Ephrussi, Boris

views updated


(b. Moscow, Russia, 9 May 1901;

d. Paris, France, 2 May 1979), embryology, development, physiology, genetics.

Ephrussi was a Russian-born French geneticist who emigrated to France through Romania, after the Bolshevik revolution. Awarded the Louisa Gross Horwitz Prize from Columbia University in 1975, he was a pioneer in developmental and physiological genetics, and made fundamental contributions to understanding the biochemical processes depending on gene action, to understanding the role of genes in biochemical processes, and in articulating the significance of cytoplasmic factors in inheritance and development. Ephrussi was perhaps the most important of several French biologists who together were crucial to the emergence and development of French genetics in the twentieth century; he held the first chair in genetics at the Sorbonne.

Ephrussi was married to Harriet Taylor Ephrussi (1918–1968), who was a biologist that worked on the structure of DNA. They had a daughter, Anne.

Early Education . Ephrussi was trained as an embryologist in the 1920s at the Marine Biological Station at Roscoff, in France. His earliest work, under the biochemist Louis Rapkine, focused on environmental factors influencing embryological development; in particular, he studied the effects of temperature on the development of sea urchin embryos. Ephrussi, in a second thesis (which was typical at the time), also did research on tissue cultures, under Emmanuel Fauré-Fremiet, with less than satisfactory results. In the end, he concluded in this project that some “intrinsic factors” played a role in development. The more fundamental interest reflected in these theses concerned the role of intracellular and extracellular processes in the regulation of development (Ephrussi, 1932). Throughout, Ephrussi later said, his interest was in understanding “the chain of reactions connecting the gene with the character” (1938, p. 6). This interest continued throughout his life, as did his commitment to the idea that both intracellular and extracellular factors were crucial to understanding development, and that, among the intracellular factors, both nuclear and extranuclear factors affected development and function.

Ephrussi’s training in morphogenesis and development, at the time, would have been thought of as embryology; that is, he was engaged in understanding the various contributions that led from a fertilized egg to an organism, where those contributions were both internal and external to the organism. In the context of French biology during the 1930s, physiological geneticists and embryologists recognized substantial cytological control of development, emphasizing pathways, and chemical determinants beyond the gene (cf. Sapp, 1987). Mendelians, by contrast, tended to assume that the nucleus contained “factors” sufficient to determine ontogeny; that is, they assumed genes were what determined development. Developmentalists, in France and elsewhere, generally felt they needed to recognize both internal and external causes: given the acknowledged fact that cells have the same genetic material whatever their morphological fates, it seemed inevitable that epigenetic factors had to govern cell fate. In France, the role of the cytoplasm was broadly assumed to be crucial to understanding development. There was, in any case, no established Mendelian tradition in France in the years immediately following World War II, and, if anything, an antagonism to Mendelian thinking before the war. Ephrussi was unusual in being sympathetic with Mendelian inheritance, and simultaneously embedded in a robust developmental tradition. Throughout his career, Ephrussi also had considerable sympathy for cytoplasmic inheritance. His Mendelian sympathies fit somewhat uncomfortably with the more conservative French sympathies with Larmarckian inheritance; his developmental sympathies fit equally uncomfortably with the prevailing influence of Mendelian genetics led by Thomas Hunt Morgan (1866–1945) and his followers.

Work in Physiological Genetics . Ephrussi began working in genetics in the 1930s after finishing his embryological training. During that decade, the best understood organism from a genetic viewpoint was clearly Drosophila. Ephrussi wanted to approach the problems of cellular differentiation from a genetic and a physiological standpoint; this dual perspective required integrating information concerning cytoplasmic and environmental effects on development with the well-known and elaborate emphasis on chromosomal genetics characteristic of Mendelian work. Because there was no established tradition of Mendelian genetics in France at this point, from 1934 to 1935, supported by the Rockefeller Foundation, Ephrussi moved to the California Institute of Technology (Caltech), to work under Morgan. (Morgan had moved from Columbia to Caltech in 1928, with Alfred Henry Sturtevant and others following him.) It was during this period that Ephrussi began his seminal work with George Wells Beadle. Beadle at that point was a Cornell-trained corn geneticist who was a postdoctoral fellow in Morgan’s lab, engaged in studying crossing over in Drosophila. When Ephrussi moved back to Paris in 1935, Beadle followed him there.

The ultimate goal of Beadle and Ephrussi’s collaborative work was to gain an increased understanding of gene expression and development. In terms that would have been natural to Ephrussi, they were interested in the physiology of gene action. There were a number of significant influences on this work, some embryological and some genetic. Within experimental embryology, the transplantation of tissues had become an accepted technique, following the work by Ernst Wolfgang Caspari in Alfred Kühn’s zoological institute at Göttingen. Essentially, the transplantation of tissue enabled experimentalists to explore the influence of surrounding tissues on developing embryos, and so to see the several factors affecting development. So if at early embryological stages, some tissue (or cell) is displaced to a different context, one can see how the development of the displaced tissue is changed. Ephrussi and Beadle extended that technique to Drosophila. Drosophila is an organism that was well understood from a genetic point of view at the time, but developmentally it was a difficult organism to work with. Ephrussi and Beadle were able to turn the micromanipulator into a device to implant imaginal disks (which develop into eyes) onto Drosophila larvae. The implantation eventually allowed them to see how development (specifically of eye color) was affected by the larger context of the developing organism. Within Drosophila genetics, the nonautonomous control of development offered a useful target. Crossbreeding had already shown that the normal, or wild-type, eye color in Drosophila was the result of two pigments, one brown and one red. Either could be inhibited by mutations, with observably different outcomes. The resulting mutant colors are vermilion and cinnabar. Sturtevant, one of the most respected of Morgan’s protégés, had shown in 1919 that eye color in Drosophila could be altered under the influence of other tissues. Its development is in this sense nonautonomous, because it is not determined entirely by the intrinsic genetic constitution. So, for example, working with a gynandromorph—a mosaic fly with both male and female constitution—whose head showed characteristics that would require it to have vermilion eyes, Sturtevant found it in fact developed a wild-type eye color. The intrinsic genetic constitution failed to determine the eye color. Sturtevant concluded that development was somehow or other modified by “diffusible substances.” The result was intriguing, but unexplained, until Beadle and Ephrussi revealed their results.

Beadle and Ephrussi implanted imaginal disks onto embryos of flies with a different genetic constitution from the donor eyes (Ephrussi and Beadle, 1936; Ephrussi and Beadle, 1937). The goal was one of illustrating various influences on eye color. They already knew from genetic studies that there are two mutant forms, each thwarting the development of the wild-type (brown) eye color in Drosophila. The actual color of the eye would then be influenced both by the intrinsic constitution of the imaginal disk, and by the surrounding tissue. It was not surprising, in light of Sturtevant’s work, that the eye color would have these dual influences. What was surprising was the pattern of outcomes.

There were, then, three genotypes in play: vermilion (v), cinnabar (cn), and wild type (v+). These were available both in the imaginal disks and the hosts. As a result, there were six possible reciprocal implantations, as exhibited in Table 1. (It was not necessary to implant disks into hosts of the same genotype.)

Implanted Imaginal Disk HostBody TypeColor of Resulting Eye
Vermilion (v)Wild Type (v+)Wild Type (v+)
Vermilion (v)Cinnabar (cn)Wild Type (v+)
Cinnabar (cn)Vermilion (v)Cinnabar (cn)
Cinnabar (cn)Wild Type (v+)Wild Type (v+)
Wild Type (v+)Cinnabar (cn)Wild Type (v+)
Wild Type (v+)Vermilion (v)Wild Type (v+)

The wild-type imaginal implants develop, as predicted, into wild-type eyes. Vermilion and cinnabar imaginal disks transplanted into a wild-type larvae likewise both develop into eyes with the wild-type color. These results are compatible with Sturtevant’s earlier observations. They are also compatible with the view that cinnabar and vermilion are simple Mendelian recessives. The wild-type dominates in development as well. The crucial cases are the reciprocal implants between cinnabar and vermilion, given in the second and third rows in the table above. When a vermilion disk is implanted onto a cinnabar larvae it develops a wild-type color, whereas when a cinnabar disk is implanted onto a vermilion larvae it retains its cinnabar color. These outcomes could be accommodated only by assuming the two genes occupy sequential positions in the metabolic pathway responsible for the production of the pigments, rather than providing independent contributions. Ephrussi wrote, somewhat later,

There are two different substances, one responsible for the change from vermilion to wild type and the other for the change from cinnabar to wild type. The wild type lymph contains both these substances. The lymph of the mutant cinnabar contains only one of them, namely the substance responsible for the change from vermilion to wild type. The mutant vermilion contains none of these substances.… [The] two substances are formed in the course of a single chain of reactions, of which the v+ substance represents the first and the cn+ the second link: ⇒ v+ ⇒ cn+. (1942, pp. 329–330)

Beadle and Ephrussi thus concluded that there are at least two diffusible substances, and, developmentally, they are organized sequentially. The vermilion mutant fails to carry out the first reaction, and therefore also cannot synthesize the second of the diffusible substances. So the vermilion host cannot supplement the cinnabar implant. The cinnabar mutant can perform the first reaction, but not the second. As a host, therefore, it would supplement the needed material for the vermilion implant, resulting in a normal eye color.

The next logical step would have been to identify the specific substances involved. Because at this time the genetic code was unknown, they could not predict the substances directly from the genetic structure. The only alternative was to begin with the phenotypic effects, and infer from them what physiological pathways produce the observed effects on development. It was clear these factors shaping development were not genes, but they were relatively simple substances. By 1939 Beadle and Ephrussi concluded that the blockage of reactions was due to a lack of specific enzymes. They also concluded that it was necessary to work with some organism that was developmentally and physiologically less complex than Drosophila. Beadle returned to the United States, where he eventually took up work on Neurospora. Ephrussi continued with the Drosophila project until the Germans invaded France. In the end, they were able to identify one of the “hormones” involved in the Drosophila cases (tryptophan, an amino acid), but the other eluded them.

Interruption during World War II . Ephrussi’s work was interrupted by World War II. During that period, Ephrussi’s work was centered at Johns Hopkins University in Baltimore, Maryland, as a refugee. By 1944 he had assumed an active role in the forces of French liberation, and was ready to return to France after the war. Following the war’s conclusion, the French undertook, under the auspices of Centre national de la recherché scientifique (CNRS), to establish an institutional profile in genetics— one that had been lacking prior to 1945. Ephrussi was named the first chair of genetics at the Sorbonne. Ephrussi applied to the Rockefeller Foundation to obtain support for an Institute of Genetics at Gif under CNRS. After some controversy, this was approved in April 1950. In the end, these funds were never expended, and Ephrussi remained in Paris at the Institut de Biologie Physicochimique (the Rothschild Institute), and later at CNRS.

Yeast and Adaptive Mutations after the War . Ephrussi’s work continued to focus on the interaction of nuclear and cytoplasmic factors on development. He was convinced that both factors were crucial to development. At this point, he abandoned the use of Drosophila, thinking that yeast (Saccharomyces) was a more promising organism, while Beadle switched to Neurospora. Yeast was a commonly used organism for biochemical studies. With Louis Pasteur’s legacy, and the lucrative benefits connected with the beer industry in France, a great deal was known about the biochemistry associated with yeast. Considerably less was known about yeast genetics, with the first systematic studies only in the late 1930s. Yeast also exhibited what were called “adaptive mutations,” but the processes of mutations in yeast were not known. Initially, Ephrussi was interested in understanding whether these mutations were induced by environmental causes or were already present in the cultures at low frequencies; that is, the question was whether environmental stress led to mutant forms that were adaptive, or whether they were already present and the environment promoted them by selection. The former view, which Ephrussi eventually embraced, would have required cytoplasmic elements playing a significant role in development.

In the end, Ephrussi’s work went in a direction rather different than did that pursued by Beadle and Edward L. Tatum in the United States, although it started out allied with that work. Ephrussi and his collaborators raised yeast on a medium containing acriflavine (an antibacterial agent), and found the resulting colonies consisted almost exclusively of a slow-growing, and respiratory-deficient, form. The strain could not use oxygen. They called these mutants petites. The changes induced were irreversible: subsequent generations did not shift back to a respiring form, even in a more benign environment. The change was inherited, but it was not inherited in a Mendelian fashion. The implication was that it was nonchromosomal inheritance. By crossing wild types with these so-called petite strains that were respiration deficient, Ephrussi was able to observe the expression of various nuclear genes in different cytoplasmic contexts. Eventually, it became clear that the slow growth of these forms was the result of a deficiency in respiratory enzymes. (Yeast is a facultative anaerobe.) The real focus of Ephrussi’s interest was in the contribution of cytoplasmic factors to development. He and his collaborators managed to show that there was crucial information contained in the cytoplasm—later identified as residing in mitochondria—that is necessary for respiration. These cytoplasmic properties were not nuclear genes, though they did appear to depend on them.

For the next decade, the mechanism responsible for this adaptive modification in yeast was the subject of considerable controversy. Ephrussi subjected the yeast to a series of anaerobic conditions. He and his collaborators were eventually able to produce adaptive responses that were reversible. In the end, Ephrussi concluded that nuclear genes, the cytoplasm, and the environment code-termine the constitution of the cell. This led Ephrussi to reject the view that cellular enzymes were exclusively under the control of nuclear genes, a view most forcibly defended by his former collaborator, Beadle. Ephrussi denied there was any evidence for the nuclear origin of the cytoplasmic factors in yeast.

It turns out that the activity of mitochondria is necessary for respiration. Without the enzymes provided by the mitochondria, respiration ceases, and the loss is irreversible. It also turns out that there are a number of distinct mitochondrial genes involved, which can be lost separately. Sometimes this shows up because cells capable of respiration occasionally yield some progeny that are competent and others that are not; this is due to the inheritance of different portions of the mitochondrial particles.

Somatic Cell Hybrids . In the 1950s Ephrussi continued his exploration of somatic cell differentiation (e.g., Ephrussi, Davidson, and Yamamoto, 1966). The general theme remained the same. As an embryologist, he was interested in the factors that contribute to cell differentiation, and how those factors limit developmental potentials. More conventional geneticists were convinced that all cells had the same potentialities for development because they are genetically identical, which meant that their nuclear genes were equally endowed (cf. Ephrussi, 1970). That, of course, again left the differences among cells unexplained.

In 1962 Ephrussi moved to Western Reserve University in Cleveland, Ohio, where he stayed until 1967, focusing on intraspecific and interspecific somatic cell hybrids. Interspecific hybrids are formed by encouraging the fusion of cells from different species, including both the cytoplasm and the nuclear materials. This often leads to chromosome loss, and allows the association of specific genes with selected chromosomes. (In this way it is fundamentally different from the more recent work using nuclear implantation.) He developed a number of techniques using hybrids formed from mouse and human cells in culture. Ephrussi’s Hybridization of Somatic Cells(1972) provides a detailed review of the work conducted in his laboratory.

Intraspecific hybrids offered the opportunity to address the same problems that had occupied Ephrussi throughout his career, though without depending on the vicissitudes of sexual reproduction (see Zallen and Burian, 1992). As the technique had originally been developed by Georges Barski, Serge Sorieul, and Francine Cornefert at the Institut Gustave Roussy in France, hybrid cells were identifiable only by chromosomal features. Ephrussi’s laboratory developed a number of other markers, including biochemical ones. Once again, he was able to return to the question of cell differentiation. He distinguished between “household” and “luxury” functions—the former being required for fundamental metabolic functions in all cells, and the latter being what distinguishes particular cells from one another. One intriguing observation will illustrate the kind of results they achieved. When they created fusions between differentiated cells from different lines, it turned out that the “luxury” functions were lost. Ephrussi took this as evidence that the hybridization repressed the “luxury” functions. Similar results were found in a number of laboratories working with hybrid cell lines.

Throughout his career, Ephrussi maintained a focus on issues of cell differentiation and development. His various shifts in the kinds of organisms he used, and his development of novel techniques, reflect his willingness to modify the tools used in his research to address the fundamental problems concerning the regulation of development and the interaction between nuclear and cytoplasmic factors.


Ephrussi resources can be found at the Rockefeller Archive Center and the Archive of the Institut Pasteur.


“D’une temperature elevée sur la mitose de segmentation des oeufs d’oursin.” Comptes rendus de l’Académie des Sciences, Paris177 (1932): 152–154.

“Contribution a l’analyse de premiers stades du développement de l’oerouf.” Action de lat termperérature. Impremature de l’Academie, Paris.

“The Absence of Autonomy in the Development of the Effects of Certain Deficiencies in Drosophila melanogaster.” Proceedings of the National Academy of Sciences of the United States of America 20 (1934): 420–422.

With George W. Beadle. “La transplantation des disques imaginaux chez la Drosophile.” Comptes rendus de l’Académie des Sciences, Paris 201 (1935): 98–100.

———. “The Differentiation of Eye Pigments in Drosophila as Studied by Transplantation.” Genetics 21 (1936): 225–247.

———. “Development of Eye Colors in Drosophila: Diffusible Substances and Their Interactions.” Genetics 22 (1937): 76–86.

“Aspects of the Physiology of Gene Action.” American Naturalist 72, no. 738 (1938): 5–23.

“Chemistry of ‘Eye Color Hormones’ of Drosophila.” Quarterly Review of Biology 17 (1942): 327–338.

“The Interplay of Heredity and Environment in the Synthesis of Respiratory Enzymes in Yeast.” Harvey Lectures 46 (1950): 45–67.

Nucleo-Cytoplasmic Relations in Microorganisms: Their Bearing on Cell Heredity and Differentiation. Oxford: Clarendon Press, 1953.

With Richard L. Davidson and Kohtaro Yamamoto. “Regulation of Pigment Synthesis in Mammalian Cells, as Studied by Somatic Hybridization.” Proceedings of the National Academy of Sciences of the United States of America 56 (1966): 1437–1440.

“Somatic Hybridization as a Tool for the Study of Normal and Abnormal Growth and Differentiation.” In Genetic Concepts and Neoplasia. Baltimore, MD: Williams and Wilkins, 1970.

Hybridization of Somatic Cells. Princeton, NJ: Princeton University Press, 1972.


Burian, Richard M., Jean Gayon, and Doris Zallen. “The Singular Fate of Genetics in the History of French Biology, 1900–1940.” Journal of the History of Biology 21 (1988): 357–402.

———. “Boris Ephrussi and the Synthesis of Genetics and Embryology.” In A Conceptual History of Embryology, edited by Scott F. Gilbert, 207–227, New York: Plenum Press, 1991.

Lwoff, André. “Recollections of Boris Ephrussi.” Somatic Cell and Molecular Genetics5 (1979): 677–679.

Roman, Herschel. “Boris Ephrussi.” Annual Review of Genetics 14 (1980): 447–450.

Sapp, Jan. Beyond the Gene: Cytoplasmic Inheritance and the Struggle for Authority in Genetics. New York: Oxford University Press, 1987.

Weiss, Mary C. “Contributions of Boris Ephrussi to the Development of Somatic Cell Genetics.” BioEssays 14 (1992): 349–353.

Zallen, Doris T., and Richard M. Burian. “On the Beginnings of Somatic Cell Hybridization: Boris Ephrussi and Chromosome Transplantation.” Genetics 132 (1992): 1–8.

Robert C. Richardson