed in the following section.

14 Introduction

1.2.1 RasGAP and Gα: the arginine finger in trans or cis

RasGAPs were discovered when it was found that Ras•GTP microinjected in cells was rapidly converted to Ras•GDP (Trahey and McCormick, 1987). The protein responsible for this activity was identified and called p120GAP. A fragment of 334 residues (GAP-334) was shown to be sufficient for catalytic activity. GAP-334 itself is a helical elongated protein with a central domain of 218 residues, which is conserved among all RasGAPs (Scheffzek et al., 1996). Later on, the complex structure (Scheffzek et al., 1997) of Ras and GAP-334 along with AlF3, a classical analogue for the gamma phosphate of ATP or GTP (Chabre, 1990; Mittal et al., 1996), revealed that the GAP interacts predominantly with the switch regions and the P-loop of Ras (Figure 1.11A). A conserved arginine residue from an exposed loop of RasGAP complements the catalytic site of Ras and the guanidium group of this arginine contacts the β-phosphate and AlF3. In addition, the main chain carbonyl oxygen of R789

Figure 1.11. The structures of (A) Ras-GAP and (B) Gαi1- RGS (Sprang, 1997). The GAPs are shown in red and green, GNBPs in yellow, switch regions are in cyan, the 2+ nucleotide, Mg and AlFx are represented in ball-and-stick. (C). Superposition of the

active site of Gαi1. GDP•AlFx (in green) onto the active site of Ras. GDP•AlF3 (in orange). The finger loop and the Arg of GAP-334 are supplied in trans and colored in

red whereas in Gαi1 it is cis and represented in cyan (Scheffzek et al., 1997). makes a hydrogen bond to the side chain amide group of the catalytic glutamine (Q61) of Ras and stabilizes its position. Due to its positioning on a flexible loop and direction towards the active site, it has been termed as ‘Arginine finger’. Thus, the principle of GTPase simulation by GAP were suggested to be (i) stabilization of the switch regions (ii) stabilization of Q61 leading to correct positioning of the nucleophilic water molecule and (iii) supplying a positive charge (arginine finger) to compensate the developing negative charge at the transition state when it is associative. This structure also explains why the G12V mutation of Ras is oncogenic. Glycine 12 is within van-der-Waals distance of the catalytic arginine and any mutation interferes sterically with the correct transition state formation, though such mutants still bind to the GAP (Polakis and McCormick, 1993). Glutamine 61 points towards the phosphate chain (Figure 1.11 C) of the nucleotide and is stabilized in its orientation by a hydrogen bond with the main chain carbonyl group of the invariant Arg789. This stabilizes the

15 Introduction transition state and mutation to any other residue-even alanine-would disturb the transition state (Scheffzek et al., 1997) of the reaction.

In the aforementioned cases, the active site is complemented by a catalytic arginine from a cognate protein, often termed ‘in trans’. Heterotrimeric G proteins are examples where the same is achieved ‘in cis’, i.e. from the same molecule (Figure 1.11 B, C). The crystal structure of Gαtα in the GTPγS bound state showed that these proteins have a Ras-like G domain fold with a helical insertion (Figure 1.11B) shortly before the switch I region (Noel et al., 1993). The structure of a Gα protein bound to GDP and aluminium fluoride revealed that the mechanism of GTP hydrolysis involves stabilization of the attacking water molecule by a catalytic glutamine (homologous to glutamine Q61 in Ras) and by a threonine from the helical domain (Coleman et al., 1994). Strikingly, the GTPase reaction was highly dependent on an arginine residue supplied by the helical domain, which makes a hydrogen bond to aluminium fluoride, mimicking the transition state of GTP hydrolysis. Nevertheless, this stabilization is dependent on helper proteins called regulators of G protein signaling (RGS), which accelerates the intrinsic reaction of Gα proteins but does not share sequence homology to any other GAPs (Druey et al., 1996). RGS has modest affinity to Gαi•GTP but binds tightly to the transition state mimic Gαi•GDP•AlFx. The structure proves that RGS does not contribute any residue to the catalytic machinery, but stabilizes the active conformation, most notably the catalytic glutamine (Tesmer et al., 1997).

1.2.2 RhoGAP and bacterial toxins

Several solved structures of Rho specific GAPs (Barrett et al., 1997; Rittinger et al., 1997a; Rittinger et al., 1997b) show that these proteins are purely helical (Figure 1.12) and as in the case of RasGAPs, they also employ an arginine finger and stabilize the catalytic glutamine residue of the GNBP (Rittinger et al., 1997b).

− Figure 1.12. Structure of RhoGAP in complex with Rho•GDP•AlF 4 (PDB code: 1AM4) and ExoS − in complex with Rac•GDP•AlF 4 (PDB code: 1HE1). The GAPs are colored in red and the GNBP counterpart in orange and blue. (The individual figures were made by Oliver Daumke.)

16 Introduction

Although no sequence similarity is observed, based on the structures it was proposed that RasGAPs and RhoGAPs have a common evolutionary ancestry (Rittinger et al., 1998).

Interestingly, bacteria have developed strategies to modulate Rho GTPase activity of host cells through toxins as Salmonella SptP (Fu and Galan, 1999), Pseudomonas ExoS, Yersinia YopE (Goehring et al., 1999) and ExoT (Krall et al., 2000). The conserved sequence patterns (Litvak and Selinger, 2003) of these small proteins (~140 residues) fold into exposed loop structures (Wurtele et al., 2001; Stebbins and Galan, 2001; Evdokimov et al., 2002), termed as bulges that carry the functionally important residues of these helical proteins (Figure 1.12B). From the structures it was suggested that bacterial GAPs and RhoGAPs have most likely evolved independently.

1.2.3 GAPs without an arginine finger

The most important members of this category are ArfGAP, RanGAP, the signal recognition particle and its receptor and RapGAP, which are discussed in the following sections.

Arf (ADP ribosylation factor) GNBPs are the major regulator of vesicle biogenesis in intracellular trafficking (reviewed in (Chavrier and Goud, 1999; Spang, 2002)). This subfamily contains Arf, Arl (Arf like), Arp (Arf-related protein) and Sar1 (Secretion-associated and Ras- related). They depart from other GTP binding proteins by a unique structural device called ‘interswitch toggle’ (Pasqualato et al., 2002) that implements communication between the N terminus and the nucleotide . Arf in the GDP bound form translocates from the cytosol to the membrane and is activated by Arf specific exchange factors (Renault et al., 2003). This process requires both the GEF Sec7 domain, which promotes the interswitch toggle (Goldberg, 1998), and membranes, which unfasten the N-terminal hasp (Antonny et al., 1997). In the active form it associates with the membrane and recruits a seven-subunit complex called coatomer, which is followed by budding of vesicles containing Arf and the coatomer complex. When the membrane curvature of the vesicle increases, the activity of ArfGAP, also present in this complex, increases dramatically, leading to GTP hydrolysis and dissociation of Arf•GDP (Bigay et al., 2003). ArfGAP contains four ankyrin domains (Mandiyan et al., 1999) and the 140 residue catalytic domain contains a zinc finger motif (Cukierman et al., 1995). The structure (Goldberg, 1999) of the GAP domain complexed with Arf•GDP (Figure 1.13 A) revealed the zinc finger of ArfGAP consisting of four strands and one helix, embedded in an irregular array of six α-helices and one β-strand, lacking similarity to any other GAP family. Interestingly, the GAP binds at a distal site (switch II) of Arf, far away from the nucleotide binding site. The arginine residue whose mutation to alanine reduces GAP activity (Mandiyan et al., 1999) is about 15 Å distance from the nucleotide binding site, which led to the proposal that the GTPase reaction is not stimulated by an ‘arginine finger’, rather by proper orientation of the catalytic glutamine through rearrangement

17 Introduction

Figure 1.13. Structural overview of (A) Arf•ArfGAP (PDB coordinates provided by J. Goldberg), where ArfGAP (in red and green) binds Arf (orange and blue) far away from the

nucleotide binding site; (B) Ran•RanGAP•RanBP1•GDP•AlF3 (PDB code: 1K5D) where RanGAP is in red and green, Ran is in orange and blue and RanBP1 is in purple. (C) SRP-SR heterodimer in GppNHp bound form (PDB code: 1OKK). SRP is shown in green and red, SR in orange and blue. (The individual figures were made by Oliver Daumke.) of switch II. It has also been shown that addition of the coatomer enhances the GTPase reaction by 1000 fold, leading to the proposal that the coatomer itself is involved in GAP catalyzed reaction and supplies an arginine finger (Goldberg, 1999).

RanGAP stimulates the GTPase reaction of the small GNBP Ran, which is involved in nuclear transport. A third protein RanBP (Ran binding protein), binds to Ran and increases its affinity for RanGAP (Seewald et al., 2003). RanGAP consists of a 330-350 residue leucine rich repeat (LRR) followed by an acidic region of approximately 40 residues (Hillig et al., 1999). A single LRR consists of a β-α hairpin consisting of a β-strand, a loop and a α-helix roughly parallel to the β-strand and this structural element is found in several proteins of different functions, e.g. ribonuclease A inhibitor or Drosophila Toll receptor (Kobe and Deisenhofer, 1995). The complex of RanGAP, RanBP1 and Ran with a GTP analogue or GDP and aluminium fluoride (Figure 1. 13 B) (Seewald et al., 2002) showed that the only arginine residue near the vicinity is flipped away from the γ-phosphate and the same was previously shown to be unimportant for catalysis (Hillig et al., 1999). Therefore, RanGAP also catalyzes without an ‘arginine finger’.

The signal recognition particle (SRP) and its receptor (SR) target newly synthesized proteins destined for secretion or membrane integration to the endoplasmic reticulum (Keenan et al., 2001). Both SRP and SR are conserved across all kingdoms of life. In prokaryotes, SRP consists of a single 48 kD GTPase called Ffh and a 110 nucleotide 4,5S RNA, whereas SR consists of the GTPase called FtsY. Structurally, they contain an IBD (insertion box domain) insertion within the G-domain and special sequence motifs ALLEADV, DARGG and DGQ. With relatively low nucleotide affinity and rapid exchange, when both proteins are loaded with

18 Introduction

GTP, they can bind to each other leading to a mutual stimulation of their GTPase activities, GTP hydrolysis and concomitant dissociation of the complex.

Very recently, the mechanism of this mutual stimulation was elucidated by solving the structure of the complex between the two GTPases in the presence of a non-hydrolysable GTP analogue (Focia et al., 2004; Egea et al., 2004). The two GTPases form a quasi- symmetric heterodimer, and the two nucleotides are aligned in a nearly symmetrical composite active site (Figure 1.13 C), sequestered from the solvent, which prevents GTP dissociation without hydrolysis. In each chain, a catalytic aspartate was identified which is thought to activate the attacking nucleophilic water molecule in cis. Additionally, an arginine and a glutamine residue are provided from both molecules into the active site and interact in cis and possibly in trans with the β- and γ-phosphate groups. Strikingly, the 3’ OH group of each GTP molecule contacts the γ-phosphate group of the opposing GTP molecule and reduces the negative charge on the γ-phosphate. This hydroxyl group was shown to be essential for association, mutual activation and catalysis (Egea et al., 2004). However, the significance of the third GTPase SR-β found in eukaryotes, is not clear.

1.2.4 GTPase activating proteins of the Rap1GAP family

RapGAPs are tumorsuppressors. Rap1GAP, the founding member of this group, encodes a polypeptide of 663 amino acids (molecular weight of 73 kD), which is abundantly expressed in brain, foetal tissue, undifferentiated cells and certain tumor cells. By deletion analysis it was shown (Rubinfeld et al., 1992) that fragments of residues 75-416 contain full catalytic activity towards Rap1. Homologous sequences have been identified from yeast, drosophila, mosquito and C. elegans to mammals and for the later, some of the important entries are SPA1 (Hattori et al., 1995), E6TP1 (Gao et al., 1999) and Tsc2 (Tuberous Sclerosis Consortium, 1993). RapGAPs are thought to be important for tumor suppression because degradation of the E6TP1 protein via the transforming papilloma virus protein E6 correlates with the development of cervical cancer (Gao et al., 2001). Deletion of the Spa1 gene in mouse on the other hand creates a spectrum of myeloid disorders that resemble chronic myeloid leukaemia (Ishida et al., 2003). The role of Tsc2 in tuberous sclerosis complex has already been described in the previous section. Figure 1.14 is an overview of the mammalian RapGAP related proteins. This is based on the presence of domains in addition to the RapGAP domain. The members of the first category consist of only RapGAP domain (KIAA1039, Rap1GAP2) or, in addition, a GoLoco domain (Rap1GAP). Two unique, biochemically uncharacterized categories are CAB66508, which contains a Citron-Kinase like domain (CHN) and KIAA1389 / Tulip where only a coiled coil domain (CC) is predicted. The third major group contains E6TP1, SPA1 and SPA1L2, all of which have RapGAP, PDZ, and coiled-coil architecture. The last example is Tuberin, the key player of the Tuberous Sclerosis

19 Introduction complex that contains part of the RapGAP domain. The members that are biochemically characterized are described in the following sections.

Figure 1.14. Domain architecture of mammalian Rap-GAPs. Abbreviations used: R1G: Rap1GAP, CC: coiled coil, TAD: Transcription activation domain, CAM: Calmoduline binding domain, CHN: Citron kinase like domain. The GoLoco domain is present in Rap1GAP but not in RapGAP2.

1.2.4.1 Rap1GAP

Rap1GAP was purified from brain extract (Polakis et al., 1991), cloned (Rubinfeld et al., 1991) and shown to stimulate specifically the GTP hydrolysis activity of Rap1, and to a smaller extent Rap2 (Janoueix-Lerosey et al., 1992). Rap1GAP is phosphorylated in vivo at four distinct sites (Polakis et al., 1992; Rubinfeld et al., 1992) and is reported to be regulated by protein degradation and relocalization from cytosol to membrane. A splice isoform of Rap1GAP, RapGAPII is expressed in brain, heart, liver and kidney and binds specifically to the α-subunit of Gi family of heterotrimeric G proteins through its N-terminal GoLoco motif

(Mochizuki et al., 1999; Willard et al., 2004). Stimulation of the Gi-coupled m2-muscarinic receptor translocates RapGAPII from the cytosol to the membrane and decreases the amount of GTP bound Rap1. This decrease of Rap1•GTP leads to an increase of ERK /

MAPK kinase activity. Rap1GAP has also been shown to bind the GTP bound form of Gαz (Meng et al., 1999) that leads to a decrease of cellular Rap1•GTP (Meng and Casey, 2002) and to the GDP bound form of Gαo that leads to an increase of cellular Rap1•GTP (Jordan et al., 1999). Rap1GAP is the only member of this family, which is well characterized in terms of in vitro biochemical data (Brinkmann et al., 2002; Kraemer et al., 2002) and more recently, structural analysis (Figure 1.15) (Daumke et al., 2004).

20 Introduction

Figure 1.15. Structure of human Rap1GAP (Daumke et al., 2004). In the dimer, molecule A is colored according to secondary structure and molecule B according to domains (PDB code: 1SRQ).

This unique protein does not share any sequence homology to any other GAP and most of the conserved lysines (except K194 and K285) and arginines (unlike RasGAPs) can be mutated without exerting dramatic effects (Brinkmann et al., 2002). The protein has distinct two-domain structures, which are called catalytic domain and dimerization domain. The functional aspects related to the hydrolysis reaction are discussed in the results section.

1.2.4.2 Signal-induced proliferation-associated protein 1 (Spa-1) and E6TP1

SPA-1 encodes a 130 kD protein, which was found to be expressed in small amounts in quiescent murine lymphoid cell lines, but expression was induced upon interleukin-2 stimulation (Hattori et al., 1995). In human, expression of SPA-1 and Rap1GAP are complementary, i.e. SPA-1 is expressed in lymphoid tissues like spleen and thymus but not in the tissues in which Rap1GAP is expressed, such as brain, kidney or pancreas (Kurachi et al., 1997). SPA-1 shows GAP activity towards both Rap1 and Rap2 and has been shown to be a tumorsuppressor gene, since Spa-1 knockout mice develop a spectrum of myeloid disorders that resemble human chronic myelogenous leukaemia (Ishida et al., 2003).

E6TP1 (E6 targeted protein 1) was identified as a 200 kD protein, which is targeted to proteasome-mediated degradation by the High-risk human papilloma virus (HPV) E6 through E6AP ubiquitin (Gao et al., 1999; Gao et al., 2002a). GAP activity towards Rap1 and Rap2 has also been demonstrated (Singh et al., 2003). SpaR is a rat homologue of E6TP1, which is reported to have an Act (actin regulatory domain) and a GKB (Guanylate kinase binding) domains (Pak et al., 2001; Pak and Sheng, 2003). SPAR was found to form a complex with the GKB domain of PSD-95 and NMDA receptors in neurons. This complex has a crucial role for establishing dendritic spines, which are postsynaptic protrusions where glutamate mediated neuronal transmission takes place (Meyer and Brose, 2003). In

21 Introduction heterologous cells, SPAR recognizes the actin cytoskeleton and recruits the scaffold protein PSD-95 to F-actin (Pak et al., 2001). Phosphorylation of SPAR1 by serum inducible kinase (SNK) leads to the degradation of SPAR, resulting in the loss of mature dendritic spines. Although the role of Rap1 regulation is unclear, SPAR is an important component of neuronal activity.

1.2.4.3 Tuberin and Hamartin

Tuberous sclerosis (Tsc) is an autosomal dominant inherited disorder affecting 1 in 6000 births, characterized by benign tumors called hamartomas with symptoms like mental retardation, epilepsy, renal failure and rarely malignancies (reviewed in (Manning and Cantley, 2003; Krymskaya, 2003)). Mutations in one of the two tumor suppressor genes Tsc1 (chromosome 9: encoding 130 kD protein Hamartin) or Tsc2 (chromosome 16: encoding 200 kD protein Tuberin) (Tuberous Sclerosis Consortium, 1993) leads to the same phenotype. Tuberin and Hamartin form a heterodimer (van Slegtenhorst et al., 1998; Nellist et al., 1999) through interaction between specific domains (reviewed in (Krymskaya, 2003)) as shown in Figure 1.16. The heterodimer is localized to membrane by means of the membrane association domain of Tsc1 (Hamartin). Interestingly, Hamartin interaction domain of Tsc2

Figure 1.16. Schematic representation of the structural-functional domains of hamartin (A) and tuberin (B). ERM: ezrin-radicin-moezin; NF-L: neurofilament-L; TAD: transcription- activating domains; CAM: calmodulin (CaM)-binding domains. (Krymskaya 2003).

22 Introduction is located at the extreme N-terminus, while the GAP domain homologous to Rap1GAP is located near the C-terminus, being separated by approximately 1000 residues, which contain several phosphorylation sites that are known to associate with adaptor proteins like 14-3-3. Multiple Akt/PKB-dependent phosphorylation sites are indicated with arrows; phosphorylation sites modulating 14-3-3 association are also indicated with arrows (reviewed in (Krymskaya, 2003)). It is known that the GAP domain is necessary but not sufficient for GAP activity.

1.2.5 Summary of biochemistry of Rap1GAP and Tsc1/2

Activation of the GTPase reaction of Rap1 by Rap1GAP appears to have several unique aspects. Firstly, a catalytic glutamine (glutamine Q61 in Ras) is crucial for the intrinsic and GAP stimulated GTP hydrolysis of Gα, Ras, Rho, Rab and Ran GNBPs. In contrast, Rap1 does not have a catalytic glutamine but a threonine at the equivalent position. As a result, the intrinsic GTP hydrolysis rate of Rap1 is 10-fold slower than for Ras but can be reconstituted by exchanging threonine T61 for a glutamine (Frech et al., 1990). Furthermore, threonine T61 is not required for the intrinsic and GAP stimulated GTP hydrolysis in Rap1 (Maruta et al., 1991). In addition to this, Glycine G12 mutations in Ras and other GNBPs render the protein inactive to GAPs since any side-chain at position 12 would interfere with the positioning of the arginine finger and the catalytic glutamine. However, Rap1GAP can still down regulate the G12V mutant of Rap1, albeit with a 8-fold reduced rate (Brinkmann et al., 2002). Again, most of the conserved arginine and lysine residues of RapGAP could be mutated without exerting a dramatic effect in catalysis (Brinkmann et al., 2002). This indicates that Rap1GAP stimulates GTP hydrolysis of Rap in a completely different way than all other GAPs. A fluorescence assay was developed to monitor a single turnover reaction cycle of the Rap1GAP catalyzed reaction in Rap1 (Kraemer et al., 2002). It was proposed that the rate limiting step of the reaction is not GTP hydrolysis but the release of inorganic phosphate, as already proposed for RasGAP-Ras system (Allin et al., 2001). However the difference between the hydrolytic cleavage step and release of GDP and Pi was small and evidences were indirect (Kraemer et al., 2002). Kraemer et al. Also proposed an additional kinetic step in comparison to RasGAP-Ras system, without explicit characterization. When this work was undertaken, at the absence of a structure of either Rap1GAP or a complex with cognate GTPase Rap1, there was absolutely no hint for a catalytic residue of Rap1GAP.

In the RheB-Tsc1/Tsc2 system, a wealth of genetic and cell biological data indicates that the Tsc1-Tsc2 heterodimer functions as a GAP for the small GNBP RheB (Tee et al., 2003) and it has been argued that RheB might function without a guanine nucleotide exchange factor (reviewed in (Manning and Cantley, 2003)). However, there is little in vitro data available for this system.

23 Introduction

1.3 Objectives of this work

The available biochemical data evidently indicate that the Rap1GAP stimulated GTPase reaction of Rap1 is fundamentally different from all other GAP stimulated reactions hitherto described. Therefore, this new mechanism was explored by newly emerging time resolved Fourier Transform Infrared spectroscopy in combination with mass spectrometry, to get important information about the mechanism of phosphoryl transfer in this system and some hint of the putative catalytic residue.

While this work was in progress, the structure of Rap1GAP was solved and the catalytic residue was identified which matched with the spectroscopic evidences. The next obvious task was to characterize the active site and the reaction in terms of mutational analysis through rapid kinetics and transition state mimicking.

Finally, it has been attempted to model the GAP activity of Tsc1-Tsc2 towards RheB under in vitro conditions. The main question has been: Why does it require a heterodimeric complex to achieve the GAP function?

24 Materials and methods 2 Materials and Methods

2.1 Materials

2.1.1 Chemicals

Chemicals from the following companies were used: Amersham-Pharmacia (Freiburg), Fluka (Neu-Ulm), Merck (Darmstadt), Qiagen (Hilden), Roche (Mannheim), Roth (Karlsruhe), Serva (Heidelberg) and Sigma-Aldrich (Deisenhofen), GERBU (Gaiberg), GIBCO (Invirtogen) and 18 Isotech (H2 O). Contributions from individuals are mentioned specifically.

2.1.2 18O labeled caged-GTPs

The 18O labeled caged GTPs used in this thesis were synthesized by Drs. Valentine Cepus, Christoph Allin, Yan Suveyzdis and Marco Blessenohl at the Lehrstuhl für Biophysik, Ruhr- Universität Bochum and had been used previously in similar studies on Ras proteins.

2.1.3 Enzymes

DNAase-I Roche (Mannheim)

Pfu DNA polymerase New England Biolabs (Schwalbach)

Restriction enzymes New England Biolabs (Schwalbach)

T4 DNA ligase New England Biolabs (Schwalbach)

Thrombin Serva (Heidelberg)

TEV-protease Gift from Dennis Fiegen and Carolin Koerner

Alkaline phosphatase Roche (Mannheim)

2.1.4 Kits

QIAprep Spin Miniprep Kit Qiagen (Hilden)

QIAquick Gel Extraction Kit Qiagen (Hilden)

BigDye Terminator Sequencing Kit Applied Biosystems (Langen)

λ-DNA standard Invitrogen (Karlsruhe)

Wide Range, SDS7 protein marker Sigma (Deisenhofen)

ECL plus Western Blotting Detection system Amersham Biosciences

25 Materials and methods

2.1.5 Columns

C-18 Ultrasphere Beckmann (Unterschleißheim)

DEAE-Sepharose Fast Flow Amersham-Pharmacia (Freiburg)

Glutathion Sepharose 4B Amersham-Pharmacia (Freiburg)

Superdex 75 und 200 Amersham-Pharmacia (Freiburg)

2.1.6 Microorganisms

E. coli CK600K supE, hsdM+, hsdR-, kanR (Hoffmann-Berling, Heidelberg)

E. coli TG1 K12, supE, hsd∆5, thi, ∆(lac-proAB), F’[traD36, proAB+, lacIq, lacZ∆M15] (Promega)

E. coli BL21 (DE3) B, F-, hsdSB (rB-, mB-), gal, dcm, ompT, λ(DE3) (Novagen)

E.coli BL21 (DE3) F– ompT hsdSB(rB– mB–) gal dcm (DE3) pRARE2 (CmR)

Rosetta (Novagen) pRARE containing the tRNA genes argU, argW, ileX, glyT,

leuW, proL, metT, thrT, tyrU, and thru

2.1.7 Mammalian cell lines

Cos7, HEK-293 (Human embryonic kidney cells).

2.1.8 Media, special reagents and antibiotics

Luria-Bertani (LB) 10 g/l Bactotryptone, 10 g/l NaCl, 5 mM NaOH, 5 g/l yeast extract

Terrific Broth (TB) 12 g/l Bactotryptone, 24 g/l Bacto-yeast-extract, 4 g/l glycerol, 17 mM

KH2PO4, 72 mM K2HPO4

Standard I 25 g/l standard-I powder (Merck, Darmstadt)

DMEM Dulbecco’s modified Eagle medium (GIBCO, Invitrogen)

OPTI-MEM 1 Reduced serum medium (Invitrogen)

Lipofectamine 2000 Cationic lipid based transfection reagent (Invitrogen)

FCS Fatal Calf Serum

Trypsine-EDTA 0,05 % (w/v) Trypsin (1:250), 0,02 % (w/v) EDTA in Puck’s Saltsolution A

26 Materials and methods

Antibiotics from GERBU (Gaiberg) were used in the concentration 100 mg/l (Ampicillin) and 50 mg/l (Kanamycin).

2.1.9 Antibodies

α-Flag M2 (ab), monoclonal (Sigma Aldrich Co. USA)

α-Myc 9E10, sc40 (Santa Cruz Biotechnology, USA)

3 antibody system 2nd antibody: biotinated mouse Ig, 3rdantibody:Streptavidine-coupled Horse radish peroxidase

Labeled 2nd antibody FITC or TRITC labeled Ig mouse (for immunohistochemistry)

2.1.10 Special Buffers

PBS 140 mM NaCl; 2,7 mM KCl; 10,1 mM Na2HPO4; 1,8 mM KH2PO4

Buffer C 64 mM Tris, 10 mM MgCl2, pH 7,6 (HCl) + 5 mM DTE

Buffer D 1 x Puffer C, 200 µM GDP, 0,4 M NaCl + 5 mM DTE

GTPase assay buffer 50 mM Hepes (pH 7.6), 100 mM NaCl, 5 mM DTE, 5 mM MgCl2

Cell Lyses Buffer 50 mM HEPES (pH 7.6), 1 mM β-Mercaptoethanol, 100 mM NaCl, 5

mM MgCl2, 1 mM PMSF in Isopropanol, 5% Glycerin, 1% NP40, 1 mM protease inhibitor cocktain without EDTA

TAE 40 mM Tris, pH 8.0; 40 mM AcOH; 1 mM EDTA

Z-buffer for β-Gal 60 mM Na2HPO4, 40 mM NaH2PO4, 10 mM KCl, 1 mM MgSO4, 2 mM DTE, 1 mg/ml ONPG (o-Nitrophenyl-β-D-galactoside), pH 7,0

10X SDS buffer 0,25 M Tris, 1,92 M Glycin, 1 % SDS

Transfer buffer (2L) 6 g Tris, 28.8 g Glycin, 2g SDS, 400 ml Methanol

2.2 Molecular biology methods

2.2.1 Agarose gels

Agarose gels were prepared and run according to standard procedure (Sambrook, 1989).

27 Materials and methods

2.2.2 Isolation of plasmid DNA

DNA isolation was done using QIAprep Spin Miniprep Kit from Qiagen (Hilden), according to manufacturer’s protocol.

2.2.3 Polymerase chain reaction (PCR)

Amplification of DNA fragments was carried out using Pfu polymerase (New England Biolabs, Schwalbach) according to standard procedures (Sambrook, 1989). Fragments were digested according to enzyme specification (NEB Handbook) and purified by 1% agarose gel and QIAquick Gel Extraction Kit (Hilden) according to manufacturer’s protocol.

2.2.4 Ligation

Vector and insert DNA were quantified in agarose gels using digested λ-marker (New England Biolabs, Schwalbach) as a reference. 10 ng of vector was ligated with a four fold molar excess of insert overnight at 8 °C using T4 ligase (Roche, Mannheim) according to the manufacturer’s protocol.

2.2.5 Competent cells

Competent cells were prepared according to (Chung et al., 1989). 200 ml LB medium was inoculated with 2 ml preculture and grown at 37 °C until an OD600 of 0,4. Bacteria were incubated for 20 min on ice, pelleted for 5 min at 1200 x g (4 °C), resuspended in 20 ml ice- cold sterile TSS buffer (85% LB medium without NaOH, 10% PEG 8000, 5% DMSO, 50 mM

MgCl2, pH 6,5), flash frozen and stored at -80 °C.

2.2.6 Transformation

The heat shock method was used according to the standard protocol (Sambrook, 1989). This includes 30 minute ice incubation of 1 µg DNA and 100-200 µl of competent cells followed by a two minute heat shock at 42°C and another ice shock for one minute. 1 ml of LB medium without antibiotic is added to this and incubated at 37°C for 40 minutes with shaking, thereafter, centrifuged carefully, discarding 900 µl supernatant, the bacteria were resuspended and plated. Ligated DNA was transformed in E. coli TG-1, amplified and isolated. The isolated DNA was transformed in BL21 (DE3) expression bacteria.

28 Materials and methods

2.2.7 Bacteria storage

50% glycerol was added to an overnight culture and bacteria stocks were stored at -80 °C.

2.2.8 Site specific mutagenesis

Site specific mutagenesis was carried out using the QuickChange kit (Stratagene, Amsterdam) according to the manufacturer’s protocol.

2.2.9 DNA sequencing

DNA sequencing was done according to (Sanger et al., 1992) using the Big Dye terminator kit (Applied Biosystems, Langen). A sequencing reaction contained 10 µl DNA Qiaprep solution, 4 µl terminator mix, 3 pmol sequencing primer in a volume of 20 µl. The sequencing PCR and DNA precipitation was carried out according to the manufacturer’s protocol. Analysis of the sequencing products was done in house on a ABI PRISM 3700 DNA Analyzer (Applied Biosystems, Langen).

2.2.10 Constructs

Construct Remark ptac hsRap1B C’1-167 Provided by D. Kühlmann pGEX 4T1 hsRap1GAP75-415 Provided O. Daumke pGEX 4T1 hsRheB1-170 Provided by D. Kühlmann pGEX4T3 hsTsc21538-1729 Provided by D. Kühlmann pGEX3T1-TEV-hsTsc21520-1763 In this study pGEX3T1-TEV-ratTsc1645-872 In this study pcDNA 3.1 Myc-Kozak In this study ratTsc12-600 pcDNA 3.1 Myc-Kozak In this study ratTsc12-672 pcDNA 3.1 Myc-Kozak In this study ratTsc12-870 pcDNA 3.1 Myc-Kozak In this study ratTsc12-1164 Flag tagged Tsc21-1784 Obtained from Dr. Kun-Liang, Michigan University. pEQ176 (β-Galactosidase) Provided by B. Voss

Note: Tsc1 and Tsc2 cDNAs were obtained from Dr. Kun-Liang Guan of Michigan University, USA and Dr. George Thomas of Friedrich Miescher Institute, Switzerland.

29 Materials and methods

2.2.11 Point mutants

Construct Point Mutants A86C (In this study, sence and antisence primers ptac Rap1B C’1-166 T61Q provided by D. Kühlmann) ptac Rap1B C’1-166 (A86C) Purified protein from A. Kraemer G12V, T61L, T61A, T61L, T61E provided by D. ptac Rap1B C’1-166 (A86C) Kühlmann

2.3 Biochemical methods

2.3.1 SDS-PAGE

Separation of proteins of different molecular weight was performed according to (Laemmli, 1970) using denaturing, discontinuous SDS-polyacrylamide gel electrophoresis (SDS- PAGE).

2.3.2 Determination of protein concentration

Protein concentration was determined according to (Bradford, 1976) using the Biorad protein assay (Biorad). The solution was calibrated using bovine serum albumin.

2.3.3 Matrix assisted laser desorption ionisation (MALDI)

The molecular weight of newly isolated proteins were determined by matrix-assisted laser desorption ionisation (MALDI). A concentrated protein solution was diluted with water and 1:1 mixed with matrix (for proteins > 10 kD: saturated sinapinic acid in acetonitrile, 0,2% TFA; for peptides < 10 kD: saturated a-cyano-4-hydroxy cinnamic acid in acetonitrile, 0,2% TFA water). Data were acquired on a Voyager DE Pro (Applied Biosystems).

2.3.4 Test expression of proteins

To test expression and solubility of proteins expressed from pGEX vectors, typically, vectors containing the desired insert were transformed in expression bacteria (BL21-DE3 and BL21-

DE3 Rosetta). A 50 ml bacteria culture in TB medium was induced at on OD600 of ~0,4-0,6 with 500 µM IPTG and grown at 37 °C overnight. Two other cultures were induced with 50 µM and 200 µM IPTG and grown at 18 °C overnight.

To test expression, 1 ml of the overnight culture was pelleted, resuspended in water and analyzed using SDS-PAGE. To test the solubility, 1 ml bacteria culture was pelleted, resuspended in 100 µl BugBuster solution (Novagen) and lysed for 5 min on ice. The solution 30 Materials and methods was centrifuged at 25000 X g for 10 min at 4 °C. Fractions of 5 and 10 µl of the supernatant were analyzed by SDS-PAGE. Alternatively, 20 ml bacteria were pelleted and resuspended in 1 ml buffer of choice. Cells were broken by 1 min sonification on ice using a Sonifier 450 (Branson Ultrasonics, Danburry, USA) at 50% duty cycle. After 30 min centrifugation at 4 °C and 25000 x g, various amounts of the supernatant were analyzed via SDS-PAGE.

2.3.5 Protein overexpression and preparation of soluble bacteria extract

Proteins were overexpressed using the parameters described in Table 1.

Table 1 Expression constructs and expression conditions

E.coli Induction [IPTG] T / Expression Plasmid Medium Antibiotics strain at OD600 / µM °C time / h

ptac Rap1B C’ CK600K Standard I Amp, Kan 0,8 500 30 16

ptac Rap1B C’ CK600K Standard I Amp, Kan 0,6 400 25 16 mutants

pGEX4T1 Rap1GAP Bl21DE3 TB Amp 0,4 50 18 16

pGEX4T1 RheB BL21DE3 TB Amp 0,6 100 25 16

pGEX4T3 Bl21DE3 TB Amp 0,6 100 25 16 hsTsc21538-1729

pGEX3T1-TEV 1520-1763 Bl21DE3 TB Amp 0,4 100 18 16 hsTsc2

pGEX3T1-TEV Bl21DE3 TB Amp 0,3 50 18 16 hsTsc1645-872

In all the cases, bacteria were grown at 37 °C until the desired OD was reached. Upon IPTG induction, temperature was reduced to increase solubility of protein. After expression, bacteria were pelleted at 4000 x g, resuspended in buffer containing 100 µM phenylmethylsulfonylfluoride (PMSF) as protease inhibitor and frozen at -20 °C. Cells were thawn, 20 mg/l DNase I (stock 20 g/l in 1 M MgCl2) and 100 µM PMSF was added and the cells incubated for 30 min on ice. Cells were broken using a micro fluidizer (Microfluidics, Newton, USA). Insoluble material was removed by centrifuging at 100.000 x g for 45 min, and the supernatant used for further purification.

31 Materials and methods

2.3.6 Purification of Rap1GAP75-415

Rap1GAP75-415 was purified as GST-fusion according to (Brinkmann et al., 2002) with a subsequent gel filtration. Bacteria were broken in PBS, 5 mM MgCl2, 5 mM DTE, 1 mM ATP, 100 µM PMSF where ATP, potassium and magnesium ions help to remove bound GroEL. The cell supernatant was applied to a GST-column (25 ml, Amersham, Freiburg) equilibrated with PBS, 5 mM MgCl2, 5 mM DTE, 1 mM ATP and excessively washed with at least 500 ml of the same buffer. The buffer was exchanged to PBS, 5 mM DTE or 50 mM Hepes (pH 7,5),

100 NaCl, 2 mM CaCl2, 5 mM DTE. GST was cleaved by addition of 200 units thrombin (Serva) and overnight incubation. Rap1GAP was eluted with PBS, 5 mM DTE or 50 mM

Hepes (pH 7,5), 100 NaCl, 2 mM CaCl2, 5 mM DTE. The protein was concentrated (30- 50 mg/ml) using an Amicon concentrator (20 kD cutoff) and further purified on a Sephadex 75 or Sephadex 200 gel filtration using 20 mM Hepes (pH 7,5), 100 mM NaCl, 5 mM DTE as running buffer. Rap1GAP eluted in two peaks from gel filtration, the first one containing bound chaperone and eluting in the exclusion volume, the second eluting as protein with an apparent molecular weight of 100 kD and containing pure Rap1GAP. Protein from the second peak was pooled, concentrated and twice washed with 20 mM Hepes (pH 7,5), 5 mM DTE using Amicon concentrators (20 kD cutoff) to remove NaCl. The protein was finally concentrated to 30-50 mg/ml and flash frozen.

2.3.7 Purification of Rap1B1-167 and mutants

C-terminally truncated Rap1B (Rap1B C’) was purified according to (Tucker et al., 1986). Bacteria were broken in 32 mM Tris (pH 7,5), 100 µM PMSF, 2 mM EDTA (see 2.3.5).

The cell supernatant was applied on a Q-sepharose column equilibrated with 0,5 x buffer C. The column was excessively washed with 0,5 x buffer C and bound proteins were eluted using a salt gradient (0-300 mM NaCl) in 0,5 x buffer C. Fractions containing Rap1B C’ were identified using SDS-PAGE and pooled. Protein was precipitated by slowly adding solid ammonium sulfate (3M final concentration). Precipitated protein was pelleted by centrifugation for 60 min at 16000 x g and resuspended in buffer D. Rap1B C’ was further purified on a Sephadex 75 gel filtration column equilibrated with buffer D. Fractions containing Rap1B C’ were pooled and concentrated using an Amicon concentrator (10kD cutoff) and flash frozen.

2.3.8 Purification of RheB1-170

RheB1-170 was expressed as described in section 2.3.5 and purified in usual procedure as described for Rap1GAP (see 2.3.7) with small modifications. After extensive wash in GSH column, the GST tag was cleaved overnight by thrombine, the protein was eluted and concentrated by using an Amicon concentrator (10kD cutoff) and further purified on

32 Materials and methods

Sephadex 75 column using 20 mM Hepes (pH 7,5), 100 mM NaCl, 5 mM DTE as running buffer. The protein eluted as monomers and was concentrated to 30-60 mg/ml and flash frozen.

2.3.9 Rap1-Aedans preparation

Rap1-Aedans•GTP wild type and mutants (2.2.11) were prepared according to (Kraemer et al., 2002). In this method, a surface-exposed cysteine is modified by a haloacetamides (here 1,5-Iaedans), which is an environmentally sensitive fluorophore and acts as an indicator of protein-protein interaction.

The Rap1B C’A86C and Rap1B C’A86C mutants (2.2.11) were prepared as described in 2.3.7. DTE interferes with the labeling reaction. Thus, DTE was exchanged to a buffer containing 2 mM ascorbate using ultrafiltration with an Amicon concentrator (10 kD cutoff). The protein was diluted to 100 µM and incubated with a 3-fold excess of 1,5-Iaedans for 12 h at 4 °C. The reaction was stopped by buffer exchange with standard buffer containing DTE. Products were analyzed by mass spectrometry (appendix 6.2). The protein was flash frozen and stored at -80 °C and later on used for nucleotide exchange.

2.3.10 Nucleotide exchange

Nucleotide exchange was performed according to (Tucker et al., 1986). This method uses EDTA, which sequesters Mg2+ ions leading to an increase in the dissociation rate of the nucleotide. The nucleotide can then be exchanged by an excess of freshly added nucleotide. A master solution containing 15 mM EDTA, 150 mM ammonium sulfate and 10 mM nucleotide (stock 100 mM nucleotide in 1 M Hepes, pH 7,5) was made in 50 mM HEPES (pH 7.5), 100 mM NaCl and 5 mM DTE and Rap1B C’ was diluted in it to obtain a concentration of 150 µM. This is incubated for 60 minutes at room temperature or overnight at 4 °C. The exchange reaction was stopped by adding 30 mM MgCl2. Non-bound nucleotide was removed by washing the protein several times with an Amicon concentrator (10 kD cutoff) at 4 °C. To confirm successful nucleotide exchange, the nucleotide concentration was determined (see 2.3.11) and compared to the protein concentration (see 2.3.2). The protein was flash frozen and stored at -80 °C.

Exchange to caged-GTP was done according to a modified protocol of (John et al., 1988). Rap•GDP was incubated with a five-fold excess of caged-GTP and catalytic amount of alkaline phosphatase in a buffer containing 50 mM Tris (pH 8.5), 5 mM DTT, 10 µM ZnSO4 and 200 mM ammonium sulfate. After 3 hour incubation at room temperature, the protein was rebuffered and excess caged nucleotide was removed by gel-filtration using NAP-5 (2 mg protein) or PD10 (5 mg protein) column. Collected fractions were checked for protein by spot test, pooled, centrifuged and the supernatant was flash frozen in stocks after protein

33 Materials and methods concentration estimation (2.3.2) and nucleotide estimation (see 2.3.11). Typically, the exchange to caged-GTP was approximately 65% complete, the rest being GDP.

2.3.11 Nucleotide detection using reversed-phase HPLC

This method was carried out according to Lenzen et al. (1995). The principle of nucleotide separation is the interaction between the hydrophobic static phase and the ion pair of nucleotide and tetrabutylammonium in the mobile phase. Depending on the number of phosphates, a variable number of tetrabutylammonium ions are bound by the nucleotide, which increases the retention time on the column.

The sample was applied on a HPLC system Gold 166 (Beckman, Palo Alto, USA) and separated via a reversed-phase column ODS Hypersil C18 (Bischoff, Leonberg). Denatured proteins were adsorbed at a nucleosil-100-C18 precolumn. The running buffer contained 10 mM tetrabutylammoniumbromide, 100 mM potassium phosphate (pH 6,5) with 7,5% acetonitrile. Nucleotide peaks were detected by measuring adsorption at 254 nm and quantified by integration. The column was calibrated by standard nucleotide solutions.

2.3.12 Fast kinetics using stopped-flow measurement

Transient kinetic experiments were carried out according to (Kraemer et al., 2002) using an SX18MV-Stopped-flow-apparatus (Applied Photophysics, Leatherhead, U.K.). Aedans labeled protein was excited with monochromatic light (λ = 350 nm, band width = 6,4 nm) and emission was followed using a cut-off filter (λ > 408 nm). For every time course, 1000 data points were measured.

2.3.12.1 Initial reaction assays

The reaction buffer was 50 mM Hepes (pH 7,5), 100 mM NaCl, 5 mM MgCl2. 2 µM Rap1- Aedans•GTP (or mutant) was 1:1 mixed with 50 µM Rap1GAP in a stopped-flow apparatus and the fluorescence was followed. All traces shown are the average of at least three (in most cases five) individual measurements.

− − 2.3.12.2 Transition state experiments (AlFx , BeF 3 , MgF 3 and vanadate)

Rap1-Aedans (or mutant) was incubated with a catalytic amount of Rap1GAP at 20 °C for 120 min to convert it in the GDP bound form. For stopped-flow measurements, 2 µM Aedans-

Rap1•GDP in 50 mM Hepes (pH 7,5), 100 NaCl, 5 MgCl2, 500 µM AlCl3 (or BeF2), 5 mM NaF was rapidly mixed with 50 µM of the corresponding Rap1GAP construct and the fluorescence

34 Materials and methods

− monitored over 500 sec. The MgF 3 experiments were done using only 5 mM NaF and

MgCl2. Similarly; the vanadate experiments were done with 500 µM vanadate prepared according to published protocol (Gordon, 1991) in the standard buffer.

2.3.12.3 Analysis of biochemical data

The fluorescence was normalized by dividing all data points by the initial value. The data were analyzed either by using the program Grafit 5.0.6 (Erithacus Software Limited) or Scientist 4.0 (Micromath Scientific, Microsoft Corp.).

2.3.13 Quenched Flow analysis

Quenched flow measurements were done in a KinTek Instrument (University Park, PA, USA), using 300 µM Rap1GAP and 50 µM Rap•GTP at 10 °C. Reactions were quenched by using 10% TFA (Trifluroacetic acid) at time points 0 ms, 5 ms, 10 ms, 15 ms, 20 ms, 30 ms, 40 ms, 50 ms, 60 ms, 70 ms, 80 ms, 90 ms, 100 ms, 200 ms, 300 ms, 400 ms, 500 ms, 1 s, 10 s and infinity time (30 minutes) and flash frozen in liquid nitrogen. After thawing, the reaction tubes were centrifuged for 10 minutes at 4 °C, the supernatants collected, neutralized with ammonium hydroxide, evaporated by speed-vac, dissolved in tetrabutylammoniumbromide solution (3.2 g/l), pre-cleaned with ZipTips™ (ZTC1 8S0, Millipore) and analyzed by MALDI (see 2.3.3) using THAP (2,4,6-trihydroxyacetophenone) (10 g/l THAP in acetonitrile and 50 g/l Citric acid, mixed in 1:1 ratio) with 0.2% TFA.

2.3.14 Electrospray ionization mass spectrometry (ESI-MS)

Electrospray ionization mass spectrometry was done for phosphate groups to confirm the incorporation of 18O in released phosphate as seen in FTIR spectra (results, sections 3.1.3, 3.1.5, 3.1.6) with the help of Dr. D. Wolters at Ruhr-Universität Bochum. After performing the 18 GTPase reaction in H2 O inside the FTIR sample chamber, the sample was extracted with 100 µl of 50% methanol water and centrifuged with a 10 kD vivaspinTM (Vivasciences, Hannover) to remove the protein. The low molecular weight flow-through was analyzed by electrospray ionization mass spectrometry between m/z 50 to 600 in a Brucker Esquire 3000 PlusTM instrument (Brucker Daltonik GmbH, Bremen, Germany), operating in negative ion mode with a spray voltage of 4 kV. The data was processed by manufacture’s software and plotted with Origin 7.0TM.

35 Materials and methods

2.4 Cell Biological methods

2.4.1 Cell culture

To recultivate cryopreserved stocks of Cos7 and HEK293 cells supplemented with 10% DMSO, the cells were thawn at 37°C in water bath or at hand and 1.5 x 105 cells were added under steryl workbench to 10 cm dishes and incubated with 10 ml Dulbecco’s modified Eagle medium (DMEM, with added L-glutamate and Pyruvate) (GIBCO, Invitrogen) supplemented with 10% (v/v) Fatal Calf Serum (FCS) for two hours at 37 °C, 7.5% carbondioxide (CO2) and relative humidity of 95%. After cell adhesion, the medium was aspirated to remove DMSO and fresh medium (DMEM + 10% FCS) containing antibiotic (P/S: Penicillin, 100 U/ml; Streptomycin, 100 µg/ml) was added and incubated under similar condition. Each plate was maintained upto a cell density of 106 (80% confluence) at 37 °C and 7.5% Carbondioxide

(CO2) with 95% humidity.

Shortly before confluence, as monitored under light microscope, the cells were passaged to new subcultures. The medium was aspirated, washed with steryl PBS and treated with

Trypsine-EDTA and incubated for 5-10 minutes at 37 °C, 7.5% carbondioxide (CO2) and relative humidity of 95%. Once the cells were detached, fresh medium was added to it to resuspend, the cell density counted using Neubauer cell counter where, the average number of cell in four wells out of total sixteen were multiplied by 1x104 to get the cell density (cells/ml). 1x105 cells were thereafter transferred to new containers or dishes.

2.4.2 Cryopreservation

Longer storage of cells of any particular passage is mediated by cryopreservation in liquid nitrogen. After passage and resuspension, the cells were centrifuged at 800g for 5 minutes and the supernatant was discarded to remove trypsin. The cells were resuspended in DMEM with 10% (v/v) FCS and 10% (v/v) DMSO as an antifreeze. Approximately 1x106 cells are stored in a cryotube and cooled in a cryocontainer (Nalgene) containing 2-propanol which ensures a slow cooling rate of 1°C/minute when kept at – 80 °C freezer. The next day, they are transferred to liquid nitrogen container.

2.4.3 Transfection

Cos7 / HEK293 cells were grown up to 70% confluence in 6 well plate (3.5 mL). 2µg plasmid DNA in 250µl of OPTI-MEM medium (without serum) was mixed gently. Lipofectamin 2000 (2 µl/µg DNA) was mixed to another in 250µl of OPTI-MEM medium. The two mixtures were incubated separately for 5 minutes at room temperature. Next, they were combined and incubate for another 20 minutes at room temperature. Meanwhile, the cells were washed

36 Materials and methods once with OPTI-MEM medium. 500µl of the DNA-Lipofectamin complex was added to the each well containing cells and incubate for 1 hr 37 °C and 7.5% Carbondioxide (CO2) with 95% humidity. Afterwards, 2.5 ml DMEM supplemented with 10% FCS and antibiotic (PS) was added to the each well incubate under aforementioned condition for 40 hrs. After 40 hrs process the cells for further studies.

2.4.4 β-Galactosidase assay

β-Galactosidase gene is not used as a primary reporter gene, but to check transfection efficiency. Therefore it serves as an internal control of transcription in transfected cells. This gene is kept under a constitutive promoter and the activity of the enzyme can be assayed by colorimetric method by the conversion of o-Nitrophenyl-β-D-galactoside (ONPG) after 20-30 minute incubation with Z-buffer (see 2.1.10) at 37 °C, 7.5% carbondioxide (CO2) and relative humidity of 95%.

2.4.5 Immunoprecipitation

The cells were washed with ice-cold PBS and lysed with 250-µl lysis buffer (2.1.10) and incubate on ice for 5 minutes. Breakage of cells was checked under light microscope. The lysed cells were scraped and pipetted in tubes. After centrifugation at 14,000 rpm for 15 minutes at 4°C, the supernatant was promptly separated from nuclei and cell debris. 50 µl of this supernatant was mixed with 16 µl 4x Lammli buffer, heated at 95 °C for 10 minutes and preserved as respective Total cell lysate (T). 1.5 µl of M2 (Ab) antibody (α-Flag) was added and incubated for two hours at 4°C. 25 µl bed volume of protein G Sepharose that has been equilibrated with lysis buffer, was added to this and incubate for one hour with rotation at 4 °C. The tubes were centrifuge at 2000 rpm for 5 minutes at 4°C and 50 µl of the supernatant was mixed with 16 µl 4x Lammli buffer, heated at 95 °C for 10 minutes and preserved as respective supernatant of immunoprecipitation (S). The rest of the supernatant was carefully discarded, washed twice with 500 µl lysis buffer and twice more with lysis buffer without NP40. 50 µl of 1x Lammli buffer was added to part of these beads, cooked and stored as respective immunoprecipitated beads (B). The fractions marked as (T), (S) and (B) were analyzed in western blot.

2.4.6 Western blot

The samples were electrophorased in 8% SDS-PAGE with pre-stained marker (Biorad, broad range) and immediately after the run, transferred into transfer buffer. The PVDF-Membrane (Hybond-P, Amersham Biosciences) was activated in methanol for 5 minutes and both membrane and 3 mm filter papers were equilibrated in transfer buffer for 30 minutes. The

37 Materials and methods proteins were transferred for one hour at 180 mA current in the cold room, with constant stirring. In the next step, the membrane was blocked with blocking buffer (1g skimmed milk in 20 ml PBS). The skimmed milk solution was diluted 1:1 with PBS, the required antibody was added (α-Myc 1:500, α-Flag 1:2500) and incubated for two 2 hours at room temperature or overnight at 4°C in 5 mL chamber (protein side should be upside). Afterwards, the membrane was washed 3 times with TPBS (PBS + 0.1% Twin 20) and 1 µl secondary antibody (mouse Ig, biotinated) was added in 5 ml PBS. After another two hours of room temperature incubation, the membrane was further washed for three times with TPBS and 1 µl of the tertiary antibody (Streptavidine coupled Horse radish peroxidase) was added in 5 mL chamber and incubated for two hours at room temperature. After final three times wash with TPBS the protein side of the membrane was treated with premixed 800 µl Solution A and 20 µl solution B (ECL plus Western Blotting Detection system, Amersham Biosciences) and incubated for 5-10 minutes before developing x-ray film.

2.4.7 Immunostaining

After transfection (2.5.4) in 3.5 mm dishes with a cover slip and desired incubation, the medium was aspirated and the cells were washed with 1 ml of steryl PBS and aspirated again. In each well, 2 ml PBS with 4% paraformaldehyde was added and incubated for 15 minutes at ambient temperature. After five fold wash with tPBS (PBS + 0.1% Triton X), the cover slips were transferred to a light protected box and 3% skimmed milk in PBS was added for blocking. After one hour of incubation at room temperature, 100 µl of 1:100 primary antibody was added in PBS containing 3% skimmed milk. After another hour of room temperature incubation, the cover slips were washed five times with tPBS with 5 minutes incubation between every wash. The secondary antibody (FITC or TRITC labeled Ig mouse) was added and re-incubated for 40 minutes. After another five times wash with tPBS with 5 minutes incubation, the cover slips were mounted on slides with mounting solution, extra water was removed and sealed. These ready samples can be kept in cold room for several days, under light protected condition. These samples were checked by confocal microscopy in a Leica DM IRBE instrument (Leica Mikroskopsysteme, Bensheim).

38 Materials and methods

2.5 Time resolved infrared spectroscopy

2.5.1 Instrumentation

A typical time resolved FTIR instrument contains a Globar (SiC [silicon carbide] heated at 1500 K) as a light source. Infrared light from the globar passes an aperture (0.25 mm to 12 mm) before entering a Michelson interferometer (Figure 2.1a). Subsequently the light passes the sample chamber, which is equipped with a thermostatic transmission cell in which the sample is placed. Finally the infrared light reaches a liquid nitrogen cooled MCT detector (mercury-cadmium-tellurium) detector. For triggering of a reaction, the sample in the cell can additionally be irradiated by a laser.

An FTIR instrument has crucial advantages over a dispersive spectrometer. By means of the incorporated Michelson interferometer (Figure 2.1a), all wavelengths can be measured in parallel (multiplex / Felgett advantage). At the beam splitter, one half of the infrared light is reflected onto a fixed mirror, while the other half is transmitted to a moving mirror. Both parts then recombine at the beam splitter. Depending on the position of the moving mirror, a path

Figure 2.1. (a) Schematic representation of a Michelson interferometer. An electromagnetic wave is splitted at the beam splitter. (b) The result of measurement of an interferogram, where intensity is plotted against the mirror position. After Fourier transformation, the intensity ( I ) is obtained as a function of the wavelength (single channel spectrum) [modified from (Kötting and Gerwert, 2005)]. difference (∆x) is created between the beam reflected by the fixed mirror and the moving mirror. The moving mirror at position x1 will lead to a path difference of zero between two waves of a monochromatic radiation and therefore to an intensity increase by a factor of 2, whereas a path difference of half a wavelength (mirror position x2) leads to extinction. This phenomenon is known as interference. The aforementioned cases are known as constructive and destructive interference, respectively. For monochromatic radiation, the Interferogram can be described as a cosine function of the mirror position x as described by equation 2.1.

39 Materials and methods

~ I(x) = I 0 []1+ cos(2πνx) Equation 2.1

I0 designates the partial radiation intensity. Polychromatic light source therefore leads to an overlap of several cosine functions. If a polychromatic radiation is used, the individual contributions are to be integrated as described by equation 2.2 below.

∞ I(x) ~ ∫ S(ν~) ⋅ exp(−2πiν~x)dν~ Equation 2.2 0

In this way, variation of path difference alters the interference pattern after the recombination, producing an interferogram in which intensity is plotted against mirror position x (Figure 2.1 b, Interferogram). Through mathematical operation known as Fourier transformation (equation 2.3), a single channel spectrum (Figure 2.1 b) is obtained, where the intensity, I, is a function of the wavelength.

∞ S(ν~) ~ ∫ I(x)exp(2πiν~x)dx Equation 2.3 −∞

An absorbance spectrum is obtained by comparison of two single channel spectra, one with sample and one without sample respectively. As an example, absorbance spectra (A) of a protein can be calculated from the single channel spectrum of the protein (I) and the single channel spectrum of the buffer as reference (I0) by equation 2.4.

A = -log (I/I0) Equation 2.4

In practice, the Interferograms are not recorded continuously, but in n discreet points n·∆x with the maximum path difference of the partial beams N·∆x. Hence interferogram consists of N discreet points at an interval of ∆x. The spectral resolution ∆ν~ is the reciprocal of the product from the point distance ∆x and the number of points N as described in equation 2.5.

1 ∆ν~ = Equation 2.5 ()N ⋅ ∆x

The calculation of the spectrum from the discreet interferogram can be done by discreet Fourier transformation as described by equation 2.6.

N −1 ⎛ nk ⎞ S(k ⋅ ∆ν~) ~ ∑ I(n ⋅ ∆x) ⋅ exp⎜2πi ⎟ Equation 2.6 n=0 ⎝ N ⎠

With modern spectrometers, a complete spectrum can be obtained within 10 milliseconds. Further advantages of FTIR-spectrometers are the absence of dispersive elements (slits combined with prisms or gratings which attenuate the signal intensity: Jaquinot advantage), and the high accuracy of the wavelength (Connes advantage).

40 Materials and methods

2.5.2 Basic principles of Infrared spectroscopy

Any molecule, including proteins can be envisioned as a system of mass points connected by springs (the bonds). Such systems can undergo distinct vibrations. A vibration can be induced by electromagnetic field, provided the energy of the field matches to the energy of the vibrational mode and the dipole moment of the molecule changes during this vibration. These interactions result in absorbance of distinct ‘quantized’ energies. In case of vibrational changes, such energies are in the infrared spectral range. Therefore, infrared spectroscopy is a vibrational spectroscopy. The stretching frequency ν can be correlated to the reduced mass (µ) and force constant (k) as described by equation 2.7.

1 k m ⋅ m ν = ⋅ where µ = 1 2 Equation 2.7 2π µ m1 + m2

If one of the atoms of such a pair is replaced by an isotope, the force constant does not change significantly but the reduced mass (µ) changes. The isotopic stretching frequency

νisotope can be easily calculated by the following equation 2.8. This is the basis of infrared band assignment which can be done either by isotopically labeled protein (Engelhard et al., 1985) or in case of nucleotides, by specific 18O labeling (Allin and Gerwert, 2001) or by amino acid exchange via site directed mutagenesis (Gerwert et al., 1989).

µ ν Isotope =ν ⋅ Equation 2.8 µ Isotope

Even a relatively small protein of 20 kD (like Ras) contains about 104 vibrational modes. In addition, hydration is necessary for activity of biomolecules. Since water is a strong absorbent in the infrared region, from a typical absorbance spectrum one can obtain global information of proteins but not about small individual changes. This problem is circumvented by difference spectroscopy where two absorbance spectrums are subtracted from each other. Typically, for a reaction A Æ B, one calculates (B – A). Thus, the vibrations of groups, which are unchanged during the reaction, are cancelled out and only the ones that change during the reaction are seen. In this process, individual absorbances of groups can be resolved from the background, which are 103 weaker. This is possible through identical measurement conditions, high sensitivity and stable instrumentation.

2.5.3 Sample preparation

Figure 2.2 shows a typical transmission cell with IR transparent windows (e.g. CaF2 or BaF2). Due to high absorptivity of water in the mid infrared spectral region, meaningful spectra of hydrated proteins are obtained only by transmission measurements through very thin film (2 – 10 µM), adjusted by a mylar-spacer. A drop of protein solution is placed on the IR- transparent window and carefully concentrated under a nitrogen stream. A typical

41 Materials and methods measurement requires about 100 – 200 µg protein and the concentration of protein in the film is 6 – 10 mM. The sample chamber is closed by a second IR-window and sealed by means

Figure 2.2. Schematic representation of a transmission cell (Kötting and Gerwert, 2005). of the O-rings. The sample holder has high thermal conductivity and it is placed inside the spectrometer for thermal equilibration.

For Rap•RapGAP system, the sample solutions were prepared essentially as described (Allin et al., 2001; Allin and Gerwert, 2001) with a slight modification for the GAP catalyzed case in that RapGAP was directly taken from a stock concentration of 25 mg/ml in 20 mM HEPES (pH 7.6), 100 mM NaCl and 5 mM DTE without further desalting and rebufferring. This turned out to be crucial for optimal reactivity of RapGAP. The exchange to caged-GTP was ca. 65% complete, 35% of Rap remaining in the GDP-bound state.

The very slow intrinsic reaction of Rap was difficult to monitor due to baseline drifts. Hence a catalytic amount of GAP has been used to accelerate the reaction for the purpose of band assignment. This procedure doesn’t change the band positions but significantly improves the baseline stability (Allin and Gerwert, 2001). For the GAP catalyzed reaction, RapGAP was used in 1.6 molar excess over Rap•caged nucleotide to achieve single turnover. All the 1% GAP catalyzed (“intrinsic”) reactions were done at 283 K and the GAP catalyzed reactions at 258 K. DTT was used to scavenge the reactive photolysis by-product 2-nitrosoacetophenone. 18 18 For the measurements in H2 O, the sample was treated with 10 µl H2 O (> 95% enrichment) and evaporated four times.

2.5.4 Initiation of the reaction and data collection

The number of functional groups involved in protein-protein interaction is small compared to the total number of groups. Therefore the absorbance changes arising from an interaction are several orders of magnitude smaller than background absorbance of the protein and buffer. Thus the sample needs to be thermally stable and activated without being removed from the sample chamber. Thus, the FTIR difference technique requires a sharp trigger. In the context of GTPase reactions this is achieved by photo labile cage compounds as shown 42 Materials and methods in Figure 2.3. The 1-(2-nitrophenyl) ethyl (NPE; Figure 2.3 a) moiety is frequently used to protect phosphate, nucleotides and nucleotide analogues. Application of UV flashes leads to photolysis, forming GTP and the by-product 2-nitrosoacetophenone (Figure 2.3 a) (McCray et al., 1980). A spectrum of caged GTP is measured prior to the photolysis as reference and after photolysis; further spectra are recorded (Figure 2.3 c, A to B part) and absorbance difference spectra are calculated (Figure 2.3 d, A to B part) through global fit (section 2.4.6). For a GNBP bound to caged GTP and even with another protein like GAP, the basic principle is the same, as shown in Figure 2.3 b. In this case, the reaction proceeds to hydrolysis, with the involvement of intermediates in case-to-case basis.

The spectra recorded in this work were collected in rapid scan mode. The principle of rapid scan FTIR method is: after taking a reference spectrum of the sample in its ground state, it is activated (e.g. by a laser flash) and interferograms are recorded in much shorter time than the half lives of the reactions (Gerwert et al., 1990). The first four reference (R) interferograms (Figure 2.3 c) represent ground state (A) (Figure 2.4 d) while the following interferograms are taken during the reaction pathway (B to C to D).

Figure 2.3. (a) Photolysis of caged [1-(2-nitrophenyl) ethyl] GTP. (b) Time course of a rapid scan FTIR experiment, investigating the interaction of two proteins (in red and blue color). (c) Time course of data acquisition. First reference spectra (R) are taken. Then a laser flash initiates the reaction. During the reaction via intermediate C through product D, interferograms are continuously recorded. (d) After kinetic analysis, e.g. by global fit amplitude spectra are obtained, showing the bands of the groups involved in the reaction. [modified from (Kötting and Gerwert, 2005)]

43 Materials and methods

Velocity of the interferometer’s moving mirror, Vmax, and the desired spectral resolution ∆ν determines the scan duration ∆t (time resolution) as described by equation 2.9.

1 1 ∆t = ⋅ Equation 2.9 2Vmax ∆ν

2.5.5 Measurement parameters

Sample composition: Protein solutions were concentrated to a final concentration of 6-10 mM on 20x2 mm CaF2 or BaF2 windows under nitrogen stream along with selected buffer system. Typical buffer composition is: 200 mM Buffer (HEPES for pH 7.6, MES for pH 6.0), 20 mM

DTE / DTT, 20 mM MgCl2. Intrinsic or 1% GAP catalyzed reactions were done at 293 K and GAP catalyzed reactions were done at 258 K, along with 12% ethylene glycol to avoid freezing.

Special instruments: IFS 66v/s FTIR spectrometer (Bruker Optics, Karlsruhe, Germany) with KBr beamsplitter, MCT detector KMPV11 – 1 – J1 (Kolmar, Newburyport, MA, USA),

Excimer-Laser LPX 240i (Lambda Physik, Göttingen, Germany) with XeCl (308 nm), CaF2 /

BaF2 windows (Korth GmbH, Kiel, Germany), Mylar film (DuPont, Circleville, OH, USA)

Operational mode: The intrinsic reactions were monitored in ‘single sided fast return’ mode with a scanner velocity of 100 KHz and the GAP catalyzed reactions were monitored in ‘double sided forward backward’ mode with a scanner velocity of 320 KHz. In both cases, the spectral resolution was 4 cm−1 and folding limit 3949.5 cm−1. The laser energy was between 90 and 120 mJ per flash with a pulse duration of ~ 20 ns. Thirty flashes were applied to achieve 70-80% conversion of caged-GTP in about 90 ms. The chosen number of flashes was a compromise between the time of photolysis and the degree of conversion. A time window of 14 ms per interferogram was allowed for the scanner to return to initial position while measuring at 320 KHz. The time resolved data were collected by OPUSTM (under OS2) and analyzed by global fit programs with IDL© and Matlab 12.1 (Math Works, Aachen, Germany).

2.5.6 Data analysis

Difference spectra for the 1% GAP catalyzed reaction (or intrinsic reaction) were made by using the software OPUSTM (Bruker). The data of the time-resolved fast scan measurements for the GAP catalyzed reactions were analyzed between 1800 and 950 cm−1 with a global fit method (Hessling et al., 1993). The global fit analysis not only fits the absorbance change at a specific wave number, but also upto 800 wave numbers in the spectrum, simultaneously. All reactions are assumed to be first order and can therefore be described as a sum of exponentials. The fit procedure minimizes the difference between the measured data

44 Materials and methods

∆Ameasured and the theoretical description ∆A, weighted according to the noise wij at the respective wave numbers, and summarized not only over time (tj) but also the wave numbers

(i). Thus, the absorbance changes ∆A in the infrared are analyzed with sums of nr exponentials with apparent rate constants kl´and amplitudes al as described in equation 2.10.

nr −kl ´t ∆A(ν ,t) = ∑ al (ν ) ⋅ e + a0 (ν ). Equation 2.10 l=1

In the analysis, the weighted sum of squared differences (f) between the fit with nr rate constants (kl´) and data points at nw measured wavenumbers (νi) and nt time-points (tj) is minimized as shown in equation 2.11.

n wtn nr 2 −kl ´t j 2 f = ∑∑(wij ) (∆Ameasured (ν i ,t j ) − ∑al (ν i ) ⋅ e + a0 (ν i )) ⋅ Equation 2.11 i==11j l = 1

For unidirectional forward reactions, the determined apparent rate constants are directly related to the respective intrinsic rate constants describing elementary reaction steps. If significant back reaction also occurs, the analysis becomes more complicated. In this case, the reaction has to be modeled until the estimated intrinsic rate constants fulfill the experimentally observed time courses described by the apparent rate constants. Because the number of intrinsic rate constants needed in a model is usually larger than the experimentally observed ones, the problem is experimentally underdetermined, and the solution is equivocal. This limitation holds for any kinetic analysis, not only for IR measurements.

An alternative method for analyzing the data is singular value decomposition, or principal component analysis (PCA) approach (Hessling et al., 1993). In this method, a set of basis difference spectra is calculated from all difference spectra measured. This procedure allows the determination of transient spectra, independent of specific kinetic models, and independent of the temporal overlap (Hessling et al., 1993; Ruckebusch et al., 2003).

2.5.7 Stability of spectral regions

Each spectrum shown in this thesis is an average of at least 3-5 reproducible measurements. In Figure 2.4 all ‘averaged’ amplitude spectra for the intermediate and product states of Rap•RapGAP reaction and corresponding double differences (results, section 3.1.3 and 3.1.6) are overlaid. Baseline correction is done at the stable spectral range between 1800 and 1700 cm−1. The region between 1690 to 1520 cm−1 is less stable due to slightly different water content and sample thickness between individual measurements. Absorbance of protein side group takes place between the stable range 1500 to 1350 cm−1 and that of phosphates between 1350 to 950 cm−1. Sample thickness changes for all wavenumbers, but in disturbed region the signal at the signal channels is much smaller.

45 Materials and methods

Figure 2.4. Stability of spectral regions. A) The spectra of the intermediate of Rap•RapGAP reaction and B) the corresponding double difference spectra (discussed in 3.1.4) C) The spectra of the product release step of Rap•RapGAP reaction and D) the corresponding double difference spectra (discussed in 3.1.5).

2.5.8 Reproducibility within the phosphate region

Figure 2.5 illustrates reproducibility and baseline stability of the spectral data for the GAP catalyzed reaction. Two representative amplitude spectra of the unlabeled GTP and with [β,γ- 18 18 O/γ- O3] label corresponding to the third rate (see 3.1.5) are superimposed in Figure 2.5A

Figure 2.5. Reproducibility in the phosphate region (1300 cm−1 to 950 cm−1). For explanation, see text.

46 Materials and methods and 2.5B. Figure 2.5C represents the double difference of the two individual spectra used in A (black line) and B (red line). This indicates baseline stability and reproducibility of individual measurements. Individual double difference spectra (B minus A) for two measurements are shown in Figure 2.5D. One can clearly see that the spectral features are conserved and the peak positions and shifts are reproducible.

47 Results 3 Results

3.1 FTIR on Rap•RapGAP Reaction

Time-resolved Fourier transform infrared (trFTIR) difference spectroscopy has been successfully applied to elucidate molecular reaction mechanisms of proteins (Gerwert, 1999; Gerwert, 2001; Kötting and Gerwert, 2005). By subtracting the spectra collected at the educt state and at different time points during the reaction, the absorbance changes of few residues involved in the reaction are selected from the quiescent background absorbance of the whole protein.

The FTIR methodologies including basic principles, sample preparation, initiation and data collection, data analysis and spectral band assignment are described in detail in materials and methods (see section 3.5). To avoid hydrolysis of GTP during stabilization of sample, the samples are made with nonhydrolyzable caged GTP containing photo labile protecting group. After triggering the GTPase reaction by UV laser flashes that removes the protecting group, the IR absorbance changes of the hydrolysis reaction were monitored between 1800 cm−1 and 950 cm−1 simultaneously, with 10 ms time-resolution in the rapid scan mode (Gerwert et al., 1990). The time resolved absorbance changes were analyzed by a multi-exponential global fit analysis (see materials and methods, section 3.5.6), which reveals the amplitude spectra and the corresponding apparent rate constants. As a convention, negative bands represent the vibrations of the educts whereas positive bands reflect vibrations of the products.

3.1.1 Summary of previous FTIR studies on Ras and RasGAP

Earlier work on Ras emphasized several important aspects in the RasGAP catalyzed GTPase reaction of Ras: (i) upon Ras binding, electron density at the β-γ bridging bond of GTP is reduced and at the same time, is enhanced at the non bridging β oxygen of GTP. FTIR spectroscopy provides information about such minute changes since the observed stretching frequency (ν) for a particular vibrational mode can be correlated to the reduced mass (µ) and force constant (k) as described previously by equation 2.7.

1 k m ⋅ m ν = ⋅ where µ = 1 2 2π µ m1 + m2

Hence, for the same pair of atoms, if there is a redistribution of electron density, that would affect the force constant (bond strength) and consequently the vibrational frequency as observed in infrared spectroscopy. The aforementioned charge redistribution effectively means that upon binding of Ras, negative charge is drawn towards the non-bridging β

48 Results oxygen of GTP from β-γ bridging bond (Cepus et al., 1998). Upon GAP binding, this effect is enhanced already at the GTP state as seen in the case of RasGAP hydrolysis catalyzed reaction of Ras (Allin et al., 2001). This indicates that GAP has a role in charge redistribution even at the ground state. ii) It was observed that an intermediate accumulates during the course of the reaction. The explicit nature of the intermediate was not understood and it was argued that either the intermediate could be transiently phosphorylated species or GDP•Pi.

(iii) Pi release from the Ras•GDP•Pi•GAP complex is rate limiting and (iv) the GTPase reaction appears to be reversible (Allin et al., 2001).

It is noteworthy that such FTIR measurements require high concentration of proteins (5-10 mM) and for the GAP catalyzed hydrolysis reaction; lower temperature is essential for resolving intermediates. While change of conditions do not change the mechanism, they are likely to affect protein protein interactions, namely dissociation of GAP from the

GNBP•GDP•Pi•GAP complex after hydrolysis. Therefore, the FTIR spectroscopic method provides more details about the molecular reaction mechanism but the description might be different or incomplete while compared to solution phase rapid kinetic experiments concerning protein protein interactions. With this consideration, the fact that Pi release appears to be the rate-limiting step in FTIR measurements (Allin et al., 2001), is in line with conclusions drawn from stopped-flow experiments (Nixon et al., 1995; Kraemer et al., 2002;

Phillips et al., 2003) where Pi release or the dissociation of NF1 (neurofibromin, a RasGAP) from the Ras•GDP•Pi•NF1 intermediate has been suggested to be rate limiting.

The charge redistribution towards the β-phosphate of GTP is probably due to the positively charged Lys16 and Mg2+, which are equidistant from the nonbridging β and γ-oxygen (Pai et al., 1990). This leads to a redistribution of excess negative charge from γ to β phosphate, as shown by QM/MM calculations (Klähn, Schlitter & Gerwert, unpublished results). In addition, the P-loop backbone N-H groups point towards the nonbridging (Sp)β-oxygen and might assist in this charge shift as proposed (Cepus et al., 1998; Allin and Gerwert, 2001). In the presence of GAP, there is an even larger charge shift towards (Sp)β oxygen of GTP and it was assumed that this is due to the presence of the additional positive charge of the arginine finger (Allin et al., 2001). Since this is the only effect on the GTP and GDP vibrations due to GAP binding, it was proposed to be crucial for the catalysis. The question arises as to whether Rap1GAP without an arginine finger (see introduction, section 1.2.4) still shifts charge towards the β-phosphate and the features described for the Ras-RasGAP system are generally applicable to other GTP-binding proteins and their respective GAPs. Here the Rap- RapGAP system appeared to be particularly interesting due to the absence of both a Gln in the GTPase and an Arg residue from the GAP that are considered to be crucial in the case of the Ras (Ahmadian et al., 1997b), Rho (Leonard et al., 1998; Graham et al., 1999b) and Rab GTPases (Albert et al., 1999).

49 Results

3.1.2 Intrinsic reaction of Rap.

To evaluate the effect of RapGAP on the stimulation of GTPase reaction of Rap, at the first hand it was necessary to characterize the intrinsic GTPase reaction of Rap without addition of GAP. Since the intrinsic reaction of Rap is very slow, it is difficult to monitor in FTIR due to baseline instability. Therefore, catalytic amount (1%) of RapGAP was used to achieve significant rate of hydrolysis. This procedure has been tested on Ras, where addition of catalytic amount of RasGAP does not change the band positions of intrinsic Ras spectra (Cepus et al., 1998), but improves the quality of spectra and therefore band assignment (Allin and Gerwert, 2001). From now on, intrinsic reaction of Rap would indicate 1% GAP catalyzed reaction.

The data for the intrinsic reaction (with 1% added RapGAP) can be fitted with one exponential function using the global fit algorithm (materials and methods), which yields amplitude spectra and apparent rate constant. Alternatively, for a slow reaction like the intrinsic hydrolysis, spectra corresponding to pre-hydrolysis and post hydrolysis states can be chosen selectively and a direct subtraction would lead to difference spectra that are nearly identical to amplitude spectra. The difference spectrum of the intrinsic reaction of Rap is compared to that of Ras in Figure 3.1b below.

Figure 3.1. Comparison of Ras and Rap structures and spectra. (a) overlay of GTP, Mg2+, Lys-16 and P-loop from the structural models of Ras-GppNHp (PDB code: 5p21) and Rap2.GTP (PDB code: 2rap). (b) Comparison of FTIR difference spectra from the intrinsic GTPase reaction of Rap (red) and Ras (black).

50 Results

The spectra agree nicely within the spectral region between 1350 cm−1 and 950 cm−1, where phosphate vibrations are expected. An intensity change at the γ vibration is seen which might be due to different interactions of the γ-phosphate with Gln61 or Thr61 in Ras and Rap, respectively. The agreement shows that the protein-bound GTP structure and charge distribution as probed by FTIR are very similar for Ras and Rap. This is in good agreement with x-ray structural models showing a very similar GTP structure and binding niche (Figure 3.1a). However, more deviations are seen for the protein part of the spectra between 1800 cm−1 and 1350 cm−1.

The bands appearing in the amplitude or difference spectra need to be assigned for interpretation. This is done with specific 18O labeled GTP as illustrated in Figure 3.2 for the intrinsic hydrolysis reaction of Rap. The four fold γ-18O label shifts the bands at 1157 cm−1 (shoulder) and 1143 cm−1 to 1119 cm−1 and 1106 cm−1 (aA) leaving all other vibrations unaffected. Therefore these specific bands correspond to γ-GTP vibration.

Figure 3.2. Assignments for the intrinsic hydrolysis reaction of Rap: (a) the difference hydrolysis spectra with normal (black lines) and labeled GTP (colored). The dotted line corresponds to the difference hydrolysis spectrum of Ras, for comparison. The 18 18 18 18 18 spectra of the labeled GTP are A) [β, γ- O/γ O3], B) H2 O, C) Sp-[β- O], D) Rp-[β- O] 18 and E) [α O2]. (b) The corresponding double difference spectra. The double 18 difference spectrum for H2 O (bB) is 2.5 fold magnified.

51 Results

The frequency shifts are illustrated by means of double difference spectra (spectrum of labeled minus spectrum of unlabeled GTP) with shaded difference bands to enhance visualization. The double difference spectrum shown in Figure 3.2 bA indicates the frequency −1 −1 shifts as difference bands at 1157/1117 cm and 1140/1103 cm . The product ‘Pi’ is also labeled and shows frequency shifts from 1078 cm−1 to 1054 cm−1 (Figure 3.2 aA), which is seen as a double difference band at 1076/1054 cm−1. (Figure 3.2, bA). The γ-labeling clearly leaves the other bands unaltered indicating isolated vibrations in Rap-bound GTP.

Ideally, in the double difference spectra only the vibrations of labeled group should appear as difference bands. Other small effects in the double difference spectra may appear due to baseline shifts and small variations between different sample preparations. Nevertheless, double differences are very useful for understanding and illustration purposes where due to spectral overlap; the assignments are sometimes difficult to follow from amplitude spectra. The band assignments presented in this work are derived the observed frequencies of the amplitude or difference spectra.

−1 −1 18 The GTP band at 1215 cm is downshifted to 1207 cm with a (Sp)-β- O label (Figure 3.2 aC), which leads to the assignment of GTP-β vibration at 1215 cm−1. Similarly, the specific α- GTP label shifts the GTP band at 1260 cm−1 to 1218 cm−1 (Figure 3.2. aE) and the GDP band from 1236 cm−1 to 1195 cm−1 (Figure 3.2. aE). This is observed as double difference band for GTP at 1261/1221 cm−1 and for GDP at 1237/1197 cm−1 (Figure 3.2 bE). Measurements with 18 labeled water (H2 O) resulted in labeling of the released phosphate groups as identified in the double difference spectra as difference bands at 1076/1054 cm−1 and as hardly resolved band at 992/972 cm−1. In this case, the original change in intensity is observed at 1078 cm−1 and 1054 cm−1 (Figure 3.2 aB) in the amplitude spectra and this is exactly the same as that of four fold γ-18O label (Figure 3.2 aA). However these small changes are better resolved in the double difference spectra (Figure 3.2 bB).

In principle, if the GTP vibrations were coupled, labeling would shift the vibrations of all groups that are vibrationally coupled to the labeled group, as observed for unbound GTP in solution. As in the case of Ras (Cepus et al., 1998; Allin and Gerwert, 2001) uncoupling of GTP vibrations due to protein binding is also observed for Rap. The assignment of GTP-γ at 1143 cm−1 with a shoulder band of GTP-γ at 1157 cm−1, GTP-β at 1215 cm−1, GTP-α at 1260 cm−1 (Figure 3.2 aE), GDP-β band at 1100 cm−1 and 1136 cm−1 and GDP-α at 1236 cm−1 (Figure 3.2 aE) are in excellent agreement with previous assignments on Ras (Allin and Gerwert, 2001). The results show that the structural arrangement and charge distribution of GTP and GDP are very similar for Ras and Rap and the intrinsic hydrolysis reaction agrees well for both proteins.

52 Results

3.1.3 RapGAP catalyzed reaction.

While a single apparent rate constant can describe the intrinsic hydrolysis reaction, at least two (leaving out the cage reaction) are required to describe the GAP catalyzed hydrolysis reaction. This immediately indicates that the reaction involves an intermediate. The IR absorbance changes for the intrinsic reaction of Ras (Figure 3.3 a), GAP catalyzed reaction of Ras (Figure 3.3 b) and Rap (Figure 3.3 c) are shown below as a function of wave number and time (logarithmic). One can clearly see the single exponential decay of the GTP-γ band in case of the intrinsic reaction of Ras at 1143 cm−1 (Figure 3.3 a).

Figure 3.3. The absorbance changes in the infrared between 1800 and 950 cm−1 for the intrinsic reaction of Ras (a), GAP catalyzed reaction of Ras (b) and Rap (c) are shown as a function of wave number and time (logarithmic). The band at 1143 cm−1 is typical

GTP-γ vibration for intrinsic and GAP catalyzed hydrolysis of Ras. Pint: intermediate.

In case of the GAP catalyzed reaction of Ras, the same band appears after photolysis and decays (Figure 3.3 b). However, in the case of GAP catalyzed reaction of Rap (c), the time

53 Results evolution of the spectra is completely different for GAP catalyzed reactions of Ras (b). The typical GTP-γ band at 1143 cm−1, which has also been observed for intrinsic reaction of Rap, is missing in the case of RapGAP catalyzed reaction. In addition, intermediate like vibration −1 −1 (Pint) is observed at 1172 cm for Rap-RapGAP system instead of 1114 cm as seen for Ras-RasGAP system. This immediately indicates that the GAP catalyzed reactions of Ras and Rap are fundamentally different.

In case of GAP catalyzed reaction of Rap, marker frequencies of GTP-β at 1128 cm−1, an −1 −1 −1 intermediate at 1172 cm , GDP-β at 1104 cm and the released Pi at 1075 cm are highlighted for illustration. The detailed assignment is discussed in later sections.

In all the cases, the absorbance changes relax to zero since the final spectrum is used as reference. For the GAP catalyzed reaction of Rap, after photolysis (described by k1´ = 5.7 s−1; not shown), the GTPase reaction takes place in the two steps with apparent rate constants of 0.6 and 0.07 s−1 respectively (at 258 K) as shown in scheme 1 below:

−1 −1 k2´= 0.6 s k3´= 0.07 s A B C (RapGTP) (Intermediate) (RapGDP + P ) i (Scheme 1)

The apparent rate constants k2´ and k3´ (at 258 K) describe the appearance of an intermediate (AÆB) and the decay to GDP and Pi (BÆC) by the slowest rate, respectively. The spectral signatures for these two steps are discussed in the following sections.

3.1.4 A to B Transition.

The amplitude spectrum of k2´, representing the transition of GTP to the intermediate, is shown in Figure 3.4a (black lines). The band assignments in the k2´ amplitude spectra of RapGAP catalyzed reaction are in general more complicated than either the intrinsic reactions of Ras and Rap or the RasGAP catalyzed hydrolysis reaction of Ras because the GTP vibrations now become highly coupled. This shows that RapGAP activates phosphoryl transfer by a different mechanism than RasGAP, by inducing a different GTP conformation, denoted from now as GTP*. Delocalized vibrations are also observed for unbound GTP in solution. A different orientation of the three phosphate groups relative to each other can cause different coupling. It is not yet clear which relative orientations of GTP* cause such delocalization. Further studies are required to clarify this phenomenon. Nevertheless, the main contributing vibration to the normal mode can still be assigned as described in the following sections. In such situations, small changes and coupled vibrations have often been described in the text in terms of double difference bands for greater visibility.

As in the case of Ras (Allin et al., 2001), two intermediate bands are identified. They appear at 1170 cm−1 and 1085 cm−1 for Rap-RapGAP system. The band at 1170 cm−1 is shifted to −1 18 18 18 1158 cm due to γ- O label [βγ- O / γ- O3] (Figure 3.4 aA, red line), which is seen in the

54 Results double difference spectra (Figure 3.4 bA, red line) as a difference band at 1172/1159 cm−1. The second band of the intermediate shifts from 1085 cm−1 to 1078 cm−1 due to [βγ-18O / γ- 18 −1 O3] labeling. This is seen as a double difference band at 1090/1072 cm . These shifts −1 indicate that the intermediate bands at 1170 and 1085 cm are either GTP-γ or cleaved Pi. In order to distinguish between these two possibilities, the same reaction was performed in 18 H2 O with unlabeled GTP. In this case, a similar shift can only be observed if cleavage has already taken place (Feuerstein et al., 1989). As for the γ-label, the intermediate band at 1170 cm−1 (Figure 3.4 aB) indeed looses intensity and another band appears at 1155 cm−1. This is more clearly visible as a difference band at 1172/1156 cm−1 in the double difference spectrum (Figure 3.4 bB). Thus we conclude that cleavage of the β-γ bond has already taken

Figure 3.4. Spectral signatures of the intermediate: a) The amplitude spectra (black lines) of the

reaction described by k2´, the A to B transition (appearance of intermediate), where the negative bands correspond to GTP and positive bands correspond to intermediate. The spectra of labeled GTP are shown in colored lines corresponding 18 18 18 18 18 to A) [β, γ- O/γ O3], B) H2 O, C) Sp-[β- O], and D) Rp-[β- O]. b) The double difference between the amplitude spectra of labeled and unlabeled GTP in panel a, with the original and shifted frequency in black and in color, respectively. place in the intermediate B. A shift of the second intermediate band is not resolved in case of 18 H2 O measurements, most likely because its shifts are not above the noise level.

The γ-18O labeling not only affects the γ vibration, but also several other phosphate bands, such as GTP-α vibrations, due to coupling (Figure 3.4 aA and 3.4 bA). This is evident from the fact that upon γ-18O labeling, the GTP-α bands are also affected as seen from the

55 Results downshift in the double difference spectra at 1262,1256 / 1240,1231 cm−1 (Figure 3.4 bA). 18 −1 The same shift is seen in the H2 O measurement at 1256/1240 cm in the double difference spectra (Figure 3.4 bB) with lower intensity. The effect of the γ-18O label on the GTP-α band at 1262,1256 / 1240,1231 cm−1 is attributed to the long-range coupled α-γ vibration in GTP.

In the case of GAP catalyzed reaction of Rap, the classical GTP-γ bands at 1144 cm−1 and 1156 cm−1 are observed only at early time points, such as in the photolysis spectra (Figure

3.5 aB and bB) but they are not observed as negative bands in the k2´ amplitude spectrum (Figure 3.4 aA). GTP vibrations are uncoupled in the photolysis and become highly coupled in the GTP state that is reflected by the negative bands in the k2´ hydrolysis amplitude spectrum (Figure 3.4 aA). Clearly the GTP bands, which appear at photolysis at 1156 cm−1 and 1144 cm−1 with uncoupled vibrations (Figure 3.5 aB and bB), are rapidly converted to another state which is seen as negative GTP bands in A to B transition (Figure 3.4 aA). Thus, the conformation of GTP seen in A to B transition is unique where phosphate vibrations are highly coupled. This state is denoted as GTP*. Because the complete photolysis of the caged compound takes about 100 ms, the reaction from GTP to GTP* is not time-resolved and must be faster than 100ms. However, comparison of the photolysis and hydrolysis amplitude spectra resolves the two different GTP conformations. This rapid conformational change also explains why the classical GTP-γ band (1143 cm−1) was not observed in time dependent spectral evolution (Figure 3.3 c).

Figure 3.5. Uncoupled GTP vibrations for a) spectra of Intrinsic reaction (aA) and amplitude spectra of first rate of GAP-catalyzed reaction (aB), inverted for the ease of comparison). b) Respective double differences. At the early time points such as photolysis (ca. 100 ms) vibrations of the GTP bands are still uncoupled in case of RapGAP catalyzed reaction as seen in bB.

18 The two stereo-specific β labels, the (Sp)-[β- O] (Figure 3.4 aC and 3.4 bC) and the (Rp)-[β- 18 18 O] (Figure 3.4 aD and 3.4 bD) show nearly the same shifts. The (Sp)-[β- O] label shifts the

56 Results

GTP-β bands 1210 cm−1 to 1202 cm−1 and from 1128 cm−1 to 1118 cm−1. This is observed in the double difference spectra as difference bands at 1216/1200 cm−1 and 1130/1119 cm−1 18 −1 (Figure 3.4 bC). Similarly, the (Rp)-[β- O] label shifts the GTP-β bands 1210 cm to 1206 cm−1 and from 1128 cm−1 to 1122 cm−1. This is observed in the double difference spectra as difference bands at 1216/1200 cm−1 and 1130/1119 cm−1 (Figure 3.4 bD). Thus the bands at 1210 cm−1 and 1128 cm−1 (Figure 3.4 aA) can be attributed to primarily a GTP-β vibration.

18 The double difference bands for (Sp)-[β- O] have higher intensity than the double difference 18 bands for (Rp)-[β- O] which indicates a greater dipole moment towards the (Sp)-β-oxygen of GTP. This might be due to the presence of additional positively charged groups in the vicinity of the (Sp)-β-oxygen. The difference between (Rp) and (Sp) label is much smaller as compared to Ras (Allin et al., 2001), and is most likely due to scrambling of the stereoselective positional isotope label due to rotation of GDP-β after cleavage and before reformation of GTP. This is explained later (see section 4.1.6; Figure 3.8).

Interestingly, the β-18O labels also shift the intermediate band, which is seen as in the double difference as a difference band at 1172/1154 cm−1 (Figure 3.4 bC and bD), just as the γ-18O label does vice versa for the β vibration at 1130 cm−1. The later is seen in the double difference spectra as a difference band at 1132/1120 cm−1 (Figure 3.4 bA). In addition, the 18 18 (Sp)-[β- O] and (Rp)-[β- O] labels shift the second intermediate band which is seen as a double difference band at 1100/1074 cm−1 (Figure 3.4 bC and bD). It seems that this second intermediate band absorbs at 1100 cm−1. In case of γ labeling, due to an additional positive band it seems to be shifted from its original position at 1100 cm−1 down to 1090 cm−1 (Figure 3.4 b: A, C and D respectively). In the third rate the intermediate band is seen at 1090 cm−1 for the γ labeling (Figure 3.9 bA).

The α-18O label also affects both intermediate vibrations (Figure 3.6) due to coupling. As seen in the amplitude spectra (Figure 3.6 aA), labeling of α-GTP shifts the bands at 1271 and 1260 cm−1 to 1250 and 1238 cm−1 respectively. This is observed as a double difference band at 1271/1242 cm−1 (Figure 3.6 bA). Since the same effect of band shift has been observed for four fold γ-18O label at 1271 cm−1 and 1260 cm−1 (Figure 3.4 aA) and a band at 1260 cm−1 corresponds to GTP-α vibration for intrinsic reaction of Ras (Allin and Gerwert, 2001) and Rap (Figure 3.2 aD) we conclude that the bands at 1271 cm−1 and 1260 cm−1 represents GTP-α vibration for RapGAP catalyzed reaction. In addition to the effect on GTP- α and the intermediate, the α-18O label also affects β-GTP bands seen as double difference bands at 1207/1189 cm−1, 1136/1122 cm−1 (Figure 3.6 bA). Nevertheless, similar effect was 18 18 also observed for γ- O4 and β- O labels previously (Figure 3.4 bA and bC). Since the same labels showed localized vibrations for the intrinsic reaction of Rap (Figure 3.2) and in the photolysis spectra of RapGAP catalyzed reaction (Figure 3.5 bB), we conclude that the α, β and γ vibrations are indeed coupled for the GTP* state.

57 Results

18 Figure 3.6. Effect of α- O labeling. aA) Amplitude spectra of k2´ for RapGAP catalyzed reaction and bA) double difference. Labeling of α-GTP affects not only the expected band at 1271/1242 cm−1, but also β-GTP bands at 1207/1189 cm−1, 1136/1122 cm−1 and intermediate bands at 1171/1156 cm−1, 1092/1072 cm−1. Similarly, aB) Amplitude

spectra of k3´ and bB) corresponding double difference (discussed in greater detail in 3.1.6). Shift of α-GDP is identified at 1235/1194 cm−1. All vibrations in the

amplitude spectra of k2´ and k3´ are coupled.

There are two possible arguments for such extensive coupling. One might argue that since the α and β labels shift the intermediate bands, the intermediate aught to represent certain vibrations of GTP (or GTP*). In other words, vibrations of γ-phosphate that is still covalently bound to the GDP counterpart. However this argument does not stand since the intermediate −1 18 band at 1172 cm is also shifted by H2 O. Since breakage of the β-γ bond of GTP is 18 18 required for incorporation of O label from H2 O, this indicates that nucleophilic attack of labeled water has already taken place leading to chemical cleavage. At the absence of covalent linkage, such long-range coupling can be induced by dipolar interaction between non-covalently bound states (Wang and Hochstrasser, 2004). For this to occur, Pi has to be less than a few Å from β-GDP. Thus, the intermediate state represents a protein-bound

GDP•Pi as in Ras•RasGAP, but with an altered geometry.

3.1.5 B to C Transition.

Figure 3.7a shows the amplitude spectrum for k3´, which describes product formation and Pi release. The former positive bands of the intermediate (e.g. 1170 cm−1, black line) appear as negative bands as it disappears with k3´ while the product peaks of the GDP and Pi appear as 18 18 positive bands. The four-fold γ label [βγ- O / γ- O3] shifts the intermediate band from 1170 cm−1 to 1157 cm−1 (Figure 3.7 aA, red line), which is seen in the double difference as a difference band at 1170/1155 cm−1 (Figure 3.7 bA, red line). The GDP band is shifted from

58 Results

1100 cm−1 to 1091 cm−1, which appears as a double difference band at 1111/1090 cm−1 in the double difference spectrum (Figure 3.7 bA, red line). In addition, due to four-fold γ 18 18 −1 −1 −1 labeling [βγ- O / γ- O3] the bands at 1075 cm and 990 cm are shifted to 1048 cm and 962 cm−1 (Figure 3.7 aA, red line). They are identified in the double difference as difference bands at 1075/1050 cm−1 and 992/962 cm−1 (Figure 3.7 bA). Such shifts are also seen for the intrinsic reaction for Rap (Figure 3.2: aA & bA) and Ras (Allin and Gerwert, 2001) and they represent released phosphate. The most notable feature of the B to C transition is the 18 18 similarity between the double difference spectra obtained with [βγ- O / γ- O3] GTP and 18 H2 O. This is explained in the next section (see 4.1.6).

18 18 Furthermore, the (Sp)-[β- O] and (Rp)-[β- O] labels show exactly the same shifts and intensities. The double difference spectra (Figure 3.7 bC and 3.7 bD) show the expected −1 effect on the ‘Pi’ intermediate band at 1170/1155 cm and on the protein-bound GDP as observed in the difference bands at 1137/1116 cm−1 and 1107/1090 cm−1 (Figure 3.7 bC and bD). These band positions are in good agreement with the band assignment of GDP-β as seen for intrinsic reactions of Ras (Allin and Gerwert, 2001), Rap (Figure 3.2) and for the GAP catalyzed reaction of Ras (Allin et al., 2001). The β-GDP vibrations are not affected by RapGAP as for RasGAP. Moreover, the GDP vibrations are not coupled anymore.

Figure 3.7. Spectral signature of product release. a) The amplitude spectra of k3´, which

describe B to C transition (the Pi release step) as explained in Figure 3.4 legend. b) The double difference between the amplitude spectra of labeled and unlabeled GDP in panel a, with the original and shifted frequency in black and color respectively.

59 Results

3.1.6 Reversibility of the GTPase reaction and 18O exchange.

18 * Using H2 O, we observe a large amount of label incorporation into GTP in the FTIR spectra (Figures 3.4 bB and 3.7 bB), which is considerably more pronounced than what has been reported for the Ras•RasGAP system by Allin et al. (Allin et al., 2001).

18 18 18 The unmistakable similarity between the four fold γ label [βγ- O / γ- O3] and H2 O (Figure 3.7 bA and bB) evoked the possibility that there might be a reverse reaction, i.e. synthesis of * GTP from GDP, Pi and H2O. Such backward reaction has long been known for ATPases such as myosin(Levy and Koshland, 1959). This was rather unexpected for GAP catalyzed hydrolysis reaction of a GNBP since the free energy of GTP hydrolysis in such system is much more favorable than ATP hydrolysis of myosin. This is the energetic aspect that makes ‘GTPases’ effective signaling rather than energy transducing molecules (Goody, 2003).

The classical approach to demonstrate back reaction is quenched flow technique (see 2.3.13) followed by mass spectrometric analysis. In the quenched flow procedure, the reaction is stopped (or “quenched”) at different time points by rapid addition of a quencher. Typically an acid like Trifluroacetic acid (TFA) or a base is used as biological quencher, which denatures the protein but does not affect the substrate significantly. The quenched reaction at different time points can thereafter be analyzed by a suitable analytical method like radioactivity, fluorescence or mass spectrometry. While radioactivity or fluorescence are quite convenient ways for mechanistic investigations, incorporation of multiple 18O label in released phosphate or unreacted GTP can only be demonstrated by mass spectrometry or 31P NMR spectroscopy (Cohn and Hu, 1978).

The RapGAP catalyzed hydrolysis reaction of Rap•GTP was investigated by using these methods. On one hand, this system was more interesting than Ras•RasGAP (Allin et al., 18 2001) since the H2 O effect was much more prominent, as seen in the FTIR data (Figure 3.7 bA and bB), one the other hand it is significantly more difficult to achieve single turnover conditions due to low affinity of RapGAP towards Rap•GTP (Brinkmann et al., 2002; Kraemer et al., 2002). Since samples obtained from quenched flow often require a pre-purification, 50 µM Rap•GTP for sensitivity reasons and at least 300 µM RapGAP for near single turnover condition. The amount of GAP required in this case, is significantly higher than what would be required for Ras•RasGAP system. Consequently, to quench such a reaction, higher amount of quencher would be required. Presumably due to the presence of quencher substrate (TFA) and high salt, sufficient signal to noise was not be achieved for either released phosphate or remaining GTP.

Methanol was chosen as an alternative quencher since even 50% methanol can denature several proteins efficiently but it can also be removed easily due to volatility. As a test, the GAP catalyzed hydrolysis reactions of Ras, Rap and Ran were performed with their 18 respective GAPs under multiple turnover conditions in H2 O. Interestingly, mass spectrometric analysis clearly indicated multiple 18O incorporation for Ran (data not shown)

60 Results and Ras (Carsten Kötting, unpublished results) but only single incorporation for Rap. Single incorporation of 18O in released phosphate would also result from the nucleophilic attack of 18 H2 O at the γ phosphate of GTP even if the reaction were irreversible. However, the experiments on Ras and Ran clearly indicated that for Rap-RapGAP system, it might be necessary to perform the reaction at very high concentration where a complex between

Rap•GTP•RapGAP and alternatively Rap•GDP•Pi•GAP might form.

Therefore, an FTIR sample (see materials and methods, section 3.5.3) was analyzed by 18 mass spectrometry after performing the reaction in H2 O. As seen in Figure 3.8 a, in addition 16 18 to O4-Pi (m/z 97), one (m/z 99), two (m/z 101), three (m/z 103) and even four (m/z 105) O atoms are incorporated into Pi. Qualitatively we can conclude, that after cleavage of protein- 18 18 bound GTP by nucleophilic attack of H2 O, bound ‘Pi’ bearing one O can reform GTP. 18 16 16 16 18 Since O and O are chemically indistinguishable, one O from O3 - O-Pi can now be transferred to water with a certain statistical probability to produce GTP containing 18O-GTP. 18 Repetition of this process leads to progressive accumulation of O into released Pi as shown in the scheme of Figure 3.8.

Figure 3.8. Reversibility of the cleavage reaction. The scheme for oxygen exchange shows how, because of reversibility of the reaction, multiple 18O atoms are incorporated

into Pi as outlined in the text. The inset (a) shows mass spectrometric peaks of the reaction product in the presence of 18O water where the 97 m/z peaks corresponds

to unlabeled Pi, whereas the peaks at 99, 101, 103 and 105 correspond to Pi containing 1 to 4 18O atoms, respectively.

In the process of FTIR sample preparation, when the protein solutions are concentrated under nitrogen stream (see materials and methods, section 2.5.3), fast exchange of water between the sample and atmospheric water takes place. Thus, the 18O label of the sample is always contaminated with 16O from atmospheric water. Therefore, the mass spectrometric 18 data cannot be used for absolute quantitative analysis of O incorporation in Pi.

61 Results

Qualititatively, the amount of double or multiple 18O incorporation is even higher than single 18O incorporation, which implies that there is significant rotational freedom of the already cleaved phosphate that resides in the Rap•GDP•Pi•RapGAP complex.

Contemporary to this work, reversibility of GTPase reaction has also been shown by quenched flow and mass spectrometric analysis for the Ras-NF1 reaction where a ratio of 88 between the forward and backward chemical step has been found (Phillips et al., 2003).

This is a further indication that in Rap•GDP•Pi•GAP complex, Pi might be closer to GDP than in Ras•GDP•Pi•RasGAP, which would facilitate the reverse reaction. If the release of Pi from

Rap•GDP•Pi•GAP complex is sufficiently slow, the cleaved γ-phosphate and GDP would remain bound to the proteins to allow an 18O-water mediated phosphate-water oxygen exchange to take place. Indeed the phosphate release is the rate-determining step for this reaction, which is described later (see 3.1.9). In spite of reversibility, the reaction does not reach a steady state since the overall equilibrium for GTP hydrolysis, due to the large free energy of hydrolysis (∆G) of GTP, is almost exclusively on the product side.

3.1.7 Summary of band assignments

The important spectral assignments for Rap•RapGAP system are summarized in Figure 3.9. The intrinsic reactions of Ras and Rap are very similar. Again, in case of GAP catalyzed reaction, after hydrolysis, the GDP and Pi bands appear at the same frequencies for both Ras and Rap. The differences are observed when the respective GAP binds to the GNBP•GTP and when the chemical cleavage takes place.

Figure 3.9. Summary of the band assignments of GTP for the intrinsic and GAP catalyzed hydrolysis reaction of Rap, in comparison to Ras, as indicated. The GDP vibrations agree in all the cases between Ras and Rap and are not shown. Downshift of the β vibration due to RasGAP and RapGAP binding is the largest effect on GTP vibration.

62 Results

In case of RasGAP catalyzed reaction, binding of the GAP resulted in a large downshift of the GTP-β vibration from 1219 cm−1 to 1140 cm−1 (Summarized in Figure 3.9). This essentially appeared as a second GTP-β vibration (Allin et al., 2001). Similar situation is also observed for RapGAP. Binding of RapGAP also induces a large downshift of a second β- vibration from 1215 cm−1 to 1128 cm−1 (Figure 3.4 aA, summarized in Figure 3.9). The frequency downshift of the second β-vibration is even larger for Rap•RapGAP system than what has been observed for Ras•RasGAP (1128 cm−1 vs. 1140 cm−1; Figure 3.9). This is the single most important difference between the GAP catalyzed reactions of Ras and Rap. Interestingly, the GTP-α band is upshifted for Rap•RapGAP. Although the mechanistic details of such an upshift of GTP-α are not clearly understood, similar combined effect on GTP α and β is also observed for GAP catalyzed reaction of another GNBP, Ran where conversely GTP-α is downshifted (1248 cm−1) and GTP-β is upshifted (1225 cm−1) (Brucker, Kötting and Gerwert, unpublished results). Unfortunately, any intermediate state is not resolved yet for Ran•RanGAP system. Therefore no detailed comparison could be made.

3.1.8 Kinetic scheme.

As described in the earlier section, for the GAP catalyzed reaction of Rap, typical GTP bands are observed at early time points, such as in the photolysis spectra. However, this is rapidly converted to a unique state that we designated as GTP*, where all the GTP vibrations are highly coupled. The chemical cleavage step of GTP to GDP and Pi is reversible and Pi release from the Rap•GDP•Pi•GAP complex is rate limiting. Based on this, the following kinetic scheme is deduced:

GAP k k * 2 * 3 RapGTP RapGTP • GAP RapGDP • Pi • GAP RapGDP + Pi + GAP k k faster than 100 ms −2 −3 (Scheme 2)

In this scheme k2, k−2, k3, and k−3 are microscopic rate constants. The nature of the intermediate in the Rap-RapGAP reaction is similar as the GAP catalyzed reaction of Ras

However, the dipolar coupling between Pi and GDP is not observed in RasGAP catalyzed reaction. Therefore Pi seems to be a shorter distance away from GDP in the Rap-RapGAP than in the Ras-RasGAP intermediate complex.

3.1.9 Time Course of The Reaction.

Based on the band assignments (summarized in Figure 3.9), the kinetics of selected groups are shown in Figure 3.10. The global fit procedure adopted in this work provides apparent rate constants (Scheme 1). This allows us to analyze the kinetics in a minimal, model independent way. The apparent rate constants of scheme 1 describe the two main reaction

63 Results steps of the GTPase reaction. They describe direct elementary reactions only if reverse reactions can be neglected. However, as shown in scheme 2, four microscopic rate constants are needed to describe the elementary steps, but only two apparent rate constants are experimentally accessible. Since the apparent rate constants differ by an order of magnitude (scheme1) they describe in a first approximation the elementary reactions. Modeling of the experimentally underdetermined reaction with estimated microscopic rate constants that are able to describe the absorbance changes, show that the back reaction leads to prolongation of the GTP decay. As an illustrative example, the disappearance of the GTP-α band at 1271 cm−1 has been chosen as marker band to monitor the disappearance of GTP since the usual marker band for GTP, the GTP-γ at 1155 cm−1 is masked by the appearing intermediate band. Since the α vibration is coupled to the γ vibration, its absorbance change is the same as for the γ vibration. Figure 3.10 shows that the marker

GTP band primarily disappears with the second rate (k2´), but k3´ prolongs the decay.

Figure 3.10. Time dependent amplitude changes at different individual IR frequencies corresponding to different bond vibrations. Shown are the data (crossed points), the fitted curves (solid curve) and the contributions from the apparent rate constants

(kl´) of the multi-exponential fit (dotted colored lines). The time axis is logarithmic. The GTP*-α at 1271 cm−1 decays primarily with second rate whereas the protein bound intermediate at 1172 cm−1 appears with the second rate and disappears with −1 the third rate. The release of the protein bound Pi to the bulk medium at 1075 cm

described by k3´ is the rate limiting step. The interaction of a carboxyamide side chain of an amino acid residue absorbing at 1703 cm−1 is also well resolved.

−1 The intermediate indicated at 1172 cm appears with the second rate (k2´) and disappears with the third rate (k3´). This shows that the cleavage reaction takes place with k2´ and the Pi

64 Results

−1 release with k3´. In agreement, the Pi release, indicated at 1075 cm appears almost solely with the third rate (k3´) as the slowest and thus rate-limiting reaction.

Interestingly, above 1350 cm−1, where different protein side groups and the protein backbone absorb, several bands appear with k2´ and disappear with k3´ (see discussion). Such kinetics is observed for example at 1703 cm−1 (Figure 3.10), which represents a carbonyl vibration, tentatively assigned to a carboxyamide (H2N-C=O) side chain of a glutamine or an asparagine residue. Interestingly, H2N-C=O of D96N mutant in bacteriorhodpsin absorbs also at 1704 cm−1 (Gerwert et al., 1989). Similar intermediate-like kinetic behavior is also observed at 1408 cm−1 (see discussion). For clear-cut assignment of the respective side chain groups, specific non-invasive mutants or site directed isototopic labels have to be used. The results presented here show that certain residues of the protein undergo a concomitant conformational change in the intermediate accompanying the cleavage of γ- phosphate.

3.1.10 Conclusion

While the key residues, Gln61 of Ras and the arginine finger of RasGAP are absent in the Rap-RapGAP system, the FTIR difference spectra show certain similar, but also clearly distinct features. On one hand, the intrinsic reactions of Ras and Rap are very similar within the phosphate region (Figure 3.1 b) though Rap has a threonine at position 61. The GAP catalyzed reactions on the other hand, are very different (Figure 3.3 b, c). In the RapGAP catalyzed case, the typical GTP vibrations are observed only at the early time points, such as in the photolysis spectra (Figure 3.5 bB). Upon GAP binding, rapidly a new GTP conformation (designated as GTP* in the text) evolves where all the vibrations of α, β and γ phosphates of GTP are coupled. Such coupled vibration has not are been observed for RasGAP catalyzed reaction of Ras. Also, it was clearly shown that the most important feature in GAP catalyzed reaction at least for Ras and Rap is downshift negative charge towards the β phosphate of GTP (Figure 3.9). In case of a nucleophilic reaction where water is the nucleophile and GDP is the leaving group, such charge movement towards non- bridging β-oxygen of GTP would be expected to accompany weakening of the β-γ bond. Therefore, the Rap•RapGAP reaction might have a transition state that is more dissociative in nature compared to Ras•RasGAP reaction.

This work also addressed two important questions concerning mechanism of phosphoryl transfer. Firstly, the FTIR data and the mass spectrometric experiments conclusively proved the reversibility of GTPase reaction that was merely suggested as a possibility in earlier work (Allin et al., 2001). The reason why this classical effect that is long known for the ATPase reaction (Levy and Koshland, 1959), has been overlooked for a long time for GTPases reaction is primarily thermodynamic (Goody, 2003). Secondly, this work also resolves the

65 Results controversy about the nature of the intermediate that is resolved in GAP catalyzed reaction. Earlier work on Ras•RasGAP indicated two possibilities to account for such an intermediate: (i) an already cleaved phosphate (Kraemer et al., 2002) or (ii) in addition to the earlier possibility, a transiently phosphorylated intermediate (Allin et al., 2001). The second possibility was solely drawn from analogy since the two intermediate bands at 1191 cm−1 and 1116 cm−1 (see Figure 3.9) agree nicely with phosphorylated aspartyl bands of Ca2+ ATPase (Barth, 1999). While appearance of two intermediate bands at very different frequencies (1172 cm−1 and 1100 cm−1) in case of Rap•RapGAP reaction does not reinforce the possibility of phosphorylated intermediate, demonstration of reversibility clearly speaks in the favor of protein bound GDP•Pi as the intermediate.

The absorbance change at 1703 cm−1 (Figure 3.10) turned out to be particularly interesting. Such a clear kinetics of a carboxyamide functionality that is concomitant to the phosphate intermediate was indicative for a functional role of this residue. Since mutation of conserved glutamines of RapGAP did not exert a dramatic effect that would be expected for a catalytic residue (Brinkmann et al., 2002), the next possibility was an asparagine. Search for another conserved carboxyamide side chain indicated the possible involvement of asparagine 290 situated in the surface exposed most conserved helix α-7 as shown in purple in Figure 3.11 (and Figure 1.15).

Figure 3.11. Sequence alignment of RapGAP family members. Sequences shown are Homo sapiens Rap1GAP (SW: P47736), Spa1 (SW: O60618), E6TP1 (SW: Q9UNU4) and Tuberin (SW: P49815), Drosophila melanogaster RapGAP (SW: O44090), Caenorhabditis elegans RapGAP (SW: P91315) and Spa1 (SW: Q20016). Helix α-7 shown in purple, contains the most conserved residues including the catalytic asparagine-290.

66 Results

This residue, in an independent study, indeed turned out to be the catalytic residue (Daumke et al., 2004). It is interesting that an Asn residue, Asn233 in Dictystelium discoideum myosin II is found in the structure of transition state analog structure of myosin, where it contacts the β-phosphate and Vanadate (Fisher et al., 1995; Smith and Rayment, 1996; Rayment, 1996). However the position of this asparagine is not conserved in other transition state structures of myosin, (Dominguez et al., 1998), and mutational analysis indicates that this Asn is involved in binding rather than hydrolysis of ATP (Sasaki and Sutoh, 1998). This suggest a different role for this residue in myosin where one or two serine residues are believed to be involved in catalysis (Fisher et al., 1995).

67 Results

3.2 Active site analysis of Rap•RapGAP system

Members of the Rap family do not possess a catalytic glutamine, which is essential for intrinsic and GAP mediated catalysis in nearly all other GNPBs examined so far. For Rap GNBPs, it had been suggested that position 61 is not essential for the intrinsic and GAP stimulated GTP hydrolysis (Maruta et al., 1991). Again, Rap1GAP, the GTPase activating protein for Rap1, has unique features: it can downregulate the G12V mutation of Rap (Brinkmann et al., 2002), which is not possible for Ras-RasGAP. Moreover, by mutational analysis it has been shown that Rap1GAP does not have an arginine finger, rather employs an ‘asparagine thumb’, which is a part of a conserved helix in Rap1GAP family ((Daumke et al., 2004), see also Figure 1.15). In this section the Rap1GAP catalyzed hydrolysis reaction was further characterized by studying the effects of mutations at position 12 and 61 on the hydrolysis rate and transition state formation. Furthermore, the active site of Rap•RapGAP system was further analyzed by site directed mutagenesis of the catalytic asparagine.

3.2.1 Effect of T61 and G12 replacement

To clarify the role of threonine 61 in Rap, this residue was mutated to alanine, leucine (oncogenic mutations for Ras) and glutamine (Ras like mutation). Furthermore glycine 12 was also mutated to valine (oncogenic for Ras) and the Rap1GAP stimulated reaction of these mutants was studied by a fluorescence reporter assay (Kraemer et al., 2002) (See 2.3.13.1). Typically, 2 µM of Aedans-Rap•GTP was mixed with 50 µM Rap1GAP at 10 °C using the stopped flow apparatus and time dependent fluorescence change was monitored (black dots in Figure 3.12). This assay gives readout of protein-protein interaction. The fluorescence increase in the first part of the transient (Figure 3.12, black dots) is due to complex formation between Aedans-Rap•GTP and Rap1GAP and the subsequent decrease is due to hydrolysis of GTP or dissociation of the proteins.

Such a fluorescence transient (fluorescence traces) can be numerically analyzed with a suitable model as shown is scheme 3 (see 2.3.13.3 and appendix 6.3). Here A and B are Aedans-Rap•GTP and Rap1GAP, C and D are the intermediates and E is Aedans-Rap•GDP.

k1 k3 k4 A + B C(+DE B ) k2 (Scheme 3)

Apart from the rate constants k1, k2, k3 and k4, three other parameters Ya, Yc and Ye are used (see appendix 6.3), which describe fluorescence coefficients of A, intermediates C and D, and product E respectively. Multiple transients can also be analyzed globally to obtain the rate constants (Kraemer et al., 2002).

68 Results

This convenient procedure provides the rate constants, which are otherwise difficult to measure due to low affinity of this system. Typical values of the rate constants obtained from such an analysis are listed in table 2 below.

Figure 3.12. Analysis of Fluorescence transient of Aedans-Rap•GTP and Rap1GAP. 1µM Aedans-Rap•GTP was mixed with 25 µM Rap1GAP at 10 °C. Black dots: measured data points; orange curve: scientist fit with scheme 3; green curve: decay of Aedans-Rap•GTP, blue and red curves: intermediates C and D; magenta: concentration of Aedans-Rap•GDP.

Table 2 Value of rate constants and fluorescence coefficients as obtained from transient kinetic analysis between Aedans-Rap•GTP and Rap1GAP at 10 ° C. For comparison, values reported in literature are cited in the third column.

Literature value (Kraemer et al., Constant Value 2002)

−1 −1 0,64 0,62 k1 (µM s )

−1 14,0 8,69 k2 (s )

−1 k3 (s ) 7,74 7,97

−1 k4 (s ) 3,13 1,70

Ya 0,983 0,99

Yc 1,584 1,69

Ye 0,983 0,99

In general, the fitted values agree to the literature (Kraemer et al., 2002). The slight deviation in k2 and k4 might also be an outcome of the fact that the Rap1GAP purification protocol has improved over the time.

69 Results

However, the fluorescence intensities of the G12V, T61A, T61L and T61Q mutants are significantly smaller than wild type Rap1, as seen in figure 3.13 A. Some fluorescence increase is still observed with T61A and to even smaller extent with G12V and T61Q but no signal is observed for the T61L mutant. After normalizing the maximal fluorescence to the same value (Figure 3.13 B), a kinetic trend is seen where the reaction of Rap1-T61Q with Rap1GAP is slower than wild type and Rap1-G12V is further slower, compared to Rap1- T61Q. While the extent of fluorescence labeling is nearly the same for all these mutants (see appendix 6.2), even the nature of the curves are different which does not allow the same kinetic analysis to be done. This leads to two possibilities, either the association process of Rap1GAP with these mutants is severely reduced, or the reporter assay employed is not suitable at all.

Figure 3.13. (A) Fluorescence reporter assay of Rap1 mutants. In each case, wild type Rap1GAP has been used. The initial values of relative fluorescence are normalized to unity. The time in the X-axis is logarithmic. (B) The same experiment as in (A) but plotted against linear a time axis, with normalization of the maximum fluorescence. R1G: Rap1GAP wild type.

To crosscheck whether the fluorescence reporter assay was suitable for analyzing these mutants, standard HPLC method (see 2.3.11) was used to analyze the effect of these mutations on GTP hydrolysis. This method is suitable for monitoring the hydrolysis reaction in multiple turnover and yields discrete datapoints. Nevertheless, such data is free from artifacts since it does not depend on fluorescence change.

3.2.2 Comparison of hydrolysis rates from HPLC measurement

In this method, 100 µM of wild type Rap1•GTP or one of the mutants were converted by using catalytic amount (50 nM in this case) of wild type Rap1GAP and the hydrolysis reaction at 20 °C was monitored at different time points. For the intrinsic reaction, an exponential fit was used. For the GAP catalyzed reaction, data points were taken up to ca. 50% hydrolysis. Within this range, these points are nearly linear as shown below (Figure 3.14 B).

70 Results

The results of the intrinsic and GAP catalyzed reaction are summarized in Table 3. The intrinsic reaction of Rap1B is about 10 times slower than Ras, while the Rap-T61Q mutant has almost similar intrinsic rate (Frech et al., 1994) as that of Ras. Surprisingly, the intrinsic hydrolysis rate of Rap-T61L is even slightly faster than that of wild type. The Rap-T61A mutant is somewhat slower but not as dramatic as would be expected for a mutation of the 61 position of any other GNBP. Taken together, this demonstrates that the position 61 has a less significant role in case of the intrinsic hydrolysis reaction of Rap.

The GAP stimulated reactions (Figure 3.14 B) show interesting features. The G12V mutation is slower, which is in line with the published results (Brinkmann et al., 2002). The T61A mutant is almost as fast as the wild type, which indicates that the hydroxyl group of threonine plays only a minor role in the GAP stimulated hydrolysis. However, for the T61L mutant of Rap, GAP stimulation is drastically reduced.

Figure 3.14. Intrinsic (A) and GAP catalyzed (B) hydrolysis reaction of Rap1 wild type and mutants as monitored by HPLC.

Table 3 Intrinsic and Rap1GAP catalyzed reaction of Rap1wild type and mutants at 20 ° C.

Intrinsic rate Ratio GAP catalyzed Ratio Fold Construct −1 −1 (min ) k(mut)/k(wt) rate (min ) k(mut)/k(wt) activation

4 Rap1B C´(wt) 0,00035 1 7,16 1 2,05 x 10

4 G12V 0,00015 0,4 4,032 0,6 2,68 x 10

T61A 0,00020 0,6 8,766 1,2 4,4 x 104

T61Q 0,0018 5 5,868 0,8 3,3 x 103

T61L 0,00052 1,5 0,108 0,015 28,8

71 Results

The rather unexpected outcome is the reduced GAP stimulation of the T61Q mutant, since the intrinsic reaction is five fold faster than Rap wild type (Table 3). This indicates a possible sterric interaction of two carboxyamide residues in the active site, one from the glutamine of the GNBP, the other being the catalytic asparagine of the GAP. Similar explanation also holds for the bigger side chain of leucine 61 in case of the T61L mutant. This makes it interesting to judge the reciprocal effect, i.e. the effect of replacement of asparagine-290 of Rap1GAP, which is described later (see 3.2.3).

It is noteworthy that with the exception of Rap1-T61A, the stimulation of GTPase reaction shows the same tendency with the decay of normalized relative fluorescence as observed in Figure 3.13 B. The GAP activation of T61Q is slower than wild type and G12V is further slower. Also the T61L mutation leads to a more severe decrease of GTPase activation and consequently, in the fluorescence based assay it shows nearly a baseline. Therefore, this assay indeed shows association process that is exploited in evaluating the effect of mutations of catalytic asparagine 290 of Rap1GAP and for ground and transition state experiments with both Rap and Rap1GAP mutants as described in the subsequent sections.

3.2.3 Replacement of catalytic aspargine 290 of Rap1GAP

To investigate the role of the catalytic asparagine-290 in detail, this residue was mutated to glutamine, which bears the same functionality in the side chain with difference in length; to arginine and lysine, two positively charged residue which might have an effect in stabilization of the transition state of the reaction and thereby enhance the rate and to aspartate, a negatively charged residue (in collaboration with Oliver Daumke). As shown in Figure 3.15, the N290D mutant turned out to be binding deficient, which might be due to repulsion of

Figure 3.15. Stopped flow kinetics showing the effect of replacement of the catalytic asparagines-290 to arginine (red), lysine (green), glutamine (blue) and aspartate (cyan). In each case, Rap1b wild type has been used. The relative fluorescences are normalized to unity. The time in X-axis is logarithmic. R1Gwt: Rap1GAP wild type.

72 Results negative charges of the nucleotide and the side chain of the residue 290. Interestingly, the two positive charges, N290R and N290K respectively, bind, but do not stimulate hydrolysis. Finally, the mutant N290Q binds and hydrolyzes, albeit very slowly. This clearly indicates that the carboxyamide side chain is essential for this system. While both asparagine and glutamine bear the same carboxyamide functionality, the difference lies in the length of the side chain.

The Rap-T61Q mutant was also tested against the different mutations of Rap1GAP. As shown below, the Rap1GAP-N290D turned out to be binding deficient (Figure 3.16, black), as observed with wild type Rap1 (Figure 3.15, cyan). Again, the Rap1GAP-N290R mutant binds Rap-T61Q but does not stimulate GTP hydrolysis. Therefore, a Ras-like glutamine- arginine combination also failed to rescue the activity. This experiment shows the uniqueness of asparagine as a catalytic residue for this system.

Interestingly, the combination of Rap1GAP-N290Q and Rap1-T61Q is again binding deficient (Figure 3.16, blue curve) as seen for Rap1-T61Q wild type Rap1GAP (Figure 3.13, red curve). This evokes a possibility that there might be not enough space for two carboxyamide residues in the active site to allow association.

Figure 3.16. Stopped flow kinetics showing the effect of replacement of the catalytic asparagines-290 to arginine (red), glutamine (blue) and aspartate (black) in combination with Rap1-T61Q mutation. The trace in cyan is for wild type Rap and RapGAP for comparison. In all other cases, Rap-T61Q is used. R1G: Rap1GAP.

3.2.4 Cross activation

The previous experiments clearly showed the necessity of a carboxyamide side chain for GAP stimulate GTP hydrolysis. The question arises: what is the role of an asparagine residue in catalysis? The role of a positively charged residue like arginine (Scheffzek et al., 1997) is attributed to stabilization of accumulated negative charge in the transition state, which is in line with proposed associative mechanism. Asparagine is not the best suited

73 Results residue for this purpose. The second possibility can be drawn from the role of the carboxyamide residue in Ras (Q61) that was attributed to positioning of the nucleophilic water molecule (Pai et al., 1990). To examine if the asparagine 290 of Rap1GAP has the same role as that of glutamine 61 of Ras in positioning the nucleophilic water, a cross activation experiment (done in collaboration with Oliver Daumke) was carried out with the combination of Rap1GAP-N290A and Rap-T61Q. For comparison, the intrinsic hydrolysis rate of Rap1-T61Q had to be determined first at 10° C, which turned out to be 0.00052 min−1 (Figure 3.17 A).

The GAP stimulated reaction (Figure 3.17 B, green line) was monitored with 50 µM Rap1GAP-N290A against 2 µM Aedans-Rap-T61Q•GTP at 10 °C. The fitted value of the rate constant is 0.0034 s−1 (0.204 min−1), which clearly indicates that there is at least 400-fold activation of the GTPase reaction with a cis-carboxyamide functionality even when the catalytic residue of Rap1GAP is mutated. It is difficult to compare the activation of wild type Rap1 and Rap1GAP (see Figure 3.12 and table 2) with the cross activation experiment done under similar condition since the data analysis of the former required two intermediate steps whereas the later can be described by a single exponential function. However, considering −1 −1 the values of k2 (7.74 s ) or k3 (3.13 s ) given in table 2, it is evident that the cross activation is still 500-2000 fold reduced than the activation of the wild type.

Figure 3.17. Cross activation experiment. Even when the catalytic asparagine-290 of Rap1GAP is mutated, Rap1-T61Q still can lead to 400-fold activation of GTPase reaction.

The results upto this point clearly shows that Rap1GAP is designed to tolerate several mutations of Rap, which in case of the nearest homologue Ras are oncogenic. Moreover, the catalytic residue employed by Rap1GAP is unique. The catalytic asparagine cannot be mutated to any other residue, including positively charged residues like arginine (Ras and RhoGAPs) or lysine (nitrogenase, (Schindelin et al., 1997)) to accelerate the hydrolysis. On 74 Results the other hand, the position 61 of the GNBP is better optimized for a threonine instead of the conventional glutamine, which is the single most interesting difference between Ras and Rap GNBPs (Bos et al., 2001). These unique aspects of GTPase activation in Rap•Rap1GAP system made it necessary to revisit transition state mimics to obtain more information, which is described below.

3.2.5 Transition state mimicking for Rap•RapGAP reaction

− BeF 3 (Schindelin et al., 1997; Miller et al., 2001), AlFx (Sternweis and Gilman, 1982; Chabre,

− 1990), MgF 3 (Antonny et al., 1990; Graham et al., 1999a; Graham et al., 2002) and vanadate (Fisher et al., 1995; Smith and Rayment, 1996) are commonly used ground state and transition state mimics for phosphoryl transfer reactions along with GDP (or ADP). Such studies have been done on myosin, F-actin (Combeau and Carlier, 1988; Combeau and Carlier, 1989), sarcoplasmic ATPases (Danko et al., 2004), nitrogenase (Schindelin et al., 1997), Gα subunits of heterotrimeric G proteins (Chabre, 1990; Antonny and Chabre, 1992), tubulin (Carlier et al., 1988) and small GNBPs with their respective GAPs (Mittal et al., 1996; Ahmadian et al., 1997a). Therefore these transition state mimics were tested on Rap•GDP•Rap1GAP as described in the methods (see 2.3.13.2.).

− − Figure 3.18. Effect of BeF 3 , AlFx, MgF 3 and vanadate on Rap•GDP•Rap1GAP (A) and Ras•GDP•NF1 (B). Time axis in each case is logarithmic.

Figure 3.18A shows the association of Rap1GAP with fluorescently labeled Rap•GDP•AlFx

(red line) followed by stopped flow measurements. Rap•GDP•AlFx associates with Rap1GAP leading to an increase in the fluorescence signal. The association process is slow, similar to

− what is seen for Ras and Rho proteins (A. Wittinghofer, unpublished). MgF 3 , which has been suggested to be a transition state analogue for certain systems (Antonny et al., 1993; Graham et al., 1999a; Graham et al., 2002) does not show any significant increase in the fluorescence intensity (Figure 3.18A, green line) within the time of measurement. Similar experiment with vanadate also does not lead any increase in fluorescence (Figure 3.18A,

75 Results violet line) indicating that vanadate is not a suitable transition state mimic for this system, unlike in the case of Dictystelium discoideum myosin II that also has an asparagine residue in the active site (Fisher et al., 1995; Smith and Rayment, 1996; Rayment, 1996).

− The situation is more interesting with BeF 3 , the widely accepted ground state analogue. It shows a distinct bi-phasic association curve. The same is true for Ras•NF1 system (Figure 3.18B), which also shows a biphasic binding curve. In a previous study with transducine Tα subunit, such an effect was attributed to the presence of two activating complexes:

− − BeF 3 •H2O and BeF2(OH) •H2O (Antonny and Chabre, 1992).

− If BeF 3 was the only beryllium species that associated with Rap•GDP and Rap1GAP, the kinetics should follow a simple exponential as for an apparent first order mechanism.

Experimentally, a fast phase of limited amplitude (Figure 3.18, τ1 in Figure 3.19) is followed by a slower enhancement (Figure 3.18, τ2 in Figure 3.19), which reaches saturation at about 50 s (black dotted line, saturation of the titration curve is not shown). This suggests the presence of two associating species in solution, one which binds rapidly either because of high kinetic binding constant (kON) or high mole fraction under the experimental fluoride concentration but the incomplete saturation with this species also suggest high dissociation rate constant (kOFF). Thus, second species is likely to have much lower equilibrium dissociation constant and able to saturate the protein despite its slow rate of binding in this range of fluoride concentration. Hence, the first phase of fluorescence increase represents concurrent binding of both species and the second phase represents slower but higher affinity binding of the second complex.

− Figure 3.19. Fluorescence evolution of Aedans-Rap•GDP•BeF 3 and Rap1GAP upon mixing in

the stopped flow setup. τ1 and τ2 are the time span of the two phases.

76 Results

Therefore, such a kinetics can be analyzed with a two component model (see 3.2.7) where the apparent activation constant (terminology adopted from (Antonny and Chabre, 1992)) for + + the high affinity species is kBF(2) = 1/ τ2 and for the low affinity species is kBF(1) = 1/ τ1 - 1/ τ2. From a double exponential fit (shown later in figure 3.21) the first association rate constant + −1 + can be calculated as: kBF(1, wild type) : 19.6 s and for the second association step, kBF(2, wild type) : 0.12 s−1. Clearly the second association step is nearly 200 fold slower than the first.

Apart from fluoride concentration (pF: -log[F−]), pH might also exert an effect (Antonny and Chabre, 1992) but this could not be tested due to stability problems of Rap1GAP in different buffer conditions. The effect of the mutations, which modulated hydrolysis rate, is judged in

− terms of AlFx and BeF 3 association as discussed in the next sections. The experiment with Ras•NF1 (Figure 3.18 B) is an important control for this purpose since the same phenomena of slow association of transition state analogue and biphasic association of the ground state analogue is observed also for this different system.

3.2.6 AlFx experiment for Rap and Rap1GAP mutants

The AlFx association experiment (3.2.5) was repeated on several mutants of either Rap1 or

Rap1GAP or in combination of the two. As seen in Figure 3.20 A, association of AlFx and wild type Rap1GAP (R1G) is only possible for wild type Rap (red line). Any mutation of either position 12 or 61 (T61A, T61Q, T61L or G12V) completely diminished the association process. If AlFx is the suitable transition state mimic for Rap•Rap1GAP system, evidently all these mutations disturb the transition state formation. This is an interesting contradiction to the hydrolysis experiments (see Table 3) since, except Rap-T61L; all the other mutants can be effectively downregulate by Rap1GAP.

Figure 3.20. A) Effect of Rap mutations on association with AlFx. R1G: Rap1GAP; wt: Rap1b, T61L: Rap1b-T61L, T61Q: Rap1b-T61Q, G12V: Rap1b-G12V and T61A: Rap1b-T61A.

B) AlFx cross activation: With Rap1GAP-N290A and Rap1-T61Q, AlFx binding can still be seen (blue line) which is not possible with wild type Rap (green) or with Rap1GAP and Rap-T61Q (violet). The time axis is logarithmic.

77 Results

Interestingly, when the asparagine residue of Rap1GAP is mutated to an alanine, the association with Rap1-T61Q is still possible (albeit slow as depicted in Figure 3.21 B blue line). Conversely, Rap1GAP wild type can not associate with the Rap-T61Q mutant in the presence of AlFx (Figure 3.20B, violet). This is the physical basis of the cross-activation experiment (3.2.4), which shows that there is space for only one carboxyamide residue in the active site. The presence of two carboxyamides might allow a transient binding (Rap1GAP can still activate Rap1-T61Q) but not the stable transition state formation under these conditions.

− 3.2.7 BeF 3 : ground stare mimic or transition state mimic?

While only two combinations of residues (see 3.2.6) allow AlFx association, Rap1GAP stimulates the GTP hydrolysis of several other mutations as shown previously (Table 3).

− Similarly, these mutations also allow association of the ground state analogue (BeF 3 ). The

− red line in Figure 3.21 represents the BeF 3 association for wild type Rap1 and Rap1GAP for

− comparison. One can clearly see the association of BeF 3 with Rap1-T61A•GDP (Figure 3.21 A, blue line) and Rap1-T61Q•GDP (Figure 3.21A, purple line) with wild type Rap1GAP although the absolute fluorescence intensities are three to five fold reduced. However, It is not understood why the second association phase is not seen for these two mutants. Similar is the situation with Rap1-T61Q and Rap1GAP-N290A, which is described in the cross- activation experiments (see 3.2.4), where a steady binding phase is observed (Figure 3.21 B, purple line). Again Rap1-T61L and Rap1-G12V, the mutants that have reduced GAP stimulation, also show negligible association (Figure 3.21 B, black and green lines respectively).

− Figure 3.21. BeF 3 association to wild type and mutants of Rap1 and Rap1GAP. A) Traces of association with wild type and mutants of Rap1 and wild type Rap1GAP. B) Traces with Rap1GAP-N290A mutant. R1G: Rap1GAP.

78 Results

− The associating kinetics with BeF 3 , Rap1GAP-N290A and wild type Rap1•GDP draws special attention. Rap1GAP-N290A binds to Rap1•GTP and but does not associate with AlFx (Daumke et al., 2004), which was the verification that this asparagine 290 is indeed the

− catalytic residue. With BeF 3 , again a biphasic curve is obtained (Figure 3.21 B, green line with the biexponential fit shown in black dash) where the first association process is similar but the second association process is at least 3 fold slower. The first association rate + −1 + constants for the wild type and N290A mutants are: kBF(1, wild type) : 19.6 s and kBF(1, N290A) : −1 + −1 + −1 11.6 s but for the second association step, kBF(2, wild type) : 0.12 s and kBF(2, N290A) : 0.04 s . This shows that even when the carboxyamide of the catalytic asparagine is missing, ground state formation is still feasible. The first and particularly the second association steps are slower for the asparagine-290 mutation. This would be expected if the carboxyamide functionality directly contacts and stabilizes the Beryllium species in ground state (or γ- phosphate in case of GTP). In this light, the further decrease of amplitude for any of the threonine-61 mutants (Figure 3.21 A, B) would indicate that the side chain hydroxyl group of threonine-61 of Rap also has a similar stabilizing role and has a predominant role compared to the asparagine in ground state stabilization.

3.2.8 GAP activation of T61Q mutant with FTIR spectroscopy

The Rap-T61Q mutant has several unique properties. This mutant has a faster intrinsic hydrolysis rate (Table 3) but showed a reduced GAP activation. Moreover, this mutant did not bind a transition state analogue (AlFx, Figure 3.20A) with wild type Rap1GAP and − showed a very different kind of behavior with the ground state analogue (BeF 3 , Figure 3.21 B) where only a steady binding phase with reduced fluorescence amplitude was observed. On the other hand, the carboxyamide functionality at position 61 allowed activation of GTPase reaction even at the absence of the catalytic asparagine 290 (see 3.2.4) and transition state formation (Figure 3.20 B). Therefore the intrinsic and GAP catalyzed reactions for this mutant were monitored with time resolved FTIR spectroscopy which provides details at atomic resolution and in real time scale. In the intrinsic reaction the T61Q mutant of Rap again turned out to be ca. 10 fold faster (data not shown). The most surprising outcome is the GAP catalyzed reaction, which turned out to be even faster than the wild type. Unfortunately it was not possible to resolve any intermediate for the GAP catalyzed reaction of Rap1-T61Q mutation, therefore no spectral comparison could be made. However, making use of the previously reported band assignments of Rap-RapGAP system (see chapter 3.1), specific kinetic information for the disappearance of GTP (at 1271 cm−1) and appearance of GDP-β (at 1101 cm−1) could be monitored. As seen in figure 3.22, the difference in disappearance of marker GTP band is nearly two fold faster for the Rap-T61Q mutant (blue line) than the wild type (red line). Similarly, the appearance of GDP is also two fold faster for the Rap-T61Q mutant (green line) compared to the wild type Rap (purple line). The deviation 79 Results from single exponential nature is an outcome of reversibility of GTPase reaction (see 3.1.5). The results are summarized in the following table 4.

Figure 3.22. Disappearance of GTP at 1271 cm−1 and appearance of GDP-β at 1101 cm−1 from single frequency FTIR kinetics at saturating concentration and 258 K. The time axis is logarithmic. The datapoints are shown without baseline correction. t61q: Rap1- T61Q; wt: Rap1 wild type.

Table 4 GAP catalyzed reaction of Rap1wt. and Rap1-T61Q with Rap1GAPwt at 258 K.

Rate constant from single Rate constant from single Rap1/mutant vs. exponential fit with marker exponential fit with marker Remark Rap1GAP GTP band at 1271 cm−1 GTP-β band at 1101 cm−1 Both steps are ca. 2 Rap1 wild type 0,37 s−1 0,13 s−1 fold faster for Rap-

T61Q than Rap wild −1 −1 Rap1-T61Q 0,66 s 0,26 s type

This clearly indicates that the Rap-T61Q mutant is a better substrate for Rap1GAP at very high concentration range (5-10 mM of both proteins as described earlier in 2.4.3).

3.2.9 Conclusion

The results presented clearly shows that the residue 61 of Rap is not essential for either intrinsic or GAP stimulated hydrolysis. However, it is best optimized as a threonine, which also distinguishes Rap from other families of small GNBPs.

A previously described fluorescence reporter assay was extended to monitor the ease of ground and transition state formation with suitable analogues. It could be shown that only the wild type Rap•Rap1GAP and Rap-T61Q•Rap1GAP-N290A combination can associate with

AlFx, the transition state analogue. Even in the absence of the catalytic residue of the GAP, 80 Results the second combination can also stimulate GTPase reaction. This is not possible with wild type Rap and Rap1GAP-N290A mutant though Rap1GAP-N290A binds to Rap•GTP (Daumke et al., 2004) and the ground state analogue. Therefore a carboxyamide residue is necessary either is cis or in trans to achieve GTPase activation. To summarize, one can conclude that the catalytic residue needs to be an asparagine and there is insufficient space for two carboxyamide functionalities.

Several mutations of Rap1 (T61A, G12V, T61Q) can be stimulated by Rap1GAP and in fact, Rap-T61Q is a better substrate for Rap1GAP at higher concentration. In any of these cases, the classical transition state formation (AlFx association) is not possible. So-called ground − state formation (BeF 3 association) is still possible though the nature of the association is

− very different. The fluorescence amplitude is reduced for BeF 3 association experiments for all these mutants and the second binding phase is not observed. Whether this might reflect a state with distorted geometry that is still partially competent for hydrolysis, is an open question, which can be addressed from structural investigation.

In principle, this might be an outcome of reduction of affinity caused by a mutation at position 61 or 12. This is not surprising because even a smaller change from one nucleotide analogue to another has been suggested to perturb the nucleotide binding site and thereby, interaction with partner proteins (Geyer et al., 1996; Kuhlmann et al., 1997; Kraemer et al., 2002). However, this might be a painstaking task to achieve saturation condition for these mutations with standard fluorescence spectroscopic method. Even with the assumption that these mutations do not change the affinity significantly and the equilibrium dissociation constant is 10 µM, which is the modest of the estimated values for this system (Brinkmann et al., 2002; Kraemer et al., 2002), millimolar concentration of GAP would be necessary to achieve near saturation of 200-500 nM of Rap (or one of the mutants; see discussion). An indirect wayout might be FTIR spectroscopy as described in the earlier chapter, since none of the T61A, T61Q or G12V mutations are invasive.

81 Results

3.3 An extension to RheB-Tsc1/Tsc2 system

This chapter extends the understanding of Rap1GAP mechanism to the Tsc1-Tsc2 complex, which is a GAP for the small GNBP RheB (Tee et al., 2003), involved in insulin dependent growth pathway (see 1.1.7). It has been demonstrated that tuberin (Tsc2) or hamartin (Tsc1) alone can stimulate the GTPase reaction of the cognate GNBP RheB by 2 fold but when they form a heterodimer, the stimulation is enhananced to 200 fold (Tee et al., 2003). However, in such experiments, typically, one of the two proteins is transiently overexpressed, immunoprecipitated and subjected to a GTPase assay. The effect of the endogenous population of the partner protein is seldom taken into consideration. Therefore the possibility that both Tsc1 and Tsc2 are required for constituting the functional GAP and the two fold activity is due to the presence of small amount of the other protein can not be ruled out.

Interestingly, as per sequence homology, the GAP domain of Tsc2 is markedly shorter (157 residues, see Figures 1.14 and 3.11) than the functional GAP domain of Rap1GAP, which comprises of 341 residues (Rubinfeld et al., 1992). It has also been demonstrated that this shortened GAP domain of Tsc2 is necessary for biological activity but it is not sufficient.

The crystal structure of Rap1GAP (Daumke et al., 2004) clarified this situation. The structure of Rap1GAP revealed two domains, which were termed catalytic domain and dimerization domain (see Figure 1.15). The catalytic domain of Rap1GAP is more conserved than the dimerization domain in RapGAP family (see in Figures 1.14 and 3.11) and harbors several Tuberous sclerosis disease mutations as shown in Figure 3.22 (Daumke et al., 2004).

Figure 3.22. Tuberin mutations found in tuberous sclerosis patients are modeled on the catalytic domain of Rap1GAP structure (PDB: 1SRQ, Daumke et al., 2004). The catalytic asparagine-1643 (N-290 for Rap1GAP) is situated on conserved helix α-7 shown in purple. Another disease mutation R1743 (R-388 for Rap1GAP) is situated on the elongated helix α-9.

82 Results

The two lysines K194 and K285, which most dramatically affect Rap1GAP activity upon mutation to alanine (Brinkmann et al., 2002), are both located in the catalytic domain to which tuberin (Tsc2) shares homology. Moreover, arginine 388 of Rap1GAP (homolog of arginine 1743 of tuberin) is a part of C-terminal helix α-9 of Rap1GAP, belonging to the GAP domain (Daumke et al., 2004). This residue is important for activity towards Rap and is a reported disease mutation in Tuberous sclerosis. This helix has extensive contacts to dimerization domain (Figure 1.15). This suggests a functional role of the dimerization domain. As judged from the sequence homology, the dimerization domain is completely absent in Tuberin (Figure 3.11). This evoked the question what the GAP domain of Tsc2 alone is doing and if the dimerization domain could be supplemented from Hamartin to reconstitute GAP activity.

3.3.1 Initial attempts of protein expression from Cos7 cells

Due to size and complexity of the proteins (see Figure 1.16), the initial idea was to express full length and shortened constructs of tuberin and hamartin in mammalian cells, purify them by immunoprecipitation (see 2.5.5) and check the GAP activity of the immunoprecipitated complexes towards RheB by a radioactive GTPase assay using 32P labeled GTP (Tee et al., 2003; Garami et al., 2003). This procedure can in principle allow the identification of minimal GAP domains. For this purpose, a transfection procedure in Cos7 cells was standardized using a β-Galactosidase (see 2.5.4) assay. For the full length Tsc1 and Tsc2 constructs (see 2.2.10), the expression was checked by immunostaining (see 2.5.7) as shown in Figure 3.23.

Figure 3.23. Expression of full length Myc-Tsc1 and Flag-Tsc2 in Cos7 cells as seen by immunostaining. (A) Flag-Tsc2 full length (Flag antibody), (B) Myc-Tsc1 full length (Myc antibody), and (C) Cotransfection of Myc-Tsc1 and Flag-Tsc2 (Flag antibody).

Expression of full length Tsc2, Tsc1 and coexpression are shown in Figures 3.23 A, B and D respectively. Clearly, Tsc2 expression is seen (A) but expression of Tsc1 (B) was significantly less. In the co-transfection experiment (D), Tsc2 staining is mainly seen at the membrane, which is an indication of Tsc1 expression since Tsc1 localizes Tsc2 to the membrane. However, western blot analysis showed only Tsc2 expression and not detectable

83 Results expression of Tsc1 (data not shown). Due to persistent difficulties in expression of Tsc1, an alternative strategy was adopted through bioinformatics and recombinant protein expression as described in the next sections.

3.3.2 Putative dimerization domain of Hamartin: a bioinformatics approach

In case of tuberin (Tsc2), a large part of the disease mutations are localized in the GAP domain and these residues, including the catalytic asparagine, are conserved between different species. Moreover the catalytic domain is homologus to the RapGAP family (see Figures 1.14 and 3.11), which made domain identification easier. However, this does not hold for Hamartin (Tsc1) where some of the mutations reported for tuberous sclerosis patients are not conserved even between human and rat. (supplement figure 1, appendix 6.4).

To find if a domain similar to the dimerization domain of Rap1GAP is embedded in Hamartin, residues 86-184 of Rap1GAP, which constitute the dimerization domain in the crystal structure (PDB: 1SRQ), and corresponding residues of other bonafide homologues E6TP1, SPAL and SPA1 (see Figure 1.14) were aligned. Such an alignment procedure gives a primary consensus sequence, which was divided into small peptide fragments of 10, 15 or 20 residues and matched against the complete Hamartin sequence. This process was repeated several times with gradual frameshift of the peptide fragments. In each step identity and homology were recorded. This procedure identified one region of Tsc1 having significant similarity as shown in Figure 3.24 below.

Figure 3.24. Sequence alignment of the dimerization domains of Homo sapiens E6TP1 (swiss prot accession number (SW): Q9UNU4), Homo sapiens SPA1 (SW: O60618), Homo sapiens Rap1GAP (SW: P47736) with residues 742-857 of Rattus norvegicus Hamartin (SW: Q9Z136). Red: identity, Green: strong homology and pink color indicates possibility of homology modeling with one or the other Rap1GAP family members.

Interestingly, this region is also conserved in hamartin sequences from different species (red dotted arrow in supplement figure 1, appendix 6.4) and no function has so far been attributed

84 Results to this part (see 1.2.4.3, Figure 1.16). Therefore this was a candidate sequence for a putative dimerization domain.

3.3.3 Fragment designing for recombinant expression and purification

To check for the role of the GAP domain of tuberin and the putative dimerization domain of hamartin, three fragments were designed from sequence alignment, Rap1GAP structure and secondary structure prediction. Tsc21538-1729: the core GAP domain as seen from the structure of Rap1GAP; Tsc21520-1763: an extended tuberin GAP domain that is conserved between different specieses (in tuberins) and contains arginine 1743 (arginine 388 of Rap1GAP), mutation of which is reported for tuberous sclerosis patients, and Tsc1645-872: the putative dimerization domain (see 2.2.10).

These proteins were expressed as described in section 2.3.5 and purified in the usual procedure as described for Rap1GAP (see 2.3.7) with small modifications. After extensive wash in the GSH column, Tsc21538-1729 was cleaved overnight by thrombine, concentrated by using an Amicon concentrator (10kD cutoff) and further purified on Sephadex 75 column using 20 mM Hepes (pH 7,5), 100 mM NaCl, 5 mM DTE as running buffer. The protein eluted as monomer (21.7 kD, 95% pure) and was concentrated to 30-60 mg/ml and flash frozen.

Tsc21520-1763 and Tsc1645-872 were purified similarly by affinity purification using GSH-column with the same buffer containing in addition 1 mM ATP and 20 mM KCl. The buffer was 1520-1763 exchanged to 20 mM HEPES (pH 7.4), 5 mM MgCl2, 5 mM DTE and for Tsc2 , GST was cleaved by TEV protease by overnight incubation. Due to stability problems, GST was not cleaved for the Tsc1645-872 construct. Tsc21520-1763 could be concentrated to 5 mg/ml (27.7 kD, ~80% pure) while Tsc1645-872 could be concentrated to 1 mg/ml (53.8 kD, 40% pure, with degradation bands) in 20 mM HEPES (pH 7.4), 5 mM MgCl2, 5 mM DTE with additional salt (300 mM NaCl) and 5% ethylene glycol. The purity of the proteins was judged from SDS- Page as shown in Figure 3.25. From now on, T2s represents the shorter and stable construct Tsc21538-1729, T2 represents the longer construct Tsc21520-1763 and H1 represents GST-Tsc1645-872. The Tsc1645-872 construct is unstable and shows degradation upto GST.

Figure 3.25. Purity of the recombinant fragments. T2s represents the shorter and stable construct Tsc21538-1729, T2 represents the longer construct Tsc21520-1763 and H1 represents GST-Tsc1645- 872. SDS7: molecular weight marker: 66, 45, 36, 29, 24, 20 and 14.4 kD.

85 Results

3.3.4 Biochemistry of Tsc21538-1729, Tsc21520-1763 and Tsc1645-872 fragments

The standard HPLC method was used to check for the activity of the purified constructs. In this procedure, 80 µM RheB•GTP was converted to RheB•GDP at 30 °C in the standard assay buffer as described in 2.3.11. To investigate the effect of the Tsc21538-1729, Tsc21520-1763 and Tsc1645-872 fragments, they were used at 2.5 µM concentration as estimated from SDS- PAGE (Figure 3.25), either stand-alone or in combination. The green line in Figure 3.26 indicates intrinsic hydrolysis reaction while the black line represents effect of both Tsc21520- 1763 (T2) and Tsc1645-872 (H1). The Tsc21520-1763 fragment alone also catalyzed the hydrolysis by two fold (red dotted line). No effect was observed for the Tsc1645-872 fragment as shown by blue dotted line. The results are summarized in Table 5.

Figure 3.26. Hydrolysis experiment of RheB•GTP at 30 °C. As earlier, H1 denotes GST- Tsc1645- 872, T2 denotes Tsc21520-1763 and T2S corresponds to Tsc21538-1729. All measurements were done at 30 °C.

Table 5. Effect of tuberin and hamartin fragments on hydrolysis of RheB•GTP

GAP −1 kobs (cat) / Construct [GAP] kobs / s Remark Fragment kobs (non-cat)

None 0 0,0024 1 Intrinsic

H1 + T2 2,5 µM each 0,0049 2 2 fold faster

T2 2,5 µM 0,0047 2 RheB1-170

Nearly H1 + T2S 2,5 µM each 0,0032 1,3 intrinsic

H1 2,5 µM 0,0024 1 No effect

86 Results

Clearly, there is only a two-fold increase of hydrolysis rate by the less stable fragment of tuberin (1520-1763) while the stable fragment (1538-1729) is nearly inactive. The putative dimerization domain does not have any effect on hydrolysis under this condition.

3.3.5 Conclusion

In this work, in vitro GAP activity of tuberin could be demonstrated. A two fold activation of GTPase reaction is also known for standalone tuberin full length protein from cell biological data (Tee et al., 2003). It is interesting to note that the less stable longer fragment of tuberin (1520-1763) which contains the arginine 1743 (arginine 388 in Rap1GAP numbering) functions better than the core GAP domain which is almost inactive. This is in line with previously reported results (Daumke et al., 2004) where it has been shown that mutation of this residue renders the protein nearly inactive. Therefore, this fragment possesses the features of the tumorsuppressor protein.

The outcome of the effect of the putative dimerization domain is disappointing. However, a similar effect is also known for Rap1GAP. In the absence of the dimerization domain, the core GAP domain Rap1GAP is completely inactive. With small parts of the dimerization domain (construct 183-398), marginal GAP activity was observed but the stimulation was drastically reduced when compared to wild type Rap1GAP (1500 fold, Oliver Daumke, unpublished results). This is expected since the helix α-9 of Rap1GAP, which bears the arginine-388 (arginine 1743 of tuberin), has extensive contacts with the dimerization domain. This leads to a possibility that even if the assignment of the putative dimerization domain of hamartin is correct, the assay might not have worked due to low affinity between the two domains. To summarize, the effect of hamartin on GAP activity remains elusive and needs to be investigated in a more systematic way (see discussion).

87 Discussion 4 Discussion

Members of Rap GNBPs do not have the crucial glutamine 61 residue that is essential for GTP hydrolysis in almost every other GNBP examined so far. Again Rap1GAP, the GTPase activating protein for Rap1, does not employ a catalytic arginine to stimulate GTP hydrolysis. The available biochemical data indicated that the Rap1GAP mediated GTPase stimulation of Rap1 is fundamentally different from all other GAP stimulated reactions described so far. Therefore investigation of this unique GTPase activation was the main theme of this work, which was accomplished by time resolved FTIR spectroscopy, rapid kinetics, mutational analysis and studies on the formation of ground and transient states with wild type and mutants to complement x-ray crystallographic work of Rap1GAP.

4.1.1 Outcome of time resolved FTIR spectroscopy

Time resolved FTIR spectroscopy provides details of the molecular reaction mechanism at atomic resolution and real time scale. This method has been successfully used for intrinsic and GAP catalyzed reactions of Ras (Allin et al., 2001; Allin and Gerwert, 2001) and in this work, it was extended to Rap•RapGAP system. The spectra of the intrinsic reactions of Ras and Rap are very similar, which also outlines the similarity of the two GNBPs at the nucleotide binding niche. Interestingly, the GAP catalyzed reactions are very different. As observed for GAP catalyzed reaction of Ras, the main feature in catalysis was movement of negative charge toward nonbridging β-phosphate. Importantly, this effect was more pronounced for Rap•RapGAP reaction than Ras•RasGAP reaction. Since greater charge shift indicates further weakening of the β-γ bond of GTP, this would imply that the RapGAP catalyzed reaction has a transient state, which is more dissociative in nature.

Previously such a charge toward the nonbridging β-phosphate was attributed to the arginine finger of RasGAP (Allin et al., 2001). In the absence of an arginine finger, one might still speculate that another positively charged residue is brought close to (Sp) β-oxygen to induce the charge shift. Interestingly, from the recent structure of Rap1GAP and related mutational analysis (Brinkmann et al., 2002; Daumke et al., 2004), it is clear that there is a cluster of positively charged residues (Lysine 285, Arginine 286 and Histidine 287) in the catalytic helix (α-7) just before the catalytic Aspargine 290 (K R H I G N D). Mutation of these residues exert medium to dramatic effect. However, due to the invasive nature of these mutants, it is still not clear how they would exert an effect of the charge shift to β-oxygen in FTIR.

A clear difference between the GAP catalyzed reaction of Rap and Ras is that RapGAP binding induces a different GTP conformation (denoted as GTP* in this work) in the educt state, which is characterized by highly coupled vibrations. This has not been observed previously in protein bound GTP. The coupled vibrations indicate a different orientation of the

88 Discussion three phosphate groups to each other as compared to GAP catalyzed hydrolysis of Ras and the intrinsic reactions. This unusual conformation of GTP might contribute to rapid phosphoryl transfer. Expectedly, after the hydrolysis, in the product state protein-bound GDP and released ‘Pi’ appear at nearly the same positions as compared to intrinsic reaction of

Ras and Rap, indicating that GAP binding has almost no influence on the final GDP and Pi state for both Ras and Rap.

As seen for the RasGAP catalyzed reaction, an intermediate is observed in which cleavage * of γ-phosphate has taken place, but Pi is in shorter distance to GDP as compared to Ras-

RasGAP. In line with the shorter GDP-Pi distance and the longer lifetime of the intermediate, we observe a more pronounced back-reaction to GTP*. In case of GTPase reaction, such a possibility was speculated earlier (Allin et al., 2001). In this work, it was proved by additional mass spectrometric experiments. The reversibility of the hydrolysis reaction and the fact that

Pi release is at least partially rate limiting are common features of GTPase and ATPase reactions.

4.1.2 Assignment of amino acid bands

In addition to the phosphate vibrations, several bands characteristic for protein groups are observed in the intermediate, which seems to indicate that conformational change of residues in the active site and are required to catalyze cleavage. This is shown in Figure 4.1.

Figure 4.1. Amino acid bands for GAP catalyzed reaction of Rap. Red line (a): AÆB transition as described in 3.1.3 and (b) BÆC transition as described in 3.1.6. When the two spectra are aligned, a definite kinetic pattern is observed for certain bands outside the phosphate region.

89 Discussion

Phosphate vibrations are observed between 950 cm−1 to 1350 cm−1 as described previously (see results, 3.1). An intermediate like kinetics is seen at several frequencies outside the phosphate region, where amino acid vibrations are expected. A challenging task is now to assign some of these protein bands. Among these, the band at 1703 cm−1 was tentatively assigned to a carboxyamide residue from literature reported values (Gerwert et al., 1989), which indeed turned out to be the catalytic asparagine residue (Daumke et al., 2004). Similarly, the band at 1408 cm−1 might be a threonine. At the absence of the structure of the complex, such information is extremely valuable for elucidating molecular mechanism of proteins and time resolved FTIR spectroscopy is the method of choice for this task. However, mutation of catalytic residues is often invasive. Side directed isotope labeling helps in such circumstances (Engelhard et al., 1985; Sonar et al., 1994; Liu et al., 1995; Tee et al., 2003). This can be done either by in vitro translation or semisynthetic methods (Bader et al., 2000; Alexandrov et al., 2002). However, considering the stability of Rap1GAP and the fact that the catalytic residues are situated in the middle of the sequence, this might be difficult to achieve.

4.1.3 The novel hydrolysis mechanism

The mechanism of GAP stimulated GTPase reaction of most Ras like proteins involves two main residues. A glutamine is located on the GNBP that is important for positioning the nucleophilic water molecule. On the other hand, GAPs, apart from stabilizing the catalytic machinery, supply a so-called arginine finger (Scheffzek et al., 1998). Mutation of the glutamine in Ras, Rho and Ran almost completely eliminates the GTPase reaction (Der et al., 1986; Xu et al., 1997; Seewald et al., 2002). However, the importance of the arginine finger varies in the different systems. It is most important for Ras (Ahmadian et al., 1997b) and Rab (Albert et al., 1999), somewhat lower for Rho (Rittinger et al., 1997a; Rittinger et al., 1997b), while Ran does not employ an arginine finger (Seewald et al., 2002), and its importance for ArfGAP is still debated (Goldberg, 1999; Mandiyan et al., 1999), see Introduction).

The results presented in this work (Chapter 3.2) clearly shows that the residue 61 of Rap is not essential for either intrinsic or GAP stimulated hydrolysis. However, it is best optimized as a threonine, which is a characteristic of Rap GNBPs. Similarly, the small GNBP RheB whose GTPase reaction is accelerated by Rap1GAP homologue Tuberin, possesses a glutamine at the equivalent position of glutamine 61 of Ras but that is also not required for Tuberin stimulated GTP hydrolysis (Li et al., 2004). Exactly similar is the situation for the residue 12 of the two GNBPs. Rap1GAP can downregulate G12V mutation of Rap1. In case of RheB, the equivalent residue is an arginine. This speaks in favor of a unique underlying mechanism for both Rap•RapGAP and RheB•Tsc1-Tsc2, which are involved in two different signaling pathways (see introduction).

90 Discussion

In this work it is also demonstrated that the catalytic residue of the GAP can only be a carboxyamide functionality and most efficiently, an asparagine. If the catalytic asparagine is mutated to a positively charged amino acid residue like arginine or lysine, it can still lead to binding, but they do not catalyze. What can be the role of catalytic asparagine? The most obvious possibility was that it helps to position the nucleophilic water molecule required for hydrolysis. This idea was tested with the cross activation experiment. This showed that even when the catalytic asparagine is mutated to alanine, carboxyamide functionality from a glutamine residue at position 61 of the GNBP could still stimulate the reaction.

Downregulation of the G12V mutant of Rap (and RheB that has an arginine in equivalent position) is another interesting aspect. In most cases, mutations at the third position of the P- loop can not be downregulated by GAPs since any side chain at this position interferes with the arrangement of the catalytic glutamine and the arginine finger in the transition state (Scheffzek et al., 1997; Rittinger et al., 1997b). This suggests that the approach of the catalytic asparagine residue of Rap1GAP is likely to be different from the classical arginine finger of RasGAP or RhoGAPs.

Since Rap and RheB seem to have the same underlying principles for downregulation, the methods developed in this work can now be used to evaluate combination of mutations on Rap, such as RapT61Q, G12R or RapT61Q, G12V, which resemble RheB or the oncogenic Ras mutant more closely. The double mutant RapT61Q, G12R could not be purified on several attempts in the usual procedure involving anion exchange chromatography (see 2.3.7). Therefore it might be necessary to purify this protein by affinity purification method. Furthermore, it would also be interesting to check the effect of the mutation of arginine 388 in greater detail, since this mutation reduces the activity of Rap1GAP towards Rap dramatically and is a reported mutation in tuberous sclerosis.

− 4.1.4 BeF 3 association for Rap mutants

− The fact that the association of the so called ground state analogue (BeF 3 ) is still possible

− for several Rap mutations that still can be stimulated by Rap1GAP, indicates that BeF 3

− might be the right “transition state analogue” for this system. The BeF 3 association studies also turned out to be useful in a different perspective, since it allowed complex formation of

− Rap•GDP•BeF 3 •Rap1GAP (Christoph Thomas, unpublished result) that might be helpful for

− future structural investigations. Nevertheless, the nature of the BaF 3 association curve changes dramatically upon mutation. In particular, the two-phase association is no longer observed for any mutation of position 61 or 12 (Rap1 T61A, G12V or T61Q). This effect can not be arrtibuted to the threonine 61 residue of Rap, since similar two-phase association is also observed for Ras•RasGAP. Two possibilities appear in mind: (i) these mutations perturb

91 Discussion the nucleotide binding site of Rap•RapGAP system for which association of one of the

− − Beryllium species (either BeF 3 •H2O or BeF2(OH) •H2O) is no longer possible. (ii) In this study, the kinetics of association has been investigated; the equilibrium saturation condition has not been reached for these mutants. In principle, fluorescence titration can answer this question. The following paragraph shows the experimental difficulty.

Assuming A and B to be the two species in equilibrium with the complex AB as shown below,

kON A + B AB kOFF

koff [A]⋅[B] The equilibrium dissociation constant KD is defined as: K D = = kon [AB]

If [A0], [B0] and [AB] represent the initial concentration of A, B and the concentration of the complex AB at a particular time, substituting [A] with ([A0] – [AB]) and [B] with ([B0] – [AB]) and rearranging variables, one obtains:

2 [A·B] – ( [A 0 ] + [B0 ] + K D ) ⋅[A·B] + [A 0 ]⋅ [B0 ] = 0 Equation 4.1

The meaningful solution of this equation is:

2 [A0 ] + [B0 ] + KD ⎛ [A0 ] + [B0 ] + KD ⎞ [AB] = − ⎜ ⎟ −[A0 ]⋅[B0 ] Equation 4.2 2 ⎝ 2 ⎠

This root of the equation can be used for calculating percentage of saturation. Few examples are shown below:

Equilibrium Amount of Amount of label Dissociation Percentage of Rap1GAP (µM) constant (K ) saturation (µM) D (µM)

10 Rap1B C´(wt) 1000,0 ~ (Kraemer et 100% 0.2 µM al., 2002)

50 Rap1B C´(wt) 1000,0 ~ (Brinkmann 95% 0.2 µM et al., 2002)

Lower affinity mutant X 1000,0 100 µM 90,9% 0.2 µM

This is the practical difficulty for which one relies on Michales Menton or transient kinetics for

− such low affinity systems. It is worth to investigate the BeF 3 association under saturation

92 Discussion condition but alternative ways, such as 31P or 19F NMR spectroscopy (Higashijima et al., 1991; Antonny et al., 1993) might be more practical and informative for this purpose.

4.1.5 Downregulation of Rap: is the strategy unified or diverse?

Ras and Rap signaling (see introduction) have certain overlapping aspects because they share a set of common effectors but there are also independent functions of these two GNBPs. In order to achieve that, conceptually it is likely that they might have different controlling features, such as the down-regulation. The work presented here elucidated some aspects of this different mechanism in Rap•RapGAP system. In general, the basic mechanism can be expected to hold for other members of the RapGAP family (see Figure 1.14).

However, ‘There are more things in Heaven and earth’ as seen in the case of dual functional GAPs that can down-regulate both Ras and Rap. This is best known for GAP1IP4BP (Cullen et al., 1995). GAP1IP4BP along with the related proteins GAP1m, RASAL (Minagawa et al., 2001) and CAPRI (Lockyer et al., 2001) is composed of tandem N-terminal C2 domains, a C- terminal pleckstrin homology domain adjacent to a Burton’s tyrosine kinase (Btk) motif and a central catalytic RasGAP related domain. This protein has high affinity for inositol 1,3,4,5- tetrakisphosphate (Ins(1,3,4,5)P4) and exquisite specificity for this isomeric configuration. This protein is known to have GAP activity for both Ras and Rap but only the RasGAP activity is inhibited by phospholipids and is specifically stimulated by (Ins(1,3,4,5)P4). The C2A and C2B domains present in GAPIP4BP were suggested to have a role in Ca2+ dependent and independent phospholipids binding respectively. Later on, it was also reported that there is no effect of GAP1IP4BP on the concentration of intracellular Ca2+ (Walker et al., 2002).

Several aspects are not understood concerning the RapGAP activity of GAP1IP4BP. These proteins do not contain a typical RapGAP domain. Therefore it is not clear which part of the protein might be necessary and sufficient for GAP activity towards Rap. Thus, it is also not understood if GAP1IP4BP might employ a catalytic asparagine to stimulate GTP hydrolysis of Rap. It is however known that the GAP activity towards Ras is dependent on the arginine finger whereas the GAP activity towards Rap is not (P. Cullen, unpublished). Last but not the least, downregulation of mutations of positions 12 and 61 for Rap is an interesting aspect of RapGAPs that is conserved for tuberin too. It is also not clear if GAP1IP4BP can mediate the same. These are important points because they can address the question if GAP1IP4BP employs the same strategy as that of RapGAP or again a divergent strategy for GTPase stimulation of Rap.

93 Discussion

4.1.6 Systematic strategy for Tuberin and Hamartin

The role of hamartin in GTPase activation of RheB is unclear. Several arguments have been presented so far. One such argument is stabilization of Tsc2 by Tsc1. A second possibility is to say that somehow direct binding of Tsc1 and Tsc2 is necessary for the function. In other words, these two proteins functionally complement each other. The third possibility is that the N-terminal of Tsc2 has an auto inhibitory function towards to the GAP domain and Tsc1 relieves it since the C-terminal GAP domain of Tsc2 partially rescued the Eker Rat from forming Renal carcinoma (Takahiro Nobukuni, personal communication). It has also been suggested that the GAP activity can be rescued completely by reintroducing the N and C- terminals in trans (Momose et al., 2002). Furthermore, evidences that the complex of Tsc1- Tsc2 might be a trimer and not a dimer (Krymskaya, 2003) makes the situation even more complicated. Research in our laboratory on both Rap1GAP (Oliver Daumke, unpublished) and the catalytic domain of tuberin (chapter 3.3) indicated that the GAP domain alone has marginal activity towards either Rap or RheB. Therefore, the role of hamartin needs to be evaluated in a more systematic way.

Transient overexpression of the full length Tsc1 and Tsc2 proteins and deletion constructs in mammalian cells, followed by a GTPase assay can be a systematic procedure. Because Tsc2 contains part of the bonafide RapGAP domain at C-terminal and the shortened GAP domain is necessary but not sufficient, the full length Tsc2 needs to be taken along with different C-terminally shortened constructs of Tsc1 that still can associate with Tsc2 (see 1.2.4.3 in introduction). For this purpose, the expression of hamartin needs to be optimized, either in a different cell line like HEK-293 or through serum starvation. Serum starvation is a standard strategy when phosphorylation and stability related problems are encountered, nevertheless, this procedure is very time consuming and often not documented in literature.

4.1.7 A possible Tuberous sclerosis Chimera

An important aspect of Rap1GAP activity is stabilization of helix α-9, which folds back from the catalytic domain to the dimerization domain and bears the crucial arginine 388 residue. The homologue of arginine 388 in tuberin is arginine 1743, which is a known Tuberous sclerosis mutation. The helix α-9 has extensive hydrophobic contacts to helix α3 and strand β1 of the dimerization domain.

The putative dimerization domain of hamartin (Chapter 3.3) was cloned, purified and its effect on the GTPase reaction of RheB was analyzed. While in our assay, the hamartin fragment did not exert any effect on the hydrolysis of RheB•GTP; this might still be an affinity problem. Tuberin and hamartin (see Figure 1.16) contain two N-terminal domains that interact with each other and form the heterodimer. These two domains are missing in the GAP (Tsc21520-1763) and putative dimerization (Tsc1645-872) constructs that were tested. For this

94 Discussion reason, the affinity of the GAP domain of Tsc2 and the putative dimerization domain of Tsc1 might be very low. Low affinity might be the barrier if these two components need to collide at RheB•GTP at the same time since reactions of molecularity three are statistically less favorable.

A possible way to verify this hypothesis is a formation of a Tsc1-Tsc2 chimera as shown in Figure 4.2. Due to the availability of cDNAs, the Tsc1 (743-857) sequence is taken from Rat while the Tsc2 (1536-1778) sequence is taken from Human. At the protein level, the sequences are nearly identical. Therefore significant difference is not expected.

Figure 4.2. Proposed Tuberous sclerosis chimera. Residues 743-857 of Rattus norvegicus Hamartin (SW: Q9Z136) and residues 1536-1778 of Homo sapience Tuberin (SW: P49815) are used. The red arrow indicates proposed point of connection, which resides on the helix α-3 that joins the two domains. Secondary structural elements are indicated from the crystal structure of Rap1GAP (PDB: 1SRQ). 95 Discussion

Helix α3 connects the two domains of Rap1GAP (see figure 1.15). The proposed connection to join the putative dimerization domain of hamartin and the shortened GAP domain of tuberin is exactly on this helix, as indicated by the red line. The connecting residue from the hamartin side is a conserved glutamate (E857) and the first residue of the tuberin side is a glutamine (Q1536), which is highly conserved in RapGAP family. Several hydrophobic residues of helix α3 and one nearly conserved arginine (colored in pink) can also be modeled from this hamartin stretch. The purple color indicates residues in the hamartin part that can be modeled from the dimerization domain of one or the other bonafied Rap1GAPs.

This approach has drawbacks. Firstly, there might be multiple ways to join two fragments and the cloning procedure for making such chimeras is often unconventional. Secondly, expression of a chimeric protein without a flexible linker involves definite amount of luck factor. Nevertheless, if this approaches leads to an active chimeric GAP that would have two implications. Firstly, it would lead to a new concept in signal transduction: Two domains come together from two different chromosomes to constitute an active GAP that is important in a cell proliferation (see 1.1.7, TOR pathway) pathway.

The second implication is directly related to therapeutics. Tuberous sclerosis complex and neurofibromatosis type 1 are the two most common neurocutaneous diseases (Kandt, 2003). Gene therapeutic approaches have been undertaken for the catalytic domain of NF1 (Weiss et al., 1999), which is also a tumorsuppressor and a GAP for Ras and for other tumorsuppressors like p53 (McCormick, 2001; McCormick, 2003). Therefore, an active Tsc- chimera would also mean reconstitution of the ‘active principle’ of this tumorsuppressor complex in a small fragment, which might be useful for gene therapy.

96 Summary 5 Summary

The aim of this work was biophysical and biochemical characterization of the GTPase activating proteins (GAPs) of the small guanine nucleotide binding proteins, Rap1 and RheB, which are best known for their respective roles in integrin mediated cell adhesion and modulation of insulin dependent cell proliferation.

Rap1 is the closest homologue of Ras with characteristic replacement of crucial catalytic glutamine residue that is essential for the intrinsic and GAP mediated GTP hydrolysis of all other small GNBPs, to a Threonine. Rap1GAP, the cognate GTPase activating protein for Rap1, is also unique since mutational analysis indicated that Rap1GAP does not employ a catalytic arginine finger like other GAPs. The Rap1GAP mediated GTPase stimulation was investigated by a combination of techniques like time resolved infrared spectroscopy (FTIR), mutational analysis and stopped flow kinetics studies that elucidated new aspects of phosphoryl transfer in the context of GTPase reaction and supplemented the structural investigation of Rap1GAP.

Time resolved FTIR spectroscopy revealed that the GAP catalyzed reactions of Ras and Rap are very different though the intrinsic reactions are very similar. In case of RapGAP catalyzed reaction, the typical GTP vibrations are observed only at the early time points and a new GTP conformation (designated as GTP*) evolves rapidly upon GAP binding, where all the vibrations of α, β and γ phosphates of GTP are coupled. The most important feature in GAP catalyzed reaction is a shift of negative charge towards the non-bridging oxygen of the β phosphate of GTP, an effect that was also observed for RasGAP catalyzed reaction, but more prominent for RapGAP stimulated hydrolysis reaction. This indicates a more dissociative transition state for the Rap•RapGAP reaction compared to Ras•RasGAP reaction. Reversibility of the GTPase reaction was conclusively demonstrated by mass spectrometry, which, taken together with the outcome that phosphate release is the rate- determining step in these reactions, underscores common aspects of phosphoryl transfer of ATPase and GTPase reactions. The FTIR difference spectra provided a hint for functional amino acid residues involved in hydrolysis, most importantly a carboxyamide functionality at 1703 cm−1, which in an independent study turned out to be the catalytic asparagine of RapGAP.

The biochemical analysis showed that the residue 61 of Rap is not essential for either intrinsic or GAP stimulated hydrolysis. However, it is best optimized as a threonine. On the other hand, the catalytic residue of the GAP needs to be a carboxyamide functionality and is most effective as an asparagine. Interestingly, a carboxyamide residue either in cis or in trans can stimulate the GTPase activation for Rap•RapGAP. It is also shown that Rap1GAP can downregulate several mutations of Rap at positions 12 and 61 that is not possible in the Ras context. In several of these situations, the association of the classical transition state

97 Summary

− analogue (AlFx) is not feasible anymore, but the so called ground state analogue (BeF 3 ) still

− binds. The explicit nature of the BeF 3 bound state for different mutations is still elusive.

Finally, the structural and biochemical understanding of Rap•RapGAP was extended to tuberin and hamartin, the key players of Tuberous sclerosis which form a heterodimer and act as a GAP for the small GNBP RheB. In this work, in vitro GAP activity of tuberin could be demonstrated. Interestingly, the core GAP domain of tuberin as designed from Rap1GAP structure, is nearly inactive. A less stable longer fragment (aa 1520-1763), which contains arginine 1743, a residue that is mutated in Tuberous sclerosis, has GAP activity. Therefore, this fragment possesses the features of the tumor suppressor protein. However, the activity is expectedly less than the heterodimer of full length tuberin and hamartin. Based on the available structural and biochemical data, a chimera of hamartin and tuberin is proposed that might be useful for rescuing the active principle of this tumor suppressor complex.

98 Zusammenfassung 6 Zusammenfassung

Das Ziel dieser Arbeit bestand in der biophysikalischen und biochemischen Charakterisierung der GTPase aktivierenden Proteine (GAPs) der kleinen Guaninnukleotid bindenden Proteine (GNBPs) Rap1 und RheB, die eine wichtige Rolle in der Integrin- vermittelten Zelladhäsion bzw. der Modulation der Insulin-abhängigen Zellproliferation spielen.

Rap1 weist unter allen GNBPs die höchste Homologie zu Ras auf, besitzt aber anstelle des katalytischen Glutamins, das essentiell für die intrinsische sowie die GAP-stimulierte GTP- Hydrolyse aller anderen kleinen GNBPs ist, ein charakteristisches Threonin. Das GAP für Rap1, Rap1GAP, verfügt aber im Gegensatz zu vielen anderen GAP-Proteinen über keinen katalytischen Argininfinger. Die Rap1GAP vermittelte GTPase-Stimulation wurde mit Hilfe verschiedener Techniken, wie zeitaufgelöster Infrarotspektroskopie (FTIR), Mutationsanalysen und Stopped Flow-Kinetiken untersucht, wobei neue Aspekte des Phosphoryltransfers im Kontext der GTPase-Reaktion aufgezeigt wurden, die die Erkenntnisse aus strukturellen Untersuchungen von Rap1GAP erweiterten.

Experimente mit zeitaufgelöster FTIR-Spektroskopie zeigten einen unterschiedlichen Mechanismus der GAP-katalysierten Hydrolyse von Ras und Rap1, die intrinsischen Reaktionen zeigen hingegen einen sehr ähnlichen Verlauf. Im Falle der Rap1GAP- katalysierten Reaktion wurden typische GTP-Schwingungen nur zu Beginn der Reaktion beobachtet. Eine neue GTP-Konformation (auch als GTP* bezeichnet), in der alle Schwingungen der α-, β- und γ-Phosphate des GTP gekoppelt sind, bildet sich nach GAP- Wechselwirkung schnell heraus. Der wichtigste Aspekt in der GAP-katalysierten Reaktion von Rap1 ist die Verlagerung von negativer Ladung zum nicht verbrückenden Sauerstoff des β-Phosphats von GTP. Dieser Effekt war zuvor auch schon in der RasGAP-katalysierten Hydrolysereaktion beobachtet worden, ist aber in der Rap1GAP- stimulierten Hydrolyse deutlicher ausgeprägt. Der Befund deutet auf einen stärker dissoziativen Charakter des Übergangszustandes der Rap1•Rap1GAP-Reaktion im Vergleich zur Ras•RasGAP-Reaktion hin. Die Reversibilität der GTPase-Reaktion, die über Massenspektrometrie demonstriert wurde, sowie die Identifizierung der Phosphat-Freisetzung als geschwindigkeitsbestimmenden Schritt in diesen Reaktionen, unterstreichen die Gemeinsamkeiten des Phosphoryltransfers von ATPase- und GTPase-Reaktion. FTIR- Differenzspektren ergaben darüberhinaus Hinweise auf Aminosäurefunktionen, die an der Hydrolysereaktion beteiligt. Hervorzuheben ist vor allem eine Amidfunktion bei 1703 cm−1, die dem Asparaginrest zugeordnet werden könnte, der in unabhängigen Studien als essentiell für die Katalyse identifiziert wurde.

Biochemische Analysen zeigten, dass der Aminosäurerest 61 von Rap1 weder für die intrinsische noch für die GAP-stimulierte Hydrolyse essentiell ist, wenngleich ein Threonin an

99 Zusammenfassung dieser Position maximale katalytische Aktivität erlaubt. Dagegen muss der katalytische Aminosäurerest des Rap1GAP - optimalerweise ein Asparagin - eine Amidfunktion aufweisen. Der Amidrest kann sowohl in cis als auch in trans die GTPase Aktivität im Rap1•Rap1GAP-Komplex stimulieren. Weiterhin konnte gezeigt werden, dass Rap1GAP in der Lage ist, die GTP-Hydrolyse von Rap1-Proteinen mit diversen Mutationen an den Positionen 12 oder 61 zu katalysieren, während das beim Ras•RasGAP-System nicht möglich ist. In vielen dieser Fälle kann mit dem klassischen Übergangszustandsanalogon

AlFx keine Assoziation beobachtet werden, während das so genannte − − Grundzustandsanalogon BeF 3 nach wie vor binden kann. Die explizite Natur des BeF 3 gebundenen Zustands ist für die unterschiedlichen Mutationen allerdings noch ungeklärt.

Basierend auf dem durch strukturelle und biochemische Untersuchungen erweiterten Verständnis des Rap1•Rap1GAP-Hydrolysemechanismus wurde in einem weiteren Ansatz die GAP-Aktivität des Tuberin/Hamartin-Komplexes untersucht. Diese Proteine bilden ein Heterodimer und fungieren als GAP für das kleine GNBP RheB. Mutationen in Hamartin und Tuberin sind Auslöser von Tuberous sclerosis. In der vorliegenden Arbeit konnte für Tuberin in vitro GAP-Aktivität demonstriert werden. Interessanterweise wies die Kern-GAP-Domäne des Tuberin, die durch Vergleich mit der Struktur von Rap1GAP bestimmt wurde, kaum GAP-Aktivität auf. Ein längeres instabileres Fragment dagegen (AS 1520-1763) war katalytisch aktiv. Dieses Tuberinkonstrukt enthielt den Argininrest 1743, welcher in Patienten mit Tuberous sclerosis oft mutiert ist, so dass dieses Fragment die Merkmale eines Tumorsuppressors aufweist. Die Aktivität war aber wie erwartet geringer als die des Heterodimers aus Tuberin der vollen Länge und Hamartin. Basierend auf den strukturellen und biochemischen Daten wurde eine Proteinchimäre vorgeschlagen, die die vollständige Katalyseaktivität des Suppressorkomplexes aufweisen sollte.

100 References 7 References

Aevarsson,A., Brazhnikov,E., Garber,M., Zheltonosova,J., Chirgadze,Y., al Karadaghi,S., Svensson,L.A., and Liljas,A. (1994). Three-dimensional structure of the ribosomal : G from Thermus thermophilus. EMBO J. 13, 3669-3677.

Ahmadian,M.R., Mittal,R., Hall,A., and Wittinghofer,A. (1997a). Aluminum fluoride associates with the small guanine nucleotide binding proteins. FEBS Lett. 408, 315-318.

Ahmadian,M.R., Stege,P., Scheffzek,K., and Wittinghofer,A. (1997b). Confirmation of the arginine-finger hypothesis for the GAP-stimulated GTP-hydrolysis reaction of Ras. Nat. Struct. Biol. 4, 686-689. al Karadaghi,S., Aevarsson,A., Garber,M., Zheltonosova,J., and Liljas,A. (1996). The structure of elongation factor G in complex with GDP: conformational flexibility and nucleotide exchange. Structure. 4, 555-565.

Albert,S., Will,E., and Gallwitz,D. (1999). Identification of the catalytic domains and their functionally critical arginine residues of two yeast GTPase-activating proteins specific for Ypt/Rab transport GTPases. EMBO J. 18, 5216-5225.

Alexandrov,K., Heinemann,I., Durek,T., Sidorovitch,V., Goody,R.S., and Waldmann,H. (2002). Intein-mediated synthesis of geranylgeranylated Rab7 protein in vitro. J. Am. Chem. Soc. 124, 5648-5649.

Allin,C., Ahmadian,M.R., Wittinghofer,A., and Gerwert,K. (2001). Monitoring the GAP catalyzed H-Ras GTPase reaction at atomic resolution in real time. Proc. Natl. Acad. Sci. U. S. A 98, 7754-7759.

Allin,C. and Gerwert,K. (2001). Ras catalyzes GTP hydrolysis by shifting negative charges from gamma- to beta-phosphate as revealed by time-resolved FTIR difference spectroscopy. Biochemistry 40, 3037-3046.

Almoguera,C., Shibata,D., Forrester,K., Martin,J., Arnheim,N., and Perucho,M. (1988). Most human carcinomas of the exocrine pancreas contain mutant c-K-ras genes. Cell 53, 549-554.

Altschuler,D.L. and Ribeiro-Neto,F. (1998). Mitogenic and oncogenic properties of the small G protein Rap1b. Proc. Natl. Acad. Sci. U. S. A 95, 7475-7479.

Antonny,B., Beraud-Dufour,S., Chardin,P., and Chabre,M. (1997). N-terminal hydrophobic residues of the G-protein ADP-ribosylation factor-1 insert into membrane phospholipids upon GDP to GTP exchange. Biochemistry 36, 4675-4684.

Antonny,B., Bigay,J., and Chabre,M. (1990). A novel magnesium-dependent mechanism for the activation of by fluoride. FEBS Lett. 268, 277-280.

Antonny,B. and Chabre,M. (1992). Characterization of the aluminum and beryllium fluoride species which activate transducin. Analysis of the binding and dissociation kinetics. J. Biol. Chem. 267, 6710-6718.

Antonny,B., Sukumar,M., Bigay,J., Chabre,M., and Higashijima,T. (1993). The mechanism of aluminum-independent G-protein activation by fluoride and magnesium. 31P NMR spectroscopy and fluorescence kinetic studies. J. Biol. Chem. 268, 2393-2402.

101 References

Asha,H., de Ruiter,N.D., Wang,M.G., and Hariharan,I.K. (1999). The Rap1 GTPase functions as a regulator of morphogenesis in vivo. EMBO J. 18, 605-615.

Astrinidis,A., Cash,T.P., Hunter,D.S., Walker,C.L., Chernoff,J., and Henske,E.P. (2002). Tuberin, the tuberous sclerosis complex 2 tumor suppressor gene product, regulates Rho activation, cell adhesion and migration. Oncogene 21, 8470-8476.

Bader,B., Kuhn,K., Owen,D.J., Waldmann,H., Wittinghofer,A., and Kuhlmann,J. (2000). Bioorganic synthesis of lipid-modified proteins for the study of signal transduction. Nature 403, 223-226.

Barford,D., Das,A.K., and Egloff,M.P. (1998). The structure and mechanism of protein phosphatases: insights into catalysis and regulation. Annu. Rev. Biophys. Biomol. Struct. 27, 133-164.

Barrett,T., Xiao,B., Dodson,E.J., Dodson,G., Ludbrook,S.B., Nurmahomed,K., Gamblin,S.J., Musacchio,A., Smerdon,S.J., and Eccleston,J.F. (1997). The structure of the GTPase- activating domain from p50rhoGAP. Nature 385, 458-461.

Barth,A. (1999). Phosphoenzyme conversion of the sarcoplasmic reticulum Ca(2+)-ATPase. Molecular interpretation of infrared difference spectra. J. Biol. Chem. 274, 22170-22175.

Beranger,F., Goud,B., Tavitian,A., and de Gunzburg,J. (1991). Association of the Ras- antagonistic Rap1/Krev-1 proteins with the Golgi complex. Proc. Natl. Acad. Sci. U. S. A 88, 1606-1610.

Berchtold,H., Reshetnikova,L., Reiser,C.O., Schirmer,N.K., Sprinzl,M., and Hilgenfeld,R. (1993). Crystal structure of active elongation factor Tu reveals major domain rearrangements. Nature 365, 126-132.

Bigay,J., Gounon,P., Robineau,S., and Antonny,B. (2003). Lipid packing sensed by ArfGAP1 couples COPI coat disassembly to membrane bilayer curvature. Nature 426, 563-566.

Bjornsti,M.A. and Houghton,P.J. (2004). The TOR pathway: a target for cancer therapy. Nat. Rev. Cancer 4, 335-348.

Blackburn,G.M., Williams,N.H., Gamblin,S.J., and Smerdon,S.J. (2003). Comment on "The pentacovalent phosphorus intermediate of a phosphoryl transfer reaction". Science 301, 1184.

Boehm,U., Guethlein,L., Klamp,T., Ozbek,K., Schaub,A., Futterer,A., Pfeffer,K., and Howard,J.C. (1998). Two families of GTPases dominate the complex cellular response to IFN-gamma. J. Immunol. 161, 6715-6723.

Bos,J.L. (1989). ras oncogenes in human cancer: a review. Cancer Res. 49, 4682-4689.

Bos,J.L., de Rooij,J., and Reedquist,K.A. (2001). Rap1 signalling: adhering to new models. Nat. Rev. Mol. Cell Biol. 2, 369-377.

Bourne,H.R. (1993). GTPases. A turn-on and a surprise. Nature 366, 628-629.

Bourne,H.R., Sanders,D.A., and McCormick,F. (1990). The GTPase superfamily: a conserved switch for diverse cell functions. Nature 348, 125-132.

Bourne,H.R., Sanders,D.A., and McCormick,F. (1991). The GTPase superfamily: conserved structure and molecular mechanism. Nature 349, 117-127.

102 References

Bradford,M.M. (1976). A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72, 248-254.

Brinkmann,T., Daumke,O., Herbrand,U., Kuhlmann,D., Stege,P., Ahmadian,M.R., and Wittinghofer,A. (2002). Rap-specific GTPase activating protein follows an alternative mechanism. J. Biol. Chem. 277, 12525-12531.

Brown,E.J., Albers,M.W., Shin,T.B., Ichikawa,K., Keith,C.T., Lane,W.S., and Schreiber,S.L. (1994). A mammalian protein targeted by G1-arresting rapamycin-receptor complex. Nature 369, 756-758.

Buglino,J., Shen,V., Hakimian,P., and Lima,C.D. (2002). Structural and biochemical analysis of the Obg GTP binding protein. Structure. (Camb. ) 10, 1581-1592.

Castro,A.F., Rebhun,J.F., Clark,G.J., and Quilliam,L.A. (2003). Rheb binds tuberous sclerosis complex 2 (TSC2) and promotes S6 kinase activation in a rapamycin- and farnesylation-dependent manner. J. Biol. Chem. 278, 32493-32496.

Cepus,V., Scheidig,A.J., Goody,R.S., and Gerwert,K. (1998). Time-resolved FTIR studies of the GTPase reaction of H-ras p21 reveal a key role for the beta-phosphate. Biochemistry 37, 10263-10271.

Chabre,M. (1990). Aluminofluoride and beryllofluoride complexes: a new phosphate analogs in enzymology. Trends Biochem. Sci. 15, 6-10.

Chavrier,P. and Goud,B. (1999). The role of ARF and Rab GTPases in membrane transport. Curr. Opin. Cell Biol. 11, 466-475.

Chung,C.T., Niemela,S.L., and Miller,R.H. (1989). One-step preparation of competent Escherichia coli: transformation and storage of bacterial cells in the same solution. Proc. Natl. Acad. Sci. U. S. A 86, 2172-2175.

Cohn,M. and Hu,A. (1978). Isotopic (18O) shift in 31P nuclear magnetic resonance applied to a study of enzyme-catalyzed phosphate--phosphate exchange and phosphate (oxygen)-- water exchange reactions. Proc. Natl. Acad. Sci. U. S. A 75, 200-203.

Coleman,D.E., Berghuis,A.M., Lee,E., Linder,M.E., Gilman,A.G., and Sprang,S.R. (1994). Structures of active conformations of Gi alpha 1 and the mechanism of GTP hydrolysis. Science 265, 1405-1412.

Combeau,C. and Carlier,M.F. (1988). Probing the mechanism of ATP hydrolysis on F-actin using vanadate and the structural analogs of phosphate BeF-3 and A1F-4. J. Biol. Chem. 263, 17429-17436.

Combeau,C. and Carlier,M.F. (1989). Characterization of the aluminum and beryllium fluoride species bound to F-actin and microtubules at the site of the gamma-phosphate of the nucleotide. J. Biol. Chem. 264, 19017-19021.

Cramer,P., Bushnell,D.A., and Kornberg,R.D. (2001). Structural basis of transcription: RNA polymerase II at 2.8 angstrom resolution. Science 292, 1863-1876.

Cukierman,E., Huber,I., Rotman,M., and Cassel,D. (1995). The ARF1 GTPase-activating protein: zinc finger motif and Golgi complex localization. Science 270, 1999-2002.

103 References

Cullen,P.J., Hsuan,J.J., Truong,O., Letcher,A.J., Jackson,T.R., Dawson,A.P., and Irvine,R.F. (1995). Identification of a specific Ins(1,3,4,5)P4-binding protein as a member of the GAP1 family. Nature 376, 527-530.

Czworkowski,J., Wang,J., Steitz,T.A., and Moore,P.B. (1994). The crystal structure of elongation factor G complexed with GDP, at 2.7 A resolution. EMBO J. 13, 3661-3668.

Danko,S., Yamasaki,K., Daiho,T., and Suzuki,H. (2004). Distinct natures of beryllium fluoride-bound, aluminum fluoride-bound, and magnesium fluoride-bound stable analogues of an ADP-insensitive phosphoenzyme intermediate of sarcoplasmic reticulum Ca2+- ATPase: changes in catalytic and transport sites during phosphoenzyme hydrolysis. J. Biol. Chem. 279, 14991-14998.

Das,A.K., Helps,N.R., Cohen,P.T., and Barford,D. (1996). Crystal structure of the protein serine/threonine phosphatase 2C at 2.0 A resolution. EMBO J. 15, 6798-6809.

Daumke,O., Weyand,M., Chakrabarti,P.P., Vetter,I.R., and Wittinghofer,A. (2004). The GTPase-activating protein Rap1GAP uses a catalytic asparagine. Nature 429, 197-201. de Bruyn,K.M., Rangarajan,S., Reedquist,K.A., Figdor,C.G., and Bos,J.L. (2002). The small GTPase Rap1 is required for Mn(2+)- and antibody-induced LFA-1- and VLA-4-mediated cell adhesion. J Biol. Chem. 2002. Aug. 16;277. (33. ):29468. -76. Epub. 2002. May. 24. 277, 29468-29476. de Rooij,J., Boenink,N.M., van Triest,M., Cool,R.H., Wittinghofer,A., and Bos,J.L. (1999). PDZ-GEF1, a guanine nucleotide exchange factor specific for Rap1 and Rap2. J. Biol. Chem. 274, 38125-38130. de Rooij,J., Zwartkruis,F.J., Verheijen,M.H., Cool,R.H., Nijman,S.M., Wittinghofer,A., and Bos,J.L. (1998). Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 396, 474-477.

Dennis,P.B., Jaeschke,A., Saitoh,M., Fowler,B., Kozma,S.C., and Thomas,G. (2001). Mammalian TOR: a homeostatic ATP sensor. Science 294, 1102-1105.

Der,C.J., Finkel,T., and Cooper,G.M. (1986). Biological and biochemical properties of human rasH genes mutated at codon 61. Cell 44, 167-176.

Dominguez,R., Freyzon,Y., Trybus,K.M., and Cohen,C. (1998). Crystal structure of a vertebrate smooth muscle myosin motor domain and its complex with the essential light chain: visualization of the pre-power stroke state. Cell 94, 559-571.

Doudna,J.A. and Batey,R.T. (2004). Structural insights into the signal recognition particle. Annu. Rev. Biochem. 73, 539-557.

Downward,J. (2003). Targeting RAS signalling pathways in cancer therapy. Nat. Rev. Cancer 3, 11-22.

Druey,K.M., Blumer,K.J., Kang,V.H., and Kehrl,J.H. (1996). Inhibition of G-protein-mediated MAP kinase activation by a new mammalian gene family. Nature 379, 742-746.

Du,X., Black,G.E., Lecchi,P., Abramson,F.P., and Sprang,S.R. (2004). Kinetic isotope effects in Ras-catalyzed GTP hydrolysis: evidence for a loose transition state. Proc. Natl. Acad. Sci. U. S. A 101, 8858-8863.

104 References

Egea,P.F., Shan,S.O., Napetschnig,J., Savage,D.F., Walter,P., and Stroud,R.M. (2004). Substrate twinning activates the signal recognition particle and its receptor. Nature 427, 215- 221.

Engelhard,M., Gerwert,K., Hess,B., Kreutz,W., and Siebert,F. (1985). Light-driven protonation changes of internal aspartic acids of bacteriorhodopsin: an investigation by static and time-resolved infrared difference spectroscopy using [4-13C]aspartic acid labeled purple membrane. Biochemistry 24, 400-407.

Evdokimov,A.G., Tropea,J.E., Routzahn,K.M., and Waugh,D.S. (2002). Crystal structure of the Yersinia pestis GTPase activator YopE. Protein Sci. 11, 401-408.

Feuerstein,J., Goody,R.S., and Webb,M.R. (1989). The mechanism of guanosine nucleotide hydrolysis by p21 c-Ha-ras. The stereochemical course of the GTPase reaction. J. Biol. Chem. 264, 6188-6190.

Fingar,D.C. and Blenis,J. (2004). Target of rapamycin (TOR): an integrator of nutrient and growth factor signals and coordinator of cell growth and cell cycle progression. Oncogene 23, 3151-3171.

Fingar,D.C., Salama,S., Tsou,C., Harlow,E., and Blenis,J. (2002). Mammalian cell size is controlled by mTOR and its downstream targets S6K1 and 4EBP1/eIF4E. Genes Dev. 16, 1472-1487.

Fisher,A.J., Smith,C.A., Thoden,J.B., Smith,R., Sutoh,K., Holden,H.M., and Rayment,I. (1995). X-ray structures of the myosin motor domain of Dictyostelium discoideum complexed with MgADP.BeFx and MgADP.AlF4-. Biochemistry 34, 8960-8972.

Focia,P.J., Shepotinovskaya,I.V., Seidler,J.A., and Freymann,D.M. (2004). Heterodimeric GTPase core of the SRP targeting complex. Science 303, 373-377.

Franke,B., Akkerman,J.W., and Bos,J.L. (1997). Rapid Ca2+-mediated activation of Rap1 in human platelets. EMBO J. 16, 252-259.

Frech,M., Darden,T.A., Pedersen,L.G., Foley,C.K., Charifson,P.S., Anderson,M.W., and Wittinghofer,A. (1994). Role of glutamine-61 in the hydrolysis of GTP by p21H-ras: an experimental and theoretical study. Biochemistry 33, 3237-3244.

Frech,M., John,J., Pizon,V., Chardin,P., Tavitian,A., Clark,R., McCormick,F., and Wittinghofer,A. (1990). Inhibition of GTPase activating protein stimulation of Ras-p21 GTPase by the Krev-1 gene product. Science 249, 169-171.

Freymann,D.M., Keenan,R.J., Stroud,R.M., and Walter,P. (1997). Structure of the conserved GTPase domain of the signal recognition particle. Nature 385, 361-364.

Fu,Y. and Galan,J.E. (1999). A salmonella protein antagonizes Rac-1 and Cdc42 to mediate host-cell recovery after bacterial invasion. Nature 401, 293-297.

Gao,Q., Kumar,A., Singh,L., Huibregtse,J.M., Beaudenon,S., Srinivasan,S., Wazer,D.E., Band,H., and Band,V. (2002a). Human papillomavirus E6-induced degradation of E6TP1 is mediated by E6AP ubiquitin ligase. Cancer Res. 62, 3315-3321.

Gao,Q., Singh,L., Kumar,A., Srinivasan,S., Wazer,D.E., and Band,V. (2001). Human papillomavirus type 16 E6-induced degradation of E6TP1 correlates with its ability to immortalize human mammary epithelial cells. J. Virol. 75, 4459-4466.

105 References

Gao,Q., Srinivasan,S., Boyer,S.N., Wazer,D.E., and Band,V. (1999). The E6 oncoproteins of high-risk papillomaviruses bind to a novel putative GAP protein, E6TP1, and target it for degradation. Mol. Cell Biol. 19, 733-744.

Gao,X., Zhang,Y., Arrazola,P., Hino,O., Kobayashi,T., Yeung,R.S., Ru,B., and Pan,D. (2002b). Tsc tumour suppressor proteins antagonize amino-acid-TOR signalling. Nat. Cell Biol. 4, 699-704.

Garami,A., Zwartkruis,F.J., Nobukuni,T., Joaquin,M., Roccio,M., Stocker,H., Kozma,S.C., Hafen,E., Bos,J.L., and Thomas,G. (2003). Insulin activation of Rheb, a mediator of mTOR/S6K/4E-BP signaling, is inhibited by TSC1 and 2. Mol. Cell 11, 1457-1466.

Garcia-Ranea,J.A. and Valencia,A. (1998). Distribution and functional diversification of the ras superfamily in Saccharomyces cerevisiae. FEBS Lett. 434, 219-225.

Gerwert,K. (1999). Molecular Reaction Mechanisms of Proteins Monitored by Time-Resolved FTIR-Spectroscopy. Biol. Chem. 380, 931-935.

Gerwert,K. (2001). Molecular Reaction Mechanisms of Proteins Monitored by Time-resolved FTIR Difference Spectroscopy. In Handbook of Vibrational Spectroscopy, Vol. 5 - Applications of Vibrational Spectroscopy in Life and Natural Sciences, J.M.Chalmers and P.R.Griffiths, eds. (Chichester: John Wiley & Sons Ltd., Chichester), pp. 1-21.

Gerwert,K., Hess,B., Soppa,J., and Oesterhelt,D. (1989). Role of aspartate-96 in proton translocation by bacteriorhodopsin. Proc. Natl. Acad. Sci. U. S. A 86, 4943-4947.

Gerwert,K., Souvignier,G., and Hess,B. (1990). Simultaneous monitoring of light-induced changes in protein side-group protonation, chromophore isomerization, and backbone motion of bacteriorhodopsin by time-resolved Fourier-transform infrared spectroscopy. Proc. Natl. Acad. Sci. U. S. A 87, 9774-9778.

Geyer,M., Schweins,T., Herrmann,C., Prisner,T., Wittinghofer,A., and Kalbitzer,H.R. (1996). Conformational transitions in p21ras and in its complexes with the effector protein Raf-RBD and the GTPase activating protein GAP. Biochemistry 35, 10308-10320.

Geyer,M. and Wittinghofer,A. (1997). GEFs, GAPs, GDIs and effectors: taking a closer (3D) look at the regulation of Ras-related GTP-binding proteins. Curr. Opin. Struct. Biol. 7, 786- 792.

Goehring,U.M., Schmidt,G., Pederson,K.J., Aktories,K., and Barbieri,J.T. (1999). The N- terminal domain of Pseudomonas aeruginosa exoenzyme S is a GTPase-activating protein for Rho GTPases. J. Biol. Chem. 274, 36369-36372.

Goldberg,J. (1998). Structural basis for activation of ARF GTPase: mechanisms of guanine nucleotide exchange and GTP-myristoyl switching. Cell 95, 237-248.

Goldberg,J. (1999). Structural and functional analysis of the ARF1-ARFGAP complex reveals a role for coatomer in GTP hydrolysis. Cell 96, 893-902.

Goldberg,J., Huang,H.B., Kwon,Y.G., Greengard,P., Nairn,A.C., and Kuriyan,J. (1995). Three-dimensional structure of the catalytic subunit of protein serine/threonine phosphatase- 1. Nature 376, 745-753.

Gomez,M., Sampson,J., and Whittemore V. (1999). The tuberous sclerosis complex. (Oxford, England: Oxford University Press).

106 References

Goody,R.S. (2003). The significance of the free energy of hydrolysis of GTP for signal- transducing and regulatory GTPases. Biophys. Chem. 100, 535-544.

Gordon,J.A. (1991). Use of vanadate as protein-phosphotyrosine phosphatase inhibitor. Methods Enzymol. 201:477-82., 477-482.

Gotoh,T., Hattori,S., Nakamura,S., Kitayama,H., Noda,M., Takai,Y., Kaibuchi,K., Matsui,H., Hatase,O., and Takahashi,H. (1995). Identification of Rap1 as a target for the Crk SH3 domain-binding guanine nucleotide-releasing factor C3G. Mol. Cell Biol. 15, 6746-6753.

Graham,D.L., Eccleston,J.F., Chung,C.W., and Lowe,P.N. (1999a). Magnesium fluoride- dependent binding of small G proteins to their GTPase-activating proteins. Biochemistry 38, 14981-14987.

Graham,D.L., Eccleston,J.F., and Lowe,P.N. (1999b). The conserved arginine in rho- GTPase-activating protein is essential for efficient catalysis but not for complex formation with Rho.GDP and aluminum fluoride. Biochemistry 38, 985-991.

Graham,D.L., Lowe,P.N., Grime,G.W., Marsh,M., Rittinger,K., Smerdon,S.J., Gamblin,S.J., and Eccleston,J.F. (2002). MgF(3)(-) as a transition state analog of phosphoryl transfer. Chem. Biol. 9, 375-381.

Harris,T.E. and Lawrence,J.C., Jr. (2003). TOR signaling. Sci. STKE. 2003, re15.

Hattori,M., Tsukamoto,N., Nur-e-Kamal MS, Rubinfeld,B., Iwai,K., Kubota,H., Maruta,H., and Minato,N. (1995). Molecular cloning of a novel mitogen-inducible nuclear protein with a Ran GTPase-activating domain that affects cell cycle progression. Mol. Cell Biol. 15, 552-560.

Herrmann,C. (2003). Ras-effector interactions: after one decade. Curr. Opin. Struct. Biol. 13, 122-129.

Hessling,B., Souvignier,G., and Gerwert,K. (1993). A model-independent approach to assigning bacteriorhodopsin's intramolecular reactions to photocycle intermediates. Biophys. J. 65, 1929-1941.

Higashijima,T., Graziano,M.P., Suga,H., Kainosho,M., and Gilman,A.G. (1991). 19F and 31P NMR spectroscopy of G protein alpha subunits. Mechanism of activation by Al3+ and F-. J. Biol. Chem. 266, 3396-3401.

Hillig,R.C., Renault,L., Vetter,I.R., Drell,T., Wittinghofer,A., and Becker,J. (1999). The crystal structure of rna1p: a new fold for a GTPase-activating protein. Mol Cell 3, 781-791.

Hinshaw,J.E. (2000). Dynamin and its role in membrane fission. Annu. Rev. Cell Dev. Biol. 16:483-519., 483-519.

Inoki,K., Corradetti,M.N., and Guan,K.L. (2005). Dysregulation of the TSC-mTOR pathway in human disease. Nat. Genet. 37, 19-24.

Inoki,K., Li,Y., Xu,T., and Guan,K.L. (2003a). Rheb GTPase is a direct target of TSC2 GAP activity and regulates mTOR signaling. Genes Dev. 17, 1829-1834.

Inoki,K., Li,Y., Zhu,T., Wu,J., and Guan,K.L. (2002). TSC2 is phosphorylated and inhibited by Akt and suppresses mTOR signalling. Nat. Cell Biol. 4, 648-657.

Inoki,K., Zhu,T., and Guan,K.L. (2003b). TSC2 mediates cellular energy response to control cell growth and survival. Cell 115, 577-590.

107 References

Ishida,D., Kometani,K., Yang,H., Kakugawa,K., Masuda,K., Iwai,K., Suzuki,M., Itohara,S., Nakahata,T., Hiai,H., Kawamoto,H., Hattori,M., and Minato,N. (2003). Myeloproliferative stem cell disorders by deregulated Rap1 activation in SPA-1-deficient mice. Cancer Cell 2003 Jul. ;4(1):55. -65. 4, 55-65.

Ito,N. and Rubin,G.M. (1999). gigas, a Drosophila homolog of tuberous sclerosis gene product-2, regulates the cell cycle. Cell 96, 529-539.

Jacinto,E. and Hall,M.N. (2003). Tor signalling in bugs, brain and brawn. Nat. Rev. Mol. Cell Biol. 4, 117-126.

Janoueix-Lerosey,I., Polakis,P., Tavitian,A., and de Gunzburg,J. (1992). Regulation of the GTPase activity of the ras-related rap2 protein. Biochem. Biophys. Res. Commun. 189, 455- 464.

John,J., Frech,M., and Wittinghofer,A. (1988). Biochemical properties of Ha-ras encoded p21 mutants and mechanism of the autophosphorylation reaction. J Biol. Chem. 263, 11792- 11799.

Jordan,J.D., Carey,K.D., Stork,P.J., and Iyengar,R. (1999). Modulation of rap activity by direct interaction of Galpha(o) with Rap1 GTPase-activating protein. J Biol. Chem. 274, 21507-21510.

Kandt,R.S. (2003). Tuberous sclerosis complex and neurofibromatosis type 1: the two most common neurocutaneous diseases. Neurol. Clin. 21, 983-1004.

Kao,S., Jaiswal,R.K., Kolch,W., and Landreth,G.E. (2001). Identification of the mechanisms regulating the differential activation of the mapk cascade by epidermal growth factor and nerve growth factor in PC12 cells. J Biol. Chem. 2001. May. 25. ;276. (21):18169. -77. Epub. 2001. Mar. 13. 276, 18169-18177.

Katagiri,K., Hattori,M., Minato,N., Irie,S., Takatsu,K., and Kinashi,T. (2000). Rap1 is a potent activation signal for leukocyte function-associated antigen 1 distinct from protein kinase C and phosphatidylinositol-3-OH kinase. Mol Cell Biol. 2000. Mar. ;20. (6. ):1956. -69. 20, 1956-1969.

Katagiri,K., Hattori,M., Minato,N., and Kinashi,T. (2002). Rap1 functions as a key regulator of T-cell and antigen-presenting cell interactions and modulates T-cell responses. Mol Cell Biol. 2002. Feb. ;22. (4):1001. -15. 22, 1001-1015.

Katagiri,K., Ohnishi,N., Kabashima,K., Iyoda,T., Takeda,N., Shinkai,Y., Inaba,K., and Kinashi,T. (2004). Crucial functions of the Rap1 effector molecule RAPL in lymphocyte and dendritic cell trafficking. Nat. Immunol. 5, 1045-1051.

Kawasaki,H., Springett,G.M., Toki,S., Canales,J.J., Harlan,P., Blumenstiel,J.P., Chen,E.J., Bany,I.A., Mochizuki,N., Ashbacher,A., Matsuda,M., Housman,D.E., and Graybiel,A.M. (1998). A Rap guanine nucleotide exchange factor enriched highly in the basal ganglia. Proc. Natl. Acad. Sci. U. S. A 95, 13278-13283.

Keenan,R.J., Freymann,D.M., Stroud,R.M., and Walter,P. (2001). The signal recognition particle. Annu. Rev. Biochem. 70, 755-775.

Kim,D.H., Sarbassov,D.D., Ali,S.M., King,J.E., Latek,R.R., Erdjument-Bromage,H., Tempst,P., and Sabatini,D.M. (2002). mTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machinery. Cell 110, 163-175.

108 References

Kitayama,H., Sugimoto,Y., Matsuzaki,T., Ikawa,Y., and Noda,M. (1989). A ras-related gene with transformation suppressor activity. Cell 56, 77-84.

Kjeldgaard,M., Nissen,P., Thirup,S., and Nyborg,J. (1993). The crystal structure of elongation factor EF-Tu from Thermus aquaticus in the GTP conformation. Structure. 1, 35-50.

Kjeldgaard,M. and Nyborg,J. (1992). Refined structure of elongation factor EF-Tu from Escherichia coli. J. Mol. Biol. 223, 721-742.

Knox,A.L. and Brown,N.H. (2002). Rap1 GTPase regulation of adherens junction positioning and cell adhesion. Science 2002. Feb. 15. ;295. (5558. ):1285. -8. 295, 1285-1288.

Kobe,B. and Deisenhofer,J. (1995). Proteins with leucine-rich repeats. Curr. Opin. Struct. Biol. 5, 409-416.

Kötting,C. and Gerwert,K. (2005). Monitoring Protein-Protein Interactions by Time-resolved FTIR Difference Spectroscopy. In Protein-Protein Interactions, Cold Spring Harbour Laboratory Press, Cold Spring Harbour, NY.), pp. 1-43.

Kraemer,A., Brinkmann,T., Plettner,I., Goody,R., and Wittinghofer,A. (2002). Fluorescently labelled guanine nucleotide binding proteins to analyse elementary steps of GAP-catalysed reactions. J Mol Biol. 2002. Dec. 6. ;324. (4):763. -74. 324, 763-774.

Krall,R., Schmidt,G., Aktories,K., and Barbieri,J.T. (2000). Pseudomonas aeruginosa ExoT is a Rho GTPase-activating protein. Infect. Immun. 68, 6066-6068.

Krengel,U., Schlichting,L., Scherer,A., Schumann,R., Frech,M., John,J., Kabsch,W., Pai,E.F., and Wittinghofer,A. (1990). Three-dimensional structures of H-ras p21 mutants: molecular basis for their inability to function as signal switch molecules. Cell 62, 539-548.

Krymskaya,V.P. (2003). Tumour suppressors hamartin and tuberin: intracellular signalling. Cell Signal. 15, 729-739.

Kuhlmann,J., Macara,I., and Wittinghofer,A. (1997). Dynamic and equilibrium studies on the interaction of Ran with its effector, RanBP1. Biochemistry 36, 12027-12035.

Kukimoto-Niino,M., Murayama,K., Inoue,M., Terada,T., Tame,J.R., Kuramitsu,S., Shirouzu,M., and Yokoyama,S. (2004). Crystal structure of the GTP-binding protein Obg from Thermus thermophilus HB8. J. Mol. Biol. 337, 761-770.

Kurachi,H., Wada,Y., Tsukamoto,N., Maeda,M., Kubota,H., Hattori,M., Iwai,K., and Minato,N. (1997). Human SPA-1 gene product selectively expressed in lymphoid tissues is a specific GTPase-activating protein for Rap1 and Rap2. Segregate expression profiles from a rap1GAP gene product. J. Biol. Chem. 272, 28081-28088.

Laemmli,U.K. (1970). Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680-685.

Lahiri,S.D., Zhang,G., Dunaway-Mariano,D., and Allen,K.N. (2003). The pentacovalent phosphorus intermediate of a phosphoryl transfer reaction. Science 299, 2067-2071.

Leipe,D.D., Wolf,Y.I., Koonin,E.V., and Aravind,L. (2002). Classification and evolution of P- loop GTPases and related ATPases. J. Mol. Biol. 317, 41-72.

Leonard,D.A., Lin,R., Cerione,R.A., and Manor,D. (1998). Biochemical studies of the mechanism of action of the Cdc42-GTPase-activating protein. J. Biol. Chem. 273, 16210- 16215. 109 References

Levy,H.M. and Koshland,D.E. (1959). Mechanism of hydrolysis of adenosinetriphosphate by muscle proteins and its relation to muscular contraction. J. Biol. Chem. 234, 1102-1107.

Li,Y., Corradetti,M.N., Inoki,K., and Guan,K.L. (2004). TSC2: filling the GAP in the mTOR signaling pathway. Trends Biochem. Sci. 29, 32-38.

Liljas,A., Aevarsson,A., al Karadaghi,S., Garber,M., Zheltonosova,J., and Brazhnikov,E. (1995). Crystallographic studies of elongation factor G. Biochem. Cell Biol. 73, 1209-1216.

Liljas,A. and al Karadaghi,S. (1997). Structural aspects of protein synthesis. Nat. Struct. Biol. 4, 767-771.

Litvak,Y. and Selinger,Z. (2003). Bacterial mimics of eukaryotic GTPase-activating proteins (GAPs). Trends Biochem. Sci. 28, 628-631.

Liu,X.M., Sonar,S., Lee,C.P., Coleman,M., RajBhandary,U.L., and Rothschild,K.J. (1995). Site-directed isotope labeling and FTIR spectroscopy: assignment of tyrosine bands in the bR-->M difference spectrum of bacteriorhodopsin. Biophys. Chem. 56, 63-70.

Lizcano,J.M., Goransson,O., Toth,R., Deak,M., Morrice,N.A., Boudeau,J., Hawley,S.A., Udd,L., Makela,T.P., Hardie,D.G., and Alessi,D.R. (2004). LKB1 is a master kinase that activates 13 kinases of the AMPK subfamily, including MARK/PAR-1. EMBO J. 23, 833-843.

Lockyer,P.J., Kupzig,S., and Cullen,P.J. (2001). CAPRI regulates Ca(2+)-dependent inactivation of the Ras-MAPK pathway. Curr. Biol. 11, 981-986.

Maegley,K.A., Admiraal,S.J., and Herschlag,D. (1996). Ras-catalyzed hydrolysis of GTP: a new perspective from model studies. Proc. Natl. Acad. Sci. U. S. A 93, 8160-8166.

Malumbres,M. and Barbacid,M. (2003). RAS oncogenes: the first 30 years. Nat. Rev. Cancer 3, 459-465.

Mandiyan,V., Andreev,J., Schlessinger,J., and Hubbard,S.R. (1999). Crystal structure of the ARF-GAP domain and ankyrin repeats of PYK2-associated protein beta. EMBO J. 18, 6890- 6898.

Manning,B.D. and Cantley,L.C. (2003). Rheb fills a GAP between TSC and TOR. Trends Biochem. Sci. 28, 573-576.

Manning,B.D., Tee,A.R., Logsdon,M.N., Blenis,J., and Cantley,L.C. (2002). Identification of the tuberous sclerosis complex-2 tumor suppressor gene product tuberin as a target of the phosphoinositide 3-kinase/akt pathway. Mol. Cell 10, 151-162.

Maruta,H., Holden,J., Sizeland,A., and D'Abaco,G. (1991). The residues of Ras and Rap proteins that determine their GAP specificities. J Biol. Chem. 266, 11661-11668.

McCormick,F. (2001). Cancer gene therapy: fringe or cutting edge? Nat. Rev. Cancer 1, 130- 141.

McCormick,F. (2003). Cancer-specific viruses and the development of ONYX-015. Cancer Biol. Ther. 2, S157-S160.

McCray,J.A., Herbette,L., Kihara,T., and Trentham,D.R. (1980). A new approach to time- resolved studies of ATP-requiring biological systems; laser flash photolysis of caged ATP. Proc. Natl. Acad. Sci. U. S. A 77, 7237-7241.

110 References

McLeod,S.J., Ingham,R.J., Bos,J.L., Kurosaki,T., and Gold,M.R. (1998). Activation of the Rap1 GTPase by the B cell antigen receptor. J. Biol. Chem. 273, 29218-29223.

Meng,J. and Casey,P.J. (2002). Activation of Gz attenuates Rap1-mediated differentiation of PC12 cells. J Biol. Chem. 2002. Nov. 8. ;277. (45. ):43417. -24. Epub. 2002. Aug. 26. 277, 43417-43424.

Meng,J., Glick,J.L., Polakis,P., and Casey,P.J. (1999). Functional interaction between Galpha(z) and Rap1GAP suggests a novel form of cellular cross-talk. J Biol. Chem. 274, 36663-36669.

Meyer,G. and Brose,N. (2003). Neuroscience. SPARring with spines. Science 2003 Nov. 21;302. (5649. ):1341. -4. 302, 1341-1344.

Miller,R.W., Eady,R.R., Fairhurst,S.A., Gormal,C.A., and Smith,B.E. (2001). Transition state complexes of the Klebsiella pneumoniae nitrogenase proteins. Spectroscopic properties of aluminium fluoride-stabilized and beryllium fluoride-stabilized MgADP complexes reveal conformational differences of the Fe protein. Eur. J. Biochem. 268, 809-818.

Minagawa,T., Fukuda,M., and Mikoshiba,K. (2001). Distinct phosphoinositide binding specificity of the GAP1 family proteins: characterization of the pleckstrin homology domains of MRASAL and KIAA0538. Biochem. Biophys. Res. Commun. 288, 87-90.

Mitra,R.S., Zhang,Z., Henson,B.S., Kurnit,D.M., Carey,T.E., and D'Silva,N.J. (2003). Rap1A and rap1B ras-family proteins are prominently expressed in the nucleus of squamous carcinomas: nuclear translocation of GTP-bound active form. Oncogene 2003 Sep. 18. ;22. (40. ):6243. -56. 22, 6243-6256.

Mittal,R., Ahmadian,M.R., Goody,R.S., and Wittinghofer,A. (1996). Formation of a transition- state analog of the Ras GTPase reaction by Ras-GDP, tetrafluoroaluminate, and GTPase- activating proteins. Science 273, 115-117.

Mochizuki,N., Ohba,Y., Kiyokawa,E., Kurata,T., Murakami,T., Ozaki,T., Kitabatake,A., Nagashima,K., and Matsuda,M. (1999). Activation of the ERK/MAPK pathway by an isoform of rap1GAP associated with G alpha(i). Nature 400, 891-894.

Momose,S., Kobayashi,T., Mitani,H., Hirabayashi,M., Ito,K., Ueda,M., Nabeshima,Y., and Hino,O. (2002). Identification of the coding sequences responsible for Tsc2-mediated tumor suppression using a transgenic rat system. Hum. Mol. Genet. 11, 2997-3006.

Morozov,A., Muzzio,I.A., Bourtchouladze,R., Van Strien,N., Lapidus,K., Yin,D., Winder,D.G., Adams,J.P., Sweatt,J.D., and Kandel,E.R. (2003). Rap1 couples cAMP signaling to a distinct pool of p42/44MAPK regulating excitability, synaptic plasticity, learning, and memory. Neuron 2003 Jul. 17;39. (2):309. -25. 39, 309-325.

Nassar,N., Horn,G., Herrmann,C., Block,C., Janknecht,R., and Wittinghofer,A. (1996). Ras/Rap effector specificity determined by charge reversal. Nat Struct. Biol. 3, 723-729.

Nellist,M., van Slegtenhorst,M.A., Goedbloed,M., van den Ouweland,A.M., Halley,D.J., and van der,S.P. (1999). Characterization of the cytosolic tuberin-hamartin complex. Tuberin is a cytosolic chaperone for hamartin. J. Biol. Chem. 274, 35647-35652.

Nixon,A.E., Brune,M., Lowe,P.N., and Webb,M.R. (1995). Kinetics of inorganic phosphate release during the interaction of p21ras with the GTPase-activating proteins, p120-GAP and neurofibromin. Biochemistry 34, 15592-15598.

111 References

Noel,J.P., Hamm,H.E., and Sigler,P.B. (1993). The 2.2 A crystal structure of transducin-alpha complexed with GTP gamma S. Nature 366, 654-663.

Ogle,J.M., Carter,A.P., and Ramakrishnan,V. (2003). Insights into the decoding mechanism from recent ribosome structures. Trends Biochem. Sci. 28, 259-266.

Ohtsuka,T., Shimizu,K., Yamamori,B., Kuroda,S., and Takai,Y. (1996). Activation of brain B- Raf protein kinase by Rap1B small GTP-binding protein. J Biol. Chem. 271, 1258-1261.

Ostermann,N., Schlichting,I., Brundiers,R., Konrad,M., Reinstein,J., Veit,T., Goody,R.S., and Lavie,A. (2000). Insights into the phosphoryltransfer mechanism of human thymidylate kinase gained from crystal structures of enzyme complexes along the reaction coordinate. Structure. Fold. Des 8, 629-642.

Pacold,M.E., Suire,S., Perisic,O., Lara-Gonzalez,S., Davis,C.T., Walker,E.H., Hawkins,P.T., Stephens,L., Eccleston,J.F., and Williams,R.L. (2000). Crystal structure and functional analysis of Ras binding to its effector phosphoinositide 3-kinase gamma. Cell 103, 931-943.

Pai,E.F., Krengel,U., Petsko,G.A., Goody,R.S., Kabsch,W., and Wittinghofer,A. (1990). Refined crystal structure of the triphosphate conformation of H-ras p21 at 1.35 A resolution: implications for the mechanism of GTP hydrolysis. EMBO J 9, 2351-2359.

Pak,D.T. and Sheng,M. (2003). Targeted protein degradation and synapse remodeling by an inducible protein kinase. Science 2003 Nov. 21;302. (5649. ):1368. -73. Epub. 2003 Oct. 23. 302, 1368-1373.

Pak,D.T., Yang,S., Rudolph-Correia,S., Kim,E., and Sheng,M. (2001). Regulation of dendritic spine morphology by SPAR, a PSD-95-associated RapGAP. Neuron 2001. Aug. 2;31. (2):289. -303. 31, 289-303.

Park,H.O., Kang,P.J., and Rachfal,A.W. (2002). Localization of the Rsr1/Bud1 GTPase involved in selection of a proper growth site in yeast. J Biol. Chem. 2002. Jul. 26. ;277. (30. ):26721. -4. Epub. 2002. Jun. 10. 277, 26721-26724.

Pasqualato,S., Renault,L., and Cherfils,J. (2002). Arf, Arl, Arp and Sar proteins: a family of GTP-binding proteins with a structural device for 'front-back' communication. EMBO Rep. 3, 1035-1041.

Pfeffer,S.R. (2001). Rab GTPases: specifying and deciphering organelle identity and function. Trends Cell Biol. 11, 487-491.

Phillips,R.A., Hunter,J.L., Eccleston,J.F., and Webb,M.R. (2003). The mechanism of Ras GTPase activation by neurofibromin. Biochemistry 42, 3956-3965.

Pizon,V., Chardin,P., Lerosey,I., Olofsson,B., and Tavitian,A. (1988). Human cDNAs rap1 and rap2 homologous to the Drosophila gene Dras3 encode proteins closely related to ras in the 'effector' region. Oncogene 3, 201-204.

Pizon,V., Desjardins,M., Bucci,C., Parton,R.G., and Zerial,M. (1994). Association of Rap1a and Rap1b proteins with late endocytic/phagocytic compartments and Rap2a with the Golgi complex. J Cell Sci. 107 ( Pt 6), 1661-1670.

Polakis,P. and McCormick,F. (1993). Structural requirements for the interaction of p21ras with GAP, exchange factors, and its biological effector target. J. Biol. Chem. 268, 9157-9160.

112 References

Polakis,P., Rubinfeld,B., and McCormick,F. (1992). Phosphorylation of rap1GAP in vivo and by cAMP-dependent kinase and the cell cycle p34cdc2 kinase in vitro. J Biol. Chem. 267, 10780-10785.

Polakis,P.G., Rubinfeld,B., Evans,T., and McCormick,F. (1991). Purification of a plasma membrane-associated GTPase-activating protein specific for rap1/Krev-1 from HL60 cells. Proc. Natl. Acad. Sci. U. S. A 88, 239-243.

Posern,G., Weber,C.K., Rapp,U.R., and Feller,S.M. (1998). Activity of Rap1 is regulated by bombesin, cell adhesion, and cell density in NIH3T3 fibroblasts. J. Biol. Chem. 273, 24297- 24300.

Potter,C.J., Pedraza,L.G., and Xu,T. (2002). Akt regulates growth by directly phosphorylating Tsc2. Nat Cell Biol. 2002. Sep. ;4(9. ):658. -65. 4, 658-665.

Powers,T. and Walter,P. (1999). Regulation of ribosome biogenesis by the rapamycin- sensitive TOR-signaling pathway in Saccharomyces cerevisiae. Mol. Biol. Cell 10, 987-1000.

Prakash,B., Praefcke,G.J., Renault,L., Wittinghofer,A., and Herrmann,C. (2000). Structure of human guanylate-binding protein 1 representing a unique class of GTP-binding proteins. Nature 403, 567-571.

Proud,C.G. (2004). mTOR-mediated regulation of translation factors by amino acids. Biochem. Biophys. Res. Commun. 313, 429-436.

Rayment,I. (1996). The structural basis of the myosin ATPase activity. J. Biol. Chem. 271, 15850-15853.

Reedquist,K.A., Ross,E., Koop,E.A., Wolthuis,R.M., Zwartkruis,F.J., van Kooyk,Y., Salmon,M., Buckley,C.D., and Bos,J.L. (2000). The small GTPase, Rap1, mediates CD31- induced integrin adhesion. J Cell Biol. 2000. Mar. 20. ;148. (6. ):1151. -8. 148, 1151-1158.

Renault,L., Guibert,B., and Cherfils,J. (2003). Structural snapshots of the mechanism and inhibition of a guanine nucleotide exchange factor. Nature 426, 525-530.

Reuther,G.W. and Der,C.J. (2000). The Ras branch of small GTPases: Ras family members don't fall far from the tree. Curr. Opin. Cell Biol. 12, 157-165.

Rittinger,K., Taylor,W.R., Smerdon,S.J., and Gamblin,S.J. (1998). Support for shared ancestry of GAPs. Nature 392, 448-449.

Rittinger,K., Walker,P.A., Eccleston,J.F., Nurmahomed,K., Owen,D., Laue,E., Gamblin,S.J., and Smerdon,S.J. (1997a). Crystal structure of a small G protein in complex with the GTPase-activating protein rhoGAP. Nature 388, 693-697.

Rittinger,K., Walker,P.A., Eccleston,J.F., Smerdon,S.J., and Gamblin,S.J. (1997b). Structure at 1.65 A of RhoA and its GTPase-activating protein in complex with a transition-state analogue. Nature 389, 758-762.

Rodnina,M.V. and Wintermeyer,W. (2001). Fidelity of aminoacyl-tRNA selection on the ribosome: kinetic and structural mechanisms. Annu. Rev. Biochem. 70:415-35., 415-435.

Rubinfeld,B., Crosier,W.J., Albert,I., Conroy,L., Clark,R., McCormick,F., and Polakis,P. (1992). Localization of the rap1GAP catalytic domain and sites of phosphorylation by mutational analysis. Mol Cell Biol. 12, 4634-4642.

113 References

Rubinfeld,B., Munemitsu,S., Clark,R., Conroy,L., Watt,K., Crosier,W.J., McCormick,F., and Polakis,P. (1991). Molecular cloning of a GTPase activating protein specific for the Krev-1 protein p21rap1. Cell 65, 1033-1042.

Ruckebusch,C., Duponchel,L., Sombret,B., Huvenne,J.P., and Saurina,J. (2003). Time- resolved step-scan FT-IR spectroscopy: focus on multivariate curve resolution. J. Chem. Inf. Comput. Sci. 43, 1966-1973.

Sabatini,D.M., Erdjument-Bromage,H., Lui,M., Tempst,P., and Snyder,S.H. (1994). RAFT1: a mammalian protein that binds to FKBP12 in a rapamycin-dependent fashion and is homologous to yeast TORs. Cell 78, 35-43.

Sahoo,T., Johnson,E.W., Thomas,J.W., Kuehl,P.M., Jones,T.L., Dokken,C.G., Touchman,J.W., Gallione,C.J., Lee-Lin,S.Q., Kosofsky,B., Kurth,J.H., Louis,D.N., Mettler,G., Morrison,L., Gil-Nagel,A., Rich,S.S., Zabramski,J.M., Boguski,M.S., Green,E.D., and Marchuk,D.A. (1999). Mutations in the gene encoding KRIT1, a Krev-1/rap1a binding protein, cause cerebral cavernous malformations (CCM1). Hum. Mol Genet. 8, 2325-2333.

Sambrook, J. Fritsch E. F. Maniatis T. Molecular Cloning. A Laboratory Manual. 1989. Cold Spring Harbor Laboratory Press. Ref Type: Generic

Sanger,F., Nicklen,S., and Coulson,A.R. (1992). DNA sequencing with chain-terminating inhibitors. 1977. Biotechnology 24, 104-108.

Sasaki,N. and Sutoh,K. (1998). Structure-mutation analysis of the ATPase site of Dictyostelium discoideum myosin II. Adv. Biophys. 35:1-24., 1-24.

Saucedo,L.J., Gao,X., Chiarelli,D.A., Li,L., Pan,D., and Edgar,B.A. (2003). Rheb promotes cell growth as a component of the insulin/TOR signalling network. Nat Cell Biol. 2003 Jun. ;5(6. ):566. -71. 5, 566-571.

Scheffzek,K., Ahmadian,M.R., Kabsch,W., Wiesmuller,L., Lautwein,A., Schmitz,F., and Wittinghofer,A. (1997). The Ras-RasGAP complex: structural basis for GTPase activation and its loss in oncogenic Ras mutants. Science 277, 333-338.

Scheffzek,K., Ahmadian,M.R., and Wittinghofer,A. (1998). GTPase-activating proteins: helping hands to complement an active site. Trends Biochem. Sci. 23, 257-262.

Scheffzek,K., Lautwein,A., Kabsch,W., Ahmadian,M.R., and Wittinghofer,A. (1996). Crystal structure of the GTPase-activating domain of human p120GAP and implications for the interaction with Ras. Nature 384, 591-596.

Schindelin,H., Kisker,C., Schlessman,J.L., Howard,J.B., and Rees,D.C. (1997). Structure of ADP x AIF4(-)-stabilized nitrogenase complex and its implications for signal transduction. Nature 387, 370-376.

Schweins,T., Geyer,M., Scheffzek,K., Warshel,A., Kalbitzer,H.R., and Wittinghofer,A. (1995). Substrate-assisted catalysis as a mechanism for GTP hydrolysis of p21ras and other GTP- binding proteins. Nat Struct. Biol. 2, 36-44.

Schweins,T., Scheffzek,K., Assheuer,R., and Wittinghofer,A. (1997). The role of the metal ion in the p21ras catalysed GTP-hydrolysis: Mn2+ versus Mg2+. J Mol Biol. 266, 847-856.

114 References

Sebzda,E., Bracke,M., Tugal,T., Hogg,N., and Cantrell,D.A. (2002). Rap1A positively regulates T cells via integrin activation rather than inhibiting lymphocyte signaling. Nat Immunol 2002. Mar. ;3(3):251. -8. 3, 251-258.

Seewald,M.J., Korner,C., Wittinghofer,A., and Vetter,I.R. (2002). RanGAP mediates GTP hydrolysis without an arginine finger. Nature 2002. Feb. 7. ;415. (6872. ):662. -6. 415, 662- 666.

Seewald,M.J., Kraemer,A., Farkasovsky,M., Korner,C., Wittinghofer,A., and Vetter,I.R. (2003). Biochemical characterization of the Ran-RanBP1-RanGAP system: are RanBP proteins and the acidic tail of RanGAP required for the Ran-RanGAP GTPase reaction? Mol Cell Biol. 2003 Nov. ;23(22. ):8124. -36. 23, 8124-8136.

Shaw,R.J., Kosmatka,M., Bardeesy,N., Hurley,R.L., Witters,L.A., DePinho,R.A., and Cantley,L.C. (2004). The tumor suppressor LKB1 kinase directly activates AMP-activated kinase and regulates apoptosis in response to energy stress. Proc. Natl. Acad. Sci. U. S. A 101, 3329-3335.

Shirouzu,M., Morinaka,K., Koyama,S., Hu,C.D., Hori-Tamura,N., Okada,T., Kariya,K., Kataoka,T., Kikuchi,A., and Yokoyama,S. (1998). Interactions of the amino acid residue at position 31 of the c-Ha-Ras protein with Raf-1 and RalGDS. J Biol. Chem. 273, 7737-7742.

Singh,L., Gao,Q., Kumar,A., Gotoh,T., Wazer,D.E., Band,H., Feig,L.A., and Band,V. (2003). The high-risk human papillomavirus type 16 E6 counters the GAP function of E6TP1 toward small Rap G proteins. J. Virol. 77, 1614-1620.

Smith,C.A. and Rayment,I. (1996). X-ray structure of the magnesium(II).ADP.vanadate complex of the Dictyostelium discoideum myosin motor domain to 1.9 A resolution. Biochemistry 35, 5404-5417.

Sonar,S., Lee,C.P., Coleman,M., Patel,N., Liu,X., Marti,T., Khorana,H.G., RajBhandary,U.L., and Rothschild,K.J. (1994). Site-directed isotope labelling and FTIR spectroscopy of bacteriorhodopsin. Nat. Struct. Biol. 1, 512-517.

Sonenberg,N. and Dever,T.E. (2003). Eukaryotic translation initiation factors and regulators. Curr. Opin. Struct. Biol. 13, 56-63.

Spang,A. (2002). ARF1 regulatory factors and COPI vesicle formation. Curr. Opin. Cell Biol. 2002. Aug. ;14(4):423. -7. 14, 423-427.

Sprang,S.R. (1997). G protein mechanisms: insights from structural analysis. Annu. Rev. Biochem. 66, 639-678.

Stebbins,C.E. and Galan,J.E. (2001). Structural mimicry in bacterial virulence. Nature 412, 701-705.

Sternweis,P.C. and Gilman,A.G. (1982). Aluminum: a requirement for activation of the regulatory component of adenylate cyclase by fluoride. Proc. Natl. Acad. Sci. U. S. A 79, 4888-4891.

Stocker,H., Radimerski,T., Schindelholz,B., Wittwer,F., Belawat,P., Daram,P., Breuer,S., Thomas,G., and Hafen,E. (2003). Rheb is an essential regulator of S6K in controlling cell growth in Drosophila. Nat Cell Biol. 2003 Jun. ;5(6. ):559. -65. 5, 559-565.

115 References

Strater,N., Klabunde,T., Tucker,P., Witzel,H., and Krebs,B. (1995). Crystal structure of a purple acid phosphatase containing a dinuclear Fe(III)-Zn(II) active site. Science 268, 1489- 1492.

Takai,Y., Sasaki,T., and Matozaki,T. (2001). Small GTP-binding proteins. Physiol Rev. 81, 153-208.

Tee,A.R., Manning,B.D., Roux,P.P., Cantley,L.C., and Blenis,J. (2003). Tuberous sclerosis complex gene products, Tuberin and Hamartin, control mTOR signaling by acting as a GTPase-activating protein complex toward Rheb. Curr. Biol. 13, 1259-1268.

Tesmer,J.J., Berman,D.M., Gilman,A.G., and Sprang,S.R. (1997). Structure of RGS4 bound to AlF4--activated G(i alpha1): stabilization of the transition state for GTP hydrolysis. Cell 89, 251-261.

Trahey,M. and McCormick,F. (1987). A cytoplasmic protein stimulates normal N-ras p21 GTPase, but does not affect oncogenic mutants. Science 238, 542-545.

Tuberous Sclerosis Consortium (1993). Identification and characterization of the tuberous sclerosis gene on chromosome 16. The European Chromosome 16 Tuberous Sclerosis Consortium. Cell 75, 1305-1315.

Tucker,J., Sczakiel,G., Feuerstein,J., John,J., Goody,R.S., and Wittinghofer,A. (1986). Expression of p21 proteins in Escherichia coli and stereochemistry of the nucleotide-binding site. EMBO J 5, 1351-1358.

Valencia,A., Chardin,P., Wittinghofer,A., and Sander,C. (1991). The ras : evolutionary tree and role of conserved amino acids. Biochemistry 30, 4637-4648. van Slegtenhorst,M., Nellist,M., Nagelkerken,B., Cheadle,J., Snell,R., van den,O.A., Reuser,A., Sampson,J., Halley,D., and van der,S.P. (1998). Interaction between hamartin and tuberin, the TSC1 and TSC2 gene products. Hum. Mol Genet. 7, 1053-1057.

Vetter,I.R. and Wittinghofer,A. (2001). The guanine nucleotide-binding switch in three dimensions. Science 294, 1299-1304.

Vivanco,I. and Sawyers,C.L. (2002). The phosphatidylinositol 3-Kinase AKT pathway in human cancer. Nat. Rev. Cancer 2, 489-501.

Volarevic,S., Stewart,M.J., Ledermann,B., Zilberman,F., Terracciano,L., Montini,E., Grompe,M., Kozma,S.C., and Thomas,G. (2000). Proliferation, but not growth, blocked by conditional deletion of 40S ribosomal protein S6. Science 288, 2045-2047.

Vossler,M.R., Yao,H., York,R.D., Pan,M.G., Rim,C.S., and Stork,P.J. (1997). cAMP activates MAP kinase and Elk-1 through a B-Raf- and Rap1-dependent pathway. Cell 89, 73-82.

Walker,S.A., Kupzig,S., Lockyer,P.J., Bilu,S., Zharhary,D., and Cullen,P.J. (2002). Analyzing the role of the putative inositol 1,3,4,5-tetrakisphosphate receptor GAP1IP4BP in intracellular Ca2+ homeostasis. J. Biol. Chem. 277, 48779-48785.

Wang,W. and Hochstrasser,R.M. (2004). Characteristics of the two dimensional infrared spectroscopy of helices from approximate simulations and analytical models. Chemical Physics 297, 195-219.

Weiss,B., Bollag,G., and Shannon,K. (1999). Hyperactive Ras as a therapeutic target in neurofibromatosis type 1. Am. J. Med. Genet. 89, 14-22.

116 References

Westheimer,F.H. (1987). Why nature chose phosphates. Science 235, 1173-1178.

Wienecke,R., Konig,A., and DeClue,J.E. (1995). Identification of tuberin, the tuberous sclerosis-2 product. Tuberin possesses specific Rap1GAP activity. J Biol. Chem. 270, 16409- 16414.

Willard,F.S., Kimple,R.J., and Siderovski,D.P. (2004). Return of the GDI: the GoLoco motif in cell division. Annu. Rev. Biochem. 73, 925-951.

Wittinghofer,A. and Waldmann,H. (2000). Ras - A molecular switch involved in tumor formation [Review]. Angewandte Chemie-International Edition 39, 4193-4214.

Woods,A., Johnstone,S.R., Dickerson,K., Leiper,F.C., Fryer,L.G., Neumann,D., Schlattner,U., Wallimann,T., Carlson,M., and Carling,D. (2003). LKB1 is the upstream kinase in the AMP- activated protein kinase cascade. Curr. Biol. 13, 2004-2008.

Wu,S.K., Zeng,K., Wilson,I.A., and Balch,W.E. (1996). Structural insights into the function of the Rab GDI superfamily. Trends Biochem. Sci. 21, 472-476.

Wurtele,M., Wolf,E., Pederson,K.J., Buchwald,G., Ahmadian,M.R., Barbieri,J.T., and Wittinghofer,A. (2001). How the Pseudomonas aeruginosa ExoS toxin downregulates Rac. Nat Struct. Biol. 2001. Jan. ;8. (1):23-6. 8, 23-26.

Xiao,G.H., Shoarinejad,F., Jin,F., Golemis,E.A., and Yeung,R.S. (1997). The tuberous sclerosis 2 gene product, tuberin, functions as a Rab5 GTPase activating protein (GAP) in modulating endocytosis. J Biol. Chem. 272, 6097-6100.

Xu,X., Wang,Y., Barry,D.C., Chanock,S.J., and Bokoch,G.M. (1997). Guanine nucleotide binding properties of Rac2 mutant proteins and analysis of the responsiveness to guanine nucleotide dissociation stimulator. Biochemistry 36, 626-632.

Zhang,Y., Gao,X., Saucedo,L.J., Ru,B., Edgar,B.A., and Pan,D. (2003). Rheb is a direct target of the tuberous sclerosis tumour suppressor proteins. Nat. Cell Biol. 5, 578-581.

Zheng,X.F., Florentino,D., Chen,J., Crabtree,G.R., and Schreiber,S.L. (1995). TOR kinase domains are required for two distinct functions, only one of which is inhibited by rapamycin. Cell 82, 121-130.

Zwartkruis,F.J., Wolthuis,R.M., Nabben,N.M., Franke,B., and Bos,J.L. (1998). Extracellular signal-regulated activation of Rap1 fails to interfere in Ras effector signalling. EMBO J 17, 5905-5912.

117 Appendix

8 Appendix

8.1 Abbreviations

Å Ångström (0.1 nm)

DTE 1,4-Dithioerythritol

EDTA Ehtylendiamintetraacetate

E. coli Escherichia coli

GAP GTPase-activating Protein

GDP, GTP Guanosine-5’-diphosphate, Guanosine-5’-triphosphate

GEF guanine nucleotide exchange factor

GNBP Guanine nucleotide binding Protein

GppNHp Guanosine-5’-(β,γ-imido)-triphosphate

GSH Reduced glutathione

GST Glutathione-S-

GTPase GTP-hydrolysyzing Enzyme

HEPES 4-2-hydroxyethyl-1-piperazineethanesulfonic acid

HPLC High performance liquid chromatography

IAEDANS 5-((((2-iodoacetyl) amino) ethyl) amino) naphthalin-1-sulfonic acid

Ins(1,3,4,5)P4 inositol 1,3,4,5-tetrakisphosphate

IPTG Isopropyl-β-D-1-thiogalactopyranosid

λ Wavelength

LRR leucine-rich repeat

MES 2-Morpholinoethanesulfonic acid

OD600 Optical density at 600 nm

ONPG O-Nitrophenyl-D-galactoside

PEG # Polyethylenglycol (Molecularweight of # Da) pF -log[F−]

Pi Orthophosphate

118 Appendix

PMSF Phenylmethylsulfonylfluoride

SDS-PAGE Sodiumdodecylsulfate-Polyacrylamide gelelectrophoresis

SIMBI signal recognition particle, MinD and BioD

THAP 2,4,6-trihydroxyacetophenone

Tris Tris-(hydroxymethyl)-aminomethan

TRAFAC Translation factor related

For amino acids, the one and three letter code was used A Ala alanine I Ile isoleucine R Arg arginine C Cys cysteine K Lys lysine S Ser serine D Asp aspartate L Leu leucine T Thr threonine E Glu glutamate M Met methionine V Val valine F Phe phenylalanine N Asn asparagine W Trp tryptophane G Gly glycine P Pro proline Y Tyr tyrosine H His histidine Q Gln glutamine x any amino acid

8.2 Mass spectra of Aedans-Rap (A86C) and mutants

Rap1B (1-167), A86C; expected mass = 19342 D wt_col #276 RT: 0.00 NL: 6.64E6 F: + p ESI Full ms [ 700.00-2000.00] 19343.0 100

95 90

85 80

75 70 65 60

55 50

45 40 Relative Abundance 35 30

25 20

15 10 19647.0 19036.0 23273.0 25108.0 25853.0 27368.0 28554.0 29112.0 5 16180.018091.0 20767.0 22621.0 0 16000 18000 20000 22000 24000 26000 28000 mass

119 Appendix

Rap1B (1-167), A86C, G12V; expected mass = 19384 D

G12V_col #706 RT: 15.34 NL: 1.28E8 F: + p ESI Full ms [ 700.00-2000.00] 19384.0 100

95 90

85 80

75 70

65 60

55 50

45 40 Relative Abundance Relative 35 30

25 20 15 10 16374.0 19353.0 15716.0 17157.0 18534.0 19643.0 21833.0 23477.0 24321.0 5 19264.0 20602.0 22575.0 0 15000 16000 17000 18000 19000 20000 21000 22000 23000 24000 mass

Rap1B (1-167), A86C, T61A; expected mass = 19312 D

T61A_col #706 RT: 15.39 NL: 7.97E7 F: + p ESI Full ms [ 700.00-2000.00] 19312.0 100 95 90 85 80 75 70 65 60 55 50

45 40 Relative Abundance 35 30 25 20 15

10 19619.0 18443.0 20834.0 21964.0 22584.0 24039.0 24736.0 5 15554.0 16312.0 17829.0 20319.0 0 15000 16000 17000 18000 19000 20000 21000 22000 23000 24000 mass

120 Appendix

Rap1B (1-167), A86C, T61L; expected mass = 19354 D

T61L_col #716 RT: 15.53 NL: 1.48E8 F: + p ESI Full ms [ 700.00-2000.00] 19354.0 100 95 90

85 80 75 70

65 60 55 50 45 40 Relative Abundance Relative 35 30 25 20

15 10 19425.0 23603.0 24286.0 5 15410.0 16373.0 17115.0 18299.0 21189.0 21636.0 23115.0 0 15000 16000 17000 18000 19000 20000 21000 22000 23000 24000 mass

Rap1B (1-167), A86C, T61Q; expected mass = 19369 D

T61Q_col #701 RT: 15.20 NL: 1.99E8 F: + p ESI Full ms [ 700.00-2000.00] 19368.0 100

95 90 85 80

75 70

65 60

55 50

45 40 Relative Abundance 35 30

25 20 15 10 19413.0 15254.0 18697.0 20800.0 5 16179.0 16964.0 18335.0 19561.0 21803.0 23572.0 23971.0 24773.0 0 15000 16000 17000 18000 19000 20000 21000 22000 23000 24000 mass

121 Appendix

8.3 Scientist model for global fit analysis

// RapGAP catalysed reaction of Rap, 3 steps, B recycled, normalized // // K1 K3 K4 // A+B = C ---> D ---> E + B [A = Rap.GTP, B = GAP, C and D = Intermediates, E = Rap.GDP] // K2 // //Variable Declaration IndVars: t DepVars: A, B, C, D, E, F25 Params: K1, K2, K3, K4, Ya, Yc, Ye // // Yc, Yb, Yc are fluorescence coefficients of A, B and C. F is the fluorescence transient. // //Differential equations & the trace A' = - K1*A*B + K2*C B' = - K1*A*B + K2*C + K4*D C' = K1*A*B - K2*C - K3*C D' = K3*C - K4*D E' = K4*D F25 = A*Ya + C*Yc + D*Yc + E*Ye // //Initial Conditions for trace 1 t = 0.0000 A = 1 B = 25 C = 0 D = 0 E = 0 ***

122 Appendix

8.4 Sequence allignment of Tsc1

↓ hsTSC1 MAQQANVGELLAMLDSPMLGVRDDVTAVFKENLNSDRGPMLVNTLVDYYLETSSQPALHI 60 RnTSC1 MAQLANIGELLSMLDSSTLGVRDDVTTIFKESLNSERGPMLVNTLVDYYLETNSQPVLHI DmTSC1 ----MVIEKIIGDLESNMTLENEEAKRKLVELLSQNKEQWVVKFMLDYFFTTGSQRILEV Sptsc1 ----MPLQSLVKALWNVLHEEESEGYPDLTELIAEVES------YQQRYPKQNPTNSQKI : .:: * . ..: : * : . . : : .. .: hsTSC1 LTTLQEPHDKHLLDRINEYVGKAATRLSILSLLGHVIRLQPSWKHKLSQAPLLPSLLKCL 120 RnTSC1 LTTLQEPHDKHLLDKMNEYVGKAATRLSILSLLGHVVRLQPSWKHKLSQAPLLPSLLKCL DmTSC1 LVKAQAPHDGYIFDKLDDCLKQSQHRVQSLQVFCFIVRHHPTWLYKIEKHRLIKSVFKLM Sptsc1 RHILDEIYEKTPFNNTRRRILWLAVLKTVIPLLILDRQAVGEWWDQI----FFP----FL : :: ::. : : :: : * :: :: : hsTSC1 KMDTDVVVLTTGVLVLITMLPMIPQSGKQHLLDFFDIFGRLSSWCLKKPGHVAEVYLVHL 180 RnTSC1 KMDTDVVVLTTGVLVLITMLPMIPQSGKQHLLDFFDIFGRLSSWCLKKPGHVTEVYLVHL DmTSC1 THEKEIVPLMSALLCIITLLPIIPNSVPNFLNDLFEVFGHLASWKLQNSNKLPDEKLVHL Sptsc1 NSPTQLKPVFSDLKSILFYILIFHDED------EWGGDLRRECAEET------. .:: : : : :: : :: :. : * * ::. •↓ ↓↓ ↓ hsTSC1 HASVYALFHRLYGMYPCNFVSFLRSHYSMKENLETFEEVVKPMMEHVRIHPELVTGSKDH 240 RnTSC1 HASVYALFHRLYGMYPCNFVSFLRSHYSMKENVETFEEVVKPMMEHVRIHPELVTGSKDH DmTSC1 QLGLQMLFHRLYGMYPCSFIAYLVEFIKRGNGGGIFQHTIKPLLNTVRVHPMLVTATPET Sptsc1 ------ITRLVDLYVSKAIENLGDVESQEQRNQTIECLVNVLVHYGIQRPKELS-S--- : ** .:* .. : * . . : :: :: ::. :* :: : hsTSC1 ELDPRRWKRLETHDVVIECAKISLDP--TEASYEDGYSVSHQISARFPHRSADVTTSPYA 298 RnTSC1 ELDPRRWKTLETHDVVIECAKISLDP--TEASYEDGDAVSHQLSACFPHRSADVTTSSYV DmTSC1 EVNNTRWKEMEPHDVVMECANLSLPVLLPETSNEDG-SYAYPMTPGYSRMTSNTSNTDYS Sptsc1 ------CFCHHFLNPPTRIPILSVMVEVIRRQGPRLYEIPQTGFYDLVLKCAEFDTS *..: :.:.: *. .:* . : . : • hsTSC1 DTQNSYGCATSTPYSTSRLMLLNMPGQLPQTLSSPSTRLITEPPQATLWSPSMVCGMTTP 358 RnTSC1 DTQNSYGGATSTPSSTSRLMLFSTPGQLPQSLSSLSTRPLPEPLQASLWSPSAVCGMTTP DmTSC1 YQLREFQQSRNVYTRFDSFASGDDVGPIWSPHNEIATTSSGIP-----LTPTTSFILPLQ Sptsc1 PILLSY--ALS----FILMILSHICNSLDDSLYRLFCIYLRFS----MIDPTSGFPSSTA .: : . : . : .. . *: . ↓ hsTSC1 PTSPGNVPPDLSHPYSKVFGTTAGGKGTPLGTPATSPPPAPLCHSDDYVHISLPQATVTP 418 RnTSC1 PTSPGNVPADLSHPYSKAFGTTTGGKGTPSGTPATSPPPAPPCPQDDCAHGPASQASATP DmTSC1 PAMNSQLMVGMTGSSPPEAAVEATPETTPLKDMRDIKQPGRAVNSHAVRAIFAVSHPSSP Sptsc1 SG-NWEVFHDFMSTYASTTTSQTDSSYNDVHDIVGSSQPD-YLESLDYSQLFSILYALYP . :: .: . . : . . * . . * hsTSC1 PRKEERMDSARPCLHRQHHLLNDRGSEEPPGSKGSVTLSDLPGFLGDLASEEDS---IEK 475 RnTSC1 PRKEERADSSRPYLPRQQDVPSDRGLEDLPGSKGSVTLRNLPDFLGDLASEEDS---IEK DmTSC1 MRKDQQSQFSFPDVSREAEESSHSYLEVNRGTAYDRRLSQVIQDRHNVERSVNTPCPSSL Sptsc1 INFLEFLRDPKLYASKHNFQIRYSFNQELLSTKSDGLLGRHLAHSNFLKYTAET-----E . : . :. : .: . * : :: hsTSC1 DKEEAAISRELSEITTAEAEPVVPRGGFDSPFYRDSLPGSQRKTHSAASSSQGASVNPEP 535 RnTSC1 DKEEAAISKELSEITTAEADPVAPRGGFDSPFYRDSLSGSQRKTHSAASGTQGFSVNPEP DmTSC1 PEINSDLSLVGGSVYPSVTQEVAAVCGECNETDRN-LCSVGGLHMPTSRSMHQLAKKRRN Sptsc1 LTDKSRWTRLDSIAVVALCNSLNAVG--IAESVMDPFGGKLPTTYEETSSATGLLAYPN- :: : . : : : . : : . : . . • ↓↓↓↓ hsTSC1 LHSSLDKLGPDTPKQAFTPIDLPCGSADESPAGDRECQTSLETSIFTPSPCKIPPPTRVG 595 RnTSC1 LHSSLDKHGPDTPKQAFTPIDPPSGSADASPAGDRDRQTSLETSILTPSPCKIPPQRGVS DmTSC1 RMASYSGNGSCADSRSSAAKKASWSTEAENPMRRTKSCSALSGMRQQHLEENDDEADCSS Sptsc1 ------ESHDIASEPFSISWP-QNPSISGSVHSATTFDKAQLSNTEDSYDN---IS : . : . . . . . :::. . .

123 Appendix

hsTSC1 FGSGQPPPYDHLFEVALPKTAHHFVIRKTEELLKKAKGNTEE-----DGVPSTSPMEVLD 650 RnTSC1 FGSGQLPPYDHLFEVALPKTACHFVSKKTEELLKKAKGNPEE-----DCVPSTSPMEVLD DmTSC1 QRQRGENGNTQKTGSRLQRSGRNLAISAPKDTARSCTHASTQTVEGLDSAPAQYENWLIE Sptsc1 HGTSYSEG------VSSIHMVKGER ..: |<------↓ hsTSC1 RLIQQGADAHSKELNKLPLPSKSVDWTHFGGSPPSDEIRTLRDQLLLLHNQLLYERFKRQ 710 RnTSC1 RLLEQGAGAHSKELSRLSLPSKSVDWTHFGGSPPSDEIRTLRDQLLLLHNQLLYERFKRQ DmTSC1 LLLECKEQRIDYERN-LLYPQDILDEYIKHAIKANESFDAEQGQLMCLQ--LEYESYRRS Sptsc1 G--SNNLELTSESLS------S-TNDTIRRLQRDLLFLQNELRFEKFVRQ ...... : : : :*: *: * :* : *. ↓ • hsTSC1 QHALRNRRLLRKVIKAAALEEHNAAMKDQLKLQEKDIQMWKVSLQKEQARYNQLQEQRDT 770 RnTSC1 QHALRNRRLLRKVIRAAALEEHNAAMKDQLKLQEKDIQMWKVSLQKEQARYSQLQQQRDT DmTSC1 IHAERNRRLMGRSRDKRSLEMERDRLREQLKNFDAKNKDLANKMDQAIRLANERQNIHQE Sptsc1 QHLQNIGKLHREHILDMAVESERQKLLLTNKRYKAQIELLNSEIDKHRSESQAALNRRVK * . :* . ::* .. : * . . : .::: . : : • • hsTSC1 MVTKLHSQIRQLQHDREEFYNQSQELQTKLEDCRNMIAELRIELKKANNKVCHTELLLSQ 830 RnTSC1 MVTQLHSQIRQLQHDREEFYNQSQELQTKLEDCRSMIAELRVELKKANSKVCHTELLLSQ DmTSC1 ELGEMRAKYQHELEEKKCLRQANDDLQTRLTSELARHKEMNYELESLRGQVFSLGTELQH Sptsc1 WENDFNNKIKALREEKKAWKSEESELKSSIESLISQLESIRNSQIDIAFSKNQLELKLQL .:. : : .::: . ..:*:: : . .:. . . . *. hsTSC1 VSQKLSNSESVQQQMEFLNRQLLVLGEVNELYLEQLQNKH--SDTTKEVEMMKAAYRKEL 888 RnTSC1 VSQKLSNSESVQQQMEFLNRQLLVLGEVNELYLEQLQSKH--PDTTKEVEMMKTAYRKEL DmTSC1 TQQQADIGLQCKQELARLEAEFIIMGEVQVRCRDRLAEIDNFRARDEELQMLQESSNLEL Sptsc1 YETKLK---EYEQHLSCVN------ISKKQVSSSSDTSFGNTKMDSSMIL . : . . :*.: :: : . . .. : * ------>| ↓ hsTSC1 EKNRSHVLQQTQRLDTSQKRILELESHLAKKDHLLLEQKKYLEDVKLQARGQLQAAESRY 948 RnTSC1 EKNRSHLLQQNQRLDASQRRVLELESLLAKKDHLLLEQKKYLEDVKSQASGQLLAAESRY DmTSC1 KDLRHSLDEKTSQLESMKHKISDLQAQLANSEKAMTEQKRLLSTVKDEYEEKFKSVNKKY Sptsc1 SNSEAVSDEQERELIESEKHRMKLESENLHLQANIELLKKDLEAINVVYEAKIFDLEKRL .. . :: .* ::: .*:: : : : *: *. :: :: :.: hsTSC1 EAQKRITQVFELEILDLYGRLEKDGLLKKLEEEKAEAAEAAEERLDCCNDGCSDSMVGHN 1008 RnTSC1 EAQRKITRVLELEILDLYGRLEKDGRLQKLEEDRAEAAEAAEERLDCCTDGCSDSLLGHN DmTSC1 DVQKKIIMQMEEKLMMMMQ------QP--- Sptsc1 SSEANAPELHNPVNLNYDA------. : . : : • hsTSC1 EEASGHNGETKTPRPSSARGSSGSRGGGGSSSSSSELSTPEKPPHQRAGPFSSRWETTMG 1068 RnTSC1 EEAAGHNGETRTSRPGGTRASCGGRVTGGSSSSSSELSTPEKPPNQR---FSSRWEPTMG DmTSC1 QGTTGHN----TCSPDTDRTDIASSIERNSPLSTSLASSESLSASLR------S Sptsc1 ------QLSKISEIKEN------YDELLTRYR------. . . * • hsTSC1 EASASIPTTVGSLPSSKSFLGMKARELFRNKSESQCDEDGMTSS-LSESLKTELGKD-LG 1126 RnTSC1 EPSSSIPTTVGSLPSSKSFLGMKTRELFRNKSESQCDEDGMTMSSFSETLKTELGKDSAG DmTSC1 TELKNLHQLVDTPTIPDVLNSMAGGAQFEDEVRPPAVDLASSASTASAINIVPHALDLPS Sptsc1 ------ELEGKFLESQAEVEELKNFQKPLVDTGSSIHSSPGLQQSKFIIRNDS : .: . : . : . . hsTSC1 VEAKIPLNLDGPHPSPPTPDSVGQLHIMDYNETHHEHS 1164 RnTSC1 MENKTPPSLDAPHPSSPSSDSMGQLHIMDYNETHHEHS DmTSC1 TSGGIGHTLTHPHPHP------HLHLQQQQQDQLQ-- Sptsc1 LHPKVGPPRRQSTDTS------RSTFRQY------. . : : : Supplement Figure 1. Alignment of Tsc1 sequences from Homo sapiens (SW: Q92574), Rattus norvegicus (SW: Q9Z136), Drosophila (SW: Q9U9A9) and Yeast (SPAC22F3.13). Red dots: reported mutations for human, red arrows: Tuberous sclerosis mutations. Dotted arrow indicates putative dimer domain.

124 Danksagung

Danksagung

I sincerely thank Prof. Dr. Alfred Wittinghofer and Prof. Dr. Klaus Gerwert for giving me an opportunity to work in their departments and for being in my examination committee. It is indeed a wonderful experience to combine different methodologies in two different but equally stimulating environments.

I thank Prof. Dr. Rolf Heumann for enlightening discussion and for being in my examination committee.

I thank Prof. Dr. Roger Goody for numerous scientific discussions and suggestions.

I thank Prof. Dr. Dr. h. c. Rolf Kinne, Dr. Jutta Roetter and International Max Planck Research School for support and inspiration.

I thank Dr. Carsten Kötting, Dr. Yan Suveyzdis, and Dr. Marco Blesshanol for introducing me to FTIR and several discussions. Axel and Harald, for all your support with the spectrometer and technical ‘fehler’.

Christian and Reza for numerous scientific discussions and personal care.

Toshi for your support and many discussions over the last year. Oli D, for being a source of inspiration all thought out my work. My utmost gratitude to both of you.

Thanks to the former members of the group and department, Astrid, Holger, Robert, Atsushi and Özkan for support and wonderful company.

Doro, for teaching me the a, b, c of several things, starting from scratch. Not to forget several clones and mutations. Thanks to Andra, Michael and all members of lab A 2.15.

Caro, thank you for all your support and an all round good feeling. Thanks to the members of lab A 1.11, Alex, Katja, Insa and Nils.

Thanks to lab A 2.12, and the people, particularly Beate, Marian and Antje. The lab hopper often hopped there.

Martin, Boriana, Arjan and Helena for a lovely atmosphere in our office A 1.23.

Rita for her assistance in bureaucratic matters. I am deeply indebted for your concern.

Agni, Sudipta, Minhaj, Vinod, Revs, Navneet, Christoph, Dennis and Torsten for discussions and good time.

To Dad, Mom, Brother and relatives who keep me in tune even if I am few thousand miles away.

125 Teilpublikationen dieser Arbeit

Teilpublikationen dieser Arbeit

1. Chakrabarti PP, Suveyzdis Y, Wittinghofer A, Gerwert K. Fourier transform infrared spectroscopy on the Rap.RapGAP reaction, GTPase activation without an arginine finger. J Biol Chem. 2004 Oct 29; 279(44): 46226-33.

2. Daumke O, Weyand M, Chakrabarti PP, Vetter IR, Wittinghofer A. The GTPase-activating protein Rap1GAP uses a catalytic asparagine. Nature. 2004 May 13; 429 (6988): 197-201.

126 Erklärung Erklärung

Ich versichere hiermit, dass ich die vorliegende Arbeit selbständig und ohne fremde Hilfe, nur unter Verwendung der angegebenen Hilfsmittel, angefertigt habe.

Dortmund, 21. März 2005

Partha Pratim Chakrabarti

127 Lebenslauf

Lebenslauf

Persönliche Daten: Partha Pratim Chakrabarti Amalienstr. 3, 44137 Dortmund geb. am 14.09.1977 in Kalkutta, Indien Staatsangehörigkeit: Indisch (Indian) Familienstatus: ledig

1994 – 1996 Ramakrishna Mission Vidyamandir, Belur, WB, India Abschluss: Higher secondary (10 + 2)

1996 – 1999 Presidency College, Calcutta University, India Abschluss: B. Sc (Hons. Chemistry)

1999 – 2001 Indian Institute of Technology, Kanpur, India Abschluss: M. Sc (Chemistry)

2001 – 2005 Promotionsarbeit am Max-Planck-Institut für molekulare Physiologie, Dortmund, Germany

128