Selective Degradation of Cytosolic Proteins by Lysosomesn J. FRED DICE Departmmt of Physiology T@s University School of Mditirzc 136 Harrison Amue Boston, Massachusetts 02111
INTRODUCTION Both cytosolic and lysosomal pathways of proteolysis operate in most cells.lJ In well-nourished cells lysosomes appear to be able to internalize proteins by a process called microautophagy (FIG. 1) in which the lysosomal membrane invaginates at multiple locations. Microautophagy appears to be nonselective in that several proteins and inert particles are internalized at similar r a t e ~ . ~ . ~ When cultured cells reach confluence an additional lysosomal pathway of proteoly~.~ ( FIG. 1)involves formation ofautophagic vacuoles sis is s t i m ~ l a t e d .Macroautophagy which sequester areas of cytoplasm.’J These vacuoles then acquire lysosomal hydrolases to form autophagosomes. Macroautophagy also appears to be nonselective in that many different organelles and proteins are sequestered at approximately the same We have studied an additional pathway of lysosomal proteolysis that is activated in confluent cell monolayers in response to serum withdrawal. This pathway is restricted to cytosolic proteins that contain peptide sequences biochemically related to Lys-Phe-Glu-Arg-Gln(KFERQ). The mechanism by which proteins with KFERQ-like peptide regions are targeted to lysosomes for degradation is similar in certain respects to the direct import of newly synthesized proteins into mitochondria or into the lumen of the endoplasmic reticulum. Therefore, proteins with KFERQ-like peptide regions may enter lysosomes by directly crossing a membrane bilayer (FIG. 1).
Ribonudease A as a Probe for Selective Lposomal Proteolpis We used red cell-mediated microinjection to introduce specific radiolabeled proteins into the cytosol of confluent cultures of human fibroblasts.l*J Radiolabeled ribonuclease A (RNase A) is degraded with a half-life of approximately 100 h in serumsupplemented cells and 50 h in serum-deprived cells. The half-lives of certain other microinjected proteins, but not others, are reduced in response to serum withdrawal
“Research in the author’s laboratory is supported by National Institutes of Health Grants AGO61 16 and AG07472. 58
DICE: SELECTIVE DEGRADATION OF PROTEINS BY LYSOSOMES
59
MICROAUTOPHAGY
FIGURE 1. Schematic representations of pathways of lysosomal proteolysis. See the text for descriptions of the pathways.
MACRoAUTOPHAGY
DIRECT PROTEIN TRANSFER
0
(TABLE 1). These data suggest that the enhanced degradation during serum withdrawal is a selective process. Several lines of evidence indicate that microinjected RNase A is degraded within lysosomes: (a) a small amount of microinjected RNase A fractionates with lysosomes,” ’ ~ degrada(b) degradation of RNase A is partially inhibited by ammonium ~ h l o r i d e , (c) , ~ ~ (d) the tion products of [3H]raffinose-RNaseA accumulate only in l y s o ~ o m e s and same peptides derived from RNase A are released from cells after microinjection and after lysosomal hydrolysis following endocytosis. l 5 The amino-terminal 20 amino acids of RNase A (RNase S-peptide) are required for the enhanced degradation in response to serum withdrawal,’ and covalent attachment of this peptide to heterologous proteins also causes their degradation rates to increase in response to serum withdrawal.16 We eventually identified amino acids 7- 11 within RNase S-peptide (KFERQ) as the region essential for enhanced degradation.” Generality of This Pathway of Protein Degradation To examine whether peptide regions similar to KFERQ exist in intracellular proteins, we raised polyclonal antibodies to KFERQ and affinity purified IgGs specifically TABLE 1.
Degradation of Long-Lived Proteins Microinjected into Human Cell9 T1/2 ( h )
Protein Group 1: Regulated Degradation RNase A RNase S-peptide Aspartate amino-transferase Pyruvate kinase Hemoglobin Group 2: Nonregulated Degradation RNase S-Protein Ovalbumin Lysozyme Insulin A-chain P-galactosidase ~~
Species
2 Serum
- Serum
cow cow Pig chicken human
100 62 80 296 213
50 30 40 121 106
cow chicken chicken cow E. coli
100 55 59 104 1756
100 49 59 104 1 75b
~
‘References to half-life values have been previously cited.” IMR-90 human diploid fibroblasts were recipient cells in all cases except for pyruvate kinase which was injected into HeLa cells. 6L. J. Terlecky and J. F. Dice, unpublished results.
ANNALS NEW YORK ACADEMY OF SCIENCES
60
-x
L
I
0
w I-
n
35
30
-
25
-
0
50
0
anti-KFERQ
0
anti-RYLPT
0
prsimmune
100
150
FIGURE 2. Immunoprecipitation of radiolabeled cytosolic proteins from fibroblasts. Cells were labeled for 2 d with [3H]leucine and cytosolic proteins prepared as described elsewhere." Mnity-purified polyclonal antibodies were used for immunoprecipitations.
200
250
ANTIBODY IpI)
directed toward the pentapeptide. l 8 The anti-KFERQ IgGs immunoprecipitated approximately 30% of [3H]leucine-labeled cytosolic proteins from human fibroblastszz (FIG.2). In pulse-chase experiments, the immunoprecipitable proteins were preferentially degraded in response to serum withdrawalla (FIG.3). In contrast, nonimmunoprecipitable proteins were degraded at the same rate in the presence and absence of serum.18
The Peptide Motif We compared amino acid sequences of four microinjected long-lived proteins whose degradation is enhanced during serum deprivation with four microinjected long lived proteins whose degradation is unaffected by serum. A common sequence motif exists in the first group of proteins but not in the second group. 19.z0 The motif appears to be a pentapeptide with a Gln at either the carboxyl or amino terminus. The pentapeptide must also contain an acidic (Glu o r Asp), a basic (Lys or Arg), and a highly hydrophobic amino acid (Phe, Leu, Ile, or Val). The fifth amino acid of the pentapeptide can be either a repeated basic or hydrophobic amino acid. With the exception of Gln, the relative positions of the amino acids do not appear to be important, but they must be contiguous. While this motif is highly redundant, of the 3.2 million possible pentapeptides, only 2304 fit the motif, that is, less than 0.1%.
FIGURE 3. Rapid degradation of immunoprecipitable proteins in response to serum withdrawal. Cells were radiolabeled for 2 d and then chased for 0, 1, or 2 d in unlabeled medium containing 10% calf serum ( + S) o r not ( - S). Cells were harvested, and radioactivity remaining in antiWERQ immunoprecipitable proteins followed. I s 2.5
TIME Idayrl
;':I
DICE: SELECTIVE DEGRADATION OF PROTEINS BY LYSOSOMES
FIGURE 4. Uptake and degradation of [3H]RNase S-peptide by isolated lysosomes. Lysosomes from 2 x lo6 cells were incubated with radiolabeled R!!ase S-peptide for 2 h at room temperature.*' Other additions were ATP and an ATP regenerating system (10 mM), Prp73 (10 pg/ml), and NH4CI (10 mM). Radioactivity converted to acid-soluble material is shown.
61
m
3
A
s
-
4
e 2 0
a
0 NONE
ATP PRP73
ATP
ATP
PRP73
PRP73
' 0
C.
NH,CI
Selective Degradation ofProteins by Isolated Lysosomes To try to elucidate the mechanism of degradation of KFERQ motif-containing proteins, we developed an in v h o assay using lysosomes isolated from IMR-90 fibroblasts over two consecutive discontinuous density gradientsZL(S. R. Terlecky and J. F. Dice, unpublished). Maximal degradation of [3H]RNase S-peptide by isolated lysosomes requires both ATP and a heat shock protein of 70 kDa (hsp70) (FIG. 4). Similar results have been obtained using native [jH]RNase A. This degradation is specific because [3H]RNase S-protein, amino acids 21-124 of RNase A which does not contain a KFERQ motif, is degraded little, if at all, under the same conditions. In addition, degradation of [jH]RNase S-peptide can be inhibited by reducing the temperature, and degradation appears to occur within acidified organelles because it is inhibited by ammonium chloride (FIG. 4). Additional experiments show that the lysosomal uptake of [jH]RNase S-peptide is saturable. Furthermore, at 0-4 "C. specific binding of [jH]RNase S-peptide to a lysosomal membrane protein occurs. Presumably, this binding component is a receptor or a peptide transport channel.
The Role ofan HSP70 in Selective Lysosomal Proteolysis We considered the possibility that an intracellular protein may recognize KFERQlike peptide regions in proteins to be degraded more rapidly in response to serum withdrawal. A protein of 73 kDa was purified on an RNase S-peptide affinity column and was designated prp73 for peptide recognition protein of 73 A monoclonal antibody that recognizes all of the members of the hsp7O familyz2reacts with prp73 purified from human fibroblast cytosol. In addition, sequence data obtained from purified prp73 tentatively identified it as the constitutively expressed heat shock cognate protein of 73 kilodaltons ( h s ~ 7 3 ) .A~ 'yeast analog of hsc73 is required for transport of proteins into mitochondria and the endoplasmic r e t i c u l ~ m . ~ ~ ~ ~ ~ Hsc73 and prp73 both bind to RNase A, RNase S-peptide, KFERQ, aspartate aminotransferase, and pyruvate kinase-proteins and peptides that contain KFERQ-like peptide motifs. Neither hsc73 nor prp73 bind to ovalbumin, lysozyme, or ubiquitinproteins that lack KFERQ-like peptide motifs.25 Finally, hsc73 and prp73 function identically in stimulating degradation of RNase S-peptide by isolated lysosomes under conditions where three other members of the hsp70 family have no Therefore, one of the hnctions of hsc73 during nutritional deprivation appears to be to recognize and target specific cytosolic proteins to lysosomes for degradation.
62
ANNALS NEW YORK ACADEMY OF SCIENCES
The mechanisms by which hsc73 promotes lysosomal degradation of proteins containing KFERQ-like pepride regions are not known. In response to serum withdrawal hsc73 may shift its subcellular distribution to the cytosol where the proteins containing the KFERQ-like regions reside.*l Activation of the pathway could also involve changes in the substrate proteins that expose the KFERQ-like regions. Finally, activation of the pathway may be due to regulation of a component distal to hsc73 binding to the protein substrate. For example, the activity of the putative receptor or peptide transporter on the lysosome surface may be regulated.
Possible Relevance of This Proteolytc Pathway to Alzheimer’s Disease It seems unlikely that the amyloid precursor protein (APP) is targeted for lysosomal proteolysis through the KFERQ-selectivepathway. This pathway applies primarily to cytosolic proteins while APP is a transmembrane protein. The APP is more likely to be exposed to lysosomal hydrolases through endocytic and recycling pathways that apply to many cell surface proteins. However, if APP (or peptides derived from APP) ever appear in the cytosol, lysosomal uptake and degradation is possible by both nonselective and selective pathways. The release of peptide fragments from proteins that are degraded within lysosomes after microinjection into the cytosol or after e n d o c y t o s i ~may ~ ~ reflect how the fiamyloid fragment is released from cells. Transfer of these peptides from the lysosome to the medium may require a single transport step, or, alternatively, may require transfer of the peptides to the cytosol or other organelles.15
SUMMARY Lysosomes are able to internalize cellular proteins in a variety ofways. One pathway is selective for cytosolic proteins containing peptide sequences biochemically related to Lys-Phe-Glu-Arg-Gln(KFERQ). This pathway is activated in confluent monolayers of cultured cells in response to deprivation of serum growth factors and applies to approximately 30% of cytosolic proteins. We have reconstituted this lysosomal degradation pathway in Pitm. ,Uptake and/or degradation is stimulated by ATP and a member of the heat shock 70-kilodalton protein family, the 73-kiloddton constitutive heat shock protein. Several possible mechanisms of selective protein transport into lysosomes and the possible relevance of this proteolytic pathway to the processing of the amyloid precursor protein are discussed.
ACKNOWLEDGMENTS I thank former and present members of my laboratory who have been responsible for much of the work presented. REPERENCES 1. DICE,J. F. 1987. Molecular determinants of protein half-lives in eukaryotic cells. FASEB J. 1: 349-357. & J. F. DICE. 1992. Pathways of intracellular protein 2. OLSON,T. S., S. R. TERLECICY degradation in eukaryotic cells. In Stability of Protein Pharmaceuticals:In Vim Pathways
DICE: SELECTIVE DEGRADATION OF PROTEINS BY LYSOSOMES
3. 4.
5. 6. 7. 8. 9. 10. 11.
12. 13.
14.
15. 16. 17. 18.
19. 20. 21. 22.
63
of Degradation and Strategies for Protein Stabilization. T. J. Ahern & M. C. Manning, Eds. In press. Plenum Publishing Corporation. New York. AHLBERG,J., L. MAFUELLA& H . GWUMANN.1982. Uptake and degradation of proteins by isolated rat liver lysosomes. Suggestion of a microautophagic pathway of proteolysis. Lab. Invest. 47: 523-532. MAFUELLA,L. & H. GWUMANN.1987. Autophagy, microautophagy, and crinophagy as mechanisms for protein degradation. In Lysosomes: Their Role in Protein Breakdown. H. Glaumann & F. J. Ballard, Eds. 319-367. Academic Press. New York. COCKLE,S. M. & R. T. DEAN.1982. The regulation of proteolysis in normal fibroblasts as they approach confluence. Evidence for participation of the lysosomal system. Biochem. J. 208: 243-249. KNECHT,E., J. HERNANDEZ-YAGO & S. GRISOLIA.1984. Regulation of lysosomal autophagy in transformed and nontransformed mouse fibroblasts under several growrh conditions. Exp. Cell Res. 154: 224-232. MORTIMORE, G. E. 1987. Mechanism and regulation of induced and basal protein degradation in liver. In Lysosomes: Their Role in Protein Breakdown. H . Glaumann & F. J. Ballard, Eds. 415-444. Academic Press. New York. PFEIFER,U. 1987. Functional morphology of the lysosomal apparatus. In Lysosomes: Their Role in Protein Breakdown H . Glaumann & F. J. Ballard, Eds. 3-59. Academic Press. New York. KOMINAMI,E., S. HASHIDA, E. A. KHAIWLAH & N. KATUNUMA. 1983. Sequestration of cytoplasmic enzymes in an autophagic vacuole-lysosomal system induced by injection of leupeptin. J. Biol. Chem. 258: 6093-6100. HENDIL,K. 1981. Autophagy of metabolically inert substances injected into fibroblasts in culture. Exp. Cell Res. 135: 157-166. KOPITZ,J., G. 0. KISEN, P. B. GORDON,P. BOHLEY& P. 0.SEGLEN.1990. Non-selective autophagy ofcytosolic enzymes in isolated rat hepatocytes. J. Cell Biol. 111: 941-954. NEFF,N. T., L. BOURRET, P. MIAO& J. F. DICE.1982. Degradation ofproteins microinjected into IMR-90 human diploid fibroblasts. J. Cell Biol. 91: 184-194. MCELLIGOTT, M. A. & J. F. DICE. 1984. Microinjection of cultured cells using red cellmediated fusion and osmotic lysis of pinosomes: a review of methods and applications. Biosci. Rep. 4: 451-466. MCELLIGOTT,M. A,, P. MIAO& J. F. DICE.1985. Lysosomal degradation of ribonuclease A and ribonuclease S-protein microinjected into human fibroblasts. J. Biol. Chem. 260: 11986-11993. ISENMAN,L. D. & J. F. DICE. 1989. Secretion of intact proteins and peptide fragments by lysosomal pathways of protein degradation. J. Biol. Chem. 264: 21591-21596. BACKER,J. M. & J. F. DICE.1986. Covalent linkage of ribonuclease S-peptide to microinjected proteins causes their intracellular degradation to be enhanced during serum withdrawal. Proc. Natl. Acad. Sci. USA 83: 5830-5834. DICE,J. F., H.-L. CHIANG,E. P. SPENSER& J. M. BACKER.1986. Regulation ofcatabolism of microinjected ribonuclease A: identification of residues 7- 11 as the essential pentapep tide. J. Biol. Chem. 262: 6853-6859. CHIANG,H.-L. & J. F. DICE. 1988. Peptide sequences that target proteins for enhanced degradation during serum withdrawal. J. Biol. Chem. 263: 6797-6805. DICE,J. F. 1990. Peptide sequences that target cytosolic proteins for lysosomal proteolysis. Trends Biochem. Sci. 15: 305-309. DICE,J. F. & H:L. CHIANG.1989. Peptide signals for protein degradation within lysosomes. Biochem. SOC. Symp. 55: 45-55. CHIANG,H.-L., S. R. TERLECKY, C. P. PLANT & J. F. DICE. 1989. A role for a 70kilodalton heat shock protein in lysosomal proteolysis of intracellular proteins. Science 2%: 282-285. KURTZ,S., J. ROW, L. PETKO& S. LINDQUIST.1986. An ancient developmental induction: heat shock proteins induced in sporulation and oogenesis. Science 231: 11541157.
64
ANNALS NEW YORK ACADEMY OF SCIENCES
23. DESHAIES,R. J., B. D. KOCH, M. WERNER-WASHBURNE, E. A. CRUG & R. SCHEKMAN. 1988. 70 kD stress protein homologues facilitatetranslocation ofsecretory and mitochondrial precursor polypeptides. Nature 332: 800-805. 24. CHIRICO,W. J., M. G. WATERS& G. BLOBEL.1988. 70K heat shock related proteins stimulate protein translocation into microsomes. Nature 332: 805-810. S. R., H.-L. CHIANG,T. S. OLSON & J. F. DICE. 1992. Protein and peptide 25. TERLECKY, binding and stimulation ofin pitro lysosomal proteolysis by the 73-kDa heat shock cognate protein. J. Biol. Chem. 267: 9202-9209.
INTRODUCTION Both cytosolic and lysosomal pathways of proteolysis operate in most cells.lJ In well-nourished cells lysosomes appear to be able to internalize proteins by a process called microautophagy (FIG. 1) in which the lysosomal membrane invaginates at multiple locations. Microautophagy appears to be nonselective in that several proteins and inert particles are internalized at similar r a t e ~ . ~ . ~ When cultured cells reach confluence an additional lysosomal pathway of proteoly~.~ ( FIG. 1)involves formation ofautophagic vacuoles sis is s t i m ~ l a t e d .Macroautophagy which sequester areas of cytoplasm.’J These vacuoles then acquire lysosomal hydrolases to form autophagosomes. Macroautophagy also appears to be nonselective in that many different organelles and proteins are sequestered at approximately the same We have studied an additional pathway of lysosomal proteolysis that is activated in confluent cell monolayers in response to serum withdrawal. This pathway is restricted to cytosolic proteins that contain peptide sequences biochemically related to Lys-Phe-Glu-Arg-Gln(KFERQ). The mechanism by which proteins with KFERQ-like peptide regions are targeted to lysosomes for degradation is similar in certain respects to the direct import of newly synthesized proteins into mitochondria or into the lumen of the endoplasmic reticulum. Therefore, proteins with KFERQ-like peptide regions may enter lysosomes by directly crossing a membrane bilayer (FIG. 1).
Ribonudease A as a Probe for Selective Lposomal Proteolpis We used red cell-mediated microinjection to introduce specific radiolabeled proteins into the cytosol of confluent cultures of human fibroblasts.l*J Radiolabeled ribonuclease A (RNase A) is degraded with a half-life of approximately 100 h in serumsupplemented cells and 50 h in serum-deprived cells. The half-lives of certain other microinjected proteins, but not others, are reduced in response to serum withdrawal
“Research in the author’s laboratory is supported by National Institutes of Health Grants AGO61 16 and AG07472. 58
DICE: SELECTIVE DEGRADATION OF PROTEINS BY LYSOSOMES
59
MICROAUTOPHAGY
FIGURE 1. Schematic representations of pathways of lysosomal proteolysis. See the text for descriptions of the pathways.
MACRoAUTOPHAGY
DIRECT PROTEIN TRANSFER
0
(TABLE 1). These data suggest that the enhanced degradation during serum withdrawal is a selective process. Several lines of evidence indicate that microinjected RNase A is degraded within lysosomes: (a) a small amount of microinjected RNase A fractionates with lysosomes,” ’ ~ degrada(b) degradation of RNase A is partially inhibited by ammonium ~ h l o r i d e , (c) , ~ ~ (d) the tion products of [3H]raffinose-RNaseA accumulate only in l y s o ~ o m e s and same peptides derived from RNase A are released from cells after microinjection and after lysosomal hydrolysis following endocytosis. l 5 The amino-terminal 20 amino acids of RNase A (RNase S-peptide) are required for the enhanced degradation in response to serum withdrawal,’ and covalent attachment of this peptide to heterologous proteins also causes their degradation rates to increase in response to serum withdrawal.16 We eventually identified amino acids 7- 11 within RNase S-peptide (KFERQ) as the region essential for enhanced degradation.” Generality of This Pathway of Protein Degradation To examine whether peptide regions similar to KFERQ exist in intracellular proteins, we raised polyclonal antibodies to KFERQ and affinity purified IgGs specifically TABLE 1.
Degradation of Long-Lived Proteins Microinjected into Human Cell9 T1/2 ( h )
Protein Group 1: Regulated Degradation RNase A RNase S-peptide Aspartate amino-transferase Pyruvate kinase Hemoglobin Group 2: Nonregulated Degradation RNase S-Protein Ovalbumin Lysozyme Insulin A-chain P-galactosidase ~~
Species
2 Serum
- Serum
cow cow Pig chicken human
100 62 80 296 213
50 30 40 121 106
cow chicken chicken cow E. coli
100 55 59 104 1756
100 49 59 104 1 75b
~
‘References to half-life values have been previously cited.” IMR-90 human diploid fibroblasts were recipient cells in all cases except for pyruvate kinase which was injected into HeLa cells. 6L. J. Terlecky and J. F. Dice, unpublished results.
ANNALS NEW YORK ACADEMY OF SCIENCES
60
-x
L
I
0
w I-
n
35
30
-
25
-
0
50
0
anti-KFERQ
0
anti-RYLPT
0
prsimmune
100
150
FIGURE 2. Immunoprecipitation of radiolabeled cytosolic proteins from fibroblasts. Cells were labeled for 2 d with [3H]leucine and cytosolic proteins prepared as described elsewhere." Mnity-purified polyclonal antibodies were used for immunoprecipitations.
200
250
ANTIBODY IpI)
directed toward the pentapeptide. l 8 The anti-KFERQ IgGs immunoprecipitated approximately 30% of [3H]leucine-labeled cytosolic proteins from human fibroblastszz (FIG.2). In pulse-chase experiments, the immunoprecipitable proteins were preferentially degraded in response to serum withdrawalla (FIG.3). In contrast, nonimmunoprecipitable proteins were degraded at the same rate in the presence and absence of serum.18
The Peptide Motif We compared amino acid sequences of four microinjected long-lived proteins whose degradation is enhanced during serum deprivation with four microinjected long lived proteins whose degradation is unaffected by serum. A common sequence motif exists in the first group of proteins but not in the second group. 19.z0 The motif appears to be a pentapeptide with a Gln at either the carboxyl or amino terminus. The pentapeptide must also contain an acidic (Glu o r Asp), a basic (Lys or Arg), and a highly hydrophobic amino acid (Phe, Leu, Ile, or Val). The fifth amino acid of the pentapeptide can be either a repeated basic or hydrophobic amino acid. With the exception of Gln, the relative positions of the amino acids do not appear to be important, but they must be contiguous. While this motif is highly redundant, of the 3.2 million possible pentapeptides, only 2304 fit the motif, that is, less than 0.1%.
FIGURE 3. Rapid degradation of immunoprecipitable proteins in response to serum withdrawal. Cells were radiolabeled for 2 d and then chased for 0, 1, or 2 d in unlabeled medium containing 10% calf serum ( + S) o r not ( - S). Cells were harvested, and radioactivity remaining in antiWERQ immunoprecipitable proteins followed. I s 2.5
TIME Idayrl
;':I
DICE: SELECTIVE DEGRADATION OF PROTEINS BY LYSOSOMES
FIGURE 4. Uptake and degradation of [3H]RNase S-peptide by isolated lysosomes. Lysosomes from 2 x lo6 cells were incubated with radiolabeled R!!ase S-peptide for 2 h at room temperature.*' Other additions were ATP and an ATP regenerating system (10 mM), Prp73 (10 pg/ml), and NH4CI (10 mM). Radioactivity converted to acid-soluble material is shown.
61
m
3
A
s
-
4
e 2 0
a
0 NONE
ATP PRP73
ATP
ATP
PRP73
PRP73
' 0
C.
NH,CI
Selective Degradation ofProteins by Isolated Lysosomes To try to elucidate the mechanism of degradation of KFERQ motif-containing proteins, we developed an in v h o assay using lysosomes isolated from IMR-90 fibroblasts over two consecutive discontinuous density gradientsZL(S. R. Terlecky and J. F. Dice, unpublished). Maximal degradation of [3H]RNase S-peptide by isolated lysosomes requires both ATP and a heat shock protein of 70 kDa (hsp70) (FIG. 4). Similar results have been obtained using native [jH]RNase A. This degradation is specific because [3H]RNase S-protein, amino acids 21-124 of RNase A which does not contain a KFERQ motif, is degraded little, if at all, under the same conditions. In addition, degradation of [jH]RNase S-peptide can be inhibited by reducing the temperature, and degradation appears to occur within acidified organelles because it is inhibited by ammonium chloride (FIG. 4). Additional experiments show that the lysosomal uptake of [jH]RNase S-peptide is saturable. Furthermore, at 0-4 "C. specific binding of [jH]RNase S-peptide to a lysosomal membrane protein occurs. Presumably, this binding component is a receptor or a peptide transport channel.
The Role ofan HSP70 in Selective Lysosomal Proteolysis We considered the possibility that an intracellular protein may recognize KFERQlike peptide regions in proteins to be degraded more rapidly in response to serum withdrawal. A protein of 73 kDa was purified on an RNase S-peptide affinity column and was designated prp73 for peptide recognition protein of 73 A monoclonal antibody that recognizes all of the members of the hsp7O familyz2reacts with prp73 purified from human fibroblast cytosol. In addition, sequence data obtained from purified prp73 tentatively identified it as the constitutively expressed heat shock cognate protein of 73 kilodaltons ( h s ~ 7 3 ) .A~ 'yeast analog of hsc73 is required for transport of proteins into mitochondria and the endoplasmic r e t i c u l ~ m . ~ ~ ~ ~ ~ Hsc73 and prp73 both bind to RNase A, RNase S-peptide, KFERQ, aspartate aminotransferase, and pyruvate kinase-proteins and peptides that contain KFERQ-like peptide motifs. Neither hsc73 nor prp73 bind to ovalbumin, lysozyme, or ubiquitinproteins that lack KFERQ-like peptide motifs.25 Finally, hsc73 and prp73 function identically in stimulating degradation of RNase S-peptide by isolated lysosomes under conditions where three other members of the hsp70 family have no Therefore, one of the hnctions of hsc73 during nutritional deprivation appears to be to recognize and target specific cytosolic proteins to lysosomes for degradation.
62
ANNALS NEW YORK ACADEMY OF SCIENCES
The mechanisms by which hsc73 promotes lysosomal degradation of proteins containing KFERQ-like pepride regions are not known. In response to serum withdrawal hsc73 may shift its subcellular distribution to the cytosol where the proteins containing the KFERQ-like regions reside.*l Activation of the pathway could also involve changes in the substrate proteins that expose the KFERQ-like regions. Finally, activation of the pathway may be due to regulation of a component distal to hsc73 binding to the protein substrate. For example, the activity of the putative receptor or peptide transporter on the lysosome surface may be regulated.
Possible Relevance of This Proteolytc Pathway to Alzheimer’s Disease It seems unlikely that the amyloid precursor protein (APP) is targeted for lysosomal proteolysis through the KFERQ-selectivepathway. This pathway applies primarily to cytosolic proteins while APP is a transmembrane protein. The APP is more likely to be exposed to lysosomal hydrolases through endocytic and recycling pathways that apply to many cell surface proteins. However, if APP (or peptides derived from APP) ever appear in the cytosol, lysosomal uptake and degradation is possible by both nonselective and selective pathways. The release of peptide fragments from proteins that are degraded within lysosomes after microinjection into the cytosol or after e n d o c y t o s i ~may ~ ~ reflect how the fiamyloid fragment is released from cells. Transfer of these peptides from the lysosome to the medium may require a single transport step, or, alternatively, may require transfer of the peptides to the cytosol or other organelles.15
SUMMARY Lysosomes are able to internalize cellular proteins in a variety ofways. One pathway is selective for cytosolic proteins containing peptide sequences biochemically related to Lys-Phe-Glu-Arg-Gln(KFERQ). This pathway is activated in confluent monolayers of cultured cells in response to deprivation of serum growth factors and applies to approximately 30% of cytosolic proteins. We have reconstituted this lysosomal degradation pathway in Pitm. ,Uptake and/or degradation is stimulated by ATP and a member of the heat shock 70-kilodalton protein family, the 73-kiloddton constitutive heat shock protein. Several possible mechanisms of selective protein transport into lysosomes and the possible relevance of this proteolytic pathway to the processing of the amyloid precursor protein are discussed.
ACKNOWLEDGMENTS I thank former and present members of my laboratory who have been responsible for much of the work presented. REPERENCES 1. DICE,J. F. 1987. Molecular determinants of protein half-lives in eukaryotic cells. FASEB J. 1: 349-357. & J. F. DICE. 1992. Pathways of intracellular protein 2. OLSON,T. S., S. R. TERLECICY degradation in eukaryotic cells. In Stability of Protein Pharmaceuticals:In Vim Pathways
DICE: SELECTIVE DEGRADATION OF PROTEINS BY LYSOSOMES
3. 4.
5. 6. 7. 8. 9. 10. 11.
12. 13.
14.
15. 16. 17. 18.
19. 20. 21. 22.
63
of Degradation and Strategies for Protein Stabilization. T. J. Ahern & M. C. Manning, Eds. In press. Plenum Publishing Corporation. New York. AHLBERG,J., L. MAFUELLA& H . GWUMANN.1982. Uptake and degradation of proteins by isolated rat liver lysosomes. Suggestion of a microautophagic pathway of proteolysis. Lab. Invest. 47: 523-532. MAFUELLA,L. & H. GWUMANN.1987. Autophagy, microautophagy, and crinophagy as mechanisms for protein degradation. In Lysosomes: Their Role in Protein Breakdown. H. Glaumann & F. J. Ballard, Eds. 319-367. Academic Press. New York. COCKLE,S. M. & R. T. DEAN.1982. The regulation of proteolysis in normal fibroblasts as they approach confluence. Evidence for participation of the lysosomal system. Biochem. J. 208: 243-249. KNECHT,E., J. HERNANDEZ-YAGO & S. GRISOLIA.1984. Regulation of lysosomal autophagy in transformed and nontransformed mouse fibroblasts under several growrh conditions. Exp. Cell Res. 154: 224-232. MORTIMORE, G. E. 1987. Mechanism and regulation of induced and basal protein degradation in liver. In Lysosomes: Their Role in Protein Breakdown. H . Glaumann & F. J. Ballard, Eds. 415-444. Academic Press. New York. PFEIFER,U. 1987. Functional morphology of the lysosomal apparatus. In Lysosomes: Their Role in Protein Breakdown H . Glaumann & F. J. Ballard, Eds. 3-59. Academic Press. New York. KOMINAMI,E., S. HASHIDA, E. A. KHAIWLAH & N. KATUNUMA. 1983. Sequestration of cytoplasmic enzymes in an autophagic vacuole-lysosomal system induced by injection of leupeptin. J. Biol. Chem. 258: 6093-6100. HENDIL,K. 1981. Autophagy of metabolically inert substances injected into fibroblasts in culture. Exp. Cell Res. 135: 157-166. KOPITZ,J., G. 0. KISEN, P. B. GORDON,P. BOHLEY& P. 0.SEGLEN.1990. Non-selective autophagy ofcytosolic enzymes in isolated rat hepatocytes. J. Cell Biol. 111: 941-954. NEFF,N. T., L. BOURRET, P. MIAO& J. F. DICE.1982. Degradation ofproteins microinjected into IMR-90 human diploid fibroblasts. J. Cell Biol. 91: 184-194. MCELLIGOTT, M. A. & J. F. DICE. 1984. Microinjection of cultured cells using red cellmediated fusion and osmotic lysis of pinosomes: a review of methods and applications. Biosci. Rep. 4: 451-466. MCELLIGOTT,M. A,, P. MIAO& J. F. DICE.1985. Lysosomal degradation of ribonuclease A and ribonuclease S-protein microinjected into human fibroblasts. J. Biol. Chem. 260: 11986-11993. ISENMAN,L. D. & J. F. DICE. 1989. Secretion of intact proteins and peptide fragments by lysosomal pathways of protein degradation. J. Biol. Chem. 264: 21591-21596. BACKER,J. M. & J. F. DICE.1986. Covalent linkage of ribonuclease S-peptide to microinjected proteins causes their intracellular degradation to be enhanced during serum withdrawal. Proc. Natl. Acad. Sci. USA 83: 5830-5834. DICE,J. F., H.-L. CHIANG,E. P. SPENSER& J. M. BACKER.1986. Regulation ofcatabolism of microinjected ribonuclease A: identification of residues 7- 11 as the essential pentapep tide. J. Biol. Chem. 262: 6853-6859. CHIANG,H.-L. & J. F. DICE. 1988. Peptide sequences that target proteins for enhanced degradation during serum withdrawal. J. Biol. Chem. 263: 6797-6805. DICE,J. F. 1990. Peptide sequences that target cytosolic proteins for lysosomal proteolysis. Trends Biochem. Sci. 15: 305-309. DICE,J. F. & H:L. CHIANG.1989. Peptide signals for protein degradation within lysosomes. Biochem. SOC. Symp. 55: 45-55. CHIANG,H.-L., S. R. TERLECKY, C. P. PLANT & J. F. DICE. 1989. A role for a 70kilodalton heat shock protein in lysosomal proteolysis of intracellular proteins. Science 2%: 282-285. KURTZ,S., J. ROW, L. PETKO& S. LINDQUIST.1986. An ancient developmental induction: heat shock proteins induced in sporulation and oogenesis. Science 231: 11541157.
64
ANNALS NEW YORK ACADEMY OF SCIENCES
23. DESHAIES,R. J., B. D. KOCH, M. WERNER-WASHBURNE, E. A. CRUG & R. SCHEKMAN. 1988. 70 kD stress protein homologues facilitatetranslocation ofsecretory and mitochondrial precursor polypeptides. Nature 332: 800-805. 24. CHIRICO,W. J., M. G. WATERS& G. BLOBEL.1988. 70K heat shock related proteins stimulate protein translocation into microsomes. Nature 332: 805-810. S. R., H.-L. CHIANG,T. S. OLSON & J. F. DICE. 1992. Protein and peptide 25. TERLECKY, binding and stimulation ofin pitro lysosomal proteolysis by the 73-kDa heat shock cognate protein. J. Biol. Chem. 267: 9202-9209.
