1
Q
What are the major pathways by which a signal can enter a cell?
A
- hydrophobic molecules can diffuse to intracellular receptor
- ion channel
- GPCR
- ligand binds to enz
2
Q
Why does the plasma membrane provide a big barrier for entering cell?
A
- not rigid, so hard to induce a switch change
3
Q
What are koff and kon?
A
- rate constants
kon
L + R ⇌ LR
koff
4
Q
What are the actual on/off rates of ligand binding to receptor?
A
- on = kon [L][R]
- off = koff [LR]
5
Q
At eq how do on and off rates relate, and how can this be rearranged?
A
- they are equal
- kon [L][R] = koff [LR]
- so koff / kon = [L][R] / [LR] = Kd (dissoc constant)
6
Q
What is the diff between k and K?
A
- k is a rate constant
- K is an eq constant
7
Q
How is fractional occupancy of ligand on receptor calc?
A
- [LR] / [R] +[LR]
- ie. fraction bound / total amount of receptor
8
Q
What can fractional occupancy demonstrate?
A
- multiply top and bottom by the ratio [L] / [LR]
- [LR] x ( [L] /[LR] )
/ [L][R] / [LR] + [L]
= [L] / Kd + [L] - plot out DIAG
- see need a LOT of ligand to get to almost complete binding
9
Q
Why is it important how tightly ligand binds receptor, and what conclusion can be drawn from this?
A
- receptor should get turned on by ligand binding
- if ligand binds v tightly, then background levels of ligand bind
- if ligand binds v weakly, need v high ligand conc
- conclusion: optimal ligand conc is approx Kd for its receptor OR optimal Kd for ligand is close to its physiological conc
- fairly strong Kd means hormone can remain at fairly low conc
10
Q
When and why is it not a problem if ligand binds receptor weakly?
A
- in autocrine/paracrine, as signal released so close to target
11
Q
What is the important consequence of the fact that Kd = koff / kon, and what conclusion does this lead to?
A
- max poss kon approx 10^8/M/s (diffusion controlled) and often lot slower, assumes right orientation
- more typical kon makes half life (log 2/k) too long
- in general cell needs to actively remove ligand from receptor, can’t wait for dissoc (receptor internalisation) OR deactivate receptor w/ arrestins
12
Q
Why do neurotransmitters need to be removed v rapidly, and how is this done?
A
- to clear way for next nerve impulse (5ms)
- ACh removed by acetylcholinesterase
- dopamine, noradrenaline and serotonin taken up by transporters
13
Q
What can inhibitors of dopamine, noradrenaline and serotonin treat?
A
- obesity and ADHD
- prozac is specific inhibitor of serotonin uptake
- but best known uptake inhibitor is cocaine
14
Q
What is alt way to get stronger binding w/o having v slow dissoc?
A
- have ligand binding by 2 weak interactions, rather than 1 strong 1
15
Q
In what instance can 1 ligand turn on a signal?
A
- ceg. 1 photon can activate rod cell in eye
- therefore signalling pathways often req amplification –> eg. in eye 1 photon leads to approx 10^5 cGMP broken down
- signal activated an enz –> eg. in eye, phosphodiesterase
16
Q
Generally, why is not good to be so sensitive that 1 ligand can turn on signal?
A
- binding and activation are random events at mol level, so would lead to random activation (too much or too little)
- proteins not rigid, so can get switched on w/ no ligand at random, or often ligand can bind and not activate it
17
Q
What does it mean to say that most signalling pathways have threshold level of signal?
A
- enough signal to lift response clear of noise
- typically cells have 10^4 - 10^6 receptors (usually need sig no. bound for signal to be transmitted)
18
Q
How do cells ignore ‘random’ signals until large enough to pass threshold and be recognised as signal?
A
- many incoming signals lead to phosphorylation
- cells have lots of nonspecific phosphatases, which go around dephosphorylating proteins
- higher rate than ‘background’ phosphorylation, so keeps signals turned off until genuine signal arises and swamps phosphatase activity
19
Q
Does a signal work as a good switch, and why?
A
- some do, some don’t
- proteins not rigid, so ‘off’ protein could randomly behave ‘on’ 0.1% of time (or often more) and vice versa
20
Q
How does myoglobin vs Hb show good signal like behaviour?
A
- DIAG*
- myoglobin = like saturation curve
- Hb = much more switch like
21
Q
How can proteins become better switches?
A
- need cooperativity
- means clustering of receptors (into lipid rafts)
- scaffold proteins to bring components together
- add domains to increase colocation
- eg. kinase cascade –> 3 kinases amplifies signal and increases switch like behaviour
22
Q
Why is Ca an unusual signal?
A
- 2nd messenger, like cyclic nucleotides
- usually a binary signal (on or off)
- not made or destroyed, just moved –> stored outside cell and in ER, moved into cyto
23
Q
How does Ca conc vary between cyto and ec space, and what does this mean for signalling?
A
- in cyto usually <10^-7M (v low)
- in ec space often much higher, around 10^-3M
- signal easy to initiate as influx when open channel, but harder to move out as against quite high conc grad
- so cells work v hard to keep Ca levels low inside cells
24
Q
What conc does Ca need to reach for signal to be prod?
A
- increase to approx 10^-6M
25
What does cell req for Ca signalling to be poss?
- Ca pumps (to pump Ca into ER and out of cells)
- Ca channels (reg by signals)
- way of recognising increase in Ca conc
26
Why is Ca a good signal?
- v quick (few ms for levels to rise enough for signal) as doesn't req enz reactions etc.
27
What is the typical Na:K in cells and how much ATP is used in maintaining this, why is Ca pumps harder work?
- in most cells [Na+] 10-30x lower inside cells and [K+] 10-30x higher inside cells
- typical euk cell uses approx 25% ATP turnover in maintaining correct ratio (65% in neurons)
- Ca2+ 10,000x lower in cyto, so cells will have to work even harder to pump out
28
What are the 2 main types of Ca pump, and where are they found?
1) Uses ATP, called P-type Ca2+ ATPase
- 1 Ca out per 1/2 ATPs
- found in all euk cells
2) Use Na grad (so indirectly use ATP)
- 1 Ca out, 3 Na in (works hard)
- OR 1 Ca and 1 K out, 3 Na in (works harder)
- found in cells that do lots of Ca signalling, eg. muscle and nerve cells, need to get rid of Ca fast
..........plus extra pumps to pump Ca into ER
29
How are Ca pumps important in sarcoplasmic reticulum in muscle?
- store Ca to be used in stimulating muscle contraction
- roughly 90% membrane protein here is P-type Ca2+ ATPase
- resets Ca conc w/in 30ms
30
What are the 2 types of Ca channels?
- IP3 gated channels (in ER membranes) --> IP3 prod from PIP2 by PLC
- voltage gated = cell membrane depolarisation (eg. from nerve impulses/in muscle) --> some Ca channels activated by Ca itself, providing amp by +ve feedback
31
What is IP3 also used for?
- fertilisation of egg by sperm
32
What is an eg. of a Ca-dep channel?
- ryanodine receptor
33
What mechanism is thought to be responsible for Ca waves and oscillations?
- cell membrane depolarisation
- travels further than simple Ca conc changes would
- can also be causes by large Ca conc grads
34
How is Ca conc important in dev oocyte?
- for dev of correct orientation
35
What helps cell maintain Ca grads?
- lots of proteins in cyto that act as Ca buffers to mop up Ca (so can't diffuse far) --> related to calmodulin
36
What is the most common mechanism for Ca action?
- binds to protein and causes conformational change, then recognised by further system
37
What is the structure of calmodulin (CaM)?
- 4 binding sites for Ca
- sort of symmetrical (result of gene dup)
- long central helix
- 4 EF hands --> helices E and F have Ca binding site and when Ca bound they close up like hand
38
What is the result of Ca binding calmodulin?
- binding to Ca exposes hydrophobic surfaces (esp Mets, not common AA, long flex side chain, so can adapt to diff target proteins)
- folds up around target helices (eg. MLCK)
- Met allows CaM to bind many diff targets
- CaM activates lots of diff downstream signals in variety of cells --> binds eg. Ca/CaM dep kinase, phosphodiesterases, NO synthase
39
Where is CaM-kinase II found?
- in brain synapses and elsewhere
40
What happens when Ca binds CaM-kinase II?
- auto-inhibited by inhibitory domain to keep it in inactive form
- Ca activates and 1st thing it phosphorylates is itself --> making it even more active (autophosphorylation)
- now fully active and Ca dissoc
- some is Ca indep (50=80% active) and shows 'memory' of having bound Ca, so prolongs signal, may be involved in learning
41
What is another common mechanism of Ca binding (ie. conformational change doesn't cause recognition by further system)?
- causes conformational change, which leads to relocation to membrane
- eg. in bacterial toxin, w/o Ca not particularly hydrophobic
- w/ Ca makes hydrophobic surface, enabling binding to membrane
- diffuses to membrane and sticks once it reaches it
- Ca often binds C2 domain --> found in many proteins (>600 in humans), key one is PKC
42
Why is signalling more complex than just linear pathways?
- not completely separate, lots of cross talk between them
43
What does steroid signalling cause?
- developmental signals
| - takes long time for change to occur, but when does causes permanent change in cell
44
What are some eg.s of hydrophobic ligands?
- steroids (testosterone, oestrogen)
- vitamin D
- retinoic acid
- thyroxine
45
How is signaling by hydrophobic ligands poss, and what happens?
- so hydrophobic can diffuse across cell membrane
- in simplest case receptors waiting in cyto
- unbound receptor inactive and usually bound by something big and cytoplasmic to inhibit it, eg. Hsp
- ligand binds, causing conformational change, releasing receptor
- receptor moves into nucleus and binds REs on DNA (ie. receptor also a TF)
46
What is a heat shock protein, and what is their role?
- overexpressed when heat shocked
- heat causes proteins to unfold
- so bind to exposed hydrophobic parts, where partially unfolded, to shield them and give them chance to refold
- = chaperone proteins
47
What do chaperone proteins do?
- stop binding to other proteins
48
What is the general structure of a homodimeric receptor for hydrophobic ligands?
- v modular, 3 domains w/ defined function
- DNA binding domain (DBD) --> binds DNA as homodimer
- ligand binding domain (LBD)
- activation domain (AD)
49
How do heterodimeric receptors differ for hydrophobic ligands?
- receptor in nucleus permanently
- in absence of ligand, usually transcrip repressors --> by acetylation
- when bind ligand, big conformational change and become transcrip activators --> by causing hyperacetylation of histones
50
What is the problem for steroid hormones, and how is this solved?
- almost insoluble in water/blood
- needs to be transported by carrier proteins
- albumin can do it, but usually use specific transporters
- eg. sex hormone binding globulin transports testosterone/oestrogen etc. --> levels increased in pregnancy, oral contraceptives, low calorie intake and levels decreased in diabetes, obesity, anabolic steroids
51
What mechanisms do receptor kinases use, and why is this a good mechanism?
- dimerisation
| - don't need rigid change like GPCRs
52
What is the mechanism for receptor kinase activation?
*DIAG*
53
What are the diff types of receptor kinases?
- ser/thr
| - tyr --> can fit into much deeper binding pockets as longer side chain
54
What are the downstream effects of receptor kinases?
- usually affect transcrip of DNA
| - ie. medium to LT effects --> differentiation and dev
55
Why is phosphorylation a good signal?
- rapid
- easily recognised
- efficient
- reversible (easy to turn on/off)
56
What are the consequences of free energy of phosphorylation?
- free energy heavily on side of dephosphorylation (default off)
- but kinetically phosphorylation v slow so need enz to make it happen
57
What is the GPCR mechanism?
- hormone binds, conf change = twists TM helix, alt structure of cyto face
- binds Gα subunit, causing conf change
- Gα dissoc from GDP and assoc w/ GTP, triggering dissoc from receptor and Gβγ
- hormone dissocs and Gα binds effector --> activating it
- hydrolysis GTP --> GDP causes Gα to dissoc from effector and reassoc w/ Gβγ
....repeats....
58
What kind of receptor kinases do Smads use?
- ser/thr
59
What is an eg. of a Smad system?
- receptor for TGFβ important for dev --> typical effect to prevent prolif, so defect leads to cancer
- type II receptor has S/T kinase domain and is constitutively active
- TGFβ binds as dimer, so has similar interactions w/ both receptors
60
How does Smad signalling work?
* DIAG*
- ligand binds type II receptor, causing dimerisation w/ type I receptor
- type II kinase can then phosphorylate type I kinase and activate it
- type I kinase then phosphorylates Smad ligand
- phosphorylated Smad dissoc and translocates to nucleus, where affects gene expression
- usually involves assoc of phosphorylated Smad w/ another diff Smad
61
How are Smads autoinhibited?
- NLS hidden in inactive Smad and only revealed after activation, so able to leave cyto and go to nucleus
62
What is the consequence in Smads, of the fact that proteins are not symmetrical, and what is the solution?
- if dimerise, must have have diff interactions w/ each half of receptor
- common solution is to make ligand a dimer too
63
How do protein:protein interactions req for Smad signalling occur, and how fast is this?
- occur v rapidly (typically <0.1s)
- even though occur by random walk (diffusion) = spread out from starting point until finds partner to bind (efficient for short distances)
64
Why is a 2D search on a membrane best?
- much faster than 3D search through cyto
| - also if attached to membrane then in correct orientation
65
Why does it not matter that type II receptor in Smads is constitutively active?
- no effect normally, as no substrate
| - ie. activity due to co-localisation
66
How is Smad signalling turned on and off?
- turned on by phosphorylation of kinase which activates it
- turned off by 1 of gene products activated by Smad3/4 complex which recruits Smad ubiquitination regulatory factor (Smurf) , which degrades Smad
- also turned off by phosphatase that deactivates Smad
67
What is the size of signal in Smad signalling dep on?
- ratio between phosphorylation and dephosphorylation
| - effectively more ligand = more signal
68
Why is Jak/STAT signalling used in cytokine signalling?
- complicated set of signals which need to interact, so system needs to be able to integrate multiple signals --> ie. prod bigger/smaller signal dep on which inputs there are
- need specificity, to respond approp to correct signal
69
How does the Jak/STAT system work?
* DIAG*
- membrane receptor recognition domain and prot Tyr kinase attached to receptor (kinase is sep)
- ligand binds to 1 receptor, quickly causes dimerisation
- -> formation of ternary complex
- cross phos
- activates JAK to activate receptor
- signal ON
70
Why is Jak/STAT not technically a RTK?
- kinase is separate protein attached to receptor
| - but works same way
71
What is a ternary complex in the Jak/STAT system?
- complex between 1 hormone and 2 receptors
72
What is the key specificity of the Jak/STAT system?
- recognition of phosphorylated receptor by diffusible signal, Stat --> done using specialised domain that recognises pY, called SH2 domain
73
What are some protein domains used widely in signalling pathways for recognising motifs?
- SH2 --> recognises pY (1000x better than Y)
- PTB --> also recognises pY, but diff
- SH3 --> recognises polyproline (v common)
- PH --> recognises membrane/lipids (mainly PI), has organisational role, not creating signalling, but in attaching signalling system in correct orientation to membrane
- PD2 --> recognises C-ter peptides
- several others, eg. DH
74
Why do many proteins use adaptor proteins?
- to link specific signal into more general pathway
- or to prod multiple inputs into same pathway
* DIAG*
75
What are common features of common protein domains used widely in signalling pathways for recognising motifs?
- small
- typically have N and C-ter close together, and on opp side to binding site --> useful as if want to bind v specific kinase, just has to insert gene in middle of linker
76
How does SH2 bind pY?
- 2 pronged binder, recognises 3 residues before pY
| - pY can poke quite far into binding pocket w/ +vely charged residues
77
How does PTB bind pY?
- recognises 3 residues before pY
78
How does SH3 bind polyproline?
- recognises PXXP
- Pro much more rigid due to structure
- PXXP folds into polyproline II helix = v extended chain, w/ 120° rotation from 1 AA to next, so Pros on same face
- big SA:vol, so rigid extension, so good signalling device for rapid on/off
79
Why do some proteins consist entirely of domains for recognising motifs, and how do they use them?
- presumably to help assemble large complexes and bring other proteins together
- some use 2 domains w/ relatively weak binding to enhance overall affinity
- others link together diff signalling pathways, eg. Grb2
- some have large amounts of chain w/ no domain at all (intrinsically disordered protein = IDP)
80
How do domains for recognising motifs aid the mechanism of Jak/STAT?
*DIAG*
81
Where does specificity of Jak/STAT pathway come from, and what consequences does this have for drug design?
- binding (not specificity of kinase)
| - trying to get good drug by inhibiting specific kinase may be wasted effort, better to go for binding interaction
82
How is Jak/STAT signal turned off?
- SOCS protein, w/ SH2 domain that binds receptor an recruits E3 ubiquitin ligase via 2nd domain called SOCS box --> ubiquitinated protein then degraded
- phosphatase that uses SH2 domain to bind receptor and dephosphorylate JAK
83
What do errors in Jak/STAT system often lead to and why?
- dev abnormality or cancer
| - as controls growth and dev
84
What is erythropoietin (EPO) and what is its role?
- a Jak/STAT ligand
- stimulates RBCs to prolif and differentiate
- so externally administered EPO illegal in sports
85
What prop of kinases are S/T?
- vast majority (>90%)
86
What is the role of VEGF (vascular endothelial growth factor) and what is it critical for?
- hormone for inducing growth of new blood vessels
| - critical for tumour growth, as need lots of oxygen
87
How does VEGF work as a RTK signalling system?
* DIAG*
- VEGF is a dimer
- 2 kinase domains close and can phosphorylate each other
- like Jak/STAT phosphorylated kinases are active and phosphorylate receptor
- this is recognised by SH2 adaptor = Grb2
88
What variations are there on VEGF signalling?
- many ligands, eg. EGF (epidermal GF)
- binds as 2 EGF --> 2 receptors, leading to conformational change, which causes dimerisation
- EGF can also form heterodimers, w/ other receptors, some of which are receptors for a diff ligand, creating potential to integrate signals from 2 diff sources
89
In what states does Grb2 exist?
- 2 states
- preferred is inactive autoinhibited state, in which SH3 domains bind to pY binding site on SH2 domain, therefore masking SH2 and SH3 sites, SH2 binding receptor exposes SH3 domain
90
What happens when Grb2 adaptor binds Sos?
- Sos normally in cyto, not close to membrane
- Pro-rich arm binds SH3 (closest domain to membrane)
- C-ter SH3 domain binds string of other parts, eventually turning on PI3K signalling
- outcome = bring Sos to membrane where can act as a GEF
91
What is the structure of the GTP bound state of a GTPase, and how does this alt when GDP bound?
- network of H bonds that fold 2 loops in protein in towards GTP, called switch I and II
- makes these 2 loops 'spring tensioned' and if GTP hydrolysed to GDP then tension released and loops spring out
- loops prefer to be in outward orientation, so default state is GDP bound
92
How are small GTPases able to act as switches in signalling pathways?
- in cell GTP approx 10x more than GDP
- G-proteins have GTPase activity and hydrolyse GTP to GDP, turning off signal
- most control comes from GEFs and GAPs
93
What is the role of GEFs?
- guanine exchange factors
| - allow bound nucleotide (normally GDP) to be released and new nucleotide (normally GTP) to be bound
94
What is the role of GAPs?
- GTPase activating protein
| - stim hydrolysis of GTP to GDP
95
How does Sos1 act as a GEF for Ras?
- cat exchange of GDP to GTP, activating Ras
96
What are defective Ras proteins often involved in, and why?
- approx 25% cancers, they are oncogenes
- as hydrolysis of GTP too slow, so left on too long
- esp Gly12 of Ras often mutated, as prevents GAP binding, so signal on much longer
97
What is cancer primarily a disease of?
- signalling
98
What is an oncogene?
- gene that, when mutated or expressed at high levels, helps turn normal cell into cancer cell
99
What is a proto-oncogene?
- normal gene that can become an oncogene due to mutations or high expression
100
What can oncogenes affect?
- any stage of signalling
| - eg. growth factors, receptors, kinases, GTP binding proteins, DNA binding, cell cycle
101
Is single oncogene enough to cause cancer, why?
- v often, need at least 2 acting at diff points to bypass failsafe mechanisms
- eg. 2 of ras, myc, p53
102
What is an eg. of a virus found to cause cancer, and how is this poss?
- Rous sarcoma virus found to encode oncogene v-src --> related to host proto-oncogene c-src, but w/ small no mutations
- v-src encodes active kinase
- gene prob picked up by virus from normal host cell and incorp into viral genome
103
What happens once Ras is activated?
- activates Raf, becomes active kinase
- Raf phosphorylates MEK
- MEK phosphorylates ERK
- phosphorylated ERK goes to nucleus
104
What is the general kinase cascade?
- MAPKKK phosphorylates MAPKK
- MAPKK phosphorylates MAPK
- MAPK = mitogen activated kinase
105
What is mitogen?
- chemical that causes cancer by triggering mitosis
106
Why are NLS/NES necessary?
- all proteins need to go through nuclear pore to get to nucleus
- proteins blocking it
- so need transporter protein, which recognises NLS/NES
107
How is ERK location controlled?
- by balance between 2 intrinsic functions
- going to nucleus directed by NLS
- leaving nucleus directed by NES
- NES > NLS when unphosphorylated, so usually in cyto, as when moves to nucleus immed leaves again due to NES
- MEK5 phosphorylates, destroying NES, so stays in nucleus
* DIAG*
108
What happens once ERK is in the nucleus?
- can phosphorylate range of proteins
- in particular phosphorylates TFs and activates them
- TFs cause transcrip of fos and other early response genes
* DIAG*
109
In what organisms is the receptor-Grb2-Sos-Ras-Raf-MEK-ERK pathway found, and what is its role?
- humans --> reg wide range of cell growth and differentiation, and commonly seen to go wrong in cancers
- Drosophila --> controls eye dev
- worms --> reg vulval dev
110
Why are fos and jun important oncogenes?
- viral expression often leads to cancer, as activates genes at wrong time
111
What happens when fos binds jun?
- forms the TF, AP1
112
What is the structure of fos and jun?
- leu zippers
113
What is the structure of a leucine zipper?
- 2x α helices, zip up and held together in middle by Leus
- pack against each other and held together by hydrophobic interactions
- consensus seq = a (φ) b c d (L) e f g
- -> Leu every 7 residues, w/ a hydrophobic residue in between
- forms coiled coil of intertwined α helices
- α helix has 3.6 res per turn, coiled helix has to be twisted a bit making it 3.5 res per turn, so Leu exactly every 2 turns
114
What dimers do fos and jun form w/ each other and alone?
* DIAG*
- fos/jun forms stable heterodimers
* DIAG*
- fof doesn't form homodimers as too unstable
* DIAG*
- jun forms homodimers --> unstable but not as much as fos, as Lys sidechain is longer and more flex than Glu
115
How do fos/jun heterodimers bind DNA?
- both have activation domains at C-ter, both need to be present of Ap1 to be active
- bind DNA using scissors grip, using basic residues at N-ter
- both bind into major groove of DNA
116
What is the common mechanism of how most kinases are activated?
- kinase has 2 domains, 1 holds ATP and other holds substrates by positioning it next to activation loop
- activation loop only in correct orientation if phosphorylated
* DIAG*
117
How are kinases often further inactivated (apart from activation loop)?
- peptide seq called pseudosubstrate loop, looks a bit like substrate and blocks active site --> roughly same seq so binds active site and moves out way once kinase activated
118
What does phosphorylation of activation loop in ERK2 lead to?
- reorg of dimerisation interface
| - dimerisation necessary for movement into nucleus
119
How are Cdk cell cycle regulatory proteins regulated?
- PSTAIRE helix has -ve Glu residue
- -ve charge involved in positioning ATP in correct orientation
- cyclin ensures helix (and ∴ Glu) in correct orientation
120
Why are nearly all kinases activated in the same way, and what is the exception?
- in evo if something works well, tends to get reused many times
- except Raf by Ras, as Raf top of kinase cascade, so last chance to stop signal being propagated (key signal input stage)
121
What is role of the protein 14-3-3?
- activates Raf
- v abundant and binds phosphorylated S/T w/ rigid structure
- dimer w/ lots of isoforms --> phosphorylation disrupts dimer and abolishes binding
- binds to Raf, but diff dep on which residues phosphorylated
- rigid enough to bind phosphates if right distance apart, or can bend them slightly to fit
122
What is the mechanism of Raf?
- resting Raf, phosphorylated at S259 and S627, clamped shut by 14-3-3 (binding site hidden)
- activated Ras binds Raf and displaces 1 end of 14-3-3 and allows PP2 to remove S259-P --> acts once Raf opened up, removes risk of spontaneously closing up as phosphate no longer there
- 14-3-3 detachment uncovers Y341, phosphorylation activates Raf and Raf starts kinase
- S259 rephosphorylates, 14-3-3 clamps it shut again and deactivates Raf
123
How do we know that kinases and phosphatases aren't v specific?
- ≈ 25,000 human genes
- recent estimate that may be 500,000 phosphorylation sites (20 per protein)
- but only 518 kinases (av. 1000 targets each)
- no. phosphatases about a third (av. 3000 target each)
124
Are S/T or Y kinases and phosphatases more specific?
- more S/T kinases, but more phosphorylation sites, so Y kinases more specific
- more Y phosphatases and fewer phosphorylation sites, so Y phosphatases more specific
125
If kinases aren't specific, where does the specificity come from?
- proximity, due to scaffolds or adaptors
126
What are the characteristics of scaffold proteins?
- flexible --> intrinsically disordered structure
| - v specific, as only certain kinases bind
127
What is the role of scaffold proteins?
- holds components close together so rate substrate gets to active site is faster --> so prefer to act w/ each other than another protein
128
Why is the insulin receptor unusual?
- 2 halves of dimeric receptor already attached by disulphide bonds, but too far apart to allow phosphorylation
129
How does insulin signalling demonstrate the effects of scaffold proteins?
- phosphorylated receptor binds to IRS1 (scaffold protein)
- IRS1 allows assembly of lots more proteins onto receptor and increases specificity and integration of signals from other pathways
- adaptors play similar role to scaffolds --> allow binding of multiple proteins around target
* DIAG*
130
Why does cell need to actively turn off kinase?
- dissoc of ec ligand is slow
131
What methods can cell use to turn off kinases?
- internalisation of ligand bound receptor (happens for most receptors, not just kinases)
- phosphatases linked to substrates via scaffolds/adaptors
* DIAG*
- degrading crucial part of system using ubiquitin systems --> E1/E2/E3 attach ubiquitin and get more specific
132
What is an eg. of a system that uses ubiquitin to turn off signal?
- Jak/STAT = activation --> expression of SOCS --> SOCS SH2 binds to phosphorylated receptor --> SOCS box domain recruits E2
- Smad signal 1 turned off by Smurf
133
How are G proteins turned off?
- GAP
- arrestin binding to GPCR --> downregs receptor and often leads to internalisation and recycling/digestion, also can be start of signalling pathways
134
Why is colocation vital?
- most steps in RTK pathway not enz reactions, they're binding events
- most of these only function to get component into diff place, usually next to cell membrane or into nucleus
135
How do PH domains demonstrate colocation?
- bind to membranes and inositol phosphates
- so attach components onto membrane in defined orientation
- can recognise specific lipids, eg. specific membranes, or specific parts of membrane (eg. membrane rafts)
136
What are membrane (lipid) rafts?
- regions of membrane rich in cholesterol and sphingolipids
- more rigid and thicker than normal regions of membrane
- occur in specific parts of membrane
137
What is the role of membrane (lipid) rafts?
- collect functionally related proteins together --> eg. signaling and cytoskeletal proteins (actin fibres, MTs)
- makes easy to link external signal to ec activity, as signals can pass rapidly down MTs to nucleus etc.
138
When does membrane (lipid) raft assembly occur?
- not permanent structures, can be assembled and taken apart as necessary
- assembly occurs as natural result of segregation of diff lipid components of membrane, ie. spontaneous event
139
What are the 3 isoforms of Ras, and how are similar/diff?
- K-Ras, N-Ras, H-Ras
- similar structure
- diff lipid anchors attached, so attached to diff membranes, so act diss as in diff places
140
Where are the diff isoforms of Ras found, and why?
- K-Ras has farnesyl charin attached, but also has basic patch, so only at cell membrane
- N-Ras and H-Ras have palmitoyl chain attached, H-Ras often has 2, palmitoyl chains more likely to be in cell membrane, so both at cell and golgi membranes in diff ratios, but palmitoylation reversible and controls where they are
141
What is the consequence for Ras from lipid anchor sbe easily removable and changed?
- can remove Ras easily by cleaving
| - or move to diff place in membrane
142
What is an important consequence of colocation?
- function of protein not determined only by activity in vitro, but also location
- can have diff functions in diff places, so simply inhibiting may not have intended effect
143
Why is autoinhibition necessary?
- equally important signal not tuned on 'by mistake'
| - ie. diff components should remain off unless activated
144
When does autoinhibition work best?
- when molecule turned off by intramolecular binding
145
Why does a linker improve binding?
* DIAG*
- local conc higher so binds tighter
- binds even if not as high affinity, as near each other, so still works w/ much higher koff
146
How is EGF receptor activated?
- all Tyrs that can be phosphorylated in flex part at end of polypeptide chain, so can easily find way into active site
- unusually, EGF binds outside faces of receptor and dimerisation interface is opp face
- activated by ligand binding, dimerises, bringing 2 domains on inside together --> bind asymmetrically, next to helix, repositions Glu, so ATP in right place
- 1 domain activates the other
147
How is EGF receptor autoinhibited?
- in resting state 2 domains fixed together, blocking active site
- this means not active kinase, as well as fact that Tyrs aren't phosphorylated
148
Is EGF a homodimer or a heterodimer?
- homodimer
| - but other things can interact and form heterodimer, eg. ErbB2
149
What can compete w/ strong intermol binding, in terms of adaptors?
- weaker intramol binding
150
What is the structure of Sos, and what are the diff domains?
* DIAG*
- DH often next to PH and regs how PH binds to membrane
- H = histone like (structurally)
151
What happens once Sos is at membrane surface, in order for it to become an effective GEF, and why is this mechanism necessary?
- a Ras-GDP binds and activates it
- Ras-GTP even better at activating Sos as GEF for further rounds
- PH binds PIP2 and increases activity even more
- ensures Sos isn't fully effective GEF until all these steps happened, so not activated 'by mistake'
152
What is the role of Src kinase?
- part of many signalling pathways and often assoc w/ receptors (Jak/STAT)
- also virally encoded oncogene
153
How is cellular Src kinase autoinhibited?
* DIAG*
- SH2 binds to pY near C-ter
- SH2-pY binding fixes linker containing Pro-rich seq
- SH3 domain binds to linker and locks everything shut
- linker locked in way that distorts active site
154
How can cellular Src kinase be activated?
- dephosphorylation of pY
- binding of SH2 to 'better' pY seq, eg. in a substrate (quite weak interaction so easily displaced)
- binding of SH3 to 'better' Pro-rich seq
155
What is the role of Abl kinase?
- similar to Src, but in diff systems
156
How is Abl kinase autoinhibited, and what happens when activated?
- kinase active site held open so groups in wrong place
- SH2 binds to back of kinase and locked by N-ter peptide -->has lipid anchor myristrol (v hydrophobic) and fits into pocket of kinase
- activation allows myristoylated N-ter to come loose and insert into membrane
157
What does the fact that the 1st thing that evolves is a binding interaction, then other features (autoinhibition/allostery) evolve later imply, and what is the consequence of this for drug intervention, and an eg.?
- implies many systems have same basic mechanism but differ in autoinhibition, scaffolds etc.
- consequence is drugs may be more specific when applied to autoinhibition, scaffold etc. rather than eg. to kinase
- wg. anticancer drug Gleevec binds to active site of Abl kinase, but not Src, cos of diff way active site locked
158
How is Raf autoinhibited, and how is this disrupted?
- by 2 halves of hairpin binding together
| - disrupted by ras binding
159
What suggests that 1st receptor kinases were S/T?
- proks have v few Y kinases, but lots of S/T kinases
| - plants also have few RTKs
160
Why bother w/ a system as complicated as RTKs?
- prob not signal amp, as scaffolds decrease amp
| - prob for more reg and input/output
161
What is the main feature of cancer, what hallmarks of this are there, and what mutations does this involve?
- uncontrolled rapid cell division and growth
- evading growth suppressors (Cdk inhibitors)
- enabling replicative immortality (telomerase inhibitors)
162
What does chemo target?
- any cells growing too fast
163
What is the 2nd most common target for cancer?
- receptor kinase pathways
164
What hallmarks of cancer, and mutations are caused by receptor kinase pathway defects?
- sustained proliferative signalling (EGFR inhibitors)
- inducing angiogenesis (VEGF inhibitors)
- activating invasion and metastasis (HGF/C-Met inhibitors)
165
What are the other hallmarks of cancer?
- avoiding immune destruction
- tumor promoting inflam
- genome instability and mutation
- resisting cell death
- dereg cellular energetics
166
What is the commonest mutation in cancer?
- p53 (regs cell cycle, not to do w/ kinases)
167
Why is hard to target cancer cells?
- as by time see it, a lot has already gone wrong
168
What are the 5 properties of a good drug?
1) specific against particular target (tight binding, ideally nM or better)
2) doesn't bind to other targets (ie. no side effects)
3) able to get to target, pref when given by mouth (small to get through membrane, soluble, fairly hydrophobic)
4) suitable pharmokinetics = good bioavailability, delivered to target, low drug metabolism, slow excretion
5) drug and metabolites not toxic
169
Why do you need to balance properties of a good drug?
- good solution to 1 often means poor at others
- good at specific binding to target and not binding to others often means bad at getting to target
- improved pharmokinetics often means bad at specific binding, not binding other targets and getting to target
170
What does the best specific binding to target usually come from?
- something that looks like natural ligand
171
How was HIV protease inhibitor drug designed, by looking at specific binding to target?
- HIV protease cleaves no. targets, inc seq Asn-Tyr-//-Pro-Ile
- 1 early drug was invirase --> changed peptide bond in substrate so no longer cleaved (looks bit like intermed after cleavage by HIV protease), but is peptide so easily metabolised
- redesigned so looks as un-peptide like as poss = nelfinavir
172
Why are small drugs usually best?
- best ligand efficiency (large as poss)
| - larger molecules often fail due to toxicity problems
173
How is ligand efficiency calc?
- binding free energy / no. heavy (non H) atoms
174
What is bioavailability?
- prop of drug that enters circulation when introd into body
175
When is getting to target in body and ideally nowhere else particularly a problem?
- for drugs targeting brain, due to blood-brain barrier
176
What do preclinical stages of drug dev involve, and best case how long do they take?
- target discovery (1 yr)
- target validation (1 yr)
- lead discovery (2-3 yrs)
- transition to dev (2 yrs) --> testing on increasingly more realistic and expensive models:
- pure proteins
- cell system
- mice/rats
- larger animals (cats/rabbits etc.)
- monkeys
177
Why are preclinical stages not always able to filter out any drugs which are going to fail?
- human metabolism diff
- human genetics v diff
- human diet v diff (smoke/drink etc.)
178
In the best case what does clinical stages of drug dev involve, and how long does this take?
- phase I = healthy volunteers (dosage, safety)
- phase II = ≈100 sick volunteers (effectiveness)
- phase III = ≈1000 patients (side effects)
- in total 6 years
179
What do clinical trials have to prove for a drug to succeed?
- statistically prove that drug has better clinical effects than existing drugs on market
180
What happens when start having problems w/ drug delivery/metabolism/excretion/toxicity?
- add bits and change things around
| - almost always means specific binding to target and getting to target get worse
181
What are 'quick to kill' procedures, and why are many drug companies trying to dev them?
- kill drug as fast as poss so not wasting time/money on it
| - many drugs fail in clinical phases (too late) --> ≈95% anticancer drugs fail here
182
What is the conventional route drug companies take when dev drugs?
- pick target
- screen large no. small molecules to come up w/ lead compound
- put through increasingly difficult set of tests to see if suitable as commercial drug
- at same time start work on successors
- can only afford this by making lots of money from few drugs that make it though ("blockbuster" drugs)
- drugs patented (20 yrs)
- drug dev long so often <10 yrs to make money before patent ends
- after this other companies can make much cheaper generics
183
What does major shake up going on in pharma industry currently involve?
- mergers/rationalisation/downsizing
- outsourcing
- increasing reliance on smaller (biotech) companies
184
Why is ideal drug often a TS analogue?
- *DIAG*
- enzs wok by stabilising TS and/or destabilising reactants (req specificity)
- hard to bind poorly to substrate but strongly to TS as look similar (but good enz will bind TS 10^10 - 10^12 x better)
185
Why are kinases a v druggable target?
- 1 substrate is ATP
| - ATP looks like good drug (small, hydrophilic, polar etc.) --> meets criteria except not binding to other targets
186
Why is it a big problem if drug is a competitive inhibitor of ATP binding?
- will interfere w/ other proteins that bind ATP --> ie. other kinases, and any enz that makes or uses ATP (glycolysis, TCA cycle)
187
Why is Gleevec a good solution to having a drug that is an ATP analogue?
- binds unusual and v specific conformation of ATP binding site, apparently only found in Abl kinase
188
What is the therapeutic index?
- ratio between toxic and therapeutic doses
| - should be as large as poss, but ≤10 for many drugs
189
Is warfarin a good drug, if judge by therapeutic index?
- no, as TI ≈ 2
190
Should drugs only be tested on 1 target?
- no, should test against lots in case strike somewhere else
| - eg. viagra
191
What is Lipinski's rule of 5?
- orally active drug violates no more than 1 of:
- -> ≤5 H bond donors
- -> ≤5 H bond acceptors
- -> mole weight < 500Da
- -> octanol/water partition coefficient log P<5 (ie. not too hydrophobic)
192
What are Lipinski's rule of 5 based on?
- empirical set of rules
| - no theoretical justification, but observed from looking at lots of compounds that work as drugs
193
What is the partition coefficient, and what does it mean?
- if log P<5, then P < 10^5, so can still be quite hydrophobic, ie. 10^5 more soluble in octanol in water
- often not met
194
What happens if a drug is too hydrophobic?
- sits in membrane, gets metabolised and causes toxicity
195
How popular are kinase inhibitors as drug targets?
- 2nd most targeted of current rule-of-5 compliant experimental and marketed drugs (GPCRs 1st)
- but represent largest prop of druggable genome --> most effort put into looking at pot drugs in this area
196
How specific are kinase inhibitor drugs?
- most are competitive inhibitors of ATP binding --> implies would target all kinases, but not as bad as this
197
Do we know the function of all human kinases?
- only about a third (of 518)
198
Why is staurosporine not a good drug?
- great kinase inhibitor, but inhibits almost every kinase well (completely nonspecific)
- so low TI
199
What can staurosporine be used for?
- as a tool compound to see bio consequence of inhibiting a kinase
200
Is imatinib (Gleevec) a good drug, why?
- yes, much more specific than staurosporine
| - but not as good an inhibitor
201
What are the targets for RTK-based drugs, and what types are there for each?
- target ec domain --> monoclonal antibody, monomeric ligand, soluble receptor
- target catalytic domain --> interfere w/ ATP binding, interfere w/ substrate phosphorylation
202
Why can't proteins be taken by mouth?
- would be degraded
203
Why are monoclonal antibodies good drugs?
- antibodies v specific and can block site for the 1 correct ligand to bind
204
What are 2 recent 'blockbuster drugs'?
- Gleevec/imatinib
| - Herceptin (monoclonal antibody)
205
What is the issue w/ proteins or 'biologic' drugs?
- expensive to prod and difficult to maintain consistency
- as have to express in cells, cell mutate/change/don't prod same amount
- so need to check purity, which can be hard to demonstrate
206
What is the adv of protein or 'biologic drugs'?
- much more specific
207
How does Herceptin work?
- works on EGF receptor, which forms heterodimer w/ HER2 (ErbB2)
- stops binding between them and activation
208
What is the importance of HER2 in cancer?
- activation of HER2 leads to cell growth and division
- inhibition stops cell in G1 phase
- HER2 overexpressed in many cancers inc several breast cancers (up to 100x)
209
What are the effects of Herceptin?
- turns off signal
- also may downreg expression of receptor, prevent proteolytic cleavage of ec domain (leading to metastasis) and induce immune response to cell
210
How effective is Herceptin, and why?
- 70% patients (w/ high HER2 expression) do not respond - epigenetics means diff cells behave diff
211
What are the problems w/ Herceptin?
- relapse (drug resistance)
- v expensive --> approx £80,000 per patient per year
- around 10% patients will dev heart disease
212
Why is acquired drug resistance a major problem for cancer therapy, and how can this be avoided?
- often means drugs only work for a few months before lose effectiveness
- need to understand resistance mechanisms and use drug combos (like in virology)
213
What is an eg. of an attempt to dev a monomeric ligand drug?
- made growth hormone covalently attached to monomeric ec receptor domain
- will bind to and cover part of GH surface, protecting it, so makes much longer lived GH (whatever effect is, will happen for longer)
- so good pharmacokinetics --> slower clearance and less immune problems, so fewer side effects and less freq administration (but still needs to be injected)
- could have 1 of 2 effects when binds:
1) block 1 side of receptor so can't dimerise, blocks ds activity = ANTAGONIST
2) receptor domain comes off, induces dimerisation = AGONIST
- turned out to be long lived agonist --> encourages growth so treatment for Dwarfism, but had lots of toxicity problems
214
What is Gleevec used to treat, and what characterises this disease?
- chronic myelogenous leukemia (CML)
- characterised by chromosome rearrangement which fuses abl kinase w/ bcr, so abl constitutively active
- Gleevec inhibits this
215
What makes Gleevec a good drug?
- v bioavailable (98%)
| - half life in body = 18 hrs and major metabolite = 40 hrs
216
Why did Gleevec put the company that prod it in a good patent position?
- approved for several other diseases
| - so could take out add patents for diff uses
217
What pot solutions are there to Gleevec resistance mutations?
- dev drug cocktails --> look like best therapy, as even if multiple redundant routes, should be able to hit them all
- dev inhibitors against mutant kinases prod by cancer cells
218
How does resistance to Gleevec dev?
- mainly by mutations --> alt binding site
219
How are notch, hedgehog and wingless similar to auxin signalling?
- use protein degrad
| - so essentially irreversible
220
What are the ligands for notch?
- Jagged, Delta, Serrate etc.
221
What is notch?
- a receptor
222
What does notch signalling cause?
- generally lateral inhibition --> differentiated cell stops neighbour differentiating
223
Where is lateral inhibition important?
- in new neurons in epithelial sheet, to prevent neighbours also becoming neurons
224
How does lateral inhibition occur?
- ligand expressed and displayed on cell surface, by interacting w/ notch (also on cell surface)
- so only works by cell to cell contact --> v local signalling, 1 cell to immediate neighbours
225
What is lateral inhibition good at initiating?
- differentiation
226
How does notch signalling occur?
- all receptors and ligands attached to cell membrane by TM seq
- notch (like all proteins expressed on cell surface) is synthesised in ER and at some point in golgi cleaved, so break in polypeptide chain
- binding ligand pulls notch away from cell membrane, cleaving TM seq
- this seq moves to nucleus and binds to TFs
- cleavage is by protease presenilin
227
How was presenilin 1st identified?
- in Alzheimer's patients
| - as cleaves Aβ from amyloid precursor protein (APP)
228
What are the functions of notch?
- controls maintenance and self renewal of stem cells
- cell cycle progression
- differentiation
- can also be oncogene or tumour suppressor (in diff cells)
229
What diff types of genes control dev of Drosophila body plan?
- egg polarity genes
- gap genes
- pair-rule genes
- segment polarity genes
- homeobox genes
230
What do egg polarity genes do?
- axis formation
- bicoid mRNAs at anterior pole and nanos mRNAs at posterior pole
- diffuse, forming opp grads
231
What do gap genes do?
- inhibited or activated and expressed in bands down body (restricted expression domain)
- eg. kruppel activated by bicoid but inhibited by high levels of hunchback
232
What do pair-rule genes do?
- expressed in bands prod by gap genes, but diff levels at each end
- initiate segmentation
- eg. eve2 req bicoid and hunchback, repressed by giant on 1 side and kruppel on other
- autoreg to make it sharper
233
What do segment polarity genes do?
- build on prior divisions to make sharply defined segment boundaries
- much narrower expression domains (single rows of cells in ring round embryo)
- eg. engrailed, hedgehog, wingless --> reinforce each other to form segment boundaries
234
What do homeobox genes do?
- target features to diff segments
235
What is the relationship between hedgehog and wingless?
- hedgehog maintains wingless transcrip and wingless maintains hedgehog transcrip
- paracrine --> diff effects on target dep on conc
236
What is the wingless signalling pathway (when present)?
- Wnt activated frizzled
- frizzled activates dishevelled (moves from internal to cell membrane)
- dishevelled inhibits GSK3/APC/Axin complex so doesn't bind to β-catenin
- so β-catenin can move to nucleus and activate transcrip of Wnt targets
237
What happens when Wnt is not present?
- in inactive cell there is constant degrad of signal
- frizzled not activated, so can't act on dishevelled
- β-catenin binds to GSK3/APC/Axin and GSK3 (kinase) phosphorylates it --> leading to recognition by ubiquitination pathway, so proteosomal degrad occurs
238
Why can't Wnt diffuse far?
- attached to ec secreted vesicles
239
What are the human equivalents to Drosophila hedgehog?
- sonic
- desert
- indian
240
Why is the hedgehog mutation named that?
- mutation makes larvae spiky
241
What do defects in hedgehog pathway cause in other animals?
- dev abnormalities, inc cyclopia
| - prob also involved in reg of adult stem cell growth, dev of hair follicles, regrowth of salamander limbs etc.
242
What is hedgehog and how is it attached to membrane?
- protein ligand
| - normally attached to cholesterol at 1 end and palmitoylated at other, attaching it to membrane
243
How might hedgehog get to its targets, and what is the result of this?
- may get to target by attachment to secreted vesicles --> makes it v short range signal, activates thin strip of cells adj to secreting cell
244
What is the hedgehog signalling pathway (when present)?
- hedgehog degrades patched receptor
- so smoothened signal transducer not activated
- so kinases inhibited
- resulting in full length Ci, which activates transcrip
245
What happens in absence of hedgehog?
- patched receptor active
- activates smoothened signal transducer
- so kinases phosphorylate Ci, resulting in cleavage
- part of cleaved protein moves to nucleus and acts as transcrip repressor
246
Where is the NF-κB signalling pathway found?
- often found in inflam and innate immune response --> ie. response to stress
- also important in memory
- also in dev --> eg. dorsal/ventral patterning
247
What can inapprop activation of NF-κB pathway lead to?
- pain and cancer
248
What makes the NF-κB pathway rapid acting?
- autoinhibited (signalling removes inhibition)
| - doesn't req new protein synthesis
249
What happens during the NF-κB pathway?
- TNFα trimer binds TNFα receptor and activates IKK complex
- NF-κB has NLS, but hidden by IκB (an inhibitor w/ NES)
- IKK complex phosphorylates IκB --> causing it to be ubiquitinated and degraded in proteasome
- this digestion of IκB req to liberate NF-κB
- NF-κB translocates to nucleus and activates transcrip of target genes --> cytokines, iNOS etc. (but also IκB, implying typically only active 1hr before shut down)
MBB343 - Biochemical Signalling
flashcards
Decks in class (2)
# Cards
