QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.090407v1 315/3/980    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Bowman, Z. S. Articles by McMillan, D. C. Search for Related Content PubMed PubMed Citation Articles by Bowman, Z. S. Articles by McMillan, D. C.

TOXICOLOGY

Primaquine-Induced Hemolytic Anemia: Role of Splenic Macrophages in the Fate of 5-Hydroxyprimaquine-Treated Rat Erythrocytes

Department of Cell and Molecular Pharmacology, Medical University of South Carolina, Charleston, South Carolina (Z.S.B., D.J.J., D.C.M.)

Received for publication June 2, 2005
Accepted August 10, 2005.

	Abstract

Top Abstract Materials and Methods Results Discussion References

 
Primaquine-induced hemolytic anemia is known to result from premature sequestration of damaged (but intact) erythrocytes by the spleen. We have shown previously that a phenolic metabolite, 5-hydroxyprimaquine (5-HPQ), is a direct-acting hemolytic agent in rats, suggesting that 5-HPQ is a mediator of the hemolytic response to primaquine. To investigate the fate of erythrocytes in vivo after in vitro exposure to 5-HPQ, rat ⁵¹Cr-labeled erythrocytes were incubated with hemolytic concentrations of 5-HPQ and then readministered intravenously to rats. The time course of loss of radioactivity from blood and uptake into the spleen and liver was measured. In rats given 5-HPQ-treated erythrocytes, an increased rate of removal of radioactivity from the circulation was observed as compared with the vehicle control. The loss of blood radioactivity was accompanied by a corresponding increase in radioactivity appearing in the spleen but not in the liver. When rats were pretreated with clodronate-loaded liposomes to deplete splenic macrophages, there was a decreased rate of removal of radioactivity from the circulation and a markedly diminished uptake into the spleen. A role for phagocytic removal of 5-HPQ-treated red cells was confirmed in vitro using the J774A.1 macrophage cell line. Furthermore, depletion of red cell GSH with diethyl maleate significantly enhanced in vitro phagocytosis of 5-HPQ-treated red cells. The data indicate that splenic macrophages are responsible for removing 5-HPQ-treated red cells and support the postulate that this metabolite is a contributor to the hemolytic anemia induced after administration of the parent compound.

One of the ring-hydroxylated metabolites, 5-hydroxyprimaquine (5-HPQ), has long been postulated to be capable of mediating primaquine hemotoxicity in vivo. In support of this postulate, we recently demonstrated that 5-HPQ is directly hemolytic in rat red cells; i.e., when rat ⁵¹Cr-labeled-red cells are incubated with 5-HPQ in vitro for 2 h at 37°C and then given intravenously to isologous rats, the tagged red cells are removed rapidly from the circulation (Bowman et al., 2004). Loss of erythrocyte viability in vivo was associated with oxidative stress responses in vitro, including methemoglobin formation, depletion of erythrocytic GSH, and formation of glutathione protein-mixed disulfides. Furthermore, when erythrocytic GSH was partially depleted (by 95%) by titration with diethyl maleate (DEM) to mimic human G6PD deficiency, the hemolytic activity of 5-HPQ was enhanced in the rat erythrocytes by more than 5-fold.

The potency and pro-oxidant nature of the hemolytic activity of 5-HPQ in rats supports its role as a major contributor to primaquine-induced hemolytic anemia; however, little is known about the fate of these erythrocytes in vivo; i.e., do they undergo intravascular lysis and uptake of the cell fragments into the liver, or do the cells remain intact and undergo selective uptake into the spleen? In addition, the role of macrophages in the process of removal of damaged erythrocytes has been presumed, yet little data exist that demonstrate a requirement for phagocytic activity in order to provoke a hemolytic response.

The present studies were undertaken to determine the fate of 5-HPQ-treated erythrocytes in rats and to clarify the role of macrophages in this process. Experimentally, we have compared uptake of 5-HPQ-treated erythrocytes into the spleens and livers of normal rats with that of rats pretreated with clodronate-loaded liposomes. This pretreatment is known to deplete splenic macrophages and hepatic Kupffer cells (Van Rooijen and Sanders, 1994). We report that 5-HPQ-treated ⁵¹Cr-labeled erythrocytes are rapidly taken up into the spleen in preference to the liver when readministered intravenously to untreated rats. In clodronate-loaded liposome-pretreated rats, the survival of the 5-HPQ-damaged red cells in the circulation was markedly prolonged, coinciding with a virtual abolition of splenic uptake and a much delayed hepatic uptake. Parallel studies using J774A.1 macrophage cultures demonstrated the uptake of 5-HPQ-damaged, but visually intact, rat red cells into "jackpot" cells in the culture (Cocco and Ucker, 2001). The uptake was concentration-dependent in regard to 5-HPQ and was enhanced markedly by prior depletion of erythrocytic GSH by DEM titration. The data indicate that the premature removal of 5-HPQ-damaged red cells in the rat results from their selective recognition and uptake by splenic macrophages. Of importance, the reproduction of the in vivo macrophage-dependent toxicity in vitro indicates that the J774A.1 macrophage culture can be used as a model to look for external cell-surface markers that initiate splenic sequestration of erythrocytes inflicted with hemolytic injury.

	Materials and Methods

Top Abstract Materials and Methods Results Discussion References

 
Chemicals and Materials. 5-HPQ was synthesized by HBr-catalyzed hydrolysis of 5-methoxyprimaquine as described previously (Bowman et al., 2004). Na ⁵¹₂CrO₄ in sterile saline (1 mCi/ml, pH 8) was obtained from PerkinElmer Life and Analytical Sciences (Boston, MA). Phosphatidylcholine (1,2-diacyl-sn-glycero-3-phosphocholine), cholesterol, and clodronate (dichloromethylene bisphosphonate) were purchased from Sigma-Aldrich (St. Louis, MO). J774A.1 cells are a monocyte-derived mouse macrophage cell line that was obtained from the American Type Culture Collection (Manassas, VA). All other reagents were of the best grade commercially available.

Animals and Erythrocyte Incubation Conditions. Male Sprague-Dawley rats (75–100 g) were purchased from Harlan (Indianapolis, IN) and maintained on food and water ad libitum. Animals were acclimated for 1 week to a 12-h light/dark cycle before their use. Blood from the descending aorta of anesthetized rats was collected into heparinized tubes and washed three times with HBSS (with Ca²⁺ and Mg²⁺) to remove the plasma and buffy coat. The packed red cells were mixed with saline containing radioactive sodium chromate (0.1 mCi/ml packed cells) and allowed to incubate for 30 min at 37°C to label the cells as described previously (Harrison and Jollow, 1986). The radiotagged erythrocytes were then washed and resuspended to a 40% hematocrit and used the same day they were collected. Stock solutions of 5-HPQ in argon-purged water were prepared to deliver the appropriate concentrations of 5-HPQ in 10 µl to erythrocyte suspensions (1–3 ml).

In some experiments, DEM was used to deplete GSH (>95%) in erythrocyte suspensions as described previously (Bolchoz et al., 2002). In brief, DEM (750 µM) dissolved in acetone was added to packed red cells. After 15 min of incubation at 37°C, the content of erythrocytic GSH was determined from an aliquot by high-pressure liquid chromatography with electrochemical detection using a standard curve.

Preparation and Administration of Clodronate-Loaded Liposomes. Clodronate-loaded liposomes were prepared as described previously (Van Rooijen and Sanders, 1994). The liposomes were washed once and centrifuged to remove free clodronate, and the liposomes were administered intravenously (0.1 ml/10 g b.wt.) via the tail vein to rats 24 h before administration of the radiolabeled erythrocytes. Intravenous administration of saline served as a control to ensure that these animals had normal, nonblocked, nonsuppressed, and nonactivated macrophages.

Measurement of Erythrocyte Disposition in Vivo. The survival and fate of erythrocytes was determined in saline-treated or clodronate-loaded liposome-treated rats after in vitro exposure to 5-HPQ as described previously (Bowman et al., 2004). In brief, the ⁵¹Cr-labeled erythrocytes were incubated with 5-HPQ (40 µM) for 2 h at 37°C. After incubation, the erythrocytes were washed once and resuspended in HBSS (40% hematocrit). Aliquots (0.5 ml) of the labeled cells were administered intravenously to isologous rats. T₀ blood samples (75 µl) were taken from the orbital sinus 1 h after administration of labeled red cells. Additional blood samples were taken at 48-h intervals for 7 days. At the end of the experiment, the samples were counted in a well-type gamma counter and the data are expressed as a percentage of the T₀ blood sample.

To determine the disposition of the ⁵¹Cr label, livers and spleens were excised at designated intervals from groups of rats that had received the labeled erythrocytes as described above. The tissues were weighed and then counted concurrently in a well-type gamma counter, and the values were corrected for ⁵¹Cr activity of residual erythrocytes within the tissue vasculature. In experiments to determine the disposition of broken red cells, untreated ⁵¹Cr-labeled erythrocytes (40% hematocrit) were subjected to three freeze-thaw cycles using a dry ice-acetone bath to induce complete cell lysis, as judged by the observed lack of red cells that pellet when the sample is centrifuged. This procedure allowed us to lyse the erythrocytes while maintaining the same hematocrit (40%) as the unbroken cells. Aliquots of the (unwashed) broken cell preparation were then administered intravenously to rats in the same manner as described above for the intact erythrocytes.

Measurement of in Vitro Erythrophagocytosis. J774A.1 macrophages were plated in 12-well culture plates at 2 x 10⁵ cells/well to prepare semiconfluent monolayers. Twenty-four hours after plating the macrophages, aliquots (50 µl) of control and 5-HPQ-treated ⁵¹Cr-labeled packed erythrocytes (treated as described above) were added to the macrophage cultures and allowed to incubate for up to 24 h at 37°C. After incubation, noningested erythrocytes were removed by washing once with NH₄Cl (0.8% w/v) followed by washing three times with HBSS. The adherent macrophages were released from the plates with 0.5 ml of NaOH (0.5 N), and the extent of phagocytosis was assessed by counting the radioactivity in macrophages using a well-type gamma counter.

	Results

Top Abstract Materials and Methods Results Discussion References

 
Disposition of ⁵¹Cr-Labeled Erythrocytes after in Vitro Exposure to 5-HPQ. Although the direct hemolytic activity of 5-HPQ in rats has been established (Bowman et al., 2004), the fate of 5-HPQ-damaged red cells in vivo (i.e., intravascular lysis versus splenic sequestration) is unknown. To determine the anatomic site(s) of sequestration of 5-HPQ-damaged red cells and the role of macrophages in this process, rat ⁵¹Cr-labeled erythrocytes were incubated with 5-HPQ (40 µM) for 2 h at 37°C. After the incubation, the erythrocytes were washed and returned to the circulation of saline-pretreated or clodronate-loaded liposome-pretreated rats. At designated intervals, a blood sample was taken from each rat followed by excision of the spleen and liver for determination of uptake of the radiolabel.

Upon readministration to saline-pretreated rats, 5-HPQ-treated erythrocytes showed an initial rapid disappearance from the circulation (Fig. 1A). This loss of blood radioactivity coincided with a marked increase in splenic radioactivity (Fig. 1B), which was not observed in rats that received saline-treated control cells. The splenic radioactivity reached an apparent steady state by day 1 corresponding to processing of the tagged red cells and subsequent release of the radioactivity for urinary excretion (data not shown). The levels of radioactivity in the livers of rats given 5-HPQ-treated erythrocytes were slightly higher than those in the controls (Fig. 1B), and these levels remained constant over the course of the experiment. On a whole organ basis, splenic radioactivity in saline-pretreated rats was four to five times that of the liver and 50 to 60 times higher when expressed per gram of tissue.

View larger version (13K): [in this window] [in a new window]   Fig. 1. Disposition of the 51Cr label in blood (A) and spleens and livers (B) of rats infused with 51Cr-labeled erythrocytes exposed in vitro to the vehicle () or 40 µM 5-HPQ () for 2 h at 37°C. Data points are means ± S.D. (n = 3).

In contrast to the rapid disappearance of blood radioactivity in saline-pretreated rats, the rate of removal of 5-HPQ-treated erythrocytes in rats pretreated with clodronate-loaded liposomes was markedly reduced (Fig. 2A) and was not accompanied by significant uptake of radioactivity into the spleen (Fig. 2B). As shown in Fig. 3A, the decreased rate of removal of blood radioactivity was associated with a time-dependent decrease in the spleen/body weight ratio (40% decrease at day 7), which presumably reflects extensive depletion of macrophages from this tissue. Administration of clodronate-loaded liposomes had no effect on liver/body weight ratio (Fig. 3B). The amount of ⁵¹Cr detected in the livers of rats pretreated with saline or clodronate-loaded liposomes was relatively low, suggesting that, under normal conditions, hepatic Kupffer cells do not have an active role in removing 5-HPQ-damaged erythrocytes. Of note, the clodronate-loaded liposome-pretreated rats showed a significant increase in liver radioactivity on day 7 (Fig. 2B), which suggests that Kupffer cells have begun to repopulate the liver and remove damaged erythrocytes in the absence of a functional splenic uptake mechanism. This observation is in agreement with previous studies that showed that macrophages repopulate the liver after liposome treatment more rapidly than they repopulate the spleen (Van Rooijen et al., 1990). Although uptake into the liver may be considered to reflect intravascular lysis of erythrocytes, it has been shown that Kupffer cells will assume the role of removing senescent and damaged erythrocytes in splenectomized animals (Ganick et al., 1977).

View larger version (14K): [in this window] [in a new window]   Fig. 2. Effect of pretreatment of rats with saline (closed symbols) or clodronate-loaded liposomes (open symbols) on the disposition of the 51Cr label in blood (A) and spleens and livers (B) of rats infused with 51Cr-labeled erythrocytes exposed in vitro to the vehicle (circles) or 40 µM 5-HPQ (squares) for 2 h at 37°C. For the purpose of comparison, the saline-treated rat data from Fig. 1A are displayed in this figure. Data points are means ± S.D. (n = 3).

View larger version (13K): [in this window] [in a new window]   Fig. 3. Effect of pretreatment of rats with clodronate-loaded liposomes on spleen (A) and liver (B) weight/body weight ratios. Rats were given saline () or clodronate-loaded liposomes () intravenously. At the indicated time points, the spleens and livers were excised and weighed. Data points are means ± S.D. *, significantly different from phosphate-buffered saline-pretreated rats (p < 0.05).

Disposition of ⁵¹Cr-Labeled Broken Cells in Vivo. To determine whether intravascular lysis could have contributed to the hemolytic response to 5-HPQ, the disposition of untreated ⁵¹Cr-labeled broken red cells was examined. Broken erythrocytes were prepared by subjecting ⁵¹Cr-tagged cells to multiple freeze-thaw cycles. As shown in Fig. 4A, rapid and complete disappearance of the broken cell ⁵¹Cr-label from the blood occurred within 2 days and the loss of blood radioactivity was accompanied by a modest increase in liver radioactivity that probably reflects uptake of red cell fragments by Kupffer cells (Fig. 4B). In contrast to 5-HPQ-treated erythrocytes, only a minor amount of radiolabel from the broken cells was taken up into the spleen. Approximately 30% of the administered radioactivity appeared in the urine during the 7-day observation period; the remainder was unaccounted for and was presumed to be occluded within the microcirculation.

View larger version (12K): [in this window] [in a new window]   Fig. 4. Disposition of the 51Cr label in blood of rats infused with 51Cr-labeled normal () and broken () erythrocytes (A) and in the spleens (), livers (), and urine () of rats infused with 51Cr-labeled broken erythrocytes (B). Data points are means ± S.D. (n = 3).

Role of Macrophages on the Hemotoxicity of 5-HPQ in Vitro. To determine whether splenic macrophage-dependent uptake of 5-HPQ-treated erythrocytes could be reproduced in vitro, ingestion of ⁵¹Cr-labeled rat erythrocytes was assessed using J774A.1 cells, a mouse monocyte-derived macrophage cell line that has been shown previously to bind and ingest both necrotic and apoptotic cells (Cocco and Ucker, 2001). Radiolabeled erythrocytes were incubated with hemolytic concentrations of 5-HPQ for 2 h at 37°C. The erythrocytes were then washed and incubated with macrophages for various time periods. At the end of each incubation period, the macrophages were rinsed with HBSS to remove nonadherent erythrocytes and then treated with ammonium chloride to remove adherent (but noningested) erythrocytes. As shown in Fig. 5, there was a concentration-dependent increase in radioactivity associated with the macrophages after 18 h of incubation. The level of radioactivity associated with the macrophages at earlier time points (6 and 12 h) was not significantly different from the controls (data not shown). Of importance, the concentration dependence for this response coincided with the concentration dependence for hemolytic activity in vivo.

View larger version (18K): [in this window] [in a new window]   Fig. 5. Effect of 5-HPQ on the uptake of rat 51Cr-labeled red cells by J774A.1 macrophages. 51Cr-labeled erythrocytes were incubated with the indicated concentrations of 5-HPQ for 2 h at 37°C. The erythrocytes were then washed and added to J774A.1 macrophage cultures for 18 h. The cultures were washed with saline and ammonium chloride to remove unbound and bound (but noningested) erythrocytes, respectively. The macrophages were then scraped from the plates, and the radioactivity was counted. Data points are means ± S.D. (n = 3). *, significantly different from controls (p < 0.01).

View larger version (145K): [in this window] [in a new window]   Fig. 6. Light micrographs of J774A.1 macrophage cultures after incubation with control (A) and 5-HPQ-treated (B) rat erythrocytes. Rat erythrocytes were treated with the vehicle or 5-HPQ (50 µM) for 2 h at 37°C before being incubated with the macrophages for 18 h. The arrows indicate the presence of intact 5-HPQ-treated erythrocytes within jackpot cells. Magnification 63x.

Of note, a subset of the J774A.1 macrophages appeared to account for a majority of the uptake of the damaged red cells. This phenomenon has been observed previously and suggests that, within the clonal population, some macrophages designated jackpot cells are more competent at erythrocyte recognition than are others. The mean values for macrophage radioactivity incorporate macrophages that have ingested no erythrocytes and others that contain several ingested erythrocytes.

Effect of GSH Depletion on Phagocytic Uptake. G6PD-deficient individuals have long been known to have an enhanced susceptibility to primaquine-induced hemolytic anemia. We have shown previously that this enhanced susceptibility can be reproduced by prior depletion of red cell GSH (Bolchoz et al., 2002; Bowman et al., 2004). In the case of 5-HPQ, the hemolytic susceptibility of GSH-depleted rat erythrocytes is approximately 5-fold greater than GSH-normal red cells (Bowman et al., 2004). To determine the effect of GSH depletion on phagocytic removal in vitro, ⁵¹Cr-labeled erythrocytes were titrated with DEM to deplete GSH (ca. 95%) before exposure to a low concentration of 5-HPQ. After 2 h of incubation at 37°C, control and GSH-depleted erythrocytes were incubated with J774A.1 macrophages for 18 h. As shown in Fig. 7, treatment of GSH-normal erythrocytes with 25 µM 5-HPQ did not induce a significant increase in phagocytic uptake by macrophages as compared with the control. However, the GSH-depleted erythrocytes showed a significant increase in phagocytic uptake as compared with the control. This observation is in agreement with the previous in vitro exposure/in vivo erythrocyte survival study that showed that 25 µM 5-HPQ was subhemolytic in GSH-normal erythrocytes but hemolytic in GSH-depleted erythrocytes (Bowman et al., 2004).

View larger version (29K): [in this window] [in a new window]   Fig. 7. Effect of depletion of erythrocytic GSH on uptake of 5-HPQ-treated erythrocytes by J774A.1 macrophages. 51Cr-Labeled erythrocytes were titrated with DEM to deplete GSH by 95%. GSH-normal and GSH-depleted erythrocytes were then incubated with the vehicle or 5-HPQ (25 µM) for 2 h at 37°C. The erythrocytes were then washed and incubated with J774A.1 macrophages for 18 h. Data points are means ± S.D. (n = 4). *, significantly different from controls (p < 0.05).

	Discussion

Top Abstract Materials and Methods Results Discussion References

On the one hand, oxidative stress may stimulate erythrocyte phagocytosis through a nonspecific mechanism referred to as culling. In this process, oxidative stress is thought to alter the membrane cytoskeleton, leading to a decrease in erythrocyte deformability, thereby restricting progress of erythrocytes through the splenic red pulp and increasing their exposure to resident macrophages (Chadburn, 2000). On the other hand, oxidative stress may induce discrete changes to the external cell surface that stimulate specific phagocytic receptors on macrophages (Beppu et al., 1987; Terpstra and van Berkel, 2000). Although what constitutes a recognition ligand on the erythrocyte surface is not yet known, extensive studies have supported several competing viewpoints, including: 1) unmasking of an antigenic site leading to opsonization by circulating autologous IgG (Kay, 1994); 2) unmasking of carbohydrate epitopes by desialylation of glycoproteins (Kay, 1993); 3) externalization of the membrane phospholipid, phosphatidylserine (Mandal et al., 2002); and 4) loss of self-recognition caused by alterations in the cell-surface expression or lateral distribution of the macrophage inhibitory ligand, CD47 (Oldenborg et al., 2000).

A major difficulty in studies on the removal of normally aged erythrocytes is that senescent cells represent only a small fraction of the total erythrocyte population. Although enrichment techniques exist, workers in this area are faced with discerning the crucial changes on the surface of a small subset of cells against a large background of nonsenescent erythrocytes. In contrast, by incubating normal erythrocytes in vitro with hemolytic agents, such as 5-HPQ, we are able to convert the normal cells to "prematurely aged" erythrocytes, which are subject to rapid removal when returned to the intact animal. Hemolytic damage is concentration-dependent in regard to 5-HPQ and can be exacerbated by prior depletion of GSH as a mimic of G6PD deficiency. This experimental system thus lends itself to the exploration of how ROS generated within erythrocytes provokes changes on the external surface that initiate the removal process. The present experiments were undertaken to answer the following questions: 1) are 5-HPQ-treated erythrocytes selectively sequestered by the spleen after their infusion into rats; 2) is sequestration mediated by splenic macrophages; and 3) can the splenic sequestration be replicated in vitro?

To determine the fate of 5-HPQ-treated erythrocytes in vivo, we compared the extent of uptake of control and broken and 5-HPQ-treated erythrocytes into the spleen and liver, which are the two major organs for erythrocyte clearance. Broken erythrocytes and intact erythrocytes exposed to 40 µM 5-HPQ in vitro were both removed rapidly from the circulation as compared with the control. However, radioactivity from the broken cells was distributed to the liver and excreted rapidly in the urine (Fig. 4B), whereas the radiolabel from 5-HPQ-treated erythrocytes was distributed primarily to the spleen (Fig. 1B). These data indicate that the spleen is the primary anatomic site of sequestration for erythrocytes exposed to 5-HPQ and suggest that intravascular lysis does not contribute significantly to the loss of erythrocytes from the circulation. A similar observation was reported in G6PD-deficient individuals receiving primaquine (Degowin et al., 1966). These workers administered primaquine to G6PD-normal and G6PD-deficient individuals who had previously had their erythrocytes tagged with ⁵¹Cr. They demonstrated that uptake of the radiolabel was associated with both the spleen and the liver; however, the role of the spleen was significantly greater when uptake was expressed relative to organ weight.

Selective depletion of macrophages from specific tissues has been shown to be beneficial for determining their role in defined biological processes (Van Rooijen and Sanders, 1994). Macrophages have long been thought to be responsible for the anemia that is observed after administration of hemolytic agents. If macrophages are the key players and erythrocyte removal is not due simply to nonspecific trapping in the splenic cords, then substances that cause a loss of macrophage viability should have an inhibitory effect on splenic sequestration. To test the dependence of the hemolytic response on the presence of functional macrophages, rats were given clodronate-loaded liposomes 24 h before administration of the erythrocytes. After being ingested, the liposomes are degraded within macrophages, releasing clodronate into the cytosol where it induces macrophage cell death by apoptosis (Van Rooijen and Sanders, 1994). A dose of 0.1 ml of clodronate-liposome suspension per 10 g body weight has been reported to be sufficient for depletion of splenic macrophages and hepatic Kupffer cells, and the decline in spleen/body weight ratio observed over the course of the experiment (Fig. 3) is consistent with selective loss of macrophages from the spleen.

The time necessary to remove 50% ⁵¹Cr-labeled erythrocytes for 5-HPQ-treated erythrocytes in clodronate-loaded liposome-pretreated rats was significantly delayed (by 3 days) in relation to that of the saline-pretreated rats (Fig. 2A). Moreover, the accumulation of radioactivity that was observed in the spleens of saline-pretreated rats was markedly reduced in clodronate liposome-pretreated rats (Fig. 2B). Interestingly, uptake of the label into the livers of clodronate liposome-pretreated animals increased significantly on day 4 (Fig. 2B). This observation is consistent with previous studies that showed that Kupffer cells repopulate the liver more rapidly than macrophages repopulate the spleen after treatment with the liposomes (Van Rooijen and Van Kesteren-Hendrikx, 2003), and it reflects the ability of Kupffer cells to increase their phagocytic activity toward erythrocytes in the absence of a functional splenic uptake mechanism (Ganick et al., 1977).

An in vitro assay for hemolytic anemia has long been desired. Although the in vitro exposure/in vivo survival assay (Fig. 1A) is useful for identifying direct-acting hemolytic agents and for studying the mechanism underlying the hemolytic response, this assay is limited by the need for large numbers of animals, it does not permit erythrocyte-macrophage interactions to be examined in detail, and it offers no way to investigate the hemolytic response in human erythrocytes. Measurement of drug-induced erythrocyte lysis in the test tube (i.e., osmotic fragility) is not a valid substitute for the hemolytic response in vivo, because the toxic endpoint is splenic sequestration. Efforts to develop cell-culture assays based on isolated splenic macrophages or Kupffer cells are hindered by the low cell yield.

To reproduce in vitro the uptake of 5-HPQ-treated erythrocytes that was observed in vivo, the phagocytic activity of mouse monocyte-derived macrophage cell lines toward control and 5-HPQ-treated erythrocytes was examined. In preliminary experiments, we used RAW 264.7 macrophages but they did not respond to 5-HPQ-treated erythrocytes (data not shown), although these cells did ingest erythrocytes that had been treated with neuraminidase (which cleaves sialic acid residues). In contrast, J774A.1 macrophages were found to respond selectively to 5-HPQ-treated erythrocytes. Uptake of the ⁵¹Cr label by these macrophages was dependent on the concentration of 5-HPQ (Fig. 5), and this concentration dependence correlated closely with that of the hemolytic response and with the formation of disulfide-linked adducts of hemoglobin to certain skeletal proteins (Bowman et al., 2004, 2005). Importantly, uptake of intact erythrocytes could clearly be observed (Fig. 6), which supports the conclusion that erythrocyte removal in vivo is due to phagocytosis of intact erythrocytes rather than to intravascular lysis. Moreover, the data suggest that restrictive passage of erythrocytes through the splenic architecture may not be necessary for removal of the damaged cells from the circulation.

To test the validity of this cell-culture system as an experimental model for hemolytic anemia, the effect of depletion of erythrocyte GSH on phagocytic uptake was examined. In agreement with previous studies (Bowman et al., 2004), GSH depletion induced the uptake of erythrocytes that were treated with a subhemolytic concentration of 5-HPQ in GSH-normal erythrocytes (Fig. 7). Overall, these data indicate that J774A.1 macrophages can be used as a valid model to identify cell-surface changes that signal macrophage recognition and phagocytosis. Future studies will examine whether these macrophages can respond to 5-HPQ-treated human erythrocytes.

In summary, we have demonstrated that splenic macrophages play a crucial role in the removal of 5-HPQ-treated rat erythrocytes from the circulation and that this response can be reproduced in vitro. This observation supports the notion that 5-HPQ is an important contributor to the hemolytic response induced by primaquine. Future studies are warranted to determine the nature of the signal that commits the damaged erythrocytes to premature splenic sequestration.

	Acknowledgements

 
We thank Jennifer Schulte and Cynthia Reich for excellent technical assistance in the preparation of this manuscript.

	Footnotes

 
This study was supported by National Institutes of Health Grants AI46424 (to D.C.M.) and C06 RR015455 to the MUSC BSB/CRI Animal Facility. The studies reported in this article were presented in part at the Experimental Biology Meeting, 2004 April 17–21; Washington, DC (Z.S.B.) and the 44th Annual Meeting of the Society of Toxicology, 2005 March 6–10; New Orleans, LA (Z.S.B.).

doi:10.1124/jpet.105.090407.

ABBREVIATIONS: G6PD, glucose-6-phosphate dehydrogenase; 5-HPQ, 5-hydroxyprimaquine; GSH, reduced glutathione; ROS, reactive oxygen species; DEM, diethyl maleate; HBSS, Hanks' buffered saline solution.

Address correspondence to: Dr. David C. McMillan, Department of Cell and Molecular Pharmacology, Medical University of South Carolina, 173 Ashley Ave., Charleston, SC 29425. E-mail: mcmilldc{at}musc.edu

	References

Top Abstract Materials and Methods Results Discussion References

 

Beppu M, Ochiai H, and Kikugawa K (1987) Macrophage recognition of the erythrocytes modified by oxidizing agents. Biochim Biophys Acta 930: 244–253.[Medline]
Beutler E (1969) Drug-induced hemolytic anemia. Pharmacol Rev 21: 73–103.[Abstract/Free Full Text]
Bolchoz LJ, Morrow JD, McMillan DC, and Jollow DJ (2002) Primaquine-induced hemolytic anemia: effect of 6-methoxy-8-hydroxylaminoquinoline on rat erythrocyte sulfhydryl status, membrane lipids, cytoskeletal proteins and morphology. J Pharmacol Exp Ther 303: 141–148.[Abstract/Free Full Text]
Bowman ZS, Morrow JD, Jollow DJ, and McMillan DC (2005) Primaquine-induced hemolytic anemia: role of membrane lipid peroxidation and cytoskeletal protein alterations on the hemotoxicity of 5-hydroxyprimaquine. J Pharmacol Exp Ther 314: 838–845.[Abstract/Free Full Text]
Bowman ZS, Oatis JE Jr, Whelan JL, Jollow DJ, and McMillan DC (2004) Primaquine-induced hemolytic anemia: susceptibility of normal versus glutathione-depleted rat erythrocytes to 5-hydroxyprimaquine. J Pharmacol Exp Ther 309: 79–85.[Abstract/Free Full Text]
Bratosin D, Mazurier J, Tissier JP, Estaquier J, Huart JJ, Ameisen JC, Aminoff D, and Montreuil J (1998) Cellular and molecular mechanisms of senescent erythrocyte phagocytosis by macrophages. A review. Biochimie 80: 173–195.[Medline]
Chadburn A (2000) The spleen: anatomy and anatomical function. Semin Hematol 37: 13–21.[CrossRef][Medline]
Cocco RE and Ucker DS (2001) Distinct modes of macrophage recognition for apoptotic and necrotic cells are not specified exclusively by phosphatidylserine exposure. Mol Biol Cell 12: 919–930.[Abstract/Free Full Text]
Degowin RL, Eppes RB, Powell RD, and Carson PE (1966) The haemolytic effects of diaphenylsulfone (DDS) in normal subjects and in those with glucose-6-phosphate-dehydrogenase deficiency. Bull World Health Organ 35: 165–179.[Medline]
Ganick DJ, Segel GB, Chamberlain J, Hirsch L, and Klemperer MR (1977) The effects of splenectomy and glucocorticoids on survival and hepatic uptake of damaged red cells in the mouse. Am J Hematol 2: 365–373.[Medline]
Harrison JH Jr and Jollow DJ (1986) Role of aniline metabolites in aniline-induced hemolytic anemia. J Pharmacol Exp Ther 238: 1045–1054.[Abstract/Free Full Text]
Jandl JH (1996) Blood: Textbook of Hematology. Little, Brown and Company, Boston.
Kay MM (1993) Generation of senescent cell antigen on old cells initiates IgG binding to a neoantigen. Cell Mol Biol 39: 131–153.[Medline]
Kay MM (1994) Regulatory autoantibody and cellular aging and removal. Adv Exp Med Biol 347: 161–192.[Medline]
Mandal D, Moitra PK, Saha S, and Basu J (2002) Caspase 3 regulates phosphatidylserine externalization and phagocytosis of oxidatively stressed erythrocytes. FEBS Lett 513: 184–188.[CrossRef][Medline]
Oldenborg P, Zheleznyak A, Fang Y, Lagenaur C, Gresham H, and Lindberg F (2000) Role of CD47 as a marker of self on red blood cells. Science (Wash DC) 288: 2051–2054.[Abstract/Free Full Text]
Terpstra V and van Berkel TJ (2000) Scavenger receptors on liver Kupffer cells mediate the in vivo uptake of oxidatively damaged red blood cells in mice. Blood 95: 2157–2163.[Abstract/Free Full Text]
Van Rooijen N, Kors N, vd Ende M and Dijkstra CD (1990) Depletion and repopulation of macrophages in spleen and liver of rat after intravenous treatment with liposome-encapsulated dichloromethylene diphosphonate. Cell Tissue Res 260: 215–222.[CrossRef][Medline]
Van Rooijen N and Sanders A (1994) Liposome mediated depletion of macrophages: mechanism of action, preparation of liposomes and applications. J Immunol Methods 174: 83–93.[CrossRef][Medline]
Van Rooijen N and Van Kesteren-Hendrikx E (2003) "In vivo" depletion of macrophages by liposome-mediated "suicide". Methods Enzymol 373: 3–16.[Medline]

This Article Abstract Full Text (PDF) All Versions of this Article: jpet.105.090407v1 315/3/980    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Bowman, Z. S. Articles by McMillan, D. C. Search for Related Content PubMed PubMed Citation Articles by Bowman, Z. S. Articles by McMillan, D. C.