Recombinant methylotrophic microorganism

Technical domain

[0001] The present disclosure concerns a metabolically modified microorganism. The present disclosure also concerns a method of converting reduced one-carbon compounds into carbon metabolites.

Related art

[0002] Biotechnology is a key sector of the 21 ^st century and is expected to expand massively in the coming decades. However, the currently applied biotransformation processes conflict with the production of human and animal foodstuffs, as the production of the used raw materials predominantly relies on the exploitation of agricultural land.

[0003] Therefore, alternative non-food and non-feed sources are required to replace sugar as the main substrate. Reduced one-carbon compounds, also called Cl compounds, are abundant in nature and as by-products of industrial processes. Their high availability, low production cost and high energy density makes them ideal source materials to produce feedstock or commodity chemicals. However, most biotechnologically relevant microorganisms lack functional assimilation pathways to convert reduced Cl compounds into metabolites and biomass.

[0004] Methanol in particular holds promise as an alternative substrate to replace sugars in the biotechnology industry. At ambient temperatures methanol is liquid with high energy storage capacity. It can be produced from carbon dioxide (CO2) or from methane and its use does not compete with food and animal feed production.

[0005] Moreover, the bioconversion of methanol also contributes to the reduction of greenhouse gases, as it can assist in reducing the amount of methane and CO2 in the atmosphere, paving the way for sustainable biotechnological processes.

ETHZ-27-PCT [0006] Organisms capable of using methanol to grow and to build their biomass, referred to as methylotrophs, are abundant in nature, however, their biotechnological application is limited due to the lack of advanced genetic tools for these organisms.

[0007] An alternative to relying on natural methylotrophs for the assimilation of reduced Cl compounds is to enable strains more commonly used in biotechnological processes, such as Escherichia coli (E. coli), to metabolize methanol. The generation of genetically engineered methylotrophs has attracted considerable attention in the past few years.

[0008] Several attempts describe this adaptation of bacterial strains, which are not naturally capable of growing on methanol, into at least partially methylotrophic strains or methanol-dependent strains.

[0009] Kim S. et al., Nat. Chem. Biol. (2020), DOI:10.1038/s41589-020-0473-5 describes a genetically altered E. coli strain, which grows on formate and methanol. The authors combined rational engineering with adaptive laboratory evolution to develop a strain capable of growth on formate via the reductive glycine pathway. Additional heterologous expression of a methanol dehydrogenase rendered this strain capable of growing on methanol, albeit with a high doubling time of about 54 hours.

[0010] WO2018148703 discloses a method for increasing production of a metabolite by a non-naturally occurring methylotroph when grown on a medium comprising methanol. However, the engineered strains described in this document require additional carbon sources, which are not Cl compounds, such as glucose or ribose, to grow on methanol. It is therefore evident, that only a small portion of the bacterial biomass and metabolites of this strain are derived from methanol.

[0011] WO2022015796 describes metabolically modified microorganisms that can grow on an organic Cl carbon source. However, it is doubted that the modified bacterial strain disclosed in this document is indeed capable of growing on methanol as a sole carbon source, since it the strain was evolved in a medium containing 50 mM 4-morpholinepropanesulfonic acid and 4 mM tricine, which may have served as additional carbon sources. The modification of the disclosed strain comprises primarily copy number variations of large parts of the genome. The copy number variations in

ETHZ-27-PCT the evolved strain resulted in genomic instability, i.e. after cultivation of this strain in rich medium lacking methanol the strain exhibited slower growth on minimal medium containing methanol.

[0012] In Keller, P. et al., 2020, Nat. Commun. 11, 5403, https://doi.org/10.1038/s41467-020-19235-5, a recombinant methanol-dependent Escherichia coli strain comprising specific gene deletions and expressing a set of heterologous polypeptides, is described. The strain requires methanol for growth but needs to be provided with further multi-carbon substrates for growth from which the majority of the biomass is formed.

[0013] The studies mentioned here were at least partially successful in generating non- naturally occurring methanol-dependent microorganisms, which are able to assimilate methanol.

[0014] However, each of the strains described in the state of the art is fraught with considerable drawbacks, such as slow growth, assimilation of only a small portion of reduced Cl compounds into biomass, or genomic instability, which impair the effectiveness of conversion of reduced Cl compounds into biomass.

[0015] It is an object of this invention to overcome the shortcomings and limitations of the state of the art.

Short disclosure of the invention

[0016] It is an aim of the present invention to render the conversion of reduced Cl- compounds, such as methanol, into biomass more efficient.

[0017] It is another aim of the present invention to provide a genetically modified methylotrophic microorganism, which is capable of producing biomass and metabolites from reduced Cl compound(s), in particular methanol. Preferably the strain should be fully methylotrophic, i.e. capable of producing all its carbon metabolites exclusively from Cl substrates. Preferably, the stain should be suitable for industrial use.

ETHZ-27-PCT [0018] It is another aim of this invention to provide a genetically modified methylotrophic microorganism with good genomic stability. The strain should be more stable than other non-naturally occurring genetically modified methylotrophs known in the art.

[0019] It is a further aim of this invention to provide an alternative genetically engineered methylotrophic microorganism.

[0020] According to the invention, one or more of these aims are attained by the object of the attached claims, and especially by the independent claims.

[0021] In particular, one or more of these aims are attained by a recombinant methylotrophic microorganism, which grows on a one-carbon (Cl) compounds as its sole source of carbon, wherein at least 50% of the carbon source consists of one or more reduced Cl carbon compounds. The recombinant methylotrophic microorganism does not require any multi-carbon sources for growth.

[0022] The recombinant methylotroph expresses or over-expresses a polypeptide having methanol dehydrogenase activity, such as Mdh, a polypeptide having 3- hexulose-6-phosphate synthase activity, such as Phi, and polypeptide having a 6- phospho 3-hexuloisomerase activity, such as Hps.

[0023] The recombinant methylotroph does either (i) not comprise, (ii) has deletions in genes expressing, (iii) has a reduction in expression of, and/or (iv) has an elimination or a reduction of enzymatic activity of both of the following polypeptides: a polypeptide having triose-phosphate-isomerase activity, such as TpiA, and a polypeptide having glutathione-dependent formaldehyde dehydrogenase activity, such as FrmA.

[0024] Each of the polypeptides having a methanol dehydrogenase activity, a polypeptide having 3-hexulose-6-phosphate synthase activity, or a polypeptide having a 6-phospho 3-hexuloisomerase activity may be expressed or over-expressed

ETHZ-27-PCT heterologously. It is also possible that one or more of these polypeptides are expressed or overexpressed homologously.

[0025] Reduced Cl-compounds consist of methanol or formaldehyde, as well as substrates which are converted into methanol or formaldehyde for biological assimilation, such as methane, dichloroethane, as well as methylated compounds, where the non-methyl moiety is not metabolized, such as sarcosine.

[0026] Heterologous expression refers to the expression of a gene in an organism which does not naturally comprise this gene. Homologous expression refers to expression or over-expression of a gene in an organism naturally comprising the gene.

[0027] When grown on methanol, the biomass and metabolites produced by the recombinant methylotroph should preferably comprise at least 70%, at least 75%, at least 80%, preferably 90% ± 5% of carbon retrieved from reduced Cl compounds, such as methanol. Preferably, the remainder of the carbon converted into biomass and metabolites is derived from another Cl source, for example from CO2.

[0028] The recombinant methylotroph of this invention is capable of growing on carbon sources consisting of Cl compounds. Different from the recombinant methylotrophs described in the prior art, the strain provided by this invention is capable of producing its entire biomass and metabolites solely from Cl compounds, such as methanol and CO2. When methanol is the only compound provided as a carbon source in the growth medium, the strain is preferably able to assimilate CO2 from atmosphere for the production of its metabolites and biomass. No multi-carbon source, i.e. a carbon source with more than one carbon atom, is required for growth.

[0029] The recombinant methylotrophic strain expresses or overexpresses, at least in part heterologously, all enzymes required for a complete ribulose monophosphate (RuMP) cycle, a pathway used for the assimilation of reduced one-carbon compounds in natural methylotrophs. Such enzymes include a polypeptide having methanol dehydrogenase activity, such as Mdh, a polypeptide having 3-hexulose 6-phosphate synthase, such as Hps, a polypeptide having 6-phospho 3-hexuloisomerase activity, such as Phi, a polypeptide having glucose-6-phosphate isomerase activity, such as Pgi, a polypeptide having transketolase activity, such as TktA, a polypeptide having

ETHZ-27-PCT ribulose-phosphate 3-epimerase, such as Rpe, a polypeptide having ribose-5- phosphate isomerase activity, such as RpiA, a polypeptide having transaldolase activity, such as TalA, a polypeptide having NADP+-dependent glucose-6-phosphate dehydrogenase, such as Zwf, a polypeptide having 2-keto-3-deoxygluconate 6- phosphate/2-keto-4-hydroxyglutarate aldolase activity, such as Eda, and a polypeptide having a phosphogluconate dehydratase, such as Edd.

[0030] In E. coli, the heterologous expression of an enzyme having a methanol dehydrogenase activity, such as Mdh, an enzyme having a 3-hexulose 6-phosphate synthase, such as Hps, and an enzyme having a 6-phospho 3-hexuloisomerase activity, such as Phi, enable the influx of Cl compounds into the RuMP cycle and complete the RuMP cycle. When expressing these enzymes, a recombinant E. coli strain can assimilate Cl substrates to some extent.

[0031] However, in E. coli and in other non-naturally occurring methylotrophs, the heterologous expression of an enzyme having a methanol dehydrogenase activity, such as Mdh, an enzyme having a 3-hexulose 6-phosphate synthase, such as Hps, and an enzyme having a 6-phospho 3-hexuloisomerase activity, such as Phi, is insufficient to allow growth of E. coli wild type strains, which are naturally non-methylotrophic, on methanol. This can be attributed to insufficient flux of formaldehyde flux into the RuMP cycle and the imbalanced efflux of RuMP cycle intermediates.

[0032] To overcome this problem, further mutations need to be introduced. Suitable mutations are mutations resulting in disruptions in expression or activity of a polypeptide having glutathione-dependent formaldehyde dehydrogenase, such as FrmA, , and of a polypeptide having triose phosphate isomerase activity, such as TpiA, as mentioned above. The mutations are preferably deletions of the gene encoding glutathione-dependent formaldehyde dehydrogenase, frmA, and the gene encoding the native triose phosphate isomerase, tpiA, of the recombinant microorganism.

[0033] A suitable mutation in or deletion of the native tpiA gene interrupts gluconeogenesis and abolishes growth on multi-carbon compounds including pyruvate, succinate, and/or acetate in the absence of methanol. The strain thereby enables methanol-dependent growth in presence of methanol and the multicarbon source pyruvate.

ETHZ-27-PCT [0034] A suitable mutation in or deletion of the native frmA gene blocks the formaldehyde detoxification pathway, ensuring high levels of formaldehyde, the one- carbon compound that enters the RuMP cycle.

[0035] To further enhance the efficiency of the recombinant methylotroph described above, and/or to achieve a stable assimilation of methanol the strain preferably comprises further mutations. Such mutations may alter the expression or activity of polypeptides coordinating, be it as enzymes or as regulators, the carbon flux through the autocatalytic RuMP cycle.

[0036] The efficiency of Cl assimilation can be increased by additionally modifying the recombinant microorganism, to express or over-express one or more polypeptides selected from the group consisting of Mdh, Hps, Phi, TktA, TktB, Rpe, RpiA, RpiB, TalA, TalB, Edd, and Eda, or any combination thereof. Preferably these polypeptides are expressed from their corresponding native genes. It is however also possible to transform the recombinant organism with one or more of the genes for heterologous expression of the encoded polypeptides.

[0037] The recombinant methylotroph may for example express or over-express a polypeptide having TktA activity and a polypeptide having RpiB activity, or a polypeptide having TktA activity and a polypeptide having Edd activity and/or polypeptide having Eda activity, or a polypeptide having RpiB and a polypeptide having Edd activity and/or polypeptide having Eda activity.

[0038] The recombinant methylotroph may comprise one or more modifications in one or more native genes increasing the enzymatic activity of the polypeptide encoded by the mutated gene. The recombinant strain may for example comprise modifications increasing activity of polypeptides encoded by one or more native genes selected from the group consisting of rpe, rpiB, tktA, and tktB, or any combination thereof.

[0039] The recombinant methylotroph may comprise one or more modifications in one or more native genes decreasing or eliminating the enzymatic activity of the polypeptide encoded by the mutated gene. This modification may for example be a point mutation, a mutation leading to a truncated polypeptide, or a deletion of the gene or portions thereof. The recombinant strain may for example comprise

ETHZ-27-PCT modifications decreasing or eliminating activity of polypeptides encoded by one or more native genes selected from the group consisting of pgi, gnd, gntR, pfkA, pgk, aroF, aroH, prs, purF, aceA, acnA, fumC, sdhB, sucA, sucC, and icd, or any combination thereof.

[0040] The recombinant methylotroph may comprise a modification in the gntR gene decreasing or eliminating the activity of DNA-binding transcriptional repressor GntR. Preferably, the modification prevents the dimerization of the repressor. The modification may, for example, be a mutation resulting in the truncation of the GntR repressor.

[0041] The efficiency of Cl assimilation in the recombinant methylotroph described above can be enhanced by introducing further modifications to reduce or eliminate the expression of one or more polypeptides from native genes selected from the group consisting of gntR, tktB, pgi, pkg, pgi, gnd, aroH, aroF, pfkA, prs, purF, a gene encoding an enzyme of the tricarboxylic acid (TCA) cycle, including aceA, acnA,fumC, sdhB, sucA, sucC, icd, aceE, aceF, sdhA, sucC, and mgo, or any combination thereof.

[0042] The recombinant methylotroph may for example comprise a reduction in expression, or a deletion of the gene encoding the Gnd polypeptide and in the gene encoding the AroH polypeptide. Alternatively, it may for example comprise a reduction in expression of the gene encoding the AcnA polypeptide and the gene encoding the Pgi polypeptide. It may comprise a reduction in the expression of the gene encoding the AcnA polypeptide and the gene encoding the Pgi polypeptide. It may comprise a reduction in the expression of the gene encoding the Pgi polypeptide and the gene encoding the Pgi polypeptide. It may comprise a reduction in the expression of the gene encoding the AcnA polypeptide, the gene encoding the Pgi polypeptide, and the gene the gene encoding the Pgi polypeptide.

[0043] An introduction of a premature stop codon in the gene encoding the Gnd polypeptide has been demonstrated to result in a significant decrease in doubling time of a recombinant methylotroph of this invention, when grown on methanol as sole carbon source. A reduction in expression, or a deletion of the gnd gene in the methylotroph of this invention is therefore particularly advantageous for its growth on Cl compounds.

ETHZ-27-PCT [0044] In addition or alternatively, the recombinant methylotroph may for example comprise a reduction in expression, or a deletion of the gene encoding the PykA and/or the PykF polypeptide. Several nonsense and missense mutations in the pykA and/or pykF genes were identified in various derivatives of a recombinant methylotroph of this invention which showed a faster growth rate on methanol, suggesting that a reduction or elimination of expression of one or both of these genes is beneficial for assimilation of Cl substrates.

[0045] The recombinant methylotroph may comprise a gene encoding a RhIB polypeptide comprising a conversion of the stop codon into a lysine-encoding codon. In E. coli MEcoli_ref_2 having the Culture Collection of Switzerland (CCOS) accession number 2080 this conversion results in expression of a polypeptide having 21 additional amino acids, which is expected to result in a loss of function. The recombinant methylotroph may comprise a reduction in the expression of the gene encoding the RhIB polypeptide, or it may express a non-functional RhIB polypeptide or a RhIB polypeptide with reduced activity compared to wild-type RhIB.

[0046] The recombinant methylotroph may express or over-express an Mdh polypeptide having at least 70%, 80%, 90%, 95%, 98% or 99% sequence identity to SEQ ID NO: 1, wherein the polypeptide preferably comprises at least an H165N mutation.

[0047] It was established as part of this invention, that a Mdh polypeptide comprising this mutation has an improved methanol dehydrogenase activity compared to the parental polypeptide. By introducing an H165N mutation in a Mdh polypeptide, the methanol dehydrogenase activity compared to the parental Mdh polypeptide can be increased by at least 1.5-fold, or at least 2-fold, or at least 2.5-fold.

[0048] Preferably, the non-naturally occurring recombinant methylotroph can tolerate methanol concentration of 1 M. The recombinant methylotroph is preferably able of growing in medium supplemented with up to 1 M M methanol, for example between 100 mM and 2 M methanol, or any value between any of the two foregoing values.

[0049] Preferably the recombinant methylotroph is grown under atmospheric conditions in or on a medium provided with methanol as sole carbon source.

ETHZ-27-PCT [0050] In one aspect of this invention the non-naturally occurring recombinant methylotroph is capable of reaching an optical density at 600 nm (ODeoo) of 2 or more, or of 2.5 or more when grown under atmospheric conditions in a medium comprising only 500 mM methanol as carbon source.

[0051] In one aspect of this invention the non-naturally occurring recombinant methylotroph is capable of growing on methanol at a doubling time of 10 hours or less, preferably of 8 hours or less.

[0052] In one embodiment, the recombinant Mdh polypeptide may further comprise a F279I mutation and/or a V359E mutation. The mutation of these amino acids is thought to contribute to an improved growth rate of the non-naturally occurring recombinant methylotroph.

[0053] In one aspect of this invention the non-naturally occurring recombinant methylotroph may have a doubling time of 5 hours or less, or of 4.5 hours or less.

[0054] The recombinant non-naturally occurring methylotroph may be obtained by genetically modifying a non-naturally occurring methylotrophic microorganism selected from a group consisting of Escherichia, in particular Escherichia coli, Bacillus, in particular Bacillus subtilis, Clostridium, Enterobacter, Klebsiella, Mannheimia, Pseudomonas, in particular, Pseudomonas putida, Acinetobacter, Shewanella, Paracoccus, Ralstonia, for example R. eutropha, Geobacter, Zymomonas, Acetobacter, Geobacillus, Lactococcus, Streptococcus, Lactobacillus, Corynebacterium, in particular Corynebacterium glutamicum, Streptomyces, Proprionibacterium, Synechocystis, Synechococcus, Cyanobacteria, Chlorobi, Deinococcus and Saccharomyces sp.

[0055] The non-naturally occurring recombinant methylotroph of this invention may be a synthetic methylotrophic microorganism.

[0056] The recombinant non-naturally occurring methylotroph may be an Escherichia coli strain. The recombinant non-naturally occurring methylotroph may be the strain E. coli strain designated MEcoli_ref_l having the Culture Collection of Switzerland (CCOS) accession number CCOS 2032.

ETHZ-27-PCT [0057] The recombinant non-naturally occurring methylotroph may be the strain E. coli strain designated MEcoli_ref_2 having the CCOS accession number CCOS 2080.

[0058] The recombinant non-naturally occurring methylotroph may be an E. coli strain having a doubling time and/or product profile which is similar to or substantially the same as the doubling time and/or the product profile of E. coli CCOS deposit having accession number CCOS 2032 or which is is similar to or substantially the same as the doubling time and/or the product profile of E. coli CCOS deposit having accession number CCOS 2080, when grown on methanol.

[0059] The invention also concerns a method for producing or for increasing production of a metabolite, comprising the steps of growing a recombinant non- naturally methylotrophic microorganism described herein on a Cl substrate and obtaining a metabolite.

[0060] The metabolite may for example be obtained from the growth supernatant of the cultured recombinant methylotroph. In addition or alternatively, the metabolite may be obtained from the biomass and/or from lysed recombinant methylotroph.

[0061] The one or more metabolites obtainable by this method will depend on the ancestral microorganism which was modified to generate the recombinant non- naturally occurring methylotroph of this invention. When modified according to this invention, the metabolites produced by these strains are generated by providing Cl compounds as principal or sole source for carbon.

[0062] The metabolites may for example be selected from a group consisting of 2- carbon compounds, 3-carbon compounds, 4-carbon compounds, higher carboxylic acids, alcohols of higher carboxylic acids, polyhydroxyalkanoates, and speciality chemicals, aromatic compounds, vitamins and their derivatives or precursors, and carotenoides.

[0063] The metabolites may for example be selected from a group consisting of formate, lactate, amino acids or sugars, such as deoxyribose.

ETHZ-27-PCT [0064] With respect to what is known in the art, the invention provides the advantage that the recombinant non-naturally methylotrophic microorganism is fully methylotroph, i.e. it uses Cl substrates as its sole source for carbon. It is therefore more efficient than the recombinant methanol-dependent and methylotrophic strains known in the art in converting Cl compounds, such as methanol, into biomass.

[0065] Moreover, a recombinant non-naturally methylotrophic microorganism disclosed according to this invention is also more efficient than previously described recombinant non-naturally methylotrophic microoorganisms in converting Cl compounds into metabolites, which may be industrially relevant metabolites. In addition, since the methylotrophy of the recombinant microorganism does not rely on gene copy number variations of the modified strain, which are notoriously instable, the recombinant methylotroph disclosed herein is also genetically more stable than previously described recombinant methylotrophic microorganisms.

[0066] Growth of the recombinant non-naturally methylotrophic microorganism disclosed according to this invention on methanol is therefore more robust than methanol-dependent growth of previously disclosed recombinant non-naturally methylotrophic strains.

Short description of the drawings

Exemplar embodiments of the invention are disclosed in the description and illustrated by the drawings in which:

Figures 1A and IB are a graphic presentation of the RuMP cycle, illustrating flux variability analysis (FVA)-predicted differences of the flux distribution of the methanol-dependent E. coli strain AfrmAAtpiA when grown on methanol and pyruvate (Figure 1A), and when grown on methanol alone (Figure IB).

Figure 2 represents schematically the adaptation of the methanol-dependent E. coli strain AfrmAAtpiA during evolution; the diagram on the left shows fluxes

ETHZ-27-PCT before long-term laboratory evolution, the diagram on the right shows fluxes following successful long-term laboratory evolution; long-term laboratory evolution is indicated by a dashed arrow;

Figure 3 depicts the culture density achieved by different generations of the evolving E. coli strain AfrmAAtpiA grown on methanol and pyruvate in continuous chemostat culture. Once methylotrophic growth was achieved after about 250 generations of evolution, the optical density increased.

Figures 4A and 4B show growth curves of the evolving E. coli strain AfrmAAtpiA in the absence of pyruvate, in minimal medium without methanol or supplemented with 500 mM methanol.

Figure 4A shows growth of the evolved population after 249 generations in the presence and absence of methanol.

Figure 4B shows a mean of 10 growth curves of the evolved strain MEcoli_ref_l in the presence and absence of methanol.

Figure 5A show genetic and proteome changes of a recombinant methylotrophic E. coli strain MEcoli_ref_l; Proteomics data, indicating the fold change in expression, are indicated left to the corresponding gene.

Figure 5B depicts a comparison of proteomes between the ancestral methanoldependent E. coli strain AfrmAAtpiA and the fully methylotrophic E. coli strain MEcoli_ref_l; the ancestral methanol-dependent E. coli was grown in minimal medium supplemented with 500 mM methanol and 20 mM pyruvate; the methylotrophic strain MEcoli_ref_l was grown in minimal medium containing only 500mM methanol as carbon source; differentially expressed genes (Iog2(fold-change) > 1.5) of important metabolic pathways and the enriched supergroup of DNA repair and replication are indicated with different symbols;

Figures 6A to 6D show methanol incorporation into protein-bound amino acids by the evolved methylotrophic E. coli strain in isotope distributions (Figures 6A and 6C) and in ¹³ C fractional distributions (Figures 6B and 6D);

ETHZ-27-PCT Figures 6A and 6B show MEcoli_ref_l grown in minimal medium supplemented with 500 mM ¹³ C methanol either under ambient CO2;

Figures 6C and 6D show MEcoli_ref_l grown in minimal medium supplemented with 500 mM ¹³ C methanol under 5% (V/V) enriched ¹³ COz atmosphere.

Figures 7A to 7D show methanol incorporation into protein-bound intermediates by the evolved methylotrophic E. coli strain in isotope distributions (Figures 7A and 7C) and in ¹³ C fractional distributions (Figures 7B and 7D);

Figures 7A and 7B show MEcoli_ref_l grown in minimal medium supplemented with 500 mM ¹³ C methanol either under ambient CO2;

Figures 7C and 7D show MEcoli_ref_l grown in minimal medium supplemented with 500 mM ¹³ C methanol under 5% (V/V) enriched ¹³ CC>2 atmosphere.

Figure 8 shows the genomic adaptations of methylotrophic E.coli strains (replicate line D), with

Figure 8A showing growth phenotypes of MEcoli_ref_l and MEcoli_ref_2 in minimal medium supplemented with 500 mM methanol,

Figure 8B showing a Venn diagram of overlap of mutations of adapted populations after 1129 (D), 1124 (E), 1118 (F) and 1125 (G) generations under serial dilution,

Figure 8C show the repeatedly mutated functional units in the context of methanol metabolism, wherein genetic element names are shown next to the associated functional metabolic step represented by an arrow.

Figure 9 shows a series of metabolites in the growth supernatant of the evolved methylotrophic E. coli strain MEcoli_ref_l.

Figures 10A to E shows bioproduction using recombinant methylotrophic E. coli strains of this invention, with

ETHZ-27-PCT Figure 10A depicting methanol metabolism in the recombinant methylotrophic E. coli strains of this invention, with metabolic routes for lactic acid, PHB and itaconic acid production based on heterogeneous expression of the indicated genes in the recombinant methylotrophic MEcoli_ref_2; dotted lines representing multiple reactions;

Figure 10B depicting three major metabolic nodes in the methanol metabolism pathway of the recombinant methylotroph, as well as examples of the derived commodity as well as speciality chemicals;

Figure IOC showing LC-MS chromatograms for MEcoli_ref_2 transformed with pljac producing lactic acid, compared to a negative control and a commercial standard;

Figure 10D showing GC-FID chromatograms for MEcoli_ref_2 transformed with pl_phb producing PHB, compared to a negative control and a commercial standard;

Figure 10E showing LC-MS chromatograms for MEcoli_ref_2 transformed with pljta producing itaconic acid, compared to a negative control and a commercial standard.

Figure 11 shows a comparison of catalytic turnover between ancestral dehydrogenase Mdh and mutated methanol dehydrogenase Mdh(H165N).

Examples of embodiments of the present invention

[0067] A "recombinant microorganism" refers to a microorganism which has been genetically modified to express or over-express endogenous polynucleotides, or to express non-endogenous nucleotide sequences, such as those included in a plasmid, or which has a reduction in expression of an endogenous gene.

ETHZ-27-PCT [0068] A "synthetic microorganism" refers to a recombinant microorganism in which a substantial portion of the genome or the entire genome has been genetically modified.

[0069] A" non-naturally occurring methylotrophic microorganism" refers to a microorganism derived from an ancestral microorganism which lacks the ability of methanol-dependent growth or of fully methylotrophic growth, but through recombinant engineering or recombinant engineering and laboratory evolution has been adapted to full methylotrophic growth.

[0070] The term "recombinant methylotroph" as used herein refers to a non-naturally occurring genetically modified methylotroph.

[0071] A "methylotrophic microorganism" is a microorganism which converts Cl compounds, in particular methanol, into biomass.

[0072] The term "methanol-dependent microorganism" refers to a microorganism which requires methanol as well as other multi-carbon substrates for growth. Even though this microorganism requires methanol for growth, it is unable to grow on Cl compounds as sole carbon source. "Methanol-dependent" therefore indicates the requirement for methanol and at least one multi-carbon compound for growth.

[0073] The term "fully methylotrophic microorganism" is a microorganism which uses exclusively reduced Cl compounds as carbon source. Preferably, the main Cl carbon source for a methylotroph is methanol. "Fully methylotroph" or "full methylotrophy" therefore indicate that only Cl-compounds, in particular methanol, are required as carbon substrates for growth.

[0074] Organisms capable of using methanol as a growth substrate, referred to as methylotrophs, are abundant in nature; however, their biotechnological application is limited due to the lack of advanced genetic tools. An alternative to relying on natural methylotrophs is to enable an already established platform organism, such as Escherichia coli, to metabolize methanol. The generation of synthetic methylotrophs has attracted considerable attention in the past few years and has mainly focused on the introduction of the ribulose monophosphate (RuMP) cycle for carbon assimilation due to its superior efficiency compared to alternative carbon assimilation pathways.

ETHZ-27-PCT [0075] In E. coli, only three genes encoding a methanol dehydrogenase (mdh), a 3- hexulose 6-phosphate synthase (hps), and a 6-phospho 3-hexuloisomerase (phi) are lacking for a complete RuMP cycle. Although the introduction of three enzymes seems straightforward, the implementation of a heterologous metabolic cycle is challenging as it requires the complete rewiring of the central metabolism of E. coli.

[0076] In particular, carbon flux through the synthetic autocatalytic RuMP cycle must be tightly coordinated with its effluxes to achieve stable methanol assimilation.

Rewiring of Central metabolic fluxes

[0077] In previous studies the inventors had generated methanol-dependent strains with a complete RuMP cycle (Keller et al., 2020, https://doi.org/10.1038/s41467-020- 19235-5). A particularly promising strain contained two deletions, a deletion in the triose phosphate isomerase (tpiA) gene and a deletion in the glutathione-dependent formaldehyde dehydrogenase (frmA) gene, abolishing expression of the respective gene products.

[0078] The engineered strain E. coli AfrmAAtpiA was strictly dependent on methanol for growth and formed about 6.5% of its biomass and about 50% of its RuMP cycle metabolites from the one-carbon substrate.

[0079] To establish a fully methylotrophic lifestyle in E. coli AfrmAAtpiA, the strain was transformed to express the genes mdh (SEQ NO ID 1, 2), hps (SEQ NO ID 52, 53), and phi (SEQ NO ID 54, 55) from a heterologous plasmid system, as described in the Materials and Methods section hereinunder. This recombinant strain was subsequently subjected to adaptive laboratory evolution (ALE), as described in more detail below. This recombinant E. coli AfrmAAtpiA strain, expressing mdh, hps, and phi is referred to as ancestral AfrmAAtpiA strain hereinafter.

[0080] Consistent with flux balance analysis (FBA), half of fructose 6-phosphate (F6P) in the ancestral AfrmAAtpiA strain was formed from methanol and dihydroxyacetone phosphate (DHAP) originated exclusively from methanol, while metabolites

ETHZ-27-PCT downstream of glyceraldehyde 3-phosphate (GAP) in glycolysis were generated from pyruvate, the second carbon source that was used as a growth substrate.

[0081] To evaluate the metabolic adaptations required for achieving growth solely on methanol, the central metabolism of the ancestral AfrmAAtpiA strain was modelled by flux variability analysis (FVA), which, in contrast to FBA, allows information regarding the overall solution space that could result in growth on methanol during evolution.

[0082] Figure 1A is flux diagram of the predicted distributions of metabolic flux during growth of the ancestral AfrmAAtpiA strain on methanol and pyruvate as co-substrates. Figure IB depicts predicted metabolic flux during growth on methanol alone based on FVA modelling. Metabolic reactions that were synthetically introduced into the metabolism of the ancestral AfrmAAtpiA strain and the corresponding enzymes are encircled. The thickness of the lines indicates the strength of the flux. Dashed lines indicate zero flux through the reaction. RuMP cycle, Entner-Doudoroff pathway and the tricarboxylis acid (TCA) cycle are indicated.

[0083] Abbreviations on Figure 1 refer to: Mdh, methanol dehydrogenase; Hps, 3- hexulose 6-phosphate synthase; Phi, 6-phospho 3-hexuloisomerase; frmA, S- formylglutathione hydrolase; tpiA, triose phosphate isomerase; CH2O, formaldehyde; H6P, arabino 3-hexulose 6-phosphate; F6P, fructose 6-phosphate; G6P, glucose 6- phosphate; 6PG, 6-phosphogluconate; Fl,6bP, fructose 1,6-bisphosphate; S7P, sedoheptulose 7-phosphate; E4P, erythrose 4-phosphate; R5P, ribose 5-phosphate; Ru5P, ribulose 5-phosphate; Xu5P, xylulose 5-phosphate; GAP, glyceraldehyde 3- phosphate; DHAP, dihydroxyacetone phosphate; l,3bPG, 1,3-bisphosphoglycerate; 3PG, 3-phosphoglycerate; 2PG, 2-phosphoglycerate; PEP, phosphoenolpyruvate.

[0084] To allow for a direct comparison of the fluxes, the target growth rate was set to 0.2 h ¹ in both conditions, a value similar to the growth rate of the native methylotroph Bacillus methanolicus at 37°C. The upper bound for the methanol uptake was set to the minimal amount of methanol uptake required (0.6 mmol gCDW ¹ h ^-1 ) to achieve the target growth rate in the presence of excess amounts of pyruvate. In the case of growth on methanol alone, the methanol uptake rate was left unbounded.

ETHZ-27-PCT [0085] For growth on pyruvate together with methanol, the highest metabolic fluxes occurred through the tricarboxylic acid (TCA) cycle, while only a small fraction of the total fluxes originated from the RuMP cycle. In contrast, the predicted flux distribution during growth on methanol as the sole carbon source was markedly different.

[0086] The fluxes through the TCA cycle were predicted to be reduced to a minimum by a factor of 16, while the fluxes in the RuMP cycle increased about 15-fold.

[0087] Furthermore, the Entner-Doudoroff pathway, which was not active under mixed growth conditions, was predicted to be essential for the synthesis of more oxidized metabolites such as pyruvate during sole growth on a Cl compound, such as methanol. Overall, the altered metabolic fluxes confirmed that a complete restructuring of the central metabolism of E. coli is required to enable growth on methanol.

[0088] To achieve full methylotrophy in the strict sense, a high selection pressure towards more efficient methanol assimilation was applied during the ALE process.

[0089] Different from targeted genetic engineering, ALE is an approach to select for random genetic modifications under selective pressure, which can lead to the enrichment of strains with substantial metabolic changes towards a particular trait.

Long-term evolution of the ancestral E-coli AfrmAAtpiA strain

[0090] The ancestral AfrmAAtpiA strain was evolved under continuous conditions in a chemostat in which the dilution rate of the culture defined its growth rate.

[0091] Figure 2 depicts schematically the targeted metabolic adaptation of the methanol-dependent ancestral AfrmAAtpiA strain towards the formation of almost its entire biomass from pyruvate (indicated in white) to growth on methanol alone, with its biomass entirely formed from Cl substrates (indicated in black). The striped part of the cycle indicates a reduced flux.

[0092] As shown in Figure 3, during the continuous experiment, methanol was present in excess at a concentration of 500 mM and pyruvate at lower concentration of 20 mM

ETHZ-27-PCT in the chemostat feed which resulted in limiting concentrations of pyruvate (< 0.01 mM) in the growth vessel to ensure a high selection pressure towards increased methanol incorporation. The pyruvate concentration was reduced to 5 mM after 98 generations. The feed medium was supplemented with 0.1 mM isopropyl-|3-D- thiogalactopyranosid (IPTG) for heterologous expression of mdh, hps and phi.

[0093] Methanol dehydrogenase 2 (mdh) variant CT4-1 from Cupriavidus necatorwas expressed from heterologously introduced plasmid pSEVA424.

[0094] 3-hexulose 6-phosphate synthase (hps) from Methylobacillusflagellatus and 6- phospho 3-hexuloisomerase (phi) from Methylobacillus flagellates were expressed from heterologously introduced plasmid pSEVA131.

[0095] Experimental boundary conditions, i.e. the concentrations of methanol and pyruvate in the feed medium and the dilution rate of the culture are shown in the top panel of Figure 3 with observations in terms of the density of the culture over time and number of generations shown in alignment underneath the panel. Due to technical issues, the reactor 1 was restarted after 200 generations (reactor 2) and reactor 2 after 223 generations (reactor 3).

[0096] During the first 150 days approximately, which corresponded to 90 ± 5 generations based on the selected dilution rate, the optical density of the culture remained rather stable. Then, to further increase the selection pressure towards a higher methanol uptake, the pyruvate concentration was gradually lowered in the feed medium first to 10 mM and ultimately to 5 mM.

[0097] While the pyruvate concentration in the growth vessel was already limiting with 20 mM of pyruvate in the feed medium, the reduction to 5 mM should lower the amplitude of the pyruvate fluctuations that arise when a droplet of feed medium enters the growth vessel. As expected, the optical density of the culture proportionally decreased by a factor of four after changing the composition of the feed medium.

[0098] A gradual increase in optical density was observed over the subsequent 100 generations. Starting from generation 115 until 223, a steady increase in optical density at 600 nm (ODeoo) from 0.23 to 0.70 was observed, indicating that the

ETHZ-27-PCT population was incorporating more methanol. After another 25 generations and a total of 249 generations, an increase in optical density to 2.5 was observed, indicating another drastic increase in methanol uptake by the population and potentially the feasibility of growth on methanol as the sole carbon and energy source.

Growth on methanol

[0099] The marked increase in yield of the chemostat culture after the long-term evolution of the ancestral AfrmAAtpiA strain indicated that the population might be capable of growing on methanol also in the absence of pyruvate. When the population was inoculated in a shake flask with medium containing only methanol (500 mM) as a carbon source, the culture was indeed able to grow, reaching an optical density ODeoo of about 0.4 at a doubling time (Td) of 60 hours (h), as shown in the growth curve of Figure 4A. In this experiment, growth of the evolved population after 249 generations in the chemostat, in the presence and absence of methanol was examined. The population was grown in minimal medium containing 500 mM methanol, as well as ampicillin and streptomycin.

[00100] At this point of the evolution, IPTG, which had been supplied to induce heterologous expression of mdh, hps and phi from their respective expression vectors, was no longer required for methylotrophic growth.

[00101] To improve the growth performance of the population, it was evolved under serial transfer conditions in medium containing only methanol as carbon source.

[00102] After 534 generations, 4 individual clones were isolated and characterized. Remarkable the growth rate increased about eightfold in all clones tested (Td 7.5 ± 0.8 h), while yields were around ODeoo 2.1.

[00103] Figure 4B shows the growth curve of the best growing isolate, MEcoli_ref_l, grown on methanol under the same conditions. Remarkably, the strain had a doubling time of around eight hours, Td = 8.1 ± 0.4 h, as shown in Figure 5B.

[00104] Recombinant methylotrophic strain MEcoli_ref_l was used for all subsequent experiments unless specified otherwise.

ETHZ-27-PCT Proteome remodelling in methylotrophic E. coli

[00105] To assess the proteome remodelling of recombinant methylotrophic strain MEcoli_ref_l, the proteome of the methanol-dependent ancestral AfrmAAtpiA strain grown on methanol and pyruvate was compared to the proteome of the methylotrophic MEcoli_ref_l grown on methanol alone.

[00106] About 20% of the detected proteins were differentially expressed between the two strains, with 1494 detected proteins, differential expression cutoff | logzffold-change) > 1.51 .

[00107] Upon evolution of the ancestral strain to a fully methylotrophic strain, enzymes of the RuMP cycle, the Entner-Doudoroff pathway as well as methanol dehydrogenase were upregulated, while enzymes of branch point reactions away from the RuMP cycle, pyruvate metabolism and the TCA cycle were downregulated, as indicated in Figure 5A.

[00108] Figure 5A schematically depicts the mutations in core metabolism identified in the methylotrophic population after 249 generations of chemostat evolution. Additional genetic changes found in MEcoli_ref_l are underlined in this Figure. Difference in expression levels (logzffold-change)) based on proteomics data are shown to the left of the corresponding gene name.

[00109] Figure 5B is a comparison chart of differential gene expression, comparing the ancestral AfrmAAtpiA E. coli strain grown in minimal medium supplemented with 500 mM methanol and 20 mM pyruvate to MEcoli_ref_l grown in medium containing only 500 mM methanol as carbon source. Q values are P values, indicating the probability that the observed result occurred by chance, adjusted for multiple hypothesis testing using the Benjamini-Hochberg procedure. Black symbols indicate differentially expressed genes (Iog2(fold-change) > 1.5) of important metabolic pathways and the enriched supergroup of DNA repair and replication. The latter is a superset of the KEGG pathways mismatch repair (eco03430), DNA replication (eco03030) and homologous recombination (eco03440). Results are also summarised in Table 1.

ETHZ-27-PCT [00110] In addition, the proteome data were contextualised by estimating relative abundances of individual proteins, which revealed that methanol dehydrogenase increased in abundance from 16% of the quantifiable proteome in the ancestral AfrmAAtpiA E. coli strain to 40% in MEcoli_ref_l. [00111] To find additional cellular functions with altered expression profiles, gene set enrichment analysis on KEGG pathway annotations were performed. Interestingly, the analysis showed that proteins related to DNA replication, mismatch repair and homologous recombination were overrepresented in the evolved strain (Table 1). [00112] Table 1: Differences in expression of relevant genes in log? fold-change, "logzFC", as well as the pathways in which these genes are implicated, are listed.

ETHZ-27-PCT

ETHZ-27-PCT

Evolutionary trajectory towards full methylotrophy and genetic modifications

[00113] To identify genetic changes acquired during evolution towards full methylotrophy, the ancestral AfrmAAtpiA E. coli strain and the fully methylotrophic reference strain MEcoli_ref_l were sequenced. In addition, the metagenomic composition of the evolving population was determined at regular time intervals to obtain temporal information of the adaptation process.

[00114] After 52 generations the number of observed and fixed mutations, i.e. mutations of an abundance greater than 90% in the population, started to increase. After 249 generations, when the population achieved methylotrophic growth in the chemostat, 564 mutations had fixed in the population.

[00115] Subsequent evolution under serial dilution regime further increased this number to about 1000. Ultimately, clones isolated after 534 generations, the same ones for which growth rates were determined, acquired on average 1155 ± 52 (mean ± standard deviation) mutations. MEcoli_ref_l carried 1231 mutations of which 55% (677 of 1231) were nonsynonymous.

[00116] To detect patterns in the evolution towards full methylotrophy, the sequence in which metabolic adaptations in methanol dehydrogenase, the RuMP cycle, RuMP cycle efflux points and the Entner-Doudoroff pathway arose in the population was further explored.

[00117] Results of this analysis are summarised in Table 2.

ETHZ-27-PCT [00118] Mutations in TCA and RuMP cycle genes dominated the adaptation process during the first 150 generations and were followed by the occurrence of a mutation in methanol dehydrogenase. In the last phase before full methylotrophy was achieved, many mutations in the analyzed metabolic pathways fixed in the population.

[00119] Particularly, the only mutation relevant of the Entner-Doudoroff pathway (gnt/?(E312*)) occurred at this timepoint. Over the course of the serial dilution, another selective sweep occurred, fixing two mutations in RuMP cycle efflux points. Interestingly, despite the higher growth rate of MEcoli_ref_l, the core set of metabolic adaptations as shown in Figure 5A was not expanded drastically over continued evolution under serial dilution regime.

[00120] Besides the core methanol metabolization pathways, a search for other functional classes that were altered during the course of evolution was conducted, in particular for KEGG pathways that were enriched in nonsynonymous mutations in MEcoli_ref_l. No significant hits were identified.

[00121] As shown in Figure 9, several of metabolites were detected in the growth supernatant of MEcoli_ref_l when grown in minimal medium supplied with 500mM of methanol under ambient conditions. Formate and deoxyribose were the most abundant natural metabolites produced and excreted by this strain.

[00122] The impact of the mutation in the key enzyme for full methylotrophy, i.e. methanol dehydrogenase was examined more closely. The mutation H165N in the Mdh CT1-4 polypeptide resulted in an about two-fold increase in its catalytic turnover number (Figure 11).

Table 2: Mutations and Proteomics data of recombinant methylotroph MEcoli_ref_l

ETHZ-27-PCT

ETHZ-27-PCT

The column "Gene" lists the name of the mutated gene of, or the genome location in the ancestral strain BW25113 (Genbank accession: NZ_CP009273). The numbers listed in column "Generation" indicate the number of generations during ALE in which the mutation was fixed. "SEQ ID NO" refers to the sequence identity number of the sequence listing filed separately for this invention.

[00123] The above-listed mutations in the genes rpe, rpiB, tktA and tktB presumably increase the metabolic flux through the RuMP cycle.

[00124] The above-listed mutations in the genes pgi, aroF, aroH, prs, and purF, presumably stabilize the metabolic flux through the RuMP cycle.

ETHZ-27-PCT [00125] The above-listed mutations in the genes pgi, aroF, aroH, prs, and purF, presumably stabilize the metabolic flux through the RuMP cycle.

[00126] The above-listed mutations in the genes ace A, acnA, fumC, sdhB, sucA and icd presumably decrease the flux through the TCA cycle.

[00127] The above-listed mutations in genes encoding the master regulators crp, rpoB and rpoC presumably cause these regulators to remodel the proteome to enable methylotrophic growth.

[00128] The above-listed mutation in the gnd gene results in a truncation of the 6-phosphogluconate dehydrogenase polypeptide, which removes the substrate binding site of the enzyme and therefore presumably abolishes its activity. This prevents flux from the Entner-Doudoroff pathway intermediate 6-phospho-D- gluconate to D-ribulose 5-phosphate under loss of CO2.

[00129] The above-listed mutation in the gntR gene results in a truncation of the DNA-binding transcriptional repressor GntR, removing a part of its effector binding and oligomerization domain. Presumably this prevents dimerization of the repressor and, in turn, repression of the Entner-Doudoroff pathway genes edd, eda. These genes are highly upregulated in MEcoli_ref_l.

[00130] The above-listed mutation in the pfkA gene removes the substrate binding site of the 6-phosphofructokinase 1 enzyme and presumably abolishes its activity. This prevents flux from the RuMP cycle into the metabolic dead end metabolite fructose 1,6-bisphosphate which cannot be further metabolized due to the tpiA deletion in MEcoli_ref_l.

[00131] The above-listed mutation in the pgk gene presumably decreases its activity such that flux coming out of the Entner-Doudoroff pathway is directed towards the TCA cycle and not back towards D-glyceraldehyde 3-phosphate and with that back into the RuMP cycle under loss of ATP.

[00132] The above listed mutation of the gene encoding the Cupriavidus necator N-l methanol dehydrogenase 2 Mdh increases catalytic turnover number of its gene

ETHZ-27-PCT product, methanol dehydrogenase, two-fold. The increased catalytic activity of this enzyme enables faster methanol oxidation.

Proteomic and genetic modifications to achieve full methylotrophy

[00133] To enable growth on methanol, the metabolism of the methanoldependent ancestral AfrmAAtpiA had to be globally rewired to achieve full methylotrophy. The observed changes concerned in methanol oxidation, the RuMP cycle, the Entner-Doudoroff pathway and the TCA cycle.

[00134] The recombinant methylotrophic microorganism of this invention does not express a polypeptide having glutathione-dependent formaldehyde dehydrogenase FrmA activity and a polypeptide having triose phosphate isomerase TpiA activity.

[00135] In addition, the recombinant methylotrophic microorganism of this invention expresses or over-express, when compared to its ancestral non-fully methylotrophic microorganism, polypeptide having methanol dehydrogenase activity, such as Mdh, a polypeptide having 3-hexulose-6-phosphate synthase activity, such as Phi, and polypeptide having a 6-phospho 3-hexuloisomerase activity, such as Hps.

[00136] This invention identifies further genes which, when modified or expressed differently compared to a non-methylotrophic parental strain, may positively contribute to achieving full methylotrophy in a non-naturally occurring methylotroph.

[00137] In one embodiment of this invention a recombinant methylotrophic microorganism expresses or over-expresses, as compared to its ancestral non- methylotrophic microorganism, one or more polypeptides selected from the group consisting of of Mdh, Hps, Phi, TktA, TktB, Rpe, RpiA, RpiB, TalA, TalB, Edd, and Eda, or any combination thereof.

[00138] A recombinant methylotrophic microorganism of this invention may include increased activity of expression of a polypeptide having ribulose-phosphate 3- epimerase (Rpe) activity and which shares at least 40%, 45%, 50%, 55%, 60%, 6%, 70%,

ETHZ-27-PCT 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% (or a value between any two of the foregoing values) or greater sequence identity , or at least 50%, 60%, 70%, 80%, 90%, 95%, 96%, 97%, 98%, 99% (or a value between any two of the foregoing values) or greater sequence similarity, as calculated by NCBI BLAST, using default parameters, to SEQ ID NO 3. Preferably the polypeptide comprises at least a H224N mutation compared to the wild-type Rpe polypeptide.

[00139] A recombinant methylotrophic microorganism of this invention may include increased activity of expression of a polypeptide having Ribose-5-phosphate isomerase B (RpiB) activity and which shares at least 40%, 45%, 50%, 55%, 60%, 6%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% (or a value between any two of the foregoing values) or greater sequence identity , or at least 50%, 60%, 70%, 80%, 90%, 95%, 96%, 97%, 98%, 99% (or a value between any two of the foregoing values) or greater sequence similarity, as calculated by NCBI BLAST, using default parameters, to SEQ ID NO 5. Preferably the polypeptide comprises at least a *150R mutation compared to the wild-type RpiB polypeptide.

[00140] A recombinant methylotrophic microorganism of this invention may include increased activity of expression of a polypeptide having Transketolase 1 (TktA) activity and which shares at least 40%, 45%, 50%, 55%, 60%, 6%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99% (or a value between any two of the foregoing values) or greater sequence identity , or at least 50%, 60%, 70%, 80%, 90%, 95%, 96%, 97%, 98%, 99% (or a value between any two of the foregoing values) or greater sequence similarity, as calculated by NCBI BLAST, using default parameters, to SEQ ID NO 7. Preferably the polypeptide comprises at least a A135E mutation compared to the wildtype TktA polypeptide.

[00141] A recombinant methylotrophic microorganism of this invention may include increased activity of expression of a polypeptide having phosphogluconate dehydratase (Edd) activity.

[00142] A recombinant methylotrophic microorganism of this invention may include increased activity of expression of a polypeptide having 2-keto-3- deoxygluconate 6-phosphate/2-keto-4-hydroxyglutarate aldolase activity (Eda) activity.

ETHZ-27-PCT [00143] A recombinant methylotrophic microorganism of this invention may include increased activity of expression of a polypeptide having TktA activity and a polypeptide having RpiB activity, or of a polypeptide having TktA activity and a polypeptide having Edd activity and/or a polypeptide having Eda activity, or of a polypeptide having RpiB and a polypeptide having Edd activity and/or polypeptide having Eda activity.

[00144] In another or further embodiment, the recombinant methylotroph may comprise one or more modifications in one or more native genes increasing the enzymatic activity of the polypeptide encoded by the mutated gene compared to the polypeptide encoded by the parental gene.

[00145] A recombinant methylotrophic microorganism of this invention may for example comprise modifications increasing activity of polypeptides encoded by one or more native genes selected from the group consisting of rpe, rpiB, tktA, and tktB, or any combination thereof.

[00146] A recombinant methylotrophic microorganism of this invention may have Rpe polypeptide having a sequence identity that is at least 95% to 100% identical to SEQ ID NO 4. Preferably, residue 224 of the Rpe polypeptide is an asparagine.

[00147] A recombinant methylotrophic microorganism of this invention may have RpiB polypeptide having a sequence identity that is at least 95% to 100% identical to SEQ ID NO 6. Preferably, residue 150 of the RpiB polypeptide is an arginine.

[00148] A recombinant methylotrophic microorganism of this invention may have TktA polypeptide having a sequence identity that is at least 95% to 100% identical to SEQ ID NO 8. Preferably, residue 135 of the TkTA polypeptide is a glutamate.

[00149] A recombinant methylotrophic microorganism of this invention may have TktB polypeptide having a sequence identity that is at least 95% to 100% identical to SEQ ID NO 10. Preferably, residue 609 of the TkTB polypeptide is a glutamate.

[00150] In another or further embodiment, a recombinant methylotrophic microorganism of this invention may comprise modifications decreasing or eliminating

ETHZ-27-PCT activity of polypeptides encoded by one or more native genes selected from the group consisting of pgi, gnd, gntR, pfkA, pgk, aroF, aroH, prs, purF, aceA, acnA, fumC, sdhB, sucA, sucC, and icd, or any combination thereof.

[00151] A recombinant methylotrophic microorganism of this invention may comprise a Glucose-6-phosphate isomerase Gpi activity, or a phosphoglycerate kinase Pgk activity, or a 3-deoxy-7-phosphoheptulonate synthase aroF activity, or a 3-deoxy- 7-phosphoheptulonate synthase aroH activity, or a ribose-phosphate diphosphokinase Prs activity, or a amidophosphoribosyltransferase PurF activity, or a isocitrate lyase AceA activity, or a aconitate hydratase A AcnA activity, or a fumarase C FumC activity, or a succinate:quinone oxidoreductase SdhB activity, or a 2-oxoglutarate decarboxylase SucA activity, or a succinyl-CoA synthetase subunit |3 Sue C activity, or a isocitrate dehydrogenase led activity that is 40% or less, for example 20%, 25%, 30%, 35% , 40% (or a value between any two of the foregoing values), of the enzymatic activity of the respective wild-type polypeptide. A recombinant methylotrophic microorganism of this invention may comprise any reduction of activity of any combination of the polypeptides listed here.

[00152] A recombinant methylotrophic microorganism of this invention may comprise a loss of 6-phosphogluconate dehydrogenase Gnd activity, or a 6- phosphogluconate dehydrogenase Gnd activity that is 40% or less, for example 20%, 25%, 30%, 35%, 40% (or a value between any two of the foregoing values), of the enzymatic activity of the respective wild-type polypeptide.

[00153] A recombinant methylotrophic microorganism of this invention may comprise a loss of 6-phosphofructokinase 1 Pfkl activity, or a 6-phosphofructokinase 1 Pfkl activity that is 40% or less, for example 20%, 25%, 30%, 35%, 40% (or a value between any two of the foregoing values), of the enzymatic activity of the respective wild-type polypeptide.

[00154] A recombinant methylotrophic microorganism of this invention may comprise a loss of DNA-binding transcriptional repressor GntR activity, or a DNA- binding transcriptional repressor GntR activity that is 40% or less, for example 20%, 25%, 30%, 35%, 40% (or a value between any two of the foregoing values), of the enzymatic activity of the respective wild-type polypeptide.

ETHZ-27-PCT [00155] The loss of activity of the GntR repressor may be due to a modification in the gntR gene preventing the dimerization of the repressor. The modification may for example be a mutation resulting in the truncation of the GntR repressor. The modification may result in an at least partial deletion of the oligomerization domain of the GntR repressor. The mutation may also alter or delete at least parts of the effector binding domain of the GntR repressor, such that the repressor no longer responds to the effector molecule.

[00156] In yet another or further embodiment, a recombinant methylotrophic microorganism of this invention may comprise further modifications to reduce or eliminate the expression of one or more polypeptides from native genes encoding an enzyme of the TCA cycle, including aceA, acnA, fumC, sdhB, sucA, sucC, icd, aceE, aceF, sdhA, sucC, and mqo, or n combination thereof.

[00157] A recombinant methylotrophic microorganism of this invention may for example comprise a reduction in expression, or a deletion of the gene encoding the Gnd polypeptide and in the gene encoding the AroH polypeptide.

[00158] Polynucleotide sequences encoding polypeptides can be derived from the relevant nucleotide sequences by using well known codon tables and the degeneracy of the genetic code. Nucleotide sequences are provided in the sequence listings as identified by their sequence identity number (SEQ ID NO).

[00159] The introduction of only two mutations (AfrmA, AtpiA) together with the heterologous expression of three genes, was sufficient to generate an E. coli strain growing on methanol after about 250 generations in a continuous chemostat culture. The evolved strain builds its entire biomass from the reduced one-carbon compound, as we demonstrate by metabolic tracer experiments, and grows at a doubling time of about 8 hours.

Methylotrophic E. coli strains with improved growth rate

[00160] The above-mentioned chemostat evolution of 249 generations and seven subsequent passages of serial dilution provided one culture, which was split into

ETHZ-27-PCT four replicate lines: lines D, E, F, and G. The synthetic methylotrophic strain MEcoli_ref_l was isolated from line D after 255 generations.

[00161] After each replicate line had evolved for more than 1200 generations, single clones from each population were isolated and their growth rates were determined. The isolates grew 21% to 46% faster than MEcoli_ref_l.

[00162] The best-growing isolate, strain MEcoli_ref_2, derived from replicate line D, which evolved for a total of 1241 generations, exhibited a doubling time of 4.3 ± 0.1 h (mean ± S.D.) and grew to higher optical density compared to strain MEcoli_ref_l as shown in Figure 8A. When grown in a bioreactor under controlled pH and aeration conditions an optical density of 24 at ODeoo was typically reached for MEcoli_ref_2.

Proteomic and genetic modifications to improve the growth rate of the synthetic methylotroph

[00163] To identify mutations correlated with faster methylotrophic growth, the metagenomic mutational makeup of the evolving serial dilution populations was determined just before isolating single clones (D, 1129; E, 1124, F: 1118; G, 1125 generations). On average, each replicate line accumulated 316 ± 57 (mean ± S.D.) new nonsynonymous and intergenic mutations compared to the initial methylotrophic population.

[00164] Mutations in the population samples from the evolving serial dilution replicate lines were filtered for nonsynonymous and intergenic mutations. Only mutations present at greater than 60% frequency in the respective population were considered. Mutations already present in the original methylotrophic MEcoli_ref_l culture were excluded from further analysis. The resulting list of genetic changes were grouped at the gene and intergenic region level. Nonsynonymous and intergenic mutations that are present in MEcoli_ref_2 but not in strain MEcoli_ref_l were filtered and grouped.

[00165] While little evolutionary parallelism was observed at the nucleotide level, it was found that the functional units methanol oxidation, RuMP cycle and pyruvate metabolism were mutated repeatedly. This was mirrored at the gene and

ETHZ-27-PCT intergenic region level where eight genes and intergenic loci were mutated in all replicate lines, as illustrated in Figure 8B.

[00166] Figure 8B depicts a Venn diagram of mutational overlap of the adapted populations D, E, F and G. Mutations present in the common starter population used for the inoculation of the four serial dilution lines were discarded for further analysis. Different mutations were grouped at the gene or intergenic region level, e.g. the gene pykA (SEQ ID NO 72; PykA polypeptide SEQ ID NO 73) aggregating four different mutations. Groups that acquired mutations not already present in MEcoli_ref_l are highlighted in bold in Figure 8B.

[00167] A recombinant methylotrophic microorganism of this invention may comprise a loss of pyruvate kinase PykA and/or pyruvate kinase PykF activity, or a pyruvate kinase PykA and/or pyruvate kinase PykF activity that is 40% or less, for example 20%, 25%, 30%, 35%, 40% (or a value between any two of the foregoing values), of the enzymatic activity of the respective wild-type polypeptide.

[00168] Methanol oxidation acquired two genetic alterations conferring amino acid substitutions in methanol dehydrogenase and several intergenic mutations in its encoding plasmid. Mutations in methanol dehydrogenase were present in three of the four replicate lines (F279I in D, F and V359E in G).

[00169] Additional nonsynonymous mutations (F279I, SEQ ID NO 66 and SEQ ID NO 67; and V359E, SEQ, ID, NO 68 and SEQ ID NO 69) of the Mdh polypeptide bearing the H165N mutation in the replicate lines D, E and G correlated with increased growth rates on methanol.4.

[00170] The average doubling time for line D was 4.8h ± 0.5h. The average doubling time for line E was 5.2h ± 0.3h. The average doubling time for line F was 5.6h ± 0.3h. The average doubling time for line G was 5.6h ± 0.5h.

[00171] The methanol dehydrogenase mutations were not fully fixed in the evolving populations at the time of sampling. Therefore, it is possible that not all isolates from replicate lines D, E and G encoded the identified amino acid substitutions. The observed improved growth rates may therefore be more

ETHZ-27-PCT pronounced in the strains bearing the identified mutations than in the evolving population lines.

[00172] The RuMP cycle gained mutations in the promoter region of the hps-phi operon.

[00173] Methanol oxidation via NAD-dependent methanol dehydrogenase is a thermodynamically disadvantaged reaction that is only forward-driven at low formaldehyde concentrations relative to the concentrations of the substrates. In addition, maintaining low intracellular formaldehyde levels is important to prevent toxic protein-protein and protein-DNA crosslinking. It is therefore interesting that the replicate lines acquired several different mutations in the promoter region of the hps- phi operon. The operon encodes the enzymes for formaldehyde condensation with ribulose 5-phosphate and its subsequent isomerization to fructose 6-phosphate (Figure 8C). The reproducible alteration of the hps-phi promoter region suggests that finetuned expression of the encoded enzymes confers a selective advantage.

[00174] The RuMP cycle also gained mutations in the gene encoding 6- phosphogluconate dehydrogenase (gnd).

[00175] 6-phosphogluconate dehydrogenase is part of the dissimilatory RuMP cycle which is responsible for formaldehyde detoxification and the generation of reduced nicotinamide adenine dinucleotide phosphate (NADPH/H ⁺ ) under loss of carbon dioxide. Two of the four observed mutations in gnd resulted in premature stop codons and the other two in radical amino acid changes.

[00176] To check if reduced flux through 6-phosphogluconate dehydrogenase is expected to be beneficial for methylotrophic growth, E. coli methanol metabolism was modelled using parsimonious flux balance analysis (pFBA). pFBA is a variant of regular flux balance analysis which first determines the maximum possible growth rate given a set substrate uptake rate and then minimizes the total flux through all reactions. Lower total flux is assumed to require less enzyme mass. Consequently, pFBA results reflect evolutionarily optimal solutions and have been shown to accurately predict fluxes in well-adapted organisms.

ETHZ-27-PCT [00177] Figure 8C shows a flow chart of the pFBA for reduced flux through 6- phosphogluconate dehydrogenase. The names of the repeatedly mutated functional units in replicate lines D, E, F and G are indicated and shown next to the associated functional metabolic step represented by an arrow. Thin lines depict reactions predicted by pFBA to carry low, for example less than 10% of incoming for gnd, or to carry no flux, as is the case for py kA/ py kF, and for part of the TCA cycle. Dotted lines represent more than one reaction.

[00178] The following acronyms were used in Figure 8C: "aKG" a-ketoglutarate, "Aco" aconitate, "Cit" citrate, "fepA" ferric enterobactin outer membrane transporter, "gnd" 6-phosphogluconate dehydrogenase, "imdh" intergenic region Mdh-encoding plasmid, “phps-phi" promoter region hps-phi operon, "ppsR" phosphoenolpyruvate synthetase regulatory protein, "pykA" pyruvate kinase 1, "pykB" pyruvate kinase 2, "rhIB” ATP-dependent RNA helicase.

[00179] pFBA predicted little flux through the gnd reaction, meaning that the reaction carries less than 10% of total flux away from the reaction's substrate, 6- phosphogluconate. Furthermore, the deletion of gnd changes the predicted growth rate on methanol only minimally (<1%).

[00180] pFBA predicted some flux through dissimilatory branch of the RuMP cycle for generation of reduced NADP. However, its deletion can be offset by membrane-bound or soluble pyridine nucleotide transhydrogenase activity and has negligible impact on the predicted growth rate.

[00181] Carbon entry into the TCA cycle occurs via a carboxylating reaction in our methylotrophic E. coli. pFBA predicted this to be achieved by converting pyruvate to phosphoenolpyruvate followed by its condensation with carbon dioxide to oxaloacetate. To avoid the creation of a futile cycle, the reverse reaction from phosphoenolpyruvate to pyruvate cannot carry flux, as illustrated in Figure 8C.

[00182] It was found that the genes encoding this reaction, pykA and pykF, acquired several nonsense and missense mutations in all replicate lines. Additionally, mutations in ppsR, a regulator of phosphoenolpyruvate synthetase, were observed.

ETHZ-27-PCT [00183] Two additional genes were mutated in all replicate lines: rhIB encoding an ATP-dependent RNA helicase an fepA encoding a ferric enterobactin outer membrane transporter.

[00184] The contribution of rhIB, encoding an ATP-dependent RNA helicase may help buffer deleterious mutations.

[00185] MEcoli_ref_2, which was isolated later (1241 generations) from replicate line D, carries mutations in the same genes and intergenic regions that repeatedly turned up in the evolving populations.

[00186] All but three of these targets of the mutations in MEcoli_ref_2, mdh, pykF and rhIB, were already mutated in the original methylotrophic reference strain, MEcoli_ref_l.

[00187] Increased methanol oxidation capacity and optimized flux from pyruvate to phosphoenolpyruvate may confer the faster methylotrophic growth of a recombinant methylotroph.

[00188] The mdh the mutation in MEcoli_ref_2 results in amino acid change F279I in the encoded polypeptide.

[00189] In addition, further mutations fixed in the intergenic regions of the Mdh- encoding plasmid may have optimized expression or plasmid copy numbers.

[00190] The pykF the mutation in MEcoli_ref_2 results in amino acid change L303H in the encoded polypeptide.

[00191] The rhIB the mutation in MEcoli_ref_2 results in a conversion of the stop codon into a lysine-encoding codon.

[00192] The improved synthetic methylotrophic reference strain, MEcoli_ref_2, exhibited a 46% reduced doubling time over MEcoli_ref_l. Furthermore, MEcoli_ref_2 grows faster (Td = 4.3 ± 0.1 h) than the comparable natural methylotroph B. methanolicus at 37°C (Td = 5 h), which also utilizes the RuMP cycle and an NAD-

ETHZ-27-PCT dependent methanol dehydrogenase. Notably, B. methanolicus is auxotrophic for vitamin B12 and biotin, while MEcoli_ref_2 can grow independently of any additional carbon supplementation. To date, MEcoli_ref_2 has the fastest reported growth of a model organism equipped with a synthetic single carbon assimilation pathway.

[00193] In summary, all four replicate lines D, E, F and G independently acquired mutations in methanol oxidation, the RuMP cycle and pyruvate metabolism. The relevance of the identified mutations was confirmed through metabolic modelling studies and pFBA.

Table 3: Mutations of derivatives of MEcoli_ref_l with reduced doubling times

Biomass and metabolite formation from methanol

[00194] To confirm complete biomass formation from methanol alone, ¹³ C metabolic tracer analysis was used. The results of these assays are shown in Figures 6A to 6D.

[00195] Figures 6A and 6D show the ¹³ C labelled fraction of protein-bound amino acids by liquid chromatography coupled mass spectrometry (LC-MS) in MEcoli_ref_l grown in medium containing no additional carbon sources. Antibiotics, IPTG and EDTA were omitted from the minimal medium in these studies to ensure that

ETHZ-27-PCT methanol and CO2 were the only available carbon sources for growing the strain. Error bars represent the standard deviation for three technical replicates of MEcoli_ref_l.

[00196] Isotopologue distributions, shown in Figures 6A and 6C, and fractional contributions, shown in Figures 6B and 6D, were determined for different protein bound amino acid by LC-MS.

[00197] For isotopologue distributions, n refers to the fully labelled molecule and n-1, n-2, etc. to molecules with one, respectively two, etc., unlabelled carbon atoms.

[00198] When grown under ambient atmosphere, appreciable amounts of ¹² C label present in protein-bound amino acids were observed (Figure 6A, 6B). Due to the absence of additional carbon sources in the medium, the remaining source of ¹² C carbon was ambient CO2. Consequently, when MEcoli_ref_l was grown under an atmosphere enriched to 5% (V/V) ¹³ CO2, protein-bound amino acids were fully labelled (Figure 6C and 6D).

[00199] In Figures 7A to 7D a similar labelling pattern is shown for extracted metabolites of methylotrophic clones from an earlier stage of the serial dilution evolution. In addition, the total biomass labelling ratio using elemental analyser/isotope ratio mass spectrometry is shown.

[00200] Label incorporation was studied in four replicates isolated from the evolving serial dilution experiment after 384 generations. Cultures were grown in minimal medium supplemented with 500 mM ¹³ C methanol either at ambient CO2, as shown in Figures 7A and B, or at 5% (V/V) enriched ¹³ CC>2 atmosphere, as shown in Figures 7C and D.

[00201] Antibiotics, IPTG and EDTA were omitted out from the minimal medium to ensure that methanol and CO2 are the only available carbon sources. The isotopologue distribution, as shown in Figures 7A and 7C, and fractional contribution, as shown in Figures 7B and 7D, was determined for different central metabolites by LC- MS. The central metabolites measured were hexose phosphates (H6P), pentose phosphates (P5P), sedoheptulose 7-phosphate (S7P), 2-phosphoglycerate/3-

ETHZ-27-PCT phosphoglycerate (2PG/3PG), asparagine (Asn), arginine (Arg), glutamine (Gin), and lysine (Lys).

[00202] For technical reasons the different hexose phosphates (glucose 6- phosphate, fructose 6-phosphate, arabino 3-hexulose 6-phosphate), pentose phosphates (ribose 5-phosphate, ribulose 5-phosphate, xylulose 5-phosphate), and 2- phosphoglycerate and 3-phosphoglycerate were summarized as one group. S7P was detected in only 3 out of 4 replicates in the ¹² CO ₂ condition.

[00203] For the isotopologue distribution, n refers to the fully labelled molecule and n-1, n-2, etc to molecules with one, respectively two, unlabelled carbon atoms.

[00204] Error bars represent the standard deviation for four biological replicates for the ¹² CO ₂ condition and for three for the ¹³ CO ₂ condition.

[00205] When grown under ambient atmosphere, 83.9 ± 4.6% (mean ± standard deviation) of the total biomass of MEcoli_ref_l was ¹³ C labelled. We found that this matches the FBA model that predicts total biomass labelling of about 83%.

[00206] When MEcoli_ref_l was grown under ¹³ CO ₂ enriched atmosphere 98.6 ± 0.3% mean ± standard deviation of its biomass was derived from ¹³ C, as expected.

[00207] Lastly, the ¹³ C fraction of protein-bound amino acids (Figure 4B) allowed to discern where CO ₂ entered metabolism. CO ₂ did not contribute to the biosynthesis of amino acids directly derived from the RuMP cycle (His, Phe, Tyr) and only little to ones (Ala, Leu, Ser, Vai) with gluconeogenic precursors. Noticeably higher levels of CO ₂ incorporation were found in amino acids derived from the TCA cycle (Arg, Asp, Glu, He, Pro, Thr).

[00208] These metabolic tracer experiments proved, that the evolved strain builds its entire biomass from the reduced one-carbon compound. The evolved E. coli strain grows at a doubling time of about 8 hours on the reduced one-carbon compound.

ETHZ-27-PCT [00209] The recombinant methylotroph proposed by this invention provides a valuable tool for microbial conversion of reduced Cl compounds, in particular methanol, into value-added compounds and for applications in industrial biotechnology.

Producing value-added compounds from methanol using synthetic methylotrophy

[00210] The synthetic methylotrophic E. coli strain MEcoli_ref_2, was engineered to provide a proof-of-principle production of value-added compounds from three key metabolic nodes: pyruvate, acetyl coenzyme A and the TCA cycle, Figure 10B. To this end, the biosynthetic pathways for lactic acid, PHB and itaconic acid were heterologously expressed in MEcoli_ref_2. The metabolic routes for the production of these compounds is shown in Figure 10A.

[00211] For the conversion of pyruvate to lactic acid only one additional enzyme is required, making it a promising first methanol-to-product target. The E. coli genome encodes D-lactate dehydrogenase (IdhA), which catalyzes pyruvate conversion to D- lactate with concomitant reduction of NAD (Figure 10A). In aerobic conditions, D- lactate dehydrogenase expression is low. E. coli also encodes an L-lactate dehydrogenase (IldD). However, the gene product only catalyzes the reverse reaction and enables lactic acid utilization as carbon source. To enable lactic acid production in an oxygen-replete setting, the highly active lactic acid dehydrogenase from Streptococcus bovis was chosen. The gene lactic acid dehydrogenase (Idh) was cloned under control of an anhydrotetracycline (aTc)-inducible promoter into an expression plasmid and transformed it into MEcoli_ref_2. The resulting strain was incubated in minimal medium supplemented with 500 mM methanol and induced L-lactate dehydrogenase expression in late-exponential phase (2 pM aTc, OD600 ~ 1.7). 21 h later, all organic acids were derivatised to their respective 3-nitrophenylhydrazones and the lactic acid derivative concentration was quantified by liquid chromatography- coupled mass spectrometry (LC-MS). When Idh is expressed, the strain produced 267.6 ± 17.6 mg L ¹ (3.0 ± 0.2 mM (mean ± S.D.)) Lactic acid, as shown in Figure 10C. In agreement with low expression of D-lactate dehydrogenase the empty vector negative control produced less than 20 pM lactic acid.

ETHZ-27-PCT [00212] Furthermore, it was observed that the detected lactic acid concentration in the medium decreased over the following days. Methanol was the only available carbon source in the medium used for cultivation of MEcoli_ref_2. To exclude unaccounted carbon sources from contributing to lactic acid formation, MEcoli_ref_2 transformed with pljac was grown in identical conditions in minimal medium containing 500 mM ¹³ C methanol (99% atomic purity). The produced lactic acid matched the expected 97% labelling. Total unlabelled methanol medium content was about 3%.

[00213] To further increase the yield of lactic acid, E. coli's native L-lactate dehydrogenase (IldD) could be deleted to prevent lactic acid reassimilation.

[00214] PHB is produced from acetyl coenzyme A, as illustrated in Figure 10A. To enable this transformation in MEcoli_ref_2 was transformed with an expression plasmid pl_phb (Table 4) which encodes the required phaCAB operon (Figure 10A). The operon was sourced from the native PHB producer Cupriavidus necator H16.

[00215] Following transformation of pl_phb into MEcoli_ref_2, the resulting strain was incubated in minimal medium supplemented with 500 mM methanol and sampled the culture after 74 h. Due to its polymeric nature, accurate direct quantification of PHB is technically challenging. To overcome this problem, PHB was depolymerized and derivatized to methyl 3-hydroxybutanoate, which was subsequently subjected to gas chromatography-coupled flame ionization detector (GC- FID) analysis. The PHB concentration was then calculated from the measured methyl 3- hydroxybutanoate concentration. Expression of the phaCAB operon resulted in about 0.32 ± 0.11 mg L ¹ PHB (Figure 10D).

[00216] The fact that the strain produced PHB, highlights the capability of MEcoli_ref_2 to express longer biosynthetic pathways. The PHB titer (0.32 mg L ^-1 ) was lower than that of lactic acid. One reason may be poor expression of the pathway genes in E. coli. Another reason may be found in limited flux towards acetyl coenzyme A, the precursor of PHB, due to the structure of E. coli metabolism on methanol. Flux away from pyruvate is expected to split roughly evenly towards phosphoenolpyruvate and acetyl coenzyme A to support efficient methanol assimilation for biomass

ETHZ-27-PCT formation. This decreases the PHB production potential. However, this is likely to be overcome by overexpressing pyruvate dehydrogenase.

[00217] The synthesis of itaconic acid proceeds from the TCA cycle intermediate cis-aconitate (Figure 10A). The reaction is catalyzed by cis-aconitate decarboxylase under loss of carbon dioxide. The cis-aconitate decarboxylase-encoding gene, cadA, was cloned from the industrial itaconic acid producer Asperg Ulus terreus into the same vector used for lactic acid production. Following a similar protocol as described above, the resulting plasmid (pljta) was transformed into MEcoli_ref_2, the strain was incubated in minimal medium supplemented with 500 mM methanol, gene expression was induced at mid-exponential phase (2 pM aTc, OD600 ~ 1.1), all organic acids were derivatized and itaconic acid was quantified by LC-MS. The strain produced 50.3 ± 3.8 mg L ¹ (392.3 ± 29.3 pM (mean ± S.D.)) itaconic acid after 70 h, when cadA was expressed (Figure 10E).

[00218] It has been shown that itaconic acid production in E. coli can be improved by coexpression of aconitase, which converts citrate to cis-aconitate, physically linked to a cis-aconitate decarboxylase variant with higher activity. Additionally, itaconic acid is converted to itaconyl-CoA by promiscuous action of succinyl-CoA synthetase, which sequesters part of the product. Tuning the activities of the two native succinyl-CoA forming complexes, 2-oxoglutarate dehydrogenase and the promiscuous succinyl-CoA synthetase, may result in an increased yield of this compound.

[00219] It was therefore established that synthetic methylotrophy provides a new mode for bioproduction from methanol.

A recombinant Methanol dehydrogenase

[00220] The ancestral strain used for generating the fully methylotrophic E. coli MEcoli_ref_l strain according to this disclosure transformed to heterologously produce the methanol dehydrogenase 2 (Mdh) variant CT4-1 from

Cupriavidus necator, SEQ ID NO 1, as described herein.

ETHZ-27-PCT [00221] The ancestral Mdh variant CT4-1 has been described previously by Wu, T.-Y. et al., 2016, Appl. Microbiol. Biotechnol. 100, 4969-4983; DOI: 10.1007/s00253- 016-7320-3.

[00222] Remarkably, following the long-term evolution study, a point mutation of the mdh gene could be identified which resulted in a significantly increased activity of the modified Mdh polypeptide (SEQ ID NO 58, 59). The point mutation caused an amino acid change H165N in the enzyme.

[00223] The mutated Mdh H165N polypeptide was isolated, and its activity was tested in vitro as described in the Material and Methods section.

[00224] To this end, the ancestral mdh CT4-1 sequence and the mutant mdh H165N sequence encoding the respective methanol dehydrogenases were cloned into a commercially available pET16b expression vector using Gibson assembly. The resulting constructs were His-tagged and were purified using nickel-immobilized metal affinity chromatography.

[00225] Catalytic activity of both enzymes was then measured in in vitro assays, and the results of these assays are shown in Figure 11. Maximum reaction speed v _m ax was determined for purified His-tagged Mdh and Mdh(H165N) on the basis of measured increase in fluorescent counts indicated in arbitrary units per second (a.u./s). Equal amounts of protein were used in both assays and methanol turnover measured by following formation of NADH/H ⁺ in a spectrophotometer, i.e. absorbance increase at 340 nm over time.

[00226] It was established as part of this disclosure, that a Mdh polypeptide bearing this mutation has an improved methanol dehydrogenase activity compared to the parental CT4-1 variant of C. necator consisting of SEQ ID NO: 1. By introducing an H165N mutation in a Mdh polypeptide, the methanol dehydrogenase catalytic turnover number compared to the parental Mdh polypeptide can be increased by at least 1.5 fold, or at least 2-fold, or at least 2.5-fold.

ETHZ-27-PCT [00227] This disclosure also concerns a modified polypeptide having at least 70%, 80%, 90%, 95%, 98% or 99% sequence identity to SEQ ID NO: 1 and comprises at least an H165N mutation.

[00228] In one embodiment, the recombinant Mdh polypeptide may further comprise a F279I mutation and/or a V359E mutation,

[00229] The Mdh polypeptide bearing the H165N mutation doubles the catalytic activity of this enzyme, as shown in Figure 11.

[00230] To further enhance the activity of this enzyme further mutations were introduced into the Mdh polypeptide bearing the H165N mutation using serial dilution evolution, resulting in two nonsynonymous mutations (F279I, V359E) of Mdh. Both, the F279I as well as the V359E mutation independently, each combined with the H165N mutation resulted in increased growth rates on methanol when expressed in synthetic methylotrophic replicate lines D, E and G of MEcoli_ref_l. A combination of all three mutations in the Mdh polypeptide is also expected to result in an increased growth rate.

[00231] A non-naturally occurring recombinant methylotroph expressing a Mdh polypeptide comprising a H165N combined with a F279I mutation and/or combined with a V359E mutation may have a reduction in doubling time of more than 30%, of more than 35%, of more than 40%, of more than 45%, for example 47% when compared to a recombinant methylotroph expressing the Mdh polypeptide comprising the H165N mutation, but neither the F279I mutation and/or combined with a V359E mutation.

[00232] When grown on methanol the doubling time of MEcoli_ref_2 expressing Mdh comprising the H165N mutation as well as F279I mutation was reduced by 47% compared to the methanol-grown MEcoli_ref_l expressing Mdh with the H165N mutation only. Similarly, the average doubling time of E coli line E expressing Mdh with the H165N and the V359E mutation was reduced by 36% compared to MEcoli_ref_l expressing Mdh with the H165N mutation only.

ETHZ-27-PCT [00233] Preferably, the modified polypeptide is engineered from a Mdh polypeptide of C. necator, or from the engineered CT4-1 Mdh variant of C. necator.

[00234] The modified polypeptide of this disclosure preferably has an improved methanol dehydrogenase activity compared to its ancestral Mdh polypeptide, for example the CT4-1 variant of C. necator.

[00235] The modified polypeptide may have an increased reaction rate v _m ax of at least 1.2-fold greater, at least 1.5-fold, at least 1.8-fold, at least 2-fold, at least 2.2-fold, at least 2.5-fold, at least 2.8-fold, or at least 3-fold greater than its parental polypeptide.

[00236] The disclosure furthermore concerns an isolated nucleic acid encoding the modified polypeptide.

[00237] The modified polypeptide may be expressed from a plasmid, preferably from an expression vector.

[00238] The modified polypeptide may be overexpressed for further processing, such as purification.

[00239] The modified polypeptide may also be expressed in a recombinant host cell to introduce or to enhance a certain metabolic trait of said microorganism.

[00240] The expression modified polypeptide in the recombinant host cell, may result in the production of a metabolite of product by the recombinant host cell, which it would not normally produce. The expression modified polypeptide may also result in an increase of production of such a metabolite or product when compared to the wildtype host cell.

[00241] The expression modified polypeptide in the recombinant host cell, may result in introducing or enhancing the capability of the recombinant host cell to metabolise a substrate, such as methanol or another Cl compound, when compared to the parental host cell.

ETHZ-27-PCT [00242] The modified polypeptide may be expressed in a host cell. The host cell may be a eukaryotic cell or a prokaryotic cell.

[00243] The parental host cell may for example be a non-naturally occurring recombinant methylotrophic microorganism. The parental host cell may for example be the AfrmAAtpiA strain.

Materials and Methods

Reagents and media:

[00244] Chemicals were obtained from Sigma-Aldrich Chemie GmbH, Buchs, Switzerland unless otherwise specified. The M9 minimal medium used for bacterial cultivation consisted of the following salts (g L ¹ ): Na ₂ HPO4 (6.78), KH2PO4 (3.0), NaCI (0.5), NH4CI (1.0), CaCI ₂ (0.735), MgSC (0.123) and trace elements. Trace elements were present in the medium at the following concentrations (mg L ^-1 ): Na ₂ EDTA (5.0), MnSO ₄ (5.0), FeSO ₄ -7 H ₂ O (1.0), Co(NO ₃ ) ₂ -6H ₂ O (1.0), ZnSO ₄ -7 H ₂ O (1.0), CuSO ₄ -5 H ₂ O (0.1), Na ₂ MoO4-2 H ₂ O (0.1), N iCI ₂ -6 H ₂ O (0.2). If indicated, antibiotics were added in the following concentrations (mg L ^-1 ): ampicillin (100), ca rbenici Hi n (50), streptomycin sulfate (20).

Primers and plasmids used in this study

[00245] Plasmids and primers used in this study are listed in Tables 4 and 5. The heterologously introduced plasmids were pSEVA424 with the methanol dehydrogenase 2 (mdh) gene variant CT4-1 from Cupriavidus necator, as described in Wu, T.-Y. et al., 2016, Appl. Microbiol. Biotechnol. 100, 4969-4983;

DOI: 10.1007/s00253-016-7320-3, and pSEVA131 with the 3-hexulose 6-phosphate synthase (hps) and the 6-phospho 3-hexuloisomerase (phi) from Methylobacillus flagellatus. The nucleotide sequence of the plasmids was confirmed by PCR and Sanger sequencing (Microsynth AG; Switzerland).

Table 4: Plasmids used in this study

ETHZ-27-PCT

Table 5: Oligonucleotides used in this study

ETHZ-27-PCT

Strains used in this study

[00246] The starting strain of the evolution experiment E. coli BW25113 AfrmAAtpiA containing pSEVA424 mdh C. necator (SEQ NO ID 1, 2), and pSEVA131 hps (SEQ NO ID 52, 53) and phi M.flagellatus (SEQ NO ID 54, 55) was described in a previous study, as disclosed in Keller, P. et al., 2020, Nat. Commun. 11, 5403, https://doi.org/10.1038/s41467-020-19235-5.

Table 6: E. coli strains used in this study

Long-term chemostat evolution

ETHZ-27-PCT [00247] The evolution experiment was conducted in a 500 mL bioreactor (Multifors, Infors-HT, Bottmingen, Switzerland) filled with 300 mL minimal medium at 37°C under constant stirring (700 revolutions per minute (r.p.m.)) and aerated with compressed air. The bioreactor was equipped with medium feed, efflux, acid, and base pump systems. The pH was kept constant at 7.1 by the addition of either hydrochloric acid (HCI) or sodium hydroxide (NaOH). The efflux pump was operated at much higher speed than the feed pump to keep the volume (292, 296, and 299 mL for reactor 1, 2, and 3, respectively) of the culture constant and to maintain the chemostat condition. The culture volume and flow rate were determined by measuring the difference in weight of the feed and waste medium over time. The dilution rates were 0.42, 0.34, and 0.62 d ¹ for reactor 1, 2, and 3, respectively. The feed medium consisted of minimal medium supplemented with 500 mM methanol, 20 mM pyruvate, 0.1 mM IPTG, ampicillin, and streptomycin for the first 90 generations. Then, the pyruvate concentration was reduced to 10 mM and after another 6 generations to 5 mM, respectively. The state of the culture in the chemostat was followed by measuring the optical density of the medium at 600 nm. The chemostat was restarted twice, once after generations 202 generations (contamination) and once after 223 generations (fresh chemostat set at faster dilution rate). The doubling time of the bacterial culture and the number of generations were calculated from standard chemostat equations. At regular intervals, the population was tested for methylotrophic and methanoldependent growth. To verify that the pyruvate concentration was limiting, sterile- filtered medium aliquots were analyzed by HPLC (UPLC Ultimate 3000, ThermoFisher Scientific, Reinach, Switzerland) equipped with an ion exclusion column (Rezex ROA- Organic Acid H+ (8%) 300 x 7.8mm, Phenomenex, Torrance, CA, United States of America) applying 5 mM H2SO4 as a mobile phase isocratica I ly. 10 pL of sample were injected at a flow rate of 0.6 mL min ¹ and the absorbance at 210 nm was recorded for 25 minutes. The pyruvate concentration was calculated based on a standard curve from samples with known pyruvate levels (0.01, 0.1, 0.5, 1, 5, 10, 20 mM pyruvate) and concentrations below 0.01 mM were considered as below the detection limit.

Serial transfer evolution

[00248] After evolution in a chemostat for 249 generations, the population was propagated under a serial transfer regime. Initially, an aliquot from the chemostat population was passaged seven times in 20 mL minimal medium containing 500 mM methanol, 0.1 mM IPTG, Amp, and Sm. This culture was then used to inoculate four

ETHZ-27-PCT replicate lineages. Each was propagated in 30 mL of the same medium, except that IPTG was omitted, and passaged during mid- or late-exponential phase. All cultures were incubated in 100 mL baffled shake flasks at 37°C, 160 r.p.m. in a Minitron shaker (Infors-HT, Bottmingen, Switzerland). To inoculate fresh medium, old cultures were diluted 1:100 (V/V).

Characterization of the growth phenotype after 534 generations

[00249] After 534 generations of evolution, the four replicate lineages, lines D, E, F, and G were streaked out on agar plates containing minimal medium supplemented with 500 mM methanol, ampicillin, and streptomycin. Four colonies were inoculated into 30 mL medium supplemented with 500 mM methanol, ampicillin, and streptomycin and cultivated in baffled shake flasks at 37°C, 160 r.p.m. in a Minitron shaker. During late-exponential growth, cultures were diluted 1:100 (V/V) in fresh medium of the same composition to assess growth. Growth of the bacterial cultures was monitored by measuring the ODeoo over time. A cryostock was generated of the fastest growing replicate (MEcoli_ref_l).

Proteome comparison between ancestral methanol-dependent E. coli and MEcoli ref 1.

[00250] MEcoli_ref_l was streaked out on agar plate containing minimal medium supplemented with 500 mM methanol and incubated at 37°C until colonies were visible. A cross-section of colonies was used to inoculate a pre-culture in 30 mL minimal medium supplemented with 500 mM methanol and cultivated in baffled shake flasks at 37°C, 160 r.p.m. until stationary phase. Next, the culture was diluted 1:100 (V/V) into fresh medium, grown until mid-exponential phase and split 1:100 (V/V) into five main-culture replicates. Once the cultures reached mid-exponential phase (ODeoo ~ 0.6), 4 OD units (1 OD unit equals 1 mL of culture at ODeoo of 1) of cells were harvested, cooled to 4°C, spun down (3220 g, 15 min), and washed once with 4 mL 10 mM MgClz, and twice with 1 mL 10 mM MgCI?. Finally, the supernatant was discarded, the cell pellet shock frozen in liquid nitrogen and frozen. The same procedure was followed for the ancestral strain except all media were additionally supplemented with 20 mM pyruvate and 0.1 mM IPTG.

[00251] Cell pellets were dissolved in 300 pLIOO mM ammonium bicarbonate, 8 M urea, lx complete EDTA-free protease inhibitor cocktail (Sigma-Aldrich, Buchs,

ETHZ-27-PCT Switzerland) and lysed by indirect sonication (3 x 1 min, 100% amplitude, 0.8 s cycle time) in a VialTweeter (HIFU, Hielscher, Teltow, Germany). Larger particles and insoluble parts were removed by centrifugation at 13,000 g, 15 min, 4°C. Protein concentrations in the lysates were determined by Pierce BCA assays (Thermo Fischer Scientific, Reinach, Switzerland). Protein disulfide bonds were reduced by adding tris(2- carboxylethyl)phosphine (TCEP, Sigma-Aldrich, Buchs, Switzerland) to a final concentration of 5 mM and incubation for 30 min at 37°C and 300 r.p.m. shaking. Cysteine residues were alkylated by adding iodoacetamide (IAA, Sigma-Aldrich, Buchs, Switzerland) to a final concentration of 10 mM for 30 min at room temperature in the dark. Prior to digestion the urea concentration was reduced below 2 M by diluting all samples 1 to 5 with freshly prepared 50 mM ammonium bicarbonate. For digestion, sequencing grade modified trypsin (Promega AG, Dubendorf, Switzerland) was added at a 1:50 (pg trypsin/pg protein) ratio. Digestion was carried out in a tabletop shaker over night at 37°C under constant shaking at 300 r.p.m. After digestion, the trypsin was inactivated using heat incubation in a tabletop shaker at 95°C for 5 min and subsequent acidification to an approximate final concentration of 1% (V/V) formic acid. The peptide samples were centrifuged at 20,000 g for 10 min to remove insoluble parts and the supernatant was taken and desalted using Sep-Pak Vak C18 reversed phase columns (Waters Corporation, Baden-Dattswil, Switzerland) as described previously and dried under vacuum. Prior to MS analysis, the samples were re-solubilized in 3% acetonitrile (ACN) containing 0.1% formic acid (FA) to a final concentration of 0.5-1 pg pL ¹ .

[00252] Mass spectrometry analyses was performed on an Orbitrap Lumos Tribrid mass spectrometer (Thermo Fischer Scientific) equipped with a digital PicoView source (New Objective, Littleton, USA) coupled to an M-Class ultraperformance liquid chromatography (UPLC) system (Waters GmbH, Wilmslow, UK). A two-channel solvent system was used with 0.1% formic acid (V/V) in water for channel A and 0.1% formic acid (V/V), 99.9% ACN (V/V) for channel B. At a peptide concentration of 0.5 pg pL ¹ for each sample 2 pL were loaded on an ACQUITY UPLC M-Class Symmetry C18 trap column (100 A, 5 pm; 180 pm x 20 mm, Waters) followed by an ACQUITY UPLC M-Class HSS T3 column (100 A, 1.8 pm; 75 pm x 250 mm, Waters). Peptide samples were separated at a flow rate of 300 nL min ¹ with an initial of 5% B for 3 min. The gradient was as follows: from 5% to 22% B in 112 min, from 22% to 32% B in 8 min, from 32% to 95% B in 5 min and from 95% to 5% B in 10 min. The mass spectrometer was operated

ETHZ-27-PCT in data-dependent acquisition (DDA) and full-scan mass spectra were acquired in the Orbitrap analyzer with a mass range of 300-2000 m/z and a resolution of 120k with an automated gain control (AGC) target value of 500,000. Fragment ion spectra (MS/MS) were acquired in the Ion trap using quadrupole isolation with a window of 1.6 Da and fragmented using higher energy collisional dissociation (HCD) with a normalized collision energy of 35%. Only precursor ions with charge states +2 to +7 and a signal intensity of at least 5000 were selected for fragmentation, and the maximum cycle time was set to 3 s. The ion trap was operated in rapid scan mode with an AGC target value of 10,000 and a maximum injection time of 50 ms. The dynamic exclusion was set to 25 s, and the exclusion window was set to 10 p.p.m. Measurements were acquired using internal lock mass calibration on m/z 371.10124 and 445.12003. Sample acquisition was performed in randomized order.

[00253] Progenesis QI (Nonlinear Dynamics, v.4.2.7207.22925) was used to process the acquired raw mass spectrometry data. Prior to the automatic alignment, 3- 5 vectors were manually seeded to aid the alignment. As alignment reference a 1:1 pool of all samples was used. After normalization, from each peptide ion a maximum of the top five tandem mass spectra were exported. The mascot generic file (* .mgf) was searched using the Mascot server (Matrix Science, v.2.7.0.1) against a decoyed and reversed protein sequence database. Two databases were constructed: For the ancestral strain, one containing the 4449 annotated proteins of the ancestral strain BW25113 (Genbank accession: NZ_CP009273) supplemented with the amino acid sequences of Mdh, Hps, Phi and concatenated with the yeast proteome (Uniprot accession: UP000002311) as well as 260 known mass spectrometry contaminants. For the evolved strain, the same database, but with modified amino acid sequences to account for observed mutations. The Mascot search parameters were as follows: Precursor ion and fragment ion tolerance were set to ± 10 ppm and ± 0.5 Da, respectively. Trypsin was selected as protease (two missed cleavages) and ions with charge state 2+, 3+ am were 4+ were selected for identification.

Carbamidomethylation of cysteine was set as fixed modification and oxidation of methionine, carbamylation of the N-terminus and lysine were set as variable modifications. The mascot search was imported into Scaffold (Proteome Software, v.5.1.0) using 5% peptide and 10% protein false discovery rate (FDR) and the resulting scaffold spectrum report were imported into Progenesis QI. For label-free protein quantification the Hi-3 approach was selected and only proteins with at least two

ETHZ-27-PCT unique peptides were considered for quantification. Statistical testing was performed directly in Progenesis with a one-way ANOVA and the resulting P-values were adjusted for multiple hypothesis testing using the Benjamini-Hochberg procedure (termed q- values). General cutoffs for significantly regulated proteins were q-values < 0.05 and | Logzffold-changes) | > 1.5.

[00254] Relative proteome contributions of individual proteins were estimated as described previously in. Only proteins for which more than three peptides were detected were considered for analysis (i.e. the quantifiable proteome). Next, for each protein the mean peak area was divided by the mean sum of all peaks. This ratio determined the relative protein abundance of a protein.

Proteomics gene set enrichment analysis of KEGG pathways

[00255] Gene set enrichment analysis of the proteomics data was conducted against the KEGG database using clusterProfiler (gseKEGG) with the following settings: organism = 'eco', minGSSize = 3, pvalueCutoff = 0.05, pAdjustMethod = 'BH'.

Characterization of the growth phenotype of MEcoli ref 1

[00256] One of the replicates of the last proteomics preculture was diluted 1:100 (V/V) into minimal medium supplemented with 500 mM methanol. In the negative control methanol was omitted. The cultures were split into 10 technical replicates of 150 pL each, transferred to a 96 well microtiter plate and incubated at 37°C, 800 r.p.m. in a LogPhase 600 reader (Agilent, Basel, Switzerland). Growth was observed by measuring absorbance at 600 nm. A calibration curve was used to convert measured absorbance values to ODeoo values corresponding to measurements with a pathlength of 10 mm.

¹³ C isotopic tracer analysis of metabolites

[00257] After 368 generations of evolution, the four replicate serial dilution lineages were streaked out on agar plates containing minimal medium supplemented with 500 mM methanol, ampicillin, and streptomycin. Four colonies were inoculated into two different conditions: (1) 30 mL of minimal medium supplemented with 500 mM ¹³ C methanol (isotopic purity 99%, Euriso-Top GmbH, Saarbrucken, Germany), but without NazEDTA and without antibiotics in baffled shake flasks, which were incubated at 37°C, 160 r.p.m. in a Minitron shaker under ambient atmosphere. (2) The same

ETHZ-27-PCT medium, which was sparged with synthetic air (80% (V/V) N2, 20% (V/V) O2) for 30 min before inoculation, in sealed shake flasks with a synthetic air atmosphere containing 5% (V/V) ¹³ CO ₂ (99% isotopic purity). In both conditions cultures were incubated at 37°C, 160 r.p.m in a Minitron shaker. At late-exponential phase cultures were diluted 1:100 (V/V) into identical conditions.

[00258] Once cultures reached optical densities between 0.5 and 1.21, metabolites were extracted from 7 to 10 OD units (1 OD unit equals 1 mL of culture at ODeoo of 1) of culture by rapid filtration. To this end, 5 OD units of culture were applied onto a 0.2 pm regenerated cellulose filter (RC58, Whatman GmbH, Dassel, Germany), filtered, washed with 10 mL, 37°C ultrapure water containing 500 mM ¹³ C methanol, quenched in 8 mL ice cold acetonitrile/methanol/0.5 M formic acid (60:20:20 (V/V/V)), vortexed for 10 s, and kept on ice for 10 min. Each culture was sampled twice in rapid succession. Following metabolite extraction, samples were lyophilized and subsequently resuspended in 250 pL solvent A/solvent B (90:10 (V/V), see below), centrifuged at 10,000 g, 4°C for 10 min, the supernatant transferred to a fresh tube and centrifuges again at 20,000 g, 4°C for 10 min and the supernatant transferred into HPLC vials. Assuming 1 OD unit corresponds to 250 pg cell dry weight, each sample was extracted from 2500 pg cell biomass dry weight, except samples from replicate 1 grown under ambient atmosphere and from replicate 4 grown under at 5% (V/V) ¹³ CO ₂ , which were extracted from 2250 pg cell biomass dry weight and 1775 pg cell biomass dry weight, respectively, because lower culture volumes were sampled.

[00259] Metabolites were analyzed using ultra-high pressure liquid chromatography (UPLC Ultimate 3000, ThermoFisher Scientific, Reinach, Switzerland) equipped with a hydrophilic interaction liquid chromatography (HILIC) column (InfinityLab Poroshell 120 HILIC-Z; 2.1x100mm, 1.9um, Agilent Technologies, Basel, Switzerland) coupled to a hybrid quadrupole-orbitrap mass spectrometer (Q Exactive Plus, ThermoFisher Scientific, Reinach, Switzerland). The solvent system consisted of 10 mM ammonium acetate, 7 pM medronic acid in ultra-pure water, pH 9 (solvent A) and 10 mM ammonium acetate, 7 pM medronic acid in acetonitrile/ultra-pure water (90:10 (V/V)), pH = 9 (solvent B). To separate metabolites, the following gradient was used for elution at a constant flow rate of 500 pL / min: 10% A for 1 minute; linearly increased to 40% A over 5 minutes; 40% A for 3 min; linearly decreased to 10% A over 0.5 minutes; and held at 10% A for 3.5 minutes. As setting for the mass spectrometry

ETHZ-27-PCT part of the method, Fourier transform mass spectrometry in negative mode with a spray voltage of -2.8 kV, a capillary temperature of 275°C, S-lens RF level of 50, an auxiliary gas flow rate of 20, and an auxiliary gas heater temperature of 350°C was applied. Mass spectra were recorded as centroids at a resolution of 70'000 at mass to charge ratio (m/z) 200 with a mass range of 75 - 800 m/z and a scan rate of ~4 Hz in full scan mode was used. Of each sample 5 uL were injected.

[00260] LC-MS results were analyzed using the emzed framework (emzed.ethz.ch). Metabolite isotopologue peaks were extracted by a targeted approach using commercial standards to define retention time - m/z peak windows applying a m/z tolerance of ±0.002 Da. In case of the metabolite P5P, a mass tolerance of 0.0015 mass units was used for the analysis. The peak area cut off was set at 20,000 counts s ¹ pL ¹ injected. For the metabolite group 2-phosphoglycerate and 3- phosphoglycerate (2PG/3PG), only 3-phosphoglycerate was verified by a commercial standard. Isotopologue fractions (Sj) and labeled fraction (LF) were determined as previously described by targeted peak integration of all detected isotopologues utilizing Eq. (1) and Eq. (2) based on m, the abundance of the respective isotopologue; n, the number of carbons in the metabolite of interest; I and j, the isotopologues.

[00261] Probably due to technical issues, the extracted metabolome of colony 4 grown at enriched ¹³ COz atmosphere exhibited low to undetectable metabolite concentrations and was not considered for further analysis.

¹³ C isotopic tracer analysis of protein-bound amino acids and total biomass

[00262] Generation of samples for tracer analysis of protein-bound amino acids was identical to ones generated for metabolite analysis. Briefly, MEcoli_ref_l was streaked out on agar plate containing minimal medium supplemented with 500 mM methanol and incubated at 37°C until colonies were visible. A cross-section of colonies was used to inoculate a preculture in 30 mL minimal medium supplemented with 500

ETHZ-27-PCT mM methanol and cultivated in baffled shake flasks at 37°C, 160 r.p.m. until stationary phase. The preculture was split into three replicates of identical conditions as described above. Cells were harvested between ODeoo 0.5 and 1.8.

[00263] Protein-bound amino acids were isolated following previously established protocols: Cell pellets were resuspended in 200 pL 6 M HCI and baked at 105°C overnight. The cell hydrolysate was dried at 95°C under constant airflow, resuspended in 1 mL water and centrifuged twice at 20,000 g for 10 min to remove insoluble debris. Samples were diluted 1:100 (V/V) in starting conditions of the LC/MS method and analyzed as described above.

[00264] ¹³ C labeling of total biomass samples was conducted by Imprint

Analytics (Imprint Analytics GmbH, Neutal, Austria) by elemental analyzer/isotope ratio mass spectrometry.

Genome resequencing

For genome resequencing, about 2 OD units of cells were sampled, centrifuged for 1 minute at 11'000 g and the supernatant discarded. Genomic DNA was extracted by MasterPure DNA purification kit (Epicentre). Purified genomic DNA was sent for Illumina NovaSeq sequencing (Novogene UK, Cambridge United Kingdom). BBMap (38.95) clumpify function was used to filter raw reads for optical and PCR duplicates. The maximum distance to consider for optical replicates was set to the appropriate value for the used sequencer (dupedist = 12000) and we allowed for one base substitution between duplicates (subs = 1). For PCR duplicates, the same substitution setting was used with two passes for error correction (passes = 2). Filtered reads were aligned to the reference genome of E. coli BW25113 (Genbank accession: CP009273) and the plasmid maps of pSEVA424 mdh2 CT4-1 Cupriavidus necator and pSEVA131 hps phi Methylobacillus flagellatus by Breseq (0.36.0) with default settings.

KEGG pathway enrichment of mutations

[00265] The set of mutations in MEcoli_ref_l was analyzed for enrichment of KEGG pathway annotations using clusterProfiler (enrichKEGG) with the following settings: organism = 'eco', pvalueCutoff = 0.05, qyvalueCutoff = 0.2, minGSSize = 3).

Serial dilution evolution for improved methylotrophic growth

ETHZ-27-PCT [00266] For the continued evolution giving rise to the methylotrophic lines and strains with improved growth rate, the methylotrophic population from the chemostat was passaged seven times in 20 mL minimal medium supplemented with 500 mM methanol, 0.1 mM isopropyl-|3-D-thiogalactopyranoside, ampicillin and streptomycin. Subsequently, this culture was split into four replicate lines (D, E, F, G). Each line was propagated at 37°C in 30 mL minimal medium containing 500 mM methanol in 100 mL baffled shake flasks and passaged into fresh medium once it reached late exponential or early stationary phase. In the beginning, cultures were incubated in a Minitron shaker (Infors-HT, Bottmingen, Switzerland) at 160 r.p.m., 50 mm throw. Later, they were moved to a Multitron shaker (Infors-HT, Bottmingen, Switzerland) at 220 r.p.m., 25 mm throw. The antibiotics ampicillin and streptomycin were added to the medium in the beginning of the evolution experiment but were later omitted. Later, antibiotics were occasionally added to test for contaminations.

Methanol dehydrogenase activity assay

[00267] The ancestral and mutated DNA sequences encoding methanol dehydrogenase were cloned into a pET16b expression vector using Gibson assembly. The resulting constructs added ten histamine residues to the N-terminus of the enzymes for nickel-immobilized metal affinity chromatography purification.

[00268] For protein expression both constructs were cultured under the same conditions: A preculture was inoculated in 10 mL LB medium supplemented with carbenicillin and incubated at 37°C, 160 r.p.m. and diluted 1:50 (V/V) in 400 mL of the same medium in 2 L baffled shake flasks on the next day. The culture was grown to mid-exponential phase (ODeoo ~ 0.7) at 37°C, 160 r.p.m. Once an ODeoo of about 0.7 was reached, the culture was induced with 0.3 mM IPTG and incubated overnight at 16°C, 160 r.p.m.

[00269] Cells were harvested by centrifugation (3250 g, 30 min, 4°C), resuspended in 10 mL lysis buffer (50 mM NaHzPO^ 300 mM NaCI, 20 mM imidazole, 2 mM dithiothreitol, Roche complete EDTA free protease inhibitor cocktail (Sigma- Aldrich Chemie GmbH, Buchs, Switzerland)) and lysed by sonication (6 mm sonication probe, amplitude 30, process time 4 min, impulse time 5 s, cool down time 15 s, Q700

ETHZ-27-PCT sonicator (Qsonica LLC, U.S.A.)). Cell debris was cleared by centrifugation (20,000 g, 40 min, 4°C). Methanol dehydrogenase was isolated from the resulting solution by fast protein liquid chromatography (AKTA, HisTrap HP, GE Healthcare, Chicago, U.S.A.) using a linear gradient from starting buffer (lysis buffer without protease inhibitor cocktail) to elution buffer (starting buffer with 500 mM imidazole) over 20 min at a flow rate of 1 mL min ^-1 . Buffer was exchanged to reaction buffer (100 mM MOPS, 5 mM MgSO4) by repeated concentration in centrifugal filter units (Amicon Ultra, 10 kDA molecular mass cutoff, Sigma-Aldrich Chemie GmbH, Buchs, Switzerland) and subsequent dilution in reaction buffer until a total dilution factor of greater than 100,000 was achieved.

[00270] Methanol dehydrogenase activity was assayed in reaction buffer supplemented with 5 mM nicotinamide adenine dinucleotide (NAD ⁺ ) as well as 500 mM methanol at 37°C and by following the formation for NADH/H ⁺ , i.e. measuring absorbance at 380 nm in a microplate reader (Tecan Infinite Pro 200, Tecan Group Ltd., Mannedorf, Switzerland). Both methanol dehydrogenase variants were added to the reaction mix at equal concentration.

Parsimonious flux balance analysis of

[00271] pFBA was conducted on the E. coli core genome model which contains 72 metabolites and 95 reactions using the cobrapy framework (0.17.1) under Python (3.7.13). To represent methanol metabolism of the methylotrophic reference strain, the RuMP cycle genes mdh, hps and phi as well as the formaldehyde detoxification genes frmA andfdh were added to the model. frmA andfdh encode S- (hydroxymethyl)glutathione dehydrogenase and formate dehydrogenase, respectively. MEcoli_ref_2 carries a deletion of tpiA, encoding triosephosphate isomerase. This deletion was introduced into the applied model.

Production of lactic acid, PHB and itaconic acid

[00272] The cryogenic glycerol stocks of MEcoli_ref_2 + pljac (lactic acid production), MEcoli_ref_2 + pl_phb (PHB production), MEcoli_ref_2 + pljta (itaconic acid production) and MEcoli_ref_2 + pl_empty (empty vector, negative control) were

ETHZ-27-PCT inoculated into 20 mL of M9 medium supplemented with 500 mM methanol, chloramphenicol, and 0.1 g L ¹ yeast extract in baffled 100 mL shake flasks. At ODeoo 0.4 to 1.0, three replicate 20 mL cultures in M9 medium containing 500 mM methanol and chloramphenicol were inoculated from the preculture. For the transfer, the number of cells required for an initial ODeoo of 0.05 were centrifuged for 2 min at 5000 g and the supernatant discarded. Subsequently, the pellet was resuspended in 900 pL M9 medium supplemented with 500 mM methanol and chloramphenicol, centrifuged again for 2 min at 5000 g, and the supernatant discarded. The remaining pellet was resuspended in the final medium. The cultures were incubated at 37°C, 220 r.p.m., 25 mm orbital shaking. At certain optical densities (pljac: ODeoo ~ 1.7, pl_phb: ODeoo ~ 0.6, pljta: OD600 ~ 1.1, pl_empty: ODeoo ~ 1.1 (lactic acid and itaconic acid); ODeoo ~ 0.9 (PHB)), production was induced by the addition of 2 pM aTc (Cayman Chemical, Ann Arbor, Michigan, USA), and the cultures subsequently incubated in the dark. Additionally, 125 mM methanol was added for lactic and itaconic acid production. Culture samples were collected after: pljac: 21 h, 45 h, 69 h; pl_phb: 74 h; pljta: 21 h, 46 h, 70 h; pl_empty: 21 h, 46 h, 70 h (lactic acid and itaconic acid); 67 h (PHB). For lactic and itaconic acid, samples were centrifuged for 2 min at 11000 g and supernatants stored at -20°C for further analysis. For PHB, cultures were centrifuged for 5 min at 11000 g. To remove any residual salt from the cells, the pellet was washed with 50 mL ultra-pure water (Mi II iQ), centrifuged again for 5 min at 11000 g and the supernatant discarded. Pellets were stored at -20°C until further analysis.

Production of ¹³ C labeled lactic acid

[00273] The same protocol as above was followed, except MEcoli_ref_2 transformed with pljac was grown in M9 medium supplemented with 500 mM ¹³ C methanol. Because the inducer aTc stock solution was prepared in methanol, the overall ¹³ C methanol atomic purity in the medium was 97.7%.

Lactic acid and itaconic acid derivatization

[00274] The organic acids in the supernatant collected from the cultivation were derivatized together with 200 pM propionate as internal standard according to a modified version of a previously reported protocols. The supernatant samples and the internal standard were mixed and diluted 20-fold in 50% (V/V) acetonitrile. Organic

ETHZ-27-PCT acids were derivatized by addition of 34 mM 3-nitrophenylhydrazine (3NPH) and 21 mM N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide (EDC), followed by incubation for 30 min at 40°C, with continuous shaking. The reaction was quenched by the addition of 0.25 volumes of 0.1% (V/V) formic acid and diluted 10-fold in 50% (V/V) acetonitrile for LC-MS measurement.

PHB extraction, depolymerization and derivatization to methyl 3-hydroxybutanoate

Cell pellets were dried by lyophilization overnight, weighed, and transferred to an airtight glass tube. 2 mL of 3% (V/V) H2SO4 in methanol containing 200 pg/mL of benzoic acid as internal standard and 2 mL of chloroform were added to the pellet. PHB standards were prepared in chloroform and mixed 1:1 (V/V) with 3% (V/V) H2SO4 in methanol containing benzoic acid. For PHB extraction, depolymerization by methanolysis, and derivatization, samples were incubated for 2.5 h in a boiling water bath. To initiate phase separation, 1 mL of ultra-pure water (MilliQ) was added to the solution and the samples subsequently incubated for 10 min in a sonication bath. The aqueous phase (upper) was discarded and the organic phase (lower) used for GC analysis. Note, different amounts of biomass were sampled of the negative control and the pl_ph b strain. To account for this, the gas chromatography traces were scaled accordingly.

Quantification of lactic acid and itaconic acid by LC-MS

The 3-nitrophenylhydrazone derivatives of lactic and itaconic acid were analyzed using ultra-high pressure liquid chromatography (UPLC Ultimate 3000, ThermoFisher Scientific, Reinach, Switzerland) with a C18 column with isobutyl side chains and tetramethylsylene end capping (Kinetex Core-Shell Technology XB-C18; 2.1 x 50 mm, 1.7 pm particle size, 100 A pore size, Phenomenex, Aschaffenburg, Germany) coupled to a hybrid quadrupole-orbitrap mass spectrometer (Q Exactive Plus, ThermoFisher Scientific, Reinach, Switzerland). Solvents were 0.1% (V/V) formic acid in ultra-pure water (solvent A) and in acetonitrile (solvent B).

To separate metabolites, the following gradient was used for elution at a constant flow rate of 500 pL/min: 100% A linearly decreased to 5% in 3 min, then held at 5% A for another two minutes. Fourier transform mass spectrometry was performed in negative mode with a spray voltage of -3.0 kV, a capillary temperature of 275°C, S-lens RF level

ETHZ-27-PCT of 50, an auxiliary gas flow rate of 20, and an auxiliary gas heater temperature of 350°C. Mass spectra were recorded as centroids at a resolution of 35,000 at mass to charge ratio (m/z) 200 with a mass range of 150-1000 m/z and a scan rate of ~4 Hz in full scan mode. Of each sample, 2 pL was injected. Lactic and itaconic acid were quantified using external standards of sodium DL-lactate solution (Sigma-Aldrich) and itaconic acid (Chemie Brunschwig AG, Basel, Switzerland), respectively, derivatized with internal standard as described above.

[00275] LC-MS results were analyzed using the emzed framework (emzed.ethz.ch). Metabolite peaks were extracted by a targeted approach using commercial standards to define retention time - m/z peak windows applying an m/z tolerance of 5 ppm. Integration was performed using trapezoid integration. Lactic acid levels below 18 mg/L were not quantifiable due to noise. The cutoff noise level was determined using peaks corresponding to the mass expected for lactic acid in the itaconic acid standards.

Methanolyzed PHB (methyl 3-hydroxybutanoate) was analyzed by gas chromatography (GC 6850, Agilent Technologies, Basel, Switzerland) equipped with a 7683B Series injector coupled to a flame ionization detector (FID). A DB-WAX column (15 m x 0.32 mm x 0.50 pm; Agilent Technologies, Basel, Switzerland) was used for metabolite separation with helium as the carrier gas at a flow of 30 mL/min. The following temperature gradient was applied: 3.5 min from 90°C to 230°C, 3 min at 230°C, 1.25 min to 90°C, 2.5 min at 90°C. 1 uL of sample was injected. The split ratio was 2.0 and the detector temperature set to 270°C. Peaks were confirmed by standards. Benzoic acid was used as internal standard to correct for methodological variation.

ETHZ-27-PCT