Revisiting the glyoxylate cycle: alternate pathways for microbial acetate assimilation

The ability to convert C2 compounds into anapleurotic compounds is a feature shared by plants and certain microorganisms and invertebrates, but conspicuously absent from vertebrates. As a consequence, all of the acetyl-CoA formed by diverse catabolic reactions in vertebrates is committed to oxidation to CO₂ via the citric acid cycle. Under conditions of starvation, acetyl-CoA formed in the liver from fatty acid oxidation is converted to the ketone bodies acetoacetyl-CoA and β-hydroxybutyrate, which are in turn exported to other tissues for reconversion to acetyl-CoA and oxidation to CO₂.

In 1957, the important questions of how some bacteria (initially Escherichia coli) grow using acetate as a sole carbon source, and how germinating seedlings (castor bean) convert fat to carbohydrate, were answered by the seminal studies of Kornberg, Krebs and Beevers (Kornberg and Beevers, 1957; Kornberg and Krebs, 1957). These studies identified two new enzymes, isocitrate lyase and malate synthase, which, in conjunction with reactions of the citric acid cycle, allowed for the net synthesis of anapleurotic succinate from two molecules of acetyl-CoA via a pathway named the glyoxylate cycle (Fig. 1A). Subsequent work showed that isocitrate lyase and malate synthase activities were present in other plants and C2-assimilating bacteria and fungi. However, other studies demonstrated that a number of C2-assimilating bacteria lacked isocitrate lyase activity, as well as the gene encoding isocitrate lyase, suggesting the need for alternative pathway(s) for C2 assimilation in these bacteria (Kornberg and Lascelles, 1960; Albers and Gottschalk, 1976; Gottschal and Kuenen, 1980). Only recently have the details of how bacteria assimilate acetate by alternative pathways begun to be fully understood.

The paper by Alber and coworkers in the current issue of Molecular Microbiology (Alber et al., 2006), together with recent work from other researchers (Korotkova et al., 2002; 2005; Meister et al., 2005), have provided important insights into novel acetyl-CoA assimilation pathways fundamentally different from the glyoxylate cycle. For their studies, Alber and coworkers focused on Rhodobacter sphaeroides, a nutritionally versatile purple non-sulphur photosynthetic bacterium capable of growth with acetate as the sole carbon source, yet known to lack isocitrate lyase. Using an elegant combination of genetics, physiology and biochemistry, the authors reveal a new pathway for acetate assimilation, key features of which are summarized in Fig. 1B. In this pathway, two molecules of acetyl-CoA are condensed to form acetoacetyl-CoA, which undergoes reduction to β-hydroxybutyryl-CoA. β-Hydroxybutyryl-CoA is in turn activated and carboxylated to form the novel intermediate mesaconyl-CoA. Hydration of mesaconyl-CoA yields β-methylmalyl-CoA, which undergoes cleavage to glyoxylate and propionyl-CoA. Condensation of glyoxylate with acetyl-CoA yields malate, while the established reactions of propionate metabolism result in carboxylation and conversion of propionyl-CoA to succinate. Thus, in this new pathway, CoA-activated esters of conventional and novel C3-C5 compounds serve as substrates for a series of condensation, rearrangement and carboxylation reactions that form two C4 gluconeogenicprecursors from three acetyl-CoAand two CO₂ molecules as summarized in Eq. 1:

(1)

Recent studies of acetate assimilation by Korotkova, Chistoserdova and Lidstrom have revealed the details of a distinct pathway for acetate assimilation in the methylotroph Methylobacterium extorquens (Korotkova et al., 2002; 2005). This pathway, named the glyoxylate regeneration cycle, also initiates acetate assimilation via formation of β-hydroxybutyryl-CoA from two acetyl-CoA (Fig. 1C). In this pathway, β-hydroxybutyryl-CoA undergoes dehydration and reduction to butyryl-CoA, which undergoes carboxylation to ethylmalonyl-CoA. A complex series of decarboxylation, rearrangement and carboxylation reactions involving newly discovered enzymes and enzymes of propionate metabolism ultimately leads to the formation of succinate, according to the net reaction shown in Eq. 2 (the same net reaction observed for the classic glyoxylate cycle):

(2)

Succinate formed in this manner is converted to malyl-CoA, which is then cleaved to acetyl-CoA and glyoxylate in a reversal of the malate synthase reaction. The glyoxylate thus formed is used to support the serine cycle for C1 assimilation, crucial to the metabolism of methylotrophs.

The studies discussed above reveal fundamentally new strategies for acetyl-CoA assimilation in which the initial reactions are not directly linked to the reactions of the citric acid cycle, as they are for the classic glyoxylate cycle. The initial reaction, condensation of 2 acetyl-CoA, is the same reaction that leads to ketone body production in mammals. Interestingly, the two new pathways diverge at the level of β-hydroxybutyryl-CoA, but both involve carboxylation reactions yielding novel C5 intermediates (Fig. 1B and C). A further distinguishing feature of the pathway described for R. sphaeroides is that it is a linear pathway with no direct links to a cyclic pathway (citric acid cycle or serine pathway). Thus, this pathway operates completely independently of other pathways in converting C2 units to gluconeogenic precursors.

While a sufficient number of the genes, enzymes and key intermediates of the alternate glyoxylate cycle of R. sphaeroides have been identified to allow the formulation of the overall pathway, the structures of two key intermediates, the substrate for and product of the crucial carboxylation reaction, remain uncharacterized. Likewise, the genes and enzymes for these transformations remain to be identified. These enzymes are likely to be quite novel in terms of their mechanisms of action and should provide interesting insights into strategies for organic substrate carboxylation once they are purified and characterized. Both the alternate glyoxylate pathway of R. sphaeroides and the glyoxylate regeneration cycle of M. extorquens also require complex carbon skeletal rearrangement reactions, presumably involving cobalamin as a cofactor, the details of which should also be quite interesting.

There is evidence that other bacteria, including other purple photosynthetic bacteria, use alternate pathways for acetate assimilation as well. The presence of the mesaconyl-CoA hydratase (mch) gene in the genomes of many of these bacteria suggest that they might use the pathway elucidated here for R. sphaeroides (Alber et al., 2006). Alternatively, some bacteria might use the reactions for C2 assimilation characterized by Korotkova et al. in methylotrophs (Korotkova et al., 2005). Of particular interest in this regard is the phototroph Rhodospirillum rubrum, for which a wholly different pathway of acetate assimilation involving citramalyl-CoA as an intermediate has been proposed (Ivanovsky et al., 1997). Thus, it is at present unclear whether all purple bacteria use the alternate glyoxylate pathway described by Alber and coworkers, the glyoxylate regeneration cycle, or if additional strategies for acetate assimilation exist.

Why certain bacteria use the very complex strategies outlined in Fig. 1B and C for acetate assimilation rather than the two simple transformations of the glyoxylate cycle remains a mystery. There are many possibilities, including the possible need for differential regulation of carbon flux through different metabolic pathways. The identification of these alternative acetate assimilation pathways reveals a surprising diversity in bacterial metabolism of C2 units, and shows that there is much left to learn about bacterial metabolism of central metabolites.