Unit: Macrophages and Development of Immunity - CNRS URA 2578

Director: Philippe Herbomel

We take advantage of the transparency of the zebrafish embryo and frye to study directly in vivo the origin, deployment in the organism, and functional properties of its first leukocytes - the 'early' macrophages - and the interaction of the latter with various microbes of medical interest.

Our model is the zebrafish, a small freshwater fish quite common in domestic aquaria. Over the last fifteen years, the zebrafish has become one of the major model species for the developmental biology of vertebrates (beside mouse, chick, frog), mainly due to two great assets :

i) The adults can be raised easily, and large numbers of embryos can easily be recovered: dawn light triggers mating, and each female lays 100-200 eggs at a time, which will develop as embryos synchronously.

ii) The embryos develop in water and are fully transparent from fertilization until well after hatching, making it possible to follow their development in great detail.

– with a stereomicroscope; many can thus be quickly examined. This made it possible to set up, for the first time among vertebrates, large-scale screens for mutants altered in the development of various organs (to date, over 1000 different mutants have been characterized).

– with a compound microscope equipped with Differential Interference Contrast (DIC or 'Nomarski'), allowing to visualize individually all cells in the embryo and follow their behaviours in a non invasive manner.

We have improved this Nomarski video-microscopic approach of cell behaviours in the developing embryo. This allowed us to discover, among other novelties, an "early" lineage of macrophages, which constitute the embryo's first leucocytes (white blood cells). Analogous early macrophages have been found in all vertebrate model species. In the zebrafish, we have found that they originate from an unexpected small region of the embryo: the cephalic lateral mesoderm, just adjacent to the cardiac territory (whereas all other leucocyte types that arise later originate from hematopoietic stem cells born at the other end of the embryo, from the caudal mesoderm (see report of the Lymphocyte Development Unit). In a collaboration with B. and C. Thisse (IGBMC, Illkirch), we have identified a series of genes whose expression marks the cells of this lineage, allowing to trace they ontogeny since gastrulation.

From the antero-cardiac mesoderm, the mesodermal precursors of the early macrophages emigrate to the neighboring yolk sac (which contains the embryo's food store), where they differentiate into "pre-macrophages" (hematopoietic-like progenitors), then into young macrophages, which show the swift, amoeboid motility typical of leukocytes. From the yolk sac, they quickly invade the embryo's head tissues; first the mesenchyme between developing organs, and from it the (still unvascularized) organs themselves. Our video sequences reveal that once in the tissues, they keep wandering without letting up, including within epthelial tissues (brain, epidermis).

At a precise developmental stage, all early macrophages present in the brain and retinas simultaneously acquire a new phenotype, which we named "early microglia" : they become much more endocytic, and their gene expression pattern shows clear changes.

Just as mammalian macrophages, zebrafish early macrophages express the M-CSF (CSF-1) receptor. We have shown that in mutant embryos that lack a functional M-CSF receptor, the early macrophages nevertheless differentiate and proliferate normally in the yolk sac. But then they do not migrate into the head tissues to any extent, and thus remain in the yolk sac and blood circulation.

We are now studying the molecular and cellular basis of the early macrophages' invasive abilities, as well as their long-term fate, in the young and in the adult.

Finally, we have shown that the early macrophages of the embryo are capable of detecting a bacterial infection focus (gram+ or gram-), go there and clear the bacteria - all this at a developmental stage when they are still the only leukocytes in the organism.

In a collaboration with L. Ramakrishnan (University of Washington, Seattle), we then studied their interactions with a pathogenic mycobacterium injected in the embryo's blood, Mycobacterium marinum, the agent of fish tuberculosis, and a close genetic relative of M. tuberculosis (the agent of human tuberculosis). M. marinum was quickly phagocytosed by the early macrophages, but survived and slowly proliferated inside them, leading to the dissemination of the infection in the embryo through the infected wandering macrophages. Unexpectedly, two days later, infected macrophages started to gather and form dense, granuloma-like aggregates, within which mycobacterial genes previously identified in adult animals as "granuloma-specific" were activated. This came as a surprise for at this developmental stage, the embryos still have no lymphocytes - which were so far considered as required for the triggering of the granulomatous response.

Photo :

Top: live zebrafish embryo; bottom: close-up on an early macrophage wandering in the yolk sac (two pictures taken 3 min. apart; Nomarski video-microscopy).

Keywords: development, macrophages, hematopoiesis, zebrafish, in vivo imaging, infection

Activity Reports 2003 - Institut Pasteur

Page Top research Institut Pasteur homepage

If you have problems with this Web page, please write to rescom@pasteur.fr