[13] Sugar transport in red blood cells

[13] Sugar transport in red blood cells

[13] SUGAR TRANSPORT IN RED BLOOD CELLS 231 less. This is due to the relatively lengthy incubations that are required for the introduction of the p...

1MB Sizes 5 Downloads 87 Views

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

231

less. This is due to the relatively lengthy incubations that are required for the introduction of the probe molecules and subsequent phospholipase treatments of the cells. The method has been successfully applied to determine phosphatidylcholine flip-flop rates under a variety of conditions in both normal and pathologic erythrocytes, such as sickle cells 25,26and hereditary pyropoikilocytes. 27 The method fulfills the requirement that the discrimination between the inner and outer pools of phosphatidylcholine is complete and that the initial distribution of the probe molecule is not disturbed by subsequent treatment of the cells with phospholipases. This is illustrated by the observation that, immediately after its insertion, 96% or more of the newly introduced [~4C]phosphatidylcholine is detected in the outer membrane layer. 8 In principle, the method is not necessarily restricted to phosphatidylcholine. Using a nonspecific phospholipid transfer protein, trace amounts of radiolabeled phosphatidylethanolamine, phosphatidylserine, and sphingomyelin can be also introduced into the outer membrane layer of the intact (human) erythrocyte. 28 It should be stressed again, however, that also in those cases the bulk phospholipid in the donor vesicles should consist of egg phosphatidylcholine. 25 p. F. H. Franck, D. T.-Y. Chiu, J. A. F. Op den K a m p , B. Lubin, L. L. M. van D e e n e n , and B. Roelofsen, J. Biol. Chem. 258, 8435 (1983). 26 p. F. H. Franck, E. M. Bevers, B. H. Lubin, P. Comfurius, D. T.-Y. Chiu, J. A. F. Op den K a m p , R. F. A. Zwaal, L. L. M. van Deenen, and B. Roelofsen, J. Clin. Invest. 75, 183 (1985). 27 p. F. H. Franck, J. A. F. Op den K a m p , B. Lubin, W. Berendsen, P. Joosten, E. Bri~t, L. L. M. van Deenen, and B. Roelofsen, Biochim. Biophys. Acta 815, 259 (1985). 28 L. Tilley, S. Cribier, B. Roelofsen, J. A. F. Op den K a m p , and L. L. M. van D e e n e n , FEBS Left. 194, 21 (1986).

[13] S u g a r T r a n s p o r t in R e d B l o o d Cells

By W. F. WIDDAS Introduction Sugars are very hydrophilic molecules and as such are quite unsuitable to cross the bimolecular lipid layers which form the basis of cell membranes but which are readily permeated by hydrophobic substances. Nevertheless, it is essential to life in mammals that glucose should enter cells METHODS IN ENZYMOLOGY, VOL. 173

Copyright © 1989by AcademicPress, Inc. All fights of reproduction in any form reserved.

232

RED BLOOD CELLS

[13]

to serve as a substrate for metabolism. Evidence has accumulated that glucose does penetrate cell membranes, but only by a special mechanism involving components of the cell membrane with which glucose (and other monosaccharides) react and which in some way facilitate the transport across the membrane. Work to study and characterize this special mechanism has chiefly been done using human red cells because they are a convenient experimental material; they also have the peculiarity of possessing a very rapid transfer rate for glucose (and closely related sugars such as 2-deoxyglucose, 3-O-methylglucose, mannose, and galactose) so that the sugar in the cells equilibrates with that in the medium. By contrast, the red blood cells of common laboratory and farm animals which one might naturally turn to as an experimental source have rates of transfer only about one-thousandth that in human red cells, and due to metabolism of the glucose which does penetrate they fail to equilibrate with the outside medium. Although this is the situation with adult laboratory animals, Widdas I showed that the red cells from fetal and newborn animals of several species have rates of transfer comparable with human blood. Red cells from fetal guinea pigs have a faster rate than human red cells but have broadly similar properties (Aubby and Widdas2).

Source o f Red Cells Human Red Cells. Fresh human red cells obtained by sterile venepuncture of the antecubital vein are ideal if adequate numbers of volunteers are available. Failing this, most laboratories have access to blood from blood banks, which are the best source when very large quantities are required for the experiments planned. The blood usually available for laboratory use is "time expired" and in such blood the rates of sugar transfer are reduced to a large and variable degree--it may be as low as half that in fresh blood. This variability is a drawback in kinetic experiments and attempts have been made to restore a more normal activity. A method used by Ginsburg 3 for a number of years consists of incubating a 10% erythrocyte suspension for 60-90 min at 37 ° in a medium consisting of NaC1 (120 mM), KCI (5 mM), NaH2PO4 (10 mM), glucose (5 mM), and inosine (15 mM), pH 7.6. The present author has confirmed the rejuvenating effect of this treatment for time-expired blood and it is recommended where the red cells are to be used for kinetic studies. W. F. Widdas, J. Physiol. (London) 127, 318 (1955). z D. S. Aubby and W. F. Widdas, J. Physiol. (London) 309, 317 (1980). 3 H. Ginsburg, personal communication (1985).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

233

Fetal Red Cells. Fetal and newborn guinea pigs have been extensively used in the author's laboratory. Newborn rabbits have been used by Augustin et al. 4 and newborn dog red cells by Lee et al. 5 Other Red Cells. Apart from the above cells, which have very fast transport rates, some studies have been made with cells having very slow rates. Chief among these have been erythrocytes from adult rabbits 6,7 and avian erythrocytes as used by Wood and Morgan. 8 Preparation of Red Cells Anticoagulants. Fresh blood tends to clot and some form of anticoagulant is needed. Citrate, 9 heparin, ~ and EDTA (1 mg/ml) 1° are all suitable, as is the acid-citrate-dextrose (ACD) mixture used in the Blood Transfusion Service. Alternatively, the blood may be defibrinated by stirring with a glass loop I and then by filtering through glass w o o l . 7 Heparinized unwashed blood can readily be kept up to 48 hr in a refrigerator at 4° and used for repeat experiments. Suspending Media. Saline media buffered with either phosphate t,9 or Tris ~j are used to dilute and wash the red cells prior to use. After the first centrifugation the buffy coat, which contains the white cells, is carefully removed and discarded. If the first dilution is about 5-fold this will reduce the normal (5 mM) blood glucose to 1 raM; two further dilutions of this order will reduce the inside concentration to under 0.1 mM, which is low enough for most experimental purposes, but some investigators have used a fourth wash to ensure that the residual glucose is insignificant. A typical suspending medium for human red cells would be NaC1 (119 raM), KCI (4.8 mM), CaClz (2.6 mM), MgCI2 (1.7 mM), Tris (35 raM), pH 7.4; tonicity, 300 mOsm. j~ For rabbit red cells the medium used was NaCI (137 mM), KCI (5.9 mM), CaCl2 (1.3 mM), MgCl2 (2.4 mM), KHzPO4 (1.2 mM), imidazole (4.2 mM), glycylglycine (7.6 mM), bovine serum albumin (0.2%), pH 7.35. 7

4 H. W. Augustin, L. van Rohden, and M. R. H~icker, Acta Biol. Med. Ger. 19, 723 (1967). 5 p. Lee, J. Auvil, J. E. Grey, and M. Smith, Fed. Proc., Fed. Am. Soc. Exp. Biol. 35, 780 (1976). 6 H. E. Morgan, C. F. Kalman, R. L. Post, and C. R. Park, Fed. Proc., Fed. Am. Soc. Exp. Biol. 14, 336 (1955). 7 D. M. Regen and H. E. Morgan, Biochim. Biophys. Acta 79, 151 (1964). 8 R. E. Wood and H. E. Morgan, J. Biol. Chem. 224, 1451 (1969). 9 p. G. LeFevre, J. Gen. Physiol. 31, 505 (1948). 10 R. D. Taverna and R. G. Langdon, Biochim. Biophys. Acta 298, 412 (1973). H p. G. LeFevre, Biochim. Biophys. Acta 120, 395 (1966).

234

RED BLOOD CELLS

[13]

Experimental Techniques In general there are three ways of investigating sugar transport in red cells but these are not usually applicable to the same type of problem. Consideration must be given both to the object of the experiment and to the source of red cells it is proposed to use when planning the most appropriate experimental approach. The three main techniques are (1) standard or modified biochemical techniques, (2) optical methods involving osmotic swelling or shrinking of erythrocytes, and (3) radioactive techniques. Standard or Modified Biochemical Techniques The biochemical estimation of sugars involves preparing a protein-free extract and, if the cellular concentration is required, a centrifugation to separate the cells is also needed. Centrifugation takes a time which is long relative to the rates of transport of sugars by ceils with fast transfer rates. Thus biochemical methods were unable to make useful contributions to the kinetics of glucose transport in human red cells. They have, however, been used in important work on sugar transport in rabbit red cells. The estimation of sugars as reducing agents is not sufficiently specific but glucose determined enzymatically12is more useful. However, in competitive studies the need arises to estimate two sugars in the presence of each other. Park and Johnson 13 used paper chromatography to separate the sugars and by eluting the spots and developing with a suitable reagent (e.g., aniline hydrogen phthalate 14for galactose) were able to estimate the sugars quantitatively. Morgan et al. 6 created conditions in rabbit red cells whereby either influx or efflux could be measured independently. They used 10-2 M NaF to inhibit utilization so that glucose was accumulated to the required amount. The cells were then washed free of external glucose at 0° and the efflux measured into a medium containing yeast hexokinase. Influx was measured into cells in the presence of 2.1 × 10-5 M Methylene Blue, which promoted phosphorylation of the glucose that penetrated, so that the internal free glucose concentration was kept less than 0.1% of the outside concentration. In this way experiments to determine the influx as a function of the outside concentration and efflux as a function of the internal concentration were carried out. These showed that in both directions a saturable transport mechanism was involved similar to 12 M. W. Slein, G. T. Coil, and C. F. Coil, J. Biol. Chem. 186, 763 (1950). 13 C. R. Park and L. H. Johnson, Am. J. Physiol. 182, 17 (1955). 14 S. M. Partridge, Nature (London) 164, 443 (1949).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

235

that shown to exist in human red cells by optical experiments but about a thousand times slower than in human cells. Using rabbit red cells Park et al) 5 were the first to demonstrate an important property of facilitated transport which has been termed "uphill transport induced by c o u n t e r f l o w . ''16 It was predicted by W i d d a s 17 that competition for a "carrier" transfer could result in one sugar being transferred against its concentration gradient if a competitive sugar was transferring down its own concentration gradient. Park et al.15 set out to test this prediction using xylose and glucose as the competing sugars. The rabbit red cells were first equilibrated with xylose, and since this is a nonmetabolizing sugar the cells approached equilibrium after a 2 hour incubation. Glucose was then added to the outside medium, and since this sugar on entering the cells was removed by metabolism its concentration gradient, inward, was maintained. The concentration of xylose in the cells was found to be reduced although its outward movement was against the concentration gradient for that sugar. The results of this now classic experiment are shown in Fig. I. It is important since the phenomenon was a prediction from the kinetics of a mobile carrier and could not arise from a mechanism involving only simple diffusion. Although the mobile carrier model may be an oversimplification of the sugar transport mechanism which is now attributed to a transmembrane protein, the phenomenon of uphill transfer by counterflow could only arise if there was some sort of conformational change within the protein which provided the equivalent to movement in the mobile carrier model. Optical Methods Involving Osmotic Swelling or Shrinking It is common experience that suspensions of intact red cells are opaque but the same suspension becomes a colored transparent liquid if the cells are hemolysed either chemically or osmotically. The opacity of intact cell suspensions is due to the higher refractive index of the inside medium containing hemoglobin at high concentration. On hemolysis the hemoglobin becomes uniformly distributed and light does not suffer refraction at the cell boundaries. The change in light transmission with hemolysis is large and can readily be detected with the naked eye and a convenient light source. In experiments the time for hemolysis can be noted with a stopwatch and if hemolysis is produced by the penetration of J5 C. R. Park, R. L. Post, C. F. Kalman, J. H. Wright, Jr., L. H. Johnson, and H. E. Morgan, Ciba Found. Colloq. Endocrinol. [Proc.] 9, 240 (1956). 16 T. Rosenberg and W. Wilbrandt, J. Gen. Physiol. 41, 289 (1957). 17 W. F. Widdas, J. Physiol. (London) 118, 23 (1952).

236

RED BLOOD CELLS

[13]

50 External Xylose x ~__.

4O . ~ . ~ . _ - - - 0 Cont r o I

30

'

Internal Xylo se

~0

- ~ X Glucose /

Glucose added

10

o

i

Fir

FIG. I. Uphill transport induced by counterflow in rabbit erythrocytes. O, Xylose concentration in control cells during a 4-hr incubation in a 40 rag/100 ml xylose medium, x, Xylose concentration in the cells after addition of glucose. Reduction in the internal xylose concentration showed that xylose had moved out into the xylose-containing medium although this was against the concentration gradient. (After Park et alJ s)

a nonelectrolyte from an isosmotic solution the speed with which hemolysis occurs is a measure of the rate of penetration of the nonelectrolyte concerned. This technique was used extensively by Jacobs and colleagues in the 1920s and 1930s and the quantitative aspects were described by Jacobs and StewartJ 8 However, it was observed by Masing j9 as long ago as 1914 that human red cells do not hemolyse in isosmotic glucose solutions, so the gross light changes produced by hemolysis are not available for following sugar penetration rates. The same light changes can be followed more accurately using photoelectric devices and 0rskov 2° observed that red cell volume changes which did not involve hemolysis could nevertheless be determined from the much smaller changes in light transmission which they caused. 18 M. H. Jacobs and D. R. Stewart, J. Cell. Comp. Physiol. 1, 71 (1932). 19 E. Masing, Pfluegers Arch. Gesamte Physiol. Menschen Tiere 156, 401 (1914). 20 S. L. ~rskov, Biochem. Z. 279, 241 (1935).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

237

Parpart, zl working in Jacob's laboratory, had also developed photoelectric techniques for following red cell volume changes and this apparatus was used by LeFevre 9 to follow the changes due to glycerol and glucose penetration. Widdas 22 modified an apparatus to follow the osmotic hemolysis in fetal sheep red cells to make it sensitive enough to follow the small changes in light transmission obtained in glucose penetration experiments. Wilbrandt and co-workers 23 had also used photoelectric techniques but the precision of their method was greatly increased by Fuhrmann's development of an apparatus for directly measuring the scattering of light at right angles to the incident beam through the suspension. 24 The three techniques have differences which will be mentioned but full details of only the author's technique will be given. An account of the Fuhrmann technique is the subject of a later chapter in this volume. 25 LeFevre's Technique. The apparatus used by LeFevre in his earliest experiments has not been published in detail but his modified apparatus as used later was fully described in 1966. lj The cuvette, of about 15-ml capacity, was enclosed in a thermostatically controlled housing and illuminated by a specially regulated 6-V tungsten lamp. The light beam was split by a glass surface to provide a reference beam containing an adjustable glass wedge while the main beam went through the cuvette. The reference and cuvette beams were detected by two separate photomultipliers supplied from the same battery source. A cathode follower coupling circuit fed into a T-Y recorder. Suspensions of cells were made up to 1/300-1/200 relative to the volume of the medium. Typically, 12 ml of cell suspension was allowed to reach a steady level of light transmission and then up to 1.5 ml of medium containing sugar at 1.7 M was added. When the cells shrank the scattering of light was increased and the forward transmission was reduced. As reswelling of the cells occurred the light transmission increased. Runs of 10 to 15 min were recorded; stirring was not specified. Widdas's Technique. Widdas 22 developed a technique which differed from LeFevre's in certain respects. The cuvette was 2 cm wide (capacity 25 ml) and was especially constructed to be boot shaped. The light beam went through the " t o e " of the boot as a 2.5-cm-diameter beam of parallel 21 A. 22 W. 23 W. 54 G. z~ G.

K. Parpart, J. F. Widdas, J. Wilbrandt, E. F. F u h r m a n n , F. F u h r m a n n ,

Cell. Comp. Physiol. 7, 153 (1935). Physiol. (London) 120, 20P (1953). GuenSberg, and H. Lauener, Heir. Physiol. Acta 5, C20 (1947). P. Liggenstorfer, and W. Wilbrandt, Experientia 27, 1428 (1971). this v o l u m e [15].

238

RED BLOOD CELLS

[13]

light while in the " l e g " of the boot a paddle stirrer was driven by a synchronous motor at about 375 rpm. The light was divided into reference and main beams which passed through Polaroid filters set mutually at right angles. The two beams converged onto a single photocell in front of which was a third Polaroid filter held in a mounting which was rotated at 12.5 Hz. A shutter in the reference beam was used to get a null (as in LeFevre's technique) but the outof-balance signal (a sine wave of 25 Hz) was amplified with a low-bandwidth ac amplifier prior to phase-sensitive rectification and display on a recording galvanometer. The 21 ml of suspension containing only 3/zl of packed cells was first allowed to reach a steady state; then sugar dissolved in the same medium was added. For entry experiments 0.5 ml of 1.67 M sugar was added in four consecutive runs, allowing equilibration to occur before each new addition. The double-beam arrangement, single photocell, and ac amplification minimized zero drift and gave good long-term stability. A second apparatus using a chopped light source in place of the Polaroid filters has also been used. In recent years a silicon photodiode incorporating a color correction filter (visible radiation) and solid-state amplifiers have replaced the original photocell and valve amplifier. A chart recorder is used in place of the recording galvanometer, but the basic optical principles have remained unchanged for over 30 years. Exit Experiments. The very low hematocrit (the cells occupied only 1 part in 7000 of the medium) used by Widdas made the technique suitable for adaptation to follow glucose exits from cells. 2627 For an exit it is not practicable to add 3/~l of packed cells to 21 ml saline in the cuvette as the packed cells take too long to be dispersed homogeneously. It was found that a preliminary suspension in 0.15-0.2 ml of the glucose-containing medium was, however, rapidly dispersed when added to the cuvette and gave satisfactory exit records. In practice the cells are preincubated in these same proportions, i.e., 10.3 ml buffered saline, 0.5 ml 1.67 M glucose in buffered saline, 0.2 ml packed cells which have been previously washed. After incubation, long enough to reach equilibrium, aliquots of 0.15-0.2 ml are tested for exits and a suitable value in this range chosen. Prior to taking an aliquot the suspension is stirred for a few seconds on a vortex mixer to ensure uniform mixing and the absence of rouleaux. When the cells are incubated at 76 mM glucose the addition of 0.15 ml to the cuvette makes the outside concentration 0.6 mM as a minimum. 26 F. Bowyer and W. F. Widdas, J. Physiol. (London) 141, 219 (1958). 27 A. K. Sen and W. F. Widdas, J. Physiol. (London) 160, 392 (1962).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

239

However, this proved to be no serious drawback as the exit times are linearly related to the outside glucose concentration (up to about 12 mM) and the time for exit into a glucose-free solution could be obtained by extrapolation (see Figs. 2 and 3). Note, however, that at the very low hematocrit (1 in 7000) the exit of even 76 m M from the cells could only have changed the outside concentration by 0.01 mM, which is negligible. At the low hematocrit used in this technique the light transmission actually increased as the cells shrank and vice versa. The cells in the 2-cm light beam would not be quite sufficient to occupy the whole area if they could be brought into the same plane; and it may be presumed that cells which shrink allow more light to pass between them and this light is not subject to scattering. Fuhrmann's Technique. As in the two methods described above the cuvette is surrounded by a water jacket except for where the light enters and leaves (in this case at right angles to the entry beam). The cuvette holds 15 ml of cell suspension which has a hematocrit of about 1 in 1000. Stirring is by means of a magnetic "flea." The light scattered at right angles increases if the cells shrink and their refractive index is increased. In this respect the changes are qualitatively similar to those in the Widdas technique although the basis is probably different. Both are in contrast to what is seen in the LeFevre technique. For further details of the Fuhrmann technique, see Chapter [15] in this volume. 25

Results from Optical Experiments Although simple osmotic and biochemical techniques had highlighted anomalies in the apparent rates of uptake of glucose at different concentrations, 28 and Klinghoffer 29 had found that equilibration, which occurred readily at low glucose concentrations, was delayed indefinitely at isosmotic concentrations, it was the optical techniques which first allowed the rates of uptake to be displayed and measured. The characteristics of the uptake curves for glucose led LeFevre 9 to conclude that simple diffusion could not be responsible and he took the important step in suggesting that transfer may involve the temporary formation of a complex with some constitutent of the cell membrane. This idea led to the formulation of simple kinetics for the transfer process; the most successful of these have been based on the model suggested by Widdas, ~7 who proposed that the rate-limiting step was a R. Ege, thesis. University of Copenhagen, Copenhagen, Denmark [cited by O. Bang and S. L. Orskov, J. Clin. Invest. 16, 279 (1937)]. 29 K. A. Klinghoffer, Am. J. Physiol. 111, 231 (1935).

240

RED BLOODCELLS

[13]

slow transfer through the membrane while the reactions with the membrane constituents at the interfaces were deemed to be rapid. In its simplest form transfer through the membrane could therefore be represented by dS/dT

= Wmax{[(S/S +

K~)] - [ S ' / ( S ' + K~)]}

(1)

where Vmaxrepresents the rate when all the components on one side are saturated and there is no backflux. S and S' represent the outside and inside sugar concentrations and Ks is the half-saturation concentration equivalent to the Michaelis constant for an enzyme-catalyzed reaction. Widdas 17,3° showed that this equation had two approximations which could be applied when Ks was either very large or very small relative to the concentrations. Thus, if S << K~ >> S' dS/dT

= (Vmax/gs)(S -

S')

(2)

(I/S)]

(3)

but if S >> Ks << S' dS/dT

= VmaxK~[(1/S') -

Equation (2) is equivalent to diffusion, but Eq. (3), which represents a near-saturation condition, suggests that the rate will be proportional to the difference of the reciprocals of the concentrations. The integrated forms of Eqs. (2) and (3) were applied to the swelling curves of red cells in glucose and sorbose solutions by Widdas. 31 The integrated form of Eq. (1) was used by LeFevre 32 to estimate the Vmax and K~ for a number of different monosaccharides. The use of optical techniques to follow exits, either of high-affinity sugars such as glucose 27 or of such sugars in the presence of inhibitors, 33 has many advantages over their use for following entries, chief among these being the greater speed of the light transmission change and the ease of quantification. Figure 2 is a reproduction of the type of records obtained from glucose exits into a series of low outside concentrations. The traces on the recorder are mostly straight lines since the efflux may be considered to be maximal for a large part of the exit. There is, of course, a small constant influx due to the glucose in the outside medium and as this is increased the exit is slowed and takes longer. To measure the records the straight line part is produced until it intersects the equilibrium value and the exit time 3o W. F. Widdas, J. Physiol. (London) 115, 36P (1951). 3i W. F. Widdas, J. Physiol. (London) 125, 163 (1954). 32 p. G. LeFevre, Am. J. Physiol. 203, 286 (1962). 33 A. K. Sen and W. F. Widdas, J. Physiol. (London) 160, 404 (1962).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

O.7~mM '

241

I 4.6mM

6.5mM

8.4mM

12.2mM

l min '

FIG. 2. Tracings of records from the photoelectric apparatus during "exit" experiments at 37°. Cells preequilibrated in 76 mM were losing glucose into medium at the concentrations shown. The linear part of each record was produced to the base line and the times of exit measured for analysis as in Fig. 3. (From Sen and Widdas. 27)

read off. This represents the time the exit would have taken if it had continued at the initial linear rate. It is found empirically, and there is theoretical justification from the integrated equations, that the exit times are a linear function of the outside concentration. Lines drawn as in Fig. 3 have two intercepts. The one with the ordinate is the extrapolated time for exit into a glucose-free solution, and knowing the concentration at which the cells were preincubated it is possible to calculate the Vmax for the cells at that concentration. The intercept on the abscissa is the outside concentration which would double the exit time compared with that into a glucose-free solution. On a simple view it can be interpreted as the half-saturation for the components on the outside of the membrane but recent work suggests the intercept is an estimate of the Km for a modifier site on the glucose transporter. 34 The use of exit experiments in the presence of inhibitors can distinguish between those which are competitive for the outside sites and those which are noncompetitive. However, care must be taken since an inhibitor which is only competitive at the inside sites would appear to be noncompetitive in exit experiments. 35,36 Drawbacks to Optical Experiments. The need for specially constructed apparatus at one time limited the number of laboratories where optical experiments could be carried out. The minimum requirement now 34 G. F. Baker and W. F. Widdas, J. Physiol. (London) 395, 57 (1988). 35 D. A. Basketter and W. F. Widdas, J. Physiol. (London) 278, 389 (1978). 36 W. F. Widdas, Curt. Top. Membr. Transp. 14, 165 (1980).

242

RED BLOOD CELLS

[13]

1.0EO.8E "0.6"~ tu 0.4-

0.2"

S

I

I

_ /

I

I

I

12

8

/

4

8

12

Conc. outside

(mM)

4

I

FIG. 3. "Exit" times obtained from records such as in Fig. 2 plotted against the outside concentration. The line (drawn by eye) has two intercepts: that on the ordinate represents the time for exit into a glucose-free medium (to) while that on the abscissa is the concentration which would have an exit time twice to. (From Sen and Widdas.27) is p r o b a b l y a specially a d a p t e d cuvette and w a t e r jacket, since Bloch37was able to use a c o m m e r c i a l l y available s p e c t r o p h o t o m e t e r and recorder. Carruthers and Melchior 38 also carried out exit experiments with a modified c o m m e r c i a l instrument. T h e swelling and shrinking of red cells call for changes in the osmotic contents relative to the m e d i u m and changes in concentration of the order of 20-30 m M are p r o b a b l y required as a minimum. Since the normal blood glucose is only 5 m M , the criticism could be levelled that the concentrations used are unphysiological. Also, since the volume changes are needed to p r o d u c e refractive index changes the methods are best used with intact red cells. Partially h e m o l y s e d red cells and resealed ghosts h a v e b e e n used in special cases (e.g., Carruthers and Melchior38). E x c h a n g e fluxes, w h e r e there is neither a volume nor a concentration change, are not m e a s u r a b l e b y either biochemical or optical techniques and it is in this field that important discoveries h a v e b e e n made using the third technique to be described and which involves the use of radioactively labeled sugars. 37R. Bloch, Biochemistry 12, 4799 (1973). 38A. Carruthers and D. L. Melchior, Biochirn. Biophys. Acta 728, 254 (1983).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

243

Radioactive Techniques 14C-Labeled glucose first became generally available in the 1950s and since then a wide variety of labeled sugars and sugar derivatives have been produced. Labeling with tritium is also obtainable and since scintillation spectrometers permit the separate determination of the tritium and 14C label in mixtures there are now many experimental possibilities, not all of which have so far been exploited. To use radioactive sugars to measure rates of transport across the red cell membrane, however, requires that a number of difficulties be overcome. Most of these arise from the great rapidity of the transfer process, the need to separate the cells from the suspending medium, and the need to deproteinize the samples. These are the same limitations as were mentioned when considering the application of biochemical techniques to red cells which have rapid rates of sugar transfers; but the increased knowledge of the transfer mechanism, gained largely from the use of optical experiments, has provided us with ways of meeting these difficulties. Temperature Effects. Although sugar transfers are very fast at 37° the process has a moderately high Q~0 of about 2.5 and consequently it is advantageous to work at lower temperatures. Many laboratories use 20° but a lower temperature of 16° was used by Baker and Widdas 39 and became the temperature of choice in the author's laboratory. Taking the temperature down to 0° caused sufficient slowing to permit the separation of rabbit red cells from their suspending media in the experiments of Morgan et al., 6 but this would not be enough for human red cells. Stopping Solutions. To prevent changes in the concentration in human red cells during centrifugation, the effect of temperature is usually combined with the use of a"stopping solution" containing powerful inhibitors of the glucose transfer mechanism. The stopping solutions are based on the inhibition produced by mercuric ions but Levine and Stein4° found empirically that the effect was improved in the presence of 1.25 mM KI, whereas Karlish et al. 4~ used a lower concentration of mercuric ions but included the inhibitor phloretin. Their stopping solution had the following composition: NaC1 (1%, w/v), HgCI2 (10 -6 M ) , KI (1.25 raM), phloretin, dissolved in ethanol to give 10-4 M phloretin, and 1% ethanol. The stopping solution used in the author's laboratory had the follow-

39 G. F. Baker and W. F. Widdas, J. Physiol. (London) 231, 143 (1973). 40 M. Levine and W. D. Stein, Biochim. Biophys. Acta 127, 179 (1966). 41 S. J. D. Karlish, W. R. Lieb, D. Ram, and W. D. Stein, Biochim. Biophys. Acta 255, 126 (1972).

244

RED BLOOD CELLS

[13]

ing composition: NaCI (0.34 M), KI (0.65 mM), HgClz (2 mM), phloretin (0.1 mM). Alternatives to the use of stopping solutions have been reported. Wilbrandt and Kotyk 42 centrifuged the suspension rapidly at 0 ° over silicone oil through which the cells passed. Mawe and Hempling 43 separated the cells from the medium by a rapid filtration through Millipore filters. Sampling Procedures. The procedures for taking and processing samples used in the author's laboratory from 1971 were based on those described by Miller. 44 Typically, samples are taken at appropriate intervals from the experimental suspension in a 1-ml automatic syringe and injected into 10 ml of ice-cold stopping solution in a conical tube. The tubes are kept on ice until they are centrifuged, which is done as soon as possible. After centrifugati0n the supernatant is carefully withdrawn so as not to disturb the cell pellet. A further 2 ml of ice-cold stopping solution is introduced from an automatic burette to wash down the sides of the tube, which is again centrifuged. After removal of the supernatant the cell pellet is resuspended in 50/xl of saline buffer with vigorous stirring on a vortex mixer while 1.2 ml of absolute alcohol is added. The cell debris is centrifuged and 1 ml of the supernatant alcoholic solution (containing labeled sugar) is carefully pipetted into a counting vial containing 10 ml of scintillation fluid. The scintillation fluid used was 0.4% (w/v) 2,5-diphenyloxazole and 0.02% (w/v) 1,4-D-[2-(5-phenyloxazolyl)]benzene in a 2 : 1 mixture of toluene and Triton X-100. Some laboratories have preferred to use 0. I ml 20% trichloroacetic acid solution in place of absolute alcohol (e.g., Karlish et al.41). The procedures outlined above presuppose that the same quantities of red cells are taken up by the automatic syringe at each sampling time. To ensure this the suspension must be adequately stirred; a Perspex apparatus was made to surround a 30-ml conical flask with a water jacket which was mounted over a magnetic stirrer so that a magnetic flea in the flask was rotated at 10 Hz. A second Perspex apparatus was designed so that the tubes holding the stopping solution were held in a repeatable position relative to the automatic syringe when samples were injected but so that stopping solution did not splash up onto the needle of the syringe. With practice, samples at 5-sec intervals could be taken but longer intervals are preferable if the experimental design permits. Treating the Results. Usually five or six samples are taken during the 42 W. Wilbrandt and A. Kotyk, Arch. Exp. Pathol. Pharmakol. 249, 279 (1964). 43 R. C. Mawe and H. G. Hempling, J. Cell. Cornp. Physiol. 66, 95 (1965). 44 D. M. Miller, Biophys. J. 8, 1329 (1968).

[13]

SUOAR TRANSPORT IN RED BLOOD CELLS

245

first 60-90 sec and two further samples are taken when it is judged the cells will h a v e c o m e to equilibrium, which m a y be from 4 to 20 min, depending on the type of e x p e r i m e n t and the presence or absence of inhibitors. The radioactivity initially used is adjusted so that o v e r the counting period (say 20 min) about 10,000 counts are recorded. The counts for the final equilibrium samples m a y be less if in the experiment radioactively is effluxing f r o m the cells. The standard error of the counts is equal to the square root of the total counted and at this level is about 1% and p r o b a b l y less than other experimental errors in the technique. In an equilibrium exchange situation where the concentrations are the same inside and out but the radioactivity starts inside (and effluxes during the experiment), the conditions are ideal in that there are neither chemical concentration nor volume changes to be considered. The radioactively labeled sugar which is leaving the cells is being replaced by an equal amount of unlabeled sugar from the outside medium which is relatively large in volume. U n d e r these circumstances the radioactivity in the samples should fall off exponentially to the final equilibrium value. This m a y be represented as - 2 . 3 0 3 log(X - X~) = kt + constant

(4)

where X is the n u m b e r of counts per minute in the sample at time t (seconds) and X= is the n u m b e r when equilibrium has been reached. Note that by subtracting these two terms the effect of any radioactivity outside the cells is also canceled out. Figure 4 taken from the results of B a k e r and

3.0

2.0

I£ 0

20

LO

60

BO

FIG. 4. Logarithmic fall-off of intracellular radioactivity during glucose exchange at 20 mM (16°) in a control experiment and in the presence of the nontransportable inhibitor ethylidene glucose. Samples taken over 60 sec in the control and 80 sec when inhibited. O, Control; ×, with 50 mM inhibitor inside the cells; C), with 200 mM inside; A, with 25 mM outside the cells. The slopes of such lines were used to derive the exchange flux rates as described in the text. (After Baker and Widdas.39)

246

RED BLOODCELLS

[13]

Widdas 39 illustrates the manner of plotting the results according to Eq. (4). Once the slope (k) of the line has been determined and irrespective of the exact volume of the cells used in the samples (provided that each sample contained a like volume), the flux represented by J is obtained from the relation J = 60kS

(5)

where S is the concentration of sugar in the cells. If S is expressed in millimolar units then the units of J are millimoles (liters cell water) -1 minute-~. Similar considerations apply to radioactive uptake experiments but in these the amount of radioactivity outside the cells will be much larger. The terms in parentheses in Eq. (4) will be reversed and the log will be positive. Although the outside radioactivity will tend to be canceled out, it is likely to give rise to greater variability between samples due to differences in washing down the centrifuge tubes and other preparative procedures. For this reason the use of radioactive efflux is to be preferred where possible, e.g., in exchanges since efflux and influx are then equal. An experiment where this is not possible and which is illustrated in Fig. 5 is one designed to show uphill transport induced by counterflow in human red cells. In this experiment cells were preincubated with 76 mM of a nonradioactive sugar and then placed in a medium containing 4 mM 14C-labeled 3-O-methylglucose at 16°. The nonradioactive sugars were glucose, 3-O-methylglucose, galactose, and sorbose. Samples were taken at the times shown and their radioactivity compared with that in the equilibrium samples when the inside concentration was the same as outside (4 mM). With glucose as the competitive sugar it will be seen that at 60 sec the inside 3-O-methylglucose had risen to over 9-fold its equilibrium value and must have been entering the cells against its concentration gradient. The other sugars were less effective in inducing the uphill transfer and the ketose sugar sorbose did not appear to induce any. The experiments in Fig. 5 were carried out by final-year honor students between 1971 and 1981 as laboratory exercises and illustrate the power and the relative simplicity of the radioactive techniques when applied to sugar transport problems. Another more difficult application of radioactive techniques has been to measure initial uptake rates under a variety of experimental conditions (e.g., Lacko e t a/.45,46). Here the uptake must be sampled at very short 45L. Lacko, B. Wittke, and P. Geck, J. Cell. Physiol. 80, 73 (1972). 46L. Lacko, B. Wittke, and P. Geck, J. Cell. Physiol. 82, 213 (1973).

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

247

Ratio

mM

50

20,

10-

Sorbose 0

I

0

:30

60

I

I

I

I

I

I

90

120 150 180 210 240 Sec FIG. 5. Uphill transport induced by counterflow in human red cells at 16°. Cells were preincubated at 76 mM in the nonradioactive sugars shown and then rapidly introduced into a medium containing 4 mM 14C-labeled3-O-methylglucose. By comparing the intracellular radioactivity at times up to 4 min with that in the cells when equilibrium had been reached the ability to induce against the gradient uptake of 3-O-methylglucose was demonstrated (except for the case of sorbose). The means and standard errors are given for over 24 experiments for each sugar carried out by honors students in the author's laboratory between 1971 and 1981.

t i m e s s i n c e if a n y r a d i o a c t i v e l y l a b e l e d sugar is a l l o w e d to a c c u m u l a t e in the cell it s o o n has a n effect u p o n the influx rate. W i t h the a v a i l a b i l i t y o f r a d i o a c t i v e l y l a b e l e d sugars of high specific a c t i v i t y it has b e e n p o s s i b l e to a d a p t fast-flow t e c h n i q u e s to m e a s u r e s u g a r fluxes. 47,48 S u c h t e c h n i q u e s c a n b e u s e d to s t u d y rates o f t r a n s p o r t 47j. Brahm, J. Physiol. (London) 339, 339 (1983). 4s A. G. Lowe and A. R. Walmsley, Biochim. Biophys. Acta 857, 146 (1986).

248

RED BLOOD CELLS

[13]

at body temperature, or above, without some of the difficulties described for other radioactive procedures. At present the experiments are limited to measurements of rates of efflux from cells loaded with radioactive sugars at various concentrations. However, the outside concentration of nonradioactive sugar can be varied. If it is the same as that in the cells, the exchange efflux is determined. At the other extreme, efflux rates into a sugar-free medium can be measured. The fast-flow technique does not lend itself to the measurement of influx rates. Reference should be made to the original articles of these and other related studies before undertaking such experiments. Results Using Radioactive Techniques

One of the earliest consequences of the use of radioactive techniques in red cell transport studies was the demonstration that the exchange of sugar can occur at a faster overall rate than is possible in either net entry or net exit experiments. The half-saturation for exchange 44was also found to be different from that determined by glucose-sorbose competition or by the Sen-Widdas exit method. The shortcomings were attributed to the simple kinetics of the mobile carrier model of Eq. (I) and developments followed rapidly in which alternative models of transfer and kinetics were advanced. The kinetics which were closest to those in the original hypothesis were derived from a carrier model with asymmetric affinities at the two sides and in which the rate constants for attaining equilibrium at the two interfaces could be finite relative to the rate constants for movement through the membrane. In this model49 the complex equation describing transport had five parameters for each sugar (and four of these were independent) and in addition there could be two parameters for diffusion inside and outside the cells. It is not possible to review these kinetics in detail here, nor is it possible to discuss analyses based on other kinetics (e.g., that by Eilam and Stein, 5° in which they used the tetramer model of Lieb and SteinSl). Suffice it to say that an experiment designed to measure an influx or an efflux under standardized conditions, e.g., as an equilibrium exchange or with defined conditions as to inside and outside concentrations, can still be analyzed on the basis that the measured flux will depend on the concentration in a simple Michaelis manner. Such a flux (J) can be represented by the following equation: J = Wmax[S/(S q- Ks)] 49 D. M. Regen and H. L. Tarpley, Biochim. Biophys. Acta 339, 218 (1974). 50 y . Eilarn and W. D. Stein, Biochim. Biophys. Acta 266, 161 (1972). 5t W. R. Lieb and W. D. Stein, Nature (London), New Biol. 230, 108 (1971).

(6)

[13]

SUGAR TRANSPORT IN RED BLOOD CELLS

249

which is a well-recognized form in enzyme kinetics and either a doublereciprocal or preferably a Hanes 52 plot can be used to determine Ks and Vmax- For the Hanes plot Eq. (6) is rewritten as S / J = (1/Vmax)(S -k-

Ks)

(7)

and plotting S / J as ordinate and S along the abscissa Ks is obtained as the negative intercept on the abscissa, grnax is the reciprocal of the slope. Such defined experiments can be used to study inhibitors of transport and will distinguish between competitive and noncompetitive substances. Other experiments which are valuable when studying inhibitors were outlined by Devrs and Krupka. 53 If, however, a more ambitious project is proposed in which fluxes are to be measured under widely different experimental conditions with a view to characterizing the overall transport process, then one of the more detailed kinetic analyses should be consulted and applied. Some of the ways in which the Regen and Tarpley49 kinetics have been used for this purpose were reviewed by Widdas. 36 The usual approach involves measuring influx into sugar-free cells, influx into cells preloaded with fixed concentrations of nonradioactive sugar, and influx (or efflux) under equilibrium exchange conditions. The great difficulty of measuring the initial influx into sugar-free cells has made this more complete analysis hard to attain except in cells with low transport rates such as avian erythrocytes as used by Cheung et al. 54 A n Indirect Technique

A novel technique was developed by Taverna and Langdon ~° which, while not applicable to intact red cells, could be used with resealed erythrocyte ghosts. These authors found that with partial hemolysis the erythrocytes were permeable to protein molecules and they were able to introduce glucose oxidase and catalase into such cells. If the cells were resealed by incubation at 37° the glucose oxidase and catalase were trapped in the cells and all the external enzymes could be removed by repeated washings. The entry of glucose into such cells induced oxygen utilization and by measuring the rate of oxygen disappearance with an oxygen electrode they could indirectly determine the rate of glucose influx. This technique allowed uptake to be measured under conditions where free glucose was not accumulating inside the cells. They were also able to use this technique with inside-out vesicles 5z C. S. Hanes, Biochem. J. 26, 1406 (1932). 53 R. Devrs and R. M. Krupka, Biochim. Biophys. Acta 510, 186 (1978). ~4 j. y . Cheung, D. M. Regen, M. E. Schworer, C. F. Whitfield, and H. E. Morgan, Biochim. Biophys. Acta 470, 212 (1977).

250

RED BLOOD CELLS

[14]

prepared from hemolysed red cells 55 and these further studies suggested that the sugar transport system in the membrane was not so asymmetric as had been presumed from the results with intact cells. Conclusion Various ways of studying glucose transport in red blood cells have been described and some examples of the results obtained by use of the three main techniques are illustrated. In general the biochemical techniques have been restricted in usefulness to cells with slow rates of transport and are not suitable for following the fast transport found in human and some fetal red cells. They might be exploited further in such cells if they could be combined with the use of "stopping solutions" as employed in the radioactive techniques. Optical methods require either the construction of a special apparatus or the modification of a suitable commercial instrument. Probably their most useful role is in checking substances which are thought to be possible inhibitors of sugar transport. Radioactive techniques are within the compass of most biological laboratories; they are easy to apply and honors students can quickly learn to use them to obtain meaningful results. They are the techniques of first choice for anyone entering this field of study. Whichever of the above methods is chosen, practice at the laboratory bench will be required to gain experience in preparing and handling the red cell suspensions. 55 R. D. Taverna and R. G. Langdon, Biochim. Biophys. Acta 298, 422 (1973).

[14] N u c l e o s i d e T r a n s p o r t a c r o s s R e d C e l l M e m b r a n e s By Z. I. CABANTCHIK Introduction Red blood cells from a variety of species are known to possess nucleoside transport system(s) 1-3 which subserve(s) manifold intracellular and i S. M. Jarvis, J. R. H a m m o n d , A. R. P. Paterson, and A. S. Clanachan, Biochem. J. 208, 83 (1982). 2 j. D. Young and S. M. Jarvis, Biosci. Rep. 3, 309 (1983). 3 p. G. W. Plagemann and R. M. Wohlhueter, Curr. Top. Membr. Transp. 14, 225 (1980).

METHODS IN ENZYMOLOGY,VOL. 173

Copyright © 1989by Academic Press, Inc. All rights of reproduction in any form reserved.