Limited Accumulation of Damaged Proteins inl-Isoaspartyl (d-Aspartyl)O-Methyltransferase-deficient Mice
2001; Elsevier BV; Volume: 276; Issue: 23 Linguagem: Inglês
10.1074/jbc.m100987200
ISSN1083-351X
AutoresJonathan D. Lowenson, Edward Kim, Stephen G. Young, Steven Clarke,
Tópico(s)Muscle metabolism and nutrition
Resumol-Isoaspartyl (d-aspartyl) O-methyltransferase (PCMT1) can initiate the conversion of damaged aspartyl and asparaginyl residues to normal l-aspartyl residues. Mice lacking this enzyme (Pcmt1−/− mice) have elevated levels of damaged residues and die at a mean age of 42 days from massive tonic-clonic seizures. To extend the lives of the knockout mice so that the long term effects of damaged residue accumulation could be investigated, we produced transgenic mice with a mouse Pcmt1 cDNA under the control of a neuron-specific promoter. Pcmt1 transgenic mice that were homozygous for the endogenous Pcmt1 knockout mutation ("transgenic Pcmt1−/− mice") had brain PCMT1 activity levels that were 6.5–13% those of wild-type mice but had little or no activity in other tissues. The transgenicPcmt1−/− mice lived, on average, 5-fold longer than nontransgenic Pcmt1−/− mice and accumulated only half as many damaged aspartyl residues in their brain proteins. The concentration of damaged residues in heart, testis, and brain proteins in transgenic Pcmt1−/− mice initially increased with age but unexpectedly reached a plateau by 100 days of age. Urine fromPcmt1−/− mice contained increased amounts of peptides with damaged aspartyl residues, apparently enough to account for proteins that were not repaired intracellularly. In the absence of PCMT1, proteolysis may limit the intracellular accumulation of damaged proteins but less efficiently than in wild-type mice having PCMT1-mediated repair. l-Isoaspartyl (d-aspartyl) O-methyltransferase (PCMT1) can initiate the conversion of damaged aspartyl and asparaginyl residues to normal l-aspartyl residues. Mice lacking this enzyme (Pcmt1−/− mice) have elevated levels of damaged residues and die at a mean age of 42 days from massive tonic-clonic seizures. To extend the lives of the knockout mice so that the long term effects of damaged residue accumulation could be investigated, we produced transgenic mice with a mouse Pcmt1 cDNA under the control of a neuron-specific promoter. Pcmt1 transgenic mice that were homozygous for the endogenous Pcmt1 knockout mutation ("transgenic Pcmt1−/− mice") had brain PCMT1 activity levels that were 6.5–13% those of wild-type mice but had little or no activity in other tissues. The transgenicPcmt1−/− mice lived, on average, 5-fold longer than nontransgenic Pcmt1−/− mice and accumulated only half as many damaged aspartyl residues in their brain proteins. The concentration of damaged residues in heart, testis, and brain proteins in transgenic Pcmt1−/− mice initially increased with age but unexpectedly reached a plateau by 100 days of age. Urine fromPcmt1−/− mice contained increased amounts of peptides with damaged aspartyl residues, apparently enough to account for proteins that were not repaired intracellularly. In the absence of PCMT1, proteolysis may limit the intracellular accumulation of damaged proteins but less efficiently than in wild-type mice having PCMT1-mediated repair. l-isoaspartyl (d-aspartyl)O-methyltransferase S-adenosyl-l-methionine S-adenosyl[methyl-14C]-l-methionine neuron-specific enolase 2,2-bis(hydroxymethyl)-2,2′,2"-nitrilotriethanol The spontaneous chemical modification of proteins by reaction with oxygen, water, sugars, and other abundant metabolites is unavoidable. The accumulation of such nonenzymatically altered proteins is associated with normal aging as well as atherosclerosis, Alzheimer's disease, and diabetes (1Brownlee M. Annu. Rev. Med. 1995; 46: 223-234Crossref PubMed Scopus (1132) Google Scholar, 2Lowenson J.D. Clarke S. Roher A.E. Methods Enzymol. 1999; 309: 89-105Crossref PubMed Scopus (23) Google Scholar, 3Stadtman E.R. Levine R.L. Ann. N. Y. Acad. Sci. 2000; 899: 191-208Crossref PubMed Scopus (941) Google Scholar). Organisms have several strategies for dealing with damaged proteins, including intracellular proteolysis mediated by proteasome and lysosome action (4Friguet B. Bulteau A.L. Chondrogianni N. Conconi M. Petropoulos I. Ann. N. Y. Acad. Sci. 2000; 908: 143-154Crossref PubMed Scopus (148) Google Scholar, 5Sano H. Nagai R. Matsumoto K. Horiuchi S. Mech. Ageing Dev. 1999; 107: 333-346Crossref PubMed Scopus (59) Google Scholar, 6Tarcsa E. Szymanska G. Lecker S. O'Connor C.M. Goldberg A.L. J. Biol. Chem. 2000; 275: 20295-20301Abstract Full Text Full Text PDF PubMed Scopus (97) Google Scholar). Some types of covalent damage, however, are simple enough to recognize and repair directly (7Visick J.E. Clarke S. Mol. Microbiol. 1995; 16: 835-845Crossref PubMed Scopus (98) Google Scholar). Enzymes such as prolyl cis-trans isomerase (8Schiene C. Fischer G. Curr. Opin. Struct. Biol. 2000; 10: 40-45Crossref PubMed Scopus (169) Google Scholar), methionine sulfoxide reductase (9Moskovitz J. Flescher E. Berlett B.S. Azare J. Poston J.M. Stadtman E.R. Proc. Natl. Acad. Sci. U. S. A. 1998; 95: 14071-14075Crossref PubMed Scopus (238) Google Scholar), and disulfide isomerase (8Schiene C. Fischer G. Curr. Opin. Struct. Biol. 2000; 10: 40-45Crossref PubMed Scopus (169) Google Scholar) can restore activity to proteins that have been chemically altered. We are interested in a common type of spontaneous protein damage in which l-aspartyl and l-asparaginyl residues undergo an intramolecular reaction that converts them tol-succinimidyl residues (10Aswad D.W. Paranandi M.V. Schurter B.T. J. Pharmacol. Biomed. Anal. 2000; 21: 1129-1136Crossref PubMed Scopus (196) Google Scholar, 11Volkin D.B. Mach H. Middaugh C.R. Mol. Biotechnol. 1997; 8: 105-122Crossref PubMed Scopus (91) Google Scholar). Nonenzymatic hydrolysis of the succinimide ring readily occurs at either carbonyl to generate both normal aspartyl residues and isoaspartyl residues, in which the peptide backbone proceeds through the β-carbonyl rather than the α-carbonyl moiety (12Geiger T. Clarke S. J. Biol. Chem. 1987; 262: 785-794Abstract Full Text PDF PubMed Google Scholar). The succinimide also racemizes more rapidly than do the open chain forms, and hydrolysis of thed-succinimide produces d-aspartyl andd-isoaspartyl residues (12Geiger T. Clarke S. J. Biol. Chem. 1987; 262: 785-794Abstract Full Text PDF PubMed Google Scholar). Local protein structure causes some l-aspartyl and l-asparaginyl residues to be especially prone to succinimide formation, and the presence of damaged aspartyl residues at these sites can significantly alter the structure, function, and immunogenicity of the protein (10Aswad D.W. Paranandi M.V. Schurter B.T. J. Pharmacol. Biomed. Anal. 2000; 21: 1129-1136Crossref PubMed Scopus (196) Google Scholar, 13Mamula M.J. Gee R.J. Elliott J.I. Sette A. Southwood S. Jones P.J. Blier P.R. J. Biol. Chem. 1999; 274: 22321-22327Abstract Full Text Full Text PDF PubMed Scopus (175) Google Scholar). To minimize the accumulation of damaged aspartyl residues in cellular proteins, all mammalian tissues possess an l-isoaspartyl (d-aspartyl) O-methyltransferase (EC 2.1.1.77; designated PCMT11 in mice) (14Clarke S. Cheng X. Blumenthal R.M. S-Adenosylmethionine-dependent Methyltransferases: Structures and Functions. World Scientific Publishing, Singapore1999: 123-148Crossref Google Scholar). This enzyme uses S-adenosyl-l-methionine (AdoMet) to methylate l-isoaspartyl (and, less efficiently,d-aspartyl) residues but not normal l-aspartyl residues (15Lowenson J.D. Clarke S. Aswad D.W. Deamidation and Isoaspartate Formation in Peptides and Proteins. CRC Press, Inc., Boca Raton, FL1995: 47-64Google Scholar). Nonenzymatic deesterification of the methylated residues returns them to the succinimide form much more rapidly than occurs in the absence of methylation, resulting in the eventual conversion of most of the damaged residues to the "repaired"l-aspartyl form (16McFadden P.N. Clarke S. Proc. Natl. Acad. Sci. U. S. A. 1987; 84: 2595-2599Crossref PubMed Scopus (162) Google Scholar, 17Johnson B.A. Murray Jr., E.D. Clarke S. Glass D.B. Aswad D.W. J. Biol. Chem. 1987; 262: 5622-5629Abstract Full Text PDF PubMed Google Scholar). The physiological importance of this pathway remained unclear untilPcmt1 knockout (Pcmt1−/−) mice were created and found to display a distinctive phenotype (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar, 19Yamamoto A. Takagi H. Kitamura D. Tatsuoka H. Nakano H. Kawano H. Kuroyanagi H. Yahagi Y. Kobayashi S. Koizumi K. Sakai T. Saito K. Chiba T. Kawamura K. Suzuki K. Watanabe T. Mori H. Shirasawa T. J. Neurosci. 1998; 18: 2063-2074Crossref PubMed Google Scholar).Pcmt1−/− mice have 2–6-fold higher levels of damaged aspartyl residues in their brain, heart, liver, and erythrocytes than are observed in wild-type tissues (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar). Furthermore,Pcmt1−/− mice are smaller than their Pcmt1+/−and Pcmt1+/+ littermates, undergo severe tonic-clonic seizures, and die at an average age of 42 days (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar, 19Yamamoto A. Takagi H. Kitamura D. Tatsuoka H. Nakano H. Kawano H. Kuroyanagi H. Yahagi Y. Kobayashi S. Koizumi K. Sakai T. Saito K. Chiba T. Kawamura K. Suzuki K. Watanabe T. Mori H. Shirasawa T. J. Neurosci. 1998; 18: 2063-2074Crossref PubMed Google Scholar). Electroencephalographic analysis shows that these mice suffer abnormal brain activity about 50% of the time, not just during the tonic-clonic seizures (20Kim E. Lowenson J.D. Clarke S. Young S.G. J. Biol. Chem. 1999; 274: 20671-20678Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar). Administration of the anti-seizure drug valproic acid enabled Pcmt1−/− mice to attain the same size and weight as their wild-type littermates, suggesting that the absence of the methyltransferase did not interfere directly with food intake or metabolism (20Kim E. Lowenson J.D. Clarke S. Young S.G. J. Biol. Chem. 1999; 274: 20671-20678Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar). The combination of valproic acid and clonazepam prolonged mean survival but only by 36 days (20Kim E. Lowenson J.D. Clarke S. Young S.G. J. Biol. Chem. 1999; 274: 20671-20678Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar). These results have raised new questions. Does the repair methyltransferase, although expressed in all tissues, have particular importance within the brain? Is it important in neurons or in nonneuronal cell types? If the seizures in the Pcmt1−/−mice were prevented, would other organs function abnormally as damaged aspartyl residues accumulated? In the current study, we have approached these questions by producing transgenic mice that express Pcmt1 solely within neurons. A rat neuron-specific enolase (NSE) promoter was used to direct the expression of mousePcmt1 cDNA in the brains of transgenic mice. The methyltransferase coding sequence (including 119 base pairs of 5′-noncoding and 777 base pairs of 3′-noncoding sequence) was obtained from a 1580-base pair murine testis cDNA clone (21Romanik E.A. Ladino C.A. Killoy L.C. D'Ardenne S.C. O'Connor C.M. Gene ( Amst. ). 1992; 118: 217-222Crossref PubMed Scopus (20) Google Scholar) and was removed from plasmid sequences with EcoRI. The proximal rat NSE promoter was isolated from the plasmid NSE-APP695 (22Higgins L.S. Catalano R. Quon D. Cordell B. Ann. N. Y. Acad. Sci. 1993; 695: 224-227Crossref PubMed Scopus (49) Google Scholar) after digestion with HindIII. After overhangs were filled with Klenow polymerase, the mouse Pcmt1 cDNA and the rat NSE vector were ligated, and the NSE-Pcmt1 transgenic construct was isolated by digestion with SalI. The transgene (2 ng/μl) was microinjected into F2 C57BL/6 × SJL fertilized mouse eggs by standard techniques (23Hogan B. Beddington R. Costantini F. Lacy E. Manipulating the Mouse Embryo: A Laboratory Manual. 2nd Ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY1994Google Scholar). From 37 microinjected eggs, 33 pups were obtained, and 4 harbored the Pcmt1 transgene. These transgenic founders were identified by polymerase chain reaction with primers corresponding to mouse Pcmt1 cDNA sequences (5′-GCCAGCCACTCGGAGCTAATCC-3′ from exon 1 and 5′-CCACTATTTCCAACCATCCGTGC-3′ from exons 4 and 5). Southern blot analysis of tail DNA confirmed the integration of the transgene (Fig.1). Two of these mice, founders 27 and 29, were bred with C57BL/6 × 129/SvJae mice that were heterozygous for a knockout mutation in the endogenous Pcmt1gene (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar). DNA samples from the tails of these mice were genotyped by polymerase chain reaction, both to detect the transgene and to detect the knockout mutation. Mice that were heterozygous for the knockout mutation and hemizygous for the transgene were selected for breeding. All mice were weaned at 21 days of age, housed in a barrier facility with a 12-h light/dark cycle, and fed a chow diet. Brain, heart, liver, kidney, and testis were removed immediately from sacrificed animals and placed in ice-cold buffer (5 ml/g wet weight) containing 250 mmsucrose, 10 mm Tris-HCl, 1 mm disodium EDTA, pH 7.4, and the protease inhibitor phenylmethylsulfonyl fluoride (25 μm). The tissues were disrupted in a glass homogenization tube with a Teflon pestle rotating at 310 rpm for four 10-s intervals. The homogenates were placed in 1.5-ml tubes and centrifuged at 20,800 × g for 10 min. The resulting supernatant fractions contained both cytosolic proteins and microsomes and were kept frozen until used. Whole blood (100–200 μl) was taken from the tail or heart and combined with 200 μl of 2 mg/ml disodium EDTA, 10 mmsodium phosphate, 137.9 mm sodium chloride, pH 7.4. Erythrocytes were collected by centrifugation at 4000 ×g for 4 min and washed four times with 1 ml of the above buffer. Pelleted erythrocytes were lysed in 5 volumes of 5 mm sodium phosphate, 1 mm disodium EDTA, pH 7.4, and 25 μm phenylmethylsulfonyl fluoride. The lysates were centrifuged in 1.5-ml conical tubes at 20, 800 ×g for 10 min to remove the membrane ghosts, and the supernatant fractions were stored at −20 °C. A modified Lowry assay was used to determine protein concentrations in the extracts. Assays were done in duplicate with bovine serum albumin as a standard (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar). Methyltransferase activity was assayed by its ability to transfer the methyl group fromS-adenosyl-l-methionine to ovalbumin. The supernatant fraction from homogenized tissues (10–60 μg of brain, heart, or testis protein; 0.6–0.8 mg of erythrocyte protein) was incubated with 0.8 mg of ovalbumin (Sigma, Grade V) in 0.2m BisTris-HCl, pH 6.0, containing 10 μm[14C]AdoMet (53 mCi/mmol; Amersham Pharmacia Biotech) in a final volume of 40 μl at 37 °C for 15 min. NaOH (80 μl of a 200 mm solution) was added to stop the reaction and to hydrolyze the [14C]methyl esters formed on ovalbumin to [14C]methanol. The reaction mixture was immediately spotted onto an 8 × 2-cm piece of thick filter paper and incubated above 5 ml of Safety-Solve scintillation fluid (Research Products International) in the neck of a sealed 20-ml scintillation vial at room temperature for 2 h to allow [14C]methanol to diffuse into the scintillation fluid. The filter paper was removed, and the scintillation fluid was counted for radioactivity. Incubations containingS-adenosyl-l-[methyl- 14C]methionine, ovalbumin, and buffer constituted the "blank" for the assay; the radioactivity in these tubes (typically <5% of the nonblank samples) was subtracted from total counts in the determination of enzyme activity. Cellular proteins were incubated at 37 °C for 2 h with 0.8 μg of recombinant human l-isoaspartyl methyltransferase (specific activity, 10,000 pmol of methyl esters formed on ovalbumin at 37 °C/min/mg protein) (24MacLaren D.C. Clarke S. Protein Expression Purif. 1995; 6: 99-108Crossref PubMed Scopus (32) Google Scholar) in 0.2 m BisTris-HCl, pH 6.0, and 10 μm [14C]AdoMet in a final volume of 40 μl. After base hydrolysis, [14C]methanol production was measured as described above to quantitate l-isoaspartyl andd-aspartyl methyl-accepting sites in cellular proteins. Incubations containing [14C]AdoMet, recombinant methyltransferase, and buffer constituted the blank for the assay; the radioactivity in these tubes was subtracted from each sample's total counts. Each sample was assayed in duplicate or triplicate, and the average value is reported. Damaged residues that are already methylated within cells by the endogenous methyltransferase andS-adenosyl-l-methionine are not measured in the assay described above but can be quantified after mild base treatment. Protein (8.3–9.6 mg) from homogenized Pcmt1−/− andPcmt1+/+ brains was incubated in 20 μl of 75 mm potassium borate, pH 10.2. After times ranging from 5 s to 360 min, 10 μl of 500 mm BisTris-HCl, pH 5.7, was added to lower the pH to about 6. Then, recombinant human methyltransferase (5 pmol/min) andS-adenosyl-l-[methyl- 14C]methionine (10 μm final concentration) in 10 μl of 150 mm BisTris-HCl, pH 6.0, was added, and these reaction mixtures were incubated at 37 °C for 135 min. The reaction was stopped by freezing on dry ice, and then the base-labile methyl esters were quantitated as described above. Urine, freshly voided on Parafilm, was collected with a pipette and stored frozen until used. Creatinine in the urine was measured by a modified form of the procedure of Bosnes and Taussky (25Bonsnes R.W. Taussky H.H. J. Biol. Chem. 1945; 109: 581-591Abstract Full Text PDF Google Scholar). An aliquot of each urine sample (0.3–1 μl) or standard creatinine (0–25 μg) was diluted to 50 μl with water in duplicate tubes. Picric acid was added (25 μl of a 40 mm solution), and the tubes were capped and incubated in a boiling water bath for 45 min. After cooling to room temperature, 25 μl of 0.75 m NaOH was added. Within 15 min, 90 μl of each sample was transferred to a flat-welled microtiter plate (Costar), and absorbance was measured at 525 nm with a Beckman DU-600 plate reader. Damaged aspartyl residues in the urine were assayed with recombinant human methyltransferase as described above. Free amino acids and isoaspartyl-containing dipeptides in urine were derivatized witho-phthalaldehyde and 2-mercaptoethanol, separated by reverse-phase high pressure liquid chromatography, and quantitated by fluorescence as described previously (26Gary J.D. Clarke S. J. Biol. Chem. 1995; 270: 4076-4087Abstract Full Text Full Text PDF PubMed Scopus (39) Google Scholar). A precipitate that formed upon mixing of the urine and derivatization reagent was removed by centrifugation at 20,800 × g for 3 min prior to injection. Urine that had been dried in 6 × 50-mm glass tubes was hydrolyzed in vaporized hydrochloric acid in vacuo at 108 °C for 3 h with a PicoTag Work Station (Waters); amino acids in the hydrolysates were quantitated as described above. The fluorescence color constants for these derivatives were determined with amino acid and dipeptide standards. We generated two Pcmt1transgenic mouse lines, lines 27 and 29, in which the murinePcmt1 methyltransferase cDNA was placed under the control of a neuron-specific promoter. We then compared the survival of "transgenic Pcmt1−/− mice" with that of "nontransgenic Pcmt1−/− mice." Whereas nontransgenicPcmt1−/− (n = 129) mice died at a median age of 44 days (with only one mouse living beyond 150 days), the transgenic Pcmt1−/− mice lived much longer (Fig.2). Of 11 line 27 transgenicPcmt1−/− mice examined in this study, 6 died between 30 and 90 days of age, but 4 lived from 549 to 757 days. Line 29 transgenic Pcmt1−/− mice (n = 19) died at a median age of 213 days (Fig. 2), and 3 lived more than 400 days. The nontransgenic Pcmt1−/− mice began to die at about 21 days. In contrast, none of the line 29 transgenic Pcmt1−/− mice died at less than 52 days of age (Fig. 2). Line 27 and most of the line 29 mice possessing one or two copies of the endogenous Pcmt1 gene were indistinguishable from comparable nontransgenic mice in size, weight, and behavior, although 15 of 94 line 29 mice ran rapidly in circles. Line 27 and line 29 mice appeared to have normal physiological functions and had unremarkable tissue histology. The transgenic Pcmt1−/− mice, however, differed from the nontransgenic Pcmt1−/− animals in several ways. First, although nontransgenic Pcmt1−/− mice weighed significantly less than age- and sex-matchedPcmt1+/− and Pcmt1+/+ littermates (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar), the weights of transgenic Pcmt1−/− mice were identical to their Pcmt1+/− and Pcmt1+/+ littermates (data not shown). Second, due to their low grade seizure activity (e.g. facial grooming and myoclonic jerks), the nontransgenic Pcmt1−/− mice could often be distinguished by observation from wild-type and heterozygous littermates. In contrast, these abnormalities were not observed in the transgenicPcmt1−/− mice. 2Nonfatal running/jumping seizures have been observed in three transgenic Pcmt1+/− andPcmt1+/+ mice but never in nontransgenic animals.Finally, nontransgenic Pcmt1−/− mice of either sex never produced litters, even when housed with Pcmt1+/+ animals and given the anti-seizure drugs valproic acid and clonazepam, and only a single mating was observed (20Kim E. Lowenson J.D. Clarke S. Young S.G. J. Biol. Chem. 1999; 274: 20671-20678Abstract Full Text Full Text PDF PubMed Scopus (56) Google Scholar). Two line 27 Pcmt1−/−animals, however, produced two small litters without the administration of any drug treatments, and 17 pairings involving male and/or female line 29 Pcmt1−/− mice have produced three litters (from two different Pcmt1−/− mothers). The Pcmt1 transgene controlled by a neuron-specific promoter appeared to rescue, at least partially, the early death phenotype seen in mice lacking the endogenous methyltransferase. We next investigated where in the mouse, and at what level, the transgene was being expressed by assaying methyltransferase activity in various mouse tissues. As expected, transgenic mice possessing one or two copies of the endogenousPcmt1 gene expressed the methyltransferase in all tissues assayed, including brain, heart, testes, erythrocytes, liver, and kidney, at levels similar to those observed in nontransgenic mice (Table I) (18Kim E. Lowenson J.D. MacLaren D.C. Clarke S. Young S.G. Proc. Natl. Acad. Sci. U. S. A. 1997; 94: 6132-6137Crossref PubMed Scopus (248) Google Scholar). In contrast, transgenicPcmt1−/− mice expressed methyltransferase activity in the brain but not in the other tissues (Table I), and Western blot analysis detected PCMT1 protein only in brain homogenates (data not shown), suggesting that the neuron-specific enolase promoter was properly directing expression to neurons. However, the activity of this transgene-derived methyltransferase in the brain was relatively low; line 27 Pcmt1−/− brains had only 6.5% and line 29Pcmt1−/− brains only 13% of the PCMT1 activity observed in wild-type brains. The amount of activity in brains from young (50 days) and older (370 days) transgenic Pcmt1−/− animals was not significantly different (data not shown). Since about half of the cells in the brain are not neurons (27Kandel E.R. Kandel E.R. Schwartz J.H. Jessell T.M. Principles Of Neural Science. 3rd Ed. Appleton & Lange, East Norwalk, CT1991: 22Google Scholar), some reduction of methyltransferase activity was expected in these brains. The low activity observed here, however, suggested that even in neurons the expression of the transgene was weaker than that of endogenousPcmt1 in wild-type neurons. Relatively low levels of neuronal expression from the neuron-specific enolase promoter have been reported by other investigators (28Forss-Petter S. Danielson P.E. Catsicas S. Battenberg E. Price J. Nerenberg M. Sutcliffe J.G. Neuron. 1990; 5: 187-197Abstract Full Text PDF PubMed Scopus (328) Google Scholar, 29Andra K. Abramowski D. Duke M. Probst A. Wiederhold K.H. Burki K. Goedert M. Sommer B. Staufenbiel M. Neurobiol. Aging. 1996; 17: 183-190Crossref PubMed Scopus (78) Google Scholar).Table Il-Isoaspartyl (d-aspartyl) O-methyltransferase activity in mice possessing the Pcmt1 transgene with a neuron-specific enolase promoterTissueGenotype of the endogenous methyltransferase genePcmt1−/−Pcmt1+/−Pcmt1+/+pmol methyl groups transferred to ovalbumin/min/mg proteinLine 27Brain1.22 ± 0.51 (12)8.25 (1)17.4 (1)Heart0.010 ± 0.001 (4)3.50 ± 2.50 (2)11.4 (1)Testis0.011 ± 0.005 (4)9.03 ± 5.77 (2)20.4 (1)Erythrocyte0.001 ± 0.000 (4)0.50 ± 0.04 (2)1.37 (1)Liver0.016 ± 0.005 (4)0.431 (1)1.1 (1)Kidney0.023 ± 0.016 (4)0.69 ± 0.43 (2)2.13 (1)Line 29Brain2.53 ± 0.87 (16)10.6 ± 0.7 (5)19.2 ± 0.5 (3)Heart0.022 ± 0.013 (4)5.5 ± 2.1 (3)11.0 ± 5.0 (3)Testis0.012 ± 0.006 (4)6.6 ± 0.6 (3)20.0 ± 3.4 (3)Erythrocyte0.002 ± 0.001 (4)0.60 ± 0.05 (3)1.3 ± 0.1 (3)Liver0.023 ± 0.008 (4)0.51 ± 0.07 (3)1.1 ± 0.1 (3)Kidney0.031 ± 0.006 (4)0.84 ± 0.11 (3)1.7 ± 0.2 (2)Methyltransferase specific activities were determined as described under "Experimental Procedures" in duplicate or triplicate in each tissue for the number of mice given in parentheses. The value of standard deviation here represents differences between individual mice; the variation of the assay within a single tissue extract was generally very low (<5%). Open table in a new tab Methyltransferase specific activities were determined as described under "Experimental Procedures" in duplicate or triplicate in each tissue for the number of mice given in parentheses. The value of standard deviation here represents differences between individual mice; the variation of the assay within a single tissue extract was generally very low (<5%). The finding that transgenic mice expressing methyltransferase solely in neurons lived longer than nontransgenic knockout mice led us to compare the accumulation of damaged aspartyl residues in the brains of these animals. Recombinant human methyltransferase was used to label these residues in cytosolic/microsomal proteins with [14C]methyl groups from [14C]AdoMetin vitro. Examining 39 transgenic and 13 nontransgenicPcmt1−/− mice, we found about 50% fewer damaged aspartyl residues when the transgene was present, indicating that the transgene-derived enzyme was repairing damaged neuronal proteins. However, the transgenic Pcmt1−/− brains still had about 4.5-fold more damaged residues than did Pcmt1+/− andPcmt1+/+ brains (Fig. 3). These damaged residues could be accumulating both in neurons, due to the relatively low methyltransferase activity, and in glia, which should not express the transgene at all. Examination of the levels of aspartyl damage in the brain with respect to age revealed several interesting points. First, these levels increased with age in young animals; 40-day-old and older mice had significantly more damaged residues per mg of protein than did the 13–21-day-old mice (p = 0.001 forPcmt1−/− mice; p = 10–8 forPcmt1+/− and Pcmt1+/+ mice; Fig. 3). Second, the 13–14-day-old Pcmt1−/− animals already possessed about 8-fold more damaged aspartyl residues per mg protein than did age-matched Pcmt1+/− and Pcmt1+/+ animals (Fig.3). The difference in the amount of damage remained 8–11-fold as these mice aged to 91 days, demonstrating that both nursing and weanedPcmt1−/− mice accumulate these residues. Finally, there was no significant increase in the level of damaged aspartyl residues in transgenic Pcmt1−/− mice after about 100 days of age (Fig. 3). In addition, although line 29 Pcmt1−/− mice averaged twice as much brain PCMT1 activity as did comparable line 27 mice, the plateau level of damaged aspartyl residues in these two lines was not significantly different. The quantity of damaged residues inPcmt1+/− and Pcmt1+/+ mice also attained an apparent steady state, although at a much lower level (Fig. 3). Because damaged residues were arising continuously in the cellular proteins, this stable level of damage probably represents a steady state between new damage, methyltransferase-linked repair, protein turnover, and perhaps unknown factors. In control experiments, we asked whether the low level of damaged residues measured in the assays of Pcmt1+/− andPcmt1+/+ mice resulted in part from the fact that some were already methylated by the endogenous enzyme. We therefore incubated brain cytosolic proteins under basic conditions wherel-isoaspartyl α-methyl esters should hydrolyze within a few minutes to generate l-isoaspartyl residues with about an 80% yield (30Stephenson R.C. Clarke S. J. Biol. Chem. 1989; 264: 6164-6170Abstract Full Text PDF PubMed Google Scholar, 31Patel K. Borchardt R.T. Pharm. Res. ( N. Y. ). 1990; 7: 703-711Crossref PubMed Scopus (243) Google Scholar). Analysis of these samples with recombinant methyltransferase (data not shown) indicated that the true number ofl-isoaspartyl residues in the Pcmt1+/− andPcmt1+/+ proteins could be as much as 2.4-fold higher than the data shown in Fig. 3. 3When damaged aspartyl residues in brain proteins from Pcmt1−/− mice were quantitated as a function of time of base treatment, a linear increase of 1.8 pmol of damaged residues/mg of protein/min was obtained. This increase resulted from the creation of new damaged residues in the proteins. In contrast, quantitation of damaged residues in brain proteins from Pcmt1+/+ mice gave a biphasic increase with time of base treatment. For the first
Referência(s)