Moss - What genes cant do - 2003, страница 10

Johannsen, with much perspicuity, anticipated the likely misconstrual of the genetic perspective. After reconstructing this tortuous journey I will return to take a closer look atJohannsen’s insightful perspective.The earliest experimental attempt to locate the Anlagen at a subcellular level began not with the chromosome or even nucleus but rather withthe nineteenth-century hypothesis of “cytoplasmic anlagen” put forwardby Wilhelm His (Gilbert 1978). C. O.

Whitman, following His, believedorganismal development could be analyzed by tracing the path of celllineages from their origins in cytoplasmic Anlagen. Not long after His,Nägeli proposed the existence of an “idioplasm,” later elaborated byWeismann, consisting of nested, hierarchically organized, particulateunits that directed development. Nägeli was nonconmittal, however, asto where the idioplasm would be located. With the accumulated cytological work on chromosomes and the strong impression made by theobservation that chromosome number and morphology remain constantacross generations, Oscar Hertwig, Weismann, Kölliker, and others postulated that the chromosomes were the site of the putative idioplasm(Sapp 1987).That party lines among embryologists in the 1890s were partitionedalong the boundary of the cell nucleus was largely contingent.

Surelysome form of nucleocytoplasmic egalitarianism could also have found aplace within the logical space of possibilities. However, the nucleusversus cytoplasm divide came to define a rather intractable oppositionover what was going to count as the proper stuff of heredity.Wilhelm Roux, following Weismann, championed a “mosaic hypothesis,” which, had it been successful, would have enabled the nuclearidioplasm theory to unite development and inheritance within asingle, particulate-preformationistic model. According to the mosaic32Chapter 1hypothesis, cellular differentiation and organismal development are regulated by the uneven distribution of idioplasm at each stage of cell division.

In this view only germ cells contain the full complement of Anlagen.The Anlagen then becomes differentially partitioned into daughter cellsduring somatic cell division. Ultimately each cell only receives thoseAnlagen required for its terminal state of differentiation. Roux hadclaimed experimental confirmation of his theory when, using a hot sterilized needle, he ablated one of the blastomeres of the two-cell-stage frogembryo. The remaining blastomere continuted to develop, albeit abnormally, resulting in something like half an embryo.

According to themosaic theory, the remaining blastomere would contain only that idioplasm capable of producing half an organism (Allen 1975).Hans Driesch, working in Naples in the 1880s and 1890s, attemptedto reproduce these findings, using an agitational method for separatingthe two blastomeres of the sea urchin. Contrary to Roux’s findings, theremaining blastomere was fully capable of producing a normal, albeitsmaller, sea urchin larvae (Allen 1975). Driesch countered Roux’s mosaicmodel with his theory of the developing organism as a “harmoniousequi-potential system.” According to this theory, cells retain the fullpotential of the organism, gaining their developmental specificity notthrough directives from within but rather externally and relationally withrespect to their position in the developing organism and the influence ofenvironment.

Driesch, in effect, proposed a nascent developmental fieldtheory. Although Roux was disinclined to concede the point, it becamegenerally accepted that his results were due to the residual effects of theablated cell matter and that Driesch’s results were truly indicative of theabsence of mosaic nuclear division.By 1910 Morgan could state unhesitatingly:We have every evidence that in embryonic development the responsive action ofthe cytoplasm is the real seat of the changes going on at this time, while the chromosomes remain apparently constant throughout the process (Morgan 1910).Had the mosaic model been successful, as suggested above, many problems would have been avoided.

The idioplasm, presumably located inthe nucleus, would have been seen to “represent” the organism as awhole and to provide the mechanism for controlling development. Theprocess of heredity and the process of development would have been sub-Genesis of the Gene33stantially united. Morgan was never sympathetic to this model becauseit smacked too much of old-fashioned preformationism.If mosaic division does not explain development then one can question whether the idioplasm in fact represents the unity of the organismat all. For those who held to an autonomous cell theory, notably E. B.Wilson, it did.

Wilson felt that every cell contained the full representation of the organism in its nucleus and that the nucleus directed the developmental epigenesis which was played out in the cytoplasm. How thenucleus could do this while remaining ostensibly constant throughoutthe organism was a question he could not answer. Driesch and Whitmanby contrast believed that only the whole organism represented the unityof the whole. Individual cells did not constitute autonomous units butrather were centers of action determined by their position in the whole(Sapp 1987). From the latter point of view heredity, in any full sense,could not be explained solely on the basis of the nucleus.

Subsequentdebate among embryologists was concerned with the relative importanceof the cytoplasm versus nucleus in heredity but also and ultimately withwhat would count as heredity.In 1903 Sutton identified the processes of Mendelian segregation (oforganismal traits) with those of chromosomal reduction (in cytologicalobservation), thus again nominating the chromosomes as the putativesite of Mendelian factors. E. B.

Wilson and Theodore Boveri endorsedSutton’s position within a year (Morgan 1910) and proceeded to performnew experiments to show the primacy of the chromosomes in inheritance. Wilson, working with protozoa, offered evidence in support of therole of the chromosomes in regeneration.Boveri provided key evidence in favor of the identification of chromosomes with Mendelian factors (the Sutton-Boveri hypothesis) bydemonstrating the “individuality” of chromosomes, i.e., that chromosomes were not homogeneous in their biological effects. By producingdispermic sea urchin eggs, ova fertilized with two sperm, he couldacquire blastomeres with different complements of chromosomes.

Separating these at the four-cell stage Boveri could show that they possesseddifferent developmental potentials ostensibly based on their differentialpossession of chromosomes (Gilbert 1978). Further, Boveri attemptedto address the cytoplasm versus nucleus debate directly by use of34Chapter 1“merogony” experiments. Working with enucleated echinoderm eggs,Boveri would replace the maternal nucleus with that from sperm of aphenotypically distinguishable variety. The relative impact of the cytoplasm versus that of the transplanted nucleus on subsequent development could then be observed. Boveri reported that it was the phenotypiceffects of the sperm which prevailed (Gilbert 1978, Sapp 1987). Adapting methods reflective of his long interest in the biology of regeneration(Allen 1985), Morgan examined the effects upon subsequent development after removing cytoplasm from an unsegmented ctenophore egg.He reported that loss of cytoplasm despite an intact nucleus resulted ina deformed embryo.

Morgan was skeptical of the findings of Boveri’smerogony experiments and during the course of extensive efforts toreproduce them found increasing evidence for the primacy of the cytoplasm, not the nucleus, in early development.Whether Mendelian factors reside on the chromosomes was not theonly pressing question for leading embryologists during the first threedecades of the twentieth century. In the absence of something like amosaic hypothesis which united idioplasm with development and withample evidence supporting the developmental efficacy of the cytoplasm,one also had to address questions pertaining to the scope and explanatory significance of identifying Mendelian factors with chromosomes.

Ifthe Sutton-Boveri hypothesis were vindicated (as of course it was), wouldthat mean that chromosomes contain a full representation of the organism as Weismann and Wilson would have it? Would chromosomes thusbe both necessary and sufficient for organizing the full development ofthe organism? This conclusion does not necessarily follow.