62, 353–359 (2001)
Copyright © 2001 by the Society of Toxicology
TOXICOLOGICAL SCIENCES
Favism: Effect of Divicine on Rat Erythrocyte Sulfhydryl Status,
Hexose Monophosphate Shunt Activity, Morphology,
and Membrane Skeletal Proteins
David C. McMillan, 1 Laura J. C. Bolchoz, and David J. Jollow
Department of Cell and Molecular Pharmacology and Experimental Therapeutics, Medical University of South Carolina,
171 Ashley Avenue, Charleston, South Carolina 29425
Received April 3, 2001; accepted May 17, 2001
Favism is an acute anemic crisis that can occur in susceptible
individuals who ingest fava beans. The fava bean pyrimidine
aglycone divicine has been identified as a hemotoxic constituent;
however, its mechanism of toxicity remains unknown. We have
shown recently that divicine can induce a favic-like response in
rats and that divicine is directly toxic to rat red cells. In the present
study, we have examined the effect of hemotoxic concentrations of
divicine on rat erythrocyte sulfhydryl status, hexose monophosphate (HMP) shunt activity, morphology, and membrane skeletal
proteins. In vitro exposure of rat red cells to divicine markedly
stimulated HMP shunt activity and resulted in depletion of reduced glutathione with concomitant formation of glutathioneprotein mixed-disulfides. Examination of divicine-treated red cells
by scanning electron microscopy revealed transformation of the
cells to an extreme echinocytic morphology. SDS-PAGE and immunoblotting analysis of the membrane skeletal proteins indicated
that hemotoxicity was associated with the apparent loss of skeletal
protein bands 2.1, 3, and 4.2, and the appearance of membranebound hemoglobin. Treatment of divicine-damaged red cells with
dithiothreitol reversed the protein changes, which indicated that
the observed alterations were due primarily to the formation of
disulfide-linked hemoglobin-skeletal protein adducts. The data
suggest that oxidative modification of hemoglobin and membrane
skeletal proteins by divicine may be key events in the mechanism
underlying favism.
Key Words: hemolytic anemia; favism; divicine; rat; glucose-6phosphate dehydrogenase deficiency; erythrocytes; glutathione.
Favism is a life-threatening hemolytic crisis that can result
from the ingestion of fava beans (Vicia faba) by susceptible
individuals who have low-activity variants of erythrocytic glucose 6-phosphate dehydrogenase (G6PD). Since G6PD regulates the production of NADPH in the red cell by the hexose
monophosphate (HMP) shunt, G6PD-deficient individuals
have a decreased capacity to maintain sufficient levels of
NADPH in response to an oxidative stress (Beutler, 1978).
Early studies identified 2 components of fava beans, divicine
1
To whom correspondence should be addressed. Fax: (843) 792–2475.
E-mail: mcmilldc@musc.edu.
and isouramil, as the probable causative agents based on their
ability to deplete reduced glutathione (GSH) in isolated suspensions of human G6PD-deficient red cells (Mager et al.,
1965). Divicine and isouramil are not present in fava beans per
se, but are aglycones of the biologically inactive fava bean
␤-glucosides, vicine, and convicine, respectively.
The mechanism underlying the onset of favism is not yet
understood. However, it has been postulated that both pyrimidine aglycones, liberated upon digestion of their parent glucosides (Hegazy and Marquardt, 1984), are absorbed into the
blood and induce oxidative damage within erythrocytes as a
consequence of their redox activity (Chevion et al., 1982;
Winterbourn and Munday, 1990). Studies of red cells withdrawn from patients during early and late stages of favic crises
have indicated that GSH depletion and HMP shunt stimulation
are key events that precede red cell loss (Gaetani et al., 1979),
and these biochemical responses are reported to be accompanied by alterations in the morphological appearance of the
erythrocytes when viewed under light microscopy (Fischer et
al., 1985).
We have shown recently that divicine can induce a favic-like
response when administered to G6PD-normal rats, and that
divicine is directly hemotoxic to rat red cells (McMillan and
Jollow, 1999). That is, when incubated with 51Cr-labeled red
cells in vitro, divicine induces alterations in the cells such that
when they are returned to the circulation of isologous rats, the
treated cells are rapidly sequestered into the spleen. In view of
the importance of sulfhydryl status in the progression of favism
and the role of membrane skeletal proteins in the maintenance
of normal red cell shape (Mohandas et al., 1983), the present
studies were undertaken to examine the sulfhydryl status, HMP
shunt activity, morphology, and membrane skeletal proteins of
rat erythrocytes exposed in vitro to hemotoxic concentrations
of divicine. We report that divicine rapidly stimulates HMP
shunt activity and depletes GSH in rat erythrocytes with concomitant formation of glutathione-protein mixed-disulfides.
These biochemical alterations are associated with profound
membrane skeletal protein damage and transformation of the
cells to an extreme echinocytic morphology.
353
354
MCMILLAN, BOLCHOZ, AND JOLLOW
MATERIALS AND METHODS
Chemicals and materials. Divicine (2,6-diamino-5-hydroxy-4[3H]-pyrimidinone) hemisulfate was synthesized as described previously (McMillan et
al., 1993). 14C-1-Glucose and Na 2 51CrO 4 in sterile saline (1 mCi/ml, pH 8) was
purchased from New England Nuclear (Billerica, MA). Rabbit anti-rat hemoglobin, HRP-conjugated goat anti-rabbit IgG, and dithiothreitol (DTT) were
purchased from Sigma (St. Louis, MO). All other chemicals and reagents were
of the best commercially available grade.
Animals. Male Sprague-Dawley rats (130 –150 g) were purchased from
Harlan Laboratories (Indianapolis, IN), and maintained on food and water ad
libitum. Animals were acclimated for 1 week to a 12-h light-dark cycle prior
to their use.
Red cell incubation conditions. Red cells were collected from anesthetized rats into heparinized tubes and washed in isotonic phosphate-buffered
saline (pH 7.4) supplemented with 10 mM D-glucose (PBSG). The red cells
were resuspended to 40% hematocrit and used the same day they were
collected. After a 5-min preincubation at 37°C, various concentrations of
divicine dissolved in PBSG were added to red cell suspensions (2 ml) and
allowed to incubate aerobically for up to 2 h at 37°C.
Analysis of red cell HMP shunt activity and sulfhydryl status. The
evolution of 14CO 2 from 14C-1-glucose was used to estimate HMP shunt
activity in rat red cell suspensions as described previously (Grossman et al.,
1995). For determination of sulfhydryl status, aliquots (200 ␮l) of the red cell
incubation mixtures were removed at various intervals after the addition of
divicine and assayed for GSH, GSSG, and protein-SSG concentration by
HPLC with electrochemical detection as described previously (Jensen et al.,
1986). The amount of sulfhydryl present in the sample was estimated by
comparison of peak height to standards prepared identically to the samples.
Morphological examination of red cells. Control and divicine-treated red
cells were prepared for scanning electron microscopy in a manner similar to
that described by Dewar (1982). After incubation, the red cells were washed
once in PBSG, and fixed and dehydrated as described previously (Grossman et
al., 1992). The cells were then cast with a thin coating of carbon and gold, and
examined in a JEOL JSM-5410LV scanning electron microscope operating at
10 kV accelerating voltage.
Electrophoretic analysis of membrane skeletal proteins. Red cell ghosts
(unsealed membrane vesicles) were prepared from control and divicine-treated
red cells by hypotonic lysis as described previously (Grossman et al., 1992).
The ghosts were washed repetitively with buffer and solubilized with SDS.
Electrophoretic and immunoblotting analysis of the solubilized membrane
ghosts was carried out as described previously (McMillan et al., 1995). The
proteins were resolved on nonreducing, continuous gels (5% monomer and
1.5% bis-acrylamide crosslinker), and the skeletal protein bands were identified according to their migration distance (Fairbanks et al., 1971).
RESULTS
Effect of Divicine on Rat Erythrocyte Sulfhydryl Status and
HMP Shunt Activity
Previous studies have shown that survival of rat 51Cr-labeled
red cells in vivo is reduced dramatically after in vitro exposure
of the radiolabeled cells to divicine (McMillan and Jollow,
1999). The response to divicine is characterized by a narrow
concentration range (1 to 2 mM), with a TC 50 of about 1.5 mM.
Since it is known that GSH depletion and HMP shunt stimulation occur in red cells of G6PD-deficient patients undergoing
a favic crisis (Gaetani et al., 1979), it was of interest to
examine these parameters in rat red cell suspensions treated
with a hemotoxic concentration of divicine. Thus, divicine (1.5
FIG. 1. Effect of divicine on rat erythrocyte sulfhydryl status. Rat red cells
were incubated at 37°C in PBSG containing divicine (1.5 mM). At the
indicated time points, aliquots were withdrawn and assayed for GSH (open
square), GSSG (filled circle), and glutathione-protein mixed-disulfides (filled
triangle); sum of the glutathione species (open diamond). Data points are
means of duplicate determinations.
mM) was added to a 40% suspension of rat red cells, and
aliquots were removed at various time points and assayed for
GSH, GSSG, and protein-SSG. Exposure of rat red cells to
divicine resulted in a very rapid decline in cellular GSH (Fig.
1), which reached a nadir within 10 min. The loss of GSH was
accompanied by a marked increase in protein-SSG formation;
GSSG content was very low and remained constant throughout
the incubation period. The sum of the sulfhydryl species was
constant during the time period of incubation, indicating that
the disappearance of GSH was due primarily to the formation
of mixed disulfides with soluble protein.
Addition of divicine to red cells (40%) also induced a
significant stimulation of HMP shunt activity, as measured by
the accumulation of 14CO 2 (derived from 14C-1-glucose) over a
1-h incubation period. Stimulation of HMP shunt activity by
divicine was concentration dependent (Fig. 2). The stimulatory
effect appeared maximal at about 0.75 mM divicine (corresponding to about an 10-fold increase in activity), and then
declined modestly at divicine concentrations above 1 mM.
Effect of Divicine on Rat Erythrocyte Morphology
Severe alterations in red-cell morphology have been described during the clinical course of favism (Weed and Reed,
1966). To investigate whether morphological transformation of
rat erythrocytes had occurred due to divicine exposure, 10 ␮l
aliquots of control and divicine-treated erythrocyte suspensions
were removed following a 2-h incubation and prepared for
DIVICINE EFFECTS ON RAT ERYTHROCYTES
FIG. 2. Effect of divicine on rat erythrocyte HMP shunt activity. Washed
red cells were suspended in PBSG and placed in flasks containing a center
well. 14C-1-Glucose was added and the flasks sealed. After 1 h of incubation
at 37°C, the reaction was terminated by injection with TCA, and the released
14
CO 2 collected in hyamine hydroxide and counted. The values are means ⫾
SD (n ⫽ 3). *Significantly different from control (p ⬍ 0.05).
355
stained gel (Fig. 4A), divicine induced a concentration-dependent alteration in the protein electrophoretic pattern of rat red
cells. As compared with the PBSG control (lane 1), red cells
exposed to increasing concentrations of divicine (lanes 2– 6)
showed decreases in bands 1 (spectrin), 2.1 (ankyrin), and 4.2,
and an increase in the amount of high molecular weight protein
(HMWP) aggregates that did not enter the gel. In addition,
divicine treatment induced the appearance of membrane-bound
hemoglobin monomer, which can be seen as a new protein
band at the lower molecular weight end of the gel.
Immunoblot analysis of the resolved skeletal proteins (Fig.
4B) revealed the concentration-dependent association of hemoglobin with the membrane. With the exception of a residual
amount of hemoglobin monomer (16 kDa), no antibody staining was observed on blotted protein from control cells (lane 1).
In contrast, protein bands consistent with formation of membrane-bound hemoglobin monomer (16 kDa), dimer (32 kDa),
and tetramer (54 kDa) (Grossman et al., 1992) were observed
in red cells treated with increasing concentrations of divicine
(lanes 2– 6). In addition, diffuse hemoglobin antibody staining
was observed in the spectrin and band 3 regions.
scanning electron microscopy. As shown in Fig. 3A, control
red cells incubated for 2 h at 37°C exhibited the biconcave
appearance of normal discocytes. Echinocytic cells were observed on occasion, but represented ⬍ 3% of the total cells. In
contrast, more than 50% of the red cells exposed to 2 mM
divicine for 2 h at 37°C had lost their discocytic morphology
and exhibited moderate to severe degrees of echinocytosis
(Fig. 3B). The cells were characterized by several moderatelysized protuberances that were asymmetrically distributed on
the cell surface. These transformations could also be observed
to a lesser degree at lower concentrations of divicine (data not
shown), and in Giemsa-stained smears viewed at the light
microscopic level, which indicated that the changes observed
in divicine-treated red cells were not an artifact of the process
of preparing the cells for scanning electron microscopy. Crossbonded erythrocytes, such as those reported to occur in divicine-treated rat red cells exposed to hypertonic media (Fischer
et al., 1985), were notably absent, as were Type III echinocytes
and spheroechinocytes, such as those generated by exposure to
the classical hemolytic agent, phenylhydrazine.
Effect of Divicine on Rat Erythrocyte Membrane Skeletal
Proteins
Membrane skeletal proteins from control and divicinetreated red cells were separated on SDS-PAGE gels and either
stained with Coomassie blue or transferred onto nitrocellulose
membranes and immunostained with rabbit polyclonal antibodies to rat hemoglobin. As shown in the Coomassie blue-
FIG. 3. Scanning electron micrograph of rat erythrocytes incubated for 2 h
at 37°C in (A) vehicle (PBSG) alone, or (B) in PBSG containing 2 mM
divicine. Magnification⫻4000.
356
MCMILLAN, BOLCHOZ, AND JOLLOW
divicine-treated cells were lysed and solubilized with SDS
prior to addition of DTT (data not shown). These data indicate
that the skeletal protein alterations induced by divicine were
due to the formation of intermolecular disulfide bonds.
DISCUSSION
The onset of favism has long been considered to be linked to
the oxidative activity of the fava bean aglycones, divicine, and
FIG. 4. Effect of divicine on rat erythrocyte membrane skeletal proteins.
Rat erythrocytes were incubated for 1 h at 37°C in PBSG containing the
vehicle (lane 1), 0.8 mM (lane 2), 1.0 mM (lane 3), 1.3 mM (lane 4), 1.5 mM
(lane 5), and 2.0 mM (lane 6) divicine. The cells were washed, and membrane
ghosts were prepared and washed exhaustively to remove unbound hemoglobin. The ghost proteins (32 ␮g) were then solubilized in SDS and subjected to
PAGE. (A) Coomassie blue-stained gel. (B) Immunoblot stained with rabbit
polyclonal antibodies to rat hemoglobin.
In view of the observed capacity of divicine to induce the
formation of disulfide bonds between GSH and sulfhydryl
groups of soluble protein in the red cell (Fig. 1), we examined
the involvement of intermolecular disulfide bond formation in
the alteration of the membrane skeletal proteins. Red cells were
incubated with divicine (1.5 mM) for 1 h and then washed and
exposed to DTT (5 mM) for an additional 1 h prior to lysis and
solubilization. As shown in Fig. 5A, exposure of untreated red
cells to DTT alone (lane 2) had no effect on the normal
electrophoretic pattern. However, post-treatment of divicinetreated red cells with DTT (lane 4) reversed the changes in the
major structural proteins and decreased the level of membranebound hemoglobin (Fig. 5B). Similar results were obtained
when cysteamine was substituted for DTT, and when the
FIG. 5. Effect of DTT on divicine-induced membrane skeletal protein
alterations in rat erythrocytes. Rat erythrocytes were incubated with the vehicle
or divicine (1.5 mM) for 1 h, then washed and incubated with DTT (5 mM) for
an additional 1 h before being subjected to SDS-PAGE. (A) Coomassie
blue-stained gel. (B) Immunoblot stained with rabbit polyclonal antibodies to
rat hemoglobin. Lane 1, control; lane 2, DTT control; lane 3, divicine alone;
lane 4, divicine ⫹ DTT.
DIVICINE EFFECTS ON RAT ERYTHROCYTES
isouramil, within erythrocytes. However, despite several decades of research in the area, the mechanism underlying the
favic response remains unclear. Two types of approaches have
been used in attempts to resolve the mechanism: (1) examination of red cells withdrawn from favic patients during a hemolytic crisis, and (2) examination of red cells incubated with
divicine or isouramil in vitro. Although examination of red
cells retrieved from favic patients has provided important information, the major limitation of this approach in regard to
understanding the mechanism is that the red cells that are most
affected, and hence of greatest interest, are rapidly removed
from the circulation by splenic and hepatic sequestration and
are thus unavailable for study. The second approach also has
had major limitations. Divicine is unstable in the presence of
oxygen at physiological pH, and many investigators have chosen to prepare it in situ by enzymatic or chemical hydrolysis of
its parent glucoside, vicine. These procedures, however, yield
only small amounts of material and, in the case of acid hydrolysis, of chemically undefined material (Pedersen et al., 1988).
Furthermore, the relevance of divicine-induced changes observed in these studies is questionable because of the lack of
correlation with the relevant toxicological end-point, premature splenic sequestration of intact red cells.
We have used a direct synthetic method (Bailey et al.,
1982) to prepare a stable hemisulfate salt of divicine (McMillan et al., 1993). This compound was able to provoke a
favic-like response in G6PD-normal rats. That is, divicine
administration reduced the survival of previously infused
51
Cr-labeled red cells and decreased the hematocrit in a
dose-dependent manner (McMillan and Jollow, 1999). The
decrease in hematocrit was matched by an increase in spleen
weight, reflective of the rapid uptake of divicine-damaged
red cells. Of importance for the present studies, divicine
hemotoxicity could be reproduced in vitro by direct exposure of 51 Cr-labeled red cells to divicine before cells were
returned to the circulation of isologous rats. Survival of
these cells was also reduced in a concentration-dependent
manner, and these cells were also found to be removed
preferentially by the spleen. These observations indicate
that the G6PD-normal rat red cell may be used as a model to
examine the relevance of divicine-induced cellular alterations to the hemotoxic response.
The present studies demonstrate that divicine hemotoxicity
in the rat is associated with the development of an oxidative
stress within the red cell. This oxidative stress response was
manifested by depletion of GSH with concomitant formation of
glutathione-protein mixed-disulfides (Fig. 1), and by stimulation of HMP shunt activity (Fig. 2). Under these experimental
conditions, alterations in the electrophoretic pattern of membrane skeletal proteins and the appearance of membrane-bound
hemoglobin were observed (Fig. 4), and the red cells were
transformed to an extreme echinocytic morphology (Fig. 3). As
357
previously reported, when rat red cells treated in this manner
are returned to isologous rats, the cells are rapidly and selectively removed by splenic sequestration (McMillan and Jollow,
1999).
The mechanism by which damaged or senescent red cells are
removed by the spleen remains poorly understood. The exposure of protein, carbohydrate, and/or lipid epitopes on the
external cell surface leading to recognition and ingestion by
splenic macrophages have all been proposed to account for the
removal of abnormal red cells from the circulation (Bratosin et
al., 1998). Although the intracellular signal that provokes the
appearance of an epitope on the external cell surface is also not
known, it has long been postulated that alterations in the
skeletal protein assembly that lies on the inner surface of the
cell membrane may be responsible for initiating this process. In
this regard, a number of studies have suggested a key role for
the binding of hemoglobin to cytoskeletal protein in the mechanism underlying normal red cell senescence. For example,
Low and colleagues (Turrini et al., 1991) showed that hemoglobin binding to the membrane of red cells stimulated the
deposition of autologous antibodies and complement, and that
these cells were ingested by cultured monocytes.
As noted above, examination of skeletal proteins from divicine-treated red cells revealed marked changes in the electrophoretic pattern. Two major features of this response were the
loss of membrane protein bands 1, 2.1, and 4.2, and the
appearance of membrane-associated aggregates of hemoglobin
monomers. The reversal of these changes by treatment of the
cells with DTT (Fig. 5) with the liberation of hemoglobin
monomer indicated that the hemoglobin addition reactions had
occurred as a result of disulfide bond formation. Although the
significance of these protein changes is not yet known, they are
quite similar to those observed previously in ghosts obtained
from rat red cells exposed to dapsone hydroxylamine (Grossman et al., 1992), which is the hemolytic metabolite of the
primary arylamine drug, dapsone.
The origin of the oxidant stress that converts the sulfhydryl
groups of GSH, hemoglobin, and skeletal protein to mixed
disulfides in these divicine-treated red cells is not yet known.
Reversible oxidation of divicine to a quinone at the expense of
molecular oxygen may generate reactive oxygen species
(Chevion et al., 1982), which have the capacity to react with
cellular thiol groups to form reactive thiyl radicals (Bradshaw
et al., 1997; Wefers and Sies, 1983). Hemoglobin thiyl radicals
could then attack the membrane proteins, resulting in the
formation of hemoglobin-skeletal protein adducts and HMWP
aggregates. Autooxidation of divicine in aerobic solution has
been shown to generate hydrogen peroxide, and in the presence
of GSH, the accumulation of GSSG has been observed (Arese,
1982). In our studies, however, GSSG was not detected in
divicine-treated red cells, suggesting that the loss of GSH was
not due to detoxification of peroxide by GSH peroxidase.
358
MCMILLAN, BOLCHOZ, AND JOLLOW
Alternatively, mixed disulfide formation could occur as a result
of attack by a compound-centered free radical (Winterbourn,
1993). In support of this concept, semiquinone radical intermediates of divicine have been detected in acellular systems by
EPR spectroscopy (Albano et al., 1984; Pedersen et al., 1988).
Whether or not these species are formed in red cells is not yet
clear and warrants further investigation.
Several reports of divicine-induced morphological alterations have appeared in the literature (Beutler, 1978; Fischer et
al., 1985; Weed and Reed, 1966). Fischer et al. (1985) described the appearance of a phenomenon referred to as plasma
membrane “crossbonding.” This altered morphology could be
observed in red cells withdrawn from G6PD-deficient individuals during the early stages of a favic crisis, and is characterized by fusion of opposing faces of the internal surface of the
plasma membrane, creating 2 compartments, one containing
hemoglobin, and one apparently devoid of cellular material.
Efforts to reproduce this response in vitro by exposure of rat
red cells to divicine was achieved only when the cells were
placed in a hypertonic medium (400 milliosmolar) to affect
shrinkage of the cells. In the present in vitro incubation studies,
intact treated cells with varying degrees of echinocytosis were
evident but changes suggestive of crossbonding and cell fragmentation were not observed. Of interest, examination of red
cells taken from rats 1 h after administration of a hemotoxic
dose of divicine revealed a high proportion of knizocytes
(pinched cells; triconcave disc) and occasional echinocytes
(McMillan and Jollow, unpublished observations); however,
crossbonded red cells were not evident under these in vivo
conditions.
In summary, the data presented in this report demonstrate
that an oxidative stress response is provoked in rat red cells
exposed to concentrations of divicine that commit the cells to
premature splenic sequestration. Under these experimental
conditions, divicine hemotoxicity in the rat is associated with
severe damage to the membrane skeleton, which is in turn
manifested by alterations in red cell morphology.
ACKNOWLEDGMENTS
This work was supported by NIH grant DK-47423. The authors wish to
thank Jennifer Schulte and Elizabeth Eagleson for their technical assistance in
the preparation of this manuscript.
REFERENCES
Albano, E., Tomasi, A., Mannuzzu, L., and Arese, P. (1984). Detection of a
free radical intermediate from divicine of Vicia faba. Biochem. Pharmacol.
33, 1701–1704.
Beutler, E. (1978). Hemolytic Anemia in Disorders of Red Cell Metabolism.
Plenum Medical, New York.
Bradshaw, T. P., McMillan, D. C., Crouch, R. K., and Jollow, D. J. (1997).
Formation of free radicals and protein mixed disulfides in rat red cells
exposed to dapsone hydroxylamine. Free Rad. Biol. Med. 22, 1183–
1193.
Bratosin, D., Mazurier, J., Tissier, J. P., Estaquier, J., Huart, J. J., Ameisen,
J. C., Aminoff, D., and Montreuil, J. (1998). Cellular and molecular mechanisms of senescent erythrocyte phagocytosis by macrophages. A review.
Biochimie 80, 173–195.
Chevion, M., Navok, T., Glaser, G., and Mager, J. (1982). The chemistry of
favism-inducing compounds. The properties of isouramil and divicine and
their reaction with glutathione. Eur. J. Biochem. 127, 405– 409.
Dewar, C. K. (1982). Preparation of red cells for scanning and transmission
electron microscopy. In Red Cell Membranes, A Methodological Approach
(J. C. Ellory and J. D. Young, Eds.), p. 13. Academic Press, New York.
Fairbanks, G., Steck, T. L., and Wallach, D. F. (1971). Electrophoretic analysis
of the major polypeptides of the human erythrocyte membrane. Biochemistry 10, 2606 –2617.
Fischer, T. M., Meloni, T., Pescarmona, G. P., and Arese, P. (1985).
Membrane cross bonding in red cells in favic crisis: A missing link
in the mechanism of extravascular haemolysis. Br. J. Haematol. 59,
159 –169.
Gaetani, G. F., Mareni, C., Salvidio, E., Galiano, S., Meloni, T., and Arese, P.
(1979). Favism: Erythrocyte metabolism during haemolysis and reticulocytosis. Br. J. Haematol. 43, 39 – 48.
Grossman, S., Budinsky, R., and Jollow, D. (1995). Dapsone-induced hemolytic anemia: Role of glucose-6-phosphate dehydrogenase in the hemolytic
response of rat erythrocytes to N-hydroxydapsone. J. Pharmacol. Exp. Ther.
273, 870 – 877.
Grossman, S. J., Simson, J., and Jollow, D. J. (1992). Dapsone-induced
hemolytic anemia: Effect of N-hydroxy dapsone on the sulfhydryl status and
membrane proteins of rat erythrocytes. Toxicol. Appl. Pharmacol. 117,
208 –217.
Hegazy, M. I., and Marquardt, R. R. (1984). Metabolism of vicine and
convicine in rat tissues: Absorption and excretion patterns and sites of
hydrolysis. J. Sci. Food Agric. 35, 139 –146.
Jensen, C. B., Grossman, S. J., and Jollow, D. J. (1986). Improved method
for determination of cellular thiols, disulfides and protein mixed disulfides using HPLC with electrochemical detection. Adv. Exp. Med. Biol.
197, 407– 413.
Mager, J., Glaser, G., Razin, A., Izak, G., Bien, S., and Noam, M. (1965).
Metabolic effects of pyrimidines derived from fava bean glycosides on
human erythrocytes deficient in glucose-6-phosphate dehydrogenase.
Biochem. Biophys. Res. Commun. 20, 235–240.
McMillan, D. C., and Jollow, D. J. (1999). Favism: Divicine hemotoxicity in
the rat. Toxicol. Sci. 51, 310 –316.
McMillan, D. C., Schey, K. L., Meier, G. P., and Jollow, D. J. (1993).
Chemical analysis and hemolytic activity of the fava bean aglycon divicine.
Chem. Res. Toxicol. 6, 439 – 444.
Arese, P. (1982). Favism—a natural model for the study of hemolytic mechanisms. Rev. Pure Appl. Pharmacol. Sci. 3, 123–183.
McMillan, D. C., Simson, J. V., Budinsky, R. A., and Jollow, D. J. (1995).
Dapsone-induced hemolytic anemia: Effect of dapsone hydroxylamine on sulfhydryl status, membrane skeletal proteins and morphology of human and rat erythrocytes. J. Pharmacol. Exp. Ther. 274,
540 –547.
Bailey, S. W., Weintraub, S. T., Hamilton, S. M., and Ayling, J. E. (1982).
Incorporation of molecular oxygen into pyrimidine cofactors by phenylalanine hydroxylase. J. Biol. Chem. 257, 8253– 8260.
Mohandas, N., Chasis, J. A., and Shohet, S. B. (1983). The influence of
membrane skeleton on red cell deformability, membrane material properties,
and shape. Semin. Hematol. 20, 225–242.
DIVICINE EFFECTS ON RAT ERYTHROCYTES
Pedersen, J. Z., Musci, G., and Rotilio, G. (1988). Electron spin resonance
characterization of the radicals produced by enzymatic or chemical cleavage
of vicine. Biochemistry 27, 8534 – 8536.
Turrini, F., Arese, P., Yuan, J., and Low, P. S. (1991). Clustering of integral
membrane proteins of the human erythrocyte membrane stimulates autologous IgG binding, complement deposition, and phagocytosis. J. Biol. Chem.
266, 23611–23617.
Weed, R. I., and Reed, C. F. (1966). Membrane alterations leading to red cell
destruction. Amer. J. Med. 41, 681– 695.
359
Wefers, H., and Sies, H. (1983). Oxidation of glutathione by the superoxide
radical to the disulfide and the sulfonate yielding singlet oxygen. Eur.
J. Biochem. 137, 29 –36.
Winterbourn, C. C. (1993). Superoxide as an intracellular radical sink. Free
Radic. Biol. Med. 14, 85–90.
Winterbourn, C. C., and Munday, R. (1990). Concerted action of reduced
glutathione and superoxide dismutase in preventing redox cycling of dihydroxypyrimidines, and their role in antioxidant defence. Free Radic. Res.
Commun. 8, 287–293.