ABSTRACT
Metabolic reprogramming is an important hallmark of cancer. Of the altered metabolic pathways associated
with malignancy, one-carbon (C1) metabolism is particularly notable. The 3-carbon of serine is the major C1
donor for de novo synthesis of purines and thymidylate in the cytosol, and the primary catabolic pathway for
serine and synthesis of glycine occurs in the mitochondria. The mitochondrial C1 pathway also generates
reducing equivalents and is an important source of ATP. The first enzyme of the mitochondrial C1 pathway,
serine hydroxymethyltransferase (SHMT) 2, is an oncodriver which is upregulated in a substantial number of
cancers. Growing evidence suggests that SHMT2 could be an independent prognostic factor and an important
therapeutic target for cancer. We discovered novel 5-substituted pyrrolo[3,2-d]pyrimidine compounds AGF291,
AGF347, and AGF359. Following their internalization by the proton-coupled folate transporter (PCFT), these
compounds inhibit mitochondrial C1 metabolism at SHMT2, with direct secondary inhibitions of cytosolic
targets in de novo purine (DNP) biosynthesis (at 5-aminoimidazole-4-carboxamide ribonucleotide
formyltransferase and glycinamide ribonucleotide formyltransferase) and SHMT1. Our compounds inhibit
proliferation of epithelial ovarian cancer, non-small cell lung cancer, colorectal cancer, and pancreatic cancer
(PaC) cells, suggesting their potential as broad-spectrum anti-tumor agents. AGF347 exhibited significant in
vivo antitumor efficacy with potential for complete responses against both early and upstage PaC xenograft
models. We posit that our novel compounds offer an entirely new approach for treating cancer. Our objective
is to optimize our lead structures for tumor targeting via PCFT and inhibition of mitochondrial and cytosolic
C1 metabolism at modest doses with minimal toxicity. We will use PaC as a disease prototype for further
development of our novel multi-targeted inhibitors. In Aim 1, we will synthesize up to 100 compounds based
on lead compounds to optimize uptake by tumors, and inhibition of SHMT2 and cytosolic pathways including
DNP biosynthesis. In Aim 2, we will test analogs from Aim 1 for antitumor potencies toward clinically relevant
PaC cell lines, tumor selectivity and plasma membrane and mitochondrial drug transport, drug metabolism, and
inhibition of SHMT2 and cytosolic pathways including DNP biosynthesis. We will measure downstream impacts
on mTOR signaling, mitochondrial respiration, glutathione pools, and reactive oxygen species. In Aim 3, we
will evaluate pharmacokinetics, tolerability, and in vivo antitumor activities of compounds from Aims 1 and 2 by
toxicity/efficacy trials with human PaC cell line xenograft and PDX models, and with the KPC mouse PaC model.
Our lead analogs are “first-in-class” and our proposed studies will afford optimized compounds with the best
balance of selective tumor targeting and anti-tumor efficacy, resulting from inhibition of SHMT2 and
downstream anabolic pathways. We anticipate developing SHMT2/DNP-targeted compounds for IND
submission and clinical trials based on our studies.