Moss - What genes cant do - 2003 (522929), страница 43
Текст из файла (страница 43)
Mantle’s liver cancer may well havebeen clonal, but it was hardly the outcome of some chancy somatic mutation occurring in isolation from the larger organizational field of his liver.Mantle’s liver underwent rounds of organizational shifts in the courseof decades of heavy alcohol exposure, which enabled him to enjoy thelongevity sufficient for a Hall of Fame baseball career and a subsequent(albeit less distinguished) run as a Manhattan restaurateur. The livercancer, clonal or otherwise, was the ultimate result of cellular dynamicsthat were highly adaptive but came at a price.The apparent paradox of the “multicentric” origins of cancer (to useSmithers’s term) and the ultimate appearance of a monoclonal tumorappear to be reconciled in the cell culture model of carcinogenesis developed by Harry Rubin.
Four decades after establishing the experimentalbasis for studying the retroviruses, Rubin continued to be engaged in anexperimental practice that he has described as his “dialogue with cultured cells” (personal communication). Rubin’s experimental system hasinvolved the use of the same cells, the NIH 3T3 line derived from mousefibroblasts, that oncogene researchers used for showing the transforming capability of viral and activated oncogenes. But Rubin found that thevery same indications of spontaneous transformation in culture are routinely obtained on the basis of metabolic stress, especially the growth ofcells under crowded postconfluent conditions. Rubin has found, overyears of exhaustive trials, that protracted exposure of cells to conditionsof confluence (when the entire floor of a tissue culture dish is filled witha complete pavement of cells one layer thick) results first in both thedeath of some cells and a population wide change in most.
These cells,when switched to optimal growth conditions, reveal a decreased growthrate, yet a greater saturation density, that is, an ability to continue toDialectics of Disorder: Normalization and Pathology as Process171grow beyond the typical contact constraints of (so-called) normal 3T3cells (Rubin, 1995a). Successive rounds of prolonged incubation of thesecells at confluence results in the appearance of transformed foci, densecolonies of cells derived from a single cell which are no longer constrained to grow in monolayer but rather form a kind of multilayered“cell culture tumor.” Where the early changes in the 3T3 cells appearto occur on something approaching a population-wide basis, it is only asmall percentage of the cells that are responsible for forming transformedfoci.
The early changes that take place on a widespread basis, whichresult in reduced rates of cell growth but higher rates of saturationdensity, have the character of adaptive epigenetic changes. And these cellsdo not immediately revert to normal growth patterns when transplantedto low-density cultures (Rubin, 1995b). A population-wide response isnot consistent with the expectations of a specific mutation, which wouldbe a very low probability event.
Yet this observed heritability would beconsistent with some form of genetic alteration. The early changes ofcells grown in postconfluent cultures have characteristics of both epigenetic adaptation to stressful conditions and some form of irreversiblegenetic alteration, but these need not be contradictory. Adaptive epigenetic changes can also result in genetic instabilities.
A possible mechanistic pathway could entail changes in DNA methylation states thatcan affect mutation rates (discussed in Chapter 3). While the widespreadcharacter of the changes seen in postconfluent cultures rules out specificpoint mutations, diffuse genetic damage need not be excluded. Infact these stages, beginning with a general decrease in growth rate andculminating in the formation of sporadic transformed foci, bear a strongresemblance to in vitro studies in which cells were initially irradiatedwith X rays or exposed to various chemical carcinogens (Rubin, 1995b).What appears to be the common denominator is that cells confrontedwith any number of stresses respond in an arguably adaptive fashionthat allows the majority of cells to continue to maintain a relativelynormal phenotype but with an increasing propensity to give rise to atumorigenic colony.
Rubin (1995b) suggests that “All the evidence pointsto the origin of cancer from a field of altered, unstable but normalappearing cells rather than from isolated mutations among otherwiseunaltered cells.”172Chapter 4When Rubin’s cells are subjected to further rounds of incubation underconditions of confluence they eventually give rise to some array of transformed foci which are each in some sense morphologically distinct. Agiven dish will thus have several different transformed populations withdistinctive phenotypes at the colony level. Rubin found that if the cellsof these dishes were to be further passaged (removed and replated ontonew dishes at initially lower densities), the cultures would eventuallybecome monoclonal cultures of transformed cells.
What this suggests isthat the ultimate appearance of a monoclonal tumor population inRubin’s experimental system is the result of the fastest growing transformed population, overwhelming the other cells over time. Rubin’sexperimental system thus provides a model for how cancer may begin asa widespread multicentric disruption of a whole cellular population (andorganizational field) yet result in a monoclonal tumor. If the eventualityof a monoclonally derived tumor need no longer appear contradictoryto the multicentric origins of carcinogenesis in the disruption of an organizational field, then the most interesting question may pertain to themicrointeractions within the disrupted field that result in the promotionand amplication of a clonal tumor.
Rubin (1993) has desribed thisprocess as that of a “progressive state selection.” “Our current workgoes beyond the monoclonal origin of cancer (Fialkow) to the fields surrounding the dominant clone to establish a cellular ecology of tumordevelopment” (personal communication).Under the rubric of “dynamic reciprocity” Mina Bissell (a formerpostdoctoral student of Rubin) and coworkers have been attempting toelucidate the molecular details of the cellular ecology of breast cancerdevelopment for more than 20 years.
Given the well-known inductivesignificance of tissue interaction in the case of embryological development, Bissell has approached breast cancer with an eye to the reciprocally stabilizing and destabilizing interactions between the glandularepithelium of the breast and the fibroblastic cells of the stroma. Inasmuch as the principal medium of communication between these twotissues is the so-called extracellular matrix (ECM) to which they bothcontribute and with which they both attach and interact, the main focusof the Bissell laboratory has been that of the relationship betweenmammary epithelium and the ECM (Werb et al.
1996). Mammary cells(as well as other types of epithelium) attach to ECM by way of cell-Dialectics of Disorder: Normalization and Pathology as Process173surface attachment proteins, especially members of the “integrin” familyof receptors. Such attachments also transmit signals to the interior ofmammary cells which then may affect many processes, including transcriptional state, and ultimately the way in which these cells in turnmodify (degradatively and/or synthetically) the ECM. Epithelial-basedchanges in the ECM will induce reactions in the cells of the stromaleading to a new volley of reciprocal cellular activities.
What Bissell andothers have observed to be a common feature of all breast cancers (andmany other types of cancer as well) is an abnormal cell-to-ECM relationship (Boudreau & Bissell 1998). Both in vitro and in vivo studies byBissell as well as others provide evidence to the effect that an abnormalECM can lead to the initiation and/or promotion of mammary tumorigenesis and that the right ECM can stabilize a nascent carcinoma(Radisky et al. 2001). Bolstered by such findings, Park, Bissell, andBarcellos-Hoff (2000) have anted up the following questions with astrong anticipation of affirmative answers in the making.1.
Can changes in the microenvironment precede the progression ofneoplastic disease?2. What features of the microenvironment promote neoplastic disease?Are these tissue-specific?3. Can the microenvironment be targeted therapeutically to preventcancer?4. Can manipulating microenvironment reverse cancer?If and when an appreciation for the role of the microdynamics of cellular ecology in tumor formation and metastatic progression becomescentral to the oncological research paradigm, a major shift of perspective toward what I have earlier referred to as a “new epigenetics” willhave been realized.From Somatic Mutation to Genetic Predisposition—the Paradoxesof ProgressMolecular analysis is not needed to distinguish what percentage ofhuman cancers are based on genetic inheritance—classical epidemiologyprovides that information.
Either cancer is seen to follow familial patterns of inheritance and is thus putatively heritable, or it is seen to arise174Chapter 4sporadically. Epidemiological studies have consistently found the vastmajority of cancers to be sporadic in nature, prompting oncologists tosee somatic as opposed to germ-line mutations as of more relevance tothe understanding of cancer etiology.Why then has so much recent attention turned toward questions ofgenetic susceptibility? Have the past appraisals of the role of inheritancein cancer morbidity been misguided? And what role has the successionof putative breakthroughs in understanding the role of somatic mutations had in leading up to the new turn toward germ-line genetics? Thislast section will address these questions and offer an answer that may besurprising.The first part of this discussion will focus on a rather remarkable andrevealing scientific review entitled Lessons from Hereditary ColorectalCancer by Kinzler and Vogelstein (1996).
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
