About

Seth Herzon is a Professor of Chemistry in the Department of Chemistry at Yale University and a member of the Yale faculty since 2008. His research interests focus on organic chemistry and chemical biology, particularly the phenomenon of emergence in bioactive natural products. His laboratory aims to formalize the study of emergence, connecting modular, evolved chemical reactivity within natural products to cellular phenotypes. This work seeks to deepen understanding of structure, bonding, and reactivity in complex systems, advance synthetic strategies and methodologies, and facilitate the creation of novel molecules with complex properties that have implications for human health. Herzon's contributions include developing innovative approaches to synthesizing molecules with potential anti-cancer effects, targeting drug-resistant brain cancers, and advancing the understanding of natural product chemistry. His work is recognized through numerous awards and honors, including the American Chemical Society Award for Creative Work in Synthetic Organic Chemistry and the Yale Faculty Innovation Award.

Research topics

• Biology
• Genetics
• Cancer research
• Stereochemistry
• Chemistry
• Microbiology
• Organic chemistry
• Crystallography
• Immunology
• Biochemistry
• Computational chemistry
• Pathology
• Medicine

Selected publications

• Practical Machine Learning Strategies. 5. Rapid, Accurate Prediction of 1H and 13C NMR Shifts in Conformationally Flexible Organic Molecules

  ChemRxiv · 2026-03-30

  articleOpen access

  Accurate prediction of NMR chemical shifts for conformationally flexible natural products remains a central challenge in structure elucidation. Traditional quantum‑chemical (QM) approaches typically achieve ≈2.0–3.0 ppm 13 C accuracy for flexible natural products but often require hours of computation per molecule. MLHF‑NMR reproduces high‑level QM 13 C accuracy for flexible molecules while reducing computation times by factors of hundreds to more than a thousand, enabling routine, high‑throughput NMR prediction for conformationally flexible natural products. Against the high‑accuracy DELTA50 experimental 1 H and 13C shift benchmark, MLHF‑NMR achieves RMS errors of 0.15 ppm ( 1 H) and 1.73 ppm ( 13 C). Across more than 8.2k 13 C chemical shifts from 420 predominantly flexible natural products, the protocol yields an average RMS error of 2.23 ppm. Practical applications include structure assignment for newly isolated natural products, structure confirmation in independently synthesized molecules, rapid detection and correction of misassignments, and strategies for improving predicted 13C shift accuracy. Validation sets and sample MLHF‑NMR 13 C shift predictions for more than 1900 synthetic drugs and drug‑like molecules are available at https://mlnmr.wavefun.com. For flexible molecules, MLHF‑NMR routinely delivers 13 C shift accuracy comparable to the best QM methods while reducing computation time by 2–3 orders of magnitude.

  PublisherDOI

• Practical Machine Learning Strategies. 5. Rapid, Accurate Prediction of 1H and 13C NMR Shifts in Conformationally Flexible Organic Molecules

  ChemRxiv · 2026-05-21

  articleOpen access

  Accurate prediction of NMR chemical shifts for conformationally flexible natural products remains a central challenge in structure elucidation. Traditional quantum‑chemical (QM) approaches typically achieve ≈2.0–3.0 ppm 13 C accuracy for flexible natural products but often require hours of computation per molecule. MLHF‑NMR reproduces high‑level QM 13 C accuracy for flexible molecules while reducing computation times by factors of hundreds to more than a thousand, enabling routine, high‑throughput NMR prediction for conformationally flexible natural products. Against the high‑accuracy DELTA50 experimental 1 H and 13C shift benchmark, MLHF‑NMR achieves RMS errors of 0.15 ppm ( 1 H) and 1.73 ppm ( 13 C). Across more than 8.2k 13 C chemical shifts from 420 predominantly flexible natural products, the protocol yields an average RMS error of 2.23 ppm. Practical applications include structure assignment for newly isolated natural products, structure confirmation in independently synthesized molecules, rapid detection and correction of misassignments, and strategies for improving predicted 13C shift accuracy. Validation sets and sample MLHF‑NMR 13 C shift predictions for more than 1900 synthetic drugs and drug‑like molecules are available at https://mlnmr.wavefun.com. For flexible molecules, MLHF‑NMR routinely delivers 13 C shift accuracy comparable to the best QM methods while reducing computation time by 2–3 orders of magnitude.

  PublisherDOI

• Three-component assembly and structure–function relationships of (–)-gukulenin A.

  ChemRxiv · 2025-07-28

  preprintOpen accessSenior author

  (–)-Gukulenin A is a cytotoxic secondary metabolite that bears two aromatic α-tropolone residues, 10 asymmetric stereocenters, functionalized cyclopentane fragments, a labile hemiketal, and an electrophilic aldehyde. Here we describe an enantioselective synthesis of (–)-gukulenin A (7) by a three-component assembly strategy. Key steps include the directed C–H arylation of a norbornyl picolinamide, a tandem Grob fragmentation–alkylation, a novel method for the synthesis of α-tropolones, a site-selective, multicomponent cross-coupling reaction, and a thermal carbonyl–ene reaction to complete the carbon skeleton. (–)-Gukulenin A (7) is efficacious and well-tolerated in murine models of ovarian cancer. Structure–function studies establish the dimeric tropolone and aldehyde substructures within (–)-gukulenin A (7) as critical drivers of cytotoxicity.

  PublisherOA PDFDOI

• Total Synthesis of Chartelline C

  Journal of the American Chemical Society · 2025-07-01 · 3 citations

  articleSenior authorCorresponding

  Chartellines are cytotoxic marine alkaloids that were isolated in the late 1980s. Their unique, heavily oxidized structures, comprising a spirocyclic β-lactam, a halogenated indolenine, a chloroenamide, and a 2-haloimidazole, have motivated extensive efforts toward their synthesis. However, only a single synthesis of any chartelline has been reported in the nearly 40 years since their isolation. Here, we describe a route to chartelline C (3) from the macrolactam 21, an intermediate we employed en route to the related securine and securamine alkaloids. The key challenges in converting macrolactam 21 to chartelline C (3) involve stereocontrolled construction of the C2–C3 enamide, isomerization of the trans-C10–C11 alkene, formation of the strained spirocyclic β-lactam, and chlorination of the enamide. Each of these challenges was addressed by an unconventional approach. The macrolactam 21 was converted to cis-enamide (2Z,10E)-24 by in situ masking of the C12 ketone, followed by stereoselective acid-catalyzed elimination of the C2 carbinolamide. The C10–C11 trans-alkene was isomerized by a photolytic process; control experiments suggest that this isomerization occurs by energy transfer. The β-lactam was constructed by a solid-state reaction involving the adsorption of the bromoindolenine 26 onto activated basic alumina. Late-stage enamide chlorination, which had been an obstacle in prior approaches, was accomplished by a novel photoredox-mediated halogenation. These latter studies suggest that N-haloanomeric amides and N-haloguanidines may act as halogen atom transfer agents under thermal or photolytic conditions. Finally, we show that chartelline C (3) reacts with sulfur-based nucleophiles, suggesting that further studies of their biological activity may be warranted.

  PublisherDOI

• Structural study on human microbiome-derived polyketide synthases that assemble genotoxic colibactin

  Structure · 2025-05-16 · 4 citations

  article
  PublisherDOI

• A Stereoselective Oxidative Dimerization En Route to (−)-Lomaiviticin A

  Organic Letters · 2025-01-23 · 1 citations

  articleOpen accessSenior authorCorresponding

  We describe a stereoselective synthesis of the dimeric diazofluorene 15, a potential precursor to the cytotoxic C2-symmetric bacterial metabolite (−)-lomaiviticin A (1). An efficient route was developed to convert the tetracyclic diol 5 to the diketone 4 (five steps, 30% overall). Oxidative dimerization of the enoxysilane 14 provided the C2-symmetric dimeric diazofluorene 15 in 56% yield and with 15:1:0 diastereoselectivity. Deprotection and 2D NMR analysis indicated that the major diastereomer possessed the (2S,2′S) configuration found in 1. This approach may ultimately be useful in the synthesis of 1 itself.

  PublisherOA PDFDOI

• Genome-scale CRISPR/Cas9 screening reveals the role of <i>PSMD4</i> in colibactin-mediated cell cycle arrest

  mSphere · 2025-02-07 · 5 citations

  articleOpen access

  ABSTRACT Colibactin is a genotoxic secondary metabolite produced by certain Enterobacteriaceae strains that populate the intestine and produces a specific mutational signature in human colonocytes. However, the host pathways involved in colibactin response remain unclear. To address this gap, we performed genome-wide CRISPR/Cas9 knockout screens and RNA sequencing utilizing live pks + bacteria and a synthetic colibactin analog. We identified 20 enriched genes with a MAGeCK score of &gt;2.0 in both screens, including proteasomal subunits (e.g., PSMG4 and PSMD4 ), RNA processing factors (e.g., SF1 and PRPF8 ), and RNA polymerase III (e.g., CRCP ), and validated the role of PSMD4 in colibactin sensitization. PSMD4 knockout in HEK293T and HT-29 cells promoted cell viability and ameliorated G2-M cell cycle arrest but did not affect the amount of phosphorylated H2AX foci after exposure to synthetic colibactin 742. Consistent with these observations, PSMD4 −/− cells had a significantly higher colony formation rate and bigger colony size than control cells after 742 exposure. These findings suggest that PSMD4 regulates cell cycle arrest following colibactin-induced DNA damage and that cells with PSMD4 deficiency may continue to replicate despite DNA damage, potentially increasing the risk of malignant transformation. IMPORTANCE Colibactin has been implicated as a causative agent of colorectal cancer. However, colibactin-producing bacteria are also present in many healthy individuals, leading to the hypothesis that some aspects of colibactin regulation or host response dictate the molecule’s carcinogenic potential. Elucidating the host-response pathways involved in dictating cell fate after colibactin intoxication has been difficult, partially due to an inability to isolate the molecule. This study provides the first high-throughput CRISPR/Cas9 screening to identify genes conferring colibactin sensitivity. Here, we utilize both bacterial infection and a synthetic colibactin analog to identify genes directly involved in colibactin response. These findings provide insight into how differences in gene expression may render certain individuals more vulnerable to colibactin-initiated tumor formation after DNA damage.

  PublisherDOI

• The Fanconi anemia pathway repairs colibactin-induced DNA interstrand cross-links

  Nature Communications · 2025-11-26

  articleOpen access

  Colibactin is a secondary metabolite produced by bacteria present in the human gut and is implicated in the development of colorectal cancer. This genotoxin alkylates deoxyadenosines on opposite strands of host cell DNA to produce DNA interstrand cross-links. While cells have evolved multiple mechanisms to resolve ("unhook") interstrand cross-links, little is known about which of these pathways promote resistance to colibactin. Here, we use Xenopus egg extracts to investigate replication-coupled repair of colibactin-induced interstrand cross-links. We show that replication fork stalling at a colibactin-induced interstrand cross-link activates the Fanconi anemia interstrand cross-link repair pathway, which unhooks the interstrand cross-link through nucleolytic incisions. These incisions generate a DNA double-strand break intermediate in one sister chromatid, which can be repaired by homologous recombination, and a monoadduct ("interstrand cross-link remnant") in the other. Translesion synthesis past the colibactin-induced interstrand cross-link remnant depends on Pol η and the Pol κ-REV1-Pol ζ polymerase complex and introduces predominantly T>A point mutations at the sites of colibactin alkylation. Taken together, our work provides a molecular framework for understanding how cells tolerate a naturally occurring and clinically relevant interstrand cross-link.

  PublisherOA PDFDOI

• Synthesis and Biological Evaluation of Colibactin Derivatives

  ACS Chemical Biology · 2025-12-19

  articleOpen accessSenior authorCorresponding

  Colibactin is a pseudo-C2-symmetric gut microbiome metabolite that induces DNA interstrand cross-links and plays a causal role in colorectal cancer. Since efforts to isolate colibactin have not been successful, we developed colibactin 742 (3a/b) as a stable colibactin mimetic. However, colibactin 742 (3a/b) exists as a mixture of ring and chain isomers, which complicates analysis of its activity. We report here the discovery of colibactin 686 (9) as a superior colibactin mimetic. Colibactin 686 (9) is more potent than colibactin 742 (3a/b) and recapitulates the bacterial genotoxic phenotype. Colibactin 686 (9) possesses a C2-symmetric structure, which will expedite its synthesis, and is incapable of ring–chain isomerization, which will simplify analysis of its biological activity. We additionally establish that colibactins do not passively diffuse into cells, and are substrates for monocarboxylate transporter pumps. These latter findings have implications for trafficking of natural colibactin, which remains poorly understood.

  PublisherOA PDFDOI

• Targeted CRISPR knockout screening identifies known and novel chemogenomic interactions between DNA damaging agents and DNA repair genes

  NAR Cancer · 2025-11-27 · 1 citations

  articleOpen access

  Genetic instability is a hallmark of cancer, often arising from mutations to DNA damage repair and response (DDR) genes. Classical genetic, biochemical, and structural approaches elucidated the foundational mechanisms of DDR pathways and provided a scientific understanding of their involvement in repair of lesions induced by broad classes of DNA-damaging agents (DDAs). However, given the chemical diversity of DDAs and resultant DNA lesions, along with the multitude of interconnected DDR factors, the chemogenomic landscape of DDA-DDR interactions remains incompletely mapped. To this end, we developed a DDR-targeted, CRISPR knockout screening approach and assessed relationships amongst 353 DNA repair genes and 15 DDAs in LN229 glioma cells. Within this dataset of 5295 DDR-related chemogenomic interactions, we identified many established interactions and discovered novel ones. For example, we observed a specific role of transcription-coupled nucleotide excision repair in the repair of adducts generated by monofunctional alkylating agents, a role for the Fanconi anemia pathway in addressing methyl lesions, overt differences in DSB repair following treatment with topoisomerase I versus II poisons, and repair dependencies associated with the imidazotetrazines temozolomide, mitozolomide, and KL-50. Future directions will continue to investigate the mechanisms of novel chemogenomic interactions that we have uncovered as well as work to identify chemogenomic interactions amenable to clinical translation.

  PublisherDOI

Recent grants

• Synthesis and Study of Complex Antiproliferative and Antimalarial Natural Products

  NIH · $1.2M · 2015–2019

• Structural and functional studies of colibactin

  NIH · $5.5M · 2017–2028

• CAREER: Total Synthesis of Complex Alkaloid Natural Products

  NSF · $550k · 2012–2017

• NIH Grant F32GM078907

  NIH · $77k · 2009

• Synthetic and translational studies of antitumor and antimicrobial natural products.

  NIH · $4.7M · 2019–2029

Frequent coauthors

• Martin Tomanik

  New York University

  25 shared

• Mengzhao Xue

  Yale University

  22 shared

• Kevin M. Wernke

  Yale University

  21 shared

• Jason M. Crawford

  Yale University

  20 shared

• Xiaoshen Ma

  Merck & Co., Inc., Rahway, NJ, USA (United States)

  19 shared

• Mingshuo Zeng

  18 shared

• Grant R. Krow

  Temple University

  15 shared

• Alan R. Healy

  14 shared

Labs

• Herzon LabPI

Awards & honors

• Petroleum Research Fund Undergraduate Research Fellowship (2…
• The F. Albert Cotton and Al Tulinsky Prize (2002)
• National Science Foundation Graduate Research Fellowship (20…
• Harvard University Certificate of Distinction in Teaching (2…
• National Institutes of Health Postdoctoral Fellowship (2006)