Proteotoxicity and the Contrasting Effects of Oxaloacetate and Glycerol on Caenorhabditis elegans Life Span: A Role for Methylglyoxal?

Introduction

A ccumulation of altered proteins is the most common biochemical signal of aging. Indeed it is possible that such accumulations, or proteotoxicity, might play causal roles in aging, especially because agents that influence cellular proteostasis by either improving or compromising polypeptide metabolism can also affect life span. Obvious examples are decreased ability to recognize and catabolize altered polypeptides, due to insufficient autophagic and proteasomal dysfunction, which decrease life span. Survival is improved by increased expression of chaperone protein/stress proteins and proteolytic function, raising cellular ability to recognize and eliminate potentially proteotoxic agents. Similarly, factors that alter generation of altered proteins (due to biosynthetic errors and postsynthetic damage) also affect aging. Indeed the deleterious effects on proteins of oxygen free radicals and related reactive oxygen species (ROS) are central to the free radical theory of aging. In addition, damage mediated by endogenously generated reactive aldehydes (e.g., malondialdehyde, hydroxynonenal, glyoxal, and methylglyoxal [MG]) provides further sources of proteostatic dysfunction.

It has been evident for some time that excessive glycolysis, as occurring in hyperglycemic conditions, causes a rise in intracellular formation of MG. There is an extensive literature on the deleterious effects of MG on cell, organelle, and molecular dysfunction. ^{1 –3} MG may also, directly or indirectly, induce generation of ROS. ⁴ That MG may influence life span was demonstrated when a raised intracellular level of the enzyme glyoxalase-1, which facilitates MG elimination, was found to increase the life span of the nematode worm Caenorhabditis elegans. ⁵ It follows that intermediary metabolites associated with energy metabolism might affect MG generation, which in turn could also influence onset of proteotoxicity and organism life span.

Two recent papers have shown that addition of glycerol and oxaloacetate (common energy metabolites) to C. elegans growth medium exert opposite effects on organism life span. Addition of glycerol to C. elegans growth medium was found to decrease life span. ⁶ Furthermore, when excess glucose (5%) was added to the growth medium, life span was similarly suppressed, apparently due to glycerol accumulation. Although life span suppression by glucose in C. elegans has been reported previously, ⁷ the explanatory mechanisms are presently elusive, but they probably involve changes in energy metabolism mediated in part by effects on signal transduction and gene expression. In contrast, dietary supplementation with oxaloacetate increases life span in C. elegans. ⁸ To explain this effect, the authors hypothesized that oxaloacetate influences the nicotinamide adenine dinucleotide (NAD⁺/NADH) ratio, which in turn could stimulate sirtuin and 5′-adenosine monophosphate (AMP)-activated protein kinase (AMPK) activities. However, the authors also noted that oxaloacetate extended the life span in a C. elegans strain possessing a mutation in the Sir-2 gene, suggesting an alternative, sirtuin-independent, mechanism of life span enhancement.

A Metabolic Explanation

Not withstanding possible involvement of sirtuins on the control of gene expression affecting C. elegans life span (whose discussion is beyond the scope of this article), it is additionally possible that both glycerol and oxaloacetate influence proteotoxicity by affecting formation of altered proteins by, respectively, increasing or decreasing the potential for MG generation.

Spontaneous decomposition of the glycolytic intermediates glyceraldehyde-3-phosphate (G3P) and dihydroxyacetone phosphate (DHAP) is a recognized route of MG formation. Dietary restriction has long been known to extend life span in a variety of organisms, although the precise mechanisms involved remain somewhat obscure. Periods of fasting are amongst the likely consequences of dietary restriction (induced either by decreased calorie intake or by intermittent feeding without any decrease in calorie intake) during which MG generation would be decreased due to lowered glycolytic flux or frequency. ⁹ Interestingly it has been shown that feeding MG-modified protein to dietary restricted mice suppresses the life span–extending effects of caloric restriction. ¹⁰

It has been proposed that MG formation, and hence formation of altered/damaged proteins, is strongly influenced by NAD⁺ availability. ^11,12 This is because metabolism of G3P (a MG precursor) down the glycolytic pathway is mediated by the NAD⁺-dependent enzyme glyceraldehyde-3-phosphate dehydrogenase (GAPDH), and anything that decreases NAD⁺ availability could increase the potential for MG formation due to G3P accumulation. It is proposed that increased glycerol levels increase MG formation by interfering with the glycerol phosphate cycle. During operation of the glycerol phosphate cycle, NAD⁺ is regenerated from the NADH (formed via the action of GAPDH on G3P) by the cytosolic conversion of DHAP to glycerol phosphate performed by glycerol phosphate dehydrogenase, which uses NADH as co-enzyme. The glycerol phosphate then enters the mitochondria to be reconverted back into DHAP using an FAD-linked glycerol phosphate dehydrogenase. The DHAP then diffuses back into the cytosol to continue operation of the cycle (see Fig. 1). Glycerol, however, can be converted into DHAP by the action of two cytosolic enzymes: First, glycerol kinase generates glycerol phosphate, and then glycerol phosphate dehydrogenase produces DHAP from glycerol phosphate using NAD⁺ as a hydrogen acceptor. But the latter reaction occurs in the reverse direction of the cytosolic component of the glycerol phosphate cycle; not only is NAD⁺ consumed but also DHAP, a potential source of MG, is generated. Therefore it is suggested that the presence of additional glycerol in the C. elegans growth medium inhibits the operation of the glycerol phosphate cycle, decreases NAD⁺ availability, causes accumulation of G3P and DHAP, which decompose into MG, and which will in turn promote ROS generation, protein glycation, and mitochondrial dysfunction (i.e., molecular symptoms of aging). That the life span can be shortened by a defect in a gene specifying an aquaporin ⁶ that is responsible for glycerol secretion into the organism's gut, ¹³ while overexpression of the functional aquaporin gene partially eliminates the life span-suppressing effects of glucose and glycerol, ⁶ are observations consistent with the proposed metabolic effects. Indeed by decreasing NAD⁺ availability, the addition of glycerol to the nematode growth medium would not only tend to increase MG formation but also limit sirtuin activity with deleterious effects on mitogenesis and life span.

OPEN IN VIEWER

FIG. 1.

Metabolism of glyceraldehyde 3-phosphate (G3P). G3P is oxidized by glyceraldehyde phosphate dehydrogenase (GADH) to 1,3-diphosphoglycerate, generating nicotinamide adenine dinucleotide (NADH) in the cytosol. This NADH is reoxidized to NAD⁺ either by malate dehydrogenase (MDH), stimulating malate/aspartate antiport to the mitochondrial matrix (which also relies on glutamate-oxaloacetate transaminase [GOT], in both compartments) and subsequent delivery of electrons to Complex I of the electron transport chain (ETC), or else by the cytosolic isoform of glycerol 3-phosphate dehydrogenase (GPDH_c), which thereby creates glycerol phosphate from dihydroxyacetone-phosphate (DHAP). DHAP is then recycled by the mitochondrial inner membrane-associated isoform of GPDH (GPDH_m), which delivers electrons to the ETC via its bound FAD cofactor. The impact of these processes on proteotoxicity is here proposed to be mediated by methylglyoxal (MG), which is a spontaneous breakdown product of both G3P and DHAP.

The beneficial effects of oxaloacetate toward life span may also be explained by effects on MG formation, which is also mediated by changes in NAD⁺ availability. As proposed by Williams et al., the oxaloacetate may be converted to malate by malate dehydrogenase (a component of the malate-oxaloacetate shuttle), which generates NAD⁺ from NADH. ⁸ The increased NAD⁺ availability would facilitate oxidation of G3P by glyceraldehyde phosphate dehydrogenase (GADH) and thereby decrease the potential for MG formation. The malate then enters the mitochondria, where it is oxidized to oxaloacetate, which is transaminated to aspartate, leaves the mitochondria, and is converted back to oxaloacetate by transamination in the cytosol, to continue operation of the shuttle (see Fig. 1). Hence supplementation of the nematode growth medium with oxaloacetate will stimulate shuttle activity, increase NAD⁺ availability, and decrease the potential for MG-induced proteotoxicity.

It is interesting to note that: (1) Operation of the glycerol phosphate cycle and the oxaloacetate-malate shuttle involve mitochondria to help maintain NAD⁺ availability, and (2) NAD⁺ also regulates sirtuins, whose activities have frequently been shown to be necessary for dietary-mediated life span extension via an upregulation of mitogenesis and mitochondrial activity. Hence it is possible that continued maintenance of NAD⁺ availability could be part of a mutually reinforcing virtuous cycle that normally helps to suppress MG formation, proteotoxicity, and aging. Indeed, it has been previously suggested that changes in MG formation could contribute to the beneficial effects of caloric restriction and intermittent feeding protocols on organism life span. ^9,11

Testing the Proposals

The proposal that both glycerol and glucose promote the accumulation of MG in C. elegans can be tested by determining whether their life span–shortening effects can be suppressed by raising glyoxalase-1 activity. ⁵ It should also be possible to determine whether oxaloacetate can ameliorate the deleterious effects of glycerol and glucose or vice versa and affects intracellular levels of MG, glycated proteins, and ROS in C. elegans. The effects of dietary supplementation with naturally occurring MG-scavengers, such as pyridoxamine, ¹⁴ spermidine, ^15,16 and carnosine, ¹⁷ could also be explored. It is interesting to note that the latter two compounds appear to be multifunctional with respect to possible roles in aging amelioration. Recently, it has been shown that spermidine increases longevity in yeast, flies, worms, and human cells as well as inhibits oxidative stress in mice ^18,19 (most probably by additionally inducing autophagy). Carnosine, a putative antioxidant and metal chelator, can also react with MG, ^20,21 inhibit its deleterious effects, ^{22 –27} and exert life span-extending effects, at least in some circumstances. ^{28 –31}

Conclusions

In conclusion, it is suggested that the direct metabolic consequences induced by the addition of either glycerol or oxaloacetate with respect to MG formation should also be considered, in addition to the well-discussed changes to gene expression mediated by sirtuins, target of rapamycin (TOR), and transcription factors that increase mitochondrial function. Changes in intracellular MG levels may help to explain the effects of these two metabolites on proteotoxicity and nematode aging, especially as MG seems to affect C. elegans life span. ^5,32 That MG-modified proteins can suppress the beneficial effects of caloric-restriction on murine life span, ¹⁰ and MG is thought responsible for much of the biochemical effects of hyperglycemia, ³³ which also seems to resemble some aspects of cellular aging, reinforces the idea that metabolically induced changes in MG formation contribute to aging and life span alteration. ^5,9,34 However because aging is multifactorial, it is unlikely that any simplistic explanation, as outlined here, can adequately and completely explain all age-related phenomena. Nonetheless, it is suggested that effects on metabolism that directly impinge upon formation of MG, a source of proteotoxicity, should be considered in addition to the well-characterized effects on intracellular proteolysis, cell signaling pathways, and gene expression, which, of course, MG can also damage. The proposal that MG generation and thereby some aspects of aging are related to NAD⁺ availability is also consistent with the ideas (“the NAD world”) recently outlined by Imai. ^35,36