Sulfate radicals enable a non-enzymatic Krebs cycle precursor

Abstract

The evolutionary origins of the tricarboxylic acid cycle (TCA), or Krebs cycle, are so far unclear. Despite a few years ago, the existence of a simple non-enzymatic Krebs-cycle catalyst has been dismissed 'as an appeal to magic', citrate and other intermediates have meanwhile been discovered on a carbonaceous meteorite and do interconvert non-enzymatically. To identify the non-enzymatic Krebs cycle catalyst, we used combinatorial, quantitative high-throughput metabolomics to systematically screen iron and sulfate reaction milieus that orient on Archean sediment constituents. TCA cycle intermediates are found stable in water and in the presence of most iron and sulfate species, including simple iron-sulfate minerals. However, we report that TCA intermediates undergo 24 interconversion reactions in the presence of sulfate radicals that form from peroxydisulfate. The non-enzymatic reactions critically cover a topology as present in the Krebs cycle, the glyoxylate shunt and the succinic semialdehyde pathways. Assembled in a chemical network, the reactions achieve more than ninety percent carbon recovery. Our results show that a non-enzymatic precursor for the Krebs cycle is biologically sensible, efficient, and forms spontaneously in the presence of sulfate radicals.

Figure 1. TCA intermediates are stable in water and show some reactivity in the presence transition metals frequent in Archean sediments.

A) Reaction scheme of the TCA cycle, including a canonical topology for the Krebs cycle the glyoxylate shunt and the succinic semialdehyde pathway. B) TCA intermediates are considerably stable in 70°C water; no reactivity is detected within 5 hours, except for oxaloacetate which forms pyruvate (Supplementary Figure 2). C) An Archean sediment-like transition metal mixture increases reactivity. Isocitrate forms α-ketoglutarate, succinate and pyruvate, while α-ketoglutarate forms succinate (Supplementary Figure 3). D) Reactions as identified (C) projected to a TCA cycle graph. α-hydroxyl and α-keto moieties that allow specific interaction with ferrous iron are indicated in red.

Figure 2. Peroxydisulfate enables the non-enzymatic interconversion of TCA intermediates.

A) Combinatorial condition screening: Five TCA intermediates were co-incubated for 0 and 300 min in combinations of eight iron and eleven sulfur sources. B) Significant reactions, expressed as c.p.s. illustrated in a heat matrix (left panel). For each possible reaction, product accumulation and substrate consumption over 300 min were calculated from integrated SRM signals. Most conditions did not reveal significant reactivity (significance threshold z > 1.6; Supplementary Table 3, Supplementary Figures 5-6). Right panel: Ten examples of significant reactions as detected upon a combination of peroxydisulfate with ferrous sulfide. left: 0 min, right: 300 min.

Figure 3. Non-enzymatic Krebs-cycle like reactions in the presence of peroxydisulfate and peroxydisulfate/ferrous sulfide.

A) Non-enzymatic TCA reactivity (relative reaction rates, normalized so that they can be compared) in the presence of peroxydisulfate and/or ferrous sulfide. Not detected (n.d.) reactions are depicted in grey (Supplementary Table 4 and Supplementary Figure 7 for rate data). B) Schematic illustration of the enzyme-catalysed Krebs cycle (black), glyoxylate shunt (orange), and succinic semialdehyde pathway (red). C) Non-enzymatic TCA-like reactions are highly efficient and replicate large parts of the reaction spectrum of the TCA cycle, the glyoxylate shunt and the succinic semialdehyde pathways. Non-enzymatic reactions were coloured according to whether they replicate the Krebs cycle (black), the glyoxylate shunt (orange), or the succinic semialdehyde route (red), as in (B). Circle diagrams illustrate the efficiency, expressed as TCA metabolite recovery as substrate (blue), newly formed TCA intermediate (red), or carbon loss (grey) indicative of the formation of non-TCA intermediates. The inner circles reflect peroxydisulfate, the outer circles the combination of peroxydisulfate and ferrous sulfide. Please note that for citrate quantifications have a higher technical variability in the combination of peroxydisulfate and ferrous sulfide, see Methods for details.

Figure 4. Peroxydisulfate enables TCA-like reactivity by providing sulfate radicals.

A) Comparison of sulfate and hydroxyl radical donors on TCA-like non-enzymatic reactions. Reaction rates were determined in the presence of hydrogen peroxide (H₂O₂) and are given relative to the respective rate in the presence of peroxydisulfate ((NH₄)S₂O₈). H₂O₂ enables a subset of reactions that are in average 91.1% slower. Insert: Total absolute cumulative reactivity comparing hydrogen peroxide, peroxydisulfate and control (water). Data is given as mean ± SD, n=3. B) Left panel: Differential scavenging capacities of 2-propanol and tert-butanol allow discriminating between reactivity mediated by sulfate or hydroxyl radicals. Right panel: The reactivity mediated by peroxydisulfate is quenched by the sulfate radical scavenger. Data is given as mean ± SD, n=3 (Supplementary Table 5 for rate data). C) The effects of 2-propanol and tert-butanol on three representative non-enzymatic reactions (black dots represent the presence of a particular scavenger). For the reaction isocitrate to α-ketoglutarate a strong effect of 2-propanol and a medium effect of tert-butanol was measured. The reactions cis-aconitate -> succinate and isocitrate -> succinic semialdehyde were mainly, but to different completion, affected by the sulfate radical scavenger. D) Differential scavenger effects of 2-propanol (red) and tert-butanol (black) confirm the reaction dependency on sulfate radicals. Values were calculated on basis of the ability of scavengers to reduce non-enzymatic reactivity compared to controls as described in (C).