We thank Charles S. Rubin, E. Richard Stanley, Luz S. Teicher, and. Allan Wolkoff for help and discussion, Linda Hall for the use of her computer, and Craig ...
Proc. Natl. Acad. Sci. USA Vol. 80, pp. 5379-5382, September 1983 Genetics
Epidermal growth factor and glucagon receptors in mice homozygous for a lethal chromosomal deletion (membrane-bound hormone receptors/gene regulation)
PHYLLIS A. SHAW AND SALOME GLUECKSOHN-WAELSCH Department of Genetics, Albert Einstein College of Medicine, Bronx, New York 10461
Contributed by Salome G. Waelsch, May 27, 1983
ABSTRACT The binding of epidermal growth factor (EGF) and of glucagon to their receptors has been examined in singlecell suspensions obtained from livers and other organs of newborn mice homozygous for a perinatally lethal deletion that includes the albino (c) locus on chromosome 7. Competition experiments with '5I-labeled and nonradioactive EGF and Scatchard analysis of equilibrium binding data showed that hepatocytes from deletion homozygotes had only -20% of the number of specific EGF receptors present in cells from normal littermates. In contrast, EGF binding to single-cell suspensions from organs other than the liver was normal in deletion homozygotes. Similar results were obtained in competitive displacement experiments with '15I-labeled and nonradioactive glucagon: hepatocytes from deletion mutants showed only -30% of the specific glucagon binding sites found in cells from normal littermates. As in the case of EGF, the decreased binding was due to decreased numbers of glucagon receptors per cell rather than alterations in receptor affinity, and glucagon binding to single-cell suspensions from organs other than the liver was normal in the deletion mutants. The reductions in numbers of EGF and glucagon receptors are liver-cell specific as are the previously described ultrastructural and biochemical abnormalities in these mutants. The significance of cell membrane integrity and hormone-receptor interactions in the control of normal liver cell differentiation is discussed.
Newborn mice homozygous for one of several overlapping perinatally lethal deletions mapping around the albino (c) locus in chromosome 7 are characterized by liver-specific multiple biochemical and ultrastructural membrane abnormalities (1). A recent investigation of mechanisms possibly responsible for these abnormalities focused on insulin binding activities, which were shown to be decreased to 20-25% of normal in liver cells of deletion homozygotes because of a reduction of the number of insulin receptors per cell (2). The pattern of binding of epidermal growth factor (EGF) is known to resemble the insulin binding pattern: both bind to the plasma membrane, and hormone-receptor complexes are formed that subsequently are internalized by the same pathway-(3). Because of this similarity, the binding of EGF to liver cells of deletion homozygotes was investigated. Furthermore, a study of glucagon binding activities was suggested by the failure of several liver-specific enzymes to be induced by glucagon in newborn deletion homozygotes in contrast to their normal littermates (1).
MATERIALS AND METHODS Normal and Mutant Mice. Mice carrying the lethal albino deletions cl4CoS and C3H are maintained in our mouse colony and bred as heterozygotes with cch (chinchilla) as the normal allele. Deletion-homozygous newborn mice are recognized as The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
albinos by their unpigmented eyes. Animals were sacrificed by decapitation within a few hours after spontaneous birth or as near term fetuses. Livers and other organs were removed for the preparation of single-cell suspensions. Preparation of Single-Cell Suspensions from Livers and Other Organs. Single-cell suspensions of livers from homozygous deletion-mutant and heterozygous and homozygous normal newborn littermates were prepared by a modification of the method of Amsterdam and Jamieson (4). Minced livers were incubated in a collagenase solution [3 mg of collagenase (Millipore) per ml of Hanks' balanced salt solution with Ca2" and Mg2e] for 15 min at 370C with shaking at 130 oscillations per min. This was followed by incubation in Hanks' solution containing 2 mM EDTA but lacking Ca2" and Mg2" for 5 min under the same conditions. The cell and tissue mixture was then washed once in McCoy's 5a medium (modified) supplemented with 10% fetal calf serum and twice in Hepes binding buffer (0.1 M Hepes/0. 12 M NaCl/0.012 M MgCl2/0.025 M KC1/0.01 M glucose/0.001 M EDTA/0.0005 M sodium acetate/10 mg of bovine serum albumin per ml, pH 7.4). After resuspension in 2 ml of binding buffer, the cells were filtered through 60Am nylon mesh and counted in a hemocytometer. Heart alone or a combination of organs-i.e., thymus, heart, lungs, spleen, and brain-were removed from mutant and normal littermates, and single-cell suspensions -were prepared by the same procedure as for liver cells. Preparations were used only when the cell viability was 90% or higher as judged by exclusion of trypan blue. Duplicates of the same number of cells (106) were used for each point in the EGF and glucagon binding experiments. Preparation of "mI-Labeled EGF ('25I-EGF). For nonradioactive EGF competitive displacement experiments, EGF (Collaborative Research, Waltham, MA) was iodinated (5) by using the metabisulfite/chloramine-T method. High-specificactivity Na'25I (1 mCi, New England Nuclear; 1 Ci = 3.7 X 1010 Bq) was immediately neutralized with 10 ,ul of 1 M potassium phosphate (pH 7). Five micrograms of EGF and 15 ,ug of chloramine T (10 ,ul) were added, and the reaction was stopped in 25 sec by the addition of 15 ,ug of sodium metabisulfite (10 ,ul). The labeled protein was separated from unreacted Na'25I by gel filtration through Sephadex G-25 with a buffer containing 0.1 M potassium phosphate (pH 7) and 2 mg of bovine serum albumin per ml. Before applying the iodinated EGF to the Sephadex column, a l-,ul aliquot was diluted and treated with trichloroacetic acid to form a precipitate for assay of percentage incorporation and specific activity, which.was 140,000-165,000 cpm/ng. l25I-EGF from New England Nuclear (131 ,uCi/,ug) was used for the saturation binding experiments. EGF Binding Assays. Time course of EGF binding. Three experiments were carried out with hepatocytes from a total of seven homozygous deletion-mutant and six normal newborn Abbreviation: EGF, epidermal growth factor.
5379
5380
Proc. Natl. Acad. Sci. USA 80 (1983)
Genetics: Shaw and Gluecksohn-Waelsch.
littermates. Hepatocytes were incubated in 0.1 ml of Hepes binding buffer with 1.7 nM 125I-EGF in the presence, for nonspecific binding computation, or absence of 125-fold excess of nonradioactive EGF for 5-240 min at 0C. At various time intervals, 85-,ul aliquots were pipetted onto an underlayer of 0.25 M sucrose in Hepes binding buffer. The unbound radioactive EGF was removed by centrifuging the cells through this underlayer for 5 min in an Eppendorf centrifuge in the cold. The supernatant was aspirated, centrifuge tube tips containing the intact cells and hormone-receptor complexes were cut, and the 15 radioactivity was determined in a gamma spectrometer. Competitive displacement of 125I-EGF from hepatocyte surface receptors by nonradioactive EGF. Five experiments were performed with mouse liver cells from a total of 11 homozygous deletion-mutant and 11 normal newborn littermates. The cells were incubated with 0.8 nM 125I-EGF in a final volume of 0.1 ml of Hepes binding buffer for 2 hr at 00C, either alone or in the presence of various concentrations of nonradioactive EGF (0.04-420 nM). After incubation, the unbound radioactive EGF was removed by pelleting the cells through an underlayer of Hepes/sucrose binding buffer as described above. The data were corrected for nonspecific binding (1 ,uM nonradioactive EGF), which amounted to z1% of the added radioactivity. Saturation binding to hepatocytes. Hepatocytes from six homozygous deletion-mutant and seven normal littermate newborn mice were used in a total of four experiments. One tenth of a milliliter containing 106 hepatocytes was incubated at 0°C for 2 hr with 0.6-4.2 nM 125I-EGF (New.England Nuclear; 131 ,uCi/,g) in the presence or absence of 125-fold excess of nonradioactive EGF (0.53 ,M). An 85-,ul aliquot was removed from each sample, and the hepatocytes were pelleted through Hepes/ sucrose as described above. Dissociation constants-(Kd) and the number of saturable binding sites (Bm.) were determined for each experiment by the method of Scatchard (6) with a TRS-80 computer (Radio Shack) for linear regression analysis of the data. EGF binding activity of cells from other organs. A total of eight homozygous deletion-mutant and seven normal littermate newborn mice were used in three experiments. These were performed with five samples each of 106 cells prepared from pools of thymus, heart, lungs, spleen, and brain in 0.1 ml of Hepes binding buffer; the cells were incubated for 2 hr at 0°C with 4.2 nM I-EGF in the presence or absence of 125-fold excess of nonradioactive EGF. An 85-±10 aliquot was removed, pelleted through Hepes/sucrose, and the radioactivity was determined. Glucagon Binding Assays. "2I-Labeled glucagon ('251-glucagon) was obtained from New England Nuclear. Competitive displacement of 1251-glucagon from hepatocyte surface receptors by nonradioactive glucagon. Three experiments were performed with hepatocytes from six homozygous deletion-mutant and six normal littermate newborn mice. The cells were incubated with 0.67 nM 125I-glucagon (New England Nuclear; 131 ,uCi/,ug), in the presence or absence of various concentrations of nonradioactive glucagon (1.9-380 nM) at 30°C for 30 min. An aliquot (85 ,u) was removed, and the cells were pelleted through Hepes/sucrose binding buffer as described for EGF binding. The data were corrected for nonspecific binding-i.e., binding in the presence of 0.76 AM nonradioactive glucagon. Binding of 1251-glucagon to cells from organs other than the liver. Five homozygous deletion-mutant and five normal littermate newborn mice were used in two experiments for the preparation of single-cell suspensions of either heart alone or pools of thymus, heart, lungs, spleen, and brain. The cell suspensions were incubated for 30 min at 300C with 0.66 nM 125Iglucagon in the presence or absence of 1 AM nonradioactive
glucagon. Cells were pelleted through an underlayer of Hepes/ sucrose binding buffer as described above. RESULTS EGF and Glucagon Binding Activities in Deletion Heterozygotes and Wild-Type Homozygotes. As shown previously, the lethal albino deletions act as recessives and have no effect on liver morphogenesis and function in the heterozygous state (1). In the present experiments also, no difference was found in binding activities of EGF or glucagon between normal deletion heterozygotes and wild-type homozygotes. Time Course of 125I-EGF Binding. The time course of binding of 125I-EGF to single-cell suspensions of livers obtained from newborn mutant and normal littermate mice is shown in Fig. 1. Maximal binding at 00C was reached at -1.5-2.5 hr in normal hepatocytes and at -1-1.5 hr ia mutant hepatocytes. There was no net loss of cell-bound radioactive EGF during the course of the experiments in which the maximal amount of EGF bound by mutant hepatocytes amounted to 16% of that bound by normal hepatocytes. Competitive Displacement of 125I-EGF from Hepatocytes. The ability was measured of increasing concentrations of nonradioactive EGF to compete with 125I-EGF for specific binding sites on the- surface of intact hepatocytes of normal and mutant newborn mice. The experiment shown in Fig. 2 demonstrates that hepatocytes of mutant mice bound -14% as much 125I-EGF as did normal hepatocytes. These results suggest that the reduction of EGF binding in mutant hepatocytes is a reflection of a reduced number of EGF receptors; nevertheless, this experiment does not exclude a possible alteration in their affinity. Saturation Binding Data. In order to confirm the suggestion derived from the competition experiments that the numbers of EGF receptors were decreased in the mutants and to examine and compare the affinities of EGF receptors, saturation binding experiments were performed. Data from four such experiments analyzed by the method of Scatchard (Table 1) show that the computer-calculated mean affinity constants (Kd) of EGF receptors are the same in hepatocytes from mutant and normal littermates, 2.5 ± 1.10 (SD) and 2.7 ± 1.05 (SD) M x 10-9, respectively. However, the numbers of receptors differ and amount to 2.43 ± 1.0 (SD) fmol per 106 normal liver cells and 0.58 ± 0.08 (SD) fmol per 106 mutant liver cells, representing a reduction in mutants of 76%. Fig. 3A shows one representative experiment of the four included in Table 1. It demonstrates that EGF receptors are saturable with maximal binding at -3-4 nM for normal and for mutant hepatocytes. Scatchard analysis of this experiment indicates that mutant hepatocytes have 27% of the number of normal hepatocyte receptors-i. e.,
"u 10 0
OD
t9N
84 8 6
-* 0
30
60
90
^
120 150 180 210 240 Time, min
FIG. 1. Time course of 125I-EGF binding at 0°C to hepatocytes from homozygous deletion-mutant (A) and normal (W) newborn mice. 125I-EGF (final concentration of 140,000 cpm/ng, 1 ng per 106 cells) was added to cells suspended in Hepes binding buffer. Duplicate samples of 106 cells per sample in the presence or absence of a 125-fold excess of nonradioactive EGF were removed at the indicated time intervals and the cell-bound radioactivity was determined.
Genetics: Shaw and Glueeksohn-Waelsch
Proc. NatL Acad. Sci. USA 80 (1983)
4t
5381
A
1.5F
-6 0
ca
C>
`43 x
s
1.0I
.o aq
_
0N
CD
Q C:
0.5
-6
C0
C._
0
u
KI
1
CZ 0
u)
0
2 3 Free EGF, nM
4
5
B
0.
0
-
-.
L
0
1
0.1
1
10
100 Total EGF, nM
1,000
FIG. 2. Competitive displacement of 125I-EGF (0.80 nM) from mutant (A) and normal (A) hepatocyte receptors with nonradioactive EGF. Binding assays were carried out with 106 cells in each assay in the presence or
0
0.!It5
0
absence of 1,000-fold excess of nonradioactive EGF.
0.67 fmol and 2.48 fmol, respectively,
per 106 cells (Fig. 3B). Calculated binding constants (Kds) were 2.9 10-9 M for normal and 3.7 x 10-9 M for mutant hepatocytes. The total number of EGF receptors derived from these data amounts to =1,500 per hepatocyte from normal newborn mice. This figure is considerably below that obtained in our own experiments with adult hepatocytes as well as those generally reported for adult liver cells and hepatoma cell lines. Two factors must be taken into account when considering these lower figures of EGF receptors in hepatocytes from newborn normal mice. First of all, it is known that the proportion of hepatocytes in the total cell population of the newborn liver is smaller than in the adult because the newborn liver includes a larger number of hematopoietic cells. Furthermore, even though no detailed studies are available, the number of liver cell receptors may well increase in the course of development and, therefore, be larger in the adult. Actually Adamson et aL (7) report EGF binding data in fetal mouse livers that suggest that numbers of receptors are even lower in fetal cells than those determined in our experiments with newborn-mouse livers. EGF Binding Activity of Cells from Organs Other Than Liver. Single-cell suspensions of thymus, spleen, heart, lungs, and brain obtained from mutant and normal littermate newborn mice were examined. The results (Table 2) show no difference between mutant and normal in the number of fmol bound per 106 cells in each of the three experiments performed. The mean values from all three experiments expressed in fmol bound per 106 cells are 0.532 0.042 (SD) and 0.519 0.082 (SD), respectively, for normal and mutant. Thus, the reduction of reX
±
±
Table 1. Saturation binding of 1251I-EGF to liver cells from newborn mice Normal cells (7) Mutant cells (6) Kd, M x 10-9 2.7 1.05 2.5 1.10 2.43 1.0 Bmax, fmol per 106 cells 0.58 0.08 ±
±
±
±
The values are means SD of four experiments. The number of animals is given in parentheses. ±
1~~
( 1 2 3 125I-EGF specifically bound, fmol per 106 cells
FIG. 3. (A) Effect of 125I-EGF concentration on binding to mutant (A) and normal (e) hepatocytes. Indicated concentrations of labeled EGF (0.6-4.2 nM; 131 ACi/,gg) were added to duplicates of 106 hepatocytes in the presence or absence of 125-fold excess nonradioactive EGF. Specific binding was determined after 2 hr of incubation at 0C. (B) Scatchard analysis of EGF saturation binding experiments to homozygous mutant (A) and normal (A) liver cells.
ceptor numbers found in mutant liver cell suspensions is not manifested in other organs. Interestingly, the nonspecific binding in cells from organs other than liver was 2-3 times higher
than that found for liver cells. Competitive Displacement of '"I-Glucagon from Hepatocytes. Competition of increasing concentrations of nonradioactive glucagon with 125I-glucagon for specific membrane binding sites was measured in intact normal and mutant hepatocytes. Mutant cells bound -30% of the amount of '"I-glucagon bound by cells from normal littermates (Fig. 4A)-i.e., 0.43 ± 0.11 (SEM) pg per 106 mutant cells and 1.43 ± 0.12 (SEM) pg per 106 normal cells, respectively (Table 3). Normalization of the data reflecting the percentage of maximal binding of '"I-glucagon as a function of nonradioactive glucagon concentration revealed that similar amounts of glucagon were required for a particular fraction of displacement by both mutant and normal hepatocytes (Fig. 4B). These results indicate that the differences observed in '"I-glucagon binding between normal and mutant hepatocytes result from a reduction of the number of glucagon receptors rather than an alteration in receptor affinity. Glucagon Binding Activity of Cells from Organs Other Than Liver. Single-cell suspensions from mutant and normal litterTable 2. 125I-EGF binding activity of cells from pools of organs other than liver Specific binding, fmol per 106 cells Normal cells Experiment Mutant cells 1 0.496 ± 0.050 (2) 0.474 ± 0.020 (3) 2 0.578 ± 0.041 (4) 0.613 ± 0.055 (4) 3 0.522 ± 0.042 (1) 0.469 ± 0.010 (1) The values are means + SD, and the number of animgls is given in parentheses. Nonspecific binding was approximately 1,100 cpm for both mutant and normal cells.
Genetics: Shaw and Gluecksohn-Waelsch
5382
Proc. Nad Acad. Sci. USA 80 (1983)
2r C.
1
bD ci-
1-
0.1
1
10
100
C)a
Cf)
A
C)
0.01
1
10
100
Total glucagon, nM FIG. 4. (A) Competitive displacement of 125I-glucagon from mutant (A) and normal (A) hepatocyte receptors with nonradioactive glucagon. Binding assays were carried out with 106 cells in each assay with 0.67 nM '251-glucagon in the presence or absence of 1,000-fold excess nonradioactive glucagon. (B) Data fromA are normalized to 100% maximal binding activity as determined in the absence of nonradioactive glucagon. A single curve is drawn through the data points obtained from mutant (A) and normal (e) hepatocytes. mate organ pools, excluding the liver, bound similar amounts of 125I-glucagon, 0.38 ± 0.02 (SEM) pg per 106 mutant cells and 0.29 ± 0.04 (SEM) pg per 106 normal cells, respectively (Table 3). Single-cell suspensions of hearts alone, known to be rich in glucagon receptors, yielded results that also showed similar values of specific glucagon binding in mutant and normal cellsi.e., 1.16 + 0.13 (SEM) and 0.99 0.21 (SEM) pg per 106 cells, respectively. ±
DISCUSSION The results show that binding of EGF by mouse hepatocytes homozygous for a deletion in chromosome 7 is decreased to -20% of normal because of a reduction of numbers of EGF receptors with, however, normal affinities. Similarly reduced is the binding of glucagon to mutant hepatocytes, which amounts to about 30% of normal. These reductions resemble those reported previously for the binding of insulin as well as a glucocorticoid hormone (2, 8). In all instances, the presence of significant levels of remaining receptors with normal affinities indicates that the respective structural
genes
encoding the
receptor proteins must
Table 3. 125I-Glucagon binding activity in single-cell suspensions of liver and other organs Specific binding, pg per 106 cells Normal cells Mutant cells Liver 1.43 ± 0.12 (6) 0.43 ± 0.11 (6) Heart 0.99 ± 0.21 (3) 1.16 ± 0.13 (3)
Organ pools* (excluding liver)
0.29 ± 0.04 (2) 0.38 ± 0.02 (2) The values are means ± SEM, and the number of animals is given
in parentheses. *
Includes brain, thymus, spleen, heart, and lungs.
be intact and, therefore, located outside the region of the deleted genome. Attempts to identify a gene-controlled mechanism that is possibly responsible for the regulation of receptor levels and hormone binding and is affected in the deletion homozygotes must take into account several facts. To begin with, one of the four receptors studied-i.e., that for glucocorticoids-is compartmentalized in the cytosol, whereas insulin, EGF, and glucagon receptors are bound to the plasma membrane. Insulin and EGF receptors are known to share the same coated pit as well as the pathway of internalization after occupancy by the respective hormones (3). Furthermore, the reductions of EGF and glucagon binding, and very likely that of insulin binding as well, are liver specific while other cell types show normal levels of hormone binding. Therefore, it is possible that specific membrane abnormalities of mutant liver parenchymal cells account for the reductions of membrane-bound receptors. Previous electron microscopic studies of deletion homozygotes actually demonstrated severe ultrastructural alterations of membrane integrity in the rough endoplasmic reticulum and the Golgi apparatus of hepatocytes in contrast to the normal ultrastructure of all other cell types (9). Even though in these studies the plasma membrane of hepatocytes appeared ultrastructurally normal, the quantitative reduction of membrane-bound receptors for insulin, EGF, and glucagon reflects a functional abnormality of this membrane. Decrease of receptor numbers could be due to defective synthesis, processing of receptors in the ultrastructurally abnormal microsomal membranes, and/or defects of their transport to the plasma membrane. It is well known that EGF exerts various biological effects that include the induction of biochemical and morphological cellular activities (10). Therefore, it is conceivable that a deficiency of available intracellular EGF resulting from the reduction of receptor numbers might interfere with normal morphological and biochemical differentiation of parenchymal liver cells. Decreased availability of glucagon resulting from a decrease in receptor numbers may be another factor contributing to the deficiencies of enzyme differentiation. It appears likely that defects of mechanisms normally involved in the regulation of specific gene expression characterizing the liver cell type are ultimately responsible for the errors of differentiation observed in this mutant system. These mechanisms include membrane-mediated receptor-hormone interactions apparently controlled by genes, some of which map in the chromosomal region covered by the lethal albino deletions. We thank Charles S. Rubin, E. Richard Stanley, Luz S. Teicher, and Allan Wolkoff for help and discussion, Linda Hall for the use of her computer, and Craig Robinson for help with the necessary breeding experiments. This work was supported by grants from the National Institutes of Health (GM27250) and the American Cancer Society (CD38) and by American Cancer Society Institutional Grant IN-28-X. 1. Gluecksohn-Waelsch, S. (1979) Cell 16, 225-237. 2. Goldfeld, A. E., Rubin, C. S., Siegel, T. W., Shaw, P. A., Schiffer, S. G. & Gluecksohn-Waelsch, S. (1981) Proc. Natl Acad. Sci. USA 78, 6359-6361. 3. Maxfield, F. R., Schlessinger, J., Schechter, Y., Pastan, I. & Willingham, M. C. (1978) Cell 14, 805-810. 4. Amsterdam, A. & Jamieson, J. D. (1972) Proc. NatL Acad. Sci. USA 69, 3028-3032. 5. Carpenter, G. & Cohen, S. (1976) J. Cell Biol 71, 159-171. 6. Scatchard, G. (1949) Ann. N.Y. Acad. Sci. 51, 660-672. 7. Adamson, E. D., Deller, M. J. & Warshaw, J. B. (1981) Nature (London) 291, 656-659. 8. Goldfeld, A. E., Firestone, G. L., Shaw, P. A. & GluecksohnWaelsch, S. (1983) Proc. Nati Acad. Sci. USA 80, 1431-1434. 9. Trigg, M. J. & Gluecksohn-Waelsch, S. (1973)J. Cell BioL 58, 549-
563. 10. Carpenter, G. & Cohen, S. (1979) Annu. Rev. Biochem. 48, 193216.
