electrodynamic stresses may also manifest themselves in an apparent viscosity ... hr in suspension for all methocel concentrations reported. the cell cytosol (Fig ...
Proc. Natl. Acad. Sci. USA Vol. 93, pp. 1786-1791, March 1996 Biophysics
The nucleation of receptor-mediated endocytosis DAVID A. EDWARDS*t, KEITH J. GOOCH*, INGRID ZHANG*, GARETH H.
MCKINLEYt, AND ROBERT LANGER§P
*Department of Chemical Engineering, Pennsylvania State University, 204 Fenske Laboratory, University Park, PA 16802; tDivision of Applied Sciences, Harvard University, Pierce Hall, Cambridge, MA 02138; §Department of Chemical Engineering, Harvard-Massachusetts Institute of Technology, Division of Health Sciences, and Whitaker College of Health Sciences, Cambridge, MA 02139; and IDepartment of Surgery, Children's Hospital, Boston, MA 02115
Contributed by Robert Langer, November 13, 1995
MATERIALS AND METHODS Reagents. 1125-labeled human transferrin (Tf) was purchased
A theory of the mechanical origins of receptorABSTRACT mediated endocytosis shows that a spontaneous membrane complex formation can provide the stimulus for a local membrane motion toward the cytosol. This motion is identified with a nucleation stage of receptor-mediated endocytosis. When membrane complexes cluster, membrane deformation is predicted to be most rapid. The rate of growth of membrane depressions depends upon the relative rates of approach of aqueous cytosolic and extracellular fluids toward the cell membrane. With cytosolic and extracellular media characterized by apparent viscosities, the rate of growth of membrane depressions is predicted to increase as the extracellular viscosity nears the apparent viscosity of the cytosol and then to decrease when the extracellular viscosity exceeds that of the cytosol. To determine whether these trends would be apparent in the overall endocytosis rate constant, an experimental study of transferrin-mediated endocytosis in two different cell lines was conducted. The experimental results reveal the same dependence of internalization rate on extracellular viscosity as predicted by the theory. These and other comparisons with experimental data suggest that the nucleation stage of receptor-mediated endocytosis is important in the overall endocytosis process.
from Amersham. All other chemicals, including human apotransferrin and methyl cellulose (80 kDa), were obtained from
Sigma. Cell Culture and Preparation. Human erythroleukemia K-562 cells were grown in RPMI 1640 medium/penicillin at 50 units/ml/streptomycin at 0.05 mg/ml/2 mM L-glutamine/ 10% fetal bovine serum. Chinese hamster ovary (CHO) cells transfected with human Tf receptor, from Timothy McGraw (Columbia University, New York), were grown in McCoy's 5A medium/penicillin at 50 units/ml/streptomycin at 0.05 mg/ ml/2 mM L-glutamine/5% fetal bovine serum. Immediately before each experiment, all cells were washed two times with 40 ml of ice cold buffer (25 mM Hepes/150 mM NaCl, dextrose at 1 mg/ml/bovine serum albumin at 1 mg/ml, pH 7.4), and centrifuged at 600 x g at 4°C for 10 min. Methyl Cellulose Preparation and Characterization. Methyl cellulose solutions were made in buffer with the dispersion technique formulated by Dow to achieve a final concentration of 0.0 to 1.8%. Briefly, after one-third volume of the buffer was heated to 90°C, methyl cellulose powder was added to the liquid and agitated until the particles were thoroughly wetted and evenly dispersed. The remainder of the buffer was then added to the methyl cellulose at 4°C. Agitation was maintained at 4°C for another 20 min. One day before the experiments, 1125-labeled Tf (0.03 ,uCi/ml; 1 Ci = 37 GBq) and nonlabeled Tf (50 nM) were added to each concentration of methyl cellulose solution and thoroughly mixed. The apparent viscosity of the methocel solutions was measured in a controlled stress rheometer (TA Instruments CSL500) using a cone-and-plate geometry. All data shown in Fig. 1 have been obtained by exponentially reducing the externally imposed stress at a constant rate from 100 Pa to 1 Pa. Endocytosis. Steady-state ratios of total internalized Tf to surface-bound Tf were determined as follows: At 4°C, 2 x 106 cells were resuspended in 1 ml of methyl cellulose solution (concentration ranging from 0.0 to 1.8%) containing radiolabeled and nonlabeled Tf. A Pasteur pipette was used to gently mix the cells and the methyl cellulose solution. Immediately after warming in a 37°C water bath for 5 min, the samples were transferred to a 37°C thermal hood, where slow rotation of the samples was maintained. After 1 hr, endocytosis was terminated by rapid addition of 12 ml of ice-cold Hank's balanced salt solution (HBSS), followed by three more washings of 12 ml of ice-cold HBSS and subsequent centrifugation (1200 x g at 4°C). After washing, cells were divided into two equal volumes. The cells from both sets of aliquots were pelleted. The samples from the first set were counted with a y counter (model 1274 Ria Gamma, LKB Wallac, Finland) to obtain the total radioactivity. The total internal (In) and surface (Sur) radioactivity levels were determined from the second set of samples using
The binding of ligands or assembly proteins to surface receptors of eukaryotic cell membranes initiates or accompanies a cascade of nonequilibrium phenomena culminating in the cellular invagination of membrane complexes within clathrincoated vesicles (1-6). This process has been referred to as receptor-mediated endocytosis (RME), at least since 1974 (1). Beyond playing a central role in cellular lipid trafficking (7), RME is the primary means by which macromolecules enter eukaryotic cells; therefore a first-principle understanding of RME can have direct bearing upon the success of gene (8) and cancer (9) therapies, as well as other biomedical applications. For these applied and other basic scientific reasons, RME is being actively studied (10-14). While considerable effort has been devoted to the identification and biochemical characterization of the early and later stages of RME, ranging from formation of a clathrin-coated pit to snap-off of a coated vesicle, less attention has been given to the process whereby a membrane depression is initially formed at the outset of RME-that is, what is the mechanism by which a spontaneous thrust of the cell membrane toward the cytosol occurs? The present article addresses this question in the context of a "nucleation stage" of RME. This terminology is intended to emphasize a conclusion that the driving force for the initial spontaneous thrust of the membrane toward the cytosol is not the overall metabolic driving force of endocytosis; rather, it is related to energy liberated by one or more (of many possible) exothermic membrane-binding reactions that precedes or accompanies formation of a membrane depression.
Abbreviations: RME, receptor-mediated endocytosis; EGF, epidermal growth factor; Tf, transferrin; WT, wild type; In radioactivity, total internal radioactivity; Sur radioactivity, total surface radioactivity. tTo whom reprint requests should be addressed.
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.
1786
Biophysics: Edwards et al.
Proc. Natl. Acad. Sci. USA 93 (1996)
100.00
10.((
1.00 1.25%,i
0.10
1.01Cc T
37C
a
0.01
100
10 Shear stress, Pa
FIG. 1. Apparent viscosity versus applied shear stress forr the different methocel solutions (0, 1, 1.25, 1.5, 1.7, and 1.8%)). A characteristic value of the cell viscosity /lint is shown. The veirtical dashed lines represent estimates of the maximum and minimum Iforce values delivered by a cell on the extracellular fluid by invagin'ating pits-that is, assuming an invaginating pit diameter of 0.1 t,m and;a cell diameter of 10 ,um, these maximum and minimum force values cain be computed by Fcell AO' X (perimeter of pit)/(area of cell membr; ane), using the membrane tension depression estimates made in the I graph preceding Eq. 6. It should be noted that these bounds estimates, the precise values of which can be expected to vary bet) cell lines and receptor-ligand pairs.
pare wreen
the method of ref. 15. Briefly, the pellets were incubated' with 0.5 ml of trypsin solution (25 mM Hepes/150 mM NaCl/ with bovine trypsin at 1 mg/ml, pH 7.4) at 4°C. The supernattants and the pellets were separated by centrifuging at 3000 x g at 4°C for 15 min. Two milliliters of HBSS was used to washi the pellets. The final pellets were counted in a y counter to give the In value, while the supernatants from each washingvwere combined to give the Sur value. Internalization values were represented in the form c f In divided by Sur per million cells [which we term steady-sstate In/Sur (16)]. Steady-state values of In/Sur for cells of a griven type in various methocel media provide relative estimate-s of endocytosis rate (16) for Tf-mediated endocytosis (17), particularly when it is assumed that Tf recycling rates are i]ndependent of the rheology of the extracellular medium. Dc)wnregulation of surface-bound Tf receptor numbers is nott expected for the duration of experiments given the prefereintial recycling of Tf receptors to the cell surface after internalization (14).
RESULTS AND DISCUSSION Theory of Nucleation of RME. Consider an immobile portion of a cell membrane whose lateral dimension (L) is much smaller than the diameter of the cell (D), while much larger than the sum of the thicknesses of the cell membrane (h) and cell coat or glycocalyx (1). In particular, if D 10 itm, h 50 A, and 1 10 nm (18), it is possible to choose L 0.4 itm, for example, such that E = L/D
