Studies on the sequestration of chemically and enzymatically modified erythrocytes

Methodologies and tools to shed light on erythrophagocytosis

Red blood cells (RBC) are the most abundant circulating cell of the human body. RBC are constantly exposed to multiple stresses in the circulation, leading to molecular and structural impairments and death. The physiological process of RBC senescence or ageing is referred to as eryptosis. At the end of their lifespan, aged RBC are recognized and removed from the blood by professional phagocytes via a phenomenon called erythrophagocytosis (EP); the phagocytosis of RBC. Some genetic and acquired diseases can influence eryptosis, thereby affecting RBC lifespan and leading to hemolytic anemia. In some diseases, such as diabetes and atherosclerosis, eryptosis and EP can participate in disease progression with both professional and non-professional phagocytes. Therefore, investigating the process of EP in vivo and in vitro, as well as in different cell types, will not only contribute to the understanding of the physiological steps of EP, but also to the deciphering of the specific mechanisms involving RBC and EP that underlie certain pathologies. In this review, the process of EP is introduced and the different methods for studying EP are discussed together with examples of the experimental procedures and materials required.

Cellular and molecular mechanisms of senescent erythrocyte phagocytosis by macrophages. A review

Human red blood cells (RBCs) have a life-span of 120 days in circulation, after which they are phagocytized by resident macrophages. Extensive studies have been undertaken by many investigators in order to elucidate the cellular and molecular mechanisms of the erythrophagocytosis. The critical questions addressed by physiologists, clinicians and biochemists are: 'which of the many traumatic blemishes that appear on the erythrocyte surface as it winds its way through the circulation is the primary signal for clearance of the effete RBC from the circulation?', or 'What is the critical signal that it, and it alone, will activate the resident macrophage to adhere to and engulf it?'. Numerous, and often conflicting, hypotheses have been proposed. Each investigator focusing on but one of the many modifications that afflict the cell surface of the ageing erythrocyte, viz changes in either or both the carbohydrate or peptidic moieties of glycoproteins; abolishment of the pre-existing asymmetry in the lipid bilayer with the exposure of phosphatidylserine residues; or alterations in spectrin, to mention but a few. Many of these investigators also have invoked an intermediary role for auto-immune antibodies that recognise the change(s) on the erythrocyte surface and thereby serve as opsonins as a prelude to the erythrophagocytosis. The objective of the present review is to evaluate the data in support of the various hypotheses, and to submit some of our own recent observations involving the use of flow cytometric procedures that: i) provide evidence that the cell surface sialic acid serves as a determinant of the life-span; ii) characterise the senescent erythrocyte population that is specifically captured and phagocytized by macrophages (utilising the rapid and sensitive procedure we developed for quantification of in vitro erythrophagocytosis); and finally iii) provide evidence for the existence of an alternative pathway that is independent of immunoglobulins.

Molecular mechanisms of erythrophagocytosis. Characterization of the senescent erythrocytes that are phagocytized by macrophages

We have recently developed a flow cytometric assay for the quantitation of erythrophagocytosis, using PKH 26-labeled erythrocytes as the target cells. Using this assay we have shown that there is extensive phagocytosis of desialylated erythrocytes. Furthermore, we have demonstrated that it is the densest population of erythrocytes obtained on a self-forming gradient of Percoll that shows the greatest susceptibility to phagocytosis. We designate this population of erythrocytes as fraction X; it is even denser than the fraction 5 found previously. This population of erythrocytes corresponds to zone X previously seen in the dot-plot of the flow cytometric analyses of human erythrocytes. Further scrutiny of this fraction indicates that a) it shows the greatest reactivity with annexin V, which is specific for the detection of phosphatidylserine (PS) exposed on the outer leaflet of the erythrocyte membrane, b) it is the most susceptible to erythrophagocytosis by resident murine peritoneal macrophages, and c) this erythrophagocytosis of PKH 26-labeled erythrocytes can be inhibited by annexin V and by liposomes containing PS. Scanning electron microscopy of fraction X shows two populations of erythrocytes: (A) spheroechinocytes with filipodes and (B) echinocytes without filipods. After a 2-h period of phagocytosis, the cells remaining in fraction X show a decrease in population A, commensurate with a decrease in reactivity with FITC-labeled annexin V from 65.5 to 24%.

Flow cytofluorimetric analysis of young and senescent human erythrocytes probed with lectins. Evidence that sialic acids control their life span

Comparing the properties of 'young' and senescent ('aged') O+ erythrocytes isolated by applying ultracentrifugation in a self-forming Percoll gradient, we demonstrate that the sialic acids of membrane glycoconjugates control the life span of erythrocytes and that the desialylation of glycans is responsible for the clearance of the aged erythrocytes. This capture is mediated by a beta-galactolectin present in the membrane of macrophages. The evidence supporting these conclusions is as follows: (1) Analysis by flow cytofluorimetry of the binding of fluorescein isothiocyanate labelled lectins specific for sialic acids shows that the aged erythrocytes bind less WGA, LPA, SNA and MAA than young erythrocytes. The binding of DSA and LCA is not modified. On the contrary, the number of binding sites of UEA-I specific for O antigen and of AAA decreases significantly. PNA and GNA do not bind to erythrocytes. (2) RCA120 as well as Erythrina cristagalli and Erythrina corallodendron lectins specific for terminal beta-galactose residues lead to unexpected and unexplained results with a decrease in the number of lectin binding sites associated with increasing desialylation. (3) The glycoconjugates from the old erythrocytes incorporate more sialic acid than the young cells. This observation results from the determination of the rate of transfer by alpha-2,6-sialyltransferase of fluorescent or radioactive N-acetylneuraminic acid, using as donors CMP-9-fluoresceinyl-NeuAc and CMP-[14C]-NeuAc, respectively. (4) Microscopy shows that the old erythrocytes are captured preferentially by the macrophages relative to the young ones. Fixation of erythrocytes by the macrophage membrane is inhibited by lactose, thus demonstrating the involvement of a terminal beta-galactose specific macrophage lectin. (5) Comparative study of the binding of WGA, LPA, SNA and MAA to the aged erythrocytes and to the in vitro enzymatically desialylated erythrocytes shows that the desialylation rate of aged cells is low but sufficient to lead to their capture by the macrophages.

The effect of glycosidases on the survival of rat erythrocytes in circulation

Enzymatic removal of sialyl groups from mammalian erythrocytes resulted in their rapid sequestration from circulation subsequent to autologous transfusion. It has been demonstrated by many investigators that the terminal beta-D-galactosyl group, exposed on red blood cell by in vitro desialosylation, is recognized by an autoimmune anti-galactosyl IgG and/or by a lectin-like receptor on monocytes and macrophages. It is demonstrated herein that the disaccharide structure beta-D-Galp-(1----3)-D-GalpNAc (a) is masked in normal rat RBC, but exposed in asialo-RBC; (b) could be detected with fluorescently-labeled peanut agglutinin; (c) could be released from the asialo-RBC with an endo-N-acetyl-alpha-D-galactosaminidase; and (d) upon its removal by treatment with the endo-N-acetyl-alpha-D-galactosaminidase, enhances the survival of the asialo-RBC in circulation.

Effect of age on some properties of mice erythrocytes

The hematological parameters of young (2-month-old) and old (2-year-old) mice were compared. No differences could be detected with the exception of an increased percentage of reticulocytes in the old animals suggesting that anemia in senescent mice does not occur. Red blood cell mean half-life in old mice was 8 +/- 0.8 days compared to 12 +/- 1 days in young mice. This reduced survival of red blood cell is not due to a different rate of cell phagocytosis in the reticulohistiocytic system of young and old animals since erythrocytes from young mice have the same mean half-life when injected both in young and old animals and vice versa. Thus, the old mice have a reduced red cell life-span but the same hematocrit of the young, suggesting that old animals possess a chronologically younger population of erythrocytes than do young animals. This has been confirmed by measuring the specific activities of some red blood cell age-dependent enzymes (hexokinase, glucose-6-phosphate dehydrogenase, pyruvate kinase) that were found to be higher in the older animals, and by the separation of erythrocytes into different density (age) groups by Percoll/albumin density gradient centrifugation. However, the erythrocytes osmotic fragility, and the cellular contents of adenine and pyridine nucleotides, as well as the content of 2,3-diphosphoglycerate and reduced glutathione, show that circulating erythrocytes in old animals constitute an heterogeneous cell population whose properties cannot be explained on the basis of a chronologically younger erythrocyte population. Furthermore, evaluation of cell components in hemopoietic tissues have shown an increased porportion of erythroid precursor cells in old animals confirming that old mice compensate for reduced red cell survival with an increased erythropoiesis.

Measurement of phagocytosis utilizing [14C]cyanate-labelled human red cells and monocytes

Human red cells were labelled with [14C]cyanate, non-oxidant, permeant reagent that binds irreversibly to amino groups in proteins. Cyanate did not modify GSH levels, nor glucose-6-phosphate dehydrogenase (G6PD) activity when added at the labelling concentration i.e. c. 0.5 mM. Phagocytes were human monocytes, isolated and plated on 16 mm diameter plastic wells. Each well was plated with 40-70 X 10(3) cells. Monocytes were quantified by DNA assay with the DNA intercalating fluorescent compound Hoechst 33,258. The basal phagocytic rate of normal red cells by monocytes was 0.34 +/- 0.21 red cell per monocyte. Treatment of normal red cells with 20 microM diamide, or 50-100 microM chromate enhanced this rate 10-15-fold. 10 microM diamide or chromate were sufficient to stimulate phagocytosis of G6PD-deficient (Mediterranean variant) red cells.

Survival of murine carrier erythrocytes injected via peritoneum

Carrier-red blood cells (C-RBC) can be administered by intraperitoneal injection (IP) in mice, and C-RBC survive in circulation as well as cells administered by intravenous injection (IV). There was no difference in the 24 hr post-injection organ distribution of normal RBC and C-RBC, an indication of the normalcy of C-RBC. Approximately 25% of the C-RBC remain in circulation 14 days post-injection indicating a 20-day half-life for C-RBC in mice. This alternate injection method for RBC also allows for extravascular targeting of RBC to peritoneal macrophages.

Use of 75Se-labeled methionine to study the sequestration of senescent red blood cells

Labeling red blood cells with Na251CrO4 enabled us to study certain aspects of red cell survival and sequestration from the circulation. As a random labeling procedure, however, the 51Cr method has certain limitations. Therefore, we developed a cohort labeling method using 75Se-methionine as a two-rat procedure. This gives us a clear pulse-labeled population of rat red cells to study the dynamics of sequestration. With this labeling procedure, it was possible to demonstrate that 1) there is an increase in the density of red cells with age, 2) a significant sequestration of red cells from the circulation is apparent at the end of 48 days and essentially is complete at the end of 60 days, 3) there is a corresponding uptake of senescent red cells in the spleen, which peaks at 55 days, and 4) the 60-day end point is sharper and is more definitive when the "specific activity" (cpm per red blood cell) of the labeled red cells in the spleen is compared to that of the red cells still in the circulation. Asialo red cells, obtained by removal of sialic acid with sialidase, frequently have been used as a model for the study of sequestration of senescent red cells. With the technique herein described, it was possible to show that while asialo red cells will inhibit the uptake of labeled asialo red cells, they have no effect on the sequestration of senescent red cells. Presumably, different sites and mechanisms of sequestration are involved.