Mammalian primitive erythroblasts undergo enucleation in the circulation, thus refuting the long-standing perception that primitive erythroblasts remain nucleated and are more similar to nucleated avian, fish, and reptile red cells than to definitive red cells of mammals.

During mammalian embryogenesis, erythropoiesis progresses through distinct phases, each phase producing cells with dramatically different characteristics. When cardiac contractions begin in mice embryos at embryonic day 8.25 (E8.25), “primitive” erythroblasts, developed in yolk sac blood islands, enter the circulation.1,2  However, by E12.5 “definitive” erythrocytes, produced in the fetal liver, begin to circulate and quickly prevail as the dominant erythroid phenotype.3  While definitive erythroblasts synthesize adult hemoglobins and enucleate, primitive red cells are larger, contain embryonic and adult hemoglobins, and have been thought not to undergo enucleation during their life span. Although several earlier observations hinted that in mouse embryos a population of large, enucleated cells might be circulating,4  the longstanding perception has been that primitive mammalian erythrocytes retain their nuclei and are thus more similar to nucleated avian, fish, and reptile red cells5  than to definitive red cells of mammals.

However, in this issue, Kingsley and colleagues (page 19) report quantitative data showing that between E12.5 and E16.5 primitive erythroblasts progressively enucleate in circulation. Further, they observed that enucleated, primitive cells can be detected up to 5 days after birth. Using antibodies to specific regions of murine embryonic βH1-globin and adult βmajor-globin, they were able to differentiate yolk sac–derived primitive red cells. These antibodies, in combination with nuclear staining, identified 3 distinct peripheral blood cell populations in E13.5 and E15.5 fetuses: a nucleated population expressing embryonic βH1-globin, an enucleated population lacking βH1-globin, and surprisingly, an enucleated population expressing βH1-globin. Small numbers of enucleated βH1-globin–expressing cells were initially detected at E12.5. By E16.5, all of the βH1-globin–expressing cells were enucleated. Morphometric analysis of cell area revealed that both nucleated and enucleated βH1-globin–expressing cells were 100 μm2 in size and about 3-fold larger than definitive erythrocytes. Importantly, the disappearance of circulating nucleated primitive cells was due to their progressive enucleation and not loss from the bloodstream.FIG1 

Immunohistochemistry with antiglobin antibodies. See the complete figure in the article beginning on page 19.

Immunohistochemistry with antiglobin antibodies. See the complete figure in the article beginning on page 19.

Close modal

These findings provide a persuasive argument refuting the currently held view that primitive mammalian erythropoiesis resembles avian and reptilian erythropoiesis more than definitive mammalian erythropoiesis. Indeed, the authors delineate a number of important similarities between murine primitive and definitive erythropoiesis. Both differentiation programs exhibit maturation with enucleation. Additionally, prior to extrusion, nuclei condense and move to the plasma membrane, coincident with loss of intermediate filaments.6  Yet one striking difference in the differentiation programs is that primitive erythroblasts appear to undergo terminal differentiation in circulation, while definitive erythroblasts mature extravascularly within 3-dimensional erythroblastic islands, closely associated with macrophages and extracellular matrix proteins. A number of new questions can now be asked. Do circulating primitive erythroblasts require contact with macrophages of the reticuloendothelial system for enucleation? What is the trigger for enucleation? Are molecular mechanisms of chromatin condensation, cytoskeletal remodeling, and nuclear extrusion similar or different in primitive and definitive erythroblasts? Do membrane mechanical properties of enucleated primitive cells differ from those of definitive cells, suggesting unique interactions among transmembrane and cytoskeletal components? The findings of Kingsley and colleagues open up many new avenues for exploration.

1
McGrath KE, Koniski AD, Malik J, Palis J. Circulation is established in a step-wise pattern in the mammalian embryo.
Blood
.
2003
;
101
:
1669
-1676.
2
Ji RP, Phoon CKL, Aristizábal O, McGrath KE, Palis J, Turnbull DH. Onset of cardiac function during early mouse embryogenesis coincides with entry of primitive erythroblasts into the embryo proper.
Circ Res
.
2003
;
92
:
133
-135.
3
Brotherton TW, Chui DHK, Gauldie J, Patterson M. Hemoglobin ontogeny during normal mouse fetal development.
Proc Natl Acad Sci U S A
.
1979
;
76
:
2853
-2857.
4
Bethlenfalvay NC, Block M. Fetal erythropoiesis. Maturation in megaloblastic (yolk sac) erythropoiesis in the C57Bl/6J mouse.
Acta Haemat
.
1970
;
44
:
240
-245.
5
Lazarides E. From genes to structural morphogenesis: the genesis and epigenesis.
Cell
.
1987
;
51
:
345
-356.
6
Sangiorgi F, Woods CM, Lazarides E. Vimentin down-regulation is an inherent feature of murine erythropoiesis and occurs independently of lineage.
Development
.
1990
;
110
:
85
-96.
Sign in via your Institution