If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
* This work was supported by National Institutes of Health Grants AG15451, AG18000, and GM26020.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
l-Isoaspartyl (d-aspartyl) O-methyltransferase (PCMT1) is a protein repair enzyme that initiates the conversion of abnormald-aspartyl and l-isoaspartyl residues to the normal l-aspartyl form. In the course of this reaction, PCMT1 converts the methyl donor S-adenosylmethionine (AdoMet) to S-adenosylhomocysteine (AdoHcy). Due to the high level of activity of this enzyme, particularly in the brain, it seemed of interest to investigate whether the lack of PCMT1 activity might alter the concentrations of these small molecules. AdoMet and AdoHcy were measured in mice lacking PCMT1 (Pcmt1−/−), as well as in their heterozygous (Pcmt1+/−) and wild type (Pcmt1+/+) littermates. Higher levels of AdoMet and lower levels of AdoHcy were found in the brains of Pcmt1−/−mice, and to a lesser extent in Pcmt1+/− mice, when compared withPcmt1+/+ mice. In addition, these levels appear to be most significantly altered in the hippocampus of the Pcmt1−/−mice. The changes in the AdoMet/AdoHcy ratio could not be attributed to increases in the activities of methionine adenosyltransferase II orS-adenosylhomocysteine hydrolase in the brain tissue of these mice. Because changes in the AdoMet/AdoHcy ratio could potentially alter the overall excitatory state of the brain, this effect may play a role in the progressive epilepsy seen in thePcmt1−/− mice.
methionine adenosyltransferase II
high performance liquid chromatography
One of the most common forms of protein damage in vitroand in vivo is deamidation and isoaspartyl formation caused by deprotonation of the peptide-bond nitrogen and its attack on the side chain carbonyl carbon of l-asparaginyl orl-aspartyl residues (
). This spontaneous reaction results in the loss of ammonia (from asparaginyl residues) or water (from aspartyl residues) and the formation of the side chain into an intermediate succinimidyl ring. If this ring is hydrolyzed at the α-carbonyl group, the peptide bond is re-routed through the side chain carbonyl resulting in the formation of anl-isoaspartyl residue. The ring is also susceptible to racemization, allowing for the formation of d-aspartyl andd-isoaspartyl residues, as well. These aberrant residues can cause drastic conformational changes in their proteins, resulting in loss of biological activity (
). Recent studies have shown that isoaspartyl formation in specific peptides or proteins may be linked to certain disease processes. For example, the presence of isoaspartyl residues in the β-amyloid peptide contributes to its insolubility, and protein isomerization is elevated in β-amyloid peptides and paired helical filaments purified from Alzheimer's disease brains (
). However, just as cells have mechanisms to combat other forms of damage, they have at least one to restrict the accumulation of isoaspartyl residues. This mechanism involves the enzyme l-isoaspartyl (d-aspartyl)O-methyltransferase (PCMT1).1
PCMT1 functions by transferring a methyl group from the moleculeS-adenosylmethionine (AdoMet) to either the α-carboxyl group of the l-isoaspartyl residue or the β-carboxyl group of the d-aspartyl residue (
). The resulting methyl ester is then susceptible to spontaneous hydrolysis leading to the loss of methanol and the re-formation of the succinimidyl ring. If the intermediate ring structure then hydrolyzes at the β-carbonyl, which typically occurs about 30% of the time at physiological pH, the residue returns to the normal l-aspartyl form (
). However, 70% of the time, the cyclic imide returns to one of the aberrant forms, which can be re-methylated by PCMT1 until the peptide or protein in which it resides is no longer a substrate for the enzyme. Therefore, one side effect of this repair mechanism may be a potentially large consumption of the methyl donor, AdoMet.
When AdoMet donates its methyl group it becomesS-adenosylhomocysteine (AdoHcy). The concentration of AdoMet compared with that of AdoHcy is often used as an index for the activity of the AdoMet-dependent methyl transfer system in living organisms (
). For example, if the AdoMet/AdoHcy ratio were lowered, it might indicate greater AdoMet “consumption” through increased methyltransferase activity. In rat studies, the administration of 3-(3,4-dihydroxyphenyl)-l-alanine (l-dopa) results in a significant increase in the activity of catecholamine-O-methyltransferase causing a reduction in the brain concentration of AdoMet, with a concomitant elevation in the level of AdoHcy, thereby lowering the AdoMet/AdoHcy ratio (
). However, if 3-(3,4-dihydroxyphenyl)-l-alanine continues to be administered, the level of AdoMet returns to normal due to the feedback regulation of methionine adenosyltransferase II (MAT II), the enzyme that produces AdoMet in the brain (
). Another enzyme that is important in maintaining a proper AdoMet/AdoHcy ratio isS-adenosylhomocysteine hydrolase (SAHH). AdoHcy must be kept from accumulating because many methyltransferases have a higher affinity for AdoHcy than AdoMet, making AdoHcy a potent inhibitor of many methylation reactions (
Although the levels of AdoMet and AdoHcy are normally well regulated, there are conditions under which the levels of these small molecules can be altered for prolonged periods, and they are often accompanied by severe physiological complications. For example, deficiencies in folic acid or vitamin B12 can cause a considerable decrease in the AdoMet/AdoHcy ratio (
). These deficiencies lead to severe neurological disorders, such as demyelination of the spinal cord characterized by vacuolar myelopathic lesions. It has been suggested that the demyelination is caused by the defect in methylation (
). In the case of deficiencies in glycine N-methyltranferase, a liver enzyme that is believed to serve as an alternate pathway for the conversion of AdoMet to AdoHcy, there is a significant increase in plasma AdoMet/AdoHcy ratios (
). Because PCMT1 may have the capacity to use relatively large amounts of AdoMet (as suggested above), it seemed of interest to investigate whether the lack of this methyltransferase results in alterations in the AdoMet and AdoHcy levels in the recently developed PCMT1-deficient mouse.
Mice lacking PCMT1 (Pcmt1−/− mice) have been generated independently by two groups in the last 5 years and exhibit the same major phenotype in each case (
). This phenotype is the development of progressive epilepsy that leads to an early death at an average of 6 weeks of age. It has been proposed that the severe neurological manifestations of these mice are due to the accumulation of one or more isoaspartyl-damaged proteins (such as tubulin, calmodulin, or synapsin) (
). Although both theories are plausible, in this study we explore yet another possibility. Specifically, we ask if the deficiency of PCMT1 causes a change in the overall AdoMet methylation flux and whether this may be contributing to the seizure phenotype seen in the Pcmt1−/− mouse.
In rat brain, there is a significant decrease in the AdoMet/AdoHcy ratio between 1 and 4 weeks of age with a more gradual decrease during maturation (
). This trend is consistent with our results fromPcmt1+/+ mice between the ages of 20 and 50 days. Because this time period also appears to be the most critical in the development of fatal seizures in the Pcmt1−/− mice, we were interested in examining the AdoMet/AdoHcy ratio inPcmt1−/− mice between these ages. Our results show that these mice, and to a limited extent their Pcmt1+/−littermates, exhibit a progressive elevation in their AdoMet/AdoHcy ratios when compared with Pcmt1+/+ mice. This elevation can either be interpreted as a decrease in the methylation flux or as an increase in AdoMet production and/or AdoHcy metabolism. However, after measuring the activities of the enzymes that regulate AdoMet production and AdoHcy metabolism in the brains of these mice, they were found to be altered in a way that would actually lead to a decrease in the AdoMet/AdoHcy ratio, not an increase. For example, in the brains of 50-day-old Pcmt1−/− and Pcmt1+/− mice, SAHH activity was either unchanged or slightly lowered, and MAT II activity was significantly reduced, perhaps due to feedback inhibition by the elevated levels of AdoMet. These findings indicate that the increase in the AdoMet/AdoHcy ratio is most likely a result of a decrease in the consumption of AdoMet as opposed to an increase in its production.
One of the objectives of this study was to determine whether there was a direct cause and effect relationship between the lack of PCMT1 and the increase in the AdoMet/AdoHcy ratio. One way we attempted to accomplish this was to see if the AdoMet/AdoHcy ratio was altered in the testes of Pcmt1−/− mice, another tissue in which PCMT1 is highly expressed in Pcmt1+/+ mice. Whereas AdoMet levels were higher in the testes of Pcmt1−/− mice at 40 and 50 days of age, we were unable to find significantly altered AdoMet/AdoHcy ratios for the ages studied here. Another way we attempted to determine whether there was a direct cause and effect relationship was to compare the localization of PCMT1 expression in the Pcmt1+/+ mouse brain with the localization of altered levels of the AdoMet/AdoHcy ratio seen in the Pcmt1−/− mouse brain. When comparing the AdoMet/AdoHcy ratio from Pcmt1−/− mice toPcmt1+/+ mice, the most significant elevation occurs in the hippocampus and cortex, and the lowest alteration occurs in the region of the thalamus. This correlates well with immunolocalization of the enzyme in rat (
) and mouse brain.3 These data suggest that the lack of the PCMT1 enzyme may be the primary cause of the altered AdoMet/AdoHcy ratios in Pcmt1−/− andPcmt1+/− mice; however, the extent of this effect may be limited to brain tissue.
In Fig. 6, we depict a model for the condition of the Pcmt1−/− mouse in which the lack of PCMT1 decreases the overall methylation flux in the brain. This deficit results in a build up of AdoMet, a diminishment of AdoHcy, an accumulation of isoaspartyl-damaged proteins (X-isoAsp), and a potential increase in the methylation of other AdoMet-dependent methyltransferase substrates (X-CH3). Any one of these effects could potentially lower the seizure threshold in the Pcmt1−/−mice. The altered AdoMet/AdoHcy ratio could affect the activity of one or more of the many other AdoMet-dependent methyltransferases, such as those involved in neurotransmitter metabolism, receptor transduction, DNA expression, or phospholipid methylation (
). Aside from altering the activities of other AdoMet-dependent methyltransferases, it was also possible that higher levels of AdoMet could lead to greater synthesis of polyamines through the action of AdoMet decarboxylase. Polyamines have been shown to have excitatory properties when injected intraventricularly into mice (
). These studies demonstrated that AdoHcy administration decreased epileptiform discharges after hippocampal stimulation and decreased the incidence of pentetrazol convulsions. AdoHcy has also been proposed as a candidate ligand for the benzodiazepine receptor based on its capacity to inhibit flunitrazepam binding to the benzodiazepine recognition site of the γ-aminobutyric acid, type A, receptor (
). Probably the most convincing study on the anticonvulsant action of AdoHcy involves the induction of seizures in mice byl-methionine-dl-sulfoximine and the inhibition of these seizures by the co-administration of adenosine and homocysteine thiolactone (
). In these studies, it was determined that the most effective anticonvulsant action of this treatment occurred when cerebral AdoHcy levels were at their highest.
While it is possible that an increase in the AdoMet/AdoHcy ratio may be responsible for lowering the seizure threshold in these mice, there is no evidence that externally administered AdoMet, which does not appear to be able to cross the plasma membrane (
), would produce the same effect. In fact, in a study of the use of AdoMet as an antidepressant for patients with chronic epilepsy, it was found that daily intravenous administration of AdoMet had no adverse effect on seizure frequency (
). Before the generation of the PCMT1-deficient mouse, it was thought that a knockout of this gene in a mammalian system might provide an advanced aging model. Due to the strong expression of the enzyme in the central nervous system, it was also anticipated that the knockout mouse might show signs of advanced neurological aging, such as neuronal degeneration or the formation of neuritic plaques. However, when the mice were finally generated, it was unanticipated how very limited their survival would turn out to be (
). Trying to make a connection between the deficiency of the methyltransferase and the development of fatal epilepsy seen in this short-lived mouse has been a challenge, especially considering the diversity and number of potential substrates for the enzyme. Theories involving the damage or alteration of various proteins, peptides, and small molecule substrates have been proposed (
). However, one previously overlooked substrate was the methyltransferase's own cofactor, AdoMet. From the studies mentioned above, we have determined that the levels of AdoMet and AdoHcy are indeed altered by the lack of PCMT1, although we do not yet know if these alterations are causing the seizures seen in thePcmt1−/− mice or if they are secondary to them. However, if the altered levels of these small molecules is influencing the seizure threshold in the Pcmt1−/− mice, there may be ways to rescue this phenotype. If we are able to prolong their lives, perhaps older Pcmt1−/− mice would exhibit phenotypes directly related to the aging of specific protein substrates, and perhaps even provide a model of advanced aging in the nervous system.
We thank Dr. Carolyn R. Houser (Department of Neurobiology and Brain Research Institute, UCLA) for guidance in brain dissections and helpful discussions. We also thank Dr. S. Harvey Mudd (National Institute of Mental Health) for helpful discussions.