Protein Aging and Repair

As proteins age, they become damaged through the racemization of aspartic and aspariginyl residues. The formation of these isoaspartyls causes the peptide backbone to kink and can result in loss of function. The primary way cells deal with this damage is with a methylation cycle by the Protein Isoaspartyl Methyltransferase (PIMT, or more commonly known as PCMT1 in certain organisms). We are presently determining the biological role of protein methyltransferases that can initiate the conversion of D-aspartyl residues to the L-configuration as well as the conversion of isopeptide linkages to normal peptide bonds. Such repair reactions may greatly increase the useful lifetime of cellular proteins and may help insure organismal survival.

The clear test of the repair hypothesis is whether it functions in vivo. In the last few years, we have taken a genetic approach where we have looked at the consequences of the genetic elimination of the methyltransferase in a variety of organisms. This work has now come to fruition, and we have recently published our work on methyltransferase-deficient bacteria (Visick and Clarke, 1998; Visick et al., 1998), nematode worms (Kagan et al., 1997, Niewmierzycka et al., 1999), and mice (Kim et al., 1997; Kim et al., 1999). We found that both the mutant bacteria and worms were more sensitive to the type of environmental stresses that can damage proteins and reduce life span. However, the most dramatic effects were seen in the repair-enzyme deficient mice where there was a large accumulation of proteins containing L-isoaspartyl residues coupled with a fatal seizure defect! This latter work (done in collaboration with Dr. Stephen Young at the Gladstone Institute in San Francisco) provided the first clear evidence for the operation of the repair pathway in animals and suggested that it was of fundamental importance to complex tissues such as brain. These discoveries have now set the stage for understanding how the accumulation of spontaneous chemical damage to proteins can affect higher mental functions.

Interestingly, in Saccharomyces cereivisiae, there is no homolog of PIMT yet the levels of isoaspartyl damage is shockingly low. We are trying to determine how yeast are able to keep the damage level so low without this essential enzyme. The current theory is that an isoaspartyl-specific metalloprotease may be responsible.


Post Translational Modifications of the Yeast Translational Apparatus:

Over 60% of the known methyltransferases in Saccharomyces cerevisiae methylated a component of the translational apparatus. Elongation factors, release factors, mRNA, tRNA, rRNA, and even the ribosomal proteins all are methylated to varying degrees. The extensive energy and resources that is put into modifying this appratus begs the question “Why methylate?”. We focus on three major areas of translation methylation:

Ribosomal Protein Methylation: a lot of the ribosomal proteins are methylated on several residues, including lysine, arginine, and histidine. Our recent work has uncovered a role for methylation involvement in affecting translation efficiency and proofreading.

Elongation Factor Methylation: yeast have three major elongation factors: EF1A, EF2, and EF3. Between these three protesin alone, there are 10 methylated lysine residues. Elongation factor methyltransferases (EFMs) have been identified for 6 of the sites. We are continuing the search for the other four enzymes and exploring the functional relevance of these modifications.

Mitochondrial Ribosomal Protein Methylation: Just as the cytoplasmic ribosomal proteins are modified, we are currently studying the methylation state of the mitochondrial ribosomal proteins through mass spectrometry.

Protein Arginine Methylation

Our laboratory has also studied and has been focused on the biochemical characterization of the family of Protein Arginine Methyltransferases, a family of 9 seven-beta strand methyltransferases that transfer methyl groups onto arginine residues. PRMTs are known to be involved in modulating transcription, signalling, DNA repair, splicing, and knockdown/overexpression of certain PRMTs has been shown to cause various types of cancer.

Our lab has recently been focused on the characterization of human PRMT7, where we show that it is the only member of the family that is able to produce monomethyl arginine, and it has a rather interesting and specific sequence specificity for RXR basic motifs, found in histone H2B among other proteins.

Our lab has also focused on characterization of PRMT9, the other outlier of the family and the final member of the PRMTs, whose activity and substrate was elusive. Using our biochemical assays, we were able to identify a substrate splicing factor, and classify it as the next symmetrically dimethylating enzyme after PRMT5. In collaboration with Mark Bedford at UT MD Anderson, we were able to link PRMT9 levels in modulating alternative splicing events.