The Tricarboxylic Acid Cycle in Cyanobacteria

See allHide authors and affiliations

Science  16 Dec 2011:
Vol. 334, Issue 6062, pp. 1551-1553
DOI: 10.1126/science.1210858


It is generally accepted that cyanobacteria have an incomplete tricarboxylic acid (TCA) cycle because they lack 2-oxoglutarate dehydrogenase and thus cannot convert 2-oxoglutarate to succinyl–coenzyme A (CoA). Genes encoding a novel 2-oxoglutarate decarboxylase and succinic semialdehyde dehydrogenase were identified in the cyanobacterium Synechococcus sp. PCC 7002. Together, these two enzymes convert 2-oxoglutarate to succinate and thus functionally replace 2-oxoglutarate dehydrogenase and succinyl-CoA synthetase. These genes are present in all cyanobacterial genomes except those of Prochlorococcus and marine Synechococcus species. Closely related genes occur in the genomes of some methanogens and other anaerobic bacteria, which are also thought to have incomplete TCA cycles.

The tricarboxylic acid (TCA) cycle, also known as the Krebs or citric acid cycle, has two functions in bacteria: It oxidizes two-carbon units derived from acetyl–coenzyme A (CoA) producing carbon dioxide and the reduced form of nicotinamide adenine dinucleotide (NADH), which provides the electrons for oxidative phosphorylation, and it provides essential precursor metabolites [e.g., oxaloacetate, 2-oxoglutarate (2-OG), and in some species succinate] that are required for biosynthesis of cellular components. In 1967 two groups reported the failure to detect 2-oxoglutarate dehydrogenase (2-OGDH), which converts 2-oxoglutarate to succinyl-CoA, in various cyanobacteria (1, 2). During the ensuing 44 years, it has become common knowledge that these organisms have an incomplete TCA cycle [see, e.g., comments in (3, 4)]. The absence of 2-OGDH has also frequently been discussed as a contributing factor to explain why most cyanobacteria are obligate photolithoautotrophs (57). Consistent with these initial observations, no fully sequenced cyanobacterial genome encodes the genes for 2-OGDH (7). The incomplete TCA cycle has recently been incorporated into metabolic models for Synechocystis sp. PCC 6803 [e.g., (810)].

Alternatives to 2-oxoglutarate dehydrogenase are known to participate in the TCA cycles of a few organisms, including those of Euglena gracilis mitochondria (11) and Mycobacterium spp. (12). In the latter organism, a thiamine pyrophosphate (TPP)–dependent enzyme, 2-oxoglutarate decarboxylase (2-OGDC), is structurally related to the large subunit of 2-OGDH and produces succinic semialdehyde (SSA). SSA is subsequently oxidized to succinate by succinic semialdehyde dehydrogenase (SSADH). Although cyanobacteria lack homologs of this type of 2-OGDC, Synechococcus sp. PCC 7002 and many other cyanobacteria encode homologs of SSADH. This observation, as well as the fact that mutants of Synechocystis sp. PCC 6803 lacking succinate dehydrogenase still synthesize succinate (13), strongly suggested that cyanobacterial genomes encode a previously unrecognized 2-OGDC. A gene neighborhood analysis revealed that the gene encoding SSADH (SynPCC7002_A2771) occurred as part of an apparent operon comprising two genes in most cyanobacteria, in which one of the genes (SynPCC7002_A2770) encodes an enzyme similar to, but phylogenetically distinct from, acetolactate synthase (IlvB). Because acetolactate synthase is a TPP-dependent enzyme, like 2-OGDH and 2-OGDC (14), it seemed possible that these two open reading frames might encode the enzymes replacing 2-OGDH. An examination of transcription data for these genes under many different growth conditions (15) strengthened this hypothesis, because the transcript levels of these two genes varied coordinately with those for other genes encoding enzymes of the TCA cycle.

Open reading frames SynPCC7002_A2770 and SynPCC7002_A2771 were separately expressed in Escherichia coli, and the resulting soluble proteins were purified (16). When the product of SynPCC7002_A2771 was incubated with SSA, succinate was produced and nicotinamide adenine dinucleotide phosphate (NADP+) was reduced (Fig. 1A). When the products from SynPCC7002_A2770 and SynPCC7002_A2771 were incubated with 2-OG, 2-OG was quantitatively converted to succinate (16) and NADP+ was reduced; however, 2-OG was consumed but no succinate was produced when NADP+ was omitted (Fig. 1B). When the product of SynPCC7002_A2770 was incubated with 2-OG alone, an aldehyde was produced that could be detected using Schiff’s reagent (Fig. 1C). These observations establish that SynPCC7002_A2770 encodes a novel 2-OGDC that produces SSA and SynPCC7002_A2771 encodes SSADH, which jointly replace 2-OGDH and succinyl-CoA synthetase (Fig. 1D).

Fig. 1

(A) High-performance liquid chromatography (HPLC) analysis of SSADH assay components (blue line) showing the formation of succinate (peak 1; see fig. S2) and NADPH (peak 3) from SSA and NADP+ (peak 4) catalyzed by SSADH (SynPCC7002_A2771). Purified SSADH was incubated with 1.5 mM SSA and 2 mM NADP+. The red line shows an HPLC analysis of 2 mM NADPH in distilled water. The green and magenta lines show the analyses of control assays lacking enzyme and NADP+, respectively. Rel. Abs., relative absorbance at 210 nm. (B) HPLC analysis of reaction components from the complete, coupled assay catalyzed by both 2-OGDC (SynPCC7002_A2770) and SSADH (blue line) and showing the conversion of 2-OG (peak 2; see fig. S2) into succinate (peak 1) and the conversion of NADP+ (peak 4) into NADPH (peak 3). The assay mixture contained 2.0 mM 2-OG and 1.0 mM NADP+. The red and magenta lines show the HPLC analyses of control assays lacking enzyme and NADP+, respectively. (C) 2-OGDC activity and detection of SSA by Schiff’s reagent. Cuvette 1 contains the complete assay mixture for 2-OGDC (16); cuvette 2, control with no enzyme added; cuvette 3, control with no 2-OG added. (D) The complete TCA cycle in Synechococcus sp. PCC 7002 showing the missing 2-OGDH enzyme (blue ×) and the 2-OGDC and SSADH shunt that converts 2-OG to succinate via SSA (red arrows).

Peptides from SynPCC7002_A2770 and SynPCC7002_A2771 were detected in the soluble proteome of Synechococcus sp. PCC 7002 (17). To investigate the function of these two enzymes in Synechococcus sp. PCC 7002, we constructed deletion mutants lacking either SynPCC7002_A2770 or SynPCC7002_A2771 (16). A deletion mutant for SynPCC7002_A1094, encoding the SdhB iron-sulfur subunit of succinate dehydrogenase, was also constructed. This sdhB mutant strain was transformed with plasmids designed to overexpress the genes for 2-OGDC or SSADH (16). Figure 2A shows the results of enzymatic assays for SSADH activity. SSA oxidation activity was present in whole-cell extracts of wild-type cells, but no activity was detected in the SynPCC7002_A2771 deletion mutant. The coupled assay for the conversion of 2-OG to succinate in the presence of NADP+ was used to demonstrate the activity of 2-OGDC. Whole-cell extracts of wild-type cells exhibited this activity, but no activity was detected in extracts prepared from either of the deletion mutants (Fig. 2B). These data establish that both 2-OGDC and SSADH are present and active in wild-type Synechococcus sp. PCC 7002 cells grown under standard photoautotrophic conditions.

Fig. 2

(A) Activity assay for SSADH in whole-cell extracts of wild-type (WT) Synechococcus sp. PCC 7002 cells at two extract concentrations (30 μg and 60 μg of protein) and for a deletion mutant 71D lacking SynPCC7002_A2771 (30 μg of protein). (B) Coupled activity assay for 2-OGDC and SSADH, which convert 2-OG to succinate in the presence of NADP+. Whole-cell extracts of wild-type Synechococcus sp. PCC 7002 cells were assayed at two extract concentrations (20 and 60 μg of protein). Extracts for deletion mutants 70D (lacking SynPCC7002_A2770) and 71D were also assayed for 20 μg of protein. Activity was monitored as NADPH production by absorption increase at 340 nm. The values plotted are the average of three independent assays; error bars indicate SE.

To assess the impact of the deletion mutations on growth, we grew cells under constant light or a 12-hour light:12-hour dark regime. During constant illumination, mutants lacking 2-OGDC and SSADH grew at a slower rate than the wild type (Fig. 3 and fig. S4); similar results were obtained when cells were grown on a light-dark cycle (fig. S4B). Relative to the wild type (doubling time = 4.05 ± 0.3 hours), the growth rate of the mutant lacking SSADH was about 17% slower (doubling time = 4.77 ± 0.13 hours), whereas that of the mutant lacking 2-OGDC was about 30% slower (doubling time = 5.23 ± 0.15 hours). A growth rate difference of 30% implies that, starting from equal cell numbers and allowing for 30 generations for the wild type (~120 hours), the mutant cells would represent only ~0.5% of the population. The sdhB mutant had a growth rate that was indistinguishable from that of the wild type (doubling time = 4.09 ± 0.25 hours). This result is similar to that obtained with a Synechocystis sp. PCC 6803 mutant lacking succinate dehydrogenase, which also grew photoautotrophically with a doubling time similar to that of the wild type (13).

Fig. 3

Growth curves for Synechococcus sp. PCC 7002 wild-type and mutant strains. Cells were grown in medium A+ (containing 10 mM nitrate as nitrogen source) under a constant irradiance of ~280 μmol photons m−2 s−1 at 38°C with sparging with 1% (v/v) CO2 in air [standard photoautotrophic growth conditions for this cyanobacterium (15)]. sdhB, sdhB deletion mutant; 70D, SynPCC7002_A2770 deletion mutant; 71D, SynPCC7002_A2771 deletion mutant; 70OVER, sdhB mutant harboring pAQ1Ex-PcpcBA::SynPCC7002_A2770 to overproduce 2-OGDC; 71OVER, sdhB mutant harboring pAQ1Ex-PcpcBA::SynPCC7002_A2771 to overproduce SSADH. The plotted data are the average values obtained from three independent experiments; error bars indicate SE. A log-linear plot of these data is shown in fig. S4.

Strains overproducing SynPCC7002_A2770 and SynPCC7002_A2771 in the sdhB mutant background were also studied in constant light (Fig. 3). The largest effect was observed for a mutant overproducing 2-OGDC, which grew much slower than the wild-type strain (doubling time ≈ 9 hours). Similar results were obtained when these strains were grown on a light-dark cycle (fig. S4B). However, the overproduction strains were unstable, and evidence for more rapidly growing suppressor mutants was readily observed when the overproduction strains were streaked on plates. Because the variants grew faster, the growth rates reported here probably underestimate the effects of overexpressing these genes. 2-OG is one of the most important metabolites in cyanobacterial cells and is important for ammonia assimilation, for regulation of nitrogen and carbon metabolism, and as a precursor metabolite for heme and chlorophyll biosynthesis (18, 19). Thus, it was not surprising that overproduction of an enzyme that can lower the intracellular levels of this critical metabolite had a severely negative impact on cell growth rate. It is also possible that SSA could accumulate and cause some toxicity effects in this mutant. Overproduction of SSADH had a noticeable but much smaller effect on cell growth (Fig. 3). In this case, SSA production presumably is still restricted by the amount of 2-OGDC in cells.

Collectively, these data demonstrate that Synechococcus sp. PCC 7002 has two enzymes that replace the activities of 2-OGDH and succinyl-CoA synthetase in the TCA cycle. The replacement enzymes produce NADPH (the reduced form of NADP+) rather than NADH and do not synthesize guanosine triphosphate by substrate-level phosphorylation (Fig. 1D). Although mutants lacking 2-OGDH and SSADH grew well under laboratory conditions as pure cultures, they would not survive for long in competition with the wild type. Nonetheless, our observations are consistent with the ability of cyanobacteria to produce energy phototrophically rather than by respiration (except during periods of darkness), and suggest that the primary function of the TCA cycle is to produce precursor metabolites for growth. Database searches showed that homologs of SynPCC7002_A2770 and SynPCC7002_A2771 occur in all cyanobacterial genomes except those of Prochlorococcus and marine Synechococcus spp., and also occur in many other bacteria, including some anaerobes (e.g., Methanosarcina and Clostridium spp.) that have been reported to have incomplete TCA cycles due to the absence of 2-OGDH.

The data presented here correct a misconception about the completeness of the cyanobacterial TCA cycle that has persisted for more than four decades. This study illustrates how the misinterpretation of negative results, whether the misannotation of a gene (20) or the failure to detect an enzyme by assay (1, 2), can have a powerful, long-lasting impact on a field.

Supporting Online Material

Materials and Methods

Figs. S1 to S4

Table S1


References and Notes

  1. See supporting material on Science Online.
  2. Acknowledgments: We thank W. R. Cannon, L. A. McCue, A. Konopka, S. Stolyar, A. Beliaev, J. K. Fredrickson, and M. Romine for encouragement and stimulating discussions. D.A.B. thanks W. Martin for long ago pointing out the unusual TCA cycle in E. gracilis. Supported by the Air Force Office of Scientific Research (MURI grant FA9550-05-1-0365) and by the Genomic Science Program of the U.S. Department of Energy, Office of Biological and Environmental Research. This contribution originates from the GSP Foundational Scientific Focus Area and Biofuels Scientific Focus Area of the Pacific Northwest National Laboratory.
View Abstract

Navigate This Article