Moss - What genes cant do - 2003 (522929), страница 35
Текст из файла (страница 35)
Inthe absence of a need for preserving the specific sequence, it is expectedthat the identity of DNA bases will mutate at a regular rate.Localization by complementary binding studies, using a src probederived from chicken to scan the DNA of a variety of other avian speciesincluding the Australian emu, indicated that the src gene was highly conserved in avian phylogeny and thus likely to be of functional significancefor the host. The level of interspecies sequence variation, as assessed bymolecular hybridization studies, was consistent with the characterizationof src as a highly conserved gene (Varmus 1989).
In order to gain furtherconfirmation of the nonviral origins of homologous src genes, fullsequence analysis would have to be performed (i.e., the gene would haveto be cloned and sequenced). The homologous genes found in thegenomes of other avian species were referred to as “c-src,” for cellularsrc. Sequence analysis of c-src demonstrated the presence of introns (noncoding intervening sequences), where “endogenous virogenes have theinsignia of provirus, being composed of continuous coding domains,flanked by repeated sequences” (Varmus 1989).
Introns are the intervening sequences of DNA which are found in eukaryotic genomes butnot prokaryotic or viral genomes. Their presence in c-src genes againsuggested that c-src was not of viral origin. Identification of the sourceof the viral src oncogene as a putatively functional host gene led theresearch program in molecular oncology back to the framework of asomatic mutation model.We said that the RSV transforming gene is indeed represented in normal cellular DNA, but not in the form proposed by the virogene-oncogene hypothesis.Instead, we argued, the cellular homolog is a normal cellular gene, which is introduced into a retroviral genome in slightly altered form during the genesis of RSV.Dialectics of Disorder: Normalization and Pathology as Process139Far from being a noxious element lying in wait for a carcinogenic signal, the progenitor of the viral oncogene appeared to have a function valued by organisms,as implied by its conservation during evolution.
Since the viral src gene allowsRSV to induce tumors, we speculated that its cellular homolog normally influenced those processes gone awry in tumorigenesis, control of cell growth ordevelopment (Varmus 1989).What enabled the retrovirus work to breath new life into the somaticmutation research program was the idea that these viruses pointed theway to the particular genes which, owing to the nature of their normalfunction, could become causes of cancer in the event of mutation. Comparison of the nucleic acid sequences of the oncogenes with the established sequence data for the genes of known proteins (Genes-D),suggested that the oncogene products fall into five protein categories. Allfive of these categories have been associated in some way with growthrelated functions.
These are the following:1. Growth factors2. Growth factor receptors3. Signal transducers4. Protein kinases5. Transcriptional activatorsAn explanatory model emerged which depicted these classes of proteins as the nodal points of a universal growth regulatory circuit. Lesionsat any of these points, according to the theory, could result in carcinogenesis. Appraisal of the significance of this model by its proponents wasnot modest.
In his 1982 review Bishop asserted that. . . it is the retroviruses that have provided the most coherent and penetratingview of tumorigenesis presently available to us. Three features of retrovirusesaccount for this sentiment. First, the oncogenes of retroviruses have provided ourfirst glimpse of enzymatic mechanisms responsible for neoplastic transformation.Second, the diversity of retrovirus oncogenes has provided a rich set of oncogenic agents whose versatility far exceeds that of DNA tumor virus oncogenes,and whose tumorigenic capacities provide separate experimental models for mostmajor forms of malignancy. Third, oncogenes appear not to be indigenous components of retrovirus genomes, but instead have been transduced from normalgenetic loci of the vertebrate hosts in which retroviruses replicate.
Moreover, wehave reasons to believe that the vertebrates from which retrovirus oncogenesderive may participate in tumorigenesis induced by agents other than viruses.Thus while tracking the evolutionary origins of oncogenes, retrovirologists have140Chapter 4been led well beyond the confines of tumor virology, to confront what may bea final common pathway of oncogenesis (Bishop 1982).In the oncogene model of Huebner and Todaro cancer is caused by genes(given the necessary level of activation), but they are genes derived froma foreign source and oriented toward a foreign mission.
Cancer wouldthen not represent an intrinsic potential of the biology of the cellorganism but rather a consequence of a foreign agent, the (virally derived)oncogene, which when activated assumes an executive function in the cell.With the recognition that retroviral oncogenes like src are derived fromnormal host genes which appear to be functional in growth-related activities, a choice had to be made as to how to interpret the manner in whichsuch genes may become involved in a process of carcinogenesis. Either thefirst or second interpretation of the autonomous cell (discussed above)could have been invoked in relation to the oncogene findings.Is cancer the original and intrinsic potential of any living cell becauseit is the unavoidable risk, the organismic downside, of totipotency? Doesthe capacity to adapt to novel situations,—the holistic-developmentalsense of autonomy—bring with it a corallary risk of cellular uncouplingand divergence from the organismic hierarchy? In this view genetic mutations in cancinogenesis would in effect be viewed as permissive and notas directive.
Mutations would serve to free the cell from those processeswhich developmentally steer it into a stable pathway of terminal differentiation. Alternatively, is the role of genetic mutations determinative ofthe cancer phenotype? Does the potential for cancer only first arise aftermutations have specified an aberrant autonomous phenotype. Does themutation code for autonomy? Bishop and Varmus ultimately opted forthe latter interpretation, selectively drawing on features of both theHuebner and Todaro and the Temin hypotheses.If cellular proto-oncogenes are normal and functional elements ofthe cell, then expression of these proto-oncogenes cannot be the causeof cancer (as they would in the Heubner and Todaro hypothesis) butrather must first be subject to some form of alteration, that is, somaticmutation (as depicted by Temin).
But unlike Temin, Bishop and Varmusdid not attempt to situate such mutations in the normal course of development. Carcinogenesis, for Bishop and Varmus, was depicted as theresult of the executive-like action of transformed proto-oncogenes (likeDialectics of Disorder: Normalization and Pathology as Process141Heubner and Todaro) which had, in effect, gained a new function forthe causation of unregulated growth.Huebner and Todaro’s oncogenes were strictly Genes-P. Postulated asthe result of the natural selection of viruses the ascription of a gene forfunctionality could have followed a Mendelian logic. If the Bishop andVarmus oncogenes could not be the result of any form of natural selection, how could they be accorded the status of a gene for functionality?Yet such status was implicit in the Mendelian (Gene-P) language whichBishop and Varmus cleaved to. “Dominantly-acting oncogenes” becametheir catch-phrase as well as the veritable motto of the oncogene researchprogram.
Bishop and Varmus spliced together pieces of the Huebner andTodaro and Temin models in such a manner as to endorse a second senseof cellular autonomy in a big way. Cells were not the repository of adaptive potential, for better or for worse, but rather were depicted as passivevehicles commandeered by phenotype-determining oncogenes.
Dominance was invoked in an ambigous double sense,—dominant withrespect to a nononcogenic allele at the same locus but also dominantwith respect to the genes of other loci. The Bishop and Varmus oncogenes, in a sense, oscillated between being Genes-D and Genes-P. Theywere like Gene-D in the sense that they were sequenced and thus molecular entities (However, the ability to distinguish clearly at the molecularlevel between normal cellular proto-oncogenes and activated oncogenesoften proved to be elusive.) And they were like Gene-P in being definedaccording to their phenotypic correlations in transformation assays.There is little doubt that the prospect for being able to unite the criteriafor being a Gene-P with the criteria for being a Gene-D (even if this wasonly tacitly understood) was responsible for generating much of theexcitement and fanfare that became associated with the oncogeneprogram.What would it actually take for the Gene-P and Gene-D criteria toproperly coincide in the case of an oncogene, and what would the largersignificance of such a coincidence be? Minimally, there would have to besome consistent difference in molecular sequence between protooncogenes and activated oncogenes and this difference would have toconsistenly map onto a difference in phenotype (of an organism, not justa cell in culture), with the later phenotype being malignant.
And yet even142Chapter 4if this correlation had become well established, and it never was, it wouldstill not settle the question as to which sense of the autonomous cell toendorse,—how to understand the fundamental nature and cause ofmalignancy.Does a mutation in a particular proto-oncogene dictate that a specificmalignant phenotype is produced, or rather does said mutation result ina cell becoming uncoupled from its surround and thereby susceptibleto embarking on a de novo path of cellular adaptation? In the latterview, malignancy is the result of complex adaptive (autopoietic) behavior which may be triggered by a mutation, but the malignant phenotypewill be largely contingent on its response to local conditions.Cancer by Decree or Cancer by Default?In contrast to the claims of genetic dominance uttered by the oncogenecommunity, a separate and unrelated program of investigation into thegenetic basis of cancer, its chronology paralleling that of the oncogeneinvestigation, appeared to identify a genetically recessive basis of carcinogenesis.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
