Brought to you by Portland Press Ltd.
Published on behalf of the International Federation for Cell Biology
Cancer Cell death Cell cycle Cytoskeleton Exo/endocytosis Differentiation Division Organelles Signalling Stem cells Trafficking
Cell Biology International (2003) 27, 747–753 (Printed in Great Britain)
Haemoglobin biosynthesis site in rabbit embryo erythroid cells
Aurora M. Cianciarullo1*, Álvaro L. Bertho2, Maurilio J. Soares3, Tânia M. Hosoda1, Simone Nogueira‑Silva1 and Willy Beçak1
1Laboratório de Genética, Instituto Butantan, Avenida Vital Brazil 1500, 05503-900 São Paulo-SP, Brazil
2Departamento de Protozoologia, Instituto Oswaldo Cruz, FIOCRUZ, Avenida Brasil 4365, 21045-900 Rio de Janeiro-RJ, Brazil
3Departamento de Ultraestrutura e Biologia Celular, Instituto Oswaldo Cruz, FIOCRUZ, Avenida Brasil 4365, 21045-900 Rio de Janeiro-RJ, Brazil


Properly metabolized globin synthesis and iron uptake are indispensable for erythroid cell differentiation and maturation. Mitochondrial participation is crucial in the process of haeme synthesis for cytochromes and haemoglobin. We studied the final biosynthesis site of haemoglobin using an ultrastructural approach, with erythroid cells obtained from rabbit embryos, in order to compare these results with those of animals treated with saponine or phenylhydrazine. Our results are similar to those obtained in assays with adult mammals, birds, amphibians, reptiles and fish, after induction of haemolytic anaemia. Therefore, the treatment did not interfere with the process studied, confirming our previous findings. Immunoelectron microscopy showed no labelling of mitochondria or other cellular organelles supposedly involved in the final biosynthesis of haemoglobin molecules, suggesting instead that it occurs free in the cytoplasm immediately after the liberation of haeme from the mitochondria, by electrostatic attraction between haeme and globin chains.

Keywords: Erythroid cells, Erythropoiesis, Mitochondria, Haemoglobin biosynthesis, Immunocytochemistry, Ultrastructure, Rabbit embryo, Oryctolagus cuniculus.

*Corresponding author

1 Introduction

In mammals, both erythropoiesis and granulopoiesis are extravascular, and cells that have matured must enter the circulation through specialized endothelial cells that regulate their egress from the bone marrow (Wilson and Tavassoli, 1994). During the developmental embryonic phase, the yolk sac is the primordial haemopoietic organ of upper vertebrate embryos. Because of its particular status, the yolk sac has been generally held as the purveyor of all stem cells, supposedly colonizing the haemopoietic organs later, as these become sequentially functional in the embryo (Dieterlen-Lièvre et al., 1994). Erythropoiesis begins in the yolk sac during the mesoblastic period of haematopoiesis in all embryonic vertebrates. At about 6 weeks of embryonic life in the human, production of erythrocytes decreases in the yolk sac and begins in the embryo. Rabbit embryos have 90%–100% of their immature erythroid cells in their peripheral blood by about 18 days or 2 1 2weeks of gestation. This is adequate for erythropoietic studies related to haeme and haemoglobin biosynthesis, as rabbits can be used to get immature erythroid cells from peripheral blood without the interference of anaemia inductor agents.

The aim of this study is to identify any detectable differences between the results obtained with our experimental model and those obtained with phenylhydrazine and saponine treatment, which are often used as haemolytic anaemia inductors. Another aim is the localization of haemoglobin antigens in situ, based on the immunogenic properties of haemoproteins, using post-embedding immunoelectron microscopy and a polyclonal anti-haemoglobin antibody revealed by protein A-gold complex. The targets are mitochondria and other cellular organelles supposedly involved in the finalbiosynthesis of haemoglobin molecules, regarding the integration of haeme and four polypeptide globin chains (Cianciarullo et al., 1999).

2 Materials and methods

2.1 Obtaining immature erythroid cells

An adult rabbit (Oryctolagus cuniculus) at 18 days of gestation, weighing 2.5 kg, was ethyl ether anaesthetized and the embryos were removed for blood collection. Blood samples were collected in 3.8% sodium citrate (10:1), and procedures for blood smears, electron microscopy and flow cytometry were immediately started.

2.2 Blood smears

Blood samples were stained with 0.5% new methylene blue dye (Sigma Chemical Co., St Louis, MO, USA) in aqueous solution (Brecher, 1949) and counter-stained with Rosenfeld's dye (Rosenfeld, 1947). Smears were observed in a Zeiss Photomicroscope and the percentage of immature erythroid cells was determined by smear scanning. Approximately 500 different erythroid cells were counted per glass slide preparation.

2.3 Scanning electron microscopy

Mature cells from adult rabbit (O. cuniculus) and amphibian (Odontophrynus americanus) were fixed for 1h with 2.5% glutaraldehyde in 0.1M sodium cacodylate buffer, pH 7.4, at room temperature. Cells were rinsed twice in the same buffer and adhered to coverslips coated with 0.1% poly-l-lysine for 10min. They were post-fixed in 0.1% OsO4in the same buffer for 10 min, rinsed twice in buffer, dehydrated in acetone series and critical point dried. Gold metallization was performed in a Balzers Union MED 010 apparatus at 25A for 2min, and a cell coating of 5 nm thickness was obtained.

Fig. 1

Mature erythrocytes of rabbit (spherical shape) and amphibian (ellipsoidal shape) processed for conventional scanning electron microscopy. Bar=5 μm.

2.4 Rhodamine 123 fluorescent probe

The cationic laser dye Rhodamine 123 (Polysciences Inc., Warrington, PA, USA) was dissolved in sterile tri-distilled water at 1 mg/ml, and stored at 4 °C in the dark. Just before use, working solutions were prepared according to Johnson et al. (1980), at 10 μg/ml in DMEM (Dulbecco's Modified Eagle Medium, Sigma). The cells were incubated in a 95%:5% air-CO2incubator at 37 °C for 30 min. The dye was removed by centrifugation at 800 rpm for 2 min. Blood cells were resuspended in 0.1 M PBS, pH 7.2, centrifuged at 800 rpm for 2 min to remove the supernatant, and resuspended in 200 μl PBS. Adhesion of erythroid cells was performed on glass slides coated with silane (aminoalkylsilane, Sigma) for 5 min in the dark at room temperature. They were mounted in 10 μl 25 mg/ml anti-fading 1,4-diazobicyclo-(2.2.2)-octane (DABCO, Sigma) (Johnson et al., 1982) in PBS, pH 8.6, plus 50 μl DMEM containing 10% foetal bovine serum. Preparations were examined immediately in a Zeiss Axiophot microscope with epifluorescence, using the filter system for fluorescein (blue excitation) to prolong observation time (Chen, 1990).

Fig. 2

Living immature erythrocytes of rabbit embryo incubated with Rhodamine 123. All white dots seen in the cytoplasm correspond to mitochondria. Bar=5 μm.

2.5 Flow cytometric analysis

Incorporation of rhodamine 123 (Rh123) by cells was determined using flow cytometric analysis (Darzynkiewicz et al., 1982). The cells were analysed to determine their size (FS-forward scatter), cytoplasmic granularity (SS-side scatter), and intensity of fluorescence incorporation, using an EPICS 751 flow cytometer (Coulter Electronics, Hialeah, FL, USA). The data were obtained using an argon laser with 488 nm excitation wavelength and red fluorescence detection (green excitation) at 600–650nm bands. Ten thousandcells were analysed in each sample and the data plotted in single and double parameter histograms. A multi-angle light scatter fingerprint was taken at each time-point, in order to establish a correlation between granularity (90°C scatter) and intensity of fluorescence incorporation by the cells (Shapiro, 1985; Watson, 1991).

Fig. 3

Graphics from flow cytometric analyses comparing mature erythroid cells of adult rabbit (a, b) and immature cells of rabbit embryos (c, d). Cells were incubated with Rh123 to select mitochondria in living cells, detected at the phycoerythrin (PE1) channel, in the mapping of each sample (MAP1). Controls were performed using blood samples without Rh123 (a, c).

2.6 Immunoelectron microscopy

Mature erythroid cells from an adult rabbit and immature erythroid cells from rabbit embryos were fixed in 2% paraformaldehyde, 0.1% glutaraldehyde and 0.01% picric acid, in 0.1 M sodium cacodylate buffer, pH 7.4, plus 3.5% sucrose, at 4 °C for 30 min. Cells were then washed in buffer, dehydrated in methanol series and embedded in Lowicryl K4M (all reagents from Sigma). Ultra-thin sections collected on nickel grids were incubated in a drop of 50 mM ammonium chloride at room temperature for 40 min. They were then transferred to a solution containing 0.02 M TBS, 1% BSA, 1% Tween 20 for 3×10 min, followed by incubation with goat anti-human haemoglobin antibody diluted 1:100 in PBS, at 37 °C in a humid chamber for 2 h. After rinsing in PBS, sections were incubated with Protein A-gold complex diluted 1:100 in PBS, at room temperature for 1 h, rinsed in PBS, then distilled water, and examined in a Zeiss EM 109 transmission electron microscope. A negative control was included using goat anti-human IgG instead of anti-human haemoglobin, and omitting the antiserum.

3 Results

3.1 Blood-smear analyses

The adult rabbit showed mature erythrocytes and about 2% reticulocytes in its peripheral blood. Rabbit embryo blood-smears showed 90–100% immature erythroid cells in peripheral blood, predominantly made up of erythroblasts. These cells presented a characteristic dense reticulum uniformly distributed around thenucleus (data not shown).

Figs 4–6

Immature erythrocytes of rabbit embryo immunolabelled with anti-human haemoglobin antibody revealed by protein A-colloidal gold particle complexes. Labelling can be observed throughout the cytoplasm of an erythroblast (Fig. 4), but is absent in cytoplasmic organelles, iron-containing vesicles (Fig. 5), endoplasmic reticulum and mitochondria (Fig. 6). Nucleus, N; mitochondria, m; polysomes, arrowheads; iron-containing vesicles, arrows; endoplasmic reticulum, er. Bars=0.5 μm (Figs 4, 6); Bars=0.25 μm (Fig. 5).

3.2 Scanning electron microscopy

Red blood cells from the adult rabbit and amphibian were collected from peripheral blood, mixed, and processed for scanning electron microscopy (Fig. 1). Rabbit erythrocytes are about half the size of amphibian erythrocytes, presenting a thick spherical profile, with a biconcave shape. Conversely, amphibian erythrocytes present a thin ellipsoidal profile, with a flattened, biconvex shape. It is interesting to observe that both cells transport a similar haemoglobin content, despite their huge difference in size. The characteristic biconcave shape of mammalian red cells from reticulocyte to haematia reveals bizarre shapes in thin sections at ultrastructural level, which needs to be considered.

3.3 Epifluorescence microscopy

Only 2% of living cells from the adult rabbit showed positive staining for Rh123, corresponding to immature erythroid cells. In the embryos, almost 100% of living erythroid cells showed positive staining for Rh123. Rod-like and filamentous mitochondria distributed around the nuclei were strongly fluorescent, as can be seen in Fig. 2. They seem to represent a significant portion of the dense reticulum found in blood smears.

3.4 Flow cytometric analysis

The histograms shown in Fig. 3(a–d) are on a logarithmic scale for fluorescence intensity, in comparison with the cell number, previously defined in 10,000 cells. A fluorescent peak for mitochondria was found in rabbit embryo blood cells (Fig. 3d), but not in adult rabbit blood cells (Fig. 3b). The distribution of cells in Fig. 3d represents differences in the maturation stages of the red blood cell population found in rabbit embryos, with respect to Rhodamine 123 uptake intensity. The larger and more granular cells were also the most intensely fluorescent (data not shown). These quantitative analyses confirm the qualitative data obtained by epifluorescence microscopy.

3.5 Ultrastructural immunolabelling analysis

Immunolabelling of immature rabbit embryo erythroid cells was detected diffusely throughout the cytoplasm, in an increasing pattern according to the cell maturation stage. Scarce intranuclear haemoglobin labelling was observed in the euchromatin region(Fig. 4). No reaction was detected in cytoplasmic structures such as the Golgi complex, ribosome, polysome, endoplasmic reticulum, ferruginous inclusions, mitochondria, or mitochondrion-like organelles(Fig. 5–8a, b). No specific labelling was seen in the negative controls, but the positive control showed intense Protein A-gold particle labelling (data not shown).

Figs 7–8

Immature erythrocytes of rabbit embryo immunolabelled with anti-human haemoglobin antibody revealed by protein A-colloidal gold particle complexes. Labelling can be observed throughout the cytoplasm of an erythroblast (see Fig. 4), but is absent in polysomes and mitochondria of a reticulocyte (Fig. 7), mitochondria with transverse cristae (Fig. 8a), and mitochondria with longitudinal cristae (Fig. 8b). Nucleus, N; mitochondria, m; polysomes, arrowheads; iron-containing vesicles, arrows; endoplasmic reticulum, er. Bars=0.5 μm (Fig. 7); Bars=0.25 μm (Figs 8a, 8b).

4 Discussion

In avian and submammalian systems, erythropoiesis is intravascular and developing red cells adhere to the luminal surface of the endothelium of venous sinuses (Tavassoli, 1988). Upon maturation, the cells lose their adhesive capacity and are released into the circulation in their nucleated form. Thus, the circulating red cells of birds and submammalian species are nucleated. Conversely, in mammals, erythropoiesis is extravascular and if a red cell begins movement into the circulation before enucleating, its nucleus is retained in the extravascular space, while the remainder of its cell body passes through the endothelium and is released as a reticulocyte (Wilson and Tavassoli, 1994). In both animal systems, erythropoietic studies using blood cells from the peripheral circulation is possible only through the induction of physiological stress, providing the release of a large number of immature cells into the peripheral blood. The most common procedure is the induction of haemolytic anaemia by phenylhydrazine hydrochloride (Cianciarullo and Meirelles, 1994; Cianciarullo et al., 1999, 2000a,b,c); and saponine is also used in some cases (Spadacci Morena et al., 1989).

We compared the experimental model of erythropoiesis by inducing agents with the peripheral blood from rabbit embryos, where 90–100% of immature erythroid cells can be obtained without drug interference or any physiological stress. For ethical purposes, we sacrificed only one adult rabbit on the 18th day of gestation, considered sufficient to reach the goals of this study. We mapped the intracellular distribution of haemoglobin in the adult rabbit and maturing embryonic erythroid cells, using a polyclonal anti-normal human Hb antibody previously assayed for interspecies cross-reactivity. The high degree of antigen-antibody reaction with heterologous Hb that we have observed can be related to the structural similarity between rabbit Hb and normal human Hb proteins. The intracellular level of Hb increased with erythrocyte maturation, detected by the increasing intensity of protein A-gold immunolabelling. The intranuclear Hbdetected mainly at the euchromatin regions seems to result from it being enclosed into newly forming nuclei during each mitotic division, since the large size of Hb molecules would hinder their crossing through the pores of the nuclear membrane (Paine and Horowitz, 1980).

The fact that there was no labelling inside intact mitochondria, Golgi apparatus, endoplasmic reticulum, or inclusions of ferruginous material confirms our hypothesis that the final integration of haemoglobin molecules occurs directly in the cytoplasm and not inside mitochondria or mitochondrion-like organelles, as proposed by other authors.

The results of this study are similar to those using phenylhydrazine hydrochloride or saponine as indirect erythropoiesis-inducing agents. However, with our method, there is no drug interference, and the physiological stress caused by the treatment had no influence on results obtained in our previous studies, with respect to erythropoiesis and haemoglobin biosynthesis.

Our immunocytochemical results obtained with rabbit embryonic erythroid cells do not support the hypothesis that mitochondria or mitochondrion-like organelles are the site of integration of haeme and the four globin chains in the final synthesis of haemoglobin molecules. These results confirm those obtained in amphibians and humans.


The authors are grateful to Fundação Butantan for financial support of this work.


Brecher G. New methylene blue as a reticulocytes stain. Am J Clin Pathol 1949:19:895-6

Chen LB. Fluorescent labelling of mitochondria. Methods Cell Biol 1990:29:103-23

Cianciarullo AM, Beçak, W, Soares, MJ. Immunocytochemical mapping of the hemoglobin biosynthesis site in amphibian erythroid cells. Tissue Cell 1999:31:342-8
Crossref   Medline   

Cianciarullo AM, Bertho, AL, Meirelles, MNL. Mitochondrial kinetics during amphibian erythropoiesis related to haeme synthesis. Cell Biol Int 2000:24:183-92
Crossref   Medline   

Cianciarullo AM, Meirelles, MNL. Comparative study of immature erythroid cells of the diploid Bufo ictericus and the tetraploid Odontophrynus americanus (Amphibia, Anura): ultrastructural cytochemical detection of nucleic acids and polysaccharides, and mapping of the element phosphorus. Cell Tissue Res 1994:278:187-95
Crossref   Medline   

Cianciarullo AM, Naoum, PC, Bertho, AL, Kobashi, LS, Beçak, W, Soares, MJ. Aspects of gene regulation in the diploid and tetraploid Odontophrynus americanus (Amphibia, Anura, Leptodactylidae). Genet Mol Biol 2000:23:357-64

Cianciarullo AM, Soares, MJ, Beçak, W. Erythropoiesis in the diploid and tetraploid Odontophrynus americanus: an evolutionary approach in these cryptic species (Amphibia, Anura, Leptodactylidae). Comp Haemat Int 2000:10:19-29

Darzynkiewicz Z, Traganos, F, Staiano-Coico, L, Kapuscinski, J, Melamed, MR. Interactions of rhodamine 123 with living cells studied by flow cytometry. Cancer Res 1982:42:799-806

Dieterlen-Lièvre F, Godin, IE, Garcia-Porrero, JA, Marcos, MAR. Initiation of hemopoiesis in the mouse embryo. Ann NY Acad Sci 1994:718:140-6

Johnson LV, Walsh, ML, Chen, LB. Localization of mitochondria in living cells with rhodamine 123. Proc Natl Acad Sci U S A 1980:77:990-4
Crossref   Medline   

Johnson GD, Davidson, RS, McNamee, KC, Russell, G, Goodwin, D, Holborow, EJ. Fading of immunofluorescence during microscopy and its remedy. J Immunol Methods 1982:55:231-42
Crossref   Medline   

Paine PL, Horowitz, SB. The movement of material between the nucleus and cytoplasm. Cell biology a comprehensive treatise 1980:299-328

Rosenfeld G. Corante pancrômico para hematologia e citologia clínica. Nova combinação do May-Grünwald e do Giemsa num só corante de emprego rápido. Mem Inst Butantan 1947:20:329-34

Shapiro HM. . Practical flow cytometry 1985:1-158

Spadacci Morena DD, Morena, P, Cianciarullo, AM, Brunner Jr, A. Hemoglobin biosynthesis in bone marrow and peripheral blood erythroid cells of snake Waglerophis merremii (Reptilia, Ophidia, Colubridae). Mem Inst Butantan 1989:51:133-9

Tavassoli M. Structure of erythroblastic islands in the bone marrow. Regulation of erythropoiesis 1988:3-13

Watson JV. . Introduction to flow cytometry 1991:1-198

Wilson JG, Tavassoli, M. Microenvironmental factors involved in the establishment of erythropoiesis in bone marrow. Ann NY Acad Sci 1994:718:271-84

Received 24 February 2003/11 April 2003; accepted 2 June 2003


ISSN Print: 1065-6995
ISSN Electronic: 1095-8355
Published by Portland Press Limited on behalf of the International Federation for Cell Biology (IFCB)