There are 20 major types of amino acids found in proteins, of which the differences are the side chains R groups that contain various chemical structures. This R group gives each amino acid and, finally, each protein-specific characteristic.

These features include size, shape, hydrophilicity, hydrophobicity, interactions, polarity, and pH level. Each of these characteristics is crucial for the stability of the proteins in the human body and environment. As building blocks of proteins, amino acids are essential for multiple biological processes, including cell growth, division, and metabolic signaling pathways.

However, the dysregulation of pathways involved in amino acid biogenesis and catabolism have been characterized in multiple inborn metabolic disorders such as phenylketonuria PKU , alkaptonuria, and maple syrup urine disease MSUD.

Gly, Ala, Val, Leu, Met, and Ile have aliphatic R groups, and Phe, Tyr, and Trp have aromatic R groups. The aliphatic or aromatic group makes the amino acids hydrophobic also called "water fear" or group with no tendency to be close to aqueous solutions.

The globular proteins will opportunistically bury the hydrophobic side chains inside the protein interior by folding into a three-dimensional shape in aqueous solutions. The amino acids with polar uncharged R groups are serine S, Ser, HO-CH2-CH NH2 -COOH , threonine T, Thr, CH3-CH OH -CH NH2 -COOH , cysteine C, Cys, HS-CH2-CH NH2 -COOH , proline P, Pro, NH- CH2 3-CH-COOH , asparagine N, Asn, H2N-CO-CH2-CH NH2 -COOH and glutamine Q, Gln, H2N-CO- CH2 2-CH NH2 -COOH.

The side chains of these amino acids possess functional spectrum groups. Most have one or more atoms, such as oxygen, nitrogen, or sulfur, with electron pairs, allowing hydrogen bonding to water or other molecules.

Also, aspartate D, Asp, HOOC-CH2-CH NH2 -COOH and glutamate E, Glu, HOOC- CH2 2-CH NH2 -COOH are amino acids with negatively charged R groups. The amino acids are linked with their neighbors in a specific order by covalent bonds, also known as peptide bonds.

These particular bonds are the amide linkages that form when the amino group reacts with the carboxylate carbon connecting two amino acids. The free amino group at one end of the polypeptide is typically called the amino-terminal or N-terminal. In contrast, the open carboxyl group at the other end is labeled as the carboxyl-terminal or C-terminal.

Protein sequences are written or read from the N-to-C terminal direction. The chains of the amino acids or progression of the amino acids distinguish exquisitely one protein from another. The organism's DNA is specific in coding a particular sequence of amino acids. Each protein consists of one or more polypeptide chains.

Proteins are polymers of 50 or more amino acids, while peptides are shorter amino acid polymers. A protease, which is also known as peptidase or proteinase, is an enzyme that catalyzes proteolysis. This phenomenon is constituted by the breakdown of proteins into smaller polypeptides and, eventually, single amino acids.

Proteases cleave the peptide bonds within proteins by hydrolysis, which is a chemical reaction where the water breaks bonds. Acids, alkalis, or enzymes may be employed to determine protein hydrolysis. The general functions of amino acids include the involvement in protein synthesis, biosynthetic products, and metabolic energy.

Essentially, there is a crucial difference between positive and negative nitrogen balance, which is critical for understanding amino acid metabolism.

In a positive balance, the nitrogen consumed is more considerable than the nitrogen excreted, while in a negative balance, the nitrogen consumed is less than the nitrogen excreted. A positive balance denotes net protein synthesis.

It occurs when the organism is recovering from starvation, growth, and pregnancy. In contrast, a negative balance entails mobilization of the amino acids, tissue necrosis, or a poor-quality condition of the human body as a consequence of 3rd-degree burns or significant surgical operations.

The amino acids subdivide into essential and non-essential. There are amino acids that need to be obtained directly diet and amino acids that can be synthesized by the organism. The amino acids the human body cannot produce are called essential amino acids, which contain His, Ile, Leu, Lys, Met, Val, Phe, Thr, and Trp.

The human body gets these nine essential amino acids from food or nutritional supplement. In specific medical conditions or at different ages, the other amino acids may be conditionally essential for the human body. Glutamate is a non-essential amino acid that can be synthesized from alpha-ketoglutaric acid in the Krebs or citric acid cycle.

In the brain and spinal cord, glutamate is synthesized from glutamine as part of the glutamate-glutamine cycle by the enzyme glutaminase.

Glutamate cannot cross the blood-brain barrier unaided and serves as a metabolic precursor for the neurotransmitter gamma-aminobutyric acid GABA via the action of glutamate decarboxylase. Methionine is converted to S-adenosylmethionine SAM by methionine adenosyltransferase.

Loss of methionine has correlated with an accumulation of hydrogen peroxide H2O2 in hair follicles, a decrease in tyrosinase effectiveness, and a gradual loss of the natural hair color. GSH is an antioxidant found in animals, plants, fungi, bacteria, and archaea.

Promoting antioxidant-mediated cell defense and redox regulation is critical in protecting cells against dopamine-induced nigral cell loss by oxidative binding metabolites. These amino acids are cysteine, carnitine, taurine, lecithin, and phosphatidylcholine. Also, methionine is medium in the biosynthesis of additional phospholipids.

Improper transformation of methionine can lead to atherosclerosis due to the accumulation of homocysteine. Moreover, this amino acid is essential to reversing the damaging methylation of the glucocorticoid receptor gene caused by repeated stress exposures, with implications for depression.

Glycine is considered to be not essential to the human diet. The body can synthesize this amino acid from the amino acid serine. However, the metabolic capacity for glycine biosynthesis does not satisfy the need for collagen synthesis in several organisms.

In the liver of some of them at the vertebrate level, glycine synthesis is catalyzed by glycine synthase, which is also known as glycine cleavage enzyme. Glycine is integral to the creation of alpha-helices in secondary protein structure, and, mainly, it is the most copious amino acid in collagen harboring triple-helices.

Glycine is also an inhibitory neurotransmitter. The interference of its release within the central nervous system spinal cord can induce spastic paralysis due to uninhibited muscle contraction. Amino acids are synthesized through different pathways.

Cys is synthesized from Met, while Tyr synthesis can occur using Phe, considering that the amino acid precursors can be available in the body. The amino acid Arg, which arises from the urea cycle, is considered "semi-essential" because the synthetic capacity of the human body is limited.

Non-essential amino acids need their precursors, which must be available in the organism. Specifically, Ala and Gly's amino acids need pyruvate to be synthesized, while aspartic acid and Asn rely on oxaloacetic acid OAA.

Thus, six essential amino acids and three non-essential are integrated from oxaloacetate and pyruvate. The transamination from Glu is vital in forming Asp and Ala from OAA and pyruvate. Asp is crucial in synthesizing Asn, Met, Lys, and Thr. OAA is critical because no Asp would form without it.

The alpha-ketoglutaric acid or 2-oxoglutaric acid is one of two ketone derivatives of glutaric acid. Its anion, alpha-ketoglutarate alpha-KG , also known as 2-oxoglutarate, is a biological compound of paramount importance.

It is the keto acid produced by the deamination of Glu and is an intermediate compound in the urea or Krebs cycle. The amino acids glutamic acid and Gln arise from alpha-KG. Finally, the amino acid Pro derives from Glu, while Ser is from 3-phosphoglyceric acid 3PG. The 3PG is the conjugate acid of glycerate 3-phosphate.

It is a biochemically significant metabolic intermediate in glycolysis and the Calvin cycle. In the Calvin cycle or photosynthetic carbon reduction PCR cycle of photosynthesis, 3PG is vital. It is the product of the spontaneous scission of an unstable 6-carbon intermediate formed upon CO fixation.

Thus, glycerate 3-phosphate is a precursor for Ser, which, in turn, can create Cys and Gly through the homocysteine cycle. Therefore, Pro arises from Glu, while Ser is from 3PG.

In the transamination reaction, an amino acid Ala or Asp exchanges its amine group for the oxy group in alpha-KG. The products are Glu and pyruvate or OAA from Ala or Asp, accordingly.

Different proteases can degrade proteins into many small peptides or amino acids by hydrolyzing their peptide bonds. The unused amino acids may degrade further to join several metabolic processes. At first, the amino acids deaminate to their metabolic intermediates.

This process is helpful for the excretion of an excessive amount of nitrogen. Subsequently, they can transform into the remaining carbon skeleton.

In particular, this deamination process contains two steps. The first part uses deamination. In this step, the aminotransferase catalyzes the -NH2 group of the amino acid to alpha-KG. After that, it produces Glu and a novel alpha-keto acid of the specific amino acid.

The Glu -NH2 group could then be transferred to OAA to form alpha-KG and Asp. This trans-amination series only degrade the primary amino acid, while the -NH2 group nitrogen does not exclude. Then, it produces ammonia and alpha-KG. In the evaluation of the biochemistry of the amino acids, seven metabolic intermediates of the aminoacidic degradation platform are paramount.

They include acetyl-CoA, pyruvate, alpha-KG, acetoacetate, fumarate, succinyl-CoA, and OAA. In the most updated classification, Leu, Ile, Thr, and Lys degrade to acetyl-CoA, while Cys, Ala, Thr, Gly, Trp, and Ser degrade to pyruvate. Glu, Arg, His, Pro, and Gln degrade to alpha-KG, while Lys, Leu, Trp, Tyr, and Phe break down to acetoacetate.

Finally, Tyr, Phe, and Asp degrade to fumarate, Val, Met, and Ile break down to succinyl-CoA, and Asp and, of course, Asn degrade to OAA.

Isoleucine is an essential nutrient because it is unsynthesized in the body. This amino acid is both a glucogenic and ketogenic amino acid. In microorganisms and plants, it is synthesized via several steps beginning with pyruvate and alpha-ketobutyrate. The enzymes involved in this biosynthesis include acetolactate synthase, acetohydroxy acid isomeroreductase, dihydroxy acid dehydratase, and valine aminotransferase.

In clinical practice, plasma or urine amino acids undergo testing to evaluate patients with possible inborn metabolism problems. They can also assess many diseases, such as liver diseases, endocrine disorders, muscular diseases, neurological disorders, neoplastic diseases, renal failure, burns, and nutritional disturbances.

Both high-performance liquid chromatography HPLC and gas chromatography GC have been used to quantitatively identify the plasma or urine amino acids in clinical settings.

Amino acid disorders are identifiable at any age; most of them become evident during infancy or early childhood. Many inborn amino metabolism diseases occur in infancy or childhood. These disorders may include cystinuria, histidinemia, phenylketonuria PKU , methyl-malonyl CoA mutase deficiency MCM deficiency , albinism, and tyrosinemia.

Other amino acid disorders may be encountered later in life, including homocystinuria, alkaptonuria, maple syrup urine disease MSUD , and cystathioninuria. These disorders lead to clinical symptoms or signs of the specific amino acid disorder, which results in the deficiency or accumulation of one or more amino acids in the body's biological fluids, such as plasma or urine.

The deficiency of Phe hydroxylase causes PKU. Currently, there are more than mutations have been identified in the gene related to the cause of PKU. Besides, the deficiency of enzymes such as dihydropteridine reductase DHPR or tetrahydrobiopterin BH4 synthesis enzymes also leads to hyperphenylalaninemia.

In the case of the classic PKU, the Phe, phenyl lactate, phenylpyruvate, and phenylacetate are increased in the plasma, urine as well as other tissue samples.

The phenyl pyruvic acid excreted in urine produces a "mousy" odor. Central nervous system symptoms, such as mental retardation, seizures, failure to walk or speak, tremors, and hyperactivity, also show in these patients.

Another characteristic of classic PKU is hypopigmentation, which is due to the deficiency in the formation of melanin, which leads to pigmentation deficiency. Usually, the patients show light skin, fair hair, and blue eyes. Temporally, low Phe content synthetic nutrient supplemented with Tyr is the treatment of the classic PKU.

Albinism is a congenital disorder that is the defect of Tyr metabolism leading to a deficiency in melanin production. The characteristics of albinism are hypopigmentation by the total or partial absence of pigment in the hair, skin, and eyes.

There is no cure for albinism because it is a genetic disorder. Alkaptonuria is a rare disease with homogentisic acid oxidase defect, an enzyme in the Tyr degradation pathway. The urine specimen of the alkaptonuria patient shows some darkening on the surface after standing for fifteen minutes, which is due to homogentisate acid oxidation.

And after two hours of standing, the patient's urine is entirely black. The characteristics of alkaptonuria include the accumulation of homogentisic aciduria, large joint arthritis, and the intervertebral disks of vertebrae deposit with dense black pigments. Tyrosinemia type 1 results from a deficiency in fumarylacetoacetate hydrolase, leading to the accumulation of fumarylacetoacetate and its metabolites especially succinylacetone in urine, which makes cabbage-like odor.

The patients show renal tubular acidosis and liver failure. MCM deficiency is a disease due to the defect of methyl malonyl CoA mutase, which catalyzes isomerization between methyl malonyl-CoA and succinyl-CoA in the pathway.

Symptoms of MCM deficiency include vomiting, dehydration, fatigue, hypotonia, fever, breathing difficulty, and infections.

Also, metabolic acidosis and developmental delay occur as long-term complications. The treatment of MCM deficiency includes a special diet with low proteins low in Ile, Met, Thr, and Val amino acids and certain fats but high in calories.

Maple syrup urine disease MSUD is a rare autosomal recessive disease with a partial or complete defect of branched-chain alpha-keto acid dehydrogenase. The enzyme can decarboxylate Leu, Ile, and Val. This deficiency leads to the accumulation of branched-chain alpha-keto acid substrates.

These three amino acids cause functional abnormalities in the brain. The urine with a classic maple syrup odor is a hallmark characteristic of MSUD. MSUD patients show symptoms such as vomiting, feeding difficulties, dehydration, and severe metabolic acidosis.

In the clinic, a synthetic formula containing a limited amount of Leu, Ile, and Val is the suggested therapy for MSUD infants.

MSUD OMIM demonstrates a disturbance of the regular activity of the branched-chain α-ketoacid dehydrogenase BCKAD complex, the second step in the catabolic trail for the branched-chain amino acids BCAAs that includes leucine, isoleucine, and valine.

MSUD can occur early in life, but late-onset MSUD is also common and include neurologic symptoms. Cystathioninuria is a rare autosomal recessive metabolic disorder due to a deficiency in cystathionase. It links with the lower activity of the enzyme cystathionase.

There are two types of primary cystathioninuria based on the inherited mutation of the CTH gene: vitamin B6 responsive and vitamin B6 unresponsive cystathioninuria. The treatment of cystathioninuria varies according to the category in different cystathioninuria patients. Increased consumption of vitamin B6 is considered the best treatment for the active form of vitamin B6.

Homocystinuria is an inherited disorder due to the defect of the metabolism of Met amino acid. The most common cause is the enzyme cystathionine beta-synthetase deficiency, which results in the elevation of Met and homocysteine and low levels of Cys in plasma and urine.

Histidinemia is a rare autosomal recessive inborn metabolic error due to the defect of the enzyme histidase. A low in His intake diet is suggested for treating histidinemia, though the restricted diet is unnecessary for most cases.

Disclosure: Fan Shen declares no relevant financial relationships with ineligible companies. Disclosure: Consolato Sergi declares no relevant financial relationships with ineligible companies.

This book is distributed under the terms of the Creative Commons Attribution-NonCommercial-NoDerivatives 4. You are not required to obtain permission to distribute this article, provided that you credit the author and journal.

Turn recording back on. National Library of Medicine Rockville Pike Bethesda, MD Web Policies FOIA HHS Vulnerability Disclosure. Help Accessibility Careers. Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation. Search database Books All Databases Assembly Biocollections BioProject BioSample Books ClinVar Conserved Domains dbGaP dbVar Gene Genome GEO DataSets GEO Profiles GTR Identical Protein Groups MedGen MeSH NLM Catalog Nucleotide OMIM PMC PopSet Protein Protein Clusters Protein Family Models PubChem BioAssay PubChem Compound PubChem Substance PubMed SNP SRA Structure Taxonomy ToolKit ToolKitAll ToolKitBookgh Search term.

StatPearls [Internet]. Treasure Island FL : StatPearls Publishing; Jan-. Show details Treasure Island FL : StatPearls Publishing ; Jan-. Search term. Biochemistry, Amino Acid Synthesis and Degradation Fan Shen ; Consolato Sergi. Author Information and Affiliations Authors Fan Shen 1 ; Consolato Sergi 2.

Affiliations 1 University of Alberta. Major genomic deletions in independent eukaryotic lineages have led to repeated ancestral loss of biosynthesis pathways for nine of the twenty canonical amino acids.

While the evolutionary forces driving these polyphyletic deletion events are not well understood, the consequence is that extant metazoans are unable to produce nine essential amino acids EAAs.

Previous studies have highlighted that EAA biosynthesis tends to be more energetically costly, raising the possibility that these pathways were lost from organisms with access to abundant EAAs.

It is unclear whether present-day metazoans can reaccept these pathways to resurrect biosynthetic capabilities that were lost long ago or whether evolution has rendered EAA pathways incompatible with metazoan metabolism.

Here, we report progress on a large-scale synthetic genomics effort to reestablish EAA biosynthetic functionality in mammalian cells. We designed codon-optimized biosynthesis pathways based on genes mined from Escherichia coli. These pathways were de novo synthesized in 3 kilobase chunks, assembled in yeasto and genomically integrated into a Chinese hamster ovary CHO cell line.

One synthetic pathway produced valine at a sufficient level for cell viability and proliferation. Increasing the dosage of downstream ilvD boosted pathway performance and allowed for long-term propagation of second-generation cells in valine-free medium at 3. This work demonstrates that mammalian metabolism is amenable to restoration of ancient core pathways, paving a path for genome-scale efforts to synthetically restore metabolic functions to the metazoan lineage.

In this report, the authors devised synthetic genomic strategies to introduce essential amino-acid biosynthetic pathways into mammalian cells. While the functionalization of methionine, threonine, and isoleucine synthesis was unsuccessful, restoration of valine synthesis rendered mammalian cells partially independent of exogenous valine.

Moreover, transcriptomes of the valine-prototrophic cell mirrored transcriptomes captured during recovery from valine deprivation in parental, valine-auxotrophic counterparts. Altogether, this work was found to be of substantial interest as it provides pioneering evidence that mammalian systems may be permissive to the restoration of essential amino acid biosynthetic pathways and is thus anticipated to have a broad impact in the fields of synthetic biology, biotechnology and beyond.

Whole genome sequencing across the tree of life has revealed the surprising observation that nine essential amino acid EAA biosynthesis pathways are missing from the metazoan lineage Payne and Loomis, Furthermore, these losses appear to have occurred multiple times during eukaryotic evolution, including in some microbial lineages Figure 1A ; Payne and Loomis, ; Guedes et al.

Branching from core metabolism, the nine EAA biosynthesis pathways missing from metazoans involve over 40 genes Figure 1B , Supplementary files that are widely found in bacteria, fungi, and plants Guedes et al. While the absence of pathways that produce essential metabolites is observed in certain bacteria Zengler and Zaramela, , which possess short generation times and high genomic flexibility to adapt to rapidly changing environments, the forces driving the loss of multiple EAA biosynthetic pathways in multicellular eukaryotes remain a great mystery.

An exception that proves the rule is the partial reacquisition of EAA biosynthetic pathways through horizontal gene transfer in certain rare insect lineages which host genome-reduced intracellular bacteria and feed on simple nutrient sources such as sap or blood Wilson and Duncan, Recent efforts in genome-scale synthesis Isaacs et al.

A Presence of amino acid biosynthesis pathways across representative diverse organisms on Earth. B Schematic of EAA biosynthesis pathway steps that require engineering in mammalian cells to enable complete amino acid prototrophy if imported from Escherichia coli.

Proline and Valine pathways shown in this work are highlighted in red. C Workflow diagram of a synthetic genomics approach involving pathway design, construction, integration and testing towards mammalian EAA restoration. We sought to explore the possibility of generating prototrophic mammalian cells capable of complete biosynthesis of EAAs using a synthetic genomics approach Figure 1C.

The Chinese hamster ovary CHO K1 cell line was chosen as a model system due to its fast generation time, amenability to genetic manipulations, availability of a whole genome sequence, and established industrial relevance for producing biologics Fischer et al.

EAA biosynthesis genes from the best characterized model organisms were considered during pathway design while optimizing for the fewest number of enzymes needed for a given EAA pathway. To avoid using multiple promoters, we introduced ribosome-skipping 2A sequences Szymczak-Workman et al.

The entire pathway was synthesized de novo by commercial gene synthesis in 3 kilobase fragments and assembled in Saccharomyces cerevisiae via homologous recombination of basepair overlaps. Subsequent antibiotic selection of cells transfected with the vector resulted in a stable cell line containing the integrated EAA pathway.

Finally, we performed a variety of phenotypic, metabolomic, and transcriptomic characterizations on the modified cell line to verify activity of the EAA biosynthesis pathway. We first confirmed that the CHO cell line was in fact auxotrophic for each of the nine EAAs.

We noted that in this cell line, canonically non-essential amino acids tyrosine and proline also exhibited EAA-like properties in dropout media. Insufficient concentrations of phenylalanine in FK medium or low expression of endogenous phenylalaninehydroxylase that converts phenylalanine to tyrosine could underlie the tyrosine limitation.

Proline auxotrophy in CHO-K1 results from epigenetic silencing of the gene encoding Δ1-pyrrolinecarboxylate synthetase P5CS in the proline pathway Hefzi et al. We therefore used proline as a test case for our synthetic genomics pipeline. We tested the P5CS-equivalent proline biosynthesis enzyme found in Escherichia coli , encoded by two separate genes, proA and proB Figure 1—figure supplement 2A.

A vector pPro carrying codon-optimized proA and proB separated by a P2A sequence was synthesized and integrated into CHO-K1 Figure 1—figure supplement 2B. CHO cells with the stably integrated pPro proline pathway showed robust growth in proline-free FK medium Figure 1—figure supplement 2C-D , thus validating a pipeline for designing and generating specific amino acid AA prototrophic cells.

To demonstrate restoration of EAA pathways lost from the metazoan lineage more than — million years ago Cunningham et al. These EAAs were chosen because their biosynthesis pathways were missing the fewest number of genes.

Valine and isoleucine collectively require four genes to recapitulate the bacterial-native pathway. Figure 2—figure supplement 1. The two remaining genes were included to test potential routes to simultaneously rescue threonine and methionine auxotrophy by selectively supplementing individual missing metabolic steps, in addition to complete pathway reconstruction for valine and isoleucine.

To biosynthesize methionine, we chose the E. coli metC gene, which encodes cystathionine-ß-lyase and converts cystathionine to homocysteine, a missing step in CHO-K1 cells in a potential serine to methionine biosynthetic pathway. Threonine production was tested using E. coli L-threonine aldolase ltaE , which converts glycine and acetaldehyde into threonine.

For branched chain amino acids BCAAs valine and isoleucine, three additional biosynthetic enzymes and one regulatory subunit are needed in theory to convert pyruvate and 2-oxobutanoate into valine and isoleucine, respectively. In the case of valine, pyruvate is converted to 2-acetolactate, then to 2,3-dihydroxy-isovalerate, then to 2-oxoisovalerate and finally to valine.

For isoleucine, 2-oxobutanoate is converted to 2-acetohydroxybutanoate, then to 2,3-dihydroxymethylpentanoate, then to 3-methyloxopentanoate, and finally to isoleucine. The final steps in the biosynthesis of both BCAAs can be performed by native CHO catabolic enzymes Bcat1 and Bcat2 Hefzi et al.

The final pMTIV construct comprises metC , itaE , ilvN, ilvB , ilvC, and ilvD organized as a single open reading frame ORF with a 2A sequence variant lying between each protein coding region Figure 2B , and driven by a single strong spleen focus-forming virus SFFV promoter.

A Three enzymatic steps encoded by E. coli genes ilvN regulatory subunit, acetolactate synthase , ilvB catalytic subunit, acetolactate synthase , ilvC ketol-acid reductoisomerase , and ilvD dihydroxy-acid dehydratase are required for valine biosynthesis in Chinese hamster ovary CHO -K1 cells.

B Schematic of pMTIV construct after genomic integration and RNA-seq read coverage showing successful incorporation and active transcription.

C Microscopy images of CHO-K1 cells with integrated pCtrl or pMTIV constructs in complete FK medium after 2 days or valine-free FK medium after 6 days. Scale bar represents µm. D Growth curve of CHO-K1 cells with pCtrl or pMTIV in complete FK medium Figure 2—source data 1.

Day-0 indicates number of seeded cells. Error bars represent data from three replicates. E Growth curve of CHO-K1 cells with pCtrl or pMTIV in valine-free FK medium Figure 2—source data 1.

Raw cell count data for pMTIV valine-free and complete FK medium tests. To test the biosynthetic capacity of pMTIV, we first introduced the construct into CHO cells. Flp-In integration was used to stably insert either pMTIV, or a control vector pCtrl into the CHO genome.

Successful generation of each cell line was confirmed by PCR amplification of junction regions formed during vector integration Figure 2—figure supplement 2A-B. RNA-seq of cells containing the pMTIV construct confirmed transcription of the entire ORF Figure 2B.

Western blotting of pMTIV cells using antibodies against the P2A peptide yielded bands at the expected masses of P2A-tagged proteins, confirming the production of separate distinct enzymes Figure 2—figure supplement 2C. In reconstituted methionine-free, threonine-free, or isoleucine-free FK medium supplemented with dialyzed FBS to reduce FBS-derived AA content Figure 2—figure supplement 3 , cells containing the pMTIV construct did not show viability over 7 days, similar to cells containing the pCtrl control vector Figure 2—figure supplement 4.

In striking contrast, however, cells containing the integrated pMTIV showed relatively healthy cell morphology and viability in valine-free FK medium Figure 2C , whereas cells containing pCtrl exhibited substantial loss of viability over 6 days.

In complete FK medium, cells carrying the integrated pMTIV construct showed no growth defects compared to control cells Figure 2D. When cultured in valine-free FK medium over multiple passages with medium changes every 2 days, pMTIV cell proliferation was substantially reduced by the 3rd passage.

We hypothesized that frequent passaging might over-dilute the medium and prevent sufficient accumulation of biosynthesized valine necessary for continued proliferation as has been demonstrated for certain non-essential metabolites which become essential when cells are cultured at low cell densities Eagle and Piez, While use of pMTIV-conditioned medium improved the survival of cells harboring the pathway, it did not completely rescue valine auxotrophy in control cells, which exhibited substantial loss of cell viability over 8 days Figure 2—figure supplement 5A.

As a control, we generated pCtrl-conditioned valine-free FK medium using the same medium conditioning regimen, which failed to enable cells to grow to the same degree as that of pMTIV-conditioned medium, suggesting that the benefit conferred by medium conditioning is valine-specific Figure 2—figure supplement 5B.

Using this regimen, we were able to culture pMTIV cells for 9 passages without addition of exogenous valine Figure 2F. The doubling time was inconsistent across the 49 days of experimentation with cells exhibiting a mean doubling time of 5.

Despite the slowed growth seen in later passages, cells exhibited healthy morphology and continued viability at day, suggesting that the cells could have been passaged even further.

The pIV construct similarly supported cell growth in valine-free FK medium, and exhibited similar growth dynamics to the pMTIV construct in complete medium Figure 2—figure supplement 6.

To confirm endogenous biosynthesis of valine, we cultured pCtrl and pMTIV cells in RPMI medium containing 13 C 6 -glucose in the place of its 12 C equivalent together with 13 C 3 -sodium pyruvate spiked in at 2 mM over three passages Figure 3—figure supplement 1A.

High-resolution MS1 of MTIV cell lysates revealed a peak at The resulting fragmentation patterns for each peak Figure 3B matched theoretical expectations for each isotopic version of valine Figure 3—figure supplement 1B.

Taken together, this demonstrates that pMTIV cells are capable of biosynthesizing valine from core metabolites glucose and pyruvate, thereby proving successful metazoan biosynthesis of valine. Over the course of 3 passages in heavy valine-free medium, the non-essential amino acid alanine, which is absent from RPMI medium and synthesized from pyruvate, was found to be Assuming similar turnover rates for alanine and valine within the CHO proteome, we expected to see similar percentages of 13 C-labeled valine.

However, just For pMTIV cells cultured in heavy but valine-replete medium, just 6. Together with the observed slow proliferation of pMTIV cells in valine-free medium, our data suggests that valine complementation is sufficient but sub-optimal for cell growth.

MS2 fragmentation patterns for each of these metabolites matched expectations Figure 3—source data 1. C RNA-seq dendrogram of pCtrl cells and pMTIV cells grown on complete FK medium or starved of valine for 4 hr or 48 hr.

D Principal Component Analysis PCA space depiction of pCtrl cells and pMTIV cells grown on complete FK medium, or starved of valine for 4 hr or 48 hr. We performed RNA-seq to profile the transcriptional responses of cells containing pMTIV or pCtrl in complete harvested at 0 hr and valine-free FK medium harvested at 4 hr and 48 hr, respectively Figure 3C , Figure 3—figure supplement 2A.

The transcriptional impact of pathway integration is modest Figure 3D. Only 51 transcripts were differentially expressed between pCtrl and pMTIV cells grown in complete medium, and the fold changes between conditions were small Figure 3E , Figure 3—figure supplement 2B.

While some gene ontology GO functional categories were enriched Figure 3—figure supplement 2C , they did not suggest dramatic cellular stress. Rather, these transcriptional changes may reflect cellular response to BCAA dysregulation due to altered valine levels Zhenyukh et al.

In contrast, comparison of 48 hr valine-starved pCtrl and pMTIV cells yielded ~ differentially expressed genes. Transcriptomes of pMTIV cells in valine-free medium more closely resembled cells grown on complete medium than did pCtrl cells in valine-free medium Figure 3D , Figure 3—figure supplement 3A.

Differentially expressed genes between pCtrl and pMTIV cells showed enrichment for hundreds of GO categories, including clear signatures of cellular stress such as autophagy, changes to endoplasmic reticulum trafficking, and ribosome regulation Figure 3—figure supplement 3B.

Most of the differentially regulated genes between pCtrl cells in complete medium, and those same cells starved of valine for 48 hr were also differentially expressed when comparing pCtrl and pMTIV cells in valine-free medium Figure 3E , supporting the hypothesis that most of the observed transcriptional changes represent broad but partial rescue of the cellular response to starvation.

We also examined the integrated stress response ISR and mTOR signaling pathways, both of which are known to modulate cellular responses to starvation Pakos-Zebrucka et al. We observed no clear signatures of mTOR activation Figure 3—figure supplement 4A , although a number of individual genes related to the mTOR pathway were significantly differentially expressed compared to pCtrl cells valine-starved for 48 hr Figure 3—figure supplement 4B.

A manually curated list of ISR genes showed signals of ISR gene activation, but showed few differences between pCtrl and pMTIV cells at 48 hr of starvation Figure 3—figure supplement 4C.

pMTIV cells grown for 5 passages over 29 days on conditioned valine-free FK medium were more similar to pMTIV cells starved for 48 hr than to pCtrl cells starved for 48 hr Figure 3—figure supplement 5A.

To improve rescue of the valine starvation phenotype, we looked for valine biosynthetic pathway intermediates in our metabolomics data that might suggest that the pathway was bottlenecked at any stage. While no signal could be detected for pyruvate, 2-acetolactate or 2-oxoisovalerate, a signal was detected for pathway intermediate 13 C 5 -2,3-dihydroxy-isovalerate, which was specific to pMTIV cells cultured in both complete and in valine-free medium Figure 3—figure supplement 1F.

To determine whether the downstream pathway gene, ilvD , which encodes the dihydroxy-acid dehydratase enzyme, might constitute a bottleneck, we generated a lentivirus encoding a puromycin resistance cassette in addition to ilvD under control of a viral MMLV promoter Figure 4A.

Both pCtrl and pMTIV cells were infected and integrants were selected for on puromycin, resulting in a population-averaged integration count of 5. This resulted in a 0. B ilvD qPCR on gDNA and cDNA from each cell line Figure 4—source data 1.

Fold change levels were relativized to pMTIV. cDNA was reverse transcribed using oligo dT primers from RNA templates collected from each cell line. Error bars show SD of three technical replicates. Error bars represent data from two replicates.

We also reduced the isoleucine content of this isotopically heavy valine-free RPMI medium to match the isoleucine content of FK medium from 0. In pMTIV cells, presence of pathway intermediate 2,3-dihydroxy-isovalerate fluctuated throughout the 24 days of culture with cells exhibiting higher concentrations in earlier time points.

pMTIV samples on average contained We quantified the functional impact of modifying flux at this pathway bottleneck by culturing both cell lines in unconditioned, reduced-isoleucine, valine-free RPMI medium containing 2 mM sodium pyruvate over 10 passages on plates not coated with gelatin.

By comparison, pMTIV cells exhibited an average doubling time of 4. Plates were not coated with gelatin. In this work, we demonstrated the successful restoration of an EAA biosynthetic pathway in a metazoan cell.

Our results indicate that contemporary metazoan biochemistry can support complete biosynthesis of valine, despite millions of years of evolution from its initial loss from the ancestral lineage. Interestingly, independent evidence for BCAA biosynthesis has also been obtained for sap-feeding whitefly bacteriocytes that host bacterial endosymbionts; metabolite sharing between these cells is predicted to lead to biosynthesis of BCAAs that are limiting in their restricted diet.

The malleability of mammalian metabolism to accept heterologous core pathways opens up the possibility of animals with designer metabolisms and enhanced capacities to thrive under environmental stress and nutritional starvation Zhang et al.

Yet, our failure to functionalize designed methionine, threonine, and isoleucine pathways highlights outstanding challenges and future directions for synthetic metabolism engineering in animal cells and animals. Other pathway components or alternative selections may be needed for different EAAs Rees and Hay, Studies to reincorporate EAAs into the core mammalian metabolism could provide greater understanding of nutrient-starvation in different physiological contexts including the tumor microenvironment Lim et al.

Emerging synthetic genomic efforts to build a prototrophic mammal may require reactivation of many more genes Supplementary files , iterations of the design, build, test DBT cycle, and a larger coordinated research effort to ultimately bring such a project to fruition.

For pathway completeness analysis, the EC numbers of each enzyme in each amino acid biosynthesis pathway excluding pathways annotated as only occurring in prokaryotes were collected from the MetaCyc database Supplementary file 4. Variant biosynthetic routes to the same amino acid were considered as separate pathways, generating distinct EC number lists.

The resulting per-pathway EC number lists were checked against the KEGG, Entrez Gene, Entrez Nucleotide, and Uniprot databases using their respective web APIs for each listed organism. CHO Flp-In cells ThermoFisher, R were used in all experiments. All cell lines tested negative for mycoplasma.

Custom amino acid dropout medium was adjusted to a pH of 7. For metabolomics experiments, medium was prepared from an amino acid-free and glucose-free RPMI powder base US Biological, R , and custom combinations of amino acids and isotopically heavy glucose and sodium pyruvate were added in to match the standard amino acid concentrations for RPMI or as specified.

pH was adjusted to 7. Where specified, cells were cultured on plates coated with 0. Plates were washed with PBS prior to use. For evaluating effects of amino acid dropout on cell growth curves, cells were seeded at 1×10 4 into 6-well plates into FK media with lowered amino acid concentrations relative to typical FK media and then allowed to grow for 5 days.

Media was then aspirated off and replaced with PBS with Hoechst live nuclear stain for automated imaging and counting using a DAPI filter set on an Eclipse Ti2 automated inverted microscope.

To count, an automated microscopy routine was used to Figure 5 random locations within each well at 10× magnification, and then the cells present in imaged frames counted using automatic cell segregation and counting software.

Given differences in cell response to starvation, segregation and counting parameters were tuned in each experiment, but kept constant between starvation conditions and cells with and without the pathway.

Conditioned medium was generated by seeding 1×10 6 pMTIV cells into 10 mL complete FK medium on 10 cm plates and replacing the medium with 10 mL freshly prepared valine-free FK medium the next day following a PBS wash step.

Cells conditioned the medium for 2 days at which point the medium was collected, centrifuged at ×g for 3 mins to remove potential cell debris, sterile filtered, and collected in mL vats to reduce batch-to-batch variation. Integrated constructs were synthesized de novo in 3 kb DNA segments with each segment overlapping neighboring segments by 80 bp.

Assembly was conducted in yeasto by co-transformation of segments into S. cerevisiae strain BY made competent by the LiOAc method Pan et al. After 2 days of selection at 30°C on SC—Ura medium, individual colonies were picked and cultured overnight.

Glass beads were added to each resuspension and the mixture was vortexed for 10 mins to mechanically shear the cells.

Next, cells were subject to alkaline lysis by adding µl of P2 lysis buffer Qiagen, for 5 mins and then neutralized by addition of Qiagen N3 neutralization buffer Qiagen, Plasmid DNA was eluted in Zyppy Elution buffer and subsequently transformed into TransforMax EPI chemically competent E.

Cell were lysed in SKL Triton lysis buffer 50 mM Hepes pH7. NuPAGE LDS sample buffer ThermoFisher, NP supplemented with 1. The membrane was incubated in the secondary antibody solution for 1. Cell pellets were generated by trypsinization, followed by low speed centrifugation, and the pellet was frozen at —80°C until further processing.

The LC column was a Millipore ZIC-pHILIC 2. Injection volume was set to 1 μL for all analyses 42 min total run time per injection. MS analyses were carried out by coupling the LC system to a Thermo Q Exactive HF mass spectrometer operating in heated electrospray ionization mode HESI.

Spray voltage for both positive and negative modes was 3. Tandem MS spectra for both positive and negative mode used a resolution of 15,, AGC target of 1e5, maximum IT of 50ms, isolation window of 0.

The minimum AGC target was 1e4 with an intensity threshold of 2e5. All data were acquired in profile mode. All valine data were processed using Thermo XCalibur Qualbrowser for manual inspection and annotation of the resulting spectra and peak heights referring to authentic valine standards and labeled internal standards as described.

QIAshredder homogenizer columns Qiagen, were used to disrupt the cell lysates. Libraries were prepared using the NEBNext Ultra RNA Library Prep Kit for Illumina New England Biolabs, E , and sequenced on a NextSeq single-end 75 cycles high output with v2.

Differential gene enrichment analysis was performed with in R with DESeq2 and GO enrichment performed and visualized with clusterProfiler against the org. db database, with further visualization with the pathview, GoSemSim, eulerr packages. Target plasmid was maintained in and purified from NEB beta electrocompetent E.

coli New England Biolabs, CK. Lentivirus was packaged by plating 4×10 6 HEKT cells on 10 cm 2 and incubating cells overnight at 37°C. Cells were transfected with a plasmid mix consisting of 3. Transfected HEKT cells were incubated for 48 hr, before medium was collected, and centrifuged at ×g for 5 mins.

The resulting supernatant was filtered using a 0. The packaged virus was applied to cells for 24 hr before the medium was exchanged for fresh medium. For RNA extraction, QIAshredder homogenizer columns Qiagen, were used to disrupt the cell lysates.

cDNA was generated from RNA using Invitrogen SuperScript IV Reverse Transcriptase Invitrogen, and oligo dT primers.

Each qPCR reaction was performed using SYBR Green Master I Roche, on a Light Cycler Roche, using the recommended cycling conditions. Primers were designed to amplify amplicons — bp in size. Sequencing data generated for this study is deposited in the NCBI SRA at accession number PRJNA Source data files have been provided for Figure 1 - figure supplement 1, Figure 1 - figure supplement 2D, Figure 2, Figure 2 - figure supplement 3, Figure 2 - figure supplement 4B, Figure 2 - figure supplement 5, Figure 2 - figure supplement 6, Figure 3, and Figure 3 - figure supplement 1, Figure 4, Figure 4 - figure supplement 1, Figure 5, and Figure 5 - figure supplement 1.

Our editorial process produces two outputs: i public reviews designed to be posted alongside the preprint for the benefit of readers; ii feedback on the manuscript for the authors, including requests for revisions, shown below.

We also include an acceptance summary that explains what the editors found interesting or important about the work. Thank you for submitting your article "Resurrecting essential amino acid biosynthesis in a mammalian cell" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, including Ivan Topisirovic as Reviewing Editor and Reviewer 1, and the evaluation has been overseen by Philip Cole as the Senior Editor.

The following individual involved in review of your submission has agreed to reveal their identity: Ran Kafri Reviewer 3. The reviewers have discussed their reviews with one another, and the Reviewing Editor has drafted this to help you prepare a revised submission.

Based on this, it was thought that more evidence is required to demonstrate that the introduction of valine biosynthetic pathway into CHO cells results in sustained proliferation and survival in the absence of valine supplementation.

Accordingly, it was deemed that the authors should monitor long-term ability of engineered CHO cells to sustain valine production and proliferate in valine-free media. To this end, monitoring flux via valine biosynthetic and degradation pathways, transcriptome and mTOR signaling at early and late time points was thought to be warranted.

These include lack of clarity pertinent to the rationale behind using "conditioned-medium" in the experiments. Moreover, potential utilization of other sources of valine e. It was appreciated that the latter cells survive in valine-free media, but it seems that their proliferation is significantly lower than in valine containing media.

Moreover, it seems that after 6 passages only a fraction of the detected valine is synthesized de novo. Would this fraction further decrease in subsequent passages? Related to this, it is not clear what is the efficiency of valine biosynthesis in CHO cells vs. a prototrophic organism.

Perhaps comparing the rates of valine synthesis in cell free extracts of CHO cells vs. those derived from a prototrophic organism may be helpful to address this. This in particular relates to amino-acid sensing pathways e.

Were the enzymes mislocalized? Are there other regulatory factors involved? Moreover, considering that the overarching tenet is that metazoans lost the ability to produce essential amino acids due to energetic restraints, it may be worthwhile noting that culturing conditions and potential differences in energy resources may impact on functionalization of essential amino acid biosynthetic pathways.

The results put forth in this manuscript suggest the authors were marginally successful in introducing a valine biosynthetic pathway into CHO cells, but fall short of demonstrating a robust, self-sustaining engineered cell line under reasonable culture conditions.

This milestone should be met prior to final acceptance at eLife. Additionally, the following revisions should be carried out prior to acceptance.

The authors should identify the timepoint at which pCTRL cells are no longer viable in dropout medium. The authors should then compare transcriptional profiles of the pMTIV cells at that timepoint to the that of pMTIV cells harvested at 4hr and 48hr.

Doing so may help identify key bottlenecks in the pathway. If a bottleneck can be identified, authors should attempt to make the pathway more efficient, either by modifying expression strategy of that enzyme or testing homologs from other hosts.

The pathway should be optimized until the major revision 1 above is achieved. For clarity, these sections should be de-emphasized in writing and figures for clarity. The task that Wang et al. We thank the reviewers for this feedback.

We understand the core issue to be the reduction in doubling time shown for later time points in Figure 2F and the suggestion that this represents a time-dependent lag in growth rate due to cumulative insufficient valine production. In response to this feedback, we set out to attain a consistent doubling time in the valine-free condition.

Importantly, this dihydroxy-acid dehydratase overexpressing cell line was passaged 10 times in the absence of valine with a consistent average doubling time of 3. Doubling time remained consistent across the 39 days of culture and no medium conditioning was required Figure 5.

Nonetheless, to alleviate concerns that the original prototrophic pMTIV cells were not able to sustain proliferation long-term in the absence of valine, we have also added additional evidence indicating that these cells retained valine prototrophy long-term:.

Given the rapid death phenotype experienced by pCtrl cells, continued survival of pMTIV cells at late passages should instead be considered an indicator of sustained prototrophy. Late time point transcriptomic data Figure 3 —figure supplement 5 for pMTIV cells demonstrating partial rescue of nutritional starvation at day 29 in conditioned valine-free FK medium.

We thank the reviewers for these comments. We have added a figure highlighting mTOR signaling differences in pMTIV and pCtrl cells at 48 h valine starvation, even though no clear signatures of mTOR activation could be detected Figure 3 —figure supplement 4.

We have also added a new supplemental figure showing transcriptomic analysis of cells grown long-term 5 passages, 29 days in conditioned valine-free FK medium Figure 3 —figure supplement 5.

Additionally, we were able to gain insight into flux through the pathway with 13 C-tracing. No signal could be detected for pyruvate, 2-acetolactate or 2-oxoisovalerate; however we were able to specifically detect pathway intermediate 2,3-dihydroxy-isoverate and have added a panel to reflect this Figure 3 —figure supplement 1F.

By the time many students get to the study of amino acid biosynthesis, they have seen so many pathways that learning new pathways for the amino acids seems daunting, even though they can be clustered into subpathways.

Most know that from a nutrition perspective, amino acids can be divided into nonessential and essential need external dietary supplementation amino acids. These are shown for humans below. Three of the essential amino acids can be made in humans but need significant supplementation.

Arginine is depleted in processing through the urea cycle. When cysteine is low, methionine is used to replace it so its levels fall. If tyrosine is low, phenylalanine is used to replace it.

For this chapter subsection, we will provide only the basic synthetic pathways in abbreviated form without going into mechanistic or structural details. Ala can easily be synthesized from the alpha-keto acid pyruvate by a transamination reaction, so we will focus our attention on the others, the branched-chain nonpolar amino acids Val, Leu, and Ile.

Since amino acid metabolism is so complex, it's important to constantly review past learning. As is evident from the figure, glutamic acid can be made directly through the transamination of α-ketoglutarate by an ammonia donor, while glutamine can be made by the action of glutamine synthase on glutamic acid.

Arginine is synthesized in the urea cycle as we have seen before. It can be made from α-ketoglutarate through the following sequential intermediates: N-acetylglutamate, N-acetylglutamate-phosphate, N-acetylglutamate-semialdehyde, N-acetylornithine to N-acetylcitruline.

The is deacetylated and enters the urea cycle. Here we present just the synthesis of lysine from aspartate and pyruvate using the diaminopimelic acid DAP pathway. Fundamentals of Biochemistry Vol. With the rare exceptions of pyrrolysine and selenocysteine, a standard set of 20 amino acid building blocks, containing a limited number of functional groups, is used by almost all organisms for the biosynthesis of proteins.

The use of Genetic Code Expansion technology to enable the site-specific incorporation of noncanonical amino acids ncAAs into proteins in living cells has transformed our ability to study biological processes and provided the exciting potential to develop modern medicines 1 , 2 , 3 , 4 , 5 , 6 , 7 , 8.

The high intracellular concentration of ncAA required to render this machinery operative has usually been achieved via chemical synthesis of the ncAA and its exogenous addition at high levels to the cell culture medium.

Although most ncAAs could penetrate cell membrane for the genetic incorporation, ncAAs bearing negative charges or polar structures normally exhibit a low cell penetration efficiency 11 , 12 , 13 , The relatively low intracellular concentrations of these ncAAs greatly limit the efficiency of ncAA incorporation into proteins using the Genetic Code Expansion technology 12 , 13 , 15 , 16 , 17 , Strategies for engineering the structures of ncAAs or ncAA-binding proteins have been employed to improve the cellular uptake of ncAAs.

In , the Schultz group adopted a dipeptide strategy to enable the cellular uptake of phosphotyrosine.

The phosphotyrosine-containing dipeptide can be synthesized and transported into cells via an adenosine triphosphate ATP -binding cassette transporter, followed by hydrolysis of the dipeptide by nonspecific intracellular peptidases In the same year, the Wang lab developed a two-step strategy for producing proteins with site-specific tyrosine phosphorylation This strategy utilized the incorporation of a phosphotyrosine anologue with a cage group, followed by chemical deprotection of the purified proteins.

However, the synthesis and purification of these dipeptides are challenging, and the required post-purification treatments limit the applicability of this methodology to efficiently incorporate phosphotyrosine in living cells.

As an alternative approach, periplasmic binding proteins PBPs have been engineered to have improved affinities for specific ncAAs These mutant PBPs enhanced uptake of the respective ncAAs up to fivefold, as evidenced by elevated intracellular ncAA concentrations and the yield of ncAA-containing green fluorescent proteins Nevertheless, the engineered PBP species are only applicable to a subset of ncAAs, and exogenous feeding of high concentrations of the ncAAs is still required.

The problem of ncAA uptake could potentially be bypassed by intracellular biosynthesis of the ncAAs from basic carbon sources 12 , 19 , 20 , 21 , 22 , 23 , 24 , The Chin group overcame the membrane impermeability of p Thr by introducing the Salmonella enterica kinase, PduX, which converts L- threonine to p Thr intracellularly A similar strategy was recently applied to the creation of autonomous bacterial cells that can biosynthesize and genetically incorporate p -amino-phenylalanine p AF , 5-hydroxyl-tryptophan 5HTP and dihydroxyphenylalanine DOPA , although no autonomous eukaryotic cells have been reported 19 , 20 , 22 , 23 , We see there that additional biosynthetic pathways for producing polar or negatively-charged ncAAs would greatly expand the utility of genetic code expansion methods.

Tyrosine sulfation is an important post-translational modification of proteins that is essential for a variety of biomolecular interactions, including chemotaxis, viral infection, anti-coagulation, cell adhesion, and plant immunity 27 , 28 , 29 , 30 , 31 , 32 , 33 , Despite its importance and ubiquity, protein sulfation has been difficult to study due to the lack of general methods for preparing proteins with defined sulfated residues 31 , To circumvent this challenge, efforts have been previously made to site-specifically incorporate sulfotyrosine sTyr using the Genetic Code Expansion technology The resulting sTyr incorporation systems have enabled several applications, including generation of therapeutic proteins with defined sulfated tyrosines, evolution of sulfated anti-gp antibodies, and confirmation of tyrosine sulfation sites 35 , 37 , 38 , 39 , To achieve reasonsble expression levels of sulfated proteins in E.

Here, we report the generation of metabolically modified prokaryotic and eukaryotic cells that can biosynthesize sTyr and incorporate it into proteins in a site-specific manner Fig.

sTyr is biosynthesized using a sulfotransferase discovered from a sequence similarity network SSN. sTyr is subsequently incorporated into proteins in response to a repurposed stop codon. The molecular properties of this the sulfotransferase were explored using bioinformatics and computational approaches, revealing a loop structure and several residues in binding pocket within this enzyme responsible for its unique specificity for tyrosine.

The further optimization of the genome and sTyr biosynthetic pathway of both prokaryotic and eukaryotic cells leads to greater expression yields of sulfated proteins than cells exogenously fed with sTyr. The utility of these sTyr autonomous cells is demonstrated by using them to produce highly potent thrombin inhibitors.

a sTyr was biosynthesized from tyrosine and PAPS in the presence of sulfotransferase identified in this study. The resulting biosynthesized sTyr was site-specifically incorporated into thrombin inhibitors, yielding enhanced thrombin inhibition b Sequence similar network SSN generated by EFI-EST server with Rn SULT1A1 as an input sequence and E value of 5.

Edges detection threshold was set at an alignment score of The upper and lower yellow representative nodes are Rn SULT1A1 P and Hs SULT1C2 O , respectively.

c Schematic representation of reported sulfation reactions of P and O d Screening of tyrosine sulfotransferases with green fluorescent protein assay. All tested proteins are included in the representative nodes of b and Nn SULT1C1 is the protein with red label b.

stands for arbitrary unit. Based on their substrate preference and cellular location, sulfotransferases can be grouped into three major families, tyrosylprotein sulfotransferase TPST , cytosolic sulfotransferase SULT , and carbohydrate sulfotransferase 43 , To identify the enzyme responsible for sulfation of cytoplasmic tyrosine, we focused on SULTs.

These enzymes catalyze sulfation of a wide variety of endogenous compounds, including hormones, neurotransmitters, and xenobiotics Based on their reported substrate specificities, we examined SULT1A1 and SULT1A3 from Homo sapiens , SULT1A1 from Rattus norvegicus , and SULT1C1 from Gallus gallus 43 , 45 , all of which are known to recognize multiple phenolic substrates.

To explore the activity of these sulfotransferases toward tyrosine, we used a green fluorescent protein assay 20 , These four sulfotransferase genes were codon-optimized for Escherichia coli and cloned into the pBad vector with DNA oligos in Supplementary Data 1.

To generate a suppression plasmid for sTyr incorporation, we used pUltra-sTyr plasmid encoding the engineered Methanococcus jannaschii tyrosyl-tRNA synthetase sTyrRS and its corresponding Mj tRNA Tyr CUA 36 , The suppressor plasmid pUltra-sTyr was used to suppress the amber codon AspTAG within a sfGFP variant encoded by the pLei-sfGFPTAG plasmid in the presence of sTyr.

Unfortunately, none of these four sulfotransferases led to sfGFP expression, indicating the failure of the biosynthesis of sTyr. To circumvent the limited substrate range of the reported sulfotransferases, we accessed the full repertoire of protein sequence diversity in nature by using a sequence similarity network SSN, Fig.

SSNs provide an effective way to visualize and analyze the relatedness of massive protein sequences on the basis of similarity thresholds of their amino acid sequences We initially created an SSN with EFI-ESI based on SULT1A1 from Rattus norvegicus as an input sequence, since its cognate substrate p -coumaric acid is similar to tyrosine Fig.

Interestingly, we found that human SULT1C2, whose substrate is tyramine, was in a different cluster of the SSN Fig. We hypothesized that enzymes with high sequence similarity to SULT1A1 from Rattus norvegicus and SULT1C2 from Homo sapiens would be potential candidates to carry out the sulfation of tyrosine.

To test this hypothesis, we selected 27 sequences from the SSN based on their similarity to both Rn SULT1A1 and Hs SULT1C2. These selected genes were cloned into the pBad vector and tested with the green fluorescent protein assay.

To our delight, a 2. Thus, we name A0AVQH7 as Nn SULT1C1 hereafter To explore the origin of the unique tyrosine specificity of Nn SULT1C1 among all the sulfotransferases tested, we analyzed the phylogenetic relationships of the enzymes.

Sulfotransferase amino acid sequences were used to generate a phylogenetic tree using the unweighted pair group method with arithmetic mean by MEGA X software package Supplementary Fig. The tree is subdivided into three major subfamilies, among which Nn SULT1C1 falls into subfamily I containing bird sulfotransferases.

Most sequences from subfamilies II and III are derived from rodent and primate groups, respectively. To further analyze the molecular basis of the unique tyrosine specificity of Nn SULT1C1, we performed a multiple sequence alignment of all sequences within subfamily I of the phylogenetic tree Supplementary Fig.

This sequence alignment revealed that most regions of Nn SULT1C1, including the PAPS-binding site, are highly conserved except for a highly variable region corresponding to Nn SULT1C1 residues SIQEPPAAS and residues likely involved in substrate binding pocket 51 , To explore the contribution of this highly variable region to substrate binding, the structure of Nn SULT1C1 was predicted via Alphafold 2.

Alphafold 2 is a machine learning approach that has been shown to predict protein structure with a high degree of accuracy 53 , 54 , 55 , This structure includes a β sheet surrounded by α-helices, giving rise to a narrow substrate-binding site Fig. We found that the highly variable region 94— residue of Nn SULT1C1 constitutes a loop for the substrate entry, which also aligns with the substrate entry loop of human SULT Supplementary Fig.

a Nn SULT1C1 structure blue predicted by AlphaFold2 and its active site consisting PAPS and Tyr. Tyr was docked into Nn SULT1C1 containing PAPS by Glide v8. d Structural similarity search of Nn SULT1C1 using the PDBeFold web server.

e Characterization of Tyr docking with Nn SULT1C1 and its structurally similar sulfotransferases via docking score and nucleophilic attack distance. Docking scores were calculated using Glide v8. Nucleophilic attack distance was defined as the distances between Tyr phenolic alcohol and PAPS sulfonate.

f Comparison of tyrosine sulfation activity of Nn SULT1C1 and its structurally similar sulfotransferases using green fluorescent protein assay. g—j Tyr docking position with Nn SULT1C1 g , mSULT1D1 h , hSULT1A3 i , and hSULT1C2 j.

PAPS and Tyr are shown as sticks with green carbon. Docking was performed by Glide v8. To further explore the other residues involved in substrate binding of Nn SULT1C1, we performed protein-ligand docking using Glide v8.

For each docking experiment, maximum output poses for each protein were set and Emodel energy was used for ranking the top 50 poses. The docking structure suggests that the α-amino group of Tyr is stabilized by Nn SULT1C1 residues Glu, Thr30, Ile33, and Trp The π-π stacking interactions between Tyr and Phe90 are likely to improve the packing interaction Fig.

The phenolic hydroxy group of Tyr is in the proper Lys-Lys-His catalytic site to engage in sulfuryl transfer. The His residue serves as a catalytic base that can remove the proton from Tyr. The Lys57 and Lys residues interact with and stabilize the sulfuryl group of PAPS and the phenolic hydroxy group of Tyr, respectively.

To validate the contribution of these residues interacting with Tyr on Nn SULT1C1 activity, Thr30, Ile33, Trp93, and Glu were mutated to alanine separately.

Alanine mutation at Thr30, Trp93, or Glu significantly decreased the activity of Nn SULT1C1 Fig. Among these residues, the EA mutation exhibits the largest decrease in activity, confirming its important interaction with Tyr.

The overall secondary structure of Nn SULT1C1 aligned well with 2zvq, which indicates its structural consistency with the other SULTs Supplementary Fig.

To further illustrate the unique specificity of Nn SULT1C1 for Tyr, dockings of Tyr to the most similar sulfotransferases, including mSULT1D1, hSULT1A3, and hSULT1C2, were carried out using Glide v8.

This result is consistent with the optimal ability of Nn SULT1C1, to generate sTyr-containing sfGFP in the green fluorescent protein assay among all tested sulfotransferases Fig. The key step of the sulfotransfer reaction involves an S N 2-type nucleophilic attack on the PAPS sulfonate by the phenoxide of Tyr.

Compared with mSULT1D1, hSULT1A3, and hSULT1C2, the docking of Tyr in Nn SULT1C1 results in the closest distance 3. Furthermore, the acceptor phenolic hydroxyl group of Tyr lies on the backside of the S-O bond of PAPS in the Tyr docking structure with Nn SULT1C1, indicating a more proper orientation for the nucleophilic attack Fig.

Having identified Nn SULT1C1 as a functional tyrosine sulfotransferase, we explored whether the biosynthesized sTyr can be genetically incorporated into proteins in E.

coli in response to the amber codon. As an initial goal, we wanted to increase sTyr production in these cells in order to optimize its availability for incorporation into proteins. Since Nn SULT1C1 utilizes tyrosine and PAPS for producing sTyr, we quantified sTyr production in five knockout E.

coli cell lines in which the gene knockout has been shown to improve the yield of either tyrosine or PAPS in E. To evaluate the effect of knocking out these genes on the biosynthesis of sTyr, we transformed the suppression plasmid pUltra-sTyr, reporter plasmid pET22b-T5-sfGFPTAG, and the biosynthesis plasmid pEvol- Nn SULT1C1 into wildtype E.

coli BW or knockout strains Fig. To our delight, we found that knockout of the cysH gene significantly improved the production of sTyr-containing sfGFP, compared to that seen in the wildtype BW strain Fig.

This observation of enhanced sfGFP-sTyr production in BWΔcysH is consistent with the previous report that knockout of cysH gene can increase cellular PAPS concentration and the production of sulfated products in E. coli 65 , Next, we examined whether manipulation of PAPS synthetic and recycling pathways in E.

coli could further enhance intracellular PAPS levels. We found that cells expressing all these genes exhibited the largest increase in fluorescence, suggesting a higher expression level of sfGFP-sTyr Fig.

The Nn SULT1C1 expression level has a significant influence on the production of sfGFP-sTyr, since we found that the concentration of Nn SULT1C1 inducer is important. To examine the contribution of the biosynthetic pathway to intracellular sTyr concentration, we measured the intracellular sTyr concentrations in cells when sTyr was either biosynthesized or delivered via exogenous feeding.

To our delight, the cellular concentration of sTyr in cells endowed with the sTyr biosynthetic pathway is Consistent with these intracellular levels of sTyr, endogenous biosynthesis of sTyr results in much higher sfGFP-sTyr expression than that produced via exogenous feeding Fig.

To further investigate the efficiency and specificity of incorporation of biosynthesized sTyr in these autonomous E. Intact sfGFP was only expressed after exogenous sTyr feeding or after induction of sTyr biosynthesis.

The yield of sfGFP-sTyr derived from biosynthetic sTyr is 5. To test the activity of Nn SULT1C1 in vitro, its kinetics values were measured. It exhibits a K m, V max , and K cat of 0.

a Schematic representation of genetic circuits used for generating completely autonomous sTyr synthesizing E. col i.

b Screening of the knockout strains for sfGFP-sTyr production after the expression of Nn SULT1C1. c The roles of PAPS recycling enzymes in producing sfGFP-sTyr using ΔcysH BW strain.

d Production of sfGFP-sTyr from cells with the addition of chemically synthesized sTyr or the biosynthesized Tyr. e Cellular concentrations of sTyr of cells with the addition of chemically synthesized sTyr or the biosynthesis of sTyr. Post-translational tyrosine sulfation occurs exclusively in eukaryotes.

One approach to determine the biological importance of protein tyrosine sulfation is to express sulfated protein in living cells in a site-specific and homogeneous fashion, a goal that is difficult to achieve by chemical synthesis or recombinant expression. Genetic code expansion based on E.

To promote the efficient expression of mammalian proteins sulfated on specific tyrosines, we have generated mammalian cells equipped with both sTyr biosynthetic and translational machinery. To generate mammalian cells capable of biosynthesizing sTyr, we used piggybac system to stably integrate Nn SULT1C1 into the genome of HEKT cells, yielding the HEKT- Nn SULT1C1 cell line Fig.

coli and Bacillus stearothermophilus tRNA CUA Tyr Fig. To evaluate the function of Nn SULT1C1 in mammalian cells, pAcBac2.

The expression of EGFP was monitored by confocal microscopy 2 days after transfection. In addition to confocal imaging, flow cytometry was used to quantify expression levels of EGFP in cells fed with exogenous sTyr and in cells biosynthesizing sTyr.

As shown in Fig. The fidelity of site-specific incorporation of sTyr was evaluated by mass spectral analysis of purified sTyr-containing EGFP proteins. These results demonstrate that the generation of mammalian cells autonomously able to biosynthesize sTyr and incorporate it into proteins significantly enhances the expression level of sTyr-containing protein in mammalian cells.

a Schematic representation of genetic circuits used for generating completely autonomous mammalian cells with sTyr-containing proteins.

b Confocal images of HEKT exogenously fed and HEKT- Nn SULT1C1 bio cells expressing sTyrRS, tRNA CUA and EGFP containing an amber codon at Tyr39 position. c Flow cytometric analysis of EGFP expression levels of HEKT exogenously fed and HEKT- Nn SULT1C1 bio cells with sTyrRS, tRNA CUA and EGFP containing an amber codon at Tyr39 position.

The normalized fluorescence was calculated by multiplying the geometric mean fluorescence by the percentage of EGFP-positive cells. Error bars represent standard deviations.

d Mass spectra analysis of EGFP with sTyr EGFPsTyr purified from HEKT- Nn SULT1C1 cells. Thrombin inhibitors represent an important class of anticoagulants used to prevent blood clotting.

In addition, several thrombin inhibitors from hematophagous organisms have been shown to facilitate the acquisition and digestion of bloodmeal 77 , 78 , Recent studies have reported that post-translational sulfation of these proteins has a dramatic effect on their inhibitory activity 31 , For example, tyrosine sulfation of hirudin increases its affinity for thrombin by more than fold 80 , Tyrosine sulfation of madanin-1 and chimadanin significantly increases their affinities for thrombin by promoting strong electrostatic interactions with positively-charged residues Fig.

Current methods for studying these site-specifically sulfated thrombin inhibitors rely heavily on solid-phase peptide synthesis and subsequent chemical ligation, processes that are time-consuming and may result in sub-optimal protein folding 31 , 82 , To explore the generation sTyr-containing thrombin inhibitors using cells endowed with autonomous sTyr biosynthetic machinery, we chose both madanin-1 and chimadanin identified in the salivary gland of haemaphysalis longicornis Fig.

To express the site-specifically sulfated thrombin inhibitors, we constructed plasmids encoding the thrombin inhibitor and substituted with amber codons at either or both of the indicated Tyr sites. sTyr-containing inhibitors were expressed by transforming ΔcysH BW cells with pEvol- Nn SULT1C1-cysDNCQ, pUltra-sTyr, and a plasmid encoding the thrombin inhibitor.

The site-specific sulfation of madanin-1 and chimadanin was further validated using SDS-PAGE and ESI-MS analysis Fig. Surface representation of positive electrostatic potential in blue and negative electrostatic potential in red.

b Amino acid sequences of madanin-1 and chimadanin. Sulfation sites are shown in red. c SDS-PAGE analysis of thrombin inhibitors with site-specific sTyr insertion expressed in completely autonomous E.

d , e Inhibition of thrombin activity by madanin-1 and chimadanin proteins. To test the thrombin inhibiting activity of the wildtype inhibitors and their sTyr-containing mutants, we performed chromogenic thrombin amidolytic activity assays in the presence of a range of concentrations of each inhibitor.

Double sulfation of chimadanin at both Tyr28 and Tyr31 further improved its Ki to 0. Furthermore, sTyr-containing thrombin inhibitors prepared using cells with completely autonomous sTyr biosynthetic machinery are more potent than chemically synthesized ones This may due to the fact that co-translational folding is more efficient than that achieved via chemical synhesis.

These data demonstrate the advantages of producing therapeutic proteins with site-specific sTyr modifications using completely autonomous cells with the ability to biosynthesize and genetically encode the sTyr.

In this research, we have generated completely autonomous bacterial and mammalian cells endowed with machinery for both sTyr biosynthesis and site-specific incorporation into proteins.

Nn SULT1C1-mediated biosynthesis of sTyr from tyrosine and PAPS was discovered using a SSN, and the unique specificity of Nn SULT1C1 for tyrosine was systematically explored using both bioinformatic and computational methods.

Use of Nn SULT1C1 and other optimized components allowed us to engineer both bacterial and mammalian cells capable of autonomously biosynthesizing sTyr and genetically incorporating it into proteins.

The value of these completely autonomous cells was further demonstrated via their use in the preparation of therapeutic sTyr-containing proteins with enhanced efficacy. More than ncAAs have been genetically incorporated into proteins in a site-specific manner, providing powerful tools for investigating protein structures and functions 1 , 2 , 3 , 6 , 85 , 86 , 87 , 88 , 89 , 90 , 91 , 92 , To date, utilizing these ncAAs in the context of Genetic Code Expansion has required both exogenous feeding and good membrane permeability of chemically-synthesized ncAAs.

Cell membranes are poorly permeable to ncAAs with charged or polar structures. Thus, intracellular biosynthesis of these ncAAs is likely to significantly expand the utility of Genetic Code Expansion technology.

Attempts to engineer cells for autonomous ncAA biosynthesis without external addition of precursors have frequently been hindered by the scarcity of verified biosynthetic pathways for producing ncAAs at high concentrations. For this reason, biosynthetic pathways for p AF, p Thr, 5HTP, and DOPA are the only ones that have been applied to bacterial cells for intracellular ncAA biosynthesis from simple carbon sources 12 , 19 , 20 , 21 , We expect that the combination of bioinformatics and ncAA screening methods reported in this work can be a powerful strategy for enlarging the repertoire of biosynthesized ncAA for Genetic Code Expansion.

Our study further reports the construction of a completely autonomous mammalian cell line capable of biosynthesizing sTyr and incorporating it into proteins in response to the amber codon.

The creation of additional mammalian cells with the endogenous ability to biosynthesize ncAAs and use them for protein synthesis will expand the preparation of therapeutic proteins, as well as allow application of the Genetic Code Expansion technology at the level of whole organisms.

The UniProt database was selected and the e value was set as 5. The resulting network was finalized by setting the alignment score threshold as to generate edges representing pairwise sequence similarities.

When the OD of the cell culture reached 0. The control cells transformed with pUltra-sTyrRS, pET22b-T5-sfGFPTAG and pEvol-empty were grown under the same condition with an indicated concentration of sTyr. The purified protein was used for SDS-PAGE and ESI-MS analysis. The structure of Nn SULT1C1 was predicted by AlphaFold2 using GitHub AlphaFold code 2.

The database, including reduced BFD, PDB70, MGnify, and Uniclust30, was used to filter structural templates. All other settings were set as default.

Based on pLDDT, the top structure was output and used in this study. The protein-ligand docking process was performed by Glide v8.

Glide uses the OPLS3 force field to evaluate the docking procedure. Four protein structures, including 2zvq, 2a3r, 2gwh, and the predicted structure of Nn SULT1C1, are taken into consideration for docking. The PAPS-binding site for the predicted Nn SULT1C1 structure is inferred by aligning with the structure of 2a3r.

For other structures, we used the original PAP sites in reported co-crystal structures to install PAPS. A short run of protein-ligand energy minimization was performed to remove the steric clashes for each of the complexes.

The docking box was inferred from the position of dopamine in 2a3r. The RMSD is set to 0. All parameters are set to default SP mode in the Glide software. The number of maximum output poses for each docking protein was set to and the top 50 poses ranked by Emodel score were picked out.

HEKT and HEKT- Nn SULT1C1 cells were transfected with pAcBac2. After being washed with PBS pH 7. The rest of cells were used for flow cytometry analysis with Sony SA Flow Cytometer where a total of 20, cells were analyzed for each sample.

Data were processed with FlowJo. Reported data are the average measurement of three independent samples prepared at the same time with the standard deviation. N- p -Tosyl -GPR- p NA acetate Cayman Chemicals was used as a chromogenic substrate to test the amidolytic activity of human α-thrombin Haematologic Technologies.

Inhibition assays were performed in the assay buffer with 0. Inhibition constants K i were determined based on a Morrison equation within GraphPad Prism. Three independent samples were prepared for each group.

All statictics analysis were performed using GraphPad Prism. Similar results were obtained from three independent experiments. Further information on research design is available in the Nature Research Reporting Summary linked to this article.

All data generated in this study are included in the paper and supplementary information. Source data are provided with this paper.

Wang, L. Expanding the Genetic Code. Article PubMed CAS Google Scholar. Ambrogelly, A. Natural expansion of the genetic code. Article CAS PubMed Google Scholar. Liu, C. Adding new chemistries to the genetic code.

Chin, J. Expanding and reprogramming the genetic code of cells and animals. Dien, V. Expansion of the genetic code via expansion of the genetic alphabet. Article CAS PubMed PubMed Central Google Scholar.

Expanding and reprogramming the genetic code. Nature , 53—60 Article ADS CAS PubMed Google Scholar. Agostini, F. et al. Biocatalysis with Unnatural Amino Acids: Enzymology Meets Xenobiology. Article CAS Google Scholar. Manandhar, M. Genetic Code Expansion: Inception, Development, Commercialization.

Expanding the Genetic Code of Escherichia coli. Science , — Furter, R. Expansion of the genetic code: Site-directed p-fluoro-phenylalanine incorporation in Escherichia coli. Protein Sci. Giese, C.

Intracellular uptake and inhibitory activity of aromatic fluorinated amino acids in human breast cancer cells. ChemMedChem 3 , — Zhang, M. Biosynthesis and genetic encoding of phosphothreonine through parallel selection and deep sequencing. Methods 14 , — Luo, X. Genetically encoding phosphotyrosine and its nonhydrolyzable analog in bacteria.

Hoppmann, C. Site-specific incorporation of phosphotyrosine using an expanded genetic code. Bundy, B. Ko, W. Construction of Bacterial Cells with an Active Transport System for Unnatural Amino Acids. ACS Synth. Burkovski, A. Bacterial amino acid transport proteins: occurrence, functions, and significance for biotechnological applications.

Palacín, M. Molecular Biology of Mammalian Plasma Membrane Amino Acid Transporters. Article PubMed Google Scholar. Mehl, R. Generation of a Bacterium with a 21 Amino Acid Genetic Code.

Chen, Y. A noncanonical amino acid-based relay system for site-specific protein labeling. Rogerson, D. Efficient genetic encoding of phosphoserine and its non-hydrolyzable analog. Creation of Bacterial Cells with 5-Hydroxytryptophan as a 21st Amino Acid Building Block.

Völler, J. Coupling genetic code expansion and metabolic engineering for synthetic cells. Wang, Y. Expanding the Structural Diversity of Protein Building Blocks with Noncanonical Amino Acids Biosynthesized from Aromatic Thiols.

Exner, M. Design of S-Allylcysteine in Situ Production and Incorporation Based on a Novel Pyrrolysyl-tRNA Synthetase Variant. ChemBioChem 18 , 85—90 Biosynthesis and Genetic Incorporation of 3,4-Dihydroxy-L-Phenylalanine into Proteins in Escherichia coli.

Veldkamp, C. Article PubMed PubMed Central CAS Google Scholar. Ludeman, J. The structural role of receptor tyrosine sulfation in chemokine recognition. Farzan, M. Tyrosine Sulfation of the Amino Terminus of CCR5 Facilitates HIV-1 Entry. Cell 96 , — Choe, H.

Tyrosine Sulfation of Human Antibodies Contributes to Recognition of the CCR5 Binding Region of HIV-1 gp Cell , — Thompson, R. Tyrosine sulfation modulates activity of tick-derived thrombin inhibitors. Somers, W. Insights into the Molecular Basis of Leukocyte Tethering and Rolling Revealed by Structures of P- and E-Selectin Bound to SLeX and PSGL Westmuckett, A.

Tyrosine Sulfation of Native Mouse Psgl-1 Is Required for Optimal Leukocyte Rolling on P-Selectin In Vivo. PLOS ONE 6 , e Article ADS CAS PubMed PubMed Central Google Scholar. Lee, S. A Type I—Secreted, Sulfated Peptide Triggers XAMediated Innate Immunity. Li, X. Characterization of a Sulfated Anti-HIV Antibody Using an Expanded Genetic Code.

Biochemistry 57 , — Recombinant expression of selectively sulfated proteins in Escherichia coli. Efficient expression of tyrosine-sulfated proteins in E.

coli using an expanded genetic code. Protein evolution with an expanded genetic code. Natl Acad. USA , — Mutagenesis and Evolution of Sulfated Antibodies Using an Expanded Genetic Code. Biochemistry 48 , — Schwessinger, B. A second-generation expression system for tyrosine-sulfated proteins and its application in crop protection.

Yang, Y.