From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
Rapid Increase in Red Blood Cell Density Driven by K:CI Cotransport in a
Subset of Sickle Cell Anemia Reticulocytes and Discocytes
By Mary E. Fabry, Jose R. Romero, Iris D. Buchanan, Sandra M. Suzuka, George Stamatoyannopoulos,
Ronald L. Nagel, and Mitzy Canessa
We have previously demonstrated that young normal (M)
and sickle cell anemia (SS)red blood cells are capable of a
volume regulatory decrease response (VRD) driven by a K:CI
cotransporter that is activated by low pH or hypotonic
conditions. We now report on the characteristics of young SS
cells (SS2, discocytes) capable of rapid increase in density in
response to swelling. We have isolated cells with high VRD
response (H-VRD) and low VRD response (L-VRD) cells by
incubation and density-gradient centrifugation under hypotonic conditions. Comparison of these cells in patients homozygous for hemoglobin (Hb)S indicated that H-VRD cells
have 91% more reticulocytes (P < 9 x IO-') than L-VRD
cells, 25% less HbF (P < 5.5 x IO-'), 106% more NEM (Nmethylma1eimide)-stimulatedK:CI cotransport activity (P <
2x
and 86% more volume-stimulated K:CI cotransport
activity (P < 1.8 x
H-VRD and L-VRD cells have similar
G I - P D and Na+/H+antiport activity. In agreement with the
reduced percent HbF in H-VRD cells, F cells (red blood cells
that contain fetal Hb) are depleted from the H-VRD population; however, F reticulocytes are enriched in the H-VRD
population to the same extent as non-F reticulocytes, which
suggests that both F and non-F reticulocytes have a similar
initial distribution of volume-sensitive K:CI cotransport activity but that it may be more rapidly inactivated in F than in S
reticulocytes. We find that H-VRD cells consist of 20%
reticulocytes (or 79% of all reticulocytes in SS2) and 80%
more mature cells. This study demonstrates the role of K:CI
cotransport in determining red blood cell density, the heterogeneity of K:CI cotransport activity in reticulocytes, and the
capacity for rapid change in the density of reticulocytes with
high K:CI cotransport activity. We speculate that the H-VRD
population may be more susceptible t o generation of dense
and irreversibly sickled cells.
o 1991b y The American Society of Hematology.
H
In the present study we ask the following questions: (1)
EMOGLOBIN (Hb) concentration plays a central
what percent of cells are capable of VRD response? (2)
role in the rate and extent of HbS polymerization in
What are the characteristics of cells capable of VRD in
the red blood cells (RBCs) of sickle cell anemia (SS)
terms of transport properties; specifically,do cells with high
patients.'" For this reason, two transport systems that can
VRD response have high KCI cotransport? (3) Are cells
change the volume of young RBCs and hence the mean
capable of VRD homogeneous in cell age? (4) What is the
corpuscular Hb concentration (MCHC) in SS RBCs are of
HbF content of high VRD cells and can it be related to the
great interest: the KCl cotransport system, which controls
formation
of dense and ISCs?
the volume regulatory decrease (VRD) response in hypoWe describe here the presence and properties of the high
tonic or acid condition^,^.^ and the Na+/H+ antiporter,
VRD (H-VRD) and low VRD (L-VRD) SS RBCs using a
which can produce a volume increase (VRl) in hypertonic
novel method of isolation (hypotonic density-gradient cenor acid condition^.^^'^ The activity of these transport systems
trifugation).
We find that H-VRD cells consist of some but
is variable among SS individuals and changes with RBC age
not
all
reticulocytes,
and that L-VRD contain a smaller
and density.5*6s1'
fraction of reticulocytes. H-VRD cells also have low HbF
Fetal Hb (HbF) is also unequally distributed as a
and high KCI cotransport activity, while exhibiting similar
function of cell density and is enriched in those sickle cells
levels
of glucose-6-phosphate dehydrogenase (G-6-PD)
of intermediate density (SS3 in our fractionation system),"
and
Na+/H+
antiport activity when compared with that of
which are denser than discocytes (SS2) and are presumably
L-VRD
cells.
We show that H-VRD SS cells can increase
older than both discocytes and the more dense SS4 cells
(irreversibly sickled cells [ISCs] and other dense cell^).'^^'^
HbF interferes with polymerization of deoxygenated HbS
From the Division of Hematology, Albert Einstein College of
in v i t r ~ ' ~and
, ' ~ reduces sickling (shape changes induced by
Medicine, Bronx, Ny;Endocrine-HypertensionDivision, Brigham and
Women's Hospital, Harvard Medical School, Boston, MA; and the
deoxygenation) in HbS containing RBCs." The putative
Division of Medical Genetics, University of Washington Medical
mechanism of enrichment is the preferential survival of F
School, Seattle.
cells.13,''
SubmittedAugust 29,1990; accepted February 28,1991.
Recent work on the characteristics of transport systems
Supported in part by Grants NHLBI 35664, SPSO-3656, HL21016,
involved in volume regulation of human RBCs will serve as
and HL38655from the National Institutes of Health.
the basis for our s t ~ d i e s . ~ - ' Two
' . ' ~ ~aspects
~ ~ merit emphasis:
Presented in part at the American Federation for Clinical Research,
(1)KCI cotransport activity appears to decrease sharply on
May 1989. Fabv ME, Romero JR, Nagel RL,Canessa M: Sickle Cells
reticulocyte
The activity of mature RBCs is
are Heterogeneous in Volume Response:A Mechanism for Irreversibly
10 to 20 times lower than that observed in young cells.6~Z0~21Sickled Cells (ISC) Formation. Clin Res 37(2):380A, 1989 (abstr).
Address reprint requests to Mary E. Fabry, PhD, Albert Einstein
(2) In contrast, the Na+/H+antiporter activity is 5 to 10
College of Medicine, Department of MedicineJUllmann Room 915,
times higher in young cells but appears to decrease rela1300 Morris ParkAve, B r o w Ny1046I.
tively slowly in SS RBCs with increased cell density."
The publication costs of this article were defrayed in part by page
We have previously demonstrated that in the densitycharge payment. This article must therefore be hereby marked
defined discocyte fraction (SS2), which contains both half
"advertisement" in accordance with 18 U.S.C. section 1734 solely to
of all cells and half of all reticulocytes, some but not all of
indicate this fact.
the cells are capable of cell volume regulatory decrease
0 I991 by The American Society of Hematology.
when swollen by exposure to hypotonic media or acid P H . ~
0006-4971I9117801-0022$3.00/0
Blood, Vol78, No 1 (July 1). 1991: pp 217-225
217
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
FABRY
218
their MCHC several grams per deciliter in less than an
hour, a factor that favors HbS polymer formation. We
suggest that the reticulocytes and young cells that exhibit
strong VRD response may become dense cells and possibly
ISCS.
MATERIALS AND METHODS
Patient material. Blood was drawn from 12 SS patients followed
in the Heredity Clinic (Director Dr H. Billett) of the Bronx
Comprehensive Sickle Cell Center (Bronx, NY) after informed
consent and characterized by two electrophoresis methods and a
solubility test for HbS. All patients included in the studies
described here were homozygous for HbS (SS) by MstII digestion.
Approximately half of the SS patients had concomitant
a-thalassemia. Reticulocytes, HbF, and G d - P D were measured on
all 12 patients; transport studies were performed on nine patients;
and F cells and F reticulocytes were determined for six patients,
which were a subset of the nine used for transport studies.
Preparation ofH-VRD and L-VRDcells. All preparations (isolation of SS2 and separation into H-VRD and L-VRD cells) were
performed on the same day as the blood was drawn. The cells were
first passed through a Pall filter (Pall Biomedicals, Inc, Fejardo,
Puerto Rico) to remove microclots and most of the white blood
cells (WBCs). The discocyte fraction of cells was then isolated from
SS cells by mixing 0.5 mL of packed SS whole blood with 5.5 mL of
Percoll (colloidal silica coated with polyvinylpyrrolidone; Pharmacia Fine Chemical, Piscataway, NJ)-Larex (arabinogalactan; Consulting Associates Inc, Tacoma, WA) gradient mix prepared as
previously describedI2and centrifuging the mixture for 30 minutes
at 38°C (gradient A). This procedure removes the remainder of the
1.
WBCs. The ionic composition of the gradient mix was millimolar: 5
glucose, 0.5 g% bovine serum albumin (BSA), 5 PO, buffer at pH
7.4, and 100 NaCl with 4 KCI and 1 MgCI,. Cells from SS patients
were separated into four density fractions'*: SSl, which consists of
reticulocytes and young cells with MCHC less than 33 g/dL (density
< 1.076); SS2, which contains reticulocytes, young RBCs, and
discocytes with an MCHC between 33 and 37g/dL (density 1.076 to
1.091); SS3, which consists of cells with an MCHC between 37 and
42 g/dL; and SS4, which consists of very dense discocytes and ISCs
with MCHCs greater than 42 g/dL. The isolated SS2 cells have
about the same percent reticulocytes as the lighter fraction, SS1,
and have a smaller percent HbF than those in the denser fraction,
SS3 (see Fig 4).
SS2 cells were isolated by pipette from isotonic Percoll-Larex
density gradients (gradient A), washed three times with isotonic
saline, resuspended in a hypotonic saline solution containing
millimolar: 10 glucose, 0.5 g% BSA, 0.1 ouabain, 5 PO, buffer at
pH 7.4, and either 100 NaCl with 4 KCI and 1 MgCI,, or 100 NaNO,
with 4 KNO, and 1 Mg(NO,),, and incubated for 30 minutes at
38°C. At the end of 30 minutes the cells were washed with fresh
hypotonic media, and then reconcentrated. The cells were then
separated by density-gradient centrifugation at 38°C for a second
time by mixing 0.5 mL of packed cells with 5.5 mL of hypotonic
Percoll-Larex mixture (gradient B, containing either CI- or NO,as the anion), which has a different density profile than that used
for preparation of SS2. This gradient mixture has a constant
concentration of Percoll and a reduced level of Stractan to allow
better separation of low MCHC cells. The procedure is illustrated
in Fig 1. The second gradient mixture (gradient B) was prepared by
adding 17.5 mL of Percoll in 32.5 mL of a solution containing
2.
P rep a r a t ive
Fractiona tion
SSI
lncu bat ion
-
ss2
-
ss3
-
3.
4.
Preparative
Fractionation
Assay
Hypotonic incubation
in CI'or NO<
pH7.4, 220mOsm
LV R
HVR
-
ss4
-
Sample: SSWB
Anion: +CI'
ss2
+
Osmolarity : +280
Gradient:A .___(
ET AL
ss2 ss2
NO; CI'
+220--r
+e+
Fig 1. Experimental method for isolation of H-VRD and L-VRD cells. (1) SS2 cells are isolated from sickle cell whole blood (SSWB) by
density-gradient centrifugation using an isotonic continuous density gradient with physiologic concentrations of Na', K', and C1- (Gradient A). (2)
SS2 cells are incubated in hypotonic CI- or NO,- media as described in the text at 37°C for 30 minutes. (3) The cells are then separated a second
time on a hypotonic gradient with a different density profile (Gradient B) as described in Materials and Methods, which contains either CI- or
NO;, into an H-VRD population which has an MCHC greater than that of cells incubated in NO; and an L-VRD population that has simply swollen
passively. (4) Finally, the separated cells are assayed for the properties discussed in the text.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
219
VOLUME REGULATION IN SICKLE CELLS
millimolar: 4 KCl or KNO,, 1 MgC1, or 1 Mg(NO,),, and 5 PO, at
pH 7.4 and adding 1 part of this mixture to 5 parts of the standard
gradient mixture to give a final osmolarity of 220 mOsm. Cells were
removed from the gradient by aspiration.
For the first series of studies the column of cells was divided into
three equal parts: the top, middle, and bottom. For cation
transport studies that required more cells, the column of cells was
equally divided into top and bottom. Cells for cation transport were
washed three times with isotonic saline to remove the gradient
mixture, then twice with preservation solution," adjusted to a
hematocrit (Hct) of 50, and packed in wet ice for overnight
shipment to Boston. Density classes prepared by this technique are
very sharply defined, as seen in Fig 1 (cells labeled SS2).
Reticulocyte, G-6-PD, and HbF determinations. Reticulocytes
were determined by resuspending aliquots of cells in centrifuged
plasma at Hct 50. Equal volumes of blood and new methylene blue
reticulocyte stain were mixed and the samples were allowed to
incubate at room temperature for at least 10 minutes, after which
smears were made for counting. For G-6-PD determinations, an
aliquot of whole blood and density-defined cells was washed three
times with isotonic saline, adjusted to an approximate Hct of 50,
and frozen at -135°C until they could be assayed. G-6-PD was
measured using Sigma kit (345-UV, Sigma, St Louis, MO) for
kinetic determination of enzyme activity at 340 nm. Complete lysis
was ensured by freezing in liquid nitrogen and thawing in water
three times. Samples were spun in a microfuge before recording
ultraviolet absorption to remove particulate matter. G-6-PD activity was recorded as Uig of hemoglobin where U is the amount of
Gd-PD activity which will convert one micromole of substrate per
minute. Hemoglobin F was determined by electrophoresis of 3 g%
hemolysate on Corning citrate agar gel (Medfield, MA), pH 6.3 for
50 minutes. After drying, the gel was scanned on a Corning 720
densitometer and the percent of the fastest moving band was
recorded.
Measurement of E C l cotransporl activiy. All transport studies
were performed on the morning after the blood was drawn. All
cells and fractions were loaded by a modified nystatin procedure to
have 11.5 f 2 Na+ and 95 5 9 K+ nmoliL cell. K' fluxes were
determined as previously de~cribed~,','~
by incubating cells at 2%
(vol/vol) Hct in the following media millimolar: A NaCl 140, pH
7.4; B: NaNO, 140, pH 7.4; C: NaCl 100, pH 7.4; and D: NaNO,
100, pH 7.4. All media contained millimolar: 0.1 ouabain (to inhibit
the Na+ pump), 1MgCI, or 1Mg(NO,),, 0.1 bumetanide (to inhibit
Na:KCI cotransport), 10 glucose, and Tris-MOPS at pH 7.4 at
37°C.
C1--dependent K+ transport was estimated by subtracting the
flux in isotonic NO,- media from that in isotonic C1- media (A-B)
or similarly in hypotonic media (C-D). The volume-dependent K+
efflux was estimated from the difference between hypotonic and
isotonic NaCl (C-A) medium. The C1- and volume-dependent K'
efflux was estimated from the difference between C1- and NO,media (C-D). The NEM (N-ethy1maleimide)-stimulatedefflux was
estimated by adding 1 mmol/L NEM to mediums A and B. The
NEM-stimulated CI--dependent K+ efflux was estimated from the
differences in K' efflux in C1- (A-B) and NO,- (B-D) media. Initial
rates of K' efflux were measured by sampling the efflux media in
duplicate samples at 1, 15, and 30 minutes, which were centrifuged
and the supernatant removed for K+ measurement by atomic
absorption. The Kt efflux was calculated from the slope of the
change in potassium concentration versus time and the Hct of the
cells in the flux media.
Measurements of Na'iH' exchange activity. Because the human
antiporter is only partially amiloride sensitive, the activity was
determined by measuring net Na+ influx into acid-loaded cells
driven by an outward H + gradient."%%RBCs were nystatin treated
to reduce intracellular Na' (Na,) content to 1.5 mmoliL cell water
and then acid loaded to an intracellular pH of 6.0 by incubation in
acid-loading solution and 50 kmol/L DIDS (4,4-diisothiocjanatestilbene-2,2'-disulfonic acid) to determine the maximum rate, as
previously described." Na' influx measurements were started by
addition of packed acid-loaded, Na,-depleted cells (1% final Hct)
to Na' media with pH, 6.0 or 8.0, preincubated at 37°C in a shaking
water bath. The difference between the Na+ influx at pH 8.0 to pH
6.0 was calculated as Na+ movement driven by an outward proton
gradient. As previously described, this component is equal to that
of H + efflux driven by an inward Na+ gradient and therefore
represents Nat/H+ exchange a~tivity.'~
Determination of F cells and F reticulocytes. HbF in erythrocytes
was detected by incubating whole blood or density-defined RBCs
with new methylene blue or brilliant cresyl blue for induction of
reticular precipitation. Films prepared from this mixture were air
dried, fixed in acetone-methanol (9:l vol/vol), rinsed in phosphatebuffered saline and then in water, dried, and then incubated with
staining solution containing anti-HbF antibodies conjugated to
fluorescein isothiocyanate (anti-HbF-FITC) and 0.01% acridine
orange. The preparations were incubated in moist chambers at
37°C for 2 hours. The acridine orange stained the precipitated
reticulum of the reticulocytes bright orange, whereas the anti-HbFFITC labeled the HbF bright green." This method permitted
distinction and quantification of reticulocytes containing HbF (F
reticulocytes) from those lacking it (S reticulocytes). This is the
most sensitive method for detection of intracellular Hbl- and is
capable of detecting 3 pg of HbF per cell or lessz6(normal cells
contain about 30 pg of Hb).
RESULTS
Isolation of H - W and L-VRD cells from SS2 cells. We
previously reported that some sickle cell discocytes (SS2)
respond to swelling induced either by hypotonic C1- or acid
pH media by activating KCI cotransport, which produces a
net loss of K+, C1-, and water and increases MCHC
(increases RBC density, reduces cell volume).6 In the
present study, we have selected hypotonic treatment because it elicits a more selective response of K C l cotransport. In contrast, acid-induced swelling modifies the activity
of other ion transport systems (ie, Na+/H+ and CI-/OHexchange) in addition to stimulating K C l cotransport. As
indicated in Materials and Methods, density-defined SS2
cells were first isolated under isotonic conditions from
whole blood of 12 SS patients and subjected to 30 minutes
of hypotonic incubation in the presence of C1- or NO,followed by separation on hypotonic continuous density
gradients in the presence of C1- or NO,- (Fig 1). Thus,
hypotonic incubation separates SS2 cells into two populations of cells: one that is capable of H-VRD response and
another that has a small or nonexistent response and swells
(L-VRD cells). The average MCHC of H-VRD cells
exceeds that of L-VRD cells by 2 to 4 gidL. To demonstrate
that this response is chloride dependent, we determined the
distribution of cells after exposure to hypotonic NO,media. We compared the distribution of cells found under
hypotonic C1- conditions to the distribution of cells under
hypotonic NO,- conditions by densitometry (Fig 2 ) and
determined the percent of cells that respond to hypotonic
CI- by shrinking. We found that 60% f 34%, n = 25
(mean ? SD, n = number of experiments, for 12 patients)
of the cells in hypotonic C1- had a density (MCHC) greater
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
220
FABRY ET AL
.
..
- Depth in tube
:
'
0
/u
m
I
I
....,- _...
I
-
I
\p
Fig 2. Density distribution of SS2 cells in a typical experiment
involving hypotonic chloride and nitrate. Depth in the tube increases
from the top on the left. The densitometer tracing is evaluated for 10
intervals as described." The MCHC indicated on the top was determined by using Pharmacia density marker beads and MCHC measurements of aspirated cells. The solid circles ( 0 )represent the density of
cells in hypotonic (220 mOsm) nitrate and the open circles ( 0 )
represent that in hypotonic chloride. In this example 60% of the cells
in hypotonic chloride have an MCHC greater than 95% of the cells in
hypotonic nitrate.
than 90% of the cells in hypotonic NO,-. The average
MCHC of SS2 cells in hypotonic NO,- is 27 & 1 g/dL, which
is comparable with that found for AA cells at the same
osmolarity (220 mOsm); the MCHC of SS2 cells increases
to 31 ? 1 g/dL in hypotonic chloride.
Activation of K:Cl cotransport results in the redistribution of
reticulocytes and HbF. To determine if the SS2 cells that
are capable of changing their density (volume) belong to a
distinct population, we compared the distribution of several
RBC characteristics after hypotonic incubation of SS2 cells
and gradient separation in the presence of C1- and NO,-.
Three density classes with approximately equal numbers of
cells in each class (top, middle, and bottom fractions) were
isolated. The top (least dense) cells in the hypotonic
chloride fractionation are more extreme representatives of
L-VRD cells and the bottom (densest) cells are a subset of
H-VRD cells. Fig 3A (solid bars) illustrates that reticulocytes are strongly enriched (79.3% t 25.9%, [bottom - top]/
bottom 2 SD, P < 1.6 X lo-', n = 9) while G-6-PD (Fig
3B, solid bars) is unchanged.
In contrast to the reticulocyte distribution, HbF is depleted in the bottom hypotonic chloride fraction
(-72.5% 2 90.5%, [bottom - top]/bottom ? SD, P < .014,
n = 13) (Fig 3C, solid bars). The behavior of F cells was
studied in a smaller sample of six patients that was also used
to study the behavior of F reticulocytes (as described
below). In this smaller sample, the trend observed (F cells
were depleted in the bottom of hypotonic chloride gradients) was consistent with that seen in the larger sample in
which HbF was measured, but was not statistically significant (Fig3D, solid bars, -40.2% & 81.2%,P < .28, n = 6).
That the density redistribution of reticulocytes, HbF, and
F cells is indeed driven by K.Cl cotransport can be demonstrated by comparing the results of experiments performed
in hypotonic chloride (where KCl cotransport is active,
solid bars) with those in hypotonic nitrate media (where
KCl cotransport is inactive, open bars). In Fig 3, the
bottom panels (delta [chloride - nitrate], hatched bars)
depicts the chloride dependent response of the three SS2
subfractions. A positive value represents enrichment in the
presence of C1- and a negative value represents depletion.
Paired t-tests indicate that the difference between C1- and
200
6
200
I
A
ul
+
V
D
150
150
U
.$
100
100
c
50
50
D
a
D
0
5.
a
u
p
-0
9
F
5.
100
7
50
100
50
I
1
50
50
-50
0
c
g -100
-150
-100
-150
T
M
B
T
M
B
-P
B
8
b
r0
o
p
moo
100
:
3
100
50
u"
+
7
a
5
;p
2
I
Y
l
-+!
150
n
-03
3.
0 : :
I
z
-50
3
2
v
-100
-100
T
M
B
T
M
B
Fig 3. Distribution of reticulocytes, G-6-PD, percent HbF, and percent F cells in hypotonic chloride and nitrate density gradients. T, top; M,
middle; 6. bottom. The cells in the bottom fraction are the H-VRD-like population. All levels are normalized t o the value of SS2 and averaged; the
error bars are the standard error of the average. (A) and ( 6 ) in the upper panels show the effect of hypotonic chloride (M) or nitrate (0)
on the
distribution of reticulocytes (A, n = 13) and G-6-PD (6,n = 13) activity in three densityfractions of SS2. The bottom panels (delta) show the paired
difference between the values in CI- and nitrate. A positive delta value indicates enrichment in the H-VRD fraction. A negative delta value
indicates that the property was depleted in the fraction cell population. (C) and (D) in the upper panels show the effect of hypotonic chloride (.I or
hypotonic nitrate (0)
on the distribution of HbF (C, n = 13) and F cells (D, n = 6) in three densityfractions of SS2. The bottom row of panels shows
the difference (delta) between the value in CI- and that in nitrate. In all cases, the top and bottom fractions were significantly enriched or depleted
with respect t o the control NO,- values when incubated in hypotonic CI- media.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
VOLUME REGULATION IN SICKLE CELLS
221
NO,- values for the top cells (L-VRD-like) and bottom
cells (H-VRD-like) are statistically significant in all cases
(P < .05 to .005).
The distribution of cells found using hypotonic chloride
gradients also differs from that found on physiologic,
isotonic density gradients. Figure 4A illustrates the distribution of reticulocytes, G-6-PD, and HbF under isotonic
conditions. SS cells with low density (MCHC) are enriched
in reticulocytes and G-6-PD activity and low in HbF. The
trends in percent reticulocytes and HbF seen under isotonic
conditions are reversed in the presence of hypotonic C1-.
This observation gives us the unique opportunity to estimate the effect of KCI cotransport on reticulocyte hydration. The movement of reticulocytes to the bottom fraction
under hypotonic chloride conditions during the 1-hour
period of the experiment implies that activation of K C l
cotransport can induce an increase of MCHC of 1 to 3
g/dL X hour. Because this is not a kinetic experiment, this
estimate represents a lower limit of the rate of dehydration.
In a separate experiment, SS2 cells were incubated for
1/2,1, and 2 hours in hypotonic media: no further change in
any of the properties was observed after the first half hour
of incubation. This indicates that all changes described in
this report were complete within the 30 minutes of incubation and the 30 minutes of separation by density-gradient
centrifugation.
Transport properties of isolated H - K W and L-KW SS2
cells. To test the hypothesis that the cells which have
migrated to higher density in hypotonic chloride gradients
have arrived there due to their elevated KCI cotransport
activity, NEM- and volume-stimulated and chloridedependent K+ efflux were measured in isolated H-VRD
and L-VRD cells. These experiments are illustrated for a
single patient in Table 1. Under isotonic conditions, there
was no chloride-dependent K+ efflux in either fraction. The
Table 1. CI-Dependent K’ Efflux From H-VRD and L-VRD Sickle Cells
Media Composition
(mmol/L)
L-VRD Cells
K+ Efflux*
H-VRDCells
K + Efflux*
3.2 f 0.2
4.8 +- 0.03
140 NaCl
7.4 f 0.17 7.1 f 0.29
140 NaNO,
-4.3
-4.2
A CI
140 NaCl + NEM
13.4 2 0.09 23.1 2 0.4
10.3 f 0.11 12.4 f 0.3
140 NaNO, + NEM
A CI, NEM stimulated
15
7.3
26.9 f 0.4 51.7 f 0.17
100 NaCl
3.4 f 0.2
3.7 2 0.14
100 NaNO,
A CI, volume stimulated
23.6
48.0
‘H-VRD) - ‘L-VRD)
LVRD %
50
None
None
105
104
Patient EM.
*K+ efflux units are in mmol/L cell x h 2 SEM. All cells were loaded
by the nystatin procedure to contain 100 mmol K’/L cell.
NEM-stimulated, chloride-dependent K+ efflux was 105%
higher in H-VRD than in L-VRD. Exposure to hypotonic
conditions showed a volume-stimulated, chloride-dependent K+ efflux that was 104% higher in H-VRD than in
L-VRD. Figure 5A (top panel) illustrates variation among
the seven patients of NEM-stimulated KCI cotransport of
H-VRD (solid bars) and L-VRD (open bars). The difference in activity between these fractions (Fig 5A, delta,
bottom panel) is positive in all cases. The mean values of
H-VRD and L-VRD are shown in Table 2. Because of
inter-individual variation in transport activity (Fig 5A,top
panels), the differences between H-VRD and L-VRD cells
were normalized before statistical analysis. Two approaches were used that yield results which are qualitatively
very similar. In Table 2 the difference was normalized to the
ion flux in SS2, using the following equation:
-
20
5 0
a
-20
.I
WE
SS1 SS2 SS3 SS4
.
.
SS1
SS2
.
.
I
SS3 SS4
Fig 5.
NEM-stimulated K:CI cotransport and Na+/H+antiport activ-
ity in L-VRD and H-VRD cells from seven SS individuals. Transport
Fig 4. Distribution of cells, percent reticulocytes, absolute percent
reticulocytes, HbF, and G-6-PD activity in isotonic density gradients of
SS whole blood. (A) Distribution of percent reticulocytes, percent
HbF, and G-6-PD activity in RBC density fractions. The results represent the average and error bars are standard errors of values obtained
in the nine patients used for transport studies. Reticulocytes (0).
G-6-PD activity (01, and HbF (A). In this panel the percent cells
expressing a property are the percent cells in each fraction. The
difference in the percent HbF in SS1 and SS4 versus that in SS2 and
SS3 is statistically significant. (B) Density distribution of cells and
absolute reticulocytes of SS whole blood under isotonic conditions.
The results are the average value and standard error for the nine
patients used in the transport studies. (0).
Percent cells in each
density fraction; (01,the absolute percent of all reticulocytes in each
fraction.
activities were expressed in flux units millimolar cells x hour. (A) The
upper panel shows the NEM-stimulated K:CI cotransport activity of
H-VRD (m)and L-VRD ( 0 )cells. The bottom panel shows (0)
the
difference between the H-VRD cells and the L-VRD cells (delta =
H-VRD - L-VRD). The standard error of the mean of individual K:CI
cotransport measurements varied between 0.1 t o 0.4 flux units (see
also Table 1). Note that NEM-stimulated K:CI cotransport was enriched (positive values) in the H-VRD cells in all cases. (B) The upper
panel shows the Na+/H+exchange activity of H-VRD (U) and L-VRD
(0)
cells. For individual measurements of Na+/H+ exchange the
standard error of the mean varied between 1 and 4 flux units. The
bottom panel shows the differences between the H-VRD cells and the
L-VRD cells (delta = H-VRD - L-VRD) (0).
Note that the Na+/H+
antiport activity did not show a consistent pattern of enrichment or
depletion between H-VRD and L-VRD cells.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
222
FABRY ET AL
Table 2. Properties of H-VRD and L-VRD HbS RBCs
Property
N'
L-VRDt
Average
H-VRDt
Average
% Reticulocytes
Yo HbF
G-6-PD (U/g Hb)
K:CI Volume stimulated
K:CI NEM stimulated§
Na+/H+exchange§
23
23
23
9
7
8
7.7 2 5.3
6.4 2 3.4
13.5 2 3.3
6.3 2 5.5
15.6 2 17.5
52.5 2 46.8
19.7 2 8.7
5.2 2 3.0
14.3 t 3.2
19.1 2 14.1
23.2 2 23.1
61.6 t 65.0
IH-VRD - L-VRD)*
ss2
+91%
-25%
+6%
106%
+86%
+18%
+
Paired 1-Test
P<
8.7 x 10-9
5.5 x 10-6
1.4 x
2 x 10-4
1.8 x 10-3
0.27
'The number of patients sampled.
tThe standard deviation for an unpaired average is large because of large inter-individualvariation.
*The values given in the last two columns are for a pairwise comparison of H-VRD and L-VRD results.
§mmol/L cells x h.
fl~~(L.,RD))/fl~~(ss2).
Alternatively, the difference was normalized to the ion flux in the L-VRD fraction using the
flux~,,,,))/flux(,,,,)
; the results of
equation: ( f l ~ ~ ( ~
. ~ ~ ~ )
this calculation are given in the text below: KCI NEMstimulated 150% ? 33%, P < .006, N = 7; KCI volumestimulated 135% ? 47.7%, P < .03, N = 9 (mean ? SE, P,
number of patients). To illustrate the average magnitude of
the differences between H-VRD and L-VRD cells, the
average values are given in the third and fourth columns of
Table 2. Both volume- and NEM-stimulated KCI cotransport were approximately two times more active in H-VRD
cells than in L-VRD cells.
The V,,, of Na+/H+antiport activity was also studied in
H-VRD and L-VRD fractions (Fig 5B, top); the difference
between H-VRD and L-VRD cells (Fig 5B, delta, bottom)
was highly variable and was not statistically significantly
(+18%, P < 0.27, Table 2) or computed as: flu^(^.,^^) flq,v,,~)/flux~L.,RD~
NaH + 36% ? 26%, P < .21, N = 8
(mean h SE, P, number of patients).
The distribution of reticulocytes, G-6-PD, and HbF in
H-T.rRD and L - W RBCs. The H-VRD and L-VRD cells
used for transport studies were also analyzed for the
distribution of reticulocytes, G-6-PD, and HbF (Table 2).
The reticulocyte count in the H-VRD subfraction exhibits a
91%enrichment over that of the L-VRD cells (P < 9 x W9,
Table 2).
Although RBC G-6-PD activity is strongly density dependent (Fig 4A, solid circles), it was only slightly enriched
(+6%,Table 2) in H-VRD cells. Another way of looking at
the relative RBC age in L-VRD and H-VRD cells is to
compare the ratio of the percent reticulocytes with the
G-6-PD (Retics/G-6-PD) activity in the same population.
The G-6-PD and percent reticulocytes for the least-dense
cells (SS1, Fig 4A) may be taken as being representative of
a young cell population. Low values of the ReticdG-6-PD
ratio imply that the cell population is more homogeneous
than high values, because a mixed population of young and
old cells would yield a high reticulocyte count but a low
average value for G-6-PD. We find that ReticsiG-6-PD is
increased from 0.47 0.26 in L-VRD cells to 1.27 ? 0.42
(n = 16) in H-VRD cells. This finding suggests that the
youngest nonreticulocytes with high G-6-PD similar to that
found in SSl (Fig 4A) were not highly enriched in H-VRD
cells. Therefore, the H-VRD population is heterogenous
with respect to cell age and contains some older cells with
high KCI cotransport, whereas the L-VRD fraction con-
tains some reticulocytes and young cells with relatively low
KCI cotransport activity.
The effect of a-thalassemia on the distribution of reticulocytes was examined and we found that the ratio
(ReticsbOtt,- Retics,,,)/Retics,,, was 210% f. 140%, N =
13 for a-thalassemia trait (aa/-a) and 440% ? 520%, N =
10 for patients with the full complement of a genes;
however, the result was not statistically significant (P < .15).
H-VRD cells have 25% less HbF than L-VRD cells
(P < 5.5 X
Table 2). This trend is the same as that
observed in cells separated into three populations (Fig 3)
and is in the opposite direction to that found in cells
separated on hypotonic NO,- gradients (Fig 3A) and to that
inferred for whole blood (Fig 4A). The ratio of
(HbFh,,,, - HbF,,,)/HbF,,, was found to be - 17% h 13%,
N = 13 for a-thalassemia trait and -22% h 21%, N = 10
for patients with all a genes. This trend was also not
statistically significant (P < .45).
Percent discocytes and reticulocytes exhibiting H- T.rRD behavior. What percent of the reticulocytes in SS2 are H-VRD
reticulocytes? The percent reticulocytes in each fraction
were converted to the absolute number of reticulocytes by
multiplying the percent cells in the fraction times the
percent reticulocytes in the fraction and normalizing to
100%. Because the average percent of H-VRD cells in SS2
is 60%
25% and 19.7% ? 8.7% of these cells are
reticulocytes, and L-VRD cells constitute 40% SS2 and are
7.7% ? 5.3% reticulocytes, performing this calculation for
each patient and averaging the results leads to the conclusion that in absolute terms 79% f 13% (mean h SD,
N = 12) of all reticulocytes present in SS2 are H-VRD
reticulocytes.
Because we have only examined SS2 in detail, we can use
this value to estimate the minimum percent of all reticulocytes that are H-VRD cells. The SS2 fraction used to
prepare H-VRD is 47% 2 5% of the total RBCs (Fig 4B,
open circles) and contains 22% ? 8% reticulocytes (Fig 4A,
open circles). When the same procedure is used to calculate
the absolute percent reticulocytes in SS2 for each patient
and is averaged, we find that SS2 contains 57% ? 9% of all
reticulocytes (Fig 4B, solid circles). Thus, at least 45%
(absolute) of all reticulocytes in whole blood are H-VRD
cells. This finding implies that, at a minimum, nearly half of
all reticulocytes have the potential for very rapid dehydration when appropriately challenged, and that at least 12%
do not respond.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
223
VOLUME REGULATION IN SICKLE CELLS
Distribution of reticulocytes, F reticulocytes, and F cells in
H - K W and L-KW RBCs. To test whether the unequal
distribution of HbF between H-VRD and L-VRD cells
reflects low KCI cotransport in F reticulocytes, the percent
reticulocytes, F reticulocytes, and F cells were determined
for six patients (Fig 6). The absolute percent reticulocytes
in the bottom fraction of hypotonic chloride gradients
(H-VRD-like cells) are enriched over the absolute percent
reticulocytes in the bottom fraction of nitrate gradients by
21.7% ? 11.9% chloride^,,,,,,, - nitrate^,,,,,,, r SD,
P < 6.5 x
n = 6, Fig 6A). The absolute percent F
reticulocytes (percent F reticulocytes multiplied by percent
reticulocytes multiplied by the number of cells, normalized
to 100%) were enriched to the same extent that reticulocytes are enriched 25.4% ? 16.4% (chloride(bottom)
nitrate^,,,,,,) 2 SD, P < .013, n = 6, Fig 6B). Another way of
examining these data is to determine the percent of
reticulocytes that are F reticulocytes; this value was constant (6.5 ? 0.8, mean ? standard deviation) in all density
fractions studied. However, F cells are selectively depleted
from the bottom fraction by -23.8%. These results indicate
that the K:Cl cotransport activity of F reticulocytes and
non-F reticulocytes are similar.
DISCUSSION
We have characterized the properties of SS cells with
high VRD response to cell swelling to define its role in
dense-cell generation in sickle cell disease. This study
demonstrates (1) the function of KCl cotransport in
70
I
3
-
I
P
I
I
20
30!
10
-40
'
I
I
T
M
B
70
1
C B
0
Ln
T
M
B
'
-40
Fig 6. Absolute percent reticulocytes and F reticulocytes in three
density fractions separated in hypotonic gradient in six HbS patients.
T, top; M, middle; B, bottom. The cells in the bottom fraction are the
H-VRD-like population. (A) The upper panel shows the absolute
percent reticulocytes in the top (L-VRD-like), middle, and bottom
(H-VRD-like) populations of SS2 separated on a hypotonic CI- gradierror bars are standard
ent (M) and hypotonic NO,- gradient (0).
deviations. The percent reticulocytes were calculated as described in
the text. The bottom panel represents the difference (delta) between
the values in chloride and nitrate. ( 6 )Absolute percent reticulocytes
normalized to 100%. which are F reticulocytes in the same three cell
populations in hypotonic chloride ( W ) and hypotonic nitrate (0).The
bottom panel (delta) represents the difference between the values in
chloride and nitrate.
determining RBC density and (2) that only a fraction of
reticulocytes have high K C l cotransport activity that results
in the capacity for extremely rapid change in density. These
findings further expand our earlier studies of SS heterogeneity under isotonic ~ o n d i t i o n s . ~ ~ ~ ' ~ ~ ~
A fraction of reticulocytes and discocytes ( H - K W ) can
rapidly increase density. Our results show that the SS2
fraction, which contains most of the discocytes and half of
all reticulocytes, has a population of cells capable of rapid
dehydration when challenged with either hypotonic or acid
conditions. The rate of dehydration is unexpectedly large
and can be estimated by comparing the final density of
reticulocytes in hypotonic chloride and nitrate. From this
comparison we conclude that H-VRD cells dehydrate as
much as 3 g/dL x hour. Because SS2 contains about half of
all reticulocytes found in whole blood, the percent H-VRD
cells allows us to estimate that at least 45% of all reticulocytes are capable of rapid density changes. Are these
reticulocytes at greater risk of dehydration? In the patient
sample studied, the average percent dense cells in whole
blood is 22%, which exceeds the percent H-VRD reticulocytes. Therefore, we conclude that while the high K C l
cotransport of H-VRD cells puts them at risk of dehydration, other factors are also crucial at determining the final
fate of these cells. One such factor is the cell's history of
deoxygenation-related changes in cation content, and another factor may be level of Na+/H+exchange activity that
can counteract the decrease in volume.
The VRD response described here was elicited by
hypotonic swelling; however, we have previously reported
that acid-induced swelling6,' also results in generation of
denser cells. Recently Brugnara et alZ9reported increased
density of SS RBCs after a 12-hour exposure to acid
conditions. However, the cells that became denser were not
characterized and the involvement of the KCI cotransporter was not demonstrated by control experiments in
nitrate. More recently, Bookchin et a130has also confirmed
our ob~ervations~,~
that acid pH increases the density of SS
RBCs after exposure to pH 7.0 for 30 minutes.
Cells with a large increase in density have high K:Cl
cotransport activiv, We have shown that KC1 cotransport
In
and Na+/H' exchange are both very active in SS
this study, we show that H-VRD cells have higher K:Cl
cotransport, both volume-stimulated and NEM-stimulated,
than L-VRD cells. In contrast, Na'/H' antiport activity is
not significantly different in H-VRD and L-VRD cells. A
high Na'/H' antiport activity may potentially counteract
the shrinking effect of the K:Cl cotransporter if intracellular pH drops below 7.2." Hence, because the antiporter is
equally active in both H-VRD and L-VRD cells, shrinkage
in H-VRD cells will be less compensated than in L-VRD
cells. That Na+/H+antiport activity did not co-enrich with
KCI cotransport further illustrates the heterogeneous distribution of transport activity among cells. We speculate
that those cells with high K C l and high Na'/H+ antiport
activity may be partially protected from density increase.
As has been demonstrated in this report and suggested
p r e v i o ~ s l y , ~ ~this
~ ~ transport
"~~'
system is capable of altering
the MCHC of SS cells by several grams per deciliter per
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
224
hour under oxygenated conditions. However, we should
note that high KCl cotransport is not solely a property of
SS reticulocytes but is also a property of AA reticuloc y t e ~7*21,31
. ~ The density increase induced by swelling may be
involved in the normal increase in cell density that occurs
during reticulocyte maturation; however, it may occur at a
slower rate in vivo because, under our experimental conditions, the cells were exposed to hypotonic media for 1 hour.
The impact of high K:CI cotransport on SS cells is
different in SS than AA cells because in SS cells the
elevation of MCHC will strongly affect both the rate and
extent of polymer formation under deoxygenated conditions,‘ thereby contributing to the pathophysiology of the
disease. The volume-stimulated, chloride-dependent VRD
response will be activated by low pH6,’’ and decreased by
the increased intracellular ionized M$’ produced by deoxy g e n a t i ~ nconditions
,~
that are found in vivo in the kidney
and ischemic tissue. Other pathways may also contribute to
shrinkage of cells from sickle cell patients in both the
oxygenated and deoxygenated states; we and others have
demonstrated that a large portion of the deoxygenationstimulated potassium efflux is chloride inde~endent.~.~’
This
observation indicates that while part of the deoxygenationstimulated K’-efflux may occur via K:CI cotransport, other
pathways play a major role.
H-KUD cells are heterogeneous in cell age. Many H-VRD
cells are young cells, as demonstrated by a strong enrichment of reticulocytes. This is consistent with our6.” and
previous observation of the marked age dependence of the KCI cotransporter in HbS and HbA RBCs.
However, two observations indicate that H-VRD cells are
not exclusively reticulocytes and young cells, nor do all
reticulocytes and young cells belong in the H-VRD population: (1) On the average, approximately 21% of all reticulocytes in SS2 have either a small or nonexistent volume
response. ( 2 ) Only a small enrichment of the cell age
marker G-6-PD activity is found in the H-VRD population.
These observations demonstrate that the KCl cotransport
activity of reticulocytes is heterogeneous, and this may arise
by two possible mechanisms: (1) when cells enter the
circulation they may have varying levels of KCI cotransport
activity; or ( 2 ) the rate of decay of KC1 cotransport activity
may vary from cell to cell. The mechanism of loss of
transport activity is unknown, but it may be due to changes
in the biochemical processes (ie, phosphorylation and
dephosphorylation) involved in regulation33or it may be
due to loss of transport sites pari passu with the reduction
in cell membrane area and shedding of membrane proteins
that takes place during reticulocyte mat~rati0n.j~
Distribution of F reticulocytes and F cells is different in
H-YRLland L-YRLlcells. The percent HbF in H-VRD
cells was consistently lower than in L-VRD cells. It has
been suggested by several investigators that F cells are on
the average older than the other SS RBCs due to the
inhibition of HbS polymerization by HbF.L3,35
Therefore,
any preparative procedure that relies on a volume change
via a cell age-dependent transporter (such as KC1 cotransport) would be expected to yield proportionately fewer cells
with high HbF. We analyzed the distribution of F reticulocytes under the influence of hypotonic chloride conditions
FABRY ET AL
and found that both F and non-F reticulocytes have an
equal probability of migrating to higher density under the
influence of these conditions. In contrast, HbF and F cells
are concentrated in L-VRD. Therefore, although the distribution of KCl cotransport for F reticulocytes is similar to
that of other reticulocytes, F cells have a less active
volume-stimulated KCl cotransport, possibly because they
are on the average older cells, or, alternatively, because
K:CI cotransport decays more rapidly in F cells.
Concomitant a-thalassemia and sickle cell disease results
in a milder clinical course: hematologic parameters are
impro~ed,36”~
survival is prolonged,” and the percent of
dense cells is r e d ~ c e d . ’ ’Embury
~ ~ ~ et a14’ have reported a
correlation between cation fluxes under deoxygenated conditions and the a-globin gene number. Our flux studies
involved too few patients in each a-globin gene class to
allow such analysis; however, the enrichment of reticulocytes in the H-VRD fraction was nearly twice as large for
patients with all four a-globin genes as for patients with
a-thalassemia trait. This trend was not statistically significant, but is consistent with the greater percent dense cells
found in patients without a-thalassemia.” A simple correlation between dense cells and H-VRD cells would not be
expected because the percent dense cells reflects a dynamic
balance between the rate of production and the rate of
destruction.
The interaction of KCI cotransport and polymerization
thus appears to be part of a vicious cycle whereby destruction of dense and irreversibly sickled cells leads to a young
RBC population, which is characterized by high activity of
transport proteins that are normally absent or inactive in
the normal mature human RBC. Our results are in agreement with the early proposal of Bertles and Milner,I3which
was recently restated by Bookchin et al? that at least some
reticulocytes may become ISCs rapidly. In the study reported here, we further dissect this phenomenon by finding
that only a fraction of reticulocytes in SS2 (79%) can
dehydrate rapidly, and that the capacity for becoming dense
is also retained in a significant fraction of older RBCs. We
also show that the presence of a KCI cotransport-mediated
system that is stimulated by both hypotonic and acid
conditions results in the capacity for generating denser
cells.
In summary, we have identified a subset of SS reticulocytes and discocytes that are capable of volume reduction
when swollen. This subset of cells (H-VRD) has a high
percent F and non-F reticulocytes, a low percent HbF,
coupled with high K:Cl cotransport activity and average
Na’/H’ exchange activity. KC1 cotransport is unequally
distributed among reticulocytes with 79% of all reticulocytes in SS2 exhibiting high K C l cotransport; however, the
majority of H-VRD cells (80%) are not reticulocytes, but
are discocytes that retain high K C l cotransport. Our
findings suggest that dense SS cells (and possibly ISCs)
might be generated shortly after egress from the marrow
from the subpopulation of SS cells with high KCl cotransport.
ACKNOWLEDGMENT
The helpful discussions of T.L. Fabry and skillful technical
assistance of Fanya Schonbuch are gratefully acknowledged.
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
225
VOLUME REGULATION IN SICKLE CELLS
REFERENCES
1. Coletta M, Hofrichter J, Ferrone FA, Eaton WA: Kinetics of
sickle haemoglobin polymerization in single red cells. Nature
300:194,1982
2. Eaton WA, Hofrichter J: Hemoglobin S gelation and sickle
cell disease. Blood 70:1245,1987
3. Noguchi CT,Torchia DA, Schechter AN: Determination of
sickle hemoglobin polymer in SS and AS erythrocytes. Blood Cells
8:225,1982
4. Brugnara C, Bunn HF, Tosteson DC: Regulation of erythrocyte cation and water content in sickle cell anemia. Science
232:388, 1986
5 . Canessa M, Spalvins A, Nagel RL: Volume-dependent and
NEM-stimulated K+,Cl- transport is elevated in oxygenated SS,
SC, and CC human red cells. FEBS Lett 200:197,1986
6. Canessa M, Fabry ME, Blumenfeld N, Nagel RL: Volumestimulated, Cl( -)-dependent K+ efflux is highly expressed in
young human red cells containing normal hemoglobin or HbS. J
Membr Biol97:97,1987
7. Canessa M, Fabry ME, Nagel RL: Deoxygenation inhibits the
volume-stimulated, C1( -)-dependent K+ efflux in SS and young
AA cells: Acytosolic Mg2+ modulation. Blood 70:1861,1987
8. Canessa M, Fabry ME, Spalvins A, Blumenfeld N, Nagel R L
Activation of a K:Cl cotransporter by cell swelling in HbAA and
HbSS red cells. Prog Clin Biol Res 240:210,1987
9. Mahnesmith RL, Aronson PS: The plasma membrane sodium/
hydrogen exchange and its role in physiologicaland pathophysiological processes. Circ Res 56:773,1985
10. Grinstein S, Rohstein A Mechanisms of regulation of the
Na+/H+ exchanger. J Membr Biol901,1986
11. Canessa M, Fabry ME, Suzuka SM, Morgan K, Nagel R L
Na+/H+ exchange is increased in sickle cell anemia and young
normal cells. J Membr Biol 116:107, 1990
12. Fabry ME, Mears JG, Pate1 P, Schaefer Rego K, Carmichael
LD, Martinez G, Nagel R L Dense cells in sickle cell anemia: The
effects of gene interaction. Blood 64:1042,1984
13. Bertles JF, Milner P F A Irreversibly sickled erythrocytes: A
consequence of the heterogeneous distribution of hemoglobin
types in sickle-cell anemia. J Clin Invest 47:1731,1968
14. Fabry ME, Buchanan ID, Nagel R L Red cell density and
rate of potassium loss are affected by hemoglobin F levels in sickle
cell anemia. Clin Res 64:551A, 1984 (abstr)
15. Singer K, Singer L Studies on abnormal hemoglobins. VIII.
The gelling phenomenon of sickle cell hemoglobin: Its biologic and
diagnostic significance. Blood 8:1008,1953
16. Nagel RL, Bookchin RM: Human hemoglobin mutants with
abnormal oxygen binding. Semin Hematol11:385,1974
17. Sewchand LS, Johnson CS, Meiselman H J The effect of
fetal hemoglobin on the sickling dynamics of SS erythrocytes.
Blood Cells 9347,1983
18. Dover GJ: Individual variation in the production and survival of F cells in sickle cell anemia. N Engl J Med 299:1428, 1978
(abstr)
19. Canessa M, Spalvins A, Nagel RL: Volume-dependent and
NEM-stimulated K+CI- transport is elevated in SS, SC and CC
human red cells. Biophys J 49:580a, 1986
20. Brugnara C, Tosteson D C Cell volume, K+ transport, and
cell density in human erythrocytes. Am J Physiol252:C269,1987
21. Hall AC, Ellory JC Evidence for the presence of volumesensitive KCl transport n ‘young’ human red cells. Biochim
Biophys Acta 858:317,1986
22. Ostrow DG, Canessa M, Sparks S, Davis GM: Preservation
of human erythrocytes for sodium-dependent lithium efflux rate
measurements. Clin Chim Acta 129:39,1983
23. Canessa M, Brugnara C, Cusi D, Tosteson D C Modes of
operation and variable stoichiometry of the furosemide-sensitive
Na and K fluxes in human red cells. J Gen Physiol87:113,1986
24. Semplicini A, Spalvins A, Canessa M: Kinetics and stoichiometry of the human red cell Na+/H+ exchanger. J Membr Biol
107:219,1989
25. Papayannopoulou T, Brice M, Stamatoyannopoulos G: Hemoglobin F synthesis in vitro: Evidence for control at the level of
primitive erythroid stem cells. Proc Natl Acad Sci USA 74:2923,
1977
26. Dover GJ, Boyer SH: Hemoglobin determinations in single
cells: Comparison of different techniques. Prog Clin Biol Res
60:115,1981
27. Fabry ME, Nagel R L Heterogeneity of red cells in the
sickler: A characteristic with practical clinical and pathophysiological implications. Blood Cells 8:9, 1982
28. Kaul DK, Fabry ME, Windisch P, Baez S, Nagel R L
Erythrocytes in sickle cell anemia are heterogeneous in their
rheological and hemodynamic characteristics. J Clin Invest 72:22,
1983
29. Brugnara C, Van Ha T, Tosteson DC: Acid pH induces
formation of dense cells in sickle erythrocytes. Blood 74:487, 1989
30. Bookchin RM, Ortiz OE, Lew V L Evidence for a direct
reticulocyte origin of dense red cells in sickle cell anemia. J Clin
Invest 87:113,1991
31. Canessa M, Fabry ME, Spalvins A, Blumenfeld N, Nagel
R L A volume-dependent KCI transporter is highly expressed in
young human red cells with AA and SS hemoglobin. J Gen Physiol
88:14a, 1986 (abstr)
32. Berkowitz LR, Orringer EP: Passive sodium and potassium
movements in sickle erythrocytes. Am J Physiol249:C208,1985
33. Jennings ML, AI Rohil N Kinetics of activation and inactivation of swelling-stimulated K+/CI- transport: The volume sensitive parameter is the rate constant for inactivation. J Gen Physiol
95:1021,1991
34. Johnstone RM, A h n J: A common mechanism may be
involved in the selective loss of plasma membrane functions during
reticulocyte maturation. Biomed Biochim Acta 49:S70,1990
35. Elseweidy MM, Abraham EC: The relationship between
fetal hemoglobin level and glycosylation in sickle cell disease. Am J
Hematol 16:375,1984
36. Embury SH, Dozy AM, Miller J, Davis JR, Kleman KM,
Preisler H, Vichinsky E, Lande WN, Lubin BH, Kan YW, Mentzer
WC: Concurrent sickle-cell anemia and alpha-thalassemia: Effect
on severity of anemia. N Engl J Med 306:270,1982
37. Higgs OR, Aldridge BE, Lamb J, Clegg JB, Weatherall DJ,
Hayes RJ, Grandison Y, Lowrie Y, Mason KP, Serjeant BE,
Serjeant GR: The interaction of alpha-thalassemia and homozygous sickle-cell disease. N Engl J Med 306:1441,1982
38. Steinberg MH, Hebbel RP: Clinical diversity of sickle cell
anemia: Genetic and cellular modulation of disease severity. Am J
Hematol 14:405,1983
39. De Ceulaer K, Higgs DR, Weatherall DJ, Hayes RJ,
Serjeant BE, Serjeant G R Alpha-thalassemia reduces the hemolytic rate in homozygous sickle-cell disease. N Engl J Med
309:189,1983
40. Embury SH, Clark MR, Monroy G, Mohandas N: Concurrent sickle cell anemia and alpha-thalassemia: Effect on pathological properties of sickle erythrocytes. J Clin Invest 73:116,1984
41. Embury SH, Backer K, Glader BE: Monovalent cation
changes in sickle erythrocytes: A direct reflection of alpha-globin
gene number. J Lab Clin Med 106:75, 1985
From www.bloodjournal.org by guest on June 16, 2017. For personal use only.
1991 78: 217-225
Rapid increase in red blood cell density driven by K:Cl cotransport in
a subset of sickle cell anemia reticulocytes and discocytes
ME Fabry, JR Romero, ID Buchanan, SM Suzuka, G Stamatoyannopoulos, RL Nagel and M
Canessa
Updated information and services can be found at:
http://www.bloodjournal.org/content/78/1/217.full.html
Articles on similar topics can be found in the following Blood collections
Information about reproducing this article in parts or in its entirety may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#repub_requests
Information about ordering reprints may be found online at:
http://www.bloodjournal.org/site/misc/rights.xhtml#reprints
Information about subscriptions and ASH membership may be found online at:
http://www.bloodjournal.org/site/subscriptions/index.xhtml
Blood (print ISSN 0006-4971, online ISSN 1528-0020), is published weekly by the American
Society of Hematology, 2021 L St, NW, Suite 900, Washington DC 20036.
Copyright 2011 by The American Society of Hematology; all rights reserved.