Researchers have made a breakthrough that could eventually lead to a more effective treatment for tuberculosis.
Tuberculosis is one of the top 10 causes of death worldwide, according to the World Health Organization.
A team of scientists, including Professor Colin Jackson from The Australian National University (ANU), has solved the mystery of how a cofactor called F420, found in the bacterium behind tuberculosis, is made.
Cofactors like F420 help enzymes to speed up chemical reactions.
Cofactor F420 is a deazaflavin that acts as a hydride carrier in diverse redox reactions in both bacteria and archaea.
While F420 structurally resembles the flavins flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), it functions as an obligate two-electron hydride carrier and hence is functionally analogous to the nicotinamides NAD+and NADP+3.
The lower reduction potential of the F420, relative tothe flavins, results from the substitution of N5 of the isoalloxazinering in the flavins for a carbon in F420.
Originally characterized from methanogenic archaea in 1972 F420 is an important catabolic cofactor in methanogens and mediates key one-carbon transformations of methanogenesis.
F420 has since been shown to be synthesized in a range of archaea and bacteria.
In Mycobacterium tuberculosis, the causative agent of tubercu-losis, F420 has been shown to contribute to persistence and to activate the new clinical antitubercular prodrugs delamanid and pretomanid.
There are also growing numbers of natural pro-ducts that have been shown to be synthesized through F420-dependent pathways, including tetra cyclines12, lincosamides, and thiopeptides
F420-dependent enzymes have also been explored for bioremediation and biocatalytic applications.
The currently accepted F420 biosynthetic pathway consists oftwo branches.
In the first branch, tyrosine is condensed with 5-amino-6-ribitylamino-2,4[1H,3H]-pyrimidinedione from the flavin biosynthetic pathway to generate the deazaflavin chro-mophore Fo (7,8-didemethyl-8-hydroxy-5-deazariboflavin) via the activity of the two-domain Fo synthase FbiC, or the CofG/H pair(where “Fbi”refers to mycobacterial proteins and “Cof”refers toarchaeal homologs).
In the second branch, it has been hypothe-sized that a 2-phospho-L-lactate guanylyltransferase (CofC inarchaea and the putative enzyme FbiD in bacteria) catalyzes the guanylylation of 2-phospho-L-lactate (2-PL) using guanosine-5ʹ-triphosphate (GTP), yielding L-lactyl-2-diphospho-5ʹ-guanosine(LPPG).
The two branches then merge at the reaction catalyzed by the transferase FbiA/CofD, where the 2-phospho-L-lactylmoiety of LPPG is transferred to Fo, forming F420.
Finally,the γ-glutamyl ligase (FbiB/CofE) catalyzes the poly-glutamylation of F42 to generate mature F420, with poly-γ-glutamate tail lengthsof ~2–8, depending on species.
There are three aspects of the F420 biosynthetic pathway that require clarification.
First, the metabolic origin of 2-PL, the proposed substrate for CofC, is unclear.
It has been assumed thata hypothetical kinase (designated CofB) phosphorylates L-lactateto produce 2-PL22.
However, no such enzyme has been identifiedin bacteria or archaea, and our genomic analysis of F420 bio-synthesis operons has failed to identify any candidate enzymeswith putative L-lactate kinase activity.
Second, the existence of FbiD has only been inferred through bioinformatics and genetic knock out studies and the enzyme has not been formally char-acterized.
Finally, the bacterial γ-glutamyl ligase FbiB is atwo-domain protein2, in which the N-terminal domain is homologous to other F420-γ-glutamyl ligases (including the archaeal equivalent, CofE) and the C-terminal domain adopts an FMN-binding nitro reductase (NTR) fold.
Although both domains are required for full γ-glutamyl ligase activity, no function has been associated with either the C-terminal domain or the FMN cofactor, given no redox reactions are known to beinvolved in F420 biosynthesis.
Here we demonstrate that 2-PL is not required for F420 bio-synthesis in pro karyotes and instead phosphoenolpyruvate (PEP),an abundant intermediate of glycolysis and gluco neogenesis, is incorporated into F420.
Mass spectrometry (MS) and protein crystallography are used to demonstrate that PEP guanylylation iscatalyzed by the FbiD/CofC enzymes that were previous lythought to act upon 2-PL.
In bacteria, the incorporation of PEP inthe pathway results in the production of the previouslyun detected intermediate dehydro-F420-0, which we identified byMS.
We then showed, with the help of ligand docking, that this intermediate is then reduced by the C-terminal domian and poly-glutamylated by the N-terminal domain.
These findings result ina substantially revised pathway for F420 biosynthesis and have allowed us to heterologously express a functional F420 biosyn-thetic pathway in Escherichia coli, an organism that does notnormally produce F420, at levels comparable to some native F420-producing organisms.
The N-terminal domain ishomologous to other F420-γ-glutamyl ligases (including thearchaeal equivalent, CofE) and the C-terminal domain adopts anFMN-binding nitroreductase (NTR) fold.
Although both domains are required for full γ-glutamyl ligase activity, nofunction has been associated with either the C-terminal domainor the FMN cofactor, given no redox reactions are known to be involved in F420 biosynthesis.
Here we demonstrate that 2-PL is not required for F420 bio-synthesis in prokaryotes and instead phosphoenolpyruvate (PEP),an abundant intermediate of glycolysis and gluconeogenesis, isincorporated into F420.
Mass spectrometry (MS) and protein crystallography are used to demonstrate that PEP guanylylation iscatalyzed by the FbiD/CofC enzymes that were previously thought to act upon 2-PL.
In bacteria, the incorporation of PEP inthe pathway results in the production of the previously undetected intermediate dehydro-F420-0, which we identified byMS
We then showed, with the help of ligand docking, that this intermediate is then reduced by the C-terminal domian and poly-glutamylated by the N-terminal domain.
These findings result in a substantially revised pathway for F420 biosynthesis and have allowed us to heterologously express a functional F420 biosyn-thetic pathway in Escherichia coli, an organism that does not normally produce F420, at levels comparable to some native F420-producing organisms.
While F420 structurally resembles the flavins flavin mononucleotide (FMN) and flavin adenine dinucleotide (FAD), it functions as an obligate two-electron hydride carrier andhence is functionally analogous to the nicotinamides NAD+and NADP+3.
The lower reduction potential of the F420, relative to the flavins, results from the substitution of N5 of the isoalloxazinering in the flavins for a carbon in F420.
Originally characterized from methanogenic archaea in 1972,F420 is an important catabolic cofactor in methanogens and mediates key one-carbon transformations of methanogenesis.
F420 has since been shownto be synthesized in a range of archaea and bacteria.
In Mycobacterium tuberculosis, the causative agent of tubercu-losis, F420 has been shown to contribute to persistence and to activate the new clinical anti tubercular prodrugs delamanid andpretomanid.
There are also growing numbers of natural pro-ducts that have been shown to be synthesized through F420-dependent pathways, including tetracyclines, lincosamides, and thiopeptides.
F420-dependent enzymes have also been explored for bioremediation and biocatalytic applications.
The currently accepted F420 biosynthetic pathway consists of two branche.
In the first branch, tyrosine is condensedwith 5-amino-6-ribitylamino-2,4[1H,3H]-pyrimidinedione fromthe flavin biosynthetic pathway to generate the deazaflavin chro-mophore Fo (7,8-didemethyl-8-hydroxy-5-deazariboflavin) via the activity of the two-domain Fo synthase FbiC, or the CofG/H pair(where “Fbi”refers to mycobacterial proteins and “Cof”refers toarchaeal homologs).
In the second branch, it has been hypothe-sized that a 2-phospho-L-lactate guanylyltransferase (CofC inarchaea and the putative enzyme FbiD in bacteria) catalyzes theguanylylation of 2-phospho-L-lactate (2-PL) using guanosine-5ʹ-triphosphate (GTP), yielding L-lactyl-2-diphospho-5ʹ-guanosine(LPPG)17.
The two branches then merge at the reaction catalyzedby the transferase FbiA/CofD, where the 2-phospho-L-lactylmoiety of LPPG is transferred to Fo, forming F420-018,19.
Finally,the γ-glutamyl ligase (FbiB/CofE) catalyzes the poly-glutamylation of F420 to generate mature F420, with poly-γ-glutamate tail lengths of ~2–8, depending on species.
There are three aspects of the F420 biosynthetic pathway that require clarification.
First, the metabolic origin of 2-PL, theproposed substrate for CofC, is unclear.
It has been assumed thata hypothetical kinase (designated CofB) phosphorylates L-lactateto produce 2-PL22.
However, no such enzyme has been identified in bacteria or archaea, and our genomic analysis of F420 bio-synthesis operons has failed to identify any candidate enzymeswith putative L-lactate kinase activity2.
Second, the existence of FbiD has only been inferred through bio informatics and gene ticknockout studies and the enzyme has not been formally char-acterized.
Finally, the bacterial γ-glutamyl ligase FbiB is atwo-domain protein, in which the N-terminal domain ishomologous to other F420-γ-glutamyl ligases (including the archaeal equivalent, CofE) and the C-terminal domain adopts anFMN-binding nitroreductase (NTR) fold20.
Although bothdomains are required for full γ-glutamyl ligase activity, no function has been associated with either the C-terminal domainor the FMN cofactor, given no redox reactions are known to beinvolved in F420 biosynthesis.
Here we demonstrate that 2-PL is not required for F420bio-synthesis in prokaryotes and instead phosphoenolpyruvate (PEP),an abundant intermediate of glycolysis and gluconeogenesis, is incorporated into F420.
Mass spectrometry (MS) and protein crystallography are used to demonstrate that PEP guanylylation iscatalyzed by the FbiD/CofC enzymes that were previously thought to act upon 2-PL.
In bacteria, the incorporation of PEP inthe pathway results in the production of the previously undetected intermediate dehydro-F420-0, which we identified by MS.
We then showed, with the help of ligand docking, that this intermediate is then reduced by the C-terminal domian and poly-glutamylated by the N-terminal domain.
These findings result ina substantially revised pathway for F420 biosynthesis and have allowed us to heterologously express a functional F420 biosyn-thetic pathway in Escherichia coli, an organism that does not normally produce F420, at levels comparable to some native F420-producing organisms.
Professor Jackson said this breakthrough could help identify new drug targets for tuberculosis.
“F420 is found in Mycobacterium tuberculosis – the bacteria which causes tuberculosis. But is not synthesised by the human body,” Professor Jackson said.
“For decades, people have been unsure about how F420, has been made.
We were able to go through and identify all the different enzymes involved in making this cofactor.
“Understanding its make-up could allow scientists to better target the disease in patients.
This is particularly significant as TB is the world’s deadliest infectious diseases, claiming over one million lives each year.”
Now researchers know how the cofactor is made in Mycobacterium tuberculosis, they can also produce it in other organisms—helping unlock safer and cleaner biotechnology applications.
“This is really important for what we call ‘green chemistry’,” Professor Jackson said.
“Instead of manufacturing a certain chemical using toxic solvents or high heat, we can now use enzymes that use this cofactor to do it in more environmentally-friendly conditions.”
Their research has been published in the journal Nature Communications.
More information: Ghader Bashiri et al. A revised biosynthetic pathway for the cofactor F420 in prokaryotes, Nature Communications (2019). DOI: 10.1038/s41467-019-09534-x
Journal information: Nature Communications
Provided by Australian National University
