Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

In Retrospect

Forty years of cellular clues from worms

The cell divisions that occur when a larva develops into an adult Caenorhabditis elegans worm were described in a cell-lineage map in 1977. The work has provided the foundation for many discoveries about developmental mechanisms.

Forty years ago, John Sulston and Robert Horvitz1 published a landmark paper that launched the nematode worm Caenorhabditis elegans on course to becoming a premier organism in which to study one of life's mysteries: how the cells of an embryo develop to form an adult animal. In the 1970s, the biggest questions of molecular biology were thought to have been answered, and the two authors joined Sydney Brenner in tackling the next big unsolved problem — how genes control development. Progress was being made at that time in understanding the life cycle of viruses called bacteriophages through genetics and elegant experimental logic, and C. elegans seemed perfect for taking a similar approach to investigating the development of a multicellular organism.

C. elegans are transparent, making them easy to observe using an optical technique, called Nomarski differential interference contrast microscopy, that highlights density differences to reveal subcellular detail, such as the nucleus and chromosomes at the mitotic stage of cell division. By means of this technique, Sulston and Horvitz made continuous observations of living worms, using hand drawings to record cell positions, divisions, migration, differentiation and, in some cases, programmed death, for each cell in the newly hatched worm until it reached adulthood two days later. This process was aided by the consistency of development that results from the genetic uniformity of the self-fertilizing C. elegans hermaphrodite and its invariant pattern of cell divisions.

The authors assembled a cellular lineage — an enormous cellular 'family tree' of all the cells in the worm. Sulston and Horvitz's work, and other studies2,3, described how the newly hatched larvae containing 550 cells develop into an adult hermaphrodite that has 959 cells, or an adult male with 1,031 cells. In 2015, a cell division in the head of the male4 was reported that Sulston and Horvitz had not spotted.

An obvious initial conclusion about the mechanisms driving worm development was that the fate of a cell probably depends on its ancestry, suggesting that daughter cells are different because of unequal inheritance of cellular components during cell division. Another insight came from the observations of similar cell-division patterns in different sections of the lineage. For example, the V1 cell divides in a repetitive pattern to form cells on the worm's outer surface. In part of the V1 lineage, cell division occurs to form an anterior hypodermal cell, which does not undergo further division, and a posterior seam cell. The posterior seam cell divides to form another anterior hypodermal cell and a posterior seam cell, and this division pattern is repeated (Fig. 1). This pattern also occurs for other cell lineages, including V2, V3, V4, V5 and V6, suggesting that a small number of subprograms might be used more than once to drive development. Deviations from this pattern, such as the formation of a neuronal precursor cell in part of the V5 cell lineage, hinted that each subprogram was also controlled by a specific genetic module.

Figure 1: Understanding development at a cellular level in the worm Caenorhabditis elegans.

How a larva develops into an adult animal at the cellular level was unknown until Sulston and Horvitz1 reported the cell lineage of post-embryonic development in C. elegans in 1977. The authors recorded the cell divisions in the worm, creating lineage trees such as that for the V1 cell (a section of which is shown), which gives rise to hypodermal cells and seam cells. Experiments in the 1970s and '80s investigated the regulation of worm development by analysing mutants that have abnormal patterns of cellular division, including mutants that undergo symmetric rather than asymmetric divisions, and by laser-based cell-ablation studies to identify the role of interactions between cells. Since the 1990s, the publication of the complete C. elegans genome sequence17 and the use of green fluorescent protein (GFP) to monitor specific cells have further aided studies of C. elegans development. Perhaps in the future, our understanding of worm development will reach the level of knowing the function of all genes in a single cell type.

The worm's invariant cell lineage offers an advantage for genetic studies because any change will stand out in stark contrast to the usual pattern. On the basis of the wild-type development, potential mechanisms of developmental control could be predicted, and these predictions suggested that mutants with specific types of abnormal development could be found. Indeed, many such mutants were identified by the authors and by their directly or indirectly trained scientific 'descendants', leading to the discovery of a staggering array of key developmental-control genes, and launching a thousand careers, including my own.

Cell-lineage mutants were identified rapidly, and many of these illuminated core developmental processes. For example, the unc-86 gene was found to have a role in mother–daughter identity in neuronal lineages and provided hints about how daughter cells become different from their mother cells5. Mutations6 of the lin-12 gene, which encodes Notch-family receptor proteins, affect cell fates in small groups of multipotent cells (those able to form more than one cell type), providing insights into how a cell's neighbours specify its fate.

The lin-4 gene7 was shown to affect the relative timing of developmental events, and was later identified8 as encoding the first known microRNA, an evolutionarily conserved non-coding RNA molecule that affects gene expression. The control of region-specific cell identity was revealed by mutations9 in the gene mab-5. And mutants in the lin-17 gene, which encodes a receptor for the intercellular signalling protein Wnt, was found to have some identical daughter cells that should have been different10, thereby revealing mechanisms of control of asymmetric divisions. The normal developmental roles of the cancer-promoting proteins epidermal growth factor (EGF) and Ras were also found by analysing mutations11 that altered cell lineages.

Another approach to investigating C. elegans development was by cell ablation using a focused laser microbeam. The concept of development regulated by cell-intrinsic mechanisms was challenged by the results of such ablation experiments12,13, in which the destruction of some cells changed the fate of their neighbouring cells. This idea was also supported by cases of indeterminate development, in which a pair of cells assumed positions at random and their fate depended on their position1,2,3.

“The Wnt signalling pathway demonstrates the interplay between intrinsic and extrinsic factors in C. elegans development.”

The pendulum swung towards the need for cell interactions in development when some of the cell-lineage mutants were found to affect signalling pathways between cells regulated by the proteins Notch, Wnt and EGF (ref. 14). Yet the pendulum swung back towards cell-intrinsic mechanisms when Wnt signalling was shown to be required in a parental cell that divides along an anterior–posterior axis to generate daughter cells that have distinct fates15. Wnt signalling in the direct parent or grandparent of a cell can affect its fate, as was shown by the finding that four cellular 'cousins' have distinct states that depend on the level of Wnt signalling received from their parents and grandparents16. This concept fits beautifully with the modularity of C. elegans lineages, because most cells undergo only one to three divisions. Thus, the role of the Wnt signalling pathway demonstrates the interplay between intrinsic and extrinsic factors in C. elegans development.

C. elegans was the first animal to be completely described at the cellular level in terms of its anatomy, development and neuronal connections, and also the first to have its genome sequenced17. We could, in principle, discover the effect of every gene on every cell in the worm. The use of green fluorescent protein as a marker has greatly aided the ability to identify and monitor cells. Next-generation DNA sequencing can define gene expression in individual cells with exquisite temporal resolution18, and, along with ongoing efforts to knock out each gene in the C. elegans genome, we can hope that we might discover the function of all genes in a single cell type. Such progress would help to answer fundamental questions about the relationship between an organism's genetic make-up (its genotype) and its physical form (phenotype). Extension of such understanding to a parent cell and its two progeny, coupled with cell-biological experiments, would exploit the potential unlocked by Sulston and Horvitz to reveal how the lineage history of a cell influences its fate. Footnote 1


  1. 1.

    See all news & views


  1. 1

    Sulston, J. E. & Horvitz, H. R. Dev. Biol. 56, 110–156 (1977).

    CAS  Article  Google Scholar 

  2. 2

    Kimble, J. & Hirsh, D. Dev. Biol. 70, 396–417 (1979).

    CAS  Article  Google Scholar 

  3. 3

    Sulston, J. E., Albertson, D. G. & Thomson, J. N. Dev. Biol. 78, 542–576 (1980).

    CAS  Article  Google Scholar 

  4. 4

    Sammut, M. et al. Nature 526, 385–390 (2015).

    CAS  ADS  Article  Google Scholar 

  5. 5

    Chalfie, M., Horvitz, H. R. & Sulston, J. E. Cell 24, 59–69 (1981).

    CAS  Article  Google Scholar 

  6. 6

    Greenwald, I. S., Sternberg, P. W. & Horvitz, H. R. Cell 34, 435–444 (1983).

    CAS  Article  Google Scholar 

  7. 7

    Ambros, V. & Horvitz, H. R. Science 26, 409–416 (1984).

    ADS  Article  Google Scholar 

  8. 8

    Lee, R. C., Feinbaum, R. L. & Ambros, V. Cell 75, 843–854 (1993).

    CAS  Article  Google Scholar 

  9. 9

    Kenyon, C. Cell 46, 477–487 (1986).

    CAS  Article  Google Scholar 

  10. 10

    Sternberg, P. W. & Horvitz, H. R. Dev. Biol. 130, 67–73 (1988).

    CAS  Article  Google Scholar 

  11. 11

    Horvitz, H. R. & Sternberg, P. W. Nature 351, 535–541 (1991).

    CAS  ADS  Article  Google Scholar 

  12. 12

    Sulston, J. E. & White, J. G. Dev. Biol. 78, 577–597 (1980).

    CAS  Article  Google Scholar 

  13. 13

    Kimble, J. Dev. Biol. 87, 286–300 (1981).

    CAS  Article  Google Scholar 

  14. 14

    Sawa, H., Lobel, L. & Horvitz, H. R. Genes Dev. 10, 2189–2197 (1996).

    CAS  Article  Google Scholar 

  15. 15

    Lin, R., Hill, R. J. & Priess, J. R. Cell 92, 229–239 (1998).

    CAS  Article  Google Scholar 

  16. 16

    Zacharias, A. L., Walton, T., Preston, E. & Murray, J. I. PLoS Genet. 11, e1005585 (2015).

    Article  Google Scholar 

  17. 17

    The C. elegans Sequencing Consortium. Science 282, 2012–2018 (1998).

  18. 18

    Schwarz, E. M., Kato, M. & Sternberg, P. W. Proc. Natl Acad. Sci. USA 109, 16246–16251 (2012).

    CAS  ADS  Article  Google Scholar 

Download references

Author information



Corresponding author

Correspondence to Paul W. Sternberg.

Related links

Related links

Related links in Nature Research

Systems biology: Molecular memoirs of a cellular family

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Sternberg, P. Forty years of cellular clues from worms. Nature 543, 628–630 (2017).

Download citation

Further reading


Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing