Moss - What genes cant do - 2003 (522929), страница 27
Текст из файла (страница 27)
In bothcases a wealth of biological complexity is made to disappear in order toprovide a provisionally useful simplification. To what extent Kauffman’sinstrumental reductionism provides practical as well as theoretical utilityis of course yet to be proven.If we begin to look at what must be the enabling background conditions for mammalian organisms to be modeled as if they were N =100,000 and K = 2 systems, we will find our way back to the wet biologyof heritable steady-state epigenetic systems. Consider the meaning ofK = 2.
In what sense can it be said that a gene only receives input fromapproximately two other genes? Strictly speaking, genes do not receiveinput from any other genes without the mediation of proteins. Andeven if the intent is just to speak of those genes which give rise to thedirectly mediating proteins, then there is still then a retinue of proteinsand thus genes which mediate the production of these mediating geneproducts.It is clear then from the outset that any talk about a highly delimitednumber of gene inputs must distinguish between a great number of geneinput candidates and find some criteria by which most of these may bebracketed and treated as if they were only background conditions. Asdiscussed in the second section above, the meaningfulness of any genedepends on the complex organizational structures and compartmentalization of the cell, which determines where the protein will be locatedand how it will be covalently and noncovalently modified.
It is only atthis level of finishing that a protein takes on biological significance.Clearly, there are many untold genes associated with the biological realization of most any protein at this level, and so these all must be factored out of what would have to count for Kauffman as “genetic inputs.”The obligatory move to make then would have to be to decree that allthose genes associated with gene products that are required for the functional realization of any gene product would not count as a genetic inputin Kauffman’s sense. This would bracket out a great many (so-calledhousekeeping) genes.
But would even this rather large concession bringKauffman’s requirements into the space of biological reality? Consideran exemplary case for the regulation of the activation or inactivation ofa particular gene sequence by a small number of particular gene prod-A Critique of Pure (Genetic) Information105ucts and see what further heuristic concessions must be made to accommodate the Kauffman model.Using recombinant techniques, Diamond et al. (1990) constructeda simplified version of the steroid-hormone regulation of the geneproliferin in order to analyze the additive and relational effects of threeregulatory proteins that we may consider to be our Kauffman inputs.Glucocorticoid steroids influence genetic transcription when joined toa soluble glucocorticoid receptor and then subsequently bound as acomplex to DNA enhancer regions known as “glucocorticoid responseelements” (GREs).
Transcriptional activation of a gene refers to the synthesis of an RNA transcript by an RNA polymerase enzyme complex,using the DNA “sense strand” as a template. The ability of the RNApolymerase to bind to the promoter region of the gene and beginsynthesis depends on the configuration of proteins that are present andthat may facilitate or inhibit transcription. “Enhancers” are regions ofDNA that often involve context-sensitive (developmental stage, tissuetype, and so forth) regulation of transcriptional activation. Diamond etal.
(1990) simplified the in vivo biology of gluccorticoid regulation byinserting an attenuated GRE sequence of only 25 nucleic acid base pairswhich, while capable of binding the hormone-receptor complex, reducesthe number of other factors that may come into play. The transcriptionalinitiation factor AP-1, which is composed of two proteins, c-fos and cjun, plays a mediating role in the transcription of many genes, but itsrole in proliferin transcription cannot be relegated to the backgroundbecause it interacts in very specific ways with the GRE and receptorhormone complex.Using the simplified GRE construct in a model system, Diamond et al.explored the relationships of the glucocorticoid receptor-hormonecomplex and the c-jun and c-fos proteins in the regulation of proliferintranscriptional activation.
We can view their results in terms of a K = 3system by expressing their findings in terms of a series of Boolean operators with the additional simplification of transforming graded transcriptional effects into an all-or-nothing binary switch. The results wouldthen be as follows. If complex and no (fos or jun), then negative.
If junand complex or no complex, then positive. If fos and jun and nocomplex, then positive. If fos and jun and complex, then negative. As an106Chapter 3artificially simplified system, this model of transcriptional regulationshould be viewed as being as simple as it gets, and yet even at first blush,it is more complicated than a Kauffman K = 2 system. In addition to thedirect affects of c-fos, c-jun, and the glucocorticoid receptor, the modelsystem also involves the glucocorticoid hormone that binds the receptor.The hormone is the product of many genes associated with the biosynthetic pathways of secretory cells in the adrenal cortex. Now, when thesteroid hormone enters the target cell, i.e., a cell that has a receptor available, it binds to the receptor in the cytoplasmic compartment.
Havingdone so the receptor hormone complex may then gain entry into thenuclear compartment, which is as much as to say that only after bindingthe hormone can the receptor pass the entry requirements administeredby the nuclear pore. And what about the regulation of the expression ofthe c-fos and c-jun proteins?To trace even the more proximately antecedent determinants of thesegenetic inputs into the regulation of the proliferin gene is to embarkquickly on an explanatory regress in which increasingly many contingentfeatures of cellular organization—surface-receptor binding status, thephosphorylation states of cell-surface receptors and signaling intermediates, the presence or absence of a host of transcriptionally active effectors, as well as others—come into play. In the words of Keith Yamamoto,the leading investigator of glucocorticoid mediated regulation, the activation state of the proliferin gene is determined by “the complex stateof the cell.” To even begin to trace back the biology that must be presupposed in treating a cell as if it were a Kauffmanian genetic regulatorysystem is to rediscover quickly the wet biology of organizational structure and compartmentalization, real steady-state systems, and theclose relationship between them.
Whether Kauffman has ultimately provided a vehicle for conceptually grappling with the intricacies of realbiology at a higher level of complexity (i.e., has given us the means tobring more biology into a single concept) or a more compelling pretextfor simply ignoring the same), is perhaps the judgment around which thevalue of his work will turn. Maybe between these two extremes lies thepossibility of conceiving of Kauffman’s model, while far too bare (anddry-) boned to represent real biology, as a sophisticated metaphor, asymbol but not a substitute, for how local regimes of order emerge fromA Critique of Pure (Genetic) Information107ensembles of multitudinously differentiated, multitudinously interactingparts.The Regional ViewKauffman offers a top-down approach to understanding the role ofself-sustaining dynamics in the achievement, propagation, and evolutionof biological order.
Beginning with formalisms at a far remove fromempirical particulars he attempts to incorporate the latter by reinterpreting it in the light of a global perspective. Differentiated cell types,for example, reappear as the allowed states of a K = 2 system. Thealternative approach to conceptualizing the role of self-sustainingdynamics as a nongenetic–template-based source of biological order isone which attempts to build up from small-scale regional processes—andare thus not at all removed from empirical wet biology. This approachwould be the one more in line with that of the Jablonka and Lamb’s(1995) notion of epigenetic inheritance systems (really, subsystems).There are many kinds of dynamic subsystems which would share theproperty of sustaining, even across generations, some physiochemicalstate based on the intrinsic ability of the system to adjust and adapt tocontingencies.Delbrück (1949) provided a small-scale theoretical model for a typeof positive-feedback mechanism in which a metabolic pattern becomesheritable.
In this model of his cyclically catalytic system there is someregulatory component (protein) the presence of which results in the promotion of its own synthesis (positive feedback). The metabolic state ofsuch a system is thus highly sensitive to fluctuations in the concentrationof that regulatory component such that the current state of the systemwould be an outgrowth of its past history. The “lac operon” ofEscherichia coli (Novick & Weiner 1957) is an example of such a system.It was found that when cultured in low levels of lactose genetically identical E. coli bacteria diverge into two different heritably stable phenotypes.
One phenotype synthesizes b-galactosidase and the other onedoesn’t. The key difference between these is based on chance fluctuationsin the intracellular concentration of the permease enzyme necessary totransport lactose inducer into the cell. The permease gene is itself partof the lac operon. When a cell by chance expresses the permease in low-108Chapter 3lactose concentration, it results in an increase of lactose take-up followedby the induction of more permease, then more lactose, uptake, whichleads to further induction until stable concentrations of permease and bgalactosidase are reached. (The latter enzyme breaks down the lactose.)New generations of E.
coli in the presence of low concentrations oflactose will thus inherit either the b-galactosidase metabolism or the nonb-galactosidase–expressing metabolism. While this is a simple model, itdoes provide some insight into how the complex state of an organismcan be the result of its dynamic (as opposed to conventionally genetic)adaptation to historical contingencies.Self-maintaining (steady-state) regulatory loops need not necessarilyentail alterations in transcriptional rate; they could also occur at the levelof posttranscriptional regulatory processes.
![](https://s.studizba.com/z.php?f=/templates/si/i/noavatar.png&w=100&h=100&t=1"){kind=avatar}
