Formate Assimilation: The Metabolic Architecture of Natural and

Subscriber access provided by University of Sussex Library

Current Topic/Perspective

Formate assimilation: The metabolic architecture of natural and synthetic pathways Arren Bar-Even Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.6b00495 • Publication Date (Web): 27 Jun 2016 Downloaded from http://pubs.acs.org on July 2, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Formate assimilation: The metabolic architecture of natural and synthetic pathways

Arren Bar-Even Max Planck Institute of Molecular Plant Physiology, Am Mühlenberg 1, 14476 Potsdam-Golm, Germany Phone: +49 331 567-8910 Email: [email protected]

† A.B.-E. is funded by the Max Planck Society.

1

ACS Paragon Plus Environment

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abbreviations FTL, formate-tetrahydrofolate ligase; THF, tetrahydrofolate; PFL, pyruvate formate-lyase; PKT, phosphoketolase.

Textual footnotes *

Throughout this manuscript I use ∆rG’m instead of ∆rG’0; the former corresponds to the reaction change in Gibbs energy under the more physiologically relevant reactant concentrations of 1 mM.

2


Page 2 of 41

Page 3 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Abstract Formate may become an ideal mediator between the physicochemical and the biological realms, as it can be produced efficiently from multiple available sources, such as electricity and biomass, and serve as one of the simplest organic compounds to provide both carbon and energy to living cells. However, limiting the realization of formate as a microbial feedstock is the low diversity of formate-fixing enzymes and thereby the low number of naturally-occurring formate-assimilation pathways. Here, the natural enzymes and pathways supporting formate assimilation are presented and discussed together with proposed synthetic routes that could enable growth on formate via existing as well as novel formate-fixing reactions. By considering such synthetic routes the diversity of metabolic solutions for formate assimilation can be expanded dramatically, such that different host organisms, cultivation conditions and desired products could be matched with the most suitable pathway. Astute application of old and new formate-assimilation pathways may thus become a cornerstone in the development of sustainable strategies for microbial production of value-added chemicals.

3


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

1. Introduction Formate is the product of CO2 reduction by a single pair of electrons, making it one of the simplest organic compounds that can provide cells with both carbon and reducing power. Studying the routes by which formate can be assimilated to biomass is therefore a study of how cells can make a living from the simplest of resources. Formate assimilation has both an ecological and a biotechnological significance. From an ecological perspective, formate is a common fermentation product of larger, more complicated carbon sources, such as sugars. As such, it tends to accumulate in environments lacking a terminal electron acceptor and could serve as a substrate for downstream metabolic processes such as acetogenesis (1), methanogenesis (2), and hydrogen production (3). In fact, it is becoming clear that formate plays an important role in anaerobic syntrophic associations, serving as a redox currency that is transferred between one microbe to another (4). From a biotechnological perspective, formate might prove itself to be of no less significance, as it can be produced using various simple routes that start from highly available resources. These include electrochemical reduction of CO2 (5-10), photoreduction of CO2 (11), hydrogenation of CO2 (12, 13), selective oxidation of biomass (14-16), partial oxidation of natural gas (17), and hydration of syngas (i.e., carbon monoxide) (18). Hence, formate could become an ideal mediator between the physicochemical and biological realms, produced efficiently from our major industrial feedstocks and serving as a sole source for microbial growth and production of value-added chemicals (19). In this manuscript I describe in detail the natural formate-fixing reactions and the formateassimilation pathways associated with them. In addition, I put forward synthetic metabolic routes that use both existing and novel formate-fixing reactions and could support innovative strategies to achieve microbial growth on formate. These synthetic pathways might prove especially interesting for biotechnological applications, as they could enable efficient formate assimilation under specific, finely-tuned conditions. Though not further discussed, I note that in order to provide the cell with reducing power and energy, most formate-assimilation pathways should be accompanied by an oxidation route, either of formate itself (via formate dehydrogenase) or of a pathway intermediate. The properties of the different pathways discussed here are summarized in Table 1.

2. Natural formate-fixing reactions There are >70 known metabolic reactions in which formate participates as a reactant (according to the KEGG database (20)). However, in almost all of these reactions formate serves either as a product of an irreversible carbon–carbon bond cleavage reaction (e.g., GTP cyclohydrolase II) or as a substrate of a reaction that does not lead, directly or indirectly, to the formation of a 4


Page 4 of 41

Page 5 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

new carbon–carbon bond (e.g., formate:quinone oxidoreductase). In fact, formate-fixing reactions are quite scarce. The best known entry-point of formate into cellular metabolism is the enzyme formatetetrahydrofolate ligase (FTL). FTL catalyzes two sequential reactions, namely a formate kinase reaction that gives rise to formyl-phosphate, followed by formylation of tetrahydrofolate (THF) by formyl-phosphate to give formyl-THF (21). While FTL does not directly generate a carboncarbon bond, it activates the formate, making it a good electrophile for downstream reactions with a nucleophilic carbon. FTL plays a significant physiological role in many organisms, some of which use formate as sole carbon source (Section 3 below), while others utilize it only as an intermediate (e.g., in many eukaryotes formate released in the mitochondria by the reverse activity of FTL is re-assimilated by a cytoplasmic FTL, directly contributing one-carbon units for biosynthetic reactions (22)). Another formate-fixing reaction is catalyzed by pyruvate formate-lyase (PFL). This enzyme is mostly known to support pyruvate cleavage, enabling the production of an extra ATP molecule during anaerobic sugar fermentation (23). However, as indicated by a ∆rG’m* value of ≈-21 kJ/mol (6≤pH≤8, 0≤I≤0.25 M (24)), the reaction is expected to be fully reversible under physiological conditions: considering that cellular metabolite concentrations mostly range between 1 µM and 10 mM (25, 26), the PFL reaction is expected to fluctuate in a range of -70 ≤ ∆rG’ ≤ +25 kJ/mol. While the in vitro reversibility of PFL has been demonstrated long ago (2729), only very recently was the enzyme shown to support the condensation of acetyl-CoA and formate in vivo, thereby supporting efficient growth of Escherichia coli on acetate and formate (30). To activate formate condensation, PFL employs a radical mechanism in which a single electron is extracted from formate by an oxygen-sensitive cysteine-radical at the active site. The resulting formyl-radical then attacks an enzyme-bound acetate moiety to generate a pyruvyl-radical which is released as pyruvate (29, 31, 32). Though not tested experimentally, it is highly likely that 2-ketobutyrate formate-lyase, catalyzing a reaction very similar to that of PFL (33), is also reversible in vivo, supporting the condensation of propionyl-CoA with formate to give 2-ketobutyrate. Phosphoribosylglycinamide formyltransferase (PurT) and 5-formaminoimidazole-4-carboxamide ribonucleotide synthase (PurP) are two enzymes that participate in purine biosynthesis in some organisms (34, 35). Both activate formate via phosphorylation and then catalyze a nucleophilic attack on the formyl-phosphate that results in the incorporation of the formate moiety into the synthesized

purine

backbone.

The

PurT

reaction,

followed

by

a

THF-dependent

phosphoribosylglycinamide formyltransferase (PurN (36)) operating in the reverse direction, *

Throughout this manuscript I use ∆rG’m instead of ∆rG’0; the former corresponds to the reaction change in Gibbs energy under the more physiologically relevant reactant concentrations of 1 mM. 5


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

could in fact substitute for the activity of FTL, supporting the ATP-dependent synthesis of formyl-THF from formate and THF. However, since, with the exception of this particular route, PurT and PurP cannot lead to assimilation of formate as a sole carbon source, they will not be discussed further in this manuscript. Finally, the enzyme lactate aldolase (also known as lactate synthase) was suggested to operate in rat’s liver and catalyze the favorable condensation of formate with acetaldehyde to give lactate (37) (∆rG’m ≈ -6 kJ/mol, 6≤pH≤8, 0≤I≤0.25 M (24)). However, as the evidence for lactate aldolase activity is quite scarce, the existence of this enzyme is unclear and as such will only be considered in the context of novel formate-fixing reactions (Section 5). Overall, when compared to carboxylation reactions, which are catalyzed by >20 different enzymes (38), formate-fixing reactions are rare. While a possible explanation for this might lie in a lower selection for the evolution of formate-fixing reactions, another reason could be the relatively low reactivity of formate: (i) Formate is a relatively strong acid (pKa = 3.75 (39)) and hence it is normally deprotonated, preventing it from serving as an electrophile. Even if formate is kept protonated, it is a much poorer electrophile than CO2, since formic acid’s OH makes a poor leaving group. Also, as formic acid is a stronger acid than most other carboxylic acids (e.g., for acetic acid pKa=4.75 (39)), formate’s carbon is a poorer electrophile than the carbon of these carboxylic acids. To address this challenge and activate formate for nueclophilic attack, enzymes such as FTL, PurT, and PurP invest ATP to generate a formyl-phosphate intermediate in which phosphate serves as a leaving group. (ii) Developing a negative charge on formate’s carbon is a challenging task, which limits its ability to serve as a nucleophile. Most other carboxylic acids (usually when activated as acylCoA) solve this problem by making the adjacent carbon nucleophilic; however, formate has no adjacent carbon. As discussed in Section 5, a thiamine-dependent mechanism has the potential to activate formate as a nucleophile. However, due to its low pKa, keeping formate in a protonated form for thiamine’s attack and activation is challenging. Alternatively, PFL addresses the difficulty of formate’s low reactivity using a radical mechanism that bypasses electrophile/nucleophile considerations.

3. Natural formate-assimilation pathways There are two main strategies by which microbes can grow on formate as sole carbon source. One strategy involves formate being oxidized to CO2, where the extracted reducing power supports carbon fixation and serves to provide the cell with energy (40). Due to its low reduction potential (-430 ≤ Eo’ ≤ -380 mV at 6≤pH≤8 and 0≤I≤0.25 M (24)), formate is an 6


Page 6 of 41

Page 7 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

ideal electron donor for reducing cellular redox carriers. The two carbon-fixation pathways known to support formatotrophic growth by the full oxidation of formate are the ATP-costly reductive pentose phosphate cycle (i.e., Calvin-Benson-Bassham Cycle) and the methanogenic version of the highly ATP-efficient but oxygen-sensitive reductive acetyl-CoA pathway (i.e., Wood-Ljungdahl pathway), where CO2 is directly reduced and activated to formylmethanofuran without a free formate intermediate (2, 25, 41, 42). The second strategy by which microbes use formate as a sole carbon source involves the condensation of formate with another metabolic intermediate (although part of it might still be oxidized to provide the cell with reducing power and energy). FTL is the only formate-fixing reaction that is known to support such growth, and in all natural formate-assimilation pathways formyl-THF is first reduced to generate the active intermediate methylene-THF. In the quite ATP-costly serine cycle (i.e., serine pathway (40), Figure 1 and Table 1) methylene-THF transfers

its

C1-moiety

to

glycine

to

generate

serine.

Serine

is

converted

to

phosphoenolpyruvate (PEP), carboxylated to oxaloacetate and reduced to malate. Malate is then activated by condensation with coenzyme A and cleaved to glyoxylate, which regenerates glycine, and acetyl-CoA, which serves as a biomass precursor. Depending on the organism, subsequent assimilation of acetyl-CoA takes place either via the common glyoxylate shunt or through the more ATP-efficient ethylmalonyl-CoA pathway (43, 44). In the acetogenic version of the reductive acetyl-CoA pathway (Figure 2), methylene-THF is further reduced to methyl-THF, which transfers its methyl moiety to a corrinoid iron–sulfur protein to react with CO2 and CoA within the CO-dehydrogenase-acetyl-CoA-synthase complex, resulting in the production of acetyl-CoA (1). Acetyl-CoA is assimilated into central metabolism via its carboxylation by a ferredoxin-dependent pyruvate synthase (45). This pathway not only avoids ATP consumption (46), but actually supports energy conservation via multiple electron bifurcation mechanisms and proton/sodium pumping complexes (47). Hence, organisms assimilating formate via this oxygen sensitive route do not require an external electron acceptor (other than the co-assimilated CO2). The reductive glycine pathway, which operates in various amino-acid- and purine-degrading microbes (48-50), provides an alternative formate-consuming acetogenic route (Figure 3). However, unlike the reductive acetyl-CoA pathway, the reductive glycine pathway is not known to support growth on formate. Rather, the pathway serves to dissipate excess reducing power and support efficient one-carbon and glycine metabolism. In this route, the reversible glycine cleavage system, operating here in the reductive direction (51-54), condenses the C1-moiety of methylene-THF with CO2 and ammonia to produce glycine. Glycine has two possible fates, depending on the organism and the availability of selenium in the medium. If selenium is available, most (but not all) of the organisms that use the pathway utilize the oxygen-sensitive

7


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 41

selenoenzyme glycine reductase to reduce glycine to acetyl-phosphate. Acetyl-phosphate is further metabolized to acetate and ATP, thereby recouping the ATP-investment by FTL and making the entire pathway ATP neutral. Other organisms use a less ATP-efficient, but oxygentolerant route (Table 1) in which glycine reacts with another methylene-THF to produce serine, which is then deaminated to pyruvate that can be further oxidized to acetate.

4. Novel pathways using existing formate-fixing reactions Known formate-fixing reactions could potentially support growth on formate via routes other than the natural formate-assimilation pathways. The simplest example is the reductive glycine pathway. While the pathway does not support growth on formate naturally, it could be established as a route for growth on formate as sole carbon source (40). Beside higher ATPefficiency, the advantage of this pathway over the serine cycle is its lower overlap with central metabolism, which makes regulation a much easier task (Table 1). On the other hand, an engineered reductive glycine pathway is less ATP-efficient than the reductive acetyl-CoA pathway because it lacks the energetic benefits of electron bifurcation mechanisms. Yet, the advantage of the serine variant of the reductive glycine pathway over the reductive acetyl-CoA pathway is that it is completely oxygen tolerant. Other FTL-dependent synthetic pathways could be derived from variations of the serine cycle, according to the general structure: (glycine + formate -> serine -> PEP/pyruvate) + CO2 -> C4 -> glycine + acetyl-CoA. For example, instead of being converted to glycerate, serine could be deaminated to pyruvate and carboxylated to oxaloacetate. Also, malyl-CoA synthetase and lyase could be replaced by oxaloacetase (55), cleaving oxaloacetate to oxalate and acetate, whereby the former could be reduced, via an oxalyl-CoA intermediate, to glyoxylate and the latter activated to acetyl-CoA. A more interesting bypass involves diverting high flux from oxaloacetate to threonine (56), which is then cleaved to regenerate glycine and produce acetylCoA (57, 58). Though longer and more ATP-costly than the natural serine cycle (Table 1), this route could be supported by endogenous enzymes that exist in most microbes. E. coli, for example, could potentially operate a variant of the serine cycle composed of only natural enzymes (besides a foreign FTL), as shown in Figure 4. While PFL is not known to naturally support growth on formate as a sole carbon source, formate-assimilation pathways could be established around this enzyme. The activity of these pathways, however, will be limited to anaerobic or microaerobic conditions, as PFL is oxygensensitive (59, 60). Interestingly, based on the general configuration of serine cycle and its variants, a family of PFL-dependent routes could be suggested, according to the general structure: (acetyl-CoA + formate -> pyruvate/PEP) + CO2 ->

C4

-> acetyl-CoA +

glyoxylate/glycine, where glyoxylate or glycine serves as biomass precursors (see, for example, Figure 5). An especially interesting option is an autocatalytic bicycle that integrates FTL and PFL 8


Page 9 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

to offer an ATP-efficient variant of the serine cycle, in which the product of the FTL-dependent cycle, acetyl-CoA, is condensed with formate by PFL, resulting in the production of pyruvate that is fed back to the (bi)cycle (Figure 6 and Table 1). FTL and PFL could be integrated in other ways to support efficient formate assimilation. For example, PFL could potentially replace pyruvate synthase in the FTL-based reductive acetyl-CoA pathway. This pathway variant might hold a significant thermodynamic as well as kinetic advantage if the microbe is cultivated under high formate concentration but limiting CO2 concentration. Indeed, the genomes of several acetogenic bacteria encode for a protein predicted to be a PFL ortholog (e.g., Acetobacterium woodii, Clostridium ljungdahlii, Clostridium aceticum, Clostridium autoethanogenum, and Clostridium ultunense). Furthermore, the FTLdependent reductive glycine pathway could be significantly upgraded if acetyl-CoA, produced via the glycine reductase pathway variant, is further metabolized to pyruvate by PFL. The resulting pathway represents the highest possible ATP-efficiency for a derivative of the reductive glycine pathway (Table 1, omitting possible contribution from electron bifurcation). As both the reductive acetyl-CoA pathway and the glycine-reductase-variant of the reductive glycine pathway are limited to anaerobic conditions, the use of PFL does not add further constraints to their activity. All formate-assimilation pathways described so far involve CO2 fixation, which might impose constraints on the concentrations of inorganic carbon in the medium in order to sustain optimal pathway activity. Therefore, establishing carboxylation-independent formate assimilation might prove beneficial for some biotechnological applications. This could be achieved via PFL in a pathway that utilizes the enzyme phosphoketolase (PKT) (61, 62). In the autocatalytic PFL-PKT cycle (Figure 7), acetyl-CoA is condensed with formate to give pyruvate, which is then metabolized by the gluconeogenic and pentose-phosphate pathways to produce sugarphosphates. Xylulose 5-phosphate (or fructose 6-phosphate) serves as a substrate for PKT, which cleaves it to glyceraldehyde 3-phosphate (or erythrose 4-phosphate) and acetylphosphate. Acetyl-phosphate is then converted to acetyl-CoA, thus completing the cycle. Finally, 2-ketobutyrate formate-lyase might be able to support growth on formate by its condensation with propionyl-CoA. Here, the metabolic options to achieve net formate assimilation are rather limited. The only potential route (Figure 8) involves the reductive carboxylation of 2-ketobutyrate to 3-methylmalate by 3-methylmalate:NAD+ oxidoreductase (63, 64); as the malic enzyme can support the in vivo reductive carboxylation of pyruvate to malate (65-69), so is this enzyme expected to support the formation of 3-methylmalate from 2ketobutyrate. A CoA-transferase enzyme could then generate 3-methylmalyl-CoA that is cleaved by 3-methylmalyl-CoA lyase (70) to propionyl-CoA and glyoxylate; the former regenerating the cycle’s substrate and the latter serving as biomass precursor. While this cycle

9


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

can be regarded as rather esoteric, and would probably depend on high CO2 concentration to support the reductive carboxylation reaction, its main advantage is that it almost does not overlap with central metabolism and could thus operate as a separate module with minimal regulation requirements.

5. Novel formate-fixing reactions Given the combinatorial nature of metabolic pathways, the addition of only one novel reaction (i.e., a reaction that is not known to be catalyzed by a naturally occurring enzyme) can dramatically expand the space of possible routes. This is especially true for formate-assimilation pathways, as the availability of naturally occurring formate-fixing reactions is so limited. In this section, I explore the possibility of evolving novel metabolic reactions to support efficient formate fixation. Broadly speaking, there are three types of pathways that could be established using such novel reactions: pathways that proceed via formate reduction to formaldehyde, pathways based on reactions in which formyl-CoA replaces acetyl-CoA, and pathways that rely on a thiamine-dependent activation of formate or formyl-CoA as nucleophiles. 5.1 Pathways that proceed via formate reduction to formaldehyde Unlike formate, formaldehyde is a highly reactive compound that could be assimilated to central metabolism rather easily. Hence, one interesting strategy to assimilate formate is to reduce it to formaldehyde and proceed via formaldehyde assimilation. However, formate reduction to formaldehyde is not known to occur in nature, and thus requires enzyme evolution. As is the case for other carboxylic acids (25), the reduction potential of formate to formaldehyde is very low, -650 ≤ E’o ≤ -450 mV (6≤pH≤8, 0≤I≤0.25 M (24)). Therefore, formate must be activated if the universal electron carrier NAD(P)H (physiological -370≤E’≤-280mV) is to be used as an electron donor. There are two major ways to activate formate. First, formate can be ligated with CoA to produce formyl-CoA. Acetyl-CoA synthetase, e.g., from E. coli (71) or Pyrobaculum aerophilum (72), can catalyze this reaction. While the reduction of formyl-CoA to formaldehyde could be supported by an acetylating acetaldehyde dehydrogenase, all enzyme variants tested so far are very slow (71, 73, 74). Hence, enzyme engineering is required to establish high reduction rate. As an alternative, formate can be activated by a kinase enzyme to formylphosphate. Acetate kinase can catalyze this reaction quite efficiently (75). However, the reduction of formyl-phosphate to formaldehyde requires enzyme engineering, probably starting from other phosphorylating aldehyde dehydrogenases, such as glyceraldehyde 3-phosphate dehydrogenase and aspartate-semialdehyde dehydrogenase. Although it was previously suggested that formaldehyde could be produced from spontaneous cleavage of methylene-THF, the kinetics of this reaction seem to be too slow to support a physiologically relevant formaldehyde production rate (40, 76, 77).

10


Page 10 of 41

Page 11 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Once produced, formaldehyde could be assimilated to central metabolism via either a natural or a synthetic route. The naturally occurring ribulose monophosphate cycle (Figure 9) and dihydroxyacetone cycle serve as good candidates for formaldehyde assimilation (78). As an alternative, a recent study demonstrated a synthetic route (Figure 10) that utilizes an evolved enzyme to catalyze a formolase reaction in which three formaldehyde molecules are condensed to form dihydroxyacetone (71). At low formaldehyde concentration, formaldehyde is instead condensed to give glycolaldehyde, which can then be oxidized to glyoxylate and further assimilated via the natural glycerate pathway (79) (Figure 10). However, the rate of both of these

formaldehyde-condensation

reactions

is

still

too

low

to

enable

growth

on

formate/formaldehyde (71, 79). A major challenge common to all possible formate-assimilation pathways that use formaldehyde as an intermediate is the very high toxicity of this compound, limiting cell growth even at low concentrations (80). Therefore, to operate such a metabolic route safely, the concentration of formaldehyde must be kept very low, which is both difficult to achieve and could also lead to kinetic limitation of the enzymes utilizing formaldehyde as a substrate. This problem becomes even more acute in the case of the formolase reaction, as it has double- or triple-dependence on the concentration of formaldehyde. Furthermore, cells have efficient formaldehyde detoxification mechanisms (e.g., glutathione). These mechanisms would have to be deleted to avoid competition with the formaldehyde assimilation routes, which in turn might lead to an even higher sensitivity towards accumulation of formaldehyde. 5.2 Formyl-CoA replacing acetyl-CoA Another general strategy for formate-fixation takes advantage of the structural similarity between formyl-CoA and acetyl-CoA. As formate can be activated to formyl-CoA (71, 72), some reactions that condense acetyl-CoA with another compound could potentially be evolved to accept formyl-CoA instead, thus giving rise to novel formate-assimilation pathways. For example, the enzyme glycine C-acetyltransferase (i.e., 2-amino-3-oxobutyrate CoA ligase) catalyzes the reversible cleavage of 2-amino-3-oxobutyrate to glycine and acetyl-CoA (mechanisms suggested in (58, 81)). As the enzyme can also accept propionyl-CoA and butyrylCoA (82), it is plausible that it could catalyze or evolve to catalyze the condensation of formylCoA with glycine to give aminomalonate semialdehyde, in spite of challenging thermodynamics (+30 < ∆rG’m < +37 kJ/mol, 6≤pH≤8, 0≤I≤0.25 M (24)). The resulting aminomalonate semialdehyde could then be favorably reduced to serine by numerous enzymes (83-86), including E. coli’s threonine dehydrogenase (83) and the gene product of YdfG (85). Notably, the formate-dependent synthesis of aminomalonate semialdehyde and its subsequent reduction to serine serve as an alternative, THF-independent route for converting glycine and formate into

11


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 41

serine. Therefore, all pathway variants of the serine cycle could potentially be converted to use the aminomalonate semialdehyde bypass instead (see schematics in Figure 12). In another example, acetyl-CoA C-acyltransferase condenses two acetyl-CoA moieties to give acetoacetyl-CoA. As this enzyme can accept acceptor substrates other than acetyl-CoA (87-91), it could potentially be evolved to catalyze the condensation of acetyl-CoA with formyl-CoA to give 3-oxopropionyl-CoA, in spite of quite unfavorable thermodynamics (+33 < ∆rG’m < +35 kJ/mol, 6≤pH≤8, 0≤I≤0.25 M (24)). 3-oxopropionyl-CoA could then be favorably reduced by a 3-hydroxyacyl-CoA

dehydrogenase

to

produce

3-hydroxypropionyl-CoA.

The

subsequent

recycling of 3-hydroxypropionyl-CoA to acetyl-CoA, e.g., via the 3-hydroxypropionate cycle, would support the net assimilation of formate (Figure 13). 5.3 Thiamine-dependent activation of formate or formyl-CoA As discussed above, the main challenge in assimilating formate is that it is both a poor electrophile and a poor nucleophile. Thiamine can address this problem by binding formate and activating it as a nucleophile (Figure 14). For example, oxalyl-CoA decarboxylase catalyzes the cleavage of oxalyl-CoA to formyl-CoA and CO2 (92). In the reverse direction, formyl-CoA is activated by thiamine as nucleophile and attacks CO2 to form oxalyl-CoA. While the condensation of formyl-CoA and CO2 is thermodynamically unfavorable (∆rG’m ≈ +37 kJ/mol, pH 7.5, I=0.25M (24)), the reaction might still take place if the partial pressure of CO2 is kept very high and formyl-CoA is synthesized irreversibly such that it accumulates within the cell to a high concentration. Alternatively, oxalyl-CoA decarboxylase might be evolved to accept formyl-CoA (or formyl phosphate) as an electrophile instead of CO2. In this case, the thiamine-activated formyl-CoA would attack the second formyl-CoA (or formyl-phosphate) to release CoA (or phosphate) and generate glyoxyl-CoA. Glyoxyl-CoA could in turn generate glyoxylate that would be assimilated into central metabolism. The recently characterized 2-hydroxyacyl-CoA lyase catalyzes the reversible cleavage of 2hydroxy long-chain fatty acids to formyl-CoA and a long-chain fatty aldehyde (93). In the reverse reaction, thiamine activates formyl-CoA as a nucleophile that attacks an electrophilic aldehyde. This enzyme could potentially be evolved to accept acetaldehyde instead of a longchain fatty aldehyde, thereby condensing formyl-CoA and acetaldehyde to produce lactoyl-CoA in a thermodynamically favorable reaction. Lactoyl-CoA would then be converted to lactate and oxidized to pyruvate, thus entering central metabolism. In an even simpler alternative, thiamine would bind formate directly and catalyze its condensation with acetaldehyde to give lactate (Figure 14). However, such a reaction is challenging as formate must be kept in a protonated form for its condensation with thiamine to occur, and even then it is considerably less susceptible to a nucleophilic attack by thiamine than formyl-CoA. Still, the existence an enzyme condensing formate with acetaldehyde was once proposed in animal liver and was termed 12


Page 13 of 41

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

lactate aldolase or lactate synthase (37). While experimental evidence for this enzyme is quite scarce, the mechanism itself is solid enough and could potentially be evolved from a thiaminedependent enzyme. Notably, such a reaction could serve as an oxygen-tolerant alternative to PFL: instead of directly condensing acetyl-CoA with formate, acetyl-CoA could be reduced to acetaldehyde and condensed with formate in a favorable reaction (∆rG’m ≈ -6 kJ/mol, 6≤pH≤8, 0≤I≤0.25 M (24)). The resulting lactate would then be oxidized to pyruvate using quinones as electron acceptors (∆rG’

Formate Assimilation: The Metabolic Architecture of Natural and

Recommend Documents