13
Begin 03/17/09
The next overhead is a repeat of the triggers of apoptotic pathways. We are interested for the most part in irradiation at the far left, under which we will subsume chemical damage, because much chemical damage is radiomimetic, and also ligand-activated receptors. As you can see from the slide, irradiation- and chemically-induced apoptosis involves p53. Though not explicitly shown in this cartoon, the ligand-activated systems can also be mediated by p53, which is why the receptors and ligands are also of interest to us. Remember that we are concerned with causes of dysfunctional or non-functional p53 stemming from chemical damage to the p53 gene, and the nature of the apoptosis triggers in the table, whether the trigger is DNA damage by chemicals or whether it is ligand-mediated, are incidental. We are describing the apoptotic pathways because the loss of such pathways is connected with p53 function and lack of apoptotic pathways is also important in the events leading to cell transformation. Major apoptotic pathways are triggered by activation of closely related cell surface receptors called Fas and TNF-R1 (TNF = tumor necrosis factor). As in the case of other signaling pathways, binding of extra cellular ligands activates both of these receptors. Transcription of Fas is induced by p53 and Fas is also activated by p53, and p53 is also involved in mediating the TNF-R1 pathway, although the mechanisms have not been well defined. The receptor pathways are illustrated on the next few slides. If you look at apoptosis in Genes VIII, there is a cartoon which obfuscates these two pathways by mixing them in the same cascade, but the overheads represent the latest versions. In both Fas and TNF-R1, the apoptotic signal is internalized through an 80-residue domain near the C-terminus called appropriately the death domain.
[TNF-R1 pathway]
The first slide in the series shows the TNF-R1 path. The receptor trimerizes and a multiprotein complex forms at the receptors. TNF-R1 binds proteins TRADD and FADD that serve as an adaptor for another protein, Caspase-8 (Genes VIII also used the name FLICE, but this name has been used less recently), which has protease activity that is activated on binding. The Caspase-8 protein belongs to a family of caspase proteases that is named from their cysteine-aspartate protease activity, that is, they hydrolyze the peptide bond betweenCys and Asp.Caspase 8 is part of a cascade that terminates with apoptosis. The slide shows non-death-domain initiated apoptotic pathways.
The next rather busy slide, which needs to be explained in sections, shows the Fas pathway.
[Fas pathway]
Fas receptors also trimerize and directly bind FADD, forming a complex which recruits Caspase-8 and initiates a cascade that eventually branches to produce several effects,with several caspase members involved in the apoptotic cascades. Caspase-8 can directly activate caspase 3, which starts a pathway to cell death. Caspase 8 can also cleave a protein called Bid. The C-terminal domain of Bid translocates to mitochondrial membranes and causes the release of cytochrome c (an electron transport-ion pump protein, heme containing but not to be confused with cytochrome P450), which complexes with a cytosolic protein Apaf-1 and another caspase, Caspase-9, to form an agglomeration called an apoptosome. The apoptosome leads to cleavage of pro-Caspase-3 to Caspase-3 leading to apoptosis. The apoptosome is illustrated in more detail on the next slide:
[Apoptosome and cascade]
Beyond this point the apoptotic pathway branches to multiple end-points as the cartoon illustrates, via pathways include activation of a number of proteases.
One target of Caspase-3 branching pathways is the DNA fragmentation factor (acronym DFF), which is activated by proteolysis. One fragment of DFF activates a nuclease that degrades DNA.
Returning to the Fas pathway, there are antiapoptotic proteins that serve to inhibit the apoptotic pathway. The Bid protein belongs to a family named from the protein Bcl2. Some Bcl2 members are required for apoptosis to proceed, while others inhibit apoptosis. Bcl2 itself inhibits apoptosis. In fact, over-expression of Bcl2 can result in tumorigenesis, which can be one consequence mutations arising from permitting the cell cycle proceed in the presence of unrepaired damage, and so Bcl2 is classified as an oncogene. Another Bcl2 protein, BAX appears to counteract Bcl2 and in fact may be necessary for p53-mediated apoptosis, as suggested in the Fas pathway slide.
As implied in the slide on the TNF-R1 pathway, there are multiple pathways to apoptosis and the next overhead shows one of the alternative pathways
[Genes VIII, Fig. 29.53]
which does not act through the death domain. A protein Daxx binds to the receptor complex and activates the kinase JNK, which in turn activates the transcription factor c-Jun, which we have encountered as an example of a b-zip motif, and was also classified as a protooncogene. The cartoon indicates that Bcl2 can inhibit this pathway also.
The next slide is a cartoon summarizing the role of p53 in apoptosis:
[Fig. 6 from CRT review, 2001]
The heavy arrows indicate sequence-specific transactivation-dependent pathways, meaning that the genes involved in those pathways are targets of p53 transcription. These include the Fas receptor ligand, the Bax protein (pro-apoptotic member of the Bcl2 family); PIG3, which we have identified the quinone oxidoreductase analogue that generates ROS; PIG8, whose function is not yet defined, but which seems to act to inhibit cell colony formation and induces apoptotic death – the latest citation I could find indicates that PIG8 operates by binding Bcl2, not as an ROS generator as implied by the slide from the CRT review; IGF-BP3, which acts to block a survival signal receptor. The transactivation-independent pathways indicated by hatched lines are, at this juncture still speculative, so we will not describe them here.
Negative alleles – mutated genes that code for non-functional proteins - exert a dominant effect, meaning that when functional and non-functional p53 proteins are present, and the non-functional protein determines cellular responses. In the case of the p53 protein, which forms tetrameric complex, this means that heteromeric wild-type:mutant complexes are non-functional.
The p53 gene itself has been shown to be altered by chemically induced mutations and half of all human tumors contain mutant forms of p53. The observation that mutated p53 is present in a wide variety of tumors indicates that its activity is a general response to damage, and not specific to a particular tissue or lesion. p53 provides a second example, in addition to ras, where chemically induced mutations can be associated mechanistically with tumorigenesis. An additional, and very interesting observation made in the CRT review, is that over-expression of mutant p53 is sufficient to immortalize primary cell lines and, in combination with one additional mutation of Ras or Myc proteins, primary cell lines can be transformed. This, in effect, establishes a kind of minimum “2-hit” requirement for malignant transformation so the Ras + p53 mutations represent at least one pathway that is completely mediated by chemically derived lesions.
The direct connection between chemically induced mutation of p53 and cell transformation lies in a study has been made of the mutations in the p53 protein isolated from human lung tumors, with the results displayed on the next overhead:
[Science1996, 274, 430-432; Fig. 1, p 53 mutations in lung cancer]
Three hotspots were specifically connected with chemical lesions. They are all G → T transversions, appearing at positions 157, 248 and 273 on the coding strand, all at Gua in a 5'-CpG-3' doublets. The hotspots all map in the DNA-binding domain of p53. These hotspots have been correlated with sites of adduct formation by benzo[a]pyrene diolepoxide through exposing cells to the diolepoxide and examining the sequence surrounding the adduct locations, which were excised (in vitro) by the UvrABC nuclease complex. The principal sites of adduct formation coincided exactly with the mutational hotspots derived from sequencing p53 from lung tumors. Selectivity for reaction at these sites does not seem to be determined by chromatin structure - this possibility was ruled out by repeating the diolepoxide exposure experiment with pure genomic DNA (i.e., DNA not associated with proteins as would be the situation in the nucleus) and finding the same hotspots. Additional support for the correlation of the mutational hotspots with sites of diolepoxide adducts derives from the observation that the G → T transversion is the predominant mutation generated by benzo[a]pyrene diolepoxide (confirm with the mutational spectrum from codon 12 of Ras in the A/J mouse lung tumors and the fact that the transversion is otherwise uncommon) and the hotspots at Arg 248 and Arg 273 are associated with other tumors in addition to lung tumors.
The mutations at residues 273, 248 and 157 are of particular interest from our perspective because they have been connected with action of BPDE. The next slide shows a crystal structure of a core fragment of a WT p53-DNA complex. As the slide shows, residues 248 and 273 are Arg, and are involved with DNA contacts, Arg 248 sits in the minor groove and Arg 273 contacts a backbone phosphate. The mutations at 248 and 273 are both Arg → Leu, which substitutes a hydrophobic residue for a cationic arginine, and obviously will obliterate the electrostatic association with DNA. The role of Val 157 seems less obvious, but it has been demonstrated that p53 is thermodynamically unstable and readily unfolds at temperatures slightly above body temperature, so the mutation at 157, Val → Phe, most likely destabilizes the structure to the point where the p53 is not functional. There are several other sites that show high levels of mutation, although the direct correlation with a BP-induced lesion has not been rigorously demonstrated. Arg 249, adjacent to Arg 248 is also in a region of DNA contact. Arg 282, Arg 175 and Gly 245 all play a role in the structural integrity of p53. The importance of His 179 is obvious, since, along with 3 Cys residues, it is a coordination site of Zn2+ which is one of the organizational components holding the whole show together. Mutations at these hotspots clearly will disrupt the affinity of p53 for the consensus sequence.
ACTIVATION OF CARCINOGENIC XENOBIOTIC CHEMICALS
This section of the course is the chemistry of activation and reactions of chemical carcinogens and used to comprise the entire body of chemical carcinogenesis. With the meteoric development of molecular biology, metabolic activation was ignored but is now becoming an area of interest again because of the realization that polymorphisms in metabolizing enzymes are important in determining sub-populations that are either highly resistant or highly susceptible to specific xenobiotic chemicals. By far and away the greatest proportion of carcinogenic compounds are not in themselves active, but must undergo metabolic transformation in order to exert carcinogenic effects. Of the metabolic processes, the most important is probably oxidation by the mixed function oxidase system. ENVR 430 gave an overview of this system, but we will take it up here againfrom a more mechanistic point-of-view.
The terminal oxidase in the MFO system and the component directly responsible for oxidizing substrates is a heme-containing protein called cytochrome P450 (written as shown, although in the past other conventions have been used).
[Overhead]
The name arises from the appearance of an intense absorbance band at 450 nm in the electronic spectrum of the CO complex of the reduced enzyme, which originally led to its discovery. As we shall see, “cytochrome P450" actually represents a family of enzymes, having molecular weights of 45-60 kDa. The cytochromes P450 are present in both prokaryotes and eukaryotes. In prokaryotes the enzyme occurs in a soluble form; in eukaryotes it is membrane-bound, located primarily in the smooth endoplasmic reticulum. At the active site of cytochrome P450 is iron protoporphyrin IX, the same iron porphyrin that serves as the oxygen carrier when it is at the active sites of hemoglobin:
[Part 4, ENVR 740,structure of PPIX]
In cytochrome P450, the iron porphyrin is noncovalently bound to the protein through the sulfhydryl group of Cys (as cysteinate anion) in a cavity of the protein that defines the active site. In the presence of an electron donor and dioxygen, the cytochrome P450 enzymes catalyze the transfer of one atom of the dioxygen to a substrate, with the release of water. Coordination of the heme by Cys is unusual among heme containing proteins and is important in the mechanistic pathway of reductive cleavage of dioxygenby cytochrome P450 the Cys coordination is critical in the oxygen transfer reactivity of the cytochromes P450. The catalytic cycle is shown on the next overhead:
[ENVR 740, cytochrome P450 cycle]
In the resting state of the enzyme iron is in the ferric (3+) valence state. The first step involves the non-covalent binding of substrate in the distal pocket at the active site. (Distal refers to the face of the protoporphyrin opposite to the face coordinated to the Cys.) This step is followed by a one-electron reduction of iron to the ferrous (2+) state, in which iron has a strong affinity for dioxygen.Dioxygen then complexes with iron to yield a ternary heme·O2·substrate complex.At this point, it is possible to uncouple the reaction sequence by reduction of the dioxygen to superoxide through a metal-to-ligand electron transfer to return the heme to the ferric (3+) resting state. The superoxide anion can dissociate, which leaves the enzyme in the resting state, but has released a progenitor of ROS. This process is referred to as futile cycling of P450, because it involves the consumption of O2 and reductant without generating any product from the substrate. (The following in red text will be treated on 03/19/09, but is posted here for completeness) If the cycle continues, reduction of the heme·O2·substrate complex by a second electron leads to a peroxoiron complex, formally O22-. Recent work supports protonation of the peroxoiron complex to generate a hydroperoxo complex, formally a complex of the mono anion of hydrogen peroxide, called cpd 0. Studies in model systems suggest this complex could oxidize unsaturated substrates predominantly by epoxidation. However, interest in cpd 0 has waned, and an alternative pathway, which is the one initially proposed and which has always represented the major reaction pathway, is addition of a second proton which results in the release of water with the transient formation of an extremely reactive monooxygen complex called compound I, that is oxidized by two electrons above the ferric resting state of the enzyme. The oxygen atom is rapidly transferred to the substrate, followed by release of the oxidized substrate and return of the cytochrome P450 to the resting state. Compound I can hydroxylation carbon and epoxidize double bonds. Investigators have used site-directed mutagenesis (mutating specific amino acid residues) to perturb the hydrogen-bonding network and proton sources at the active site of cytochrome P450 to alter the balance of the two pathways for oxidation of specific probes having both a double bond and an activated C–H bond. These experiments are interesting in terms of their implications for the effects of single nucleotide polymorphisms, and we will discuss them again at in that context.
1