Effects of canavanine on the secretion of plasma proteins by Hep G2 cells

198

Bioehimica et Biophysica Acta 847 (1985) 198-206 Elsevier

BBA 11587

E f f e c t s o f c a n a v a n i n e o n t h e s e c r e t i o n o f p l a s m a p r o t e i n s b y H e p G 2 cells C o l v i n M. R e d m a n

* and Gabriella Avellino

The Lindslev F. Kimball Research Institute, New York Blood Center, 310 East 67th Street, New York, N Y 10021 (U.S.A.)

(Received June 24th, 1985)

Key words: Secretion: Plasma protein: Canavanine: (Hep G2 cell)

Many secretory proteins contain an amino-terminal propeptide extension which is removed prior to secretion. The point of cleavage is usually marked by a basic pair of amino acids containing arginine. Canavanine, an analogue of arginine, is incorporated into protein and has been shown to inhibit the proteolytic processing of several of these prosecretory proteins. The addition of 3 mM canavanine to Hep G2 cells incubated with L-[3SSlmethionine inhibited the secretion of 11 plasma proteins studied. Of the secretory proteins studied only albumin is thought to contain a propeptide, which is marked by a pair of arginine residues at its point of proteolytic processing. Canavanine had varying effects on the secretion of plasma proteins; ranging from a 43-53% inhibition of secretion of a t antitrypsin and ~xI anti-chrymotrypsin to nearly abolishing (93% inhibition) secretion of transferrin. Canavanine also caused most of the proteins studied to migrate slower on sodium dodecyl sulfate polyacrylamide gel electrophoresis. Two of the canavanine-treated proteins (albumin and transferrin) which underwent marked changes in electrophoretic mobility were more sensitive than untreated proteins to proteolysis by Staphylococcus A ureus V8 proteinase. The slower electrophoretic migration and the greater sensitivity to proteolysis of these proteins may be attributed to marked structural changes caused by the incorporation of canavanine. This suggests that the inhibition of plasma protein secretion by canavanine is not only due to an inhibition of the processing of proteins but may be caused by structural distortions of the secretory proteins.

Introduction The proteolytic processing of secretory proteins, which exist intracellularly as proproteins, is usually marked by a basic pair of amino acids, one or both of which may be arginine [1,2]. Because of this, canavanine, an analogue of arginine, has been used to study the secretion and processing of secretory proteins. The proteolytic cleavage of procollagen [3], proinsulin and proglucagon [4] and proopiomelanocortin [5] are inhibited when the appropriate cells are incubated with canavanine. We have shown that when Hep G2 cells are treated

* To whom correspondence should be addressed.

with canavanine there is an inhibition of albumin secretion and that proalbumin, instead of processed albumin, is released into the medium. Canavanine, however, also inhibits the secretion of other plasma proteins [6]. Many of the plasma proteins, produced by hepatocytes, such as transferrin, are not thought to exist intracellularly as prosecretory proteins [7]. Because of this we have studied the effect of canavinine on the secretion of 11 plasma proteins known to be secreted by Hep G2 cells. The results indicate that canavanine, in addition to inhibiting the processing of proproteins whose point of cleavage is marked by arginine residues, also inhibits secretion of other proteins probably by causing a structural distortion to the proteins. The affected proteins are marked by a retardation

0167-4889/85/$03.30 © 1985 Elsevier Science Publishers B.V. (Biomedical Division)

199 of their electrophoretic mobilities on SDS-polyacrylamide gel electrophoresis and by an increased sensitivity to proteolysis. Materials and Methods

Materials. I_-Canavanine sulfate was purchased from Sigma Chemical Co. St. Louis, MO. Rabbit immunoglobulins monospecific to human albumin, transferrin, fibrinogen, haptoglobin, cqfetoprotein and C3c complement were obtained from Accurate Chemicals, Westbury, NY. Rabbit antiserum to apolipoprotein A, apolipoprotein B, cq-antichymotrypsin, cq-antitrypsin and ceruloplasmin were bought from Calbiochem-Behring Corp., La Jolla, Ca. Protein A-Sepharose CL-4B was obtained from Pharmacia, Piscataway, N J, tissue culture medium and fetal calf serum from GIBCO Laboratories, Grand Island, NY, Trasylol (apatrinin) from Mobay Chemical Corp. NY, and phenylmethylsulfonyl fluoride (PMSF) from Calbiochem-Behring, La Jolla, CA. Hep G2 cells were kindly provided by Drs. B.B. Knowles and D.P. Aden of the Wistar Institute, Phidelphia, PA. Incubation of Hep G2 cells with L-FSS]methionine and canavanine. Hep G2 cells, grown to near confluency in polystyrene petri dishes (60 × 15 mm) (Corning, NY), were incubated at 37°C in a humid a i r / C O 2 atmosphere with Eagle's essential medium supplemented with 10% fetal calf serum. Before the introduction of L-[35S]methionine to the incubation medium the cells were washed with Dulbecco's phosphate-buffered saline and the cells were preincubated for 30 min in minimal essential medium free of methionine. Canavanine-treated cells were preincubated, as described earlier [6], in a similar medium except that 3 mM canavanine was substituted for L-arginine. At the end of the preincubation period the medium was changed, L-[35S]methionine was added, and the cells incubated for periods of up to 4 h. Isolation of radioactive intracellular and secreted plasma proteins. The radioactive plasma proteins secreted into the medium or retained by cells, were isolated by immunoprecipitation and SDS-polyacrylamide gel electrophoresis as previously described for albumin [6]. Prior to treatment with antibodies the washed cells (about 1 . 1 0 6 ) were homogenized in 2 ml of 0.15 M NaC1, 5 mM

EDTA, 50 mM Tris-HCl (pH 7.4), 0.5% SDS, 2.5% Triton X-100, and 100 units/ml Trasylol. The mixture was heated for 3 min in a boiling water bath to solubilize protein and then diluted 5 times with buffer 1 (0.15 M NaC1, 5 mM EDTA, 50 mM Tris-HCl (pH 7.4), 100 units/ml Trasylol and 10 mM PMSF). The radioactive medium was also treated to obtain a final concentration of 0.5% Triton X-100, 50 mM Tris-HCl (pH 7.4), 5 mM EDTA, 0.1% SDS, 100 U / m l Trasylol and 10 mM PMSF. The radioactive plasma proteins present in both the cell extract and the treated medium were then isolated by immunoprecipitation, using monospecific antibodies and protein A-Sepharose, followed by SDS-polyacrylamide gel electrophoresis as previously described [6]. The total radioactivity in the immunoprecipitate was determined by eluting the immune complex from protein A-Sepharose with 8 M u r e a / l % SDS in 0.47 M Tris-phosphoric acid (pH 6.7). The mixture was heated in boiling water for 3 min and the radioactivity in the solubilized proteins was determined in a liquid scintillation counter. When the radioactivity in proteins isolated by SDS-polyacrylamide gel electrophoresis was determined, the radioactive areas, detected by autoradiography, were excised and solubilized with 20% H202 and 20% HC104 and by heating at 70 75°C (for 3 h). The samples were then cooled to 4°C and the radioactivity was determined following the addition of a liquid scintillation phosphor (Hydrofluor) purchased from National Diagnostics, Summervile, NJ. Proteolysis with S. aureus V8. Control and canavanine-treated Hep G2 cells were incubated in the presence of L-[35S]methionine for 4 h at 37°C as described above. The radioactive medium was removed, and both the control and the canavanine-treated cells were washed with Dulbecco's phosphate buffer and recovered from the petri dishes by scraping, and the cells (about 1 • 106) were homogenized in 2 ml 0.15 M NaC1/5 mM E D T A / 5 0 mM Tris-HC1 (pH 7.4)/1% Triton X - 1 0 0 / 0 . 5 % sodium deoxycholate. Insoluble materials were removed by centrifugation and the soluble proteins were divided into two parts. One part was treated at 37°C with 0.02 /~g/ml, 50 u n i t s / m g of S. aureus V 8 proteinase (Miles Scientific) for varying periods of time up to 10 min and

200

the other part was incubated for the same length of time without proteinase. The reaction was stopped by the addition of SDS to a final concentration of 1% and by heating in a boiling water bath for 2 min. The solution was then diluted 10-fold with 0.15 M N a C 1 / 5 m M E D T A / 5 0 mM Tris-HC1 (pH 7.4) and radioactive albumin and transferrin were isolated by immunoprecipitation and SDS-polyacrylamide gel electrophoresis. The radioactivity remaining in intact albumin and transferrin was determined by excising the pertinent portions of the gels and measuring radioactivity. Electron microscopy. The Hep G2 cells, grown in monolayer, were rinsed with 0.1 M sodium cacodylate buffer (pH 7.2), and fixed in situ for 30 rain at room temperature with 3% glutaraldehyde in cacodylate buffer. The fixed cells were washed with cacodylate buffer, scraped from the petri dishes and recovered by centrifugation. The cells were post-fixed with OsO4, dehydrated with graded ethanol solutions and embedded in Epon 812. Thin sections were made and double stained with uranyl acetate and lead citrate. Electron microscopy was performed with a Philips 201 electron microscope. Results

Inhibition of protein synthesis and secretion by' canauanine To determine the optimal concentration of canavanine that inhibits protein secretion, various amounts of canavanine were added to arginine-free incubation medium and Hep G2 cells were incubated with L-[3-~S]methionine for 2 h. The radioactivity of intracellular and secreted total proteins and albumin was measured. Previously it had been shown that under these conditions the incorporation of L-[35S]methionine into total proteins and albumin was linear for up to 4 h and that 3 mM canavanine slightly inhibited (20-25%) protein synthesis but had a more pronounced effect on the secretion of albumin [7]. At 3 mM, canavanine effectively substitutes for arginine [3]. The incorporation of radioactivity into both secreted and intracellular proteins was inhibited by about 25% at 2 6 mM canavanine. Increasing the canavanine concentration to 30 mM inhibited protein synthe-

sis and secretion by 42% (Fig. 1). The effect of canavanine on albumin radioactivity was different. At concentrations up to 30 mM canavanine, the amount of radioactivity in intracellular albumin remained unchanged. Canavanine, however, at concentrations as low as 1 mM, inhibited the secretion of albumin (Fig. 1). Previous pulse-chase studies showed that canavanine inhibits albumin secretion and that the

i0 ~, _o t3 v

g

8t\ x X ~ x ~ r\

Q.

x ~ / / ~

u~ Z

x

6~

61- ~ - - e " ° ~ O ~ o _ # ~ o

0 n~ 13_

¢r"

Z

4o

_d J W ¢r" F"Z

2 ~tsJ

'°I\ I

f0 >F.--

hi

W n~

l

~

,

I

I

I

E

I

I

I

II

~x--#.

p(..) 0

4 • ~

Z

• ~ll ~

/

o

2 ..J

o

i

2, 3, 4 5, 6i 7, e, 9i

l'Oll

3'0

CANAVANINE (mM) Fig. 1. Effect of different concentrations of canavanine on synthesis and secretion of total protein and albumin. Control cells were incubated for 2 h at 37°C with minimal essential medium (free of L-methionine) but containing L-[ 35S]methionine (1126 C i / m tool, 12.5 ~Ci/ml). The canavanine-treated cells were similarly incubated but the minimal essential medium was free of L-arginine and contained different amounts of canavanine as indicated in the above figures. The radioactivities of trichloroacetic acid-precipitable proteins and of albumin, in both the incubation medium and within the cells, was determined. The top panel shows the radioactivity in total trichloroacetic acid precipitable proteins and the bottom panel that in albumin. × , secreted protein radioactivity; O, intracellular protein radioactivity.

201

nonsecreted albumin is retained intracellularly [6]. Since, near maximal inhibition of total protein, secretion was obtained with 3 mM canavanine, this concentration was used in subsequent experiments in which the secretion of individual plasma proteins was measured. Canavanine (3 mM) inhibited the secretion of all 11 plasma proteins studied. However, there were marked differences in the extent to which the secretion of various proteins was inhibited. For example, the secretion a~-antichymotrypsin and c~l-antitrypsin were inhibited by only 40-50%, while that of transferrin and complement C3 were nearly completely abolished (97%). The secretion of other plasma proteins studied was inhibited by intermediate amounts (Table I).

Effect of canaoanine on the electrophoretic mobiHties of plasma proteins on SDS polyacrylamide gels" Four nascent plasma proteins, which are produced by Hep G2 cells in substantial amounts and whose secretion is affected to different extents by canavanine, were analyzed by SDS-polyacrylamide gel electrophoresis and autoradiography. Both the secreted proteins and those which remained intracellularly were investigated. As previously

TABLE l T H E E F F E C T OF C A N A V A N I N E ON T H E I N H I B I T I O N OF S E C R E T I O N OF D I F F E R E N T P L A S M A P R O T E I N S Hep G-2 cells were incubated with L-[35Slmethionine for 4 h at 37°C in the presence and absence of 3 m M canavanine as described in Fig. 1. The proteins secreted into the medium were isolated by immunoprecipitation and their radioactivity was determined. The values given are the averages of three experiments. Plasma proteins

% inhibition

Albumin cq-Antichymotrypsin cq-Antitrypsin Apolipoprotein A Apolipoprot'ein B Ceruloplasmin Complement C3c cq-Fetoprotein Fibrinogen Haptoglobin Transferrin Trichloroacetic acid-precipitable proteins

65.1 _ 4.3 43.3 ___15.0 53.4+ 3.4 65.8 _+_+12.1 82.2 + 0.7 66.1 + 8.5 96.5+ 0.7 81.6+ 2.9 75.3 _+11.2 70.7 _+ 2.8 97.0 + 2.0 62.0 + 9.0

shown, canavanine treatment caused both intracellular and secreted nascent albumin to migrate more slowly upon electrophoresis. Albumin, containing canavanine, migrated with an apparent molecular size of 70000 instead of 66000 (Fig. 2A). Previous studies showed that canavaninetreated albumin is mostly proalbumin [6]. However proalbumin, which contains a hexapeptide extension at the amino terminal sequence, does not separate on SDS-polyacrylamide gel electrophoresis from processed albumin. Thus the slower migration is probably due to a structural conformation change in the protein. A similar pattern was seen with canavanine-treated transferrin. When isolated from the intracellular fraction, transferrin containing canavanine migrated more slowly than intracellular transferrin isolated from control cells (Fig. 2A). In the incubation medium radioactive transferrin was present in detectable amounts only when cells were not treated with canavanine. Canavanine virtually abolished the secretion of transferrin (Fig. 2A, lane 6). In control cells the secreted transferrin migrated as a larger protein than intracellular transferrin (Fig. 2A, lanes 5 and 7). This is probably due to incomplete glycosylation of the intracellular protein as compared to the fully glycosylated secreted proteins. In these studies we have not, however, measured the extent of glycosylation of the various glycoproteins nor the effect of canavanine on this process. When proteins whose secretion are least affected by canavanine, cq-antichymotrypsin and a~-antitrypsin were analyzed by a similar procedure it was noted that there was little difference in the electrophoretic mobilities of canavanine-treated and control proteins. Nascent secreted a~-antichymotrypsin from both control and canavaninetreated cells migrated similarly (lanes 1 and 2 Fig. 2B). lntracellular % antichymotrypsin could be separated into two distinct proteins, one with a size similar to that of the secreted protein, and the other (approx. 56 000) migrating as a smaller protein. Both intracellular species migrated the same in control and canavanine-treated cells. Radioactive %-antitrypsin was not present intracellularly in amounts sufficient to be detected by this procedure but, when isolated from the medium, both the secreted canavanine-treated and the control

202

A

B

9

m

66

-

45

-

32

-

i

....

21 14-

I

2

3

ALBUMIN

4

5

6

7

8

TRANSFERRIN

I

2

5

4

(~1 onfichymotrypsin

5

6

(~1ontitrypsin

Fig. 2. Effect of canavanine on the secretion and electrophoretic mobility on SDS-Polyacrylamide gel electrophoresis, of albumin, cq-antichymotrypsin, al-antitrypsin and transferrin. Hep G-2 cells were preincubated for 30 rain at 37°C in methionine-free minimal essential medium. The canavanine-treated cells were preincubated in arginine-free medium with 3 mM canavanine. At the end of the preincubation period, fresh medium (with and without 3 mM canavinine), which contained 62.5 /*Ci/ml of 1,-[35S]methionine (100 Ci/mmol), was added and the cells were further incubated for 4 h at 37°C. Radioactive cq-antichymotrypsin, ~l-antitrypsin, albumin and transferrin were isolated by immunoprecipitation and 7.5% SDS-polyacrylamide gel electrophoresis, from the secreted medium and from an intracellular detergent-soluble fraction. Autoradiograms of the SDS-polyacrylamide gel are shown. Panel A, lane 1 contains secreted albumin from the control cells; lane 2, secreted albumin from canavanine-treated cells; lane 3, intracellular albumin from control cells; lane 4, intracellular albumin from canavanine-treated cells; lane 5, secreted transferrin from control cells; lane 6, secreted transferrin from canavanine treated cells; lane 7, intracellular transferrin from control cells and lane 8, intracellular transferrin from canavanine-treated cells. In panel B, lane 1 contains cq-antichymotrypsin secreted into the medium by canavanine-treated cells; lane 2, cq-antichrymotrypsin secreted by control cells; lane 3, intracellular al-antichymotrypsin from canavanine-treated cells; lane 4, cq-antichymotrypsin from control cells; lane 5, a~-antitrypsin secreted by canavanine-treated cells and lane 6, al-antitrypsin secreted by control cells.

proteins had a similar electrophoretic migration ( F i g . 2B, l a n e s 5 a n d 6). An unidentified radioactive protein, with a m o l e c u l a r size o f o v e r 2 0 0 0 0 0 often co-imm u n o p r e c i p i t a t e d w i t h t h e v a r i o u s a n t i b o d i e s (see l a n e 5, Fig. 2A).

Sensitivity of canavanine-treated proteins to proteolysis Proteins be more or ing on the t e i n s to t h e

in d i f f e r e n t c o n f o r m a t i o n a l s t a t e s m a y less s u s c e p t i b l e t o p r o t e o l y s i s d e p e n d e x p o s u r e o f p e r t i n e n t sites o f t h e p r o proteolytic enzyme. Thus sensitivity to

203

I00

80

60

xx

o.

I.tJ I-O OC rl

"'o-__

40

~- 20 C)

0

Z

_z I00=

.

.

.

.

N

Morphological studies

w 60

40

o

20

"-c) .......

0

p r o t e o l y s i s has b e e n used as a m e a s u r e of structural changes of p r o t e i n s [8]. T o d e t e r m i n e w h e t h e r o r not two of the secretory proteins, whose secretion was i n h i b i t e d b y c a n a v a n i n e treatment, have a l t e r e d structures, a l b u m i n a n d transferrin were t r e a t e d with a p r o t e o l y t i c enzyme, f r o m S. A ureus V8, for various p e r i o d s of time a n d the degree of p r o t e o l y s i s was m e a s u r e d b y d e t e r m i n i n g the percent of i n t a c t p r o t e i n s recovered after imm u n o p r e c i p i t a t i o n a n d S D S - p o l y a c r y l a m i d e gel electrophoresis. Both a l b u m i n a n d transferrin isolated from c a n a v a n i n e - t r e a t e d cells were m u c h m o r e susceptible to proteolysis than the corres p o n d i n g p r o t e i n s isolated from u n t r e a t e d cells (Fig. 3). This, together with the altered m i g r a t i o n of these p r o t e i n s on S D S - p o l y a c r y l a m i d e gel electrophoresis, indicates that c a n a v a n i n e t r e a t m e n t causes a m a r k e d structural change in these p r o teins.

i

I

2

I

3

i

4

I

I

67

I

-~

;b,'o

MINS

Fig. 3. Increased sensitivity of canavanine-treated transferrin and albumin to proteolysis by V8 proteinase. Hep G-2 cells were incubated in the presence of e-[35S]methionine for 4 h at 37°C. One set of cells contained 3 mM canavanine instead of L-arginine. The radioactive medium was removed, the cells washed in phosphate-buffered saline, recovered from the petri dishes and homogenized in the presence of 1% Triton X-100 and 0.5% sodium deoxycholate. Insoluble materials were removed by centrifugation and the supernatant fraction was treated at room temperature with 0.02 /Lg/ml of S. aureus V8 proteinase for various periods of time up to 10 min. The reaction was stopped by the addition of SDS to a final concentration of 1% and heating in a boiling water bath for 2 min. The solution was then diluted with buffer 1 to obtain an SDS concentration of 0.1%, and transferrin and albumin were isolated by immunoprecipitation and SDS-polyacrylamide gel electrophoresis. The radioactivity remaining in intact transferfin and albumin was measured. The top panel shows the rate of proteolysis of albumin and the bottom panel that of transferrin. e, control protein; O, canavanine-treated protein.

T h e m o r p h o l o g y , at the level of the electron m i c r o s c o p e , of c a n a v a n i n e - t r e a t e d a n d u n t r e a t e d H e p G 2 cells was c o m p a r e d . A l t h o u g h the biochemical evidence indicates that a great p a r t of the secretory p r o t e i n s synthesized is not secreted, there was no evidence of an increased n u m b e r of secretory vesicles within the cells, n o r of an enlargem e n t of the e n d o p l a s m i c reticulum or of the G o l g i a p p a r a t u s (Fig. 4). The n u m b e r of lysosomes or a u t o p h a g o l y s o s o m e s also a p p e a r e d to be n o r m a l in c a n a v a n i n e - t r e a t e d cells. It has been p r o p o s e d that the i n c o r p o r a t i o n of c a n a v a n i n e into secretory p r o t e i n s causes t h e m to a p p e a r ' a b n o r m a l ' a n d thus these p r o t e i n s are m o r e susceptible to internal d e g r a d a t i o n [9-12], p r e s u m a b l y via l y s o s o m a l enzymes. In general, the c a n a v a n i n e - t r e a t e d cells app e a r e d to have n o r m a l gross m o r p h o l o g y (Fig. 4). Discussion

C a n a v a n i n e , an arginine analogue, c o m p e t i tively inhibits the charging of t - R N A Arg [33] a n d is itself i n c o r p o r a t e d into p r o t e i n s [3,16,13,14]. It has b e e n used to s t u d y the p r o t e o l y t i c processing of a n u m b e r of p r o s e c r e t o r y proteins, since the p o i n t of cleavage of these p r o t e i n s is frequently m a r k e d by an arginine residue [3-6]. However, t r e a t m e n t of a variety of cells with c a n a v a n i n e shows that it has

204

Fig. 4. Electron micrographsof canavanine-treated Hep G-2 cells. Hep G-2 cells were incubated with or without 3 mM canavanine for 4 h as described in Fig. 2. Panel a shows a large portion of the cytoplasm of a control cell and panel b that of a canavanine-treated cell. Magnification 7850×. Bar is approx. 500 nm.

multiple cellular effects, which include causing many intracellular proteins to undergo changes in electrophoretic mobility [3,15 17] and increased degradation [9 12]. In these latter studies the effect of canavanine on total cellular proteins rather than on individual proteins was described. Transferrin, one of the main secretory proteins produced by hepatocytes, is secreted, together with a number of other plasma proteins, by Hep G2 cells [18]. Although transferrin, like other plasma proteins, is first synthesized as a larger precursor protein with a pre-(signal) extension, it does not occur intracellularly with an N-terminal prosegment [7]. Yet, canavanine treatment of Hep G2 cells almost completely abolishes the secretion, but not the synthesis, of this protein. This indicates that canavanine inhibits secretion by mechanisms other than by affecting proteolytic cleavage of proproteins. As described for other proteins, canavanine incorporation into transferrin causes this protein to migrate more slowly on SDS-polyacrylamide gel electrophoresis. The reasons why proteins which contain canavanine instead of

arginine have different electrophoretic mobilities, even though these proteins are reduced and treated with 6M urea and 1% SDS, are not understood. This artifact of electrophoretic migration could be confusing when canavanine is used, as an arginine analogue, to study proteolytic processing of secretory protein. Slower migration on SDS-polyacrylamide gel electrophoresis is usually equated with the occurrence of the larger proprotein. Nterminal amino sequencing, as was performed for canavanine-treated pro-albumin [6], in addition to migration on SDS-polyacrylamide gel electrophoresis, must be performed in such studies to determine whether or not canavanine affects proteolytic processing. Canavanine inhibited the secretion of all plasma proteins studied, but it affected the secretion of some proteins more than others. The secretion of ch-antichymotrypsin and ~l-antitrypsin was minimally affected (40-50%), while that of other proteins such as complement C3c and transferrin was nearly abolished. The secretion of albumin, the main secretory protein produced by these cells,

205 was inhibited by 65%. Presumably, all of these proteins contain arginine residues which may be substituted with canavanine and this by itself may sufficiently change the structure of the proteins to enable them not to be secreted efficiently. Based on changes in electrophoretic mobility it appears that transferrin and albumin undergo greater structural changes than al-antichymotrypsin and a~-antitrypsin. It is not possible, however, to correlate strictly the inhibition of secretion with changes in electrophoretic mobility. That canavanine may cause a structural change that leads to an unfolding or denaturing of proteins is supported by the fact that canavanine-treated albumin and transferrin, which have slower migration on SDS-polyacrylamide gel electrophoresis, are more susceptable to limited proteolysis by S. a u r e u s V8 proteinase than are the normal proteins. Since canavanine causes structural changes in proteins and the proteins are isolated by immunoprecipitation, it is possible that the altered proteins may not immunoprecipitate quantitatively. If this were the case, secretion of canavanine-modified proteins would appear, wrongfully, to be inhibited. However, the antibodies used are monospecific, but polyclonal, and should react to many different epitopes on the proteins. These antibodies do immunoprecipitate canavanine-modified proteins, as is shown by their reaction with intracellular nonsecreted proteins. For example in both control and canavaninetreated cells anti-albumin immunoprecipitates the same amount of nascent intracellular albumin, yet differentiates between different amounts of albumin secreted. We cannot, however, rule out the possibility that different proteins may be affected to differing extents and that some of the inhibition of secretion may be attributed to incomplete immunoprecipitation of the structurally modified proteins. In vivo, transferrin and albumin are transported intracellularly and secreted from the liver at different rates [19]. Studies with hepatocytes in culture confirm that the various plasma proteins are secreted at different rates and that the ratelimiting step may be the movement of these proteins from the rough endoplasmic reticulum to the Golgi apparatus [20-23]. This had led to the postulation that the cisternal surface of the endo-

plasmic reticulum contains receptors which regulate the rate at which secretory proteins are transported from one compartment to another [22]. The incorporation of canavanine into secretory proteins may distort the structure of the secretory proteins, some undergoing greater changes than others, and these ' a b n o r m a l ' proteins may no longer be recognized by the receptors which are involved in intracellular transport. Secretory proteins which are greatly modified by canavanine, and are not secreted, may be shunted from the secretory route to a degradation pathway. The mechanism by which this may be accomplished' is not understood. In this regard it should be noted that proteins made by fibroblasts grown in the presence of canavanine are degraded twice as rapidly as normal proteins [12]. Depending on the concentration used and on the availability of arginine, canavanine may substitute for all or only some of the arginine residues in proteins which are actively being synthesized. This may cause multiple effects, both on individual proteins and on various aspects of the physiology of the cell. In spite of this, Hep G2 cells, grown in the presence of canavanine for 4 h, show normal morphology with no apparent disruption to the organelles involved in protein synthesis and secretion. Canavanine may be incorporated to a greater extent into secretory proteins than into other cellular proteins, since secretory proteins undergo more active synthesis. Thus treatment of cells with canavanine may show a preferential effect on secretory proteins, and yet the effect of canavanine on secretion may not be due to its expected inhibition of processing of proproteins but may be due to a more general effect in which the structures of the secretory proteins are altered.

Acknowledgements We thank Tellervo Huima for her work on electron microscopy and for photography. This work was supported by grant HL09011 from the National Institutes of Health.

References 1 Potts, J.T., Kronenberg, H.M., Habener, J.F. and Rich, A. (1980) Ann. NY Acad. Sci. 343, 38-54

206 2 Steiner, D.F., Quinn, P.S., Patzelt, C., Chan, S.J., Marsh, J. and Tager, H.S. (1980) in Cell Biology: A Comprehensive Treatise (Prescott, D.M. and Goldstein, L., eds.), Vol. 4, 175 202, Academic Press, New York 3 Schein, J., Harsch, M., Cywinski, A. and Rosenbloom, J. (1980) Arch. Biochem. Biophys. 203, 572-579 4 Noe, B.D. (1981) J. Biol. Chem. 256, 4940-4946 5 Hoshina, H., Hortin, G. and Boime, I. (1982) Science 217, 63-64 6 Redman, C.M., Avellino, G. and Yu, S. (1983) J. Biol. Chem. 258, 3446-3452 7 Schreiber, G., Dryburgh, H., Weigand, K., Schreiber, M., Witt, I., Seydewitz, H. and Howlett, G. (1981) Arch. Biochem. Biophys. 212, 319-328 8 Scheele, G. and Jacoby, R. (1982) J. Biol. Chem. 257, 12277-12282 9 Knowles, S.E., Gunn, J.M., Hanson, R.W. and Ballard, F.J. (1975) Biochem. J. 146, 595 600 10 Hendil, K.B. (1976) J. Cell. Physiol. 87, 289-296 11 Dean, R.T. and Riley, P.A. (1978) Biochim. Biophys. Acta 539, 230-237 12 Fong, D. and Poole, B. (1982) Biochim. Biophys. Acta 696, 193-200

13 Bradley, M.O., Hayflick, L. and Schimke, R.T. (1976) J. Biol. Chem. 251, 3521-3529 14 Prouty, W.F. (1976) J. Cell Physiol. 88, 371-382 15 Attias, J., Schlesinger, M.J. and Schlesinger, S. (1969) J. Biol. Chem. 244, 3810-3817 16 Murphy, E.C. and Arlinghaus, R.B. (1980) J. Biol. 33, 954-961 17 Pines, M., Rosenthal, G.A. and Rosenbloom, J. (1980) Proc. Natl. Acad. Sci. USA 78, 5480-5483 18 Knowles, B.B., Howe, C.A. and Aden, D.P. (1980) Science 209, 497-499 19 Morgan, E.H. and Peters, T., Jr. (1971) J. Biol. Chem. 246, 3508-3511 20 Strous, G.J. and Lodish, H.F. (1980) Cell 22, 709-717 21 Fitting, T. and Kabat, D. (1982) J. Biol. Chem. 25 14011 14017 22 Lodish, H.F., Kong, N., Snider, M. and Strous, G.J., (1983) Nature 304, 80-83 23 Ledford, B.E. and Davies, D.F. (1983) J. Biol. Chem. 258, 3304-3308