- Email: [email protected]
Turnover of cell-surface macromolecules in cultured dog tracheal epithelial cells
336
Biochimica et Biophysica Acta, 966 (1988) 336-346
Elsevier BBA 22955
Turnover of cell-surface macromolecules in cultured dog tracheal epithelial cells
Itsuo Iwamoto, Jay A. Nadel, Shabtai Varsano and Lennart S. Forsberg Cardiovascular Research Institute and Departments of Medicine and Physiology, University of California, San Francisco, San Francisco, CA (U.S.A.)
(Received 21 December 1987) (Revised manuscript received11 April 1988)
Key words: Cell surface macromolecule;Glycocalyx;Sulfated macromolecule;Epithelial cell; (Dog trachea) We studied the metabolism of sulfated cell-surface macromolecules in dog tracheal epithelial cells in primary culture. To examine the time-course and rate of appearance of sulfated macromolecules at the cell surface, the cells were pulsed with 35SO4 for short periods (5-15 min), and the incubation medium was sampled for spontaneously released macromolecules (basal secretions) and for release induced by trypsin (trypsin-accessible secretions). Trypsin-accessible ass-labeled macromolecules appeared on the cell surface within 5-10 min, increased linearly, and plateaued by 40 min; the median transit time for 35S-labeled macromolecules to reach the cell surface was 21 min. 35S-labeled macromolecules in basal secretions increased with a similar time-course, reaching a plateau by 40 rain. Incorporation of [3Hlserine into the protein moiety of trypsin-accessible macromolecules occurred more slowly; trypsin-accessible 3H-labeled macromolecules were barely detectable at 1 h and increased to a maximum after 2 h, suggesting the presence of a preformed pool of nonsulfated core protein. Pretreatment with cycloheximide, an inhibitor of protein synthesis, decreased trypsin-accessible 35S-labeled macromolecules log-linearly depending on the duration of pretreatment providing an estimate of the rate of depletion of the core protein pool ( i l l 2 = 32 min). During continuous exposure to 3SSO4, ass-labeled macromolecules accumulated on the cell surface (trypsin-accessible compartment) for 16 h, at which point the cell-surface pool was saturated ( t l / 2 = 7.5 h). After pulse-labeling the cells with 35SO4 for 15 min, the 35S-labeled macromolecules disappeared continuously from the cell surface ( t l / 2 = 4.6 h), and 79% of the radioactivity was recovered in the medium as nondialyzable macromolecules. Release of the 3sS-labeled macromolecules from the cell surface was abolished at 4 ° C , indicative of an energy-dependent process, but multiple proteinase inhibitors did not affect the release. We conclude that sulfate is metabolized rapidly into epithelial cell-surface macromolecules, which accumulate continuously into a relatively large cell-surface pool, before they are released by an undefined energy-dependent mechanism.
Abbreviation: DME, Dulbecco's modified Eagle's medium. Correspondence: J.A. Nadel, Cardiovascular Research Institute, Box 0130, Universityof California, San Francisco, CA 94143-0130, U.S.A.
Introduction
Cell-surface macromolecules, which consist mainly of proteoglycans and mucin-type glycopro-
0304-4165/88/$03.50 © 1988 ElsevierSciencePublishers B.V. (BiomedicalDivision)
337 teins, are present at the apical surface of the airway epithelium [1-5] and are believed to have many biological functions, including maintenance of mucociliary flow and protection from infection [6]. In the respiratory tract, the cell-surface macromolecules form a discrete region, or glycocalyx, which is readily distinguished from the mucus layer by histocytochemical and autoradiographic procedures [2-6]. Many studies have focused on the biochemistry of the mucus layer, but many aspects of the underlying epithelial glycocalyx remain unclear. We have recently shown that primary cultures of dog tracheal epithelial cells, which lack prominent secretory storage granules, incorporate [35S]sulfate into an extensive cellsurface glycocalyx [7,8]. The major component is a sulfated mucin glycoprotein (molecular weight >/ 10 6) which has several unique structural features, including a high proportion of O-glycosidically linked glycosaminoglycans of the poly(N-acetyllactosamine) type [7,8]. We found that this sulfated mucin accounted for over 70% of the cell-surface macromolecular sulfate in the cultured cells. The sulfated macromolecules appear to be available at the cell surface in a trypsin-accessible form, because treating the cells with trypsin causes an immediate 5-fold increase in the recovery of these macromolecules in the medium. However, the rates of synthesis, degradation, and release of this and other cell-surface macromolecules from airway epithelial cells are unknown. In the present study, we designed pulse-chase experiments to examine the metabolism of the sulfated cell-surface macromolecules in primary cultures of dog tracheal epithelial cells. Using [35S]sulfate as precursor, we determined the rates of appearance, accumulation, and release of sulfated macromolecules at the cell surface. In addition, dual-labeling studies with [3H]serine and [35S]sulfate were used to examine in detail the time-course of synthesis of the protein moiety of the macromolecules. Measurement of [35S]sulfate incorporation in the presence of cycloheximide, an inhibitor of protein synthesis, provided an estimate of the rate of depletion of the intracellular protein pool for the macromolecules. We also examined possible mechanisms of release of the sulfated macromolecules from the cell surface.
Methods
Materials Dulbecco's modified Eagle's medium (DME), Ham's F-12 medium (F-12), fetal calf serum, nonessential amino acids (alanine, asparagine, aspartic acid, glutamic acid, glycine, proline and serine), and antibiotics (penicillin, streptomycin, gentamicin and fungizone) were obtained from the Cell Culture Facility, University of California, San Francisco. Crude collagenase from Clostridium h&tolyticum (type I), human placental collagen, bovine pancreatic trypsin, cycloheximide, soybean trypsin inhibitor, aprotinin, leupeptin, pepstatin, phosphoramidon, captopril and bestatin were purchased from Sigma Chemical Co. (St. Louis, MO). Polycarbonate filters (0.8 /xm pore size, 25 mm diameter) were purchased from Nucleopore Corp. (Pleasanton, CA). Dialysis membranes (molecular weight cutoff, 12000-14000) were purchased from Spectrum Medical Industries (Los Angeles, CA). Na235SO4 (spec. act., 43 Ci/mg sulfate) and L-[3H]serine (spec. act. 8 Ci/mmol) were purchased from ICN Radiochemicals (Irvine, CA). Isolation and culture of tracheal epithelial cells and conditions of incubation Dog tracheal epithelial cells were isolated and cultured as previously described [9]. Briefly, the trachea was removed from mongrel dogs (18-30 kg) after intravenous sodium pentobarbital anesthesia (30 mg/kg). The epithelium was stripped from the submucosa with forceps, minced with scissors, and digested with 0.02% crude collagenase at room temperature for 3 h. The epithelial cells in the collagenase solution were collected by centrifugation and suspended in medium consisting of 50% DME and 50% F12 (DME/F12) that contained 5% fetal calf serum, 1% nonessential amino acids, and antibiotics (105 U/I penicillin, 100 mg/1 streptomycin, 100 mg/1 gentamicin, and 500 ~g/1 fungizone). We refer to this medium as complete medium. The viability of the cells recovered was 88.5 + 2.2% (mean + S.D.) (n = 24 tracheas) as determined by erythrosin B dye exclusion.
338
The isolated cells were plated on collagencoated polycarbonate filters (diameter, 25 mm) at a density of 4 . 1 0 5 viable cells/cm 2 in 2.5 ml of complete medium in 35-mm plastic dishes. Dishes were placed at 3 7 ° C in 6% CO 2 in air in a CO 2 incubator. The complete medium in the dish was changed after the first 24 h and then every 2 - 3 days. Cell growth and monolayer formation were assessed by observing the cells on the transparent area of the dish surrounding the filter. The cells became confluent 4 - 5 days after the start of culture, and 7- to 9-day-old cells were used for experiments. The cells had typical characteristics of epithelial cells and contained no secretory granules in their cytoplasm [9]. Unless otherwise stated, the experiments were performed at 37 ° C, and the cells were incubated in D M E / F 1 2 medium throughout the experiment. Each experiment was performed in duplicate; reported values are the average of the two measurements. We refer to the time at the start of radiolabeling as time 0 rain.
viously shown that the conditions used for trypsin incubation do not cause cell leakage or detachment of cells from the filter disks [7]. To determine the time-course of radiolabel incorporation and macromolecule accumulation on the cell surface, it was necessary to isolate an entire cell-surface pool of completed macromolecules instantly, at any series of time points following introduction of radiolabel, without waiting for the metabolic release of these components from the cell surface (which occurs slowly, tl/2 = 4.6 h; see Results). Thus, 'trypsin-accessible' secretions were analyzed, because treating the cells with trypsin proved to be an effective procedure in which an entire pool of cell-surface macromolecules can be harvested at any one instant and their amounts quantitated. This allows measurement of the rates of incorporation and accumulation and, at the same time, eliminates interference due to the ongoing basal rate of degradation (which includes release from the cell surface and internalization).
'Basal" and "trypsin-accessible' secretions
Time-course of 35S04 incorporation into cell-surface macromolecules
Previous studies have shown that sulfated macromolecules are released spontaneously from cultured airway epithelial cells, giving rise to what we refer to as 'basal' secretions. The major sulfated macromolecule of these basal secretions is a mucin glycoprotein (molecular weight > / 1 0 6) [7,8]. Prior to release, the sulfated macromolecule resides on the cell surface in a trypsin-accessible domain, or compartment; treatment of the cells with trypsin causes an immediate release of this macromolecule into the medium, yielding the 'trypsin-accessible' secretions. To study the turnover rates of epithelial cell-surface macromolecules, both 'basal' and 'trypsin-accessible' secretions were analyzed. We obtained basal secretions by incubating epithelial cells with radiolabeled precursors (35SO4 or [3H]serine), and harvesting radiolabeled macromolecules in the culture medium at various subsequent times. Trypsin-accessible secretions were obtained by incubating the cells with radiolabeled precursors as in the basal studies, but we added 125 U / m l of trypsin at 3 7 ° C for 5 - 1 0 min at the end of each chasing time. Radiolabeled macromolecules present in the culture medium following this treatment were then harvested. We have pre-
To assess the time-course and rate of appearance of sulfated cell-surface macromolecules, cells were pulsed briefly with 35SO4, and the trypsin-accessible 35S-labeled macromolecules released into the medium were collected at various chase times. Specifically, cells grown on polycarbonate filters were exposed to 1 5 0 / ~ C i / m l of 35SO4 in 2 ml of D M E / F 1 2 in a 35-mm dish at 37 ° C for 5 min and then washed four times in D M E / F 1 2 medium. Basal and trypsin-accessible 35S-labeled macromolecules were then collected by incubating the cells for 5 min with trypsin (trypsin-accessible) or without trypsin (basal) after chase times of 5, 10, 15, 20, 30, 40 and 60 rain.
Time-course of [3H]serine incorporation into trypsin-accessible macromolecules To determine the time-course of core protein synthesis for the cell-surface macromolecules, the rate of incorporation of [3H]serine into trypsinaccessible macromolecules was determined. Cells were washed once in serine-free D M E before incubation, exposed simultaneously to 50 ~tCi/ml of 35SO4 and 40 /~Ci/ml of [3H]serine in serine-free
339 D M E for 15 min and then washed in D M E / F 1 2 medium. Trypsin-accessible secretions were collected after chase times of 0.5, 1, 2, 3 and 5 h. The trypsin-accessible secretions from each time point were concentrated by lyophilization and analyzed on a column of Sepharose CL-4B as described previously [7,8]. The amounts of 3H-labeled macromolecules and 35S-labeled macromolecules were determined by monitoring the activity of each isotope recovered in the void volume fraction of the column. We have previously shown that the major macromolecular species in this void volume fraction is a sulfated mucin glycoprotein ( M r >~ 106), which contains alkaline-borohydride-labile O-glycosidically linked glycans of the poly(Nacetyllactosamine) type [7,8]. In these experiments, basal secretions were not evaluated.
Time-course of depletion of core protein pool To determine the pool size of the core protein of the sulfate cell-surface macromolecules, we took advantage of the fact that cycloheximide inhibits protein synthesis [10]. Because the core protein pool is depleted continuously by synthesis and secretion of the sulfated macromolecules, the longer periods of incubation with cycloheximide resulted in progressive depletion of this pool. Therefore, to determine the time-course of depletion of the core protein, we pretreated the cells with 0.5 mM cycloheximide at 37 ° C for 0, 0.5, 1, 1.5, 2, 3, 4 or 6 h. Then we incubated the cells with 150 /~Ci/ml 35SO4 for 15 rain and washed the cells. Basal and trypsin-accessible 35S-labeled macromolecules were harvested after a 60-min chase (the time at which all of the 35S-labeled macromolecules were shown to reach the cell surface after a 15-min pulse; see Fig. 2).
Time-course of accumulation of sulfated macromolecules on the cell surface To assess the time-course of accumulation and total pool size of sulfated macromolecules on the cell surface (trypsin-accessible compartment), we determined the amounts of 35S-labeled macromolecules after prolonged labeling with 35SO4. The cells were incubated with 50 # C i / m l of 35SO4 in complete medium in a CO 2 incubator (6% CO 2 in air) for various periods (4, 8, 12, 16, 20 or 24 h). The cells were washed four times and chased for 1
h. The cells were then incubated with trypsin for 10 min, and trypsin-accessible 35S-labeled macromolecules were collected in the culture medium. In parallel experiments, trypsin was omitted and basal (spontaneously released) secretions were measured in the medium.
Time-course of disappearance of sulfated cell-surface macromolecules We studied simultaneously the time-course of disappearance of sulfated macromolecules from the cell surface and the time-course of recovery of sulfated macromolecules in the incubation medium. The cells were incubated with 150 /~Ci/ml of 35SO4 for 15 min and washed. Then, at 60 rain after the beginning of incubation (by which time we assumed that 35SO4 label was fully incorporated into cell-surface macromolecules), we incubated the cells in 2.5 ml of D M E / F 1 2 at 37 ° C in a CO 2 incubator for various times (0, 2.5, 5, 10, 15, 20 or 25 h). At the end of each chase time, we collected the medium to measure total 3SS as well as 35S-labeled macromolecules. Next we measured either basal secretion from the cells (one series of experiments) or trypsin-accessible secretions after adding trypsin for 10 min (another series of experiments). The amount of total 35S recovered in the medium (in all molecular species) at the end of the 25 h chase period was compared to the total amount of 35S present in the cells immediately prior to the first chase period (at time 0 h).
Effect of temperature and of proteinase inhibitors on release of sulfated cell-surface macromolecules To determine whether active cell metabolism is required for the basal release of the sulfated cellsurface macromolecules, we examined the effect of temperature. Cells preincubated with 35SO4 (150 /~Ci/ml for 15 min) were incubated in D M E / F 1 2 either at 3 7 ° C or at 4 ° C from time 60 min for 6 h. At the end of incubation, medium was collected for measurement of released 35S-labeled macromolecules. Then the cells incubated at 4 ° C were rewarmed to 37 ° C. Trypsin was then added to all cells for 10 min, and the trypsin-accessible 35Slabeled macromolecules were measured. Because exogenous proteinases cleave the cellsurface macromolecules [7], we also examined the
340 effect of proteinase inhibitors on the disappearance of the sulfated macromolecules from the cell surface. We reasoned that if airway epithelial cell-surface macromolecules are normally released by proteolytic cleavage, then the appropriate proteinase inhibitor should inhibit basal secretion and cause accumulation of the macromolecules, thus increasing subsequent release by trypsin. Cells preincubated with 35SO4 (150 /~Ci for 15 min) were incubated at 37 °C from time 60 min for 6 h. The seven proteinase inhibitors added were soybean trypsin inhibitor (100 ~ g / m l ) and aprotinin (30/~g/ml) to inhibit serine proteinases, leupeptin (10 ~ g / m l ) to inhibit thiol proteinases, pepstatin (10 /~g/ml) to inhibit carboxyl proteinases, phosphoramidon (10 -5 M) to inhibit neutral endopeptidase, captopril (10 -5 M) to inhibit kininase II, and bestatin (10 -5 M) to inhibit aminopeptidases. After the cells were washed four times, trypsin-accessible 35S-labeled macromolecules were measured. As controls, the same study was performed in the absence of inhibitors.
Measurements of ~sS-labeled maeromolecules and of total 35S The amounts of 35S-labeled macromolecules were measured by counting the activity of nondialyzable 35S in the samples or by measuring radioactivity eluting in the void volume fractions of Sepharose CL-4B chromatography (see below). For dialysis, the sample was dialyzed against 5 1 of distilled water (the water was changed every 6 h) for 3 days at 4 ° C using dialysis membranes (molecular weight cutoff, 12 000-14000). Then the activity of 35S in the dialyzed sample was counted by a beta scintillation counter. The amounts of total 35S in the incubation medium were measured by counting the radioactivity of 3SS in the incubation medium without dialyzing. The amounts of total 35S in the pulselabeled cells were measured by counting the radioactivity of 35S of the cells after disrupting the cells by adding distilled water. The amounts of 35S-labeled and 3H-labeled macromolecules after 35SO4 and [3H]serine duallabeling were determined by counting the activity of 35S and 3H recovered in the void volume fraction of Sepharose CL-4B gel filtration chromatog-
raphy, performed as described previously [7]. The incubation medium collected at each chasing time was concentrated by lyophilization and applied to a column (1 × 20 cm) of Sepharose CL-4B. Chromatography was performed under associative conditions (0.1 M NaOAc buffer (pH 5.8)/0.02% sodium azide), and aliquots of fractions were assayed for radioactivity in the scintillation counter with windows optimized for dual-label counting. The amount of 35S-labeled or 3H-labeled macromolecules was calculated from the sum of radioactivities of 35S and 3H recovered in the Sepharose CL-4B void volume peak, representing macromolecules having an apparent molecular weight >/ 1 • 1 0 6.
The amounts of 35S-labeled and 3H-labeled macromolecules and of total 35S are expressed as the activity of 35S or 3H in dpm produced by epithelial cells grown on one polycarbonate filter. Counting efficiencies for 3H in dilute buffers and culture mediums were approx. 31% as determined with quenched 3H standards. 35S counting efficiency was over 95% by assay of newly purchased radionuclide.
Analysis of data Data are summarized as mean_+ S.D. The half-time of turnover of trypsin-accessible 35Slabeled macromolecules in the epithelial cells was determined as the time when the amount of the macromolecules became one half of the maximum amount in the experiment. Data were analyzed by Student's paired t-test. Results
Time-course of 35S04 incorporation into cell-surface macromolecules 35S-labeled macromolecules were detectable in the cell-surface trypsin-accessible pool within 5-10 min following the introduction of radiolabel (Fig. 1). Trypsin-accessible 35S-labeled macromolecules increased linearly with a plateau at 40 min; the half-time of appearance of 35S-labeled macromolecules at the cell surface was 21 _+ 1 min. 35S-labeled macromolecules present in basal secretions also increased with a similar time-course, reaching a plateau after 40 rain.
341 1500-
a n a p p a r e n t m o l e c u l a r w e i g h t > / 1 0 6 as a s s e s s e d by Sepharose CL-4B chromatography. However, trypsin-accessible 3H-labeled macromolecules, of similar apparent molecular weight, appeared more slowly: t h e y w e r e d e t e c t a b l e a f t e r a 1 h chase, r e a c h i n g a m a x i m u m at 2 h (Fig. 2).
1250-
~,
•
750E
0")
250-
o
o
10
20
30
Duration
of Chase
40
50
60
(min)
Fig. 1. Time-course of appearance of sulfated macromolecules on the cell surface in cultured dog tracheal epithelial cells. Cultured cells were pulsed with 35SO4 for 5 min from time 0 min, and 35S-labeled macromolecules were sampled for 5 min in the culture medium after chasing for 5, 10, 15, 20, 30, 40 and 60 min. In one series of experiments, the samples were obtained without adding trypsin (basal secretion; ©). In a second series of experiments, trypsin was added for 5 min at the various times (trypsin-accessible secretions; e). Data are means _+S.D. for three experiments at each time point.
Time-course of [3H]serine incorporation into trypsin-accessible macromolecules After a 15-min trypsin-accessible peared rapidly on trypsin-accessible
p u l s e w i t h [ 3 H ] s e r i n e a n d 35SO4, 35S-labeled m a c r o m o l e c u l e s a p t h e cell s u r f a c e as e x p e c t e d . T h e 35S-labeled m a c r o m o l e c u l e s h a d
1500~ ~.~ 1250 1000 oJ 750 oE o 250
Time-course of core protein depletion in sulfated cell-surface macromolecules In the presence of cycloheximide, trypsin-accessible 35S-labeled m a c r o m o l e c u l e s d e c r e a s e d loglinearly, depending on the duration of cyclohexim i d e p r e t r e a t m e n t (Fig. 3). T h e h a l f - t i m e o f d e c r e m e n t o f t r y p s i n - a c c e s s i b l e 35S-labeled m a c r o m o l e c u l a r s e c r e t i o n (35SO4 i n c o r p o r a t i o n ) w a s 32 +4 min.
Time-course of accumulation of sulfated macromolecules on the cell surface W h e n t h e cells w e r e i n c u b a t e d w i t h 35SO4 for p r o l o n g e d p e r i o d s ( 4 - 2 4 h), 35S-labeled m a c r o m o l e c u l e s ( b o t h b a s a l a n d t r y p s i n - a c c e s s i b l e ) inc r e a s e d l i n e a r l y for 1 2 - 1 6 h a n d t h e n r e a c h e d a
•~. 5000"
~ i
1000 500
o =
o
E
s00
100.
0
i
0
1
2
3
4
.
i
.
i
•
i
5
Duration of Chase (hi
Fig. 2. Comparison of time-course of appearance of trypsinaccessible 35S-labeled and 3H-labeled macromolecules in cultured dog tracheal epithelial cells. Cultured cells were pulsed simultaneously with 3SSO4 and [3H]serine for 15 min from time 0 rain to assess the relative rates of appearance of sulfated macromolecules and of macromolecules with newly synthesized protein core. Then, after a chase period of 0.5, 1, 2, 3 and 5 h, trypsin was added for 10 rain, and the culture medium was analyzed for macromolecules containing 35S (O) and for those containing 3H (©) by Sepharose CL-4B chromatography. Data are means + S.D. for three experiments at each time point.
Duration of Cycloheximifle
Pretreatment (h)
Fig. 3. Effect of pretreatment with the protein synthesis inhibitor cycloheximide on secretion of sulfated macromolecules in cultured dog tracheal epithelial ceils. Cells were preincubated with cycloheximide for various periods of 0.5-6 h. Then the cells were pulsed with 3SSO4 for 15 min, and 35S-labeled macromolecules were measured in the culture medium at time 60 min. In one series of experiments, the samples were obtained without adding trypsin (basal secretion; ©). In a second series of experiments, trypsin was added for 10 rain (trypsinaccessible secretion, O). Data are means + S.D. for three experiments at each time point.
342
o
x
E=
Basal secretions, like trypsin-accessible secretions, decreased gradually over several hours (Fig. 5A, open squares).
70 60 50
-5
40
Effect of temperature and of proteinase inhibitors on sulfated cell-surface macromolecules
30 E o
20
E
10
Cells which were incubated at 4 ° C for 6 h showed essentially no release of sulfated macro-
0 0
4
8
12
16
20
24
Duration of Incubation of 35 SO 4 (h)
Fig. 4. Time-course of accumulation of sulfated macromolecules on the cell surface (trypsin-accessible compartment) in cultured dog tracheal epithelial cells. Cells were incubated with 3SSO4 for 4 - 2 4 h, and then 3sS-labeled macromolecules were measured. In one series of experiments, no trypsin was added (basal secretion; ©). In a second series of experiments, trypsin was added for 10 rain after the incubation with 35SO4 (trypsin-accessible secretion; O). Data are means_+S.D, for three experiments at each time point.
A. Disappearance from Cell Surface
4000-
3000 1
2000 o
1000 (13
i
0 0
steady-state level (Fig. 4). The half-time for accumulation of trypsin-accessible 35S-labeled macromolecules was 7.5 +_ 0.6 h of exposure to 35SO4.
Time-course of disappearance of sulfated macromolecules from the cell surface Trypsin-accessible 35S-labeled macromolecules decreased over 20 h with a half-time of 4.6 _+ 0.4 h (Fig. 5A, solid squares). The data were found to approximate a first order exponential decay, with a correlation coefficient ( r ) = 0.97 (determined by a least-squares fit to the transformed exponential curve). As the trypsin-accessible 35S-labeled macromolecules disappeared from the cell surface, nondialyzable 35S-labeled macromolecules recovered in the incubation medium increased proportionately (Fig. 5B, solid circles). More than 90% of the total 35S that had been present in the 35SO4-pulsed cells before the incubation was recovered in the medium after 25 h of incubation of the cells. In addition, 79 + 4% of the total 35S recovered in the medium after each incubation period was nondialyzable 35S-labeled macromolecules. These findings suggest that most of the trypsin-accessible 3sS-labeled macromolecules were released from the cell surface as whole macromolecules.
5
10
15
20
q 25
B. Recovery from Incubation Medium 6000-
4000-
2000-
0
5
1'0
1'5
2'0
2'5
Duration of Incubation (h)
Fig. 5. Time-course of disappearance of sulfated macromolecules from the cell surface in cultured dog tracheal epithelial cells. We studied simultaneously the time-course of disappea~a~,ce of~rypsin-accessible 3sS-labeled macromolecules (A, B) and of reco~;ery of 35S in the incubation medium (B). Cells were pulsed with 35SO4 for 15 min and rested until 60 min had elapsed, at which time m a x i m u m 35S was shown to be incorporated into cell-surface macromolecules. This is labeled as time zero. At various times (2.5-25 h), trypsin was added for 10 min and samples of culture m e d i u m were removed for measurement of 35S-labeled macromolecules. These trypsin-accessible macromolecules decreased with incubation time (A, B). This was cofffirt~ed by simultaneous measurement of the 35S that had accumulated in the culture medium during the time of incubation (B). Data for 35S-labeled macromolecules (o) and total 3SS (©) are shown separately. Data for basal secretion of 3sS-labeled macromolecules (sampled for 10 min after various times of incubation) are also shown (A, D). Data are means _+ S.D. for four experiments at each time point.
343 A. Trypsin-accessible Secretion 100 " 80" _e
60
0
E 0
40'
E 20'
cubated for 6 h at 4 ° C and rewarmed produced twice as many 35S-labeled macromolecules after addition of trypsin as the cells maintained at 3 7 ° C ( P < 0.025) (Fig. 6A). The addition of proteinase inhibitors to the incubation medium (see Methods for description) had no effect on the disappearance of trypsinaccessible 35S-labeled macromolecules from the cell surface ( P > 0.5).
0
37 ° C
4 ° C, followed by 3 7 ° C
B. Recovery from Incubation Medium A
100" 80"
a ~' 15 o
60
/ / / /
/ / / /
/ / / /
/ / / /
/ / / /
/ / z /
/ / / /
/////// / / / / / / / / / / / / / / /11/11/ I I / 1 1 / i II/1111 / / / / / / /
///////
o
40'
15
/ / I / I / / / / / / / / / / / i / / / ~
20 0
~.ztzz,.z
37°C
4° C
Fig. 6. Effect of temperature on release of sulfated macromolecules in cultured dog tracheal epithelial cells. Cells were pulsed with 35SO4 for 15 rain, rested for an additional 45 min, and then maintained for an additional 6 h at either 3 7 ° C (left columns) or at 4 ° C (right columns). At the end of the 6 h incubation, the cooled cells were rewarmed to 3 7 ° C , trypsin was added to all cells for 10 min, and the trypsin-accessible 35S-labeled macromolecules were measured (A). In cells that were incubated at 4 ° C, practically no basal secretion of 35Slabeled macromolecules occurred compared to the cells incubated at 3 7 ° C (B). The cells incubated for 6 h at 4 ° C and rewarmed produced twice as m a n y 35S-labeled macromolecules after the addition of trypsin as the cells maintained at 3 7 ° C (A). Data are means__+ S.D. in four experiments. * Significantly different from the mean obtained after incubation of the cells at 37 o C, P < 0.025.
molecules from the cell surface into the medium (basal secretion; Fig. 6B, right). The amount of 35S-labeled macromolecules recovered in the incubation medium after a 6 h incubation at 4 ° C was only 9 _+ 1% of that recovered in the medium after a 37°C incubation ( P < 0.025) (Fig. 6B, left). Following the 6 h incubation at 4 ° C, the cells were rewarmed and trypsin was added. The trypsin-accessible 35S-labeled macromolecules obtained were 90 _+ 6% of those obtained before incubation in the cold (Fig. 6A, right). Cells in-
Discussion Previously, we showed that dog tracheal epithelial cells in primary culture synthesize a sulfated mucin glycoprotein ( M r > / 1 0 6 ) , which accounts for over 70% of macromolecular sulfate when 35SO4 is used as the labeling precursor [7,8]. Ultrastructural and biochemical studies suggested that the sulfated mucin is located at the cell surface in a trypsin-accessible form [7,8]. In the present study, using the trypsin-accessible 35S-labeled macromolecules as an indicator of cell-surface mucin, we show that the sulfated macromolecules are rapidly produced, accumulate continuously into a large cell-surface pool, then undergo continuous release into the medium by an energy dependent mechanism. Incorporation of sulfate and transport to the cell surface of the sulfated macromolecules began within 5-10 min after the introduction of 35SO4 label (Fig. 1). The median transit time for newly sulfated macromolecules to reach the cell surface was estimated at 21 _+ 1 rain. These results show that sulfation and transport to the cell surface occur rapidly in tracheal epithelial cells and suggest that there is a preformed pool of nonsulfated macromolecules. The protein moiety of macromolecules eluting at the void volume of Sepharose CL-4B chromatography was synthesized and detectable at the cell surface within 2 h after uptake of an amino acid, as indicated by the time-course of [3H]serine incorporation into trypsin-accessible macromolecules (Fig. 2). The difference in the time-course of [3H]serine and 35SO4 incorporation further suggests the existence of an intracellular pool of nonsulfated precursors at various stages of completion, and could also reflect the transit time of this pool to the site of sulfation (presumably the Golgi region). A slower equilibration time for
344
the intracellular serine pool with [3H]serine could also contribute to this difference. The incorporation of both radiolabels into a single molecular species was not verified; however, both 3H and 3~S activities co-eluted at the void volume during Sepharose CL-4B chromatography. It is noteworthy that the major component of the Sepharose void volume fraction is a sulfated mucin, containing a high proportion of sulfated glycans which can be released from protein by alkalineborohydride treatment [8]. Following label incorporation, completed macromolecules accumulate on the cell surface into a trypsin-accessible pool, where they are probably intercalated into the membrane. The cell-surface pool was saturated after 16 h of continuous production of macromolecules during prolonged exposure to 35SO4 (Fig. 4). This suggests a relatively large pool of the sulfated macromolecules, consistent with our previous ultrastructural observations which showed a dense cell-surface coat containing anionic glycoconjugate [7]. Continuous release of the sulfated macromolecules from the cell surface is shown by the disappearance of trypsin-accessible 35S-labeled macromolecules from the surface over 20 h (tl/2 = 4.6_+ 0.4 h) and by the recovery of the macromolecules in the medium (Fig. 5). Thus, turnover of sulfated cell-surface macromolecules in tracheal epithelial cells is rapid and continuous. Our finding of rapid and continuous turnover, that is, time-courses and rates of production, accumulation on the cell surface, and release of cellsurface macromolecules in cultured tracheal epithelial cells, is in general agreement with previous autoradiographic studies of airway epithelia in situ. Thus in cat, ferret and rabbit tracheal organ culture, autoradiographic studies have shown that 35SO4-1abeled materials are transported from supranuclear regions to the epithelial surface within 1 h [11]. In intestinal columnar epithelial cells in rat and rabbit, cell-surface macromolecules were transported from the Golgi region to the apical surface within 30 min after pulsing with various radiolabeled sugars [12-14]. In ferret tracheal organ cultures, turnover of sulfated macromolecules in the epithelium was considerably more rapid (t~/2 = 2 h) than the turnover rates observed for submucosal glands (containing a high percentage of cells with prominent secretory stor-
age granules) or for cartilage (in which proteoglycans are deposited in the chondrocyte extracellular matrix) [15]. In nonepithelial cell cultures, the reported rates of synthesis, accumulation, and release for cellsurface macromolecules [16-19] are in the same range as those found for the cell-surface macromolecules in cultured tracheal epithelial cells. For example, in mouse 3T3 cell lines, most of the cell-surface proteoglycan was released from the cells with an approximate tl/2 of 6 h, and the majority of the macromolecules released were recovered in the dialyzed medium [17]. In cultured rat ovarian granulosa cells, the cell-surface dermatan sulfate proteoglycan accumulated on the cell surface with an approximate tl/2 of 7 h, and over 90% of the macromolecules was released into the medium with a tl/2 of 4-6 h [16,18]. The rate of 35SO4 incorporation into tracheal cell-surface macromolecules is consistent with biochemical [18] and autoradiographic [20] studies in other cell types which indicate that sulfation of high-molecular-weight glycoconjugates occurs in the Golgi region, immediately prior to secretion or appearance on the cell surface. For example, the cell-surface heparan sulfate and dermatan sulfate proteoglycans of cultured rat ovarian granulosa cells were transported to the cell surface with a i l l 2 of 13 min following introduction of 35804 label [18]. The time-course of synthesis and depletion of the core protein of cell-surface macromolecules in airway epithelial cells has not previously been studied. By comparison, rat chondrocyte cultures incorporated [3H]serine into proteoglycans within 3 - 4 h after pulsing, and the pool size of the core protein decreased with a t~/2 of 60 min of cycloheximide pretreatment [21,22], compared to apparent incorporation within 2 h and a depletion rate with a tl/2 of approx. 30 min for the tracheal epithelial cell-surface macromolecules. The apparent specific activities and sizes of the intracellular serine and sulfate pools were not calculated in this study; however, studies in other systems have indicated that the intracellular sulfate pool is extremely small or nonexistent [18]. The results of the present study, in which sulfate is incorporated into macromolecules within 5-10 min after labeling, suggest that the cultured dog tracheal epi-
345
thelial cells also lack a significant intracellular pool of free sulfate. Although the mechanism of continuous release of the macromolecules in the tracheal epithelial cells is unknown, our findings indicate that the process must be energy-dependent because the release of the macromolecules was virtually abolished at 4 ° C (Fig. 6). In some other cells, surface glycoproteins have been shown to be released by proteolytic cleavage, a process inhibited by serine-proteinase inhibitors [23,24]. However, none of the proteinase inhibitors that we tested, including serine-proteinase inhibitors, prevented the spontaneous release of the macromolecules in the tracheal epithelial cells. It is possible that proteinases were inaccessible to the exogenous inhibitors due to factors such as compartmentalization or conformational and steric restraints; alternatively, a still undiscovered membrane-bound proteinase could be responsible for release. The function of cell-surface macromolecules in the airway epithelial cell glycocalyx is still largely unknown. However, our finding of rapid and continuous turnover of sulfated cell-surface macromolecules in tracheal epithelial cells suggests that these macromolecules may play an important role in protecting the mucosal surface of the airways. To prevent microorganisms or toxins from causing infections or injuries in the mucosal surface, these materials must be removed before they reach the mucosal surface [25]. If microorganisms or toxins are only trapped in the mucus, the protection of mucosal surface is insufficient. The mucus must also be removed continuously and replaced by fresh mucus [26]. Thus, the rapid and continuous turnover of the macromolecules in tracheal epithelial cells might provide a highly effective protective mechanism in the airways. Furthermore, in some way, epithelial cellsurface macromolecules may be involved in bacterial and inflammatory cell adhesion to the airway epithelium. It is now known, for example, that certain bacteria produce lectin-like proteins which bind with high affinity to mammalian cellsurface carbohydrates [27]. In the respiratory tract, Streptococcus pneumoniae adherence to human pharyngeal epithelial cells was proposed to involve glycoconjugate receptors containing the lactosamine repeating unit [28]. We have recently identi-
fied O-linked oligosaccharides containing the lactosamine sequence on the cell-surface mucin glycoprotein of dog tracheal epithelial cells [7,8]. If these cell-surface macromolecules are involved in adhesion, cleavage of the macromolecules by bacterial enzymes or by enzymes released by inflammatory cells (e.g., neutrophil elastase, mast cell chymase, and tryptase), as we have demonstrated previously [7], might modify the adhesion process, and this modification could play an important role in prevention of infections and in the inflammatory process. Other studies have suggested that the sulfated glycosaminoglycans of the urothelial luminal surface act as an antiadhesion barrier, precluding bacterial and protein adhesion [29]. In summary, we have shown that cell-surface macromolecules in tracheal epithelial cells are rapidly and continuously produced, accumulate on the cell surface, and are continuously released from the cell surface. Our finding of rapid and continuous turnover of cell-surface macromolecules suggests a possible functional importance of these macromolecules in protecting the mucosal surface of the airways.
Acknowledgments The writers thank Dr. C.B. Basbaum for useful discussions, Ms. M. Zeiger for editorial assistance, and Ms. B. Cost and Ms. P. Snell for preparing the manuscript. This study was supported in part by N I H Program Project Grant HL-24136.
References 1 Spicer, S.S., Chakrin, L.W., Wardell, J.R., Jr. and Kendrick, W. (1971) Lab. Invest. 25, 483-490. 2 Gallagher, J.T., Hall, R.L., Jeffery, P.K., Phipps, R.J. and Richardson, P.S. (1978) J. Physiol. (Lond.) 275, 36P-37P. 3 Jeffery, P.K. (1978) in Respiratory Tract Mucus (Ciba Foundation Symposium 54 (New Series), ed.), pp. 5-20, Elsevier/Excerpta Medica/North-Holland, Amsterdam. 4 Spicer, S.S., Mochizuki, I., Setser, M.E. and Martinez, J.R. (1980) Am. J. Anat. 158, 93-109. 5 Basbaum, C.B. (1986) in Clinics in Chest Medicine (Widdicombe, J.H., ed.), Vol. 7, pp. 231-237, W.B. Saunders, Philadelphia. 6 Reid, L., Bhaskar, K. and Coles, S. (1982) in Mucus in Health and Disease (Chantler, E.N., Elder, J.B. and Elstein, M., eds.), Vol. II, pp. 369-391, Plenum Press, New York.
346 7 Varsano, S., Basbaum, C.B., Forsberg, L.S., Borson, D.B., Caughey, G. and Nadel, J,A. (1987) Exp. Lung Res. 13, 157-183. 8 Forsberg, L.S., Varsano, S., lwamoto, 1. and Nadel, J.A. (1987) J. Cell Biochem. (Suppl.) l l D , 142. 9 Widdicombe, J.H., Coleman, D.L., Finkbeiner, W.E. and Friend, D.S. (1987) Cell Tissue Res. 247, 95-103. 10 Oleinick, N.L. (1977) Arch. Biochem. Biophys. 182, 171-180. 11 Gashi, A.A., Benaron, L. and Basbaum, C.B. (1982) Fed. Proc. 41, 666 12 Neutra, M. and Leblond, C.P. (1966) J. Cell Biol .30, 137-150. 13 Bennett, G. and Leblond, C.P. (1970) J. Cell Biol .46, 409-418. 14 MacDermott, R.P., Donaldson, R.M., Jr. and Trier, J.S. (1974) J. Clin. Invest. 54, 545-554. 15 Gashi, A.A., Nadel, J.A. and Basbaum, C.B. (1987) Exp. Lung Res. 12, 83-96. 16 Yanagishita, M. and Hascall, V.C. (1979) J. Biol. Chem. 254, 12355-12364. 17 Roblin, R., Albert, S.O., Gelb, N.A. and Black, P.H. (1975) Biochemistry 14, 347-357.
18 Yanagishita, M. and Hascall, V.C. (1984) J. Biol. Chem. 259, 10270-10283. 19 Iozzo, R.V. (1984) J. Cell Biol. 99, 403-417. 20 Young, R.W. (1973) J. Cell Biol. 57, 175-189. 21 Kimura, J.H., Caputo, C.B. and Hascall, V.C. (19810 J. Biol. Chem. 256, 4368-4376. 22 Mitchell, D. and Hardingham, T. (1981) Biochem. J. 196, 521-529. 23 Little, P.S. and Huang, A.S. (1978) J. Virol. 27, 330-339. 24 Howard, S.C., Sherblom, A.P., Huggins, J.W., Carraway, C.C. and Carraway, K.L. (1981) Arch. Biochem. Biophys. 207, 40-50. 25 Gibbons, R.J. (1982) in Recent Advances in Mucosal Immunity (Strober, W., Hanson, L.A. and Sell, K.W., eds.), pp. 343-351, Raven Press, New York. 26 Clamp, J.R. (1984) Biochem. Soc. Trans. 12, 754-756. 27 Beachey, E.H. (1981) J. Infect. Dis. 143, 325-345. 28 Andersson, B., Dahmen, J., Frejd, T., Leffler, H., Magnusson, G., Noori, G. and Eden, C.S. (1983) J. Exp. Med. 158, 559-570. 29 Parsons, C.L., Stauffer, C.C. and Schmidt, J.D. (1980) Science 208, 605-607.
Biochimica et Biophysica Acta, 966 (1988) 336-346
Elsevier BBA 22955
Turnover of cell-surface macromolecules in cultured dog tracheal epithelial cells
Itsuo Iwamoto, Jay A. Nadel, Shabtai Varsano and Lennart S. Forsberg Cardiovascular Research Institute and Departments of Medicine and Physiology, University of California, San Francisco, San Francisco, CA (U.S.A.)
(Received 21 December 1987) (Revised manuscript received11 April 1988)
Key words: Cell surface macromolecule;Glycocalyx;Sulfated macromolecule;Epithelial cell; (Dog trachea) We studied the metabolism of sulfated cell-surface macromolecules in dog tracheal epithelial cells in primary culture. To examine the time-course and rate of appearance of sulfated macromolecules at the cell surface, the cells were pulsed with 35SO4 for short periods (5-15 min), and the incubation medium was sampled for spontaneously released macromolecules (basal secretions) and for release induced by trypsin (trypsin-accessible secretions). Trypsin-accessible ass-labeled macromolecules appeared on the cell surface within 5-10 min, increased linearly, and plateaued by 40 min; the median transit time for 35S-labeled macromolecules to reach the cell surface was 21 min. 35S-labeled macromolecules in basal secretions increased with a similar time-course, reaching a plateau by 40 rain. Incorporation of [3Hlserine into the protein moiety of trypsin-accessible macromolecules occurred more slowly; trypsin-accessible 3H-labeled macromolecules were barely detectable at 1 h and increased to a maximum after 2 h, suggesting the presence of a preformed pool of nonsulfated core protein. Pretreatment with cycloheximide, an inhibitor of protein synthesis, decreased trypsin-accessible 35S-labeled macromolecules log-linearly depending on the duration of pretreatment providing an estimate of the rate of depletion of the core protein pool ( i l l 2 = 32 min). During continuous exposure to 3SSO4, ass-labeled macromolecules accumulated on the cell surface (trypsin-accessible compartment) for 16 h, at which point the cell-surface pool was saturated ( t l / 2 = 7.5 h). After pulse-labeling the cells with 35SO4 for 15 min, the 35S-labeled macromolecules disappeared continuously from the cell surface ( t l / 2 = 4.6 h), and 79% of the radioactivity was recovered in the medium as nondialyzable macromolecules. Release of the 3sS-labeled macromolecules from the cell surface was abolished at 4 ° C , indicative of an energy-dependent process, but multiple proteinase inhibitors did not affect the release. We conclude that sulfate is metabolized rapidly into epithelial cell-surface macromolecules, which accumulate continuously into a relatively large cell-surface pool, before they are released by an undefined energy-dependent mechanism.
Abbreviation: DME, Dulbecco's modified Eagle's medium. Correspondence: J.A. Nadel, Cardiovascular Research Institute, Box 0130, Universityof California, San Francisco, CA 94143-0130, U.S.A.
Introduction
Cell-surface macromolecules, which consist mainly of proteoglycans and mucin-type glycopro-
0304-4165/88/$03.50 © 1988 ElsevierSciencePublishers B.V. (BiomedicalDivision)
337 teins, are present at the apical surface of the airway epithelium [1-5] and are believed to have many biological functions, including maintenance of mucociliary flow and protection from infection [6]. In the respiratory tract, the cell-surface macromolecules form a discrete region, or glycocalyx, which is readily distinguished from the mucus layer by histocytochemical and autoradiographic procedures [2-6]. Many studies have focused on the biochemistry of the mucus layer, but many aspects of the underlying epithelial glycocalyx remain unclear. We have recently shown that primary cultures of dog tracheal epithelial cells, which lack prominent secretory storage granules, incorporate [35S]sulfate into an extensive cellsurface glycocalyx [7,8]. The major component is a sulfated mucin glycoprotein (molecular weight >/ 10 6) which has several unique structural features, including a high proportion of O-glycosidically linked glycosaminoglycans of the poly(N-acetyllactosamine) type [7,8]. We found that this sulfated mucin accounted for over 70% of the cell-surface macromolecular sulfate in the cultured cells. The sulfated macromolecules appear to be available at the cell surface in a trypsin-accessible form, because treating the cells with trypsin causes an immediate 5-fold increase in the recovery of these macromolecules in the medium. However, the rates of synthesis, degradation, and release of this and other cell-surface macromolecules from airway epithelial cells are unknown. In the present study, we designed pulse-chase experiments to examine the metabolism of the sulfated cell-surface macromolecules in primary cultures of dog tracheal epithelial cells. Using [35S]sulfate as precursor, we determined the rates of appearance, accumulation, and release of sulfated macromolecules at the cell surface. In addition, dual-labeling studies with [3H]serine and [35S]sulfate were used to examine in detail the time-course of synthesis of the protein moiety of the macromolecules. Measurement of [35S]sulfate incorporation in the presence of cycloheximide, an inhibitor of protein synthesis, provided an estimate of the rate of depletion of the intracellular protein pool for the macromolecules. We also examined possible mechanisms of release of the sulfated macromolecules from the cell surface.
Methods
Materials Dulbecco's modified Eagle's medium (DME), Ham's F-12 medium (F-12), fetal calf serum, nonessential amino acids (alanine, asparagine, aspartic acid, glutamic acid, glycine, proline and serine), and antibiotics (penicillin, streptomycin, gentamicin and fungizone) were obtained from the Cell Culture Facility, University of California, San Francisco. Crude collagenase from Clostridium h&tolyticum (type I), human placental collagen, bovine pancreatic trypsin, cycloheximide, soybean trypsin inhibitor, aprotinin, leupeptin, pepstatin, phosphoramidon, captopril and bestatin were purchased from Sigma Chemical Co. (St. Louis, MO). Polycarbonate filters (0.8 /xm pore size, 25 mm diameter) were purchased from Nucleopore Corp. (Pleasanton, CA). Dialysis membranes (molecular weight cutoff, 12000-14000) were purchased from Spectrum Medical Industries (Los Angeles, CA). Na235SO4 (spec. act., 43 Ci/mg sulfate) and L-[3H]serine (spec. act. 8 Ci/mmol) were purchased from ICN Radiochemicals (Irvine, CA). Isolation and culture of tracheal epithelial cells and conditions of incubation Dog tracheal epithelial cells were isolated and cultured as previously described [9]. Briefly, the trachea was removed from mongrel dogs (18-30 kg) after intravenous sodium pentobarbital anesthesia (30 mg/kg). The epithelium was stripped from the submucosa with forceps, minced with scissors, and digested with 0.02% crude collagenase at room temperature for 3 h. The epithelial cells in the collagenase solution were collected by centrifugation and suspended in medium consisting of 50% DME and 50% F12 (DME/F12) that contained 5% fetal calf serum, 1% nonessential amino acids, and antibiotics (105 U/I penicillin, 100 mg/1 streptomycin, 100 mg/1 gentamicin, and 500 ~g/1 fungizone). We refer to this medium as complete medium. The viability of the cells recovered was 88.5 + 2.2% (mean + S.D.) (n = 24 tracheas) as determined by erythrosin B dye exclusion.
338
The isolated cells were plated on collagencoated polycarbonate filters (diameter, 25 mm) at a density of 4 . 1 0 5 viable cells/cm 2 in 2.5 ml of complete medium in 35-mm plastic dishes. Dishes were placed at 3 7 ° C in 6% CO 2 in air in a CO 2 incubator. The complete medium in the dish was changed after the first 24 h and then every 2 - 3 days. Cell growth and monolayer formation were assessed by observing the cells on the transparent area of the dish surrounding the filter. The cells became confluent 4 - 5 days after the start of culture, and 7- to 9-day-old cells were used for experiments. The cells had typical characteristics of epithelial cells and contained no secretory granules in their cytoplasm [9]. Unless otherwise stated, the experiments were performed at 37 ° C, and the cells were incubated in D M E / F 1 2 medium throughout the experiment. Each experiment was performed in duplicate; reported values are the average of the two measurements. We refer to the time at the start of radiolabeling as time 0 rain.
viously shown that the conditions used for trypsin incubation do not cause cell leakage or detachment of cells from the filter disks [7]. To determine the time-course of radiolabel incorporation and macromolecule accumulation on the cell surface, it was necessary to isolate an entire cell-surface pool of completed macromolecules instantly, at any series of time points following introduction of radiolabel, without waiting for the metabolic release of these components from the cell surface (which occurs slowly, tl/2 = 4.6 h; see Results). Thus, 'trypsin-accessible' secretions were analyzed, because treating the cells with trypsin proved to be an effective procedure in which an entire pool of cell-surface macromolecules can be harvested at any one instant and their amounts quantitated. This allows measurement of the rates of incorporation and accumulation and, at the same time, eliminates interference due to the ongoing basal rate of degradation (which includes release from the cell surface and internalization).
'Basal" and "trypsin-accessible' secretions
Time-course of 35S04 incorporation into cell-surface macromolecules
Previous studies have shown that sulfated macromolecules are released spontaneously from cultured airway epithelial cells, giving rise to what we refer to as 'basal' secretions. The major sulfated macromolecule of these basal secretions is a mucin glycoprotein (molecular weight > / 1 0 6) [7,8]. Prior to release, the sulfated macromolecule resides on the cell surface in a trypsin-accessible domain, or compartment; treatment of the cells with trypsin causes an immediate release of this macromolecule into the medium, yielding the 'trypsin-accessible' secretions. To study the turnover rates of epithelial cell-surface macromolecules, both 'basal' and 'trypsin-accessible' secretions were analyzed. We obtained basal secretions by incubating epithelial cells with radiolabeled precursors (35SO4 or [3H]serine), and harvesting radiolabeled macromolecules in the culture medium at various subsequent times. Trypsin-accessible secretions were obtained by incubating the cells with radiolabeled precursors as in the basal studies, but we added 125 U / m l of trypsin at 3 7 ° C for 5 - 1 0 min at the end of each chasing time. Radiolabeled macromolecules present in the culture medium following this treatment were then harvested. We have pre-
To assess the time-course and rate of appearance of sulfated cell-surface macromolecules, cells were pulsed briefly with 35SO4, and the trypsin-accessible 35S-labeled macromolecules released into the medium were collected at various chase times. Specifically, cells grown on polycarbonate filters were exposed to 1 5 0 / ~ C i / m l of 35SO4 in 2 ml of D M E / F 1 2 in a 35-mm dish at 37 ° C for 5 min and then washed four times in D M E / F 1 2 medium. Basal and trypsin-accessible 35S-labeled macromolecules were then collected by incubating the cells for 5 min with trypsin (trypsin-accessible) or without trypsin (basal) after chase times of 5, 10, 15, 20, 30, 40 and 60 rain.
Time-course of [3H]serine incorporation into trypsin-accessible macromolecules To determine the time-course of core protein synthesis for the cell-surface macromolecules, the rate of incorporation of [3H]serine into trypsinaccessible macromolecules was determined. Cells were washed once in serine-free D M E before incubation, exposed simultaneously to 50 ~tCi/ml of 35SO4 and 40 /~Ci/ml of [3H]serine in serine-free
339 D M E for 15 min and then washed in D M E / F 1 2 medium. Trypsin-accessible secretions were collected after chase times of 0.5, 1, 2, 3 and 5 h. The trypsin-accessible secretions from each time point were concentrated by lyophilization and analyzed on a column of Sepharose CL-4B as described previously [7,8]. The amounts of 3H-labeled macromolecules and 35S-labeled macromolecules were determined by monitoring the activity of each isotope recovered in the void volume fraction of the column. We have previously shown that the major macromolecular species in this void volume fraction is a sulfated mucin glycoprotein ( M r >~ 106), which contains alkaline-borohydride-labile O-glycosidically linked glycans of the poly(Nacetyllactosamine) type [7,8]. In these experiments, basal secretions were not evaluated.
Time-course of depletion of core protein pool To determine the pool size of the core protein of the sulfate cell-surface macromolecules, we took advantage of the fact that cycloheximide inhibits protein synthesis [10]. Because the core protein pool is depleted continuously by synthesis and secretion of the sulfated macromolecules, the longer periods of incubation with cycloheximide resulted in progressive depletion of this pool. Therefore, to determine the time-course of depletion of the core protein, we pretreated the cells with 0.5 mM cycloheximide at 37 ° C for 0, 0.5, 1, 1.5, 2, 3, 4 or 6 h. Then we incubated the cells with 150 /~Ci/ml 35SO4 for 15 rain and washed the cells. Basal and trypsin-accessible 35S-labeled macromolecules were harvested after a 60-min chase (the time at which all of the 35S-labeled macromolecules were shown to reach the cell surface after a 15-min pulse; see Fig. 2).
Time-course of accumulation of sulfated macromolecules on the cell surface To assess the time-course of accumulation and total pool size of sulfated macromolecules on the cell surface (trypsin-accessible compartment), we determined the amounts of 35S-labeled macromolecules after prolonged labeling with 35SO4. The cells were incubated with 50 # C i / m l of 35SO4 in complete medium in a CO 2 incubator (6% CO 2 in air) for various periods (4, 8, 12, 16, 20 or 24 h). The cells were washed four times and chased for 1
h. The cells were then incubated with trypsin for 10 min, and trypsin-accessible 35S-labeled macromolecules were collected in the culture medium. In parallel experiments, trypsin was omitted and basal (spontaneously released) secretions were measured in the medium.
Time-course of disappearance of sulfated cell-surface macromolecules We studied simultaneously the time-course of disappearance of sulfated macromolecules from the cell surface and the time-course of recovery of sulfated macromolecules in the incubation medium. The cells were incubated with 150 /~Ci/ml of 35SO4 for 15 min and washed. Then, at 60 rain after the beginning of incubation (by which time we assumed that 35SO4 label was fully incorporated into cell-surface macromolecules), we incubated the cells in 2.5 ml of D M E / F 1 2 at 37 ° C in a CO 2 incubator for various times (0, 2.5, 5, 10, 15, 20 or 25 h). At the end of each chase time, we collected the medium to measure total 3SS as well as 35S-labeled macromolecules. Next we measured either basal secretion from the cells (one series of experiments) or trypsin-accessible secretions after adding trypsin for 10 min (another series of experiments). The amount of total 35S recovered in the medium (in all molecular species) at the end of the 25 h chase period was compared to the total amount of 35S present in the cells immediately prior to the first chase period (at time 0 h).
Effect of temperature and of proteinase inhibitors on release of sulfated cell-surface macromolecules To determine whether active cell metabolism is required for the basal release of the sulfated cellsurface macromolecules, we examined the effect of temperature. Cells preincubated with 35SO4 (150 /~Ci/ml for 15 min) were incubated in D M E / F 1 2 either at 3 7 ° C or at 4 ° C from time 60 min for 6 h. At the end of incubation, medium was collected for measurement of released 35S-labeled macromolecules. Then the cells incubated at 4 ° C were rewarmed to 37 ° C. Trypsin was then added to all cells for 10 min, and the trypsin-accessible 35Slabeled macromolecules were measured. Because exogenous proteinases cleave the cellsurface macromolecules [7], we also examined the
340 effect of proteinase inhibitors on the disappearance of the sulfated macromolecules from the cell surface. We reasoned that if airway epithelial cell-surface macromolecules are normally released by proteolytic cleavage, then the appropriate proteinase inhibitor should inhibit basal secretion and cause accumulation of the macromolecules, thus increasing subsequent release by trypsin. Cells preincubated with 35SO4 (150 /~Ci for 15 min) were incubated at 37 °C from time 60 min for 6 h. The seven proteinase inhibitors added were soybean trypsin inhibitor (100 ~ g / m l ) and aprotinin (30/~g/ml) to inhibit serine proteinases, leupeptin (10 ~ g / m l ) to inhibit thiol proteinases, pepstatin (10 /~g/ml) to inhibit carboxyl proteinases, phosphoramidon (10 -5 M) to inhibit neutral endopeptidase, captopril (10 -5 M) to inhibit kininase II, and bestatin (10 -5 M) to inhibit aminopeptidases. After the cells were washed four times, trypsin-accessible 35S-labeled macromolecules were measured. As controls, the same study was performed in the absence of inhibitors.
Measurements of ~sS-labeled maeromolecules and of total 35S The amounts of 35S-labeled macromolecules were measured by counting the activity of nondialyzable 35S in the samples or by measuring radioactivity eluting in the void volume fractions of Sepharose CL-4B chromatography (see below). For dialysis, the sample was dialyzed against 5 1 of distilled water (the water was changed every 6 h) for 3 days at 4 ° C using dialysis membranes (molecular weight cutoff, 12 000-14000). Then the activity of 35S in the dialyzed sample was counted by a beta scintillation counter. The amounts of total 35S in the incubation medium were measured by counting the radioactivity of 3SS in the incubation medium without dialyzing. The amounts of total 35S in the pulselabeled cells were measured by counting the radioactivity of 35S of the cells after disrupting the cells by adding distilled water. The amounts of 35S-labeled and 3H-labeled macromolecules after 35SO4 and [3H]serine duallabeling were determined by counting the activity of 35S and 3H recovered in the void volume fraction of Sepharose CL-4B gel filtration chromatog-
raphy, performed as described previously [7]. The incubation medium collected at each chasing time was concentrated by lyophilization and applied to a column (1 × 20 cm) of Sepharose CL-4B. Chromatography was performed under associative conditions (0.1 M NaOAc buffer (pH 5.8)/0.02% sodium azide), and aliquots of fractions were assayed for radioactivity in the scintillation counter with windows optimized for dual-label counting. The amount of 35S-labeled or 3H-labeled macromolecules was calculated from the sum of radioactivities of 35S and 3H recovered in the Sepharose CL-4B void volume peak, representing macromolecules having an apparent molecular weight >/ 1 • 1 0 6.
The amounts of 35S-labeled and 3H-labeled macromolecules and of total 35S are expressed as the activity of 35S or 3H in dpm produced by epithelial cells grown on one polycarbonate filter. Counting efficiencies for 3H in dilute buffers and culture mediums were approx. 31% as determined with quenched 3H standards. 35S counting efficiency was over 95% by assay of newly purchased radionuclide.
Analysis of data Data are summarized as mean_+ S.D. The half-time of turnover of trypsin-accessible 35Slabeled macromolecules in the epithelial cells was determined as the time when the amount of the macromolecules became one half of the maximum amount in the experiment. Data were analyzed by Student's paired t-test. Results
Time-course of 35S04 incorporation into cell-surface macromolecules 35S-labeled macromolecules were detectable in the cell-surface trypsin-accessible pool within 5-10 min following the introduction of radiolabel (Fig. 1). Trypsin-accessible 35S-labeled macromolecules increased linearly with a plateau at 40 min; the half-time of appearance of 35S-labeled macromolecules at the cell surface was 21 _+ 1 min. 35S-labeled macromolecules present in basal secretions also increased with a similar time-course, reaching a plateau after 40 rain.
341 1500-
a n a p p a r e n t m o l e c u l a r w e i g h t > / 1 0 6 as a s s e s s e d by Sepharose CL-4B chromatography. However, trypsin-accessible 3H-labeled macromolecules, of similar apparent molecular weight, appeared more slowly: t h e y w e r e d e t e c t a b l e a f t e r a 1 h chase, r e a c h i n g a m a x i m u m at 2 h (Fig. 2).
1250-
~,
•
750E
0")
250-
o
o
10
20
30
Duration
of Chase
40
50
60
(min)
Fig. 1. Time-course of appearance of sulfated macromolecules on the cell surface in cultured dog tracheal epithelial cells. Cultured cells were pulsed with 35SO4 for 5 min from time 0 min, and 35S-labeled macromolecules were sampled for 5 min in the culture medium after chasing for 5, 10, 15, 20, 30, 40 and 60 min. In one series of experiments, the samples were obtained without adding trypsin (basal secretion; ©). In a second series of experiments, trypsin was added for 5 min at the various times (trypsin-accessible secretions; e). Data are means _+S.D. for three experiments at each time point.
Time-course of [3H]serine incorporation into trypsin-accessible macromolecules After a 15-min trypsin-accessible peared rapidly on trypsin-accessible
p u l s e w i t h [ 3 H ] s e r i n e a n d 35SO4, 35S-labeled m a c r o m o l e c u l e s a p t h e cell s u r f a c e as e x p e c t e d . T h e 35S-labeled m a c r o m o l e c u l e s h a d
1500~ ~.~ 1250 1000 oJ 750 oE o 250
Time-course of core protein depletion in sulfated cell-surface macromolecules In the presence of cycloheximide, trypsin-accessible 35S-labeled m a c r o m o l e c u l e s d e c r e a s e d loglinearly, depending on the duration of cyclohexim i d e p r e t r e a t m e n t (Fig. 3). T h e h a l f - t i m e o f d e c r e m e n t o f t r y p s i n - a c c e s s i b l e 35S-labeled m a c r o m o l e c u l a r s e c r e t i o n (35SO4 i n c o r p o r a t i o n ) w a s 32 +4 min.
Time-course of accumulation of sulfated macromolecules on the cell surface W h e n t h e cells w e r e i n c u b a t e d w i t h 35SO4 for p r o l o n g e d p e r i o d s ( 4 - 2 4 h), 35S-labeled m a c r o m o l e c u l e s ( b o t h b a s a l a n d t r y p s i n - a c c e s s i b l e ) inc r e a s e d l i n e a r l y for 1 2 - 1 6 h a n d t h e n r e a c h e d a
•~. 5000"
~ i
1000 500
o =
o
E
s00
100.
0
i
0
1
2
3
4
.
i
.
i
•
i
5
Duration of Chase (hi
Fig. 2. Comparison of time-course of appearance of trypsinaccessible 35S-labeled and 3H-labeled macromolecules in cultured dog tracheal epithelial cells. Cultured cells were pulsed simultaneously with 3SSO4 and [3H]serine for 15 min from time 0 rain to assess the relative rates of appearance of sulfated macromolecules and of macromolecules with newly synthesized protein core. Then, after a chase period of 0.5, 1, 2, 3 and 5 h, trypsin was added for 10 rain, and the culture medium was analyzed for macromolecules containing 35S (O) and for those containing 3H (©) by Sepharose CL-4B chromatography. Data are means + S.D. for three experiments at each time point.
Duration of Cycloheximifle
Pretreatment (h)
Fig. 3. Effect of pretreatment with the protein synthesis inhibitor cycloheximide on secretion of sulfated macromolecules in cultured dog tracheal epithelial ceils. Cells were preincubated with cycloheximide for various periods of 0.5-6 h. Then the cells were pulsed with 3SSO4 for 15 min, and 35S-labeled macromolecules were measured in the culture medium at time 60 min. In one series of experiments, the samples were obtained without adding trypsin (basal secretion; ©). In a second series of experiments, trypsin was added for 10 rain (trypsinaccessible secretion, O). Data are means + S.D. for three experiments at each time point.
342
o
x
E=
Basal secretions, like trypsin-accessible secretions, decreased gradually over several hours (Fig. 5A, open squares).
70 60 50
-5
40
Effect of temperature and of proteinase inhibitors on sulfated cell-surface macromolecules
30 E o
20
E
10
Cells which were incubated at 4 ° C for 6 h showed essentially no release of sulfated macro-
0 0
4
8
12
16
20
24
Duration of Incubation of 35 SO 4 (h)
Fig. 4. Time-course of accumulation of sulfated macromolecules on the cell surface (trypsin-accessible compartment) in cultured dog tracheal epithelial cells. Cells were incubated with 3SSO4 for 4 - 2 4 h, and then 3sS-labeled macromolecules were measured. In one series of experiments, no trypsin was added (basal secretion; ©). In a second series of experiments, trypsin was added for 10 rain after the incubation with 35SO4 (trypsin-accessible secretion; O). Data are means_+S.D, for three experiments at each time point.
A. Disappearance from Cell Surface
4000-
3000 1
2000 o
1000 (13
i
0 0
steady-state level (Fig. 4). The half-time for accumulation of trypsin-accessible 35S-labeled macromolecules was 7.5 +_ 0.6 h of exposure to 35SO4.
Time-course of disappearance of sulfated macromolecules from the cell surface Trypsin-accessible 35S-labeled macromolecules decreased over 20 h with a half-time of 4.6 _+ 0.4 h (Fig. 5A, solid squares). The data were found to approximate a first order exponential decay, with a correlation coefficient ( r ) = 0.97 (determined by a least-squares fit to the transformed exponential curve). As the trypsin-accessible 35S-labeled macromolecules disappeared from the cell surface, nondialyzable 35S-labeled macromolecules recovered in the incubation medium increased proportionately (Fig. 5B, solid circles). More than 90% of the total 35S that had been present in the 35SO4-pulsed cells before the incubation was recovered in the medium after 25 h of incubation of the cells. In addition, 79 + 4% of the total 35S recovered in the medium after each incubation period was nondialyzable 35S-labeled macromolecules. These findings suggest that most of the trypsin-accessible 3sS-labeled macromolecules were released from the cell surface as whole macromolecules.
5
10
15
20
q 25
B. Recovery from Incubation Medium 6000-
4000-
2000-
0
5
1'0
1'5
2'0
2'5
Duration of Incubation (h)
Fig. 5. Time-course of disappearance of sulfated macromolecules from the cell surface in cultured dog tracheal epithelial cells. We studied simultaneously the time-course of disappea~a~,ce of~rypsin-accessible 3sS-labeled macromolecules (A, B) and of reco~;ery of 35S in the incubation medium (B). Cells were pulsed with 35SO4 for 15 min and rested until 60 min had elapsed, at which time m a x i m u m 35S was shown to be incorporated into cell-surface macromolecules. This is labeled as time zero. At various times (2.5-25 h), trypsin was added for 10 min and samples of culture m e d i u m were removed for measurement of 35S-labeled macromolecules. These trypsin-accessible macromolecules decreased with incubation time (A, B). This was cofffirt~ed by simultaneous measurement of the 35S that had accumulated in the culture medium during the time of incubation (B). Data for 35S-labeled macromolecules (o) and total 3SS (©) are shown separately. Data for basal secretion of 3sS-labeled macromolecules (sampled for 10 min after various times of incubation) are also shown (A, D). Data are means _+ S.D. for four experiments at each time point.
343 A. Trypsin-accessible Secretion 100 " 80" _e
60
0
E 0
40'
E 20'
cubated for 6 h at 4 ° C and rewarmed produced twice as many 35S-labeled macromolecules after addition of trypsin as the cells maintained at 3 7 ° C ( P < 0.025) (Fig. 6A). The addition of proteinase inhibitors to the incubation medium (see Methods for description) had no effect on the disappearance of trypsinaccessible 35S-labeled macromolecules from the cell surface ( P > 0.5).
0
37 ° C
4 ° C, followed by 3 7 ° C
B. Recovery from Incubation Medium A
100" 80"
a ~' 15 o
60
/ / / /
/ / / /
/ / / /
/ / / /
/ / / /
/ / z /
/ / / /
/////// / / / / / / / / / / / / / / /11/11/ I I / 1 1 / i II/1111 / / / / / / /
///////
o
40'
15
/ / I / I / / / / / / / / / / / i / / / ~
20 0
~.ztzz,.z
37°C
4° C
Fig. 6. Effect of temperature on release of sulfated macromolecules in cultured dog tracheal epithelial cells. Cells were pulsed with 35SO4 for 15 rain, rested for an additional 45 min, and then maintained for an additional 6 h at either 3 7 ° C (left columns) or at 4 ° C (right columns). At the end of the 6 h incubation, the cooled cells were rewarmed to 3 7 ° C , trypsin was added to all cells for 10 min, and the trypsin-accessible 35S-labeled macromolecules were measured (A). In cells that were incubated at 4 ° C, practically no basal secretion of 35Slabeled macromolecules occurred compared to the cells incubated at 3 7 ° C (B). The cells incubated for 6 h at 4 ° C and rewarmed produced twice as m a n y 35S-labeled macromolecules after the addition of trypsin as the cells maintained at 3 7 ° C (A). Data are means__+ S.D. in four experiments. * Significantly different from the mean obtained after incubation of the cells at 37 o C, P < 0.025.
molecules from the cell surface into the medium (basal secretion; Fig. 6B, right). The amount of 35S-labeled macromolecules recovered in the incubation medium after a 6 h incubation at 4 ° C was only 9 _+ 1% of that recovered in the medium after a 37°C incubation ( P < 0.025) (Fig. 6B, left). Following the 6 h incubation at 4 ° C, the cells were rewarmed and trypsin was added. The trypsin-accessible 35S-labeled macromolecules obtained were 90 _+ 6% of those obtained before incubation in the cold (Fig. 6A, right). Cells in-
Discussion Previously, we showed that dog tracheal epithelial cells in primary culture synthesize a sulfated mucin glycoprotein ( M r > / 1 0 6 ) , which accounts for over 70% of macromolecular sulfate when 35SO4 is used as the labeling precursor [7,8]. Ultrastructural and biochemical studies suggested that the sulfated mucin is located at the cell surface in a trypsin-accessible form [7,8]. In the present study, using the trypsin-accessible 35S-labeled macromolecules as an indicator of cell-surface mucin, we show that the sulfated macromolecules are rapidly produced, accumulate continuously into a large cell-surface pool, then undergo continuous release into the medium by an energy dependent mechanism. Incorporation of sulfate and transport to the cell surface of the sulfated macromolecules began within 5-10 min after the introduction of 35SO4 label (Fig. 1). The median transit time for newly sulfated macromolecules to reach the cell surface was estimated at 21 _+ 1 rain. These results show that sulfation and transport to the cell surface occur rapidly in tracheal epithelial cells and suggest that there is a preformed pool of nonsulfated macromolecules. The protein moiety of macromolecules eluting at the void volume of Sepharose CL-4B chromatography was synthesized and detectable at the cell surface within 2 h after uptake of an amino acid, as indicated by the time-course of [3H]serine incorporation into trypsin-accessible macromolecules (Fig. 2). The difference in the time-course of [3H]serine and 35SO4 incorporation further suggests the existence of an intracellular pool of nonsulfated precursors at various stages of completion, and could also reflect the transit time of this pool to the site of sulfation (presumably the Golgi region). A slower equilibration time for
344
the intracellular serine pool with [3H]serine could also contribute to this difference. The incorporation of both radiolabels into a single molecular species was not verified; however, both 3H and 3~S activities co-eluted at the void volume during Sepharose CL-4B chromatography. It is noteworthy that the major component of the Sepharose void volume fraction is a sulfated mucin, containing a high proportion of sulfated glycans which can be released from protein by alkalineborohydride treatment [8]. Following label incorporation, completed macromolecules accumulate on the cell surface into a trypsin-accessible pool, where they are probably intercalated into the membrane. The cell-surface pool was saturated after 16 h of continuous production of macromolecules during prolonged exposure to 35SO4 (Fig. 4). This suggests a relatively large pool of the sulfated macromolecules, consistent with our previous ultrastructural observations which showed a dense cell-surface coat containing anionic glycoconjugate [7]. Continuous release of the sulfated macromolecules from the cell surface is shown by the disappearance of trypsin-accessible 35S-labeled macromolecules from the surface over 20 h (tl/2 = 4.6_+ 0.4 h) and by the recovery of the macromolecules in the medium (Fig. 5). Thus, turnover of sulfated cell-surface macromolecules in tracheal epithelial cells is rapid and continuous. Our finding of rapid and continuous turnover, that is, time-courses and rates of production, accumulation on the cell surface, and release of cellsurface macromolecules in cultured tracheal epithelial cells, is in general agreement with previous autoradiographic studies of airway epithelia in situ. Thus in cat, ferret and rabbit tracheal organ culture, autoradiographic studies have shown that 35SO4-1abeled materials are transported from supranuclear regions to the epithelial surface within 1 h [11]. In intestinal columnar epithelial cells in rat and rabbit, cell-surface macromolecules were transported from the Golgi region to the apical surface within 30 min after pulsing with various radiolabeled sugars [12-14]. In ferret tracheal organ cultures, turnover of sulfated macromolecules in the epithelium was considerably more rapid (t~/2 = 2 h) than the turnover rates observed for submucosal glands (containing a high percentage of cells with prominent secretory stor-
age granules) or for cartilage (in which proteoglycans are deposited in the chondrocyte extracellular matrix) [15]. In nonepithelial cell cultures, the reported rates of synthesis, accumulation, and release for cellsurface macromolecules [16-19] are in the same range as those found for the cell-surface macromolecules in cultured tracheal epithelial cells. For example, in mouse 3T3 cell lines, most of the cell-surface proteoglycan was released from the cells with an approximate tl/2 of 6 h, and the majority of the macromolecules released were recovered in the dialyzed medium [17]. In cultured rat ovarian granulosa cells, the cell-surface dermatan sulfate proteoglycan accumulated on the cell surface with an approximate tl/2 of 7 h, and over 90% of the macromolecules was released into the medium with a tl/2 of 4-6 h [16,18]. The rate of 35SO4 incorporation into tracheal cell-surface macromolecules is consistent with biochemical [18] and autoradiographic [20] studies in other cell types which indicate that sulfation of high-molecular-weight glycoconjugates occurs in the Golgi region, immediately prior to secretion or appearance on the cell surface. For example, the cell-surface heparan sulfate and dermatan sulfate proteoglycans of cultured rat ovarian granulosa cells were transported to the cell surface with a i l l 2 of 13 min following introduction of 35804 label [18]. The time-course of synthesis and depletion of the core protein of cell-surface macromolecules in airway epithelial cells has not previously been studied. By comparison, rat chondrocyte cultures incorporated [3H]serine into proteoglycans within 3 - 4 h after pulsing, and the pool size of the core protein decreased with a t~/2 of 60 min of cycloheximide pretreatment [21,22], compared to apparent incorporation within 2 h and a depletion rate with a tl/2 of approx. 30 min for the tracheal epithelial cell-surface macromolecules. The apparent specific activities and sizes of the intracellular serine and sulfate pools were not calculated in this study; however, studies in other systems have indicated that the intracellular sulfate pool is extremely small or nonexistent [18]. The results of the present study, in which sulfate is incorporated into macromolecules within 5-10 min after labeling, suggest that the cultured dog tracheal epi-
345
thelial cells also lack a significant intracellular pool of free sulfate. Although the mechanism of continuous release of the macromolecules in the tracheal epithelial cells is unknown, our findings indicate that the process must be energy-dependent because the release of the macromolecules was virtually abolished at 4 ° C (Fig. 6). In some other cells, surface glycoproteins have been shown to be released by proteolytic cleavage, a process inhibited by serine-proteinase inhibitors [23,24]. However, none of the proteinase inhibitors that we tested, including serine-proteinase inhibitors, prevented the spontaneous release of the macromolecules in the tracheal epithelial cells. It is possible that proteinases were inaccessible to the exogenous inhibitors due to factors such as compartmentalization or conformational and steric restraints; alternatively, a still undiscovered membrane-bound proteinase could be responsible for release. The function of cell-surface macromolecules in the airway epithelial cell glycocalyx is still largely unknown. However, our finding of rapid and continuous turnover of sulfated cell-surface macromolecules in tracheal epithelial cells suggests that these macromolecules may play an important role in protecting the mucosal surface of the airways. To prevent microorganisms or toxins from causing infections or injuries in the mucosal surface, these materials must be removed before they reach the mucosal surface [25]. If microorganisms or toxins are only trapped in the mucus, the protection of mucosal surface is insufficient. The mucus must also be removed continuously and replaced by fresh mucus [26]. Thus, the rapid and continuous turnover of the macromolecules in tracheal epithelial cells might provide a highly effective protective mechanism in the airways. Furthermore, in some way, epithelial cellsurface macromolecules may be involved in bacterial and inflammatory cell adhesion to the airway epithelium. It is now known, for example, that certain bacteria produce lectin-like proteins which bind with high affinity to mammalian cellsurface carbohydrates [27]. In the respiratory tract, Streptococcus pneumoniae adherence to human pharyngeal epithelial cells was proposed to involve glycoconjugate receptors containing the lactosamine repeating unit [28]. We have recently identi-
fied O-linked oligosaccharides containing the lactosamine sequence on the cell-surface mucin glycoprotein of dog tracheal epithelial cells [7,8]. If these cell-surface macromolecules are involved in adhesion, cleavage of the macromolecules by bacterial enzymes or by enzymes released by inflammatory cells (e.g., neutrophil elastase, mast cell chymase, and tryptase), as we have demonstrated previously [7], might modify the adhesion process, and this modification could play an important role in prevention of infections and in the inflammatory process. Other studies have suggested that the sulfated glycosaminoglycans of the urothelial luminal surface act as an antiadhesion barrier, precluding bacterial and protein adhesion [29]. In summary, we have shown that cell-surface macromolecules in tracheal epithelial cells are rapidly and continuously produced, accumulate on the cell surface, and are continuously released from the cell surface. Our finding of rapid and continuous turnover of cell-surface macromolecules suggests a possible functional importance of these macromolecules in protecting the mucosal surface of the airways.
Acknowledgments The writers thank Dr. C.B. Basbaum for useful discussions, Ms. M. Zeiger for editorial assistance, and Ms. B. Cost and Ms. P. Snell for preparing the manuscript. This study was supported in part by N I H Program Project Grant HL-24136.
References 1 Spicer, S.S., Chakrin, L.W., Wardell, J.R., Jr. and Kendrick, W. (1971) Lab. Invest. 25, 483-490. 2 Gallagher, J.T., Hall, R.L., Jeffery, P.K., Phipps, R.J. and Richardson, P.S. (1978) J. Physiol. (Lond.) 275, 36P-37P. 3 Jeffery, P.K. (1978) in Respiratory Tract Mucus (Ciba Foundation Symposium 54 (New Series), ed.), pp. 5-20, Elsevier/Excerpta Medica/North-Holland, Amsterdam. 4 Spicer, S.S., Mochizuki, I., Setser, M.E. and Martinez, J.R. (1980) Am. J. Anat. 158, 93-109. 5 Basbaum, C.B. (1986) in Clinics in Chest Medicine (Widdicombe, J.H., ed.), Vol. 7, pp. 231-237, W.B. Saunders, Philadelphia. 6 Reid, L., Bhaskar, K. and Coles, S. (1982) in Mucus in Health and Disease (Chantler, E.N., Elder, J.B. and Elstein, M., eds.), Vol. II, pp. 369-391, Plenum Press, New York.
346 7 Varsano, S., Basbaum, C.B., Forsberg, L.S., Borson, D.B., Caughey, G. and Nadel, J,A. (1987) Exp. Lung Res. 13, 157-183. 8 Forsberg, L.S., Varsano, S., lwamoto, 1. and Nadel, J.A. (1987) J. Cell Biochem. (Suppl.) l l D , 142. 9 Widdicombe, J.H., Coleman, D.L., Finkbeiner, W.E. and Friend, D.S. (1987) Cell Tissue Res. 247, 95-103. 10 Oleinick, N.L. (1977) Arch. Biochem. Biophys. 182, 171-180. 11 Gashi, A.A., Benaron, L. and Basbaum, C.B. (1982) Fed. Proc. 41, 666 12 Neutra, M. and Leblond, C.P. (1966) J. Cell Biol .30, 137-150. 13 Bennett, G. and Leblond, C.P. (1970) J. Cell Biol .46, 409-418. 14 MacDermott, R.P., Donaldson, R.M., Jr. and Trier, J.S. (1974) J. Clin. Invest. 54, 545-554. 15 Gashi, A.A., Nadel, J.A. and Basbaum, C.B. (1987) Exp. Lung Res. 12, 83-96. 16 Yanagishita, M. and Hascall, V.C. (1979) J. Biol. Chem. 254, 12355-12364. 17 Roblin, R., Albert, S.O., Gelb, N.A. and Black, P.H. (1975) Biochemistry 14, 347-357.
18 Yanagishita, M. and Hascall, V.C. (1984) J. Biol. Chem. 259, 10270-10283. 19 Iozzo, R.V. (1984) J. Cell Biol. 99, 403-417. 20 Young, R.W. (1973) J. Cell Biol. 57, 175-189. 21 Kimura, J.H., Caputo, C.B. and Hascall, V.C. (19810 J. Biol. Chem. 256, 4368-4376. 22 Mitchell, D. and Hardingham, T. (1981) Biochem. J. 196, 521-529. 23 Little, P.S. and Huang, A.S. (1978) J. Virol. 27, 330-339. 24 Howard, S.C., Sherblom, A.P., Huggins, J.W., Carraway, C.C. and Carraway, K.L. (1981) Arch. Biochem. Biophys. 207, 40-50. 25 Gibbons, R.J. (1982) in Recent Advances in Mucosal Immunity (Strober, W., Hanson, L.A. and Sell, K.W., eds.), pp. 343-351, Raven Press, New York. 26 Clamp, J.R. (1984) Biochem. Soc. Trans. 12, 754-756. 27 Beachey, E.H. (1981) J. Infect. Dis. 143, 325-345. 28 Andersson, B., Dahmen, J., Frejd, T., Leffler, H., Magnusson, G., Noori, G. and Eden, C.S. (1983) J. Exp. Med. 158, 559-570. 29 Parsons, C.L., Stauffer, C.C. and Schmidt, J.D. (1980) Science 208, 605-607.