When Good Signals Go Bad

BCHM2072 2006 Lecture 18

WHEN GOOD SIGNALS GO BAD:

Signal Transduction and Disease

G-protein coupled receptors.

Cholera and pertussis: Examples of excess signal due to deficient termination.

The actions of these two toxins are similar in many ways, but there are differences too….

When cholera toxin is released from Vibrio cholera bacteria in the infected intestine, it binds to the enterocytes (intestinal cells), triggering endocytosis of the toxin. Next, the cholera toxin cleaves to become an active enzyme. The enzymatic fragment of the toxin is an ‘ADP-ribosyltransferase’, an enzyme that catalyses the transfer of the ADP ribosyl part of NAD to the stimulatory G-protein subunit (Gsa). (This reaction is called ADP-ribosylation). Normally, a G-protein mediated signal is terminated by the GTPase activity (GTPà GDP) of the G-protein itself. Ribosylation interferes with the GTPase activity of Gsa. Thus, GTP cannot be hydrolysed, the G protein is locked in its active GTP-bound form; thereby it can continually stimulate adenylate cyclase to produce cAMP. (à the accelerator is stuck ON). The high cAMP levels activate the cystic fibrosis transmembrane conductance regulator (CFTR), causing a dramatic efflux of ions and water from infected intestinal cells, leading to watery diarrhoea (yuk!).

There are also other types of a subunits of G proteins that inhibit adenylate cyclase activity (= inhibitory G-proteins (Gia). This is an additional method of terminating the G-protein mediated signal.

These Gia proteins are also activated by a G-protein coupled receptor.(in this case, its an inhibitory-type receptor that binds an inhibitory ligand) The mechanism of control of these is the same as the stimulatory proteins, but the active protein has the opposite effect: Gia proteins are active, and inhibit AC, when bound to GTP, and inactivated (not inhibiting AC) when GTP is hydrolysed to GDP.

The symptoms of Whooping Cough (Pertussis) result from the action of a toxin produced by the bacteria Bordetella pertussis. Like Cholera toxin, Pertussis toxin is an ADP- ribosyltransferase, but it adds ADP-ribose to Gia. This prevents GDP release so that the inhibitory G protein is locked in the inactive state (That is, the inhibitory protein cannot bind GTP, cannot be activated so cannot perform its normal function to inhibit adenylate cyclase). Here, the brakes are disabled. The conversion of ATP to cyclic AMP cannot be stopped à ñ cAMP. B. pertussis also produces another toxin that acts as an adenylate cyclase itself.

Receptor Tyrosine Kinases (RTKs):

Her-2+ breast cancer: an example of excess, unregulated (constitutive) initiation of signal due to receptor overexpression.

Revision: Normally, binding of a ligandà dimerization of receptor chains à activation of their intrinsic kinase activity à phosphorylation of specific tyrosine residues in the receptor cytoplasmic tails. These phosphorylated tyrosines are recognition sites (eg by SH2 domains) for intracellular signaling intermediates (eg enzymes like PLCg and IP3 or adaptors that recruit Ras), which link receptors to downstream transduction cascades, including activation of MEKs and MAPKs.

Receptor Tyrosine Kinases (= receptors with intrinsic tyrosine kinase activity) are often receptors for growth factors. It makes sense that overexpression of growth factor receptors à cell proliferationà tumours. Many breast cancers (~20% to 30% ) are associated with overexpression of an epidermal growth factor (EGF) receptor chain called Her-2 (this chain is also called erbB-2, and the dimeric receptor is called erbB just to be confusing! )[1] The Her-2 protein actually is one chain in the dimeric receptors for EGF-related ligands (Receptors for different EGF-related ligands derive their specificity from the identity of their individual chains: different chains form different homodimeric and heterodimeric complexes) Overexpression of this receptor protein promotes spontaneous receptor dimerization in the absence of a ligand à constitutive receptor activation.

Although the signaling pathways induced by the erbB receptor are incompletely characterized, it is thought that, like other RTKs, it leads to activation of MAPKs (mitogen-activated protein kinases) that stimulate cell division by phosphorylating intranuclear transcription factors (remember myc, Jun and Fos?).

It has also been suggested that the mechanism that causes increase in proliferation is phosphorylation of a cdk2 inhibitor which prevents this inhibitor from entering the nucleus. (phoshorylation of the cdk inhibitor requires the phosphoinosityl 3-kinase/ protein kinase C pathway)

Cdk inhibitors normally arrest the cell cycle in response to anti-proliferative signals. (Most of you have heard of cyclin dependent kinases (cdk). They are required for the cell to pass through checkpoints in the cell cycle. Cdk2 is responsible for controlling entry into mitosis (as part of the maturation promoting factor (MPF: Remember this from Dr Nichols lecture in MBLG2071?) If the inhibitor of cdk2 is phosphorylated, it cannot enter the nucleus to inhibit cdk2, cdk2 remains active and the cell can continue unchecked into mitosis àuncontrolled cell proliferation.

Her-2-positive cancer is bad news, as Her-2 overexpression often results in more aggressive breast cancer cells that may not be as responsive to standard breast cancer treatments, including certain regimens of chemotherapy. However, its not all bad news. Herceptin® is a monoclonal antibody that binds to the extracellular part of Her-2 and prevents dimerisation with the other chain in the receptor. This prevents the signal being initiated à prevents activation of MAPKs and phosphorylation of the cdk2-inhibitorà cdk2-inhibitor can enter the nucleus and inhibit cdk2 à cell cycle arrest in G1. Herceptin therefore blocks the pro-proliferative signals initiated by Her-2 overexpression.

Dr Easterbrook-Smith also mentioned that ~15% cancers (and 66% of melanomas) involve a mutation in Ras. Active Ras stimulates downstream effectors including MEK and the MAPKs. Ras, like other G-proteins, is active when bound to GTP, and also is a GTPase, which inactivates itself by hydrolysing GTPà GDP. Many Ras mutations prevent this hydrolysis, so then Ras is constitutively active (by that, we mean it is ON all the time, not only when it gets a signal to be on)à uncontrolled proliferation.

Achondroplasia: an example of deficient signal due to receptor mutation.

If excess signal through an RTK à excess cell proliferation, it makes sense that a deficient signal à deficient cell proliferation. One such example if achondroplasia (a = without, chondro = cartilage, plasia = cell proliferation/growth). This condition is due to a mutation in the Fibroblast Growth Factor (FGF) receptor. Chondrocytes in the growth plates of long bones do not receive a signal to proliferate, and the bone growth is diminishedà short stature. The mutation is at the same nucleotide in ~98% of cases à glycine (neutral) to arginine (positive) substitution in the transmembrane domain of the receptor. Interestingly, this mutation arises spontaneously in ~80% of cases.

Cytokine receptors:

Revision: Cytokine receptors are those that are not tyrosine kinases themselves (like the RTKs), but are associated with intracellular tyrosine kinases (JAKs) which phosphorylate tyrosines in the receptor tail to facilitate binding of STATs (Signal Transducers and Activators of Transcription) via an SH2 domain in STAT (remember SH2 domains can bind to only to tyrosines that are phosphorylated) STATs then become phosphorylated themselves, and can then dimerise, translocate to, and enter the nucleus where they activate transcription. The genes they activate depends on the specific receptor that initiated the signal (specific receptors associate with specific subsets of the many different JAKs and STATs). Cytokines are important mostly for signals in the immune system eg pro- or anti-inflammatory, activation or inhibition and development of immune cells.

Rheumatoid arthritis: an example of excess initiation of signal due to ligand overexpression.

Rheumatoid arthritis is characterised by severe, destructive inflammation of synovial cells in the joints. One proposed mechanism for this disease is an excess of the pro-inflammatory cytokine TNFa (tumour necrosis factor). Much research is currently focused on developing drugs to manipulate this pro-inflammatory pathway. One treatment in use is a monoclonal antibody (infliximab, but you don’t need to remember the name) that binds specifically to TNFa, preventing it binding to the receptor. This obviously prevents all downstream events. Another target of research is SOCS (Suppressor Of Cytokine Signals). These proteins are JAK inhibitors that normally function as negative feedback in the pathway, as SOCS expression is stimulated by the same cytokine that initiated the pathway. A drug that could stimulate SOCS would make a useful anti-inflammatory. What other targets/mechanisms for anti-inflammatory drugs could you envision?

Severe combined immunodeficiency (SCID): an example of deficient signal due to receptor mutation.

Cytokine receptors are usually heterodimers (2 different chains). The identity of the chains determines specificity. Many cytokine receptors have one chain in common: the ‘common gamma chain’ (gc). In X-linked severe combined immunodeficiency syndrome (XL-SCID), the gc chain is mutated, and receptors for many cytokines are absent or inactive. Autosomal recessive SCID (AR-SCID) is the result of a mutation in JAK or in the ILx receptor chain (the ‘x’ is specific to the ligand: eg IL7 (IL=interleukin) uses the IL7 chain in its receptor). SCID causes failure of development of T cells and NK cells of the immune systemà severe infections early in life. Treatment for these syndromes was originally strict isolation (‘the bubble boy!) and later, bone marrow transplants to replace the defective cells. Recently, there has been a successful gene therapy trial to cure this syndrome. Gene therapy supplies a functional copy of the gene (delivered in a viral vector) to replace the defective gene. The diagram in the Powerpoint slide outlines gene therapy for a different type of SCID due to deficiency of an enzyme called ADA (adenosine deaminase), but the principle is the same for each mutation.

Intracellular tyrosine kinases:

Chronic Myelogenous Leukaemia (CML): an example of unregulated (constitutive) excess signal due to mutation of a kinase (abl).

abl is an intracellular (non-receptor) tyrosine kinase. Like other TKs, it catalyses the transfer of a phosphate from (usually) ATP to a tyrosine residue in its targets. It interacts in pathways that respond to pro-proliferative signals in myeloid cells. Its activity is controlled by another protein that binds to its SH3 domain. In CML, a part of chromosome 22 carrying the gene for bcr (bcr = breakpoint cluster region) translocates to chromosome 9, right in front of the gene for abl. (this is now called the Philadelphia chromosome). The result is that a ‘fusion’ protein is translated, called bcr-abl. The fusion protein is mainly cytoplasmic, compared with abl, which is mainly nuclear. The bcr domain interferes with the SH3 domain of abl, preventing regulation of abl’s TK activity. bcr abl is therefore a constitutive TK. This means it rampages around, phosphorylating, and activating, its pro-proliferative / anti apoptotic targets even when the cell is not receiving a proliferative signal. (eg bcr-abl activates RAS signaling through the Grb2 adaptor molecule which interacts specifically with a phosphorylated tyrosine in bcr à activation of MAPKs. The phosphatidyl inositol 3' kinase (PIP3K) pathway is also activated, as is transcription initiation by myc. bcr-abl also recently been found to activate STATs). This of course causes uncontrolled proliferation of myeloid cells à leukaemia.

Recently a treatment has been developed for CML, which previously had a poor prognosis (median survival: 4 yrs with conventional therapy). Glivec® (or Gleevec ®) binds at the kinase active site, and competes with ATP. It is quite specific for abl, but not other TKs. Developed in the late 1990s, Glivec was one of the first examples of rational drug design, where the structure of the active site was known, and suitable inhibitors were designed and tested. It is also considered to be one of a new generation of ‘magic bullet’ drugs (as is Herceptin®) that targets only cancer cells, rather than targeting all rapidly dividing cells.

X-linked agammaglobulinaemia: an example of deficient signal due to mutation that prevents binding of a kinase (BTK = B-cell tyrosine kinase = Brutons tyrosine kinase).

BTK is a non receptor TK that is part of a complex pathway that initiates the gene rearrangements that are required to make antibodies. (I won’t go into that here, but it’s very interesting if you’d like to read about it. Gene rearrangement means that the many 1000s of different antibodies we produce (each one unique to a specific antigen) do not need to be made from 1000s of different genes. This phenomenon was a huge mystery for many years). This gene rearrangement is also required for B-cells maturation. BTK is one partner in the pathway initiated by the B cell receptor, and intersects with the phosphoinositide / PK3 pathway à IP3 à ñCa2+ (revise Dr Easterbrook-Smiths lecture). When BTK is mutated, it cannot interact with its partners, and signal transduction fizzles out. This mutation responsible for X-linked agammaglobulinaemia (a = without, gammaglobulin = the old name for antibodies. aemia = blood). Because gene rearrangement cannot proceed, B-cells do not mature and antibodies cannot be produced. Babies born with this disorder have recurrent severe bacterial infection, but are not overly susceptible to viral infections.

For those of you who haven’t done any immunology: Basically (a huge simplification…) B-cells and antibodies fight extracellular bacteria, T-cells fight intracellular bacteria and viruses.

So, you can see that, like Goldilocks, too much or too little signal transduction is not a good thing. Revise these notes in conjunction with Dr Easterbrook-Smith’s lectures. Note how the signal is initiated, what subsequent targets are, and how they are activated, and what the final consequence is (eg transcription) and how the signal is terminated. If you understand the pathways first, you can deduce (usually) the consequences of an interruption to the pathway. You should also be able to envision where drugs might be targeted to treat these diseases.

[1] To really be confusing, it’s also known as Her-2/neu, MLN 19, NEU proto-oncogene, NGL (neuroblastoma- or glioblastoma-derived), p185erbB2 or TKR1…..you can just remember Her-2!).