Susumo Ohno Chromosomes Herman J. Muller

Ohno’s hypothesis and Muller’s paradox:

sex chromosome dosage compensation may serve collective gene functions

Donald R. Forsdyke  (2012) BioEssays 34 (11) 930-933 

 [This pre-peer-reviewed version, with subsequent elaborations (photos, coloured text, etc.) closely approximates the copyright holder's (Wiley Periodicals Inc.) final version, which may be accessed at the Wiley-Blackwell website: ]

Abstract Muller found halving gene dosage, as in males with one X chromosome, did not affect specific gene function. Why then was dosage “compensated?” This paradox was solved by invoking collective gene functions, such as the self/not-self discrimination afforded by protein aggregation pressure. This predicts female susceptibility to autoimmune disease.








End-Note_(Nov_2013)_Concentration_Conservation:_Proteins v mRNAs v DNAs




There has been much recent debate on whether new technologies reveal that the expression levels of genes on X chromosomes accord with Ohno’s hypothesis for how mammals compensate for the degeneration of the male Y chromosome [1-7]. In his famous treatise Ohno in 1967 noted [8]:

"In fishes, amphibians, and most reptiles, the X and the Y ... of each species are still largely homologous to one other. Thus, even the heterogametic sex maintains two doses of each sex-linked gene. In the case of placental mammals, however, the Y has shed all the Mendelian genes which were allelic to the genes on the X. As a result, most, if not all, of the X-linked genes exist in the hemizygous state in the male. Each X-linked gene must have accommodated itself to this hemizygous state by doubling the rate of product output. Once this doubling in efficiency was accomplished, the genetic disparity between the male with one X and the female with two Xs became very great. A need arose to adjust the dosage effect of X-linked genes between the two sexes. In mammals, the dosage compensation is accomplished by random inactivation of one or the other X in individual female somatic cells. Consequently, phenotypic expression of X-linked genes in individual somatic cells of both sexes is hemizygous, and the mammalian female is a natural mosaic with regard to the activity of X-linked genes."

Thinking in terms of individual, rather than of collective, gene functions, Ohno supposed that as, one by one, genes disappeared from the original proto-Y in males, there would have been pressure for males to increase the output from the corresponding allelic genes encoded by their solitary X. This would have sorted out the males, but the poor females that inherited such hyperactive X chromosomes would have had to find a way to decrease the output (since they had two X chromosomes) on a gene-by-gene basis. Females seemed to have solved the problem by turning off one of the X's.

Muller’s paradox

H. J. Muller’s seminal 1948 study of dosage compensation in fruit flies [9] does not form part of the modern debate. Yet it was Muller who gave critical evidence on the dosages of some sex chromosome gene products that differed between the sexes. These values corresponded in both sexes, not to the ascending limb, but to the plateau of dose-response curves (i.e. plots of functional expression against quantity of gene product). In other words, halving a gene’s dosage in males did not affect its functional expression (Fig. 1).

Figure 1. Dose-response curve showing a measure of conventional phenotype plotted against the quantity of a gene product that contributes to that phenotype. The thick vertical arrow indicates the normal concentration (Y) of a non-rate-limiting gene product in a diploid homozygous cell. If gene product concentration is directly proportional to gene dosage, halving gene dosage will halve gene product dosage (Y to Y’).  This does not affect phenotype because concentrations corresponding to the plateau of the dose-response curve are not rate-limiting (do not change phenotype value). X and X’ indicate the corresponding points for a rate-limiting gene product (big change in corresponding phenotype value). Also shown is the hierarchical flow from gene to phenotype. Reproduced from ref. 25 with permission of McGill-Queen’s University Press.

However, like those engaged in the modern debate, Muller persisted in seeking an explanation in terms of specific gene function:

"The effects of individual genes, whether on the X or other chromosomes, are so near their saturation levels as to make direct [functional] discrimination between one and two doses impossible. Should not the very fact that most of these genes are so near their saturation level make dosage compensation unnecessary? Why should there be a perceptible advantage in going through the motions of equalizing them still further?"

To resolve the paradox of apparent haplosufficiency, Muller supposed that:

"The compensation mechanism must be concerned with the equalization of exceedingly minute differences. Dosage compensation has in fact become established because of its advantage in regulating more precisely the grade of characters whose variations in grade, even without it, would be exceedingly minute. … The selective forces that established it must depend on such minute advantages" [Muller's italics].

Thus, while appearing haplosufficient, the genes were actually haploinsufficient.

Collective gene functions

Emphasizing his focus on specific gene function, Muller entitled his paper “Evidence on the precision of genetic adaptation.” But why, given that specific functions barely change, should differences in protein doses revealed by some modern technologies, matter? To answer this we must recall that genes have both specific and collective functions. For example, it has been known for nearly a century that cytosolic proteins collectively affect the distribution of ions across the cell membrane (“Donnan equilibrium”) [10-11]. While decreases in doses of the products of a few genes might only marginally affect collective functions, the progressive loss of genes from the degenerating Y chromosome would be expected eventually to evoke compensatory adaptations to maintain collective functions. The genes responsible for such adaptations need not necessarily be the same as the genes that were lost. Indeed, highly expressed genes – expected to best maintain collective functions – would be selected on this basis [12, 13].

Thus, having lost a set of genes from the shrinking proto-Y chromosome, males would not necessarily have to increase the output from the same set of X-borne allelic genes. To maintain overall protein concentrations, it would suffice marginally to increase the output of some X-borne genes, perhaps by tinkering with some global output-controlling mechanism. If we label X chromosome genes A to Z, a loss of A from the proto-Y (halving the concentration of the A gene product) could either be ignored by males or compensated by a marginal tweak of some or all of A-Z on their solitary X. The actual concentration of the A gene product (already halved) would scarcely change. This X chromosome would be handed on to females, who could progressively adapt using a similar global mechanism.

Ohno and Muller rejected

On the assumption that prior to the degeneration of the Y chromosome the average outputs of all chromosome pairs were comparable, Lin et al. in 2012 [7] have, as in the previous work from their laboratory [2], used the autosomes as a yardstick against which to compare sex chromosome expression levels. Furthermore, they have now compared mammalian genes with their orthologs in close outgroup species that had diverged prior to the emergence of mammalian sex chromosomes. This provides "unambiguous evidence for expression halving (i.e. no change in per-allele expression level) of X-linked genes during evolution." In other words, Ohno’s hypothesis is falsified. Thinking, like Muller [9], in terms of individual gene functions, they conclude:

"A natural question is how such chromosome-wide expression halving has been tolerated, given that autosomal monosomy [loss of one chromosome of a pair] is lethal in humans? Because Y degeneration is stepwise, expression reduction happened gradually to more and more X-linked genes during evolution. Thus, a possible explanation is that, at any time in evolution, an organism is faced with the expression halving of only one additional gene, which might have been slightly deleterious and thus can be fixed. This evolutionary process contrasts the sudden loss of an entire chromosome in monosomy that causes a large fitness reduction at once. Consistent with the above explanation is the observation that up to 97% of yeast genes have no detectable fitness effect when one allele [at a time] is deleted from a diploid cell. Although haplosufficiency has not been systematically examined in mammals, it is probable that, for most genes, expression halving has little fitness effect [by virtue of loss of an individual gene-encoded function]."

However, unlike Muller, who in fruit flies looked for explanation of his apparent haplosufficiency paradox to “exceedingly minute differences” in specific function [9], Lin et al. suppose haploinsufficiency not to be a problem in mammals. Most genes are deemed haplosufficient. Muller, like Ohno, is rejected. The only exceptions admitted are genes – approximately 5% of the total – whose products, by virtue of their specific functions, need to form complexes with the products of other genes [7, 14]. Still arguing in terms of specific gene function, others have postulated that, while small in number, these genes might have included “critical hub genes” that formed a regulatory network controlling a multiplicity of other genes [15].  

"Monosomy or trisomy for a whole large chromosome usually presents a major problem; in humans, for example, either is usually lethal. This is probably because differences over a large genomic segment are more likely to involve differences in dose of critical hub genes. The more hub genes involved, the larger the problem. ... This view suggests that the reduction in dose of a minority of X genes drove the selection of complex molecular mechanisms to adjust X dosage in one sex or the other."

A new evolutionary force?

Unwilling to explain their results in such terms, Lin et al. now repeat their laboratory’s call for a "reopening" of "the search for the evolutionary force driving the origin of chromosome-wide X inactivation in female mammals” [7]. Collective gene functions could provide such a force, highly expressed genes playing a major role. In particular, there is aggregation pressure. In the crowded cytosol individual protein species are held to both contribute to, and be acted upon by, this pressure, to which their concentrations have been fine-tuned over evolutionary time [13, 16-17]. To resolve “Muller’s paradox,” in 1994 I proposed that, if aggregation pressure is excessive, individual proteins would more readily approach their solubility limit and their specific peptides would then become available for presentation as MHC complexes to T lymphocytes [18]. Normally, susceptibility to differential aggregation would aid the intracellular detection of either abnormal self, or foreign (not-self), proteins. New studies of the role of X chromosomes in the high susceptibility of females to autoimmune disease have cast fresh light on this [19]. Evidence that failure to inactive an X chromosome predisposes both XX females and XXY males to autoimmune disease (reactivity with normal self), suggests that aggregation pressure could have been a major selective force driving the evolution of dosage compensation [20-26]. The failure would have created an excess of X-linked proteins, so increasing aggregation pressure beyond acceptable limits.

Because the initiating factor is construed as degeneration of genes on the proto-Y chromosome, it is convenient to suppose, like Ohno, that the male adapted first and then the female was forced to counter-adapt. In males the collective function would slowly decrease (because of a fall in protein concentration). If this collective function were aggregation pressure, then the resulting immunological impairment would force males to adapt by increasing protein concentration. By the same token, females receiving adapted chromosomes from males would tend to have had excessive protein concentrations, which would have provoked autoimmune disease. We do not know which of these selective forces would have been more powerful in achieving the dosage balance that we see today. But we should note that if it is concentration, rather than absolute dosage, of protein that is important, then other means of upward adjustment – such as slightly decreasing cell volume – might suffice in males. If such a volume adjustment were transferred to females, then the primary onus for adjusting protein dosage, in this case downwards, would rest with them.


New technologies for determination of gene expression levels have been used to examine the evolutionary basis of mammalian sex chromosome dosage compensation [1-7]. Muller’s invocation of “exceedingly minute differences” in specific gene functions as the evolutionary force driving X chromosome dosage compensation in fruit flies [9], and Ohno’s hypothesis in mammals [8], combine to confuse those who seek to understand seemingly conflicting results. Instead of specific gene functions, perhaps we should think more in terms of collective gene functions. New work on the enhanced susceptibility of females to autoimmune disease suggests that a dimly recognized collective function – aggregation pressure – may have played a major role in the evolution of dosage compensation [16-26]. Two apparently disparate areas – dosage compensation and immunology – may be profoundly related.  


For over a decade Queen’s University has hosted my dosage compensation educational web-pages.


  1. Nguyen DK and Disteche CM. 2006. Dosage compensation of the active X chromosome in mammals. Nat Genet 38: 47-53.

  2. Xiong Y. et al. 2010. RNA sequencing shows no dosage compensation of the active X-chromosome. Nat Genet 42: 1043-1047.

  3. Deng X. et al. 2011. Evidence for compensatory upregulation of expressed X-linked genes in mammals, Caenorhabditis elegans, and Drosophila melanogaster. Nat Genet 43: 1179-1185.

  4. Kharchenko PV, Xi R, Park PJ. 2011. Evidence for dosage compensation between X chromosome and autosomes in mammals. Nat Genet 43: 1167-1169.

  5. Lin H. et al. 2011. Relative overexpression of X-linked genes in mouse embryonic stem cells is consistent with Ohno's hypothesis. Nat Genet 43: 1169-1170.

  6. He X et al. 2011. He et al. reply. Nat Genet 43: 1171-1172.

  7. Lin F, Xing K, Zhang J, He X. 2012. Expression reduction in mammalian X chromosome evolution refutes Ohno's hypothesis of dosage compensation. Proc Natl Acad Sci USA 109: 11752-11757.

  8. Ohno S. 1967. Sex Chromosomes and Sex-linked Genes. Berlin, Germany: Springer-Verlag.

  9. Muller HJ. 1948. Evidence on the precision of genetic adaptation. Harvey Lect 43: 165-229.

  10. Hitchcock DI. 1924. Proteins and the Donnan equilibrium. Physiol Rev 4: 505-531.

  11. Kurbel S. 2008. Are extracellular osmolality and sodium concentration determined by Donnan effects of intracellular protein charges and of pumped sodium? J Theor Biol 252: 769-772.

  12. McConkey EH. 1982. Molecular evolution, intracellular organization, and the quinary structure of proteins. Proc Natl Acad Sci USA 79: 3236-3240.

  13. Yang J-R, Liao B-Y, Zhuang S-M, Zhang J. 2012. Protein misinteraction avoidance causes highly expressed proteins to evolve slowly. Proc Natl Acad Sci USA 109: E831-E840. doi:10.1073/pnas.1117408109.

  14. Pessia E. et al. 2012. Mammalian X chromosome inactivation evolved as a dosage-compensation mechanism for dosage-sensitive genes on the X chromosome. Proc Natl Acad Sci USA 109: 5346-5351.

  15. Arnold AP, Itoh Y, Melamed E. 2008. A bird’s eye view of sex chromosome dosage compensation. Ann Rev Genomics Hum Genet 9: 109-127.

  16. Forsdyke DR. 1995. Entropy-driven protein self-aggregation as the basis for self/not-self discrimination in the crowded cytosol: what the Greeks did not know. J  Biol Sys 3: 273-287.

  17. Forsdyke DR. 2012. Functional constraint and molecular evolution. In Encyclopedia of Life Sciences (John Wiley & Sons, Chichester, 2012) doi:10.1002/9780470015902.a0001804.pub3

  18. Forsdyke DR. 1994. Relationship of X-chromosome dosage compensation to intracellular self/not-self discrimination: A resolution of Muller’s paradox? J Theor Biol 167: 7-12.

  19. Forsdyke DR. 2009. X chromosome reactivation perturbs intracellular self/not-self discrimination. Immun Cell Biol 87: 525-528.

  20. Pan H-F. et al. 2009. Susceptibility to systemic lupus erythematosus may be related to gene dosage effect on the X chromosome. Med Hypoth 72: 104-105.

  21. Smith-Bouvier DL. et al. 2008. A role for sex chromosome complement in the female bias in autoimmune disease. J Exp Med 205: 1099-1108.

  22. Scofield RH. et al. 2008. Klinefelter’s syndrome (47, XXY) in male systemic lupus erythematosus patients. Support for the notion of a gene-dose effect from the X chromosome. Arth Rheum 58: 2511-2517.

  23. Dillon SP. et al. 2012. Sex chromosome aneuploidies among men with systemic lupus erythematosus. J Autoimm 38: J129-J134.

  24. Selmi C, Brunetta E, Raimondo MG, Meroni PL. 2012. The X chromosome and the sex ratio in autoimmunity. Autoimm Rev 11: A531-A537.

  25. Forsdyke DR. 2001. Sex chromosomes. In The Origin of Species Revisited. Montreal: McGill-Queen’s University Press, p. 183-199.

  26. Forsdyke DR. 2011. Sex chromosome dosage compensation. In Evolutionary Bioinformatics. 2nd edition. New York: Springer, p. 357-360.

End-Note (Nov 2013) Concentration Conservation: Proteins v mRNAs v DNAs

As set out elsewhere on these web-pages (Click Here), new technology is supporting McConkey (1982) in showing greater conservation of protein concentrations as compared with the corresponding mRNAs (Laurent et al. 2010; Khan et al. 2013). This is in accord with views on the greater selection pressures, both specific and collective (Forsdyke 2012), acting on proteins. Yet, there is growing evidence for selection pressure on nucleic acid structure, conventionally regarded as acting at the cytoplasmic level on mRNAs. This seeming paradox would be explained if the selective pressure on nucleic acid structure acted primarily at the DNA level, and RNA structure was often a secondary consequence of this (Click Here).

Forsdyke DR (2012) Functional constraint and molecular evolution. Encyclopedia of Life Sciences. John Wiley, Chichester.
Khan et al. (2013)  Primate transcript and protein expression levels evolve under compensatory selection pressures. Science 342, 1100-1104.

End-Note (Jan 2014) Method of Detecting Double Synthesizing Cells

A new method of differential staining of proteins encoded by the paternal or maternal X chromosomes became available (Wu et al. 2014). My remark on Zimmer's "Loom" blog is self explanatory: "One possible explanation of the extreme susceptibility of human females to autoimmune disease has been that they fail, partially or completely, to inactivate one X-chromosome, so that the concentration of X-chromosome gene products is increased in females, and this makes them susceptible to attack by their own lymphocytes. Scanning the beautiful pictures of Nathans and colleagues, one seldom comes across a cell of intermediate color (not pure red or pure green). And those that are seen are explained as due to the super-imposition of two cells. It would be of great interest to apply this fascinating new technology to organisms where the sex that dosage compensates is prone to autoimmune disease."

Wu et al. (2014) Cellular Resolution Maps of X Chromosome Inactivation: Implications for Neural Development, Function, and Disease. Neuron 81, 103-119.


End-Note (July 2017) Universality of Dosage Compensation

Although at the time resting on fragmentary evidence, the prediction was made (Forsdyke 1994) that "dosage compensation will be found in Z/W chromosome-bearing species." Subsequently the evidence grew stronger, but there were reports of incompleteness for Z/W species, with local rather than general dosage compensation. Huylmans et al. (2017) now provide a possible explanation:

"We find that, while some species and tissues seem to have incomplete dosage compensation, this is in fact due to the accumulation of male-biased genes and the depletion of female-biased genes on the Z chromosome. Once this is accounted for, the Z chromosome is fully compensated in all four species, through the up-regulation of Z expression in females and in some cases additional downregulation in males. ... Taken together, these results show that the uneven distribution of sex-biased genes on sex chromosomes can confound conclusions about dosage compensation and that Z chromosome-wide dosage compensation is not only possible but ubiquitous among Lepidoptera."

Huylmans AK, Macon A, Vicoso B (2017) Global dosage compensation is ubiquitouw in Lepidoptera, but counteracted by the masculinization of the Z chromosome. Mol. Biol. Evol (in press).

Go to: X-Chromosome and self/not-self discrimination (Click Here)

Go to: Main X-Chromosome page (Click Here)

Go to: Videolectures (Click Here)

Return to: Evolution Index (Click Here)

Return to: HomePage (Click Here)

 Posted on internet October 2012 after formal publication, and last edited 07 July, 2017 by Donald Forsdyke