- Email: [email protected]
Effect of Treatment with Phospholipase A2, Colchicine, and β-Galactosidase on Agglutination of Rat Hepatocytes
Effect of Treatrneint with Phospholipase A*, Colchicine, and P-Galactosidase on Agglutination of Rat Hepatocytes ANWARB. BIH;HAZI*'*'X,SAM H. SAAD*, AND TONYABBAS GHALEB* Received July 211, 1987, from the *Department of Physiology, Faculty of Medicine, American Universiv of Beirut, Beirut, Lebanon. Accepted for publication December 1, 19813. *Present address: Office of the American University of Beirut, 850 Third Avenue, New York, NY 10022. Abstract 0 The effects of phospholipase A,, colchicine, and pgalactosidase on concanavalin A-induced agglutination of viable hepatocytes isolated irom normal and diabetic rats are reported. Phospholipase A, (0.92 pg'mL), colchicine (400 pg/mL), and p-galactosidase (300 p,g/mL)treatments caused a significant increase in the agglutination rate of hepatocytes. These findings suggest that phospholipase 4 treatment resulted in the unshielding of lectin receptors. Colchicine treatment seemed to release those receptors from cellular restraints which tend to separate and/or direct them. The promoting effect of p-galactosidase could be attributed to a deci?ease in repulsive forces due to a reduction in net negative charge density after removal of N-acetylneuraminic acid residues. Normal rat hepatocytes seem to be richer in galactosides, phospholipids, and the microfilament-microtubule network than their diabetic counterparts.
Several of the biological responses of cells are initiated by the binding of external molecules to specific receptors o n the cell surface. An analysis o f t he interaction of ligands with cell surface receptors is particularly important for an understandi n g of the biological respone,es concerned. Factors regulating the ability of cells t o agglutinate with plant lectins include the n u m b e r of lectin receptor siteq1-3 availability a n d motili t y of these sites: a n d t h e gross morphology of the cell surface.5-9 Foir example, normal a d u l t cells are less able to agglutinate with lectins such as concanavalin A when compared w i t h transformed a n d embryonic cells.10Jl Pretreatment of norrnal and transformed cells with proteolytic enzymes,12,13 P-galactosidase,l4 phospholipase A2,15J6 a n d other substancesl7.18 prior to agglutination have elucidated the mode and extent of cell membrane a n d surface membrane receptor transformations w i t h cellular changes. Recently, Bikhazi et a1 12 reported tlhat the rate of agglutination of collagenase-dispersed rat h epatocytes w a s twice as fast in normal cells i%S in cells obtained from streptozocin-treated diabetic animals. Treatimen t of dispersed normal hepatocytes w i t h trypsin and neuraminidase h a d n o effect on the rate of concanavalin A-induced agglutination, b u t increased the rate of agglutination in hepatocytes isolated from diabetic rats.13 F u r t h e r work is needed iin order t o explain the roles of membrane glycoproteins a n d phospholipids a n d the microfilamentous network in hepakocyte agglutination. Therefore, t h i s paper reports the results of t r e a t i n g dispersed hepatocytes from normal a n d diabetic rats with phospholipase A,, colchicine, a n d /3-galactosidase on concanavalin A-induced cytoagglutination. The objective is to relate cell surface membrane alterations in disease states, such as diabetes, when compared w i t h normals, in cell-to-cell agglutination kinetrc experiments.
Experimental Section Preparation 'of Diabetic Rats-Diabetes mellitus was induced in adult Sprague-Dawley rats by iv injection with 65 mgikg of streptozocin (U-9889, lot no. 60,2'73-9;courtesy of Upjohn, Kalamazoo, MI). 0022-3549/89/0700-054 7$01.00/0 0 1989, American Pharmaceuti8calAssociation
The procedure and blood glucose level tests were followed as previously described by Bikhazi et a1.12,13,19 Preparation of Rat Hepatocytes-Liver parenchymal cells were dispersed by collagenase perfusion, and cells were isolated and checked for viability by the trypan blue exclusion test, and for respiration rate according to a method described previously by Bikhazi et al.12J9 Only preparations showing ,908 viability were used. Agglutination of Hepatoeytes from Normal and Diabetic RatsFor the control experiments, -9 x 10" hepatocytes were suspended in 225 mL of Ca2+-and Mgz+-freephosphate buffered saline (pH 7.4). Cell counts were recorded kinetically by an automated counter (Coulter Counter model A; Coulter Electronics, Hialeah, FL) a t 0,5, 10, 20, and 30 min.12.19 Treatment with Concanavalin A-Two millimeters of Ca2'- and Mg"-free phosphate buffered solution containing concanavalin A was added to 225 mL of buffer to obtain a lOO-pg/mL solution of the le~tin.12~13 The hepatocytes were then dispersed in this solution and cell counts were obtained as decribed above. Treatment with Phospholipase A,, Colchicine, or PGalactosidase-Hepatocytes were treated with phospholipase A,, colchicine, or P-galactosidase and agglutination was measured in the absence of lectin. For phospholipase A,, cells were dispersed in Ca2+- and Mg2'-free phosphate buffer and treated with 0.92 pg/mL of the enzyme for 10 min.15J6 (Experiments were done utilizing three concentrations of each of phospholipase A,, colchicine, and pgalactosidase, and the reported concentrations in this paper are those that produced a maximal effect after cellular treatments while maintaining cellular viability constant.) For colchicine. hepatocytes were treated with the alkaloid for 20 min at a concentration of 400 pg/ mL.18 For p-galactosidase, treatment was for 10 min a t a concentration of 300 CLgimL.14 After each of the above-mentioned treatments, a cellular viability test with trypan blue was routinely performed, and preparations showing >90% viability were used. Cell counts were recorded at 0,5, 10, 20, and 30 min with the automated counter. In another experiment, concanavalin A (100 pg/mL in the suspension) was added after the above treatment to induce agglutination. Data Analysis-The physical model on irreversible agglutination12s20was used in the data analysis and interpretation. The general second-order rate relationship derived by Higuchi et a1.21 and supported by Bikhazi et a1.12,20was used to describe the agglutination rate. The following equations were used:
l/HZ
+ K,
1/H,
=
K
16yK,,r,Tl3~
(1)
where: =
(2)
and H, is the number of hepatocyte singlets at any time t , Hg is the number of singlets at t = 0, y is the sticking probability constant (which measures the stickiness of cells), KOis the Boltznian constant, rs is the hepatocyte radius, T is the temperature in degrees Kelvin, and 7) is the viscosity of the medium.12.20
Results A v e r a g e N u m b e r , M e a n Diameter, and Viability of Rat Hepatocytes-The average number of isolated cells per liver Journal of Pharmaceutical Sciences / 547 Vol. 78, No. 7, Ju/y 1989
was ~ 1 . x2 lo8 hepatocytes. The mean diameter was taken as 20 pm,12 and cellular viability was tested by the trypan blue exclusion test. The rate of oxygen production was estimated to be 1.0 pUmg dry weightk, which is within normal range for mammalian cells.19 Second-OrderRate Plots for Agglutination-The value of H, (eq 1)was estimated from the cumulative size distribution curves exemplified in Figure 1. The cell counts (H,) at threshold 30 (volume 4.2 x lo3 pm3) and a t 0, 5, 10, and 20 min were used to draw the linear curves of lIH, versus time (Figure 2). All data were treated similarly to those in Figure 2, which represents the least-squares line fitted to the experimental data. The slopes of the least-squares lines were estimated, and y values (eq 2) were calculatedl2.20 (Table I). Hepatocyte Agglutination-Effect of Treatment with Phospholipase A,, Colchicine, or P-Galactosidase-Agglutination was not observed after treatment with phospholipase A,, colchicine, or Pgalactosidase in the absence of concanavalin A. Therefore, agglutination observed after addition of concanavalin A (after treatment with the above substances) is due to the lectin. General Agglutination Data-Figure 1A represents typical time-dependent cumulative size distribution plots for diabetic rat hepatocytes with respect to time in the presence of 100 pgl mL of concanavalin A. Similar curves were obtained for agglutination of hepatocytes isolated from normal rats. These curves were used as controls for all subsequent data for comparison after treatment with phospholipase A, (Figure lB), colchicine, and P-galactosidase. Curves similar to those in Figures 1 and 2 were observed after treatment of normal and diabetic rat hepatocytes with the latter substances. The values of the sticking probability factor (7) for hepatocytes isolated from normal and diabetic rats (Table I)
= I 6
5
10
15 20 TIME (MINUTES)
Figure 2-Representative second-order plot of diabetic rat hepatocytes agglutination (lihepatocytecount versus time). Straight lines are leastsquares fit to the experimental data in the presence of 0.92 pgimL of phospholipase A, and 100 pgimL of concanavalin A. Key: (B) 100 pghL of concanavalin A; (0)0.92 pg/mL of phospholipase A,.
showed no significant difference from those reported by Bikhazi et al.,12J3 supporting the reproducibility of the data. Table I reports the sticking probability factor data for the various treatments. Treatment of the hepatocytes from normal and diabetic rats with phospholipase A, followed by addition of concanavalin A resulted in a threefold increase in the yvalues for normal and diabetic cells when compared with
W
3 ln
2000
CK W
>
0
m
w IS00 b > 0
2
i 3
a w I
W
I
10000
LL
0
0
Lz W
W
a m
m
5
1000
LL
r 3
5000
z
Z
0 0
3
6
9
1 2 1 5
Figure 1-Representative cumulative size distribution curves for diabetic rat hepatocytes. (A) Treatment with concanavalin A alone. (B) Treatment with phospholipase A, and concanavalin A. Key: (0)0 min, without concanavalin A; (e)0 min, with concanavalin A; (A)5 min; (A) 10 rnin; (B) 20
min.
548 i Journal of Pharmaceutical Sciences Vol. 78,No. 7, July 1989
Table I-Sticking Probability Factors for Normal and Diabetic Rat Hepatocyte Agglutination with Concanavalin A (Con A), Phospholipase A,, Colchi’clne,and pGalactosidase
Treatmenta
-Concanavalin A (Control) Phospholipase A, Phospholipase PI, + Con A Colchicine Colchicine + Con A p-Galactosidase 8-Galactosidase + Con A
!Sticking Probability Factor 2 SEMb (y x lo7 cm-’*hepatocyte-’) Normal Rats 7.42 t 0.13 (c) 0 24.21 0.12 (e) 0 24.02 2 0.10 (9) 0 15.95 5 0.1 1 (i)
*
Diabetic Rats 3.84
* 0.67 (d)
0 12.26 t 1.50 (f) 0 10.53 2 1.42 (h) 0 8.85 2 1.20 (j)
a Concentrations:concanavalin A, 100 pg/mL; phospholipase A*, 0.92 pgimL; colchicine, 400 pghnL; &galactosidase, 300 pg/mL. Mean 2 SEM; II = 6; lelters in parentheses are codes for statistical analysis; significant differences: p < 0.00’1 (c, d ) , (c, e), (c, g), (c, i), (d, f), (d, h), (d, j), (e, f ) , (g, PO, 0, j).
the controls. The same effect was observed with colchicine treatment followed by the #additionof conanavalin A. However, treatment with p-galactosidase followed by concanavalin A resulted in a twofold increase in y values. Statistical Analysis d Data-The t test method was used for data analysis since a normal population distribution existed for the normal and diabetic animal groups.
Di!rcussion Surface glycoproteins, which are randomly distributed over the cell surface, have been reported to be connected to a microfilament-microtubule network i n the cytoplasmic matrix.22 Concanavaliri A, by binding to surface receptors, may induce a nonrandom clustered distribution of these receptors.23 The role of the microtubules in the capping of the concanavalin A receptors i s c o n t r o v e r s i a l . Some investigators24 suggested that the microtubules are responsible for the directed movement of the receptors. Thus, treatment of cells with colchicine, which disrupts the microtubules, may show a hindering effect on the clustering of these receptors. However, other investigators18s22suggested that after disruption of the microtubules with colchicine, the receptors are no longer restrained so they can be clustered by concanavalin A. The data in Table I supports the latter suggestion that clustering of receptors requires release from cellular restraints (such as by colchicine treatment) which tend to separate and/or direct them. The increase in agglutination (Table I) after pretreatment with phospholipase A, may be due to the hydrolysis and removal of membrane phospholipids. It is therefore possible that membrane phospholipids may be directly involved in the shielding of concanavalin A. receptors, or that the removal of these phospholipids may alter the structure of other membrane components which in turn are responsible for the unmasking process. The data also indicate that phospholipids are more abundant in normal hepatocytes than in their diabetic counterparts, :supporting the results by Bikhazi et a1.12 p-Galactosidase with concanavalin A causes a n increase in the agglutination of both normal and diabetic rat hepatocytes (Table I). Galactosides have been shown to be responsible for the existence of the net negative charge on the cell surface and always occupy distal positions to the polypeptide chain of the
membrane glycoproteins.6 P-Galactosidase may act by removing these terminal galactosides which could cause a reduction in net negative membrane charge density, thereby reducing the repulsive forces and enhancing cytoagglutination. Neuraminidasel3 and Pgalactosidase seem to have the same effect on concanavalin A-induced agglutination of diabetic rat hepatocytes. However, neuraminidase specifically cleaves off N-acetylneuraminic acid terminal groups, while pgalactosidase hydrolyzes all galactoside groups. Since both enzymes have the same effects on diabetic hepatocytes, it may be deduced that N-acetylneuraminic acid is the only galactoside present in the structure of the concanavalin A receptors on the surface of diabetic rat hepatocytes. However, treatment of normal cells with p-galactosidase gave higher sticking probability constants (yvalues) than cells treated with neuraminidase, suggesting that normal cell membranes contain not only N-acetylneuraminic acid, but a n abundance of Blinked galactosides as well. Cell membranes of diabetic rat hepatocytes may be richer in N-acetylneuraminic acid than their normal counterparts (Table I), contrary to what has been reported by Chandramouli and Carter.25 In conclusion, most of the receptors in concanavalin Ainduced cytoagglutination of normal and diabetic hepatocytes from rats are embedded in the cell membrane, and their exposure for increased agglutination would need enzymatic andtor drug treatments.
References and Notes Inbar, M.; Sachs, L. Proc. Natl. Acad. Sci. 1969, 63, 1418. Burger, M. M. Proc. Nutl. Acad. Sci. 1969, 62, 994. Inbar, M.; Sachs, L. Nature 1969,223, 710. Marikovsky, Y.;Inbar, M.; Danon, D.; Sachs, L. Exp. Cell Res. 1974, 89, 359. 5. Weiss, L. J. Cell Biol. 1965,26, 735. 6. Porter, K.; Prescot, D.; Frye, J. J . Cell Biol. 1973, 57, 815. 7. Willingham, M. C.; Pastan, I. Proc. Natl. Acad. Sci. 1975, 72, 1263. 8. Bretcher, M. S.; Raff, M. C. Nature 1975,258,43. 9. Glenney, J. R., Jr.; Hixson, D. C.; Walborg, E. F., Jr. Exp. Cell Res. 1979, 118, 353. 10. Nicholson, G. Nature, New Biol. 1971,233, 244, 11. Rosenblith, J. Z.; Ukena, T. E.; Yin, H. H.; Berlin, R. D.; Karnovsky, M. J. Proc. Natl. Acad. Sci. 1971, 70, 1625. 12. Bikhazi, A. B.; Abboud, E. H.; Agulian, S. K., Nassar, C. F. Pflugers Archiu. 1980,386, 245. 13. Bikhazi, A. B.; Turyaki, L. H. J . Pharm. Sci. 1984, 73, 628. 14. Cuatrecasas, P. Fed. Proc. 1973, 32, 1838. 15. Cuatrecasas, P. J . Biol. Chem. 1971.246. 6532. 16. Cuatrecasas, P. J. Biol. Chem. 1971; 246; 6522. 17. Edelman, G. M.; Yahara, I.; Wang, J. L. Proc. Natl. Acad. Sci. 1973. 70. 1442. 18. Oppenheimer, S. B.; Bales, B. L.; Brenneman, G.; Knapp, L.; Lesin, E. S.;Neri, A,;Pollock, E. G. Exp. Cell Res. 1977,105,291. 19. Bikhazi, A. B.; Noubani, H. M.; Coe, E. L. J . Phurm. Sci. 1983, 72, 116. 20. Bikhazi, A. B.; Ayyoub, G. E. J . Phurm. Sci. 1978, 67,939. 21. Higuchi, W. I.; Okada, R.; Stelter, G. A.; LembGger, A. P. J . Phnrm. Sci. 1963,52,49. 22. Fontana. F. Acta Embrvol. E m . 1979.3. 225. 1. 2. 3. 4.
I
,
25. Chandramouli, V.; Carter, J. R., Jr. Diabetes 1975,24, 257
Acknowledgments This work was supported by the Nadim Andraos Foundationand by the Lebanese National Council for Scientific Research.
Journal of Pharmaceutical SciencesI 549 Vol. 78,No. 7,July 1989
Several of the biological responses of cells are initiated by the binding of external molecules to specific receptors o n the cell surface. An analysis o f t he interaction of ligands with cell surface receptors is particularly important for an understandi n g of the biological respone,es concerned. Factors regulating the ability of cells t o agglutinate with plant lectins include the n u m b e r of lectin receptor siteq1-3 availability a n d motili t y of these sites: a n d t h e gross morphology of the cell surface.5-9 Foir example, normal a d u l t cells are less able to agglutinate with lectins such as concanavalin A when compared w i t h transformed a n d embryonic cells.10Jl Pretreatment of norrnal and transformed cells with proteolytic enzymes,12,13 P-galactosidase,l4 phospholipase A2,15J6 a n d other substancesl7.18 prior to agglutination have elucidated the mode and extent of cell membrane a n d surface membrane receptor transformations w i t h cellular changes. Recently, Bikhazi et a1 12 reported tlhat the rate of agglutination of collagenase-dispersed rat h epatocytes w a s twice as fast in normal cells i%S in cells obtained from streptozocin-treated diabetic animals. Treatimen t of dispersed normal hepatocytes w i t h trypsin and neuraminidase h a d n o effect on the rate of concanavalin A-induced agglutination, b u t increased the rate of agglutination in hepatocytes isolated from diabetic rats.13 F u r t h e r work is needed iin order t o explain the roles of membrane glycoproteins a n d phospholipids a n d the microfilamentous network in hepakocyte agglutination. Therefore, t h i s paper reports the results of t r e a t i n g dispersed hepatocytes from normal a n d diabetic rats with phospholipase A,, colchicine, a n d /3-galactosidase on concanavalin A-induced cytoagglutination. The objective is to relate cell surface membrane alterations in disease states, such as diabetes, when compared w i t h normals, in cell-to-cell agglutination kinetrc experiments.
Experimental Section Preparation 'of Diabetic Rats-Diabetes mellitus was induced in adult Sprague-Dawley rats by iv injection with 65 mgikg of streptozocin (U-9889, lot no. 60,2'73-9;courtesy of Upjohn, Kalamazoo, MI). 0022-3549/89/0700-054 7$01.00/0 0 1989, American Pharmaceuti8calAssociation
The procedure and blood glucose level tests were followed as previously described by Bikhazi et a1.12,13,19 Preparation of Rat Hepatocytes-Liver parenchymal cells were dispersed by collagenase perfusion, and cells were isolated and checked for viability by the trypan blue exclusion test, and for respiration rate according to a method described previously by Bikhazi et al.12J9 Only preparations showing ,908 viability were used. Agglutination of Hepatoeytes from Normal and Diabetic RatsFor the control experiments, -9 x 10" hepatocytes were suspended in 225 mL of Ca2+-and Mgz+-freephosphate buffered saline (pH 7.4). Cell counts were recorded kinetically by an automated counter (Coulter Counter model A; Coulter Electronics, Hialeah, FL) a t 0,5, 10, 20, and 30 min.12.19 Treatment with Concanavalin A-Two millimeters of Ca2'- and Mg"-free phosphate buffered solution containing concanavalin A was added to 225 mL of buffer to obtain a lOO-pg/mL solution of the le~tin.12~13 The hepatocytes were then dispersed in this solution and cell counts were obtained as decribed above. Treatment with Phospholipase A,, Colchicine, or PGalactosidase-Hepatocytes were treated with phospholipase A,, colchicine, or P-galactosidase and agglutination was measured in the absence of lectin. For phospholipase A,, cells were dispersed in Ca2+- and Mg2'-free phosphate buffer and treated with 0.92 pg/mL of the enzyme for 10 min.15J6 (Experiments were done utilizing three concentrations of each of phospholipase A,, colchicine, and pgalactosidase, and the reported concentrations in this paper are those that produced a maximal effect after cellular treatments while maintaining cellular viability constant.) For colchicine. hepatocytes were treated with the alkaloid for 20 min at a concentration of 400 pg/ mL.18 For p-galactosidase, treatment was for 10 min a t a concentration of 300 CLgimL.14 After each of the above-mentioned treatments, a cellular viability test with trypan blue was routinely performed, and preparations showing >90% viability were used. Cell counts were recorded at 0,5, 10, 20, and 30 min with the automated counter. In another experiment, concanavalin A (100 pg/mL in the suspension) was added after the above treatment to induce agglutination. Data Analysis-The physical model on irreversible agglutination12s20was used in the data analysis and interpretation. The general second-order rate relationship derived by Higuchi et a1.21 and supported by Bikhazi et a1.12,20was used to describe the agglutination rate. The following equations were used:
l/HZ
+ K,
1/H,
=
K
16yK,,r,Tl3~
(1)
where: =
(2)
and H, is the number of hepatocyte singlets at any time t , Hg is the number of singlets at t = 0, y is the sticking probability constant (which measures the stickiness of cells), KOis the Boltznian constant, rs is the hepatocyte radius, T is the temperature in degrees Kelvin, and 7) is the viscosity of the medium.12.20
Results A v e r a g e N u m b e r , M e a n Diameter, and Viability of Rat Hepatocytes-The average number of isolated cells per liver Journal of Pharmaceutical Sciences / 547 Vol. 78, No. 7, Ju/y 1989
was ~ 1 . x2 lo8 hepatocytes. The mean diameter was taken as 20 pm,12 and cellular viability was tested by the trypan blue exclusion test. The rate of oxygen production was estimated to be 1.0 pUmg dry weightk, which is within normal range for mammalian cells.19 Second-OrderRate Plots for Agglutination-The value of H, (eq 1)was estimated from the cumulative size distribution curves exemplified in Figure 1. The cell counts (H,) at threshold 30 (volume 4.2 x lo3 pm3) and a t 0, 5, 10, and 20 min were used to draw the linear curves of lIH, versus time (Figure 2). All data were treated similarly to those in Figure 2, which represents the least-squares line fitted to the experimental data. The slopes of the least-squares lines were estimated, and y values (eq 2) were calculatedl2.20 (Table I). Hepatocyte Agglutination-Effect of Treatment with Phospholipase A,, Colchicine, or P-Galactosidase-Agglutination was not observed after treatment with phospholipase A,, colchicine, or Pgalactosidase in the absence of concanavalin A. Therefore, agglutination observed after addition of concanavalin A (after treatment with the above substances) is due to the lectin. General Agglutination Data-Figure 1A represents typical time-dependent cumulative size distribution plots for diabetic rat hepatocytes with respect to time in the presence of 100 pgl mL of concanavalin A. Similar curves were obtained for agglutination of hepatocytes isolated from normal rats. These curves were used as controls for all subsequent data for comparison after treatment with phospholipase A, (Figure lB), colchicine, and P-galactosidase. Curves similar to those in Figures 1 and 2 were observed after treatment of normal and diabetic rat hepatocytes with the latter substances. The values of the sticking probability factor (7) for hepatocytes isolated from normal and diabetic rats (Table I)
= I 6
5
10
15 20 TIME (MINUTES)
Figure 2-Representative second-order plot of diabetic rat hepatocytes agglutination (lihepatocytecount versus time). Straight lines are leastsquares fit to the experimental data in the presence of 0.92 pgimL of phospholipase A, and 100 pgimL of concanavalin A. Key: (B) 100 pghL of concanavalin A; (0)0.92 pg/mL of phospholipase A,.
showed no significant difference from those reported by Bikhazi et al.,12J3 supporting the reproducibility of the data. Table I reports the sticking probability factor data for the various treatments. Treatment of the hepatocytes from normal and diabetic rats with phospholipase A, followed by addition of concanavalin A resulted in a threefold increase in the yvalues for normal and diabetic cells when compared with
W
3 ln
2000
CK W
>
0
m
w IS00 b > 0
2
i 3
a w I
W
I
10000
LL
0
0
Lz W
W
a m
m
5
1000
LL
r 3
5000
z
Z
0 0
3
6
9
1 2 1 5
Figure 1-Representative cumulative size distribution curves for diabetic rat hepatocytes. (A) Treatment with concanavalin A alone. (B) Treatment with phospholipase A, and concanavalin A. Key: (0)0 min, without concanavalin A; (e)0 min, with concanavalin A; (A)5 min; (A) 10 rnin; (B) 20
min.
548 i Journal of Pharmaceutical Sciences Vol. 78,No. 7, July 1989
Table I-Sticking Probability Factors for Normal and Diabetic Rat Hepatocyte Agglutination with Concanavalin A (Con A), Phospholipase A,, Colchi’clne,and pGalactosidase
Treatmenta
-Concanavalin A (Control) Phospholipase A, Phospholipase PI, + Con A Colchicine Colchicine + Con A p-Galactosidase 8-Galactosidase + Con A
!Sticking Probability Factor 2 SEMb (y x lo7 cm-’*hepatocyte-’) Normal Rats 7.42 t 0.13 (c) 0 24.21 0.12 (e) 0 24.02 2 0.10 (9) 0 15.95 5 0.1 1 (i)
*
Diabetic Rats 3.84
* 0.67 (d)
0 12.26 t 1.50 (f) 0 10.53 2 1.42 (h) 0 8.85 2 1.20 (j)
a Concentrations:concanavalin A, 100 pg/mL; phospholipase A*, 0.92 pgimL; colchicine, 400 pghnL; &galactosidase, 300 pg/mL. Mean 2 SEM; II = 6; lelters in parentheses are codes for statistical analysis; significant differences: p < 0.00’1 (c, d ) , (c, e), (c, g), (c, i), (d, f), (d, h), (d, j), (e, f ) , (g, PO, 0, j).
the controls. The same effect was observed with colchicine treatment followed by the #additionof conanavalin A. However, treatment with p-galactosidase followed by concanavalin A resulted in a twofold increase in y values. Statistical Analysis d Data-The t test method was used for data analysis since a normal population distribution existed for the normal and diabetic animal groups.
Di!rcussion Surface glycoproteins, which are randomly distributed over the cell surface, have been reported to be connected to a microfilament-microtubule network i n the cytoplasmic matrix.22 Concanavaliri A, by binding to surface receptors, may induce a nonrandom clustered distribution of these receptors.23 The role of the microtubules in the capping of the concanavalin A receptors i s c o n t r o v e r s i a l . Some investigators24 suggested that the microtubules are responsible for the directed movement of the receptors. Thus, treatment of cells with colchicine, which disrupts the microtubules, may show a hindering effect on the clustering of these receptors. However, other investigators18s22suggested that after disruption of the microtubules with colchicine, the receptors are no longer restrained so they can be clustered by concanavalin A. The data in Table I supports the latter suggestion that clustering of receptors requires release from cellular restraints (such as by colchicine treatment) which tend to separate and/or direct them. The increase in agglutination (Table I) after pretreatment with phospholipase A, may be due to the hydrolysis and removal of membrane phospholipids. It is therefore possible that membrane phospholipids may be directly involved in the shielding of concanavalin A. receptors, or that the removal of these phospholipids may alter the structure of other membrane components which in turn are responsible for the unmasking process. The data also indicate that phospholipids are more abundant in normal hepatocytes than in their diabetic counterparts, :supporting the results by Bikhazi et a1.12 p-Galactosidase with concanavalin A causes a n increase in the agglutination of both normal and diabetic rat hepatocytes (Table I). Galactosides have been shown to be responsible for the existence of the net negative charge on the cell surface and always occupy distal positions to the polypeptide chain of the
membrane glycoproteins.6 P-Galactosidase may act by removing these terminal galactosides which could cause a reduction in net negative membrane charge density, thereby reducing the repulsive forces and enhancing cytoagglutination. Neuraminidasel3 and Pgalactosidase seem to have the same effect on concanavalin A-induced agglutination of diabetic rat hepatocytes. However, neuraminidase specifically cleaves off N-acetylneuraminic acid terminal groups, while pgalactosidase hydrolyzes all galactoside groups. Since both enzymes have the same effects on diabetic hepatocytes, it may be deduced that N-acetylneuraminic acid is the only galactoside present in the structure of the concanavalin A receptors on the surface of diabetic rat hepatocytes. However, treatment of normal cells with p-galactosidase gave higher sticking probability constants (yvalues) than cells treated with neuraminidase, suggesting that normal cell membranes contain not only N-acetylneuraminic acid, but a n abundance of Blinked galactosides as well. Cell membranes of diabetic rat hepatocytes may be richer in N-acetylneuraminic acid than their normal counterparts (Table I), contrary to what has been reported by Chandramouli and Carter.25 In conclusion, most of the receptors in concanavalin Ainduced cytoagglutination of normal and diabetic hepatocytes from rats are embedded in the cell membrane, and their exposure for increased agglutination would need enzymatic andtor drug treatments.
References and Notes Inbar, M.; Sachs, L. Proc. Natl. Acad. Sci. 1969, 63, 1418. Burger, M. M. Proc. Nutl. Acad. Sci. 1969, 62, 994. Inbar, M.; Sachs, L. Nature 1969,223, 710. Marikovsky, Y.;Inbar, M.; Danon, D.; Sachs, L. Exp. Cell Res. 1974, 89, 359. 5. Weiss, L. J. Cell Biol. 1965,26, 735. 6. Porter, K.; Prescot, D.; Frye, J. J . Cell Biol. 1973, 57, 815. 7. Willingham, M. C.; Pastan, I. Proc. Natl. Acad. Sci. 1975, 72, 1263. 8. Bretcher, M. S.; Raff, M. C. Nature 1975,258,43. 9. Glenney, J. R., Jr.; Hixson, D. C.; Walborg, E. F., Jr. Exp. Cell Res. 1979, 118, 353. 10. Nicholson, G. Nature, New Biol. 1971,233, 244, 11. Rosenblith, J. Z.; Ukena, T. E.; Yin, H. H.; Berlin, R. D.; Karnovsky, M. J. Proc. Natl. Acad. Sci. 1971, 70, 1625. 12. Bikhazi, A. B.; Abboud, E. H.; Agulian, S. K., Nassar, C. F. Pflugers Archiu. 1980,386, 245. 13. Bikhazi, A. B.; Turyaki, L. H. J . Pharm. Sci. 1984, 73, 628. 14. Cuatrecasas, P. Fed. Proc. 1973, 32, 1838. 15. Cuatrecasas, P. J . Biol. Chem. 1971.246. 6532. 16. Cuatrecasas, P. J. Biol. Chem. 1971; 246; 6522. 17. Edelman, G. M.; Yahara, I.; Wang, J. L. Proc. Natl. Acad. Sci. 1973. 70. 1442. 18. Oppenheimer, S. B.; Bales, B. L.; Brenneman, G.; Knapp, L.; Lesin, E. S.;Neri, A,;Pollock, E. G. Exp. Cell Res. 1977,105,291. 19. Bikhazi, A. B.; Noubani, H. M.; Coe, E. L. J . Phurm. Sci. 1983, 72, 116. 20. Bikhazi, A. B.; Ayyoub, G. E. J . Phurm. Sci. 1978, 67,939. 21. Higuchi, W. I.; Okada, R.; Stelter, G. A.; LembGger, A. P. J . Phnrm. Sci. 1963,52,49. 22. Fontana. F. Acta Embrvol. E m . 1979.3. 225. 1. 2. 3. 4.
I
,
25. Chandramouli, V.; Carter, J. R., Jr. Diabetes 1975,24, 257
Acknowledgments This work was supported by the Nadim Andraos Foundationand by the Lebanese National Council for Scientific Research.
Journal of Pharmaceutical SciencesI 549 Vol. 78,No. 7,July 1989