In the OSTI Collections: Biosynthesis

Dr. Watson computer sleuthing scientist

Article Acknowledgement:

Dr. William N. Watson, Physicist

DOE Office of Scientific and Technical Information

Separator line

By sets of interrelated processes that appear far more complex than any of human design, living cells use raw materials from their environment to synthesize the substances that they and other cells use.  Since the processes are so complex (and microscopic), they have been hard to discover and are still only partially understood.  But since understanding them improves our appreciation of how living things function, as well as our ability to manage and use them for nutrition, medicine, energy, and other purposes, researchers continue to gradually work out precisely how the many organisms in Earth’s ecosystem synthesize the huge variety of molecules they produce.   A sample of papers about such investigations available through DoE PAGES suggests the wide range of questions they address. 


Separator line


Proline, one of the amino acids that organisms synthesize into proteins, is itself synthesized in complex plants by two different sets of reactions that have a common last step—one catalyzed[Wikipedia] by an enzyme (P5C reductase) whose activity appears to have a complex regulation mechanism.  As with other chemical processes, the way this regulation works is largely determined by how the atoms are arranged in the molecules involved.  To figure out the enzyme’s atomic arrangements both during and between catalyses, researchers at the National Cancer Institute, Argonne National Laboratory, and the University of Ferrara in Italy examined something that depended on those arrangements:  how crystals of P5C-reductase complexes[Wikipedia] and of pure P5C reductase affected x-rays of different wavelengths.  P5C forms complexes with proline and other molecules when it’s catalyzing the synthesis of proline; at other times, P5C molecules’ atoms are arranged differently.  The different crystals’ effects on x-rays provided enough clues to infer the atomic arrangements of both the pure and complexified enzyme.[DoE PAGES]  These findings did explain some previously-observed facts about proline synthesis, and also suggested some possible means of controlling enzyme action. 

The correlation of structural form and function is not confined to individual biomolecules.  The organelles[Wikipedia] that constitute functional parts of many cells have their own structures, which can alter as the organelles perform various functions.  The endoplasmic reticulum[Wikipedia], an organelle that consists of a network of membranes, is known to be a site for biosynthesizing many primary metabolites—substances that an organism needs for reproduction or for long-term survival.  The localized biosynthesis of secondary metabolites, which are important but serve less immediate needs, has received less attention, but some recent experiments[DoE PAGES] showed that the endoplasmic reticula in cells of Fusarium graminearum fungi, which produce toxins that can devastate wheat and barley, reorganized their endoplasmic reticula when stimulated to produce those toxins.  In part of the investigation, a particular strain of F. graminearum, stained with a blue fluorescent dye that accumulated on the fungal cells’ endoplasmic reticula, was grown in different media, one of which induced the fungus to produce the toxins.  The F. graminearum strains used include some that have been genetically altered to synthesize one of the key enzymes in its toxin-producing reaction sequence with a green fluorescent tag.  Fluorescence of both the dye and the enzyme tag showed that the enzyme localized on the cells’ endoplasmic reticula, and that when toxin synthesis was induced, the endoplasmic reticula changed shape. 




Figure 1


Figure 1.  Fusarium graminearum fungi grown in media that contain minimal amounts of nutrients for growth (left) and in media that stimulate the fungi to produce trichothecene[Wikipedia] toxins (right).  Different strains of the fungus, each genetically altered to produce a green-fluorescing version of one of the enzymes involved in the toxin production, are shown in parts a, b, and c; parts d and e show fungi of the same wild strain.  The fungi are also stained by a blue-fluorescing dye that accumulates on their endoplasmic reticula.  The fluorescence of the protein and the dye show that the enzyme localized on the endoplasmic reticula, which changed shape when toxin synthesis was induced.  (From “Structural reorganization of the fungal endoplasmic reticulum upon induction of mycotoxin biosynthesis”[DoE PAGES], p. 4.) 



Many studies explore how cells’ metabolite production is regulated.  The three studies described below were conducted by different research groups, but focused on different biosyntheses that occur in the same plant species, Arabidopsis thalianaA. thaliana is generally considered a weed, though its combination of complexity with a relatively small set of genetic material have made it useful to biologists.[Wikipedia] 


The title of the Plant Cell paper “Phosphorylation of WRINKLED1 by KIN10 Results in its Proteasomal Degradation, Providing a Link Between Energy Homeostasis and Lipid Biosynthesis”[DoE PAGES] summarizes the findings of one investigation that added a few pieces of information about how a cell’s metabolism is regulated.  “WRINKLED1” is the name given to one the many proteins that control when portions of DNA are transcribed[Wikipedia] to RNA, which is the first step in translating the DNA’s information into cell activity.  In A. thaliana, WRINKLED1 regulates both the biosynthesis of lipids and a reaction sequence that releases energy from the simple sugar glucose[Wikipedia] to form molecules of two important substances, one of which is adenosine triphosphate (ATP), the medium of energy transfer within cells[Wikipedia].  “KIN10” is an enzyme that can catalylze the transfer of phosphate groups from ATP to other molecules.  The investigators found that KIN10 catalyzes phosphate-group transfers to two sites of the WRINKLED1 protein, which was found to result in its degradation; otherwise-similar proteins with different amino-acid residues at the two sites weren’t so degraded in the presence of KIN10.  Thus when sugar levels in A. thaliana cells are high and KIN10 is inhibited, WRINKLED1 proteins are degraded less and the biosynthesis of lipids[Wikipedia] is favored; but when sugar levels are low, KIN10 regains activity, WRINKLED1 proteins are more degraded, curtailing both lipid biosynthesis and energy release from glucose. 


A different study, reported in BMC Plant Biology[DoE PAGES], examined one substance of a type whose biosynthesis and function remain little understood.  The researchers produced various lines of evidence that a particular gene affects A. thaliana’s biosynthesis of two types of side chain in the structure of rhamnogalacturonan-I (RG-I), a complex polysaccharide[Wikipedia] found in the walls of plant cells that’s critical for the growth of pollen tubes.  The researchers found that mutations of this gene weren’t transmitted by pollen, so they couldn’t produce A. thaliana plants that completely lacked the normal gene, but in plants germinated in vitro that had one normal and one mutated version of the gene, they found reduced rates of pollen-tube formation.  Silencing the corresponding genes[Wikipedia] in different plants, of the species Nicotiana benthamiana, resulted in reduced expansion of their internodes and petioles, and less galactose in their cell wall materials.  Other analyses indicated that the N. benthamiana plants had less branching and less of one type of side chain in their RG-I molecules, while A. thaliana lines that overexpressed the gene had more of the other side-chain type, grew their central stems less dominantly over their side stems, and exhibited several other alterations. 


Figure 2


Figure 2.  A gene that appears to affect how Arabidopsis thaliana plants biosynthesize the cell-wall constituent rhamnogalacturonan-I also seems to have a similar importance in Nicotiana benthamiana plants, as seen by how normal N. benthamiana growth (left) is affected when that gene is silenced[Wikipedia] (right).  (From “A DUF-246 family glycosyltransferase-like gene affects male fertility and the biosynthesis of pectic arabinogalactans”[DoE PAGES], p. 7 of 17.)



The third study deals with suberin[Wikipedia], one of the most abundant lipid-based polymers[Wikipedia] in nature.  Suberin controls water and solute movement in plants, particularly in A. thaliana seed coats, where it also protects the seeds from infection.  According to a Plant Physiology paper[DoE PAGES], more is known about genes that are responsible for synthesizing suberin than about the regulatory mechanism that controls its biosynthesis.  The authors show that in the A. thaliana seed coat, a particular protein (MYB107) regulates transcription of genes for suberin biosynthesis into RNA.  Disrupting this protein suppresses the expression of genes involved in the biosynthesis of suberin but not cutin, a main component of leaves’ protective film.  MYB107 disruption also lowers the accumulation of seed coat suberin and alters its structure, thus making the seed coats more permeable and susceptible to stresses.  MYB107 was also found to bind directly to portions of DNA that initiate the expression of suberin-biosynthesis genes, which further verified its primary role in regulating that expression. 



Figure 3

Figure 3.  The disruption of MYB107 protein suppresses the expression of genes involved in the biosynthesis of suberin, which controls water and solute movement in the seed coats of A. thaliana plants.  Above are transmission electron micrographs of suberin ultrastructures in A. thaliana seed coats of the wild type (WT) and in plants in which the action of the protein MYB107 is disrupted (myb107).  Mature seeds were ultrathin sectioned, and the sections (90–100 nm) were treated with 10% H2O2 and then stained with 1% aqueous uranyl acetate and Sato lead. Enlarged images from the squared areas in A are shown in B, and enlarged images from the squared areas in B are shown in C. bpl, Brown pigment layer; oi, outer integument. Arrows point to suberin lamellae in C. Bars = 2 mm (A), 500 nm (B), and 100 nm (C).  (After “The MYB107 Transcription Factor Positively Regulates Suberin Biosynthesis”[DoE PAGES], p. 1050.)



Separator line



Other explorations focus on biosynthetic processes or products that are more directly useful to human beings.  Some of these deal only with how a process works, while others are oriented more toward how we might use it. 


One investigation reported in BMC Genomics[DoE PAGES] is expected to facilitate the genetic improvement of Bixa orellana, whose seeds produce and store the commercially important orange-red pigment bixin[Wikipedia].  RNA from young leaves of B. orellana and two different developmental stages of seeds were used to construct libraries of the RNA that the plant’s cells use to convey biosynthesis instructions from their DNA to the ribosomes[Wikipedia] where bixin (and other proteins) are synthesized.  Combining this data with information from other genetic databases led to a hypothetical model of bixin biosynthesis that involves coordinated activation of genes for synthesizing bixin, carotenoid (another pigment), and methylerythritol phosphate (a substance formed during one step of a major lipid-synthesis reaction sequence[Wikipedia]). 


Experiments with the bacterial strain Rhodococcus jostii RHA1 under various conditions revealed genes that are involved in the strain’s biosynthesis of triacylglycerols[DoE PAGES]  The results may inform the design of other R. jostii strains that might produce more of these substances, whose various industrial uses include the manufacture of biodiesel fuels.  Analyzing the RNA of RHA1 bacteria grown on media that provided widely differing amounts of nitrogen (in ammonium chloride) showed that some genes were expressed more by RHA1 in limited-nitrogen media, including one gene whose overexpression was associated with a 20% increase of triacylglycerol accumulation, while other genes were dysregulated. 


One organism that already has strains designed for fuel and chemical production, the yeast Saccharomyces cerevisiae, was recently experimented with to address an inherent limitation on its ability to continuously convert sugars into products derived from one if its central metabolites and molecular building blocks, acetyl coenzyme A (acetyl-CoA)[Wikipedia].  Metabolism of glucose produces pyruvate, which gets converted into acetyl-CoA, but in yeast cells that are respiring this process is limited as pyruvate gets directed toward the yeast cells’ mitochondria, organelles which generate ATP.  And in yeasts that are fermenting instead of respiring, pyruvate is directed towards producing ethanol, again limiting its conversion to acetyl-CoA and its derivatives.  But an alternative way to synthesize acetyl-CoA was recently tested[DoE PAGES] as part of an effort to increase S. cerevisiae’s output of the acetyl-CoA derivative mevalonate, a negatively charged ion of mevalonic acid[Wikipedia] that is pharmaceutically important (and formed in one step of another major lipid-synthesis reaction sequence[Wikipedia]).  Researchers at Lawrence Berkeley National Laboratory, the University of California at Berkeley, and the Joint BioEnergy Institute modified S. cerevisiae in three ways:  having the cells catalyze more production of acetyl-CoA from citrate instead of pyruvate by keeping the citrate from being diverted into a different reaction (which is accomplished by deleting an enzyme that facilitates that reaction); increasing the throughput of the reaction sequence that turns acetyl-CoA into mevalonate; and replacing the enzyme that promotes the further conversion of mevalonate into other substances with a different enzyme so that mevalonate would accumulate in the cell.  The researchers showed that these measures significantly increased mevalonate production, and expected that the same strategy could improve the efficiency of other S. cerevisiae strains’ production of fuels and other chemicals derived from acetyl-CoA. 


While S. cerevisiae yeast metabolizes sugar into various other substances, some bacterial species can similarly make new molecules by metabolizing methane[Wikipedia]Methylomicrobium buryatense is a promising species for making valuable fatty-acid derivatives starting from methane, since it’s easy and cheap to cultivate, grows fast, and can be operated on by established genetic tools.  Researchers at San Diego State University and UC San Diego investigated different ways to improve fatty-acid accumulation in M. buryatense[DoE PAGES] and found limitations from fatty-acid degradation in the bacteria and from the levels of acetyl-CoA and malonyl-CoA.  The researchers also identified and studied a particular gene for regulating M. buryatense’s fatty-acid biosynthesis, observing how differently the bacteria behaved when the gene was deleted or overexpressed, and thus discovering a set of enzymes involved in fatty-acid biosynthesis as well as other regulator genes. 


Microbial metabolism of methane produces useful outputs, but the existence of this metabolism itself is useful to us since it limits the amount of atmospheric methane that slows the radiation of heat from Earth into space.  In particular, microbial communities that inhabit undersea methane seeps[Wikipedia] limit the flow of methane from beneath the ocean floor into the water, from which it can escape into the atmosphere and trap heat.  To trace how proteins are synthesized in these ecosystems and understand them better, researchers at Caltech and Oak Ridge National Laboratory fed mixed microbial communities with ammonium ions and methane molecules whose nitrogen and carbon atoms, respectively, had one more neutron than average to make them more massive.[DoE PAGES]  The ammonium and methane taken up by the microbes were metabolized into various proteins that contained the more massive nitrogen and carbon atoms, thus allowing those proteins to be separated from others by weight[Wikipedia] and enabling the microbes’ protein syntheses to be tracked.  The 3,495 proteins identified in the experiments, including 26 detected in all the experiments whose function was unknown, suggested several characteristics of methane seep ecosystems, and demonstrated their heterogeneity and broad functional diversity. 



Figure 4

Figure 4.  A schematic diagram of the sampling approach used for the proteomic stable-isotope probing[Wikipedia] in this study.  Sediment was collected from active seep areas at Hydrate Ridge North[Wikipedia] for incubation #5133 and Hydrate Ridge South for #3731.  Sediment from both sample sites was used as inoculum in parallel methane-infused incubations, with one bottle from each set receiving NH4Cl made with nitrogen-14 at a concentration of 1 millimole per cubic meter, and the other receiving an equal concentration of NH4Cl made with nitrogen-15.  Both #3731 incubations were sampled for proteomic stable-isotope probing after 17 and 326 days; both #5133 incubations were sampled for proteomic stable-isotope probing after 160 days.  (After “Proteomic Stable Isotope Probing Reveals Biosynthesis Dynamics of Slow Growing Methane Based Microbial Communities”[DoE PAGES], p. 4.)




Figure 5

Figure 5.  Data for enzymes involved in the metabolic pathway that removes methane from the environment anaerobically[Wikipedia].  Filled boxes indicate an enzyme constituent’s presence in the set of proteins expressed by the associated sample.  (From “Proteomic Stable Isotope Probing Reveals Biosynthesis Dynamics of Slow Growing Methane Based Microbial Communities”[DoE PAGES], p. 13.) 




Separator line



When you check your idea of reality against reality and the two don’t match, you know that at least something is wrong with your idea.  With enough checking, you might even be able to determine exactly what’s wrong with it.  Evidence that something was wrong with various mathematical models of bacterial metabolism came when their predictions of energy yields proved inaccurate.  This inaccuracy has been traced to the models’ poor representations of key chemical reaction pathways in energy biosynthesis and the multistep transfer of electrons[Wikipedia] from donors to acceptors.  To overcome this, researchers at the University of Chicago and Argonne National Laboratory developed means for building more accurate models based on a well-studied set of diverse organisms—over 8,000 microbial genomes.[DoE PAGES]  They compared the models to explore core metabolic pathways across all microbial life, and examined what the models implied about microbes’ biosynthesis of ATP and essential precursors of biomass.  6,600 of the models implied the existence of some type of electron-transporting reaction sequence in which oxygen is the final electron acceptor.  5,100 of the models had an electron-transport sequence in which oxygen wasn’t an electron acceptor, and 1,279 models didn’t include any sequence of electron-transporting reactions.  While “nearly 5,586” of the models predicted accurate yields of ATP according to their report, 2,495 models of microbial core metabolism predicted no ATP production under any of the conditions tested, thus revealing that some information gaps about central reaction pathways remained.  The researchers then established a way to systematically identify and correct inconsistent information in the models, leading to accurate predictions of energy yields.  They have also made their analyses available for others to explore. 


Although knowing which molecules interact in what reactions would represent an extensive understanding of biosynthesis, it doesn’t address all the details of how the interactions happen.  Indeed, as one other paper indicates[DoE PAGES], one may need to account for fine details of molecular structure to even be sure of which molecules are involved in a metabolic pathway.  Such a detailed model of a single catalyst, without reference to entire genomes of any species or large set of species, was examined by researchers with the University of Tennessee, Oak Ridge National Laboratory, and China’s Shandong Agricultural University.  The first transfer of methyl (CH3) in the reaction series that produces caffeine is catalyzed by an enzyme called xanthosine methyltransferase (XMT) by a mechanism whose nature was not clear before the researchers mathematically simulated its action using known quantum-mechanical laws.  Contrary to what has been widely assumed, the model indicates that electrically neutral XMT cannot catalyze the methyl transfer.  On the other hand, the model indicates that a slightly different molecule can catalyze it—namely, XMT ionized to have an extra electron’s worth of charge.  The paper about these findings cites additional evidence from measurements and other calculations that corroborate this conclusion, and discusses the implications for caffeine biosynthesis. 



Figure 6


Figure 6.  The first transfer of methyl (CH3) in the reaction series that produces caffeine is catalyzed by the enzyme xanthosine methyltransferase (XMT).  A mathematical model of two candidate transfer processes shows that neutral XMT would be greatly hindered from catalyzing the transfer, since the energy of the reactants would have to be quite high during the process, while XMT with an electron’s worth of negative charge could catalyze the transfer with relative ease.  (From “Understanding the catalytic mechanism of xanthosine methyltransferase in caffeine biosynthesis from QM/MM molecular dynamics and free energy simulations”[DoE PAGES], page A.)




Separator line



The complexity of a living cell’s interrelated metabolic processes seems only dimly conveyed by the brief descriptions above.  The individual research reports give a somewhat clearer impression, but even they deal mostly with the particular features of the metabolic network that were addressed by the research.  A better idea of existing discoveries about how the various biosynthetic and other processes interact to regulate an organism’s metabolism may be gained from more extensive sets of information, like the numerous diagrams of individual metabolic pathways for several species at the metaTIGER website of the University of Leeds Bioinformatics Group, or the online overview map of Escherichia coli K-12 substr. MG1655 metabolism at



Separator line







Separator line

Papers available through DoE PAGES















Separator line

Additional references



  • has a database about the bacterial substrain Escherichia coli K-12 MG1655 that includes an overview map of the bacterium’s metabolism.