Eur. J. Biochem. 58,9- 14 (1975)
Lysosomal Enzymes as Agents of Turnover of Soluble Cytoplasmic Proteins Roger T. DEAN Department of Experimental Pathology, University College Hospital Medical School, London (Received February 5 / April 28, 1975)
The degradation of cytosol proteins in vitro by purified cathepsin D and cathepsin B, and by mixtures of lysosomal enzymes, was studied. By means of a double-labelling method, it was shown that the relative rates of degradation of cytosol proteins by the purified enzymes and by mixtures of enzymes under a wide range of conditions in vitro correlated well with their relative rates of turnover in vivo. The complex mixture of cytosol proteins was degraded less rapidly after denaturation than in the native state, both by the purified proteases and by the mixture of lysosomal enzymes. This contrasts with previous results on proteolysis of single purified proteins. The possible role of lysosoma1 enzymes in turnover in vivo was discussed.
The double-labelling method of Arias etal. [l] distinguishes proteins which turn over rapidly in vivo from those which turn over slowly, by virtue of their isotope ratios : proteins turning over fast show a high ratio of 3H to 14C radioactivity. Using such double-labelled proteins, it has been found that the rates of degradation of cytosol proteins by trypsin and pronase in vitro correlate well with their rates of turnover in vivo [ 2 ] . It has been suggested on the basis of studies on enzyme inactivation in vitro [3 - 61 and proteolysis during liver perfusion [7,8],that turnover of intracellular proteins may be largely a function of the lysosomal system ; although some work [9] indicates the involvement of at least two systems. Clearly, the process of autophagy [lo] provides a mechanism for internalising any cellular component within lysosomes, allowing subsequent digestion. Thus the present work describes the degradation in vitro by lysosomal proteases of double-labelled cytoplasmic proteins, in order to asses further the possible involvement of such proteases in turnover of such proteins in vivo. MATERIALS AND METHODS In the double-labelling procedure El], rats (150 g) were given intraperitoneal injections of 50 pCi [14C]leucine, and, 96 h later, 100 pCi ['Hlleucine. 4 h later Enzymes. Cathepsin D (EC 3.4.23.5); cathepsin B, 3.4.22.1).
(EC
the animals were killed, and livers homogenised (five passes of a tight Dounce) in two volumes of 0.15 M NaCI- 50mM Tris-HC1, pH 7.8, a medium providing good osmotic protection for lysosomes [ll]. The homogenates were centrifuged at 10000 x g for 10 min, and the resulting supernatants centrifuged at 100000 x g for 30 min to give a soluble cytoplasmic fraction. This fraction, will be referred to as cytosol protein mixture but it must be noted that it is not the same as the substrate used by Ansorge etal. 1121, which is a mixture of high-molecular-weight cytosol proteins. Cytosol protein mixture was dialysed for 24 h against 50 vol. of 50 mM Tris-HC1, pH 7.8, with two changes, and the dialysate was used for proteolytic digestions. A portion of cytosol protein mixture was gel-filtered on Sepharose 6B in the same buffer ( V , = 320 ml, V, = 105 ml). Three fractions, each highly heterogeneous, were obtained, by pooling material : fraction 1, emerging in a peak at Vo (molecular weights greater than about 1.5 x lo6); fraction 2, emerging between 175-225 ml (molecular weights 6 x lo52 x lo5; fraction 3, between 225 - 275 ml (molecular weights 2 x lo5- 5 x lo4). Molecular weights (except for fraction l), were estimated from the elution positions of proteins of known molecular weight. The correlation between subunit size and 'H/14C ratio of cytosol protein mixture components [2) was observed in sodium dodecylsulphate - polyacrylamide gel electrophoresis [13]. Soluble mixtures of lysosomal enzymes were obtained by extracting a total lysosomal-mitochondria1
10
fraction [14]; or by a preparation of lysosomes loaded with Triton WR-1339 [15-171; with 0.1 % Triton X-100, and centrifuging at 100000 x g for 60 min, to remove particulate material. In each case livers were extensively perfused with saline before homogenisation to reduce the amount of serum (containing proteinase inhibitors) present. For some experiments, the resuspended particulate material was used as a source of enzymes, and in others the Triton-treated suspensions were used without centrifugation. Purified rabbit liver cathepsin D [18] and human liver cathepsin B, [19] were kindly given by Dr A. J. Barrett. Assays for cathepsin D and cathepsin B, were as described by Barrett [20]: method 1 in each case. The unit of cathepsin B, activity is that amount of enzyme which will hydrolyse 1 pmole of substrate per hour under assay conditions. Proteolytic digestions were conducted at 37 “C for up to 96 h, in stoppered tubes, with the following buffers (100 mM final): pH 2.5, glycine-HC1; pH 3.0 sodium formate; pH 4.3, 5.0 sodium acetate; pH 6.0, 7.0 potassium phosphate; pH 8.0, 9.0 Tris-HCl. Substrate concentration was 10 mg/ml, 14C 31 000 dis. x min-’ x ml-’, 3H 68000 dis. x min-’ x ml-’, and the amounts of the purified enzymes used were: cathepsin D : 10 U/ml; cathepsin B, : 1.4 U/ml(l mM dithiothreitol added [21]). 10 mg of lysosomal proteins was present in incubations with crude lysosomal components, while 1 mg of lysosomal protein was used in those with tritosomal components. Incubations were usually in a total volume of 1 ml, and samples were removed at various times for determination (by scintillation counting) of radioactivity soluble in 5 % trichloroacetic acid. There was no loss of cathepsin D activity during incubation for up to 96 h, while cathepsin B1 fell to 70% of the original activity by 96 h. In certain experiments in which “intact” lysosomes were required, short incubations (up to 1 h) the presence of 0.25 M sucrose (for osmotic protection) were used, to minimise disruption of lysosomes. When suspensions of broken lysosomes or extracts thereof were used, Triton X-100 was maintained at 0.1 % to ensure disruption. RESULTS General Characteristics of the Degradation of Cytosol Proteins by Lysosomal Enzymes
The release of trichloroacetic-acid-soluble radioactivity from the cytosol protein mixture was never linear with time for more than 30 min. Thus, in order to assess the pH-dependence of the digestions, incubations at various pH values were compared after 1 h, 4 h and 24 h. pH optima observed were not dependent on length of incubation or on the isotope observed. Autodigestion of the cytosol protein mix-
Lysosomes and Turnover
w Fig. 1. p H dependence of hydrolysis of cytosol protein mixture by carhepin B, rrnd D. Dialysed cytosol protein mixture (10mg/ml) W d S incubated at the specified pH values with either cathepsin B, (1.4 U/ml; A) or cathepsin D (11 U/ml; 0)and trichlorodceticacid-soluble radioactivity measured at 4 h
ture, in incubations without added lysosomal enzymes, was very slow (at pH 3 or 5, always less than 10% of that by 11 U/ml of cathepsin D), as reported previously [22]. The pH optimum for autoproteolysis was 5.0, and there was slight activation by 1 mM dithiothreitol, which was most pronounced at pH 5.0 (in a 24-h incubation, pH 5.0: 25 %). Cathepsin D gave fastest digestion of the cytosol protein mixture at pH 5, with considerable activity at pH 3 (Fig. 1). No activity was detected at pH 7.0, allowing for the slight auto-proteolysis of the cytosol protein mixture at this pH. Furthermore, the very slow release of trichloroacetic-acid-soluble radioactivity during incubations at pH 7.0, was not affected by the presence of pepstatin (an inhibitor of cathepsin D and other acid proteases [23-251 at 10 pg/ml, either in the presence or absence of rabbit cathepsin D. At pH 5.0, pepstatin (10 pg/ml) completely abolished digestion of the cytosol protein mixture by cathepsin D, while dithiothreitol was also inhibitory (up to 20% inhibition by 10 mM dithiothreitol in 3-h incubations). Liberation of trichloroacetic-acid-soluble radioactivity at pH 7.0 by cytosol peptidases from the products of a brief cathepsin D digestion (2 h; stopped by the addition of pepstatin) was not significantly more rapid than from undigested cytosol protein. Thus effects of such cytosol peptidases were small, in agreement with [12]. Cathepsin B, showed a pH optimum of 4.3, with significant activity between pH 3 and 7 (Fig. 1). Release of radioactivity in 12 h by 1.4 U/ml of cathepsin B1 (pH 4.3) was 60% of total, whereas that by 10 U of cathepsin D was only 45%. This difference in rates of hydrolysis was apparent between 2 h and 24 h (Fig. 2). The optimum for digestion by extracts of both crude and purified lysosomes (see Methods) was pH 3, with activity detectable between pH 2.5 and pH 9. The
R. T. Dean
11
Time of digestion ( h )
Fig. 2. Hydrolysis oj’native and denatured cytosolprotein mixture by carhepsin B , , cathepsin D and a tritosome extract. Cytosol protein mixture (dialysed) (10 mg/ml) was incubated at the specified pH with eithercathepsinB, (1.4 U/ml)orcathepsinD(ll U/ml)oratritosome extract (1 mg/ml protein) and trichloroacetic-acid-soluble radioactivity measured at various times. Symbols: (0)cathepsin B,, at pH 4.3 with native cytosol protein mixture; (A) cathepsin B,, at pH 4.3 with denatured cytosol protein mixture; (0)cathepsin D at pH 5.0 with native cytosol protein mixture; (A) cathepsin D at pH 3.0 with denatured cytosol protein mixture; (0)crude lysosomal extract at pH 3.0, 0.1 % Triton X-100, with native cytosol protein. Cytosol protein was denatured by treatment at pH 1.5 (see text)
Table 1 . Eflect of dithiorhreitol on digestion of the cylosol protein mixture by a tritosomal extract, at various p H values. Incubations were as described in legend to Fig.2, for 24 h either without dithiothreitol (controls) or with 1 mM dithiothreitol (tests). It should be noted that the “activations” observed may represent stimulation or stabilisation of enzyme activities, amongst other possibilities. The results expressed are the difference between digestion in control and test incubations as a percentage of the control value (measured as liberation of trichloroacetic-acid-soluble radioactivity)
PH
Effect of dithiothreitol
that obtained with lysosomes disrupted by 0.1 % Triton X-100 before incubation. Very low rates of digestion (less than 0.5% of total radioactivity released in 1 h) were observed in incubations at pH 5 and pH 7. Thus there was no satisfactory evidence for the digestion of cytosol-protein by intact lysosomes, as these low rates of digestion can probably be explained by lysosoma1 breakage. Experiments comparing hydrolysis at pH 5.0 by cathepsins D and B1 with that by mixtures of the two enzymes (both at the usual concentrations) gave no evidence for synergism: after 2, 6, 24 and 48 h of digestion release by the mixture of enzymes was equal to the sum of the releases by the enzymes separately. Because it has been noted that some proteins are hydrolysed by proteases more rapidly after denaturation in the native state (see for instance [27]), the generality of this was assessed using the cytosol protein mixture as substrate. The cytosol protein mixture was denatured by two means, by boiling for 15 min at pH 7.0 or by titrating to pH 1.5 with HCl, and after 48 h a t 4 “C,retitrating to pH 7.0. While the pH optima for digestion of denatured cytosol protein mixture by cathepsin B, and a tritosomal extract rcinained pH 4.3 and pH 3.0, respectively, that for cathepsin D was changed to pH 3.0. When digestions of native and denatured cytosol protein mixture by cathepsins B, and D and by tritosomal extracts (under optimal conditions of pH) were compared (Fig. 2) it was found that in each case, degradation of the denatured proteins was the slower. Whereas cathepsin B, digested native cytosol protein mixture substantially more rapidly than cathepsin D, the two enzymes degraded denatured cytosol protein mixture at similar rates.
~
% control 3 5 6 7 9
- 25
+ 35 + 23 + 22
+ 10
time-course of digestion by a lysosomal extract is shown in Fig.2. Activation by 1 mM dithiothreitol was detectable between pH 5 and pH 9 (Table 1). Similar pH dependences were observed for digestion by resuspended particulate fractions from the crude lysosome preparations: activity at pH 3 (2-h incubations) of the soluble fraction was 160% of that of the particulate fraction (with the crude preparations much cathepsin B1 was solubilised by Triton X-100, but this was not true of tritosomes [17]). In 30-min incubations with osmotic protection, “intact” crude lysosomes were again most active at pH 3.0: however lysosome rupture at low pH is rapid [26], and the release of radioactivity was similar to
Correlation of Digestion in vitro with Turnover in vivo
The susceptibility of native cytosol proteins to trypsin and pronase at pH 7.0 correlates well with their turnover rates in vivo [2] : thus trichloroaceticacid-soluble fragments released initially have high 3H/14Cratios, and later fragments have lower ratios. This correlation was initially only sought in experiments at neutral pH with commercially available enzymes. In the present work intrinsic liver lysosomal enzymes have been used under a wider variety of conditions. The correlation was clearly demonstrated with cathepsin D at pH 3, pH 5 (Fig. 3 ) and pH 6, and with cathepsin B, at pH 4.3, pH 5 , and pH 6 . With lysed suspensions of tritosomes or of crude lysosomes, the correlation was observed between pH 3 and 8. This characteristic of digestion with extracts of lysosomes was first pointed out by Bohley et al. [28] and recently confirmed [29]. Even in digestions with denatured cytosol protein mixture, an initial release of high
12
Lysosomes and Turnover
’6 0 ._
2!
I
I
0.4;
€0
I
I
240
120 180 Time of digestion (min)
Fig. 3 . Hydrolysis of’native cytosol proteins by cathepsin D. Cytosol protein mixture (10 mg/ml) was incubated at pH 5.0 with cathepsin D (1 1 Ujrnl) and trichloroacetic-acid-soluble 3H and 14C radioactivity determined at _various times, and their ratios calculated. The isotope ratio of the cytosol protein used, was 0.55 (arbitrary units)
0” 0
I
4
20
40 Time of digestion (min)
I
100
Fig. 4. Susceptibility of cytosol proteins of various size ranges to cathepsin D.Hydrolyses were performed at pH 3.0 with 11 U/ml of cathepsin D. Substrate concentration in each case was adjusted to 1 mg/ml protein. Symbols: (0) large proteins (see Methods); (0)intermediate proteins; (A) small proteins
ratio material by cathepsin D was observed, as noted previously with trypsin [30]. The digestibility of molecules of different size ranges was also compared using fractions of the cytosol protein mixture obtained by gel filtration, each at 1 mg/ml. The correlation of digestibility with molecular size was very clear (Fig.4). As mentioned above, in experiments with “intact” lysosomes, no evidence was obtained for digestion in vitro by intact lysosomes. It has been claimed that this process occurs [31], and that proteins can be internalised by lysosomes in vitro, although this has been
strongly disputed [32,33]. Since the published experiments have involved crude lysosomal fraction, and have merely measured co-sedimentability of radioactive protein with lysosomes, I have attempted to demonstrate uptake of proteins into the soluble phase of the purified lysosomes i. e. to sites from which they can be released by rupturing lysosomes with 0.1 ”/, Triton X-100. Thus after incubation of lysosomes with cytosol protein mixture and also with highly labelled serum, and repeated washing of lysosomes, the amount of non-sedimentable radioactivity released by Triton X-100-0.25 M sucrose was compared with that released by sucrose alone. Numerous controls for release from membrane-bound sites, including those generated during the incubations (30 min at 37 “C or 4 “C) were performed. No evidence for uptake was obtained either in presence or absence of ATP, in agreement with results of Huisman et al. [32,33].
DISCUSSION The degradation of several purified proteins by extracts of lysosomes has been studied previously (see for instance [34- 36]), and some correlation between susceptibility to denaturation in vitro and susceptibility to lysosomal enzymes noted [34]. In addition a few cytosol enzymes have been used as experimental substrates for lysosomal enzymes (see review in [22]), while the use of mixed cytosol proteins as substrates has been pioneered by Bohley et al. [12,22]. The degradition of mixed cytosol proteins by highly purified lysosomal enzymes has however, received little attention. In the present work the digestion of this substrate by cathepsins D and B1 has been outlined. Whereas it has been previously claimed that purified chicken cathepsin D [37] does digest the cytosol protein substrate of Ansorge et al. [12] at pH 6.9, this was not observed here with rabbit cathepsin D acting on cytosol protein mixture. The optimum of digestion by cathepsin B1 was 4.3, which is lower than its optimum (pH 6 ) against synthetic substrates and azo-haemoglobin [19]. The relative contribution of the various lysosomal proteases to the digestion of cytosol proteins remains to be established, although the efficiency of digestion by cathepsin B, and the stimulation of proteolysis by dithiothreitol suggests that thiol proteinases may well be important. Whereas previous work suggests that proteins are more susceptible to proteolysis after denaturation than in the native state [27], this was not observed in the present work. The pH optimum of cathepsin D versus denatured cytosol protein mixture was lowered to pH 3, but those of cathepsin B, and a tritosomal extract were as for native cytosol protein mixture. Even when di-
R. T. Dean
gestions were conducted at the relevant pH optimum, the denatured cytosol protein mixture was not more susceptible to either the purified or the crude mixture of lysosomal enzymes. A crucial difference between most published experiments and those reported now is in the complexity of the substrate used. Previous work has considered the digestion of a limited number of purified proteins, used singly; in contrast, the present work concerns proteolysis of a complex mixture of proteins, and is thus a more general test of the idea that denatured proteins are more susceptible to digestion than are native proteins. It is notable that in a previous experiment using such a complex mixture of proteins, it was found that the denatured proteins were less susceptible to trypsin than the native mixture [12], in agreement with the present results. Thus the idea that denatured proteins are more susceptible to hydrolysis by lysosomal or other proteases [27,34] is not generally applicable. Indeed, it has been found that although molecules of tyrosine aminotransferase containing certain amino acid analogues are much more easily denatured than normal enzyme, the turnover rates in vivo of modified and normal enzyme are the same [38]. Although Linderstrom-Lang (see [39]) has suggested that initial proteolytic attack catalyses subsequent rapid denaturation of the substrate, there is evidence against this in the case of subtilisin acting on human haemoglobin (see [40]). Thus only a broad scheme such as that of Green and Neurath [41] will include all the presently likely mechanisms of digestion of proteins. The clear correlation between digestion by lysosoma1enzymes in vitro and turnover of cytosol proteins in vivo indicates that lysosomal enzymes possess some of the characteristics requisite for a system of turnover in vivo. Furthermore, the range of conditions under which these characteristics apply is wide, and in view of recent estimates of intralysosomal pH [42,43], probably includes most conditions which apply within lysosomes. Even with denatured cytosol protein mixture, fragments of high 3H/’4C ratio can still be released preferentially in the early stages of digestion. The only cytosol protein which has been demonstrated within lysosomes is ferritin [44]; thus direct evidence for a major role of lysosomes in turnover of cytosol proteins is very limited. Recent work [45] on the inhibition of intracellular protein turnover by chloroquine provides some indirect evidence for this, and the present work gives information on the characteristics of lysosomal enzymes as agents of turnover. Autophagy (see [10,44]) provides a mechanism for the random uptake of cytosol proteins into lysosomes. Thus, as has been noted previously [3] if the lysosomes are a major agent of their turnover, for the relative degradation rates of cytosol proteins within lysosomes
13
to be reflected in their relative turnover rates, a mode of selective escape from lysosomes of intact proteins (or their subunits) is necessary. There is presently no evidence for this. Alternatively, some lysosomal proteases may be permanently or transiently exposed on the external surface of lysosomes ; or some extra-lysosomal molecules of “lysosomal” proteases may be involved, perhaps on the surface of the endoplasmic reticulum (the “diffuse” cathepsin D distribution demonstrated immunohistochemically in macrophages may be due to such microsomal enzyme [46]). Finally, since the relative turnover rates of the cytosol protein mixture in vivo can be mimicked in vitro not only with liver lysosomal proteases, but also with enzymes of varied specificities from other tissues or species, the possibility remains that liver cytosol protein turnover may be the function of some non-lysosomal liver proteases. In investigating further the role of lysosomes in the process, a direct demonstration of the uptake of cytosol proteins in vivo is required, and the use of specific inhibitors of lysosomal enzymes is also indicated. R.T.D. thanks the Medical Research Council for support
REFERENCES 1. Arias, J. M., Doyle, D. & Schimke, R. T. (1969) J . Biol. Chem. 244, 3303-3315. 2. Dice, J. F., Dehlinger, P. J. & Schimke, R. T. (1973) J . Biol. Chem. 248,4220 -4228. 3. Segal, H. L., Matsuzawa, T., Haider, M. & Abraham, G. J. (1969) Biochem. Biophys. Res. Commun. 36, 764-770. 4. Haider, M. & Segal, H. L. (1972) Arch. Biochem. Biophys. 148, 228- 237. 5. Auricchio, F. & Liquori, A. (1971) FEBS Lett. 12, 329-332. 6. Auricchio, F., Mollica, L. & Liquori, A. (1972) Biochem. J. 129, 1131- 1138. 7. Neely, A. N. & Mortimore, G. E. (1973) Fed. Proc. 32, 299. 8. Mortimore, G. E., Neely, A. N., Cox., J. R. & Guinivan, R. A. (1973) Biochem. Biophys. Res. Commun. 54, 89-95. 9. Poole, B. & Wibo, M. (1973) J. Biol. Chem. 248, 6221 -6226. 10. De Duve, C. & Wattiaux, R. (1966) Annu. Rev. Physiol. 28, 435 - 492. 11. Canonico, P. G. & Bird, J. W. C. (1969) Cytobios, I , 23-31. 12. Ansorge, S., Bohley, P., Kirschke, H., Langner, J. & Hanson, H. (1971) Acta Biol. Med. Germ. 26, 953-962. 13. Neville, D. M. (1971) J. Biol. Chem. 246, 6328-6335. 14. De Duve, C., Pressman, B. C., Gianetto, R., Wattiaux, R., Appelmans, F. (1955) Biochem. J . 60, 604-617. 15. Wattiaux, R.,Wibo, M. & Baudhuin, P. (1963) in CIBA Symposium on Lysosomes (de Reuck, A. V. S. & Cameron, M. P., eds) pp. 176-200, Churchill, London. 16. Beaufay, H. (1972) in Lysosomes: u Laboratory Hundbook (Dingle, J. T., ed.) pp. 46- 135, North Holland, Amsterdam. 17. Dean, R.T. (1975) Biochem. Soc. Trans. 3, 250-252. 18. Barrett, A. J. (1967) Biochern. J. 104, 601-608. 19. Barrett, A. J. (1973) Biochem. J . 131, 809-822. 20. Barrett, A. J. (1972) in Lysosomes: a Loboratory Handbook (Dingle, J. T., ed.) pp. 1-45, North Holland, Amsterdam. 21. Barrett, A. J. (1972) Anal. Biochem. 47, 280-293. 22. Bohley, P., Kirschke, H., Langner, J., Ansorge, S., Wiederanders, B. & Hanson, H. (1971) in Tissue Proteinases (Barrett,
R. T. Dean: Lysosomes and Turnover
14
23.
24. 25. 26. 27. 28.
A. J. & Dingle, J. T., eds) pp. 187-219,North Holland, Amsterdam. Morishima, H., Takita, T., Aoyagi, T., Tdkeuchi, T. & Urnezawa, H. (1970) J. Antibiot. (Tokyo) Ser. A , 23 263265. Barrett, A. J. & Dingle, J. T. (1972) Biochem. J. 127,439-441. Woessner, J. F. (972) Biochem. Biophys. Res. Commun. 47, 965- 970. Shibko, S. &Tappel, A. L. (1965) Biochem. J. 95,731 -741. Goldberg, A. L. & Dice, J. F. (1974) Annu. Rev. Biochem. 43, 835-869. Bohley, P., Miehe, C., Miehe, M., Ansorge, S., Kirschke, H., Langner, J. & Wiederanders, B. (1972) Acta Biol. Med. Germ. 28,323 330. Segal, H. L., Winkler, J. R. & Miyagi, M. P. (1974) J . Biol. Chem. 249,6364- 6365. Taylor, J. M., Dehlinger, P. J., Dice, J. F. & Schimke, R. T. (1973) Drug Metah. Disposition, 1, 84-91. Hayashi, M., Hiroi, Y. & Natori, Y. (1973) Nut. New Biol.242 163- 166. Huisman, W., Lanting, L., Bouma, J. M. W. & Gruber, M. (1974) FEBS Left.45, 129-131. Huisman, W., Bouma, J. M. W. & Gruber, M. (1974) Nature (Lond.) 250,428 - 429. ~
29. 30. 31. 32. 33.
34. Coffey, J. W. & de Duve, C. (1968) J. Bid. Chem. 243, 32553263. 35. Huisman, W., Bouma, J. M. W. & Gruber, M. (1973) Biochim. Biophys. Acra, 297,98- 109. 36. Huisman, W., Lanting, L., Doddema, H. J., Bourna, J. M. W. & Gruber, M. (1974) Biochim. Biophys. Acra, 370, 297-307. 37. Barrett, A. J. (1970) Biochem. J. 117, 601 -607. 38. Johnson, R. W. & Kenney, F. T. (1973) J. Biol. Chem. 248, 4528 -4531. 39. Linderstrom-Lang, K. (1950) Cold Spring Harbour Symp. Quant. Biol. 14, 117-126. 40. Hill, R. L. (1965) Adv. Pro?. Chem. 20, 37 - 107. 41. Green, N. H. & Neurath, H. (1954) in The Proteins B(Neurath, H. & Bailey, K., eds), vol. 2, pp. 1059- 1198, Academic Press, New York. 42. Goldman, R. & Rottenberg, H. (1973) FEBS Lett. 233-238. 43. Reijngoud, D. J. & Tager, J. M. (1973) Biochim. Biophys. A m , 297, 174- 178. 44. Trump, B. F., Vdligosky, J. M., Arstila, A. V., Mergner, W. J. & Kinney, T. D. (1973) Am. J. Pathol. 72, 295-336. 45. Wibo, M. & Poole, B. (1974) J. Cell Biol. 63, 430-440. 46. Dingle, J. T., Poole, A. R., Lazarus, G . S. & Barrett, A. J. (1973) J . Exp. Med. 137, 1124-1141.
R. T. Dean, Department of Experimental Pathology, University College Hospital Medical School, University Street, London, Great Britain WClE 633
Note Added in Proof(September 2, 1975). Direct evidence for a role of lysosomes in turnover of intracellular proteins has now been obtained in experiments with the carboxyl proteinase inhibitor, pepstatin [R. T. Dean (1975) Nature, in press].
Lysosomal Enzymes as Agents of Turnover of Soluble Cytoplasmic Proteins Roger T. DEAN Department of Experimental Pathology, University College Hospital Medical School, London (Received February 5 / April 28, 1975)
The degradation of cytosol proteins in vitro by purified cathepsin D and cathepsin B, and by mixtures of lysosomal enzymes, was studied. By means of a double-labelling method, it was shown that the relative rates of degradation of cytosol proteins by the purified enzymes and by mixtures of enzymes under a wide range of conditions in vitro correlated well with their relative rates of turnover in vivo. The complex mixture of cytosol proteins was degraded less rapidly after denaturation than in the native state, both by the purified proteases and by the mixture of lysosomal enzymes. This contrasts with previous results on proteolysis of single purified proteins. The possible role of lysosoma1 enzymes in turnover in vivo was discussed.
The double-labelling method of Arias etal. [l] distinguishes proteins which turn over rapidly in vivo from those which turn over slowly, by virtue of their isotope ratios : proteins turning over fast show a high ratio of 3H to 14C radioactivity. Using such double-labelled proteins, it has been found that the rates of degradation of cytosol proteins by trypsin and pronase in vitro correlate well with their rates of turnover in vivo [ 2 ] . It has been suggested on the basis of studies on enzyme inactivation in vitro [3 - 61 and proteolysis during liver perfusion [7,8],that turnover of intracellular proteins may be largely a function of the lysosomal system ; although some work [9] indicates the involvement of at least two systems. Clearly, the process of autophagy [lo] provides a mechanism for internalising any cellular component within lysosomes, allowing subsequent digestion. Thus the present work describes the degradation in vitro by lysosomal proteases of double-labelled cytoplasmic proteins, in order to asses further the possible involvement of such proteases in turnover of such proteins in vivo. MATERIALS AND METHODS In the double-labelling procedure El], rats (150 g) were given intraperitoneal injections of 50 pCi [14C]leucine, and, 96 h later, 100 pCi ['Hlleucine. 4 h later Enzymes. Cathepsin D (EC 3.4.23.5); cathepsin B, 3.4.22.1).
(EC
the animals were killed, and livers homogenised (five passes of a tight Dounce) in two volumes of 0.15 M NaCI- 50mM Tris-HC1, pH 7.8, a medium providing good osmotic protection for lysosomes [ll]. The homogenates were centrifuged at 10000 x g for 10 min, and the resulting supernatants centrifuged at 100000 x g for 30 min to give a soluble cytoplasmic fraction. This fraction, will be referred to as cytosol protein mixture but it must be noted that it is not the same as the substrate used by Ansorge etal. 1121, which is a mixture of high-molecular-weight cytosol proteins. Cytosol protein mixture was dialysed for 24 h against 50 vol. of 50 mM Tris-HC1, pH 7.8, with two changes, and the dialysate was used for proteolytic digestions. A portion of cytosol protein mixture was gel-filtered on Sepharose 6B in the same buffer ( V , = 320 ml, V, = 105 ml). Three fractions, each highly heterogeneous, were obtained, by pooling material : fraction 1, emerging in a peak at Vo (molecular weights greater than about 1.5 x lo6); fraction 2, emerging between 175-225 ml (molecular weights 6 x lo52 x lo5; fraction 3, between 225 - 275 ml (molecular weights 2 x lo5- 5 x lo4). Molecular weights (except for fraction l), were estimated from the elution positions of proteins of known molecular weight. The correlation between subunit size and 'H/14C ratio of cytosol protein mixture components [2) was observed in sodium dodecylsulphate - polyacrylamide gel electrophoresis [13]. Soluble mixtures of lysosomal enzymes were obtained by extracting a total lysosomal-mitochondria1
10
fraction [14]; or by a preparation of lysosomes loaded with Triton WR-1339 [15-171; with 0.1 % Triton X-100, and centrifuging at 100000 x g for 60 min, to remove particulate material. In each case livers were extensively perfused with saline before homogenisation to reduce the amount of serum (containing proteinase inhibitors) present. For some experiments, the resuspended particulate material was used as a source of enzymes, and in others the Triton-treated suspensions were used without centrifugation. Purified rabbit liver cathepsin D [18] and human liver cathepsin B, [19] were kindly given by Dr A. J. Barrett. Assays for cathepsin D and cathepsin B, were as described by Barrett [20]: method 1 in each case. The unit of cathepsin B, activity is that amount of enzyme which will hydrolyse 1 pmole of substrate per hour under assay conditions. Proteolytic digestions were conducted at 37 “C for up to 96 h, in stoppered tubes, with the following buffers (100 mM final): pH 2.5, glycine-HC1; pH 3.0 sodium formate; pH 4.3, 5.0 sodium acetate; pH 6.0, 7.0 potassium phosphate; pH 8.0, 9.0 Tris-HCl. Substrate concentration was 10 mg/ml, 14C 31 000 dis. x min-’ x ml-’, 3H 68000 dis. x min-’ x ml-’, and the amounts of the purified enzymes used were: cathepsin D : 10 U/ml; cathepsin B, : 1.4 U/ml(l mM dithiothreitol added [21]). 10 mg of lysosomal proteins was present in incubations with crude lysosomal components, while 1 mg of lysosomal protein was used in those with tritosomal components. Incubations were usually in a total volume of 1 ml, and samples were removed at various times for determination (by scintillation counting) of radioactivity soluble in 5 % trichloroacetic acid. There was no loss of cathepsin D activity during incubation for up to 96 h, while cathepsin B1 fell to 70% of the original activity by 96 h. In certain experiments in which “intact” lysosomes were required, short incubations (up to 1 h) the presence of 0.25 M sucrose (for osmotic protection) were used, to minimise disruption of lysosomes. When suspensions of broken lysosomes or extracts thereof were used, Triton X-100 was maintained at 0.1 % to ensure disruption. RESULTS General Characteristics of the Degradation of Cytosol Proteins by Lysosomal Enzymes
The release of trichloroacetic-acid-soluble radioactivity from the cytosol protein mixture was never linear with time for more than 30 min. Thus, in order to assess the pH-dependence of the digestions, incubations at various pH values were compared after 1 h, 4 h and 24 h. pH optima observed were not dependent on length of incubation or on the isotope observed. Autodigestion of the cytosol protein mix-
Lysosomes and Turnover
w Fig. 1. p H dependence of hydrolysis of cytosol protein mixture by carhepin B, rrnd D. Dialysed cytosol protein mixture (10mg/ml) W d S incubated at the specified pH values with either cathepsin B, (1.4 U/ml; A) or cathepsin D (11 U/ml; 0)and trichlorodceticacid-soluble radioactivity measured at 4 h
ture, in incubations without added lysosomal enzymes, was very slow (at pH 3 or 5, always less than 10% of that by 11 U/ml of cathepsin D), as reported previously [22]. The pH optimum for autoproteolysis was 5.0, and there was slight activation by 1 mM dithiothreitol, which was most pronounced at pH 5.0 (in a 24-h incubation, pH 5.0: 25 %). Cathepsin D gave fastest digestion of the cytosol protein mixture at pH 5, with considerable activity at pH 3 (Fig. 1). No activity was detected at pH 7.0, allowing for the slight auto-proteolysis of the cytosol protein mixture at this pH. Furthermore, the very slow release of trichloroacetic-acid-soluble radioactivity during incubations at pH 7.0, was not affected by the presence of pepstatin (an inhibitor of cathepsin D and other acid proteases [23-251 at 10 pg/ml, either in the presence or absence of rabbit cathepsin D. At pH 5.0, pepstatin (10 pg/ml) completely abolished digestion of the cytosol protein mixture by cathepsin D, while dithiothreitol was also inhibitory (up to 20% inhibition by 10 mM dithiothreitol in 3-h incubations). Liberation of trichloroacetic-acid-soluble radioactivity at pH 7.0 by cytosol peptidases from the products of a brief cathepsin D digestion (2 h; stopped by the addition of pepstatin) was not significantly more rapid than from undigested cytosol protein. Thus effects of such cytosol peptidases were small, in agreement with [12]. Cathepsin B, showed a pH optimum of 4.3, with significant activity between pH 3 and 7 (Fig. 1). Release of radioactivity in 12 h by 1.4 U/ml of cathepsin B1 (pH 4.3) was 60% of total, whereas that by 10 U of cathepsin D was only 45%. This difference in rates of hydrolysis was apparent between 2 h and 24 h (Fig. 2). The optimum for digestion by extracts of both crude and purified lysosomes (see Methods) was pH 3, with activity detectable between pH 2.5 and pH 9. The
R. T. Dean
11
Time of digestion ( h )
Fig. 2. Hydrolysis oj’native and denatured cytosolprotein mixture by carhepsin B , , cathepsin D and a tritosome extract. Cytosol protein mixture (dialysed) (10 mg/ml) was incubated at the specified pH with eithercathepsinB, (1.4 U/ml)orcathepsinD(ll U/ml)oratritosome extract (1 mg/ml protein) and trichloroacetic-acid-soluble radioactivity measured at various times. Symbols: (0)cathepsin B,, at pH 4.3 with native cytosol protein mixture; (A) cathepsin B,, at pH 4.3 with denatured cytosol protein mixture; (0)cathepsin D at pH 5.0 with native cytosol protein mixture; (A) cathepsin D at pH 3.0 with denatured cytosol protein mixture; (0)crude lysosomal extract at pH 3.0, 0.1 % Triton X-100, with native cytosol protein. Cytosol protein was denatured by treatment at pH 1.5 (see text)
Table 1 . Eflect of dithiorhreitol on digestion of the cylosol protein mixture by a tritosomal extract, at various p H values. Incubations were as described in legend to Fig.2, for 24 h either without dithiothreitol (controls) or with 1 mM dithiothreitol (tests). It should be noted that the “activations” observed may represent stimulation or stabilisation of enzyme activities, amongst other possibilities. The results expressed are the difference between digestion in control and test incubations as a percentage of the control value (measured as liberation of trichloroacetic-acid-soluble radioactivity)
PH
Effect of dithiothreitol
that obtained with lysosomes disrupted by 0.1 % Triton X-100 before incubation. Very low rates of digestion (less than 0.5% of total radioactivity released in 1 h) were observed in incubations at pH 5 and pH 7. Thus there was no satisfactory evidence for the digestion of cytosol-protein by intact lysosomes, as these low rates of digestion can probably be explained by lysosoma1 breakage. Experiments comparing hydrolysis at pH 5.0 by cathepsins D and B1 with that by mixtures of the two enzymes (both at the usual concentrations) gave no evidence for synergism: after 2, 6, 24 and 48 h of digestion release by the mixture of enzymes was equal to the sum of the releases by the enzymes separately. Because it has been noted that some proteins are hydrolysed by proteases more rapidly after denaturation in the native state (see for instance [27]), the generality of this was assessed using the cytosol protein mixture as substrate. The cytosol protein mixture was denatured by two means, by boiling for 15 min at pH 7.0 or by titrating to pH 1.5 with HCl, and after 48 h a t 4 “C,retitrating to pH 7.0. While the pH optima for digestion of denatured cytosol protein mixture by cathepsin B, and a tritosomal extract rcinained pH 4.3 and pH 3.0, respectively, that for cathepsin D was changed to pH 3.0. When digestions of native and denatured cytosol protein mixture by cathepsins B, and D and by tritosomal extracts (under optimal conditions of pH) were compared (Fig. 2) it was found that in each case, degradation of the denatured proteins was the slower. Whereas cathepsin B, digested native cytosol protein mixture substantially more rapidly than cathepsin D, the two enzymes degraded denatured cytosol protein mixture at similar rates.
~
% control 3 5 6 7 9
- 25
+ 35 + 23 + 22
+ 10
time-course of digestion by a lysosomal extract is shown in Fig.2. Activation by 1 mM dithiothreitol was detectable between pH 5 and pH 9 (Table 1). Similar pH dependences were observed for digestion by resuspended particulate fractions from the crude lysosome preparations: activity at pH 3 (2-h incubations) of the soluble fraction was 160% of that of the particulate fraction (with the crude preparations much cathepsin B1 was solubilised by Triton X-100, but this was not true of tritosomes [17]). In 30-min incubations with osmotic protection, “intact” crude lysosomes were again most active at pH 3.0: however lysosome rupture at low pH is rapid [26], and the release of radioactivity was similar to
Correlation of Digestion in vitro with Turnover in vivo
The susceptibility of native cytosol proteins to trypsin and pronase at pH 7.0 correlates well with their turnover rates in vivo [2] : thus trichloroaceticacid-soluble fragments released initially have high 3H/14Cratios, and later fragments have lower ratios. This correlation was initially only sought in experiments at neutral pH with commercially available enzymes. In the present work intrinsic liver lysosomal enzymes have been used under a wider variety of conditions. The correlation was clearly demonstrated with cathepsin D at pH 3, pH 5 (Fig. 3 ) and pH 6, and with cathepsin B, at pH 4.3, pH 5 , and pH 6 . With lysed suspensions of tritosomes or of crude lysosomes, the correlation was observed between pH 3 and 8. This characteristic of digestion with extracts of lysosomes was first pointed out by Bohley et al. [28] and recently confirmed [29]. Even in digestions with denatured cytosol protein mixture, an initial release of high
12
Lysosomes and Turnover
’6 0 ._
2!
I
I
0.4;
€0
I
I
240
120 180 Time of digestion (min)
Fig. 3 . Hydrolysis of’native cytosol proteins by cathepsin D. Cytosol protein mixture (10 mg/ml) was incubated at pH 5.0 with cathepsin D (1 1 Ujrnl) and trichloroacetic-acid-soluble 3H and 14C radioactivity determined at _various times, and their ratios calculated. The isotope ratio of the cytosol protein used, was 0.55 (arbitrary units)
0” 0
I
4
20
40 Time of digestion (min)
I
100
Fig. 4. Susceptibility of cytosol proteins of various size ranges to cathepsin D.Hydrolyses were performed at pH 3.0 with 11 U/ml of cathepsin D. Substrate concentration in each case was adjusted to 1 mg/ml protein. Symbols: (0) large proteins (see Methods); (0)intermediate proteins; (A) small proteins
ratio material by cathepsin D was observed, as noted previously with trypsin [30]. The digestibility of molecules of different size ranges was also compared using fractions of the cytosol protein mixture obtained by gel filtration, each at 1 mg/ml. The correlation of digestibility with molecular size was very clear (Fig.4). As mentioned above, in experiments with “intact” lysosomes, no evidence was obtained for digestion in vitro by intact lysosomes. It has been claimed that this process occurs [31], and that proteins can be internalised by lysosomes in vitro, although this has been
strongly disputed [32,33]. Since the published experiments have involved crude lysosomal fraction, and have merely measured co-sedimentability of radioactive protein with lysosomes, I have attempted to demonstrate uptake of proteins into the soluble phase of the purified lysosomes i. e. to sites from which they can be released by rupturing lysosomes with 0.1 ”/, Triton X-100. Thus after incubation of lysosomes with cytosol protein mixture and also with highly labelled serum, and repeated washing of lysosomes, the amount of non-sedimentable radioactivity released by Triton X-100-0.25 M sucrose was compared with that released by sucrose alone. Numerous controls for release from membrane-bound sites, including those generated during the incubations (30 min at 37 “C or 4 “C) were performed. No evidence for uptake was obtained either in presence or absence of ATP, in agreement with results of Huisman et al. [32,33].
DISCUSSION The degradation of several purified proteins by extracts of lysosomes has been studied previously (see for instance [34- 36]), and some correlation between susceptibility to denaturation in vitro and susceptibility to lysosomal enzymes noted [34]. In addition a few cytosol enzymes have been used as experimental substrates for lysosomal enzymes (see review in [22]), while the use of mixed cytosol proteins as substrates has been pioneered by Bohley et al. [12,22]. The degradition of mixed cytosol proteins by highly purified lysosomal enzymes has however, received little attention. In the present work the digestion of this substrate by cathepsins D and B1 has been outlined. Whereas it has been previously claimed that purified chicken cathepsin D [37] does digest the cytosol protein substrate of Ansorge et al. [12] at pH 6.9, this was not observed here with rabbit cathepsin D acting on cytosol protein mixture. The optimum of digestion by cathepsin B1 was 4.3, which is lower than its optimum (pH 6 ) against synthetic substrates and azo-haemoglobin [19]. The relative contribution of the various lysosomal proteases to the digestion of cytosol proteins remains to be established, although the efficiency of digestion by cathepsin B, and the stimulation of proteolysis by dithiothreitol suggests that thiol proteinases may well be important. Whereas previous work suggests that proteins are more susceptible to proteolysis after denaturation than in the native state [27], this was not observed in the present work. The pH optimum of cathepsin D versus denatured cytosol protein mixture was lowered to pH 3, but those of cathepsin B, and a tritosomal extract were as for native cytosol protein mixture. Even when di-
R. T. Dean
gestions were conducted at the relevant pH optimum, the denatured cytosol protein mixture was not more susceptible to either the purified or the crude mixture of lysosomal enzymes. A crucial difference between most published experiments and those reported now is in the complexity of the substrate used. Previous work has considered the digestion of a limited number of purified proteins, used singly; in contrast, the present work concerns proteolysis of a complex mixture of proteins, and is thus a more general test of the idea that denatured proteins are more susceptible to digestion than are native proteins. It is notable that in a previous experiment using such a complex mixture of proteins, it was found that the denatured proteins were less susceptible to trypsin than the native mixture [12], in agreement with the present results. Thus the idea that denatured proteins are more susceptible to hydrolysis by lysosomal or other proteases [27,34] is not generally applicable. Indeed, it has been found that although molecules of tyrosine aminotransferase containing certain amino acid analogues are much more easily denatured than normal enzyme, the turnover rates in vivo of modified and normal enzyme are the same [38]. Although Linderstrom-Lang (see [39]) has suggested that initial proteolytic attack catalyses subsequent rapid denaturation of the substrate, there is evidence against this in the case of subtilisin acting on human haemoglobin (see [40]). Thus only a broad scheme such as that of Green and Neurath [41] will include all the presently likely mechanisms of digestion of proteins. The clear correlation between digestion by lysosoma1enzymes in vitro and turnover of cytosol proteins in vivo indicates that lysosomal enzymes possess some of the characteristics requisite for a system of turnover in vivo. Furthermore, the range of conditions under which these characteristics apply is wide, and in view of recent estimates of intralysosomal pH [42,43], probably includes most conditions which apply within lysosomes. Even with denatured cytosol protein mixture, fragments of high 3H/’4C ratio can still be released preferentially in the early stages of digestion. The only cytosol protein which has been demonstrated within lysosomes is ferritin [44]; thus direct evidence for a major role of lysosomes in turnover of cytosol proteins is very limited. Recent work [45] on the inhibition of intracellular protein turnover by chloroquine provides some indirect evidence for this, and the present work gives information on the characteristics of lysosomal enzymes as agents of turnover. Autophagy (see [10,44]) provides a mechanism for the random uptake of cytosol proteins into lysosomes. Thus, as has been noted previously [3] if the lysosomes are a major agent of their turnover, for the relative degradation rates of cytosol proteins within lysosomes
13
to be reflected in their relative turnover rates, a mode of selective escape from lysosomes of intact proteins (or their subunits) is necessary. There is presently no evidence for this. Alternatively, some lysosomal proteases may be permanently or transiently exposed on the external surface of lysosomes ; or some extra-lysosomal molecules of “lysosomal” proteases may be involved, perhaps on the surface of the endoplasmic reticulum (the “diffuse” cathepsin D distribution demonstrated immunohistochemically in macrophages may be due to such microsomal enzyme [46]). Finally, since the relative turnover rates of the cytosol protein mixture in vivo can be mimicked in vitro not only with liver lysosomal proteases, but also with enzymes of varied specificities from other tissues or species, the possibility remains that liver cytosol protein turnover may be the function of some non-lysosomal liver proteases. In investigating further the role of lysosomes in the process, a direct demonstration of the uptake of cytosol proteins in vivo is required, and the use of specific inhibitors of lysosomal enzymes is also indicated. R.T.D. thanks the Medical Research Council for support
REFERENCES 1. Arias, J. M., Doyle, D. & Schimke, R. T. (1969) J . Biol. Chem. 244, 3303-3315. 2. Dice, J. F., Dehlinger, P. J. & Schimke, R. T. (1973) J . Biol. Chem. 248,4220 -4228. 3. Segal, H. L., Matsuzawa, T., Haider, M. & Abraham, G. J. (1969) Biochem. Biophys. Res. Commun. 36, 764-770. 4. Haider, M. & Segal, H. L. (1972) Arch. Biochem. Biophys. 148, 228- 237. 5. Auricchio, F. & Liquori, A. (1971) FEBS Lett. 12, 329-332. 6. Auricchio, F., Mollica, L. & Liquori, A. (1972) Biochem. J. 129, 1131- 1138. 7. Neely, A. N. & Mortimore, G. E. (1973) Fed. Proc. 32, 299. 8. Mortimore, G. E., Neely, A. N., Cox., J. R. & Guinivan, R. A. (1973) Biochem. Biophys. Res. Commun. 54, 89-95. 9. Poole, B. & Wibo, M. (1973) J. Biol. Chem. 248, 6221 -6226. 10. De Duve, C. & Wattiaux, R. (1966) Annu. Rev. Physiol. 28, 435 - 492. 11. Canonico, P. G. & Bird, J. W. C. (1969) Cytobios, I , 23-31. 12. Ansorge, S., Bohley, P., Kirschke, H., Langner, J. & Hanson, H. (1971) Acta Biol. Med. Germ. 26, 953-962. 13. Neville, D. M. (1971) J. Biol. Chem. 246, 6328-6335. 14. De Duve, C., Pressman, B. C., Gianetto, R., Wattiaux, R., Appelmans, F. (1955) Biochem. J . 60, 604-617. 15. Wattiaux, R.,Wibo, M. & Baudhuin, P. (1963) in CIBA Symposium on Lysosomes (de Reuck, A. V. S. & Cameron, M. P., eds) pp. 176-200, Churchill, London. 16. Beaufay, H. (1972) in Lysosomes: u Laboratory Hundbook (Dingle, J. T., ed.) pp. 46- 135, North Holland, Amsterdam. 17. Dean, R.T. (1975) Biochem. Soc. Trans. 3, 250-252. 18. Barrett, A. J. (1967) Biochern. J. 104, 601-608. 19. Barrett, A. J. (1973) Biochem. J . 131, 809-822. 20. Barrett, A. J. (1972) in Lysosomes: a Loboratory Handbook (Dingle, J. T., ed.) pp. 1-45, North Holland, Amsterdam. 21. Barrett, A. J. (1972) Anal. Biochem. 47, 280-293. 22. Bohley, P., Kirschke, H., Langner, J., Ansorge, S., Wiederanders, B. & Hanson, H. (1971) in Tissue Proteinases (Barrett,
R. T. Dean: Lysosomes and Turnover
14
23.
24. 25. 26. 27. 28.
A. J. & Dingle, J. T., eds) pp. 187-219,North Holland, Amsterdam. Morishima, H., Takita, T., Aoyagi, T., Tdkeuchi, T. & Urnezawa, H. (1970) J. Antibiot. (Tokyo) Ser. A , 23 263265. Barrett, A. J. & Dingle, J. T. (1972) Biochem. J. 127,439-441. Woessner, J. F. (972) Biochem. Biophys. Res. Commun. 47, 965- 970. Shibko, S. &Tappel, A. L. (1965) Biochem. J. 95,731 -741. Goldberg, A. L. & Dice, J. F. (1974) Annu. Rev. Biochem. 43, 835-869. Bohley, P., Miehe, C., Miehe, M., Ansorge, S., Kirschke, H., Langner, J. & Wiederanders, B. (1972) Acta Biol. Med. Germ. 28,323 330. Segal, H. L., Winkler, J. R. & Miyagi, M. P. (1974) J . Biol. Chem. 249,6364- 6365. Taylor, J. M., Dehlinger, P. J., Dice, J. F. & Schimke, R. T. (1973) Drug Metah. Disposition, 1, 84-91. Hayashi, M., Hiroi, Y. & Natori, Y. (1973) Nut. New Biol.242 163- 166. Huisman, W., Lanting, L., Bouma, J. M. W. & Gruber, M. (1974) FEBS Left.45, 129-131. Huisman, W., Bouma, J. M. W. & Gruber, M. (1974) Nature (Lond.) 250,428 - 429. ~
29. 30. 31. 32. 33.
34. Coffey, J. W. & de Duve, C. (1968) J. Bid. Chem. 243, 32553263. 35. Huisman, W., Bouma, J. M. W. & Gruber, M. (1973) Biochim. Biophys. Acra, 297,98- 109. 36. Huisman, W., Lanting, L., Doddema, H. J., Bourna, J. M. W. & Gruber, M. (1974) Biochim. Biophys. Acra, 370, 297-307. 37. Barrett, A. J. (1970) Biochem. J. 117, 601 -607. 38. Johnson, R. W. & Kenney, F. T. (1973) J. Biol. Chem. 248, 4528 -4531. 39. Linderstrom-Lang, K. (1950) Cold Spring Harbour Symp. Quant. Biol. 14, 117-126. 40. Hill, R. L. (1965) Adv. Pro?. Chem. 20, 37 - 107. 41. Green, N. H. & Neurath, H. (1954) in The Proteins B(Neurath, H. & Bailey, K., eds), vol. 2, pp. 1059- 1198, Academic Press, New York. 42. Goldman, R. & Rottenberg, H. (1973) FEBS Lett. 233-238. 43. Reijngoud, D. J. & Tager, J. M. (1973) Biochim. Biophys. A m , 297, 174- 178. 44. Trump, B. F., Vdligosky, J. M., Arstila, A. V., Mergner, W. J. & Kinney, T. D. (1973) Am. J. Pathol. 72, 295-336. 45. Wibo, M. & Poole, B. (1974) J. Cell Biol. 63, 430-440. 46. Dingle, J. T., Poole, A. R., Lazarus, G . S. & Barrett, A. J. (1973) J . Exp. Med. 137, 1124-1141.
R. T. Dean, Department of Experimental Pathology, University College Hospital Medical School, University Street, London, Great Britain WClE 633
Note Added in Proof(September 2, 1975). Direct evidence for a role of lysosomes in turnover of intracellular proteins has now been obtained in experiments with the carboxyl proteinase inhibitor, pepstatin [R. T. Dean (1975) Nature, in press].
