Targeting Bacillosamine Biosynthesis in Bacterial Pathogens: Development of Inhibitors to a Bacterial Amino-Sugar Acetyltransferase from Campylobacter jejuni

Abstract

The glycoproteins of selected microbial pathogens often include highly modified carbohydrates such as 2,4-diacetamidobacillosamine (diNAcBac). These glycoconjugates are involved in host cell interactions and may be associated with the virulence of medically-significant Gram-negative bacteria. In light of genetic studies demonstrating the attenuated virulence of bacterial strains in which modified carbohydrate biosynthesis enzymes have been knocked out, we are developing small molecule inhibitors of selected enzymes as tools to evaluate whether such compounds modulate virulence.

We performed fragment-based and high-throughput screens against an amino-sugar acetyltransferase enzyme, PglD, involved in biosynthesis of UDP-diNAcBac in C. jejuni. Herein we report optimization of the hits into potent small molecule inhibitors (IC₅₀ <300 nM). Biophysical characterization shows that the best inhibitors are competitive with acetyl coenzyme A and an X-ray co-crystal structure reveals that binding is biased towards occupation of the adenine sub-pocket of the AcCoA binding site by an aromatic ring.

Graphical Abstract

Introduction

Most clinically relevant antibiotics are targeted at essential bacterial survival functions including replication, transcription, translation, and cell wall biosynthesis.¹ This is an outcome of in vitro antibiotic screening and discovery, which has relied on observing cell death in culture in laboratory settings. The emergence of resistance against all major mechanisms of antibiotic action requires new paradigms for combating infectious disease. A potentially promising approach includes the development of agents that target bacterial virulence in vivo in human hosts. Such approaches may mitigate the effects of infectious disease, while potentially resulting in less selective pressure for resistance development.² Virulence factors are implicated in many bacterial processes including host-cell adhesion, invasion, and colonization, as well as quorum sensing and biofilm formation.^2–5 In order to develop antivirulence agents, it is critical to identify validated pathogen-specific processes that cause virulence in the targeted human hosts.

Protein glycosylation is widespread in nature and regulates a variety of cellular functions including protein folding, cell-cell interactions, cell signaling, and the host immune response.⁶ Glycans are attached to proteins via either serine/threonine (O-linked) or the amide nitrogen of asparagine (N-linked). It is now recognized that selected bacteria possess the biosynthetic machinery for O- and/or N-glycosylation and that this modification may play a role in pathogenicity.^7–11 N-glycosylation was first discovered in C. jejuni in 1999 and the protein glycosylation (pgl) pathway has been characterized in detail for this organism (Figure 1).^12,13 In C. jejuni, more than 60 proteins, including confirmed virulence factors, are modified with a conserved heptasaccharide.¹⁴

Figure 1.

Schematic representation of the C. jejuni N-linked protein glycosylation pathway. Enzymes are shown in italics with the oligosaccharyl transferase PglB shown as determined in PDB 3RCE. Also shown is an N-linked glycosylation substrate PEB3 (PDB: 2HXW), which is a virulence factor in periplasm that is modified by N-linked glycosylation. Inset highlights the three sugar-modifying enzymes that convert UDP-GlcNAc to UDP-diNAcBac.

In a significant divergence between prokaryotes and eukaryotes, bacteria and archaea have specialized enzymatic processes to modify the structures of selected carbohydrates for incorporation into glycoconjugates. Furthermore, the discovery of unique prokaryote-specific sugars is continuing with the pace of bacterial genome sequencing and bioanalytical methods development.⁷ In contrast to the glycosyltransferase enzymes, which assemble complex glycans and share common folds and mechanisms across domains of life, the specialized sugar-modifying enzymes are attractive targets for developing targeted antivirulence agents because they tend not to have mammalian homologs and because the associated glycoconjugates are linked with bacterial pathogenicity.¹⁵ Of particular interest is di-N-acetylbacillosamine (diNAcBac),¹⁶ which is derived from N-acetylglucosamine (GlcNAc). DiNAcBac is found, for example, at the reducing end of O-linked glycans in selected strains of N. gonorrhoeae and A. baumannii, and N-linked glycans in C. jejuni, highlighting the importance of diNAcBac biosynthesis pathways in these Gram-negative pathogens.¹⁷ Intriguingly, while protein glycosylation pathways from these pathogens glycosylate diverse proteins with different glycans, the reducing-end sugar is diNAcBac.^18–20 The N-linked protein glycosylation (pgl) pathway of C. jejuni and steps leading to diNAcBac biosynthesis are illustrated in Figure 1. The first two steps of UDP-diNAcBac biosynthesis utilize an NAD⁺-dependent dehydratase (PglF) followed by a pyridoxal phosphate-dependent aminotransferase (PglE) to produce a UDP-4-amino-sugar, which is then acetylated by PglD using acetyl coenzyme A (AcCoA) as a co-substrate (Figure 1 inset).¹⁸ Subsequent glycan assembly onto an undecaprenyl-diphosphate carrier, is catalyzed by a series of glycosyl transferases. After assembly, the completed heptasaccharide is translocated across the inner membrane and the glycan is transferred to protein substrates in the bacterial periplasm by the oligosaccharyl transferase PglB.

Studies have shown that disruption of genes responsible for diNAcBac biosynthesis (pglF, pglE, pglD) perturb production of the native heptasaccharide glycosyl donor in C. jejuni.²¹ In addition, ΔpglD and ΔpglE strains show greatly reduced colonization of the gastrointestinal tract of 1-day-old chicks, thereby establishing a link between protein N-glycosylation and C. jejuni pathogenicity in host cells.²² Further insight into these effects came from transposon mutagenesis experiments, which identified pglF and pglE as essential genes for colonization. In mice, mutation of pglE impaired invasion of intestinal epithelial cells and colonization of the gut.²³ The causative glycoconjugates underpinning these findings remain unknown, but several molecular associations between N-glycosylation and C. jejuni virulence have been defined. For example, VirB10, a structural component to the type IV secretion system (TFSS), needs to be glycosylated at Asn97, otherwise a 10-fold decrease in natural competency results.²⁴ Recently, 16 N-linked glycoproteins were identified and found to be associated with C. jejuni outer membrane vesicles (OMVs) including the PEB3 adhesin.²⁵ Pathogens deploy OMVs to deliver bacterial proteins into host cells, making this an important finding in the relationship between periplasmic glycoproteins and virulence.²⁶ Protein O-glycosylation is also associated with virulence; for example, loss of glycosylation of PilE, a constituent of the type IV pilin in N. gonorrhoeae, leads to a significant decrease in epithelial cell adhesion.²⁷

Herein we report the development of inhibitors targeting the C. jejuni PglD, a UDP-amino-sugar acetyltransferase, which catalyzes the third step in diNAcBac biosynthesis. PglD represents an attractive target for inhibitor development as it is well understood from a structural and mechanistic perspective.^28,29 Additionally, PglD is a soluble, well-expressed enzyme, which makes it tractable for structure/activity-driven inhibitor discovery. Crystallographic analysis of PglD reveals a homotrimeric structure with three equivalent active sites formed at the interfaces between adjacent protomers.^28,29 Like many other bacterial UDP-amino-sugar acetyltransferases, PglD is a member of the left-handed β-helix family comprising two domains (Figure 2a); a N-terminal domain with a β-α-β-α-β-α Rossmann fold motif that binds the UDP-4-amino-sugar, and a C-terminal hexapeptide repeat motif that defines the left-handed β-helix that contributes to the AcCoA binding site. PglD has been co-crystallized in the presence of the AcCoA and UDP-4-amino-sugar substrates (Figure 2b).²⁸ The structural features of PglD and related homologs differentiate these prokaryotic AcCoA-dependent acetyltransferases from their mammalian counterparts, including for example HAT1 of the GCN5-related N-acetyltransferase (GNAT) superfamily.³⁰

Figure 2.

Structure analysis of C. jejuni PglD. (a) Crystal structure of the PglD homotrimer (PDB 3BSY) in cartoon representation; individual protomers are colored in green, blue and orange. (b) Expanded of view of the PglD active site protein surface with the AcCoA substrate shown in stick representation (magenta, PDB 3BSY) as well as the UDP-4-amino-sugar substrate (yellow, PDB 3BSS). (c) Crystal structure of fragment 1 (magenta) bound to PglD with a semi-transparent surface and selected side-chain residues shown in stick representation and annotated (PDB 5TYH). (d) Docking output pose generated for compound 3 (Glide, Maestro).³⁸

Currently, many questions remain as to why modified carbohydrates, such as diNAcBac, are incorporated into bacterial glycoconjugates. Therefore, chemical tools such as small molecule inhibitors that selectively block diNAcBac biosynthesis would be valuable agents for understanding the significance of glycan diversification in selected bacteria. In addition, such tools could help to validate the pgl pathway enzymes as potential antivirulence targets. Consequently, we have established a program to develop inhibitors of diNAcBac biosynthesis with a focus on the well-characterized C. jejuni pathway. This Gram-negative pathogen is a common cause of gastroenteritis in humans and may also result in the development of Guillain-Barré syndrome, an auto-immune disease in which the peripheral nerves are attacked resulting in damage to the myelin insulation.^31,32 In recent years, C. jejuni has shown increased resistance towards front-line antibiotics including the macrolides and fluoroquinolones, which inhibit protein synthesis and DNA unwinding, respectively.³³

Results

Screening

Initial hits for the development of inhibitors for PglD were identified by employing fragment-based screening and high-throughput molecular library approaches.³⁴ The fragment-based screen was performed with a fragment set that included a selection of the Maybridge Ro3 chemical library as described previously.³⁵ The primary screen involved differential scanning fluorimetry (DSF) analysis against PglD to identify fragments that bind to and stabilize the natively-folded protein. Following this, hits were confirmed by NMR spectroscopy (WaterLOGSY and STD) and further validated using a UV-based biochemical assay, which measures CoASH release following acetyl transfer to the UDP-4-amino-sugar.³⁶ Using this workflow, 10 compound hits with IC₅₀ values in the 1–10 mM range and ligand efficiencies in the 0.29 – 0.32 range were validated. For example, compound 1 (Figure 2), which was analyzed by X-ray in complex with PglD, binds to the central AcCoA binding groove occupying the pantetheine site proximal to active-site residues involved in amino-sugar activation and thioester attack (Figure 2c). Unfortunately, efforts to build upon this, and other validated fragments yielded only modest improvements in affinity. Coincident with these studies a high-throughput screen was performed in collaboration with the Broad Institute, using a Diversity-Orientated Synthesis (DOS) compound collection (83,000) and the NIH Molecular Libraries Probe Production Centers Network (MLPCN) library (276,000) of small molecules.³⁷ Compounds were screened at 10 μM using a PglD end-point assay monitoring the generation of CoASH from the acetyltransferase reaction. Despite the low hit rate (defined as >20% inhibition) of 0.02% and 0.12% respectively, the screen led to the discovery of thienopyrimidinone-6-carboxylate compound 2 as a promising hit for further development. For direct comparison with the validated fragment hits, 2 was deconstructed to the core structure 3, which demonstrated a validated IC₅₀ value of 860 μM and LE of 0.31. In light of these results and the tractability of the core towards synthetic manipulation at four points of diversification (C2, C4, C5, and C6; Scheme 1) this compound class became the focus of our continuing medicinal chemistry efforts.

Scheme 1.

General structure and numbering system of the thienopyrimidine core.

Preliminary structure activity studies to vary substitution at C2 generated analog 4 as the best inhibitor with an IC₅₀ value of 1 μM (LE 0.33). However, although this compound represented a good improvement in potency, the candidate suffered from some disadvantages. Specifically, 4 features a planar structure with poor solubility properties and a metabolic liability due to the 1,3-unsubsituted indole and conjugated alkene systems. Furthermore, 4 and related analogs showed poor long-term stability in solution. A significant effort was also made to crystallize PglD with several of the thienopyrimidinone hits, however, this proved unsuccessful. Therefore, we performed a docking study on fragment 3 to generate a computational model as a guide for future synthetic efforts aimed at eliminating the compound liabilities while improving potency.³⁸ The thienopyrimidone-6-carboxylate core was predicted to occupy a binding site overlapping with the panthetheine moiety of AcCoA, as had been previously observed in the co-crystal structure of fragment 1 (PDB 5TYH), with similar placement of the carboxylate moiety and close agreement between the positions of the furan oxygen of 1 and exocyclic oxygen of 2 (compare Figure 2c and d). Given these observations, we exploited the computational model as a guide for the design of the next generation of analogs, which recognized the opportunity to extend the molecule both at the C2 position to interact with a hydrophobic region on the β-helix, and at the C4 position, by transitioning to the thienopyrimidine core, to further occupy the AcCoA binding groove (Scheme 1).

In order to reach cytoplasmic targets in Gram-negative pathogens, compounds may enter cells through passive diffusion or pass through the bacterial porins to reach the periplasm and then reach the cytoplasm via passive or active mechanisms.^39,40 With these parameters in mind, our overall goal was to develop potent (IC₅₀ <500 nM) inhibitors of C. jejuni PglD based on the thienopyrimidine-6-carboxylate core with good ligand efficiency (LE >0.30) together with balanced physicochemical properties including cLogP values between 2 – 4.⁴¹

Chemistry

We developed modular synthetic routes to provide access to the target thienopyrimidine carboxylic acids with readily-modifiable substitution at the C2 and C4 positions (Schemes 2 and 3).

Scheme 2.

Synthesis of inhibitors with thienopyrimidine-6-carboxylate core 11

Conditions: (a) 1.2 equiv. R₂CN, 4 M HCl in 1,4-dioxane, 90°C (50–90%); (b) 1.5 equiv. R₄NH₂, 1.3 equiv. PyBOP, 3 equiv. DBU, MeCN, RT (40–90%); (c) 1.2 equiv. ClCH₂CN, 4 M HCl in 1,4-dioxane, 90°C (50–90%) (d) 20 equiv. P(OEt)₃, 145°C (85%); (e) 1.5 equiv. NaH, 2 equiv. ArCHO, THF, RT (70–90%); (f) 10% Pd(OH)₂/C, THF/MeOH 1:1, 60°C (quant.); (g) 1.5 equiv. R₄NH₂, 1.3 equiv. PyBOP, 3 equiv. DBU, MeCN, RT (40–90%) (h) 3 equiv. NaOH, THF/MeOH/H₂O 2:1:1, RT (50–80%).

Scheme 3.

Synthesis of inhibitors with thienopyrimidine-6-carboxylate core 16

Conditions: (a) 4 equiv. SOCl₂, MeOH, 50°C (99%); (b) 2 equiv. HNO₃, H₂SO₄, RT; (c) 10% Pd(OH)₂/C, THF/MeOH 1:1, RT (93% over 2 steps); (d) 1.2 equiv. R₂CN, 4 M HCl in 1,4-dioxane, 90°C (60–80%); (e) 1.5 equiv. R₄NH₂, 1.3 equiv. PyBOP, 3 equiv. DBU, MeCN, RT (40–90%); (f) 3 equiv. NaOH in THF/MeOH/H₂O 2:1:1, RT (50–80%).

For modification at the C2-position (Scheme 2, Route I), the synthesis was initiated by condensation of commercially available thiophene, 5, with nitriles bearing the desired R₂ substituent, under strongly acidic conditions. In most cases, the thienopyrimidone products of general structure 6, proved highly insoluble, and therefore chromatography and NMR characterization was carried out after the subsequent step. In cases where nitriles with the appropriate R₂ moiety were inaccessible, an alternative approach (Route II) was employed. This route is somewhat longer and includes a Michaelis-Arbuzov reaction with triethyl phosphite and analog 6 (Core structure: R₂ = -CH₂Cl) to furnish phosphonate 8. Subsequent Horner-Wadsworth-Emmons (HWE) olefination and hydrogenation generated thienopyrimidinones of general structure 9 with a variety of phenylethyl substituents at the C2-position.

Installation of a substituent at the C4-position of thienopyrimidines is typically accomplished by transforming the C4 oxygen of the thienopyrimidone into chloride or bromide, with POCl₃ and PBr₃ respectively, followed by a S_NAr reaction, both performed at high temperatures. To avoid use of these harsh conditions and strong lachrymators, we instead employed a one-step, room temperature reaction modified from chemistry developed by researchers at Wyeth.⁴² Utilization of a sufficiently strong base, such as 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU), in excess solubilized the thienopyrimidinones in acetonitrile and replacing the toxic and expensive BOP reagent for the readily available PyBOP, allowed for robust formation of the desired pyrimidines of general structure 10 in good yields (40–90%). These thienopyrimidines could be isolated by column chromatography in excellent purity (>95% homogeneity as determined by LC-MS). The final saponification of thioenopyrimidine esters represented by 7 and 10 was achieved in a THF/MeOH/H₂O mixture containing NaOH; occasionally mild heating was required to achieve complete conversion. The final inhibitors with core scaffold 11 were isolated by trituration as easy to handle solids.

A related thienopyrimidine isomer with general structure 16, was accessed as described in Scheme 3. In this case, the thiophene starting material 14 is not commercially available but could be readily prepared by the esterification of dicarboxylic acid 12 followed by nitration and hydrogenation. Condensation with nitriles provided thienopyrimidinones of general structure 15, which differ from the scaffold 11 series in the relative location of the sulfur atom and the absence of a methyl group on the thiophene ring. Installation of substituents at C4 and saponification were performed as described previously, to provide inhibitors with core scaffold 16.

Initial SAR analysis

All target thienopyrimidine carboxylic acids 11 and 16 (Figure 3) were assayed for inhibition of recombinantly-expressed and purified C. jejuni NCTC 11168 PglD acetyltransferase.¹⁸ The continuous assay monitors the release of CoASH using 5,5′-dithiobis-(2-nitrobenzoic acid,³⁶ and was carried out at 300 μM AcCoA (K_m of 295 μM), and 500 μM UDP-4-amino-sugar (K_m of 274 μM). IC₅₀ values of selected compounds are summarized in Table 1. The AcCoA concentration was aligned with reported physiologic concentrations in typical Gram-negative bacteria.⁴³

Figure 3.

Representative examples of thienopyrimidine-6-carboxylate inhibitors of C. jejuni PglD; Core 11: R₂ = a – z; R₄ = a – z; Core 16: R₂ = d, h; R₄ = a.

Table 1.

Experimental data obtained for selected inhibitors (Figures 3 and 4): IC₅₀ values determined with the continuous C. jejuni PglD inhibition assay, calculated ligand efficiency (LE), thermal stabilization of the PglD/ligand complex with the DSF assay, and cLogP values calculated with built-in functionality of the ChemDraw^® commercial software package. Compounds 51–54 (Figure 4): 2MPE: 2-methoxy-2-phenylethyl and 2PP: 2-phenylpropyl.

Compound #	Scaffold	R2	R4	IC50/(μM)	LE	ΔTm/(°C)	cLogP
17	11	d	a	2.2 ± 0.4	0.35	7.4	4.44
18	11	d	b	1.6	0.32	3.6	1.48
19	11	d	c	4.5	0.26	nd	5.89
20	11	d	d	1.4	0.27	13.2	6.54
21	11	d	g	0.54	0.27	nd	6.46
22	11	d	h	0.45	0.28	nd	6.46
23	11	d	i	0.48	0.28	13.7	6.46
24	11	d	j	0.42	0.29	nd	6.68
25	11	d	k	0.39	0.29	nd	6.68
26	11	d	l	0.48	0.29	13.2	6.68
27	11	d	m	0.72	0.28	nd	6.99
28	11	d	n	0.42	0.29	nd	6.99
29	11	d	o	0.25	0.30	14.3	6.99
30	11	d	p	0.39	0.30	12.0	5.04
31	11	d	q	0.34	0.30	12.0	5.04
32	11	d	r	0.42	0.30	12.5	5.04
33	11	s	a	2.4	0.29	nd	3.46
34	11	t	a	3.5	0.28	nd	3.46
35	11	u	a	9.0	0.26	nd	3.46
36	11	d	x	1.0	0.27	nd	6.42
37	11	d	y	0.37	0.26	nd	6.55
38	11	d	z	0.59	0.26	11.5	6.50
39	11	e	a	4.4	0.31	7.9	4.97
40	11	f	a	4.1	0.30	7.4	5.50
41	11	h	a	1.1 ± 0.2	0.33	8.4	4.36
42	11	k	a	2.3	0.33	8.9	4.58
43	11	q	a	0.46 ± 0.05	0.39	8.4	2.94
44	11	s	b	1.4	0.28	5.3	0.50
45	11	s	r	0.28	0.27	11.5	4.06
46	11	v	a	2.1	0.33	9.4	5.15
47	11	w	a	4.6	0.28	7.4	5.32
48	11	z	a	1.8	0.31	7.4	4.41
49	16	d	a	3.9	0.34	4.8	4.45
50	16	h	a	1.4	0.34	5.9	4.37
51	11	2PP(S)	2MPE(R)	2.0	0.24	9.4	6.50
52	11	2PP(S)	2MPE(S)	1.1	0.25	10.4	6.50
53	11	2PP(R)	2MPE(R)	0.87	0.27	nd	6.50
54	11	2PP(R)	2MPE(S)	0.42	0.27	9.9	6.50
55	11	s	2MPE(S)	0.27 ± 0.09	0.26	12.5	5.12

Our initial strategy was to explore the C2- and C4-substituents individually and then combine the optimal structural components from each series to capture advantageous synergistic binding effects. We observed that short aliphatic chains and saturated carbocycles at C2 (with R₄ = Me) provided weak inhibitors with IC₅₀ values in the 10 μM to 1 mM range (data not shown). Based on these studies we selected an unsubstituted phenylethyl moiety (Figure 3, substituent d) as a favorable C2-substituent in terms of potency gain versus added mass (e.g. 17, IC₅₀ = 2.2 μM, LE 0.35, Table 1). However, when we explored C4-substituents (with R₂ = H), these analogs proved to be poor inhibitors (IC₅₀ > 500 μM, data not shown) and SAR trends could not be well defined from the data. Accordingly, we elected to directly prepare disubstituted analogs, with R₂ = phenylethyl (d), and iteratively build up the R₄ substituent in terms of length, sterics and heteroatom substitution (e.g. inhibitors 18–35; shorter and smaller substituents omitted for brevity). In general, only modest gains in potency could be achieved, and all the gains were at the cost of poorer ligand efficiencies. Ultimately, we observed that a phenylethyl moiety was also well tolerated at the C4-position of the thienopyrimidine core (20, R₂ = R₄ = phenylethyl, IC₅₀ = 1.4 μM).

In order to gain a better understanding of the binding behaviour of the inhibitors, we examined the results of the phenyl scan on the R₄ substituent, which positioned moieties with different characteristics (sterics, polarity and hydrogen-bonding capabilities) ortho, meta or para on the aryl ring (inhibitors 21–35). However, while we had predicted that these substitutions would either complement or clash with the protein surface resulting in a measurable difference in IC₅₀ values, we instead observed a “flat” SAR with inhibitors exhibiting IC₅₀ values around 0.5 μM (e.g. 21–35). Indeed, only substitution with an acetamido group (33–35) elicited clear variances, with the ortho substitution being the most disfavored (35, IC₅₀ = 9.0 μM). A potential explanation for the flat SAR could be that if the aryl group occupies the AcCoA binding groove then one edge would face the protein surface and the other would be solvent exposed, thus precluding any interaction between the moieties on the R₄-aryl ring and the protein surface. This hypothesis was tested by preparing analogs bearing two moieties on the aryl ring (inhibitors 36–38), however, potency remained at 0.5 μM.

In light of these data, we developed two hypotheses. One potential hypothesis was that the compounds might have a different mode of action other than competitive inhibition with the AcCoA substrate. We explored this by applying several biochemical and biophysical approaches, which are presented in the next section (vide infra). The second possibility was that the inhibitor binding mode differed significantly from that predicted by the computational model; for example, either the ligands bind in a different location on the PglD protein surface or the enzyme itself undergoes a significant conformational change. We also noted that the C2, C4-disubstituted inhibitors were inherently quite symmetrical, with a pseudo C₂ axis of rotation in the plane of the thienopyrimidine core (Figure 4a). This feature might explain the flat SAR observed for the phenyl scan of the R₄-substituent since the effect of any moiety on the aryl group could potentially be masked if the ligand flipped 180° and positioned the R₂-substituent in the AcCoA binding groove instead. To test this hypothesis, we synthesized inhibitors 49 and 50, which included the quasi-regioisomeric thienopyrimidine core 16 and, if the binding orientation remains true, should position the sulfur atom towards the protein surface. Indeed, these inhibitors had similar potency to their parent analogs (compare 49 IC₅₀ = 3.9 μM vs. 17 IC₅₀ = 2.2 μM, and 50 IC₅₀ = 1.4 μM vs. 41 IC₅₀ = 1.1 μM), suggesting the thienopyrimidine carboxylic core can bind in either of two orientations into the AcCoA binding groove. To further validate this hypothesis, we developed inhibitors that would be locked in a fixed conformation by introducing rigidifying elements in the alkyl linkers between the thienopyrimidine core and substituent aryl groups. Synthetically it was straightforward to add a methyl group on the second carbon of the R₂-substituent and a methoxy group on the second carbon of the R₄-substituent; the prerequisite starting materials were commercially available as both stereoisomers and could be easily transformed into the corresponding nitrile or amine for the modular syntheses (Figure 4b). With the four rigidified analogs we would expect to see data consistent with a matched/mismatched phenomenon; either substituent should have an enantiomer that complements the protein surface better than the other and when combined these effects should result in one poor inhibitor, two moderate ones and a single preferred analog. Indeed, this was observed with compounds 51–54 (Figure 4b), albeit in a modest IC₅₀ range. These combined results strongly suggested that the inhibitors were capable of binding to the PglD active site with either the R₂-, or R₄-substituent in the AcCoA groove.

Figure 4.

(a) Illustration of the quasi C₂ symmetry axis of the disubstituted inhibitors (b) SAR of inhibitors 51 – 54 with rigidified R₂ and R₄ substituents; magenta symbols represent stereochemical preference of R₂ and R₄ substituents with respect to the PglD protein surface and highlight the mismatched/matched pattern.

Overall, the SAR data (Figure 4) shows that while binding may be somewhat promiscuous, the R₂-substituent was far more important for binding potency. This experimental observation was therefore inconsistent with our initial computational model, which had oriented this substituent over the β-helix (and quite solvent exposed) and placed the R₄-substituent in the AcCoA groove. These findings prompted us to investigate the binding mode in more detail, to provide a foundation for the next synthetic and biochemical efforts.

Biophysical characterization and inhibitor optimization

With the knowledge gained from SAR studies up to this point, we generated a potent C. jejuni PglD inhibitor, 55 (Figure 5), with an IC₅₀ of 270 nM; however, this was at the cost of decreased ligand efficiency (0.26) and a high cLogP (5.12). In order to further optimize the potency and physicochemical properties of these compounds, we needed to better characterize their binding characteristics. Towards this end, we employed several complementary approaches including the PglD inhibition assay with varying substrate concentrations, DSF analysis, dynamic light scattering (DLS) and X-ray crystallography.

Figure 5.

Structures of key inhibitors discussed in this section: 17, 43 and 55.

We had first initiated inhibitor development based on modeling studies, which suggested that the thienopyrimidine carboxylate core competed with the AcCoA substrate for binding to the PglD active site. To test this experimentally, we monitored inhibition in the presence of varying concentrations of the two substrates. We observed that significantly higher concentrations of the UDP-4-amino-sugar substrate had no measurable effect on the inhibition by a representative analog of 11 (compound 37, supporting information Table S2). In contrast, a 2-, and 5-fold increase in AcCoA substrate concentration resulted in measured IC₅₀ values of 0.63 μM and 1.4 μM, respectively. These values represent a 2- and 4-fold decrease in potency and strongly suggest that the thienopyrimidine carboxylate inhibitors compete with AcCoA for binding to the PglD active site.

All of the data thus far has been based on the enzymatic inhibition assay and, despite the judicious use of detergent and BSA to prevent aggregation phenomena,⁴⁴ may be prone to artefacts that obscure binding trends. We therefore employed DSF as a qualitative secondary assay to assess binding of inhibitors to the C. jejuni PglD. DSF employs a fluorescent dye, Sypro Orange, to report on the thermal denaturation of a protein or protein/ligand complex.⁴⁵ Potent inhibitors bind to their target enzyme more strongly that weak ones, resulting in a more stable protein/ligand complex and higher measured melting temperature (T_m). Therefore, the thermal stabilization (ΔT_m) of PglD can be used to provide an indirect measure of the relative binding affinity of the related family of compounds. We selected a subset of inhibitors with a range of IC₅₀ values and structural characteristics to analyze by DSF and plotted the ΔT_m vs IC₅₀ values (Figure 6). The graph shows a positive correlation (R² = 0.72) between inhibition potency and thermal stabilization over a range of IC₅₀ values indicating that the compounds stabilize the native PglD homotrimer and that the more potent compounds, based on IC₅₀ values, indeed stabilize PglD more effectively.

Figure 6.

Correlation between the in vitro potency against PglD and the T_m (calculated from the first derivative of the thermal melting curve) of the PglD/inhibitor complexes with core 11; thermal melt assays were performed with 10 μM PglD and 100 μM compound. Correlation coefficient provided for all data indicated by red diamond. Inhibitors with R₄ = b (ethylamine) (blue diamond, circled) appear as outliers and are omitted from the calculation of the correlation coefficient.

Next, we analyzed the enzyme quaternary structure in solution by dynamic light scattering (DLS) and found the calculated hydrodynamic radius of apo PglD to be ~34 Å, consistent with the crystal structures of trimeric PglD (supporting information Figure S1). In the presence of 10-fold excess of inhibitor 55, the measured radius was marginally larger, ~36 Å, yet still consistent with the known trimeric structure. We also noted that the polydispersity index (PDI) of the PglD/55 complex (24 % PDI) was lower than apo PglD (31% PDI), providing strong support that the inhibitor does not destabilize quaternary structure and indeed binds to the homotrimeric enzyme.

Throughout these efforts we also continued to pursue crystallization attempts to gain structural insight on representative thienopyrimidine analogs. X-ray analysis of the C. jejuni NCTC 11168 PglD enzyme has been successfully carried out in our group and others.^28,29 Typically, the reported precipitant solutions were either high ionic strength (1.9 M ammonium sulfate) or contained acidic buffer components that were frequently observed to bind in the enzyme active site (sulfate, phosphate, citrate).^28,29 Despite numerous attempts, application of previously published protocols did not yield co-crystal structures with the current thienopyrimidine-6-carboxylate inhibitors and thus we screened for new conditions that would be more compatible with the binding of small organic molecules. Ultimately, a crystallization hit with PEG-3350 was successfully optimized to reliably produce high quality crystals. The crystals develop as hollow, hexagonal rods that fill in from one end as they mature; producing either “vase-like” or solid crystals (Figure S3). The diffraction quality did not appear to differ between a solid or hollow end of such crystals. The crystals were tolerant of DMSO in concentrations up to 10%, but tended to be rather fragile and could break upon looping. Therefore, we employed an “in-drop soaking” protocol, in which a solution of the inhibitor is directly added to the drop containing the crystals.

Using these crystallization conditions, we were able to obtain high-quality diffraction data and successfully solve the co-crystal structure of the PglD/17 complex (Figure 7 and S4). From the structure, two significant observations were immediately evident. Firstly, as anticipated from the SAR studies, the thienopyrimidine core was flipped 180° compared to the computational model developed with the core fragment, with the R₂-subsituent binding in the AcCoA groove (Figure 7, panel b). This position of the thienopyrimidine core directs the R₄-subsituent away from the AcCoA groove and into solvent where it makes no significant interactions with the PglD protein surface (Figure S5, LigPlot⁴⁶). This observation is consistent with our SAR findings that R₄-subsituent makes small contributions to inhibitor potency at the expense of ligand efficiency (Table 1). Secondly, the most surprising observation was that the ligand was bound 3.9 Å further along the AcCoA groove, relocating the carboxylate distal to the acetyltransferase active site where the AcCoA thioester is normally accommodated. This observation was rather unexpected as we assumed that the Asn118, H134 and S136 residues would be preferred hydrogen-bonding partners for the inhibitor carboxylic acid moiety. Furthermore, this displacement precludes a T-shaped π-stacking between the side chain of Phe152 and the thienopyrimidine core; instead the binding mode positions Phe152 at an offset angle over the inhibitor carboxylate (Figure S4). To compensate, inhibitor 17 makes several other key interactions with the PglD surface; the carboxylic acid moiety is positioned at a favorable distance and orientation to form hydrogen bonds with the Ser136 side chain and the backbone amide NH of Ile155. Similarly, the thienopyrimidine N1 atom is well placed to interact with Gly173. From the structure it appears that a definitive interaction may be the placement of the R₂ phenyl ring within the AcCoA adenine binding sub-pocket, displacing solvent waters from a relatively hydrophobic protein cavity. The ligand binding mode is also internally favorable (low torsion) as the bound conformation adopts an anti-periplanar angle of approximately 160° along the benzylidene bond between the thienopyridimine core and R₂ phenyl ring (Figure 7c).

Figure 7.

Co-crystal structure of the PglD/17 complex (PDB 5T2Y). (a) Inhibitor 17 shown in sticks presentation (magenta) and overlayed with AcCoA substrate in lines (cyan, PDB 3BSY) and sugar substrate in sticks (yellow, PDB 3BSS). (b) overlay with the computational docking pose of compound 2 in line representation (orange). (c) Side-view highlighting the AcCoA adenine sub-pocket and interaction between 17 and residues Gly173, Ile155 and Ser136. White dashed lines indicate distances between 17 and PglD side-chain residues; yellow dashed line shows distance between the carboxylate computational docking pose and crystal structure. (d) Composite omit map 2F_o –F_c contoured at 1.0 σ generated with the built in functionality of the PHENIX software and image generated in PYMOL.

Armed with this new structural information we set out to further optimize the inhibitor series. First we investigated the possibility of simultaneously placing the R₂ phenyl ring in the adenine sub-pocket and the carboxylate in the active site hot spot by increasing the linker length between the inhibitor core and R₂-substituent. This did not prove successful as analogs 39 and 40 (with a 3- and 4-carbon linker respectively) demonstrated lower in vitro potency; presumably due to internal strain of unfavorable torsion angles of the alkyl linker. Secondly we noted that although the R₂ phenyl group fits well in the adenine sub-pocket, it does not fully occupy the volume of the [6,5]-fused ring cavity and there was unoccupied space at the meta position relative to the linker (Figure 7c). We explored several moieties at this position (41 –OMe, 42 –F, 46 –Cl and 47 –CF₃, Table 1), but only gained a minor improvement in potency with compound 41 (IC₅₀ = 1.1 μM). Finally upon closer inspection of the adenine sub-pocket we speculated that if the aromatic carbon at a position ortho to the linker is replaced by a nitrogen and protonated, it would be well-oriented to form a hydrogen bond with the backbone oxygen of Gly173, as it does with the exocyclic amine of the adenine of the AcCoA substrate (PDB ID 3BSY).²⁸ While the resultant pyridine nitrogen is only weakly basic, it could potentially be protonated in the microenvironment of the enzyme and lead to an order of magnitude improvement in potency. For example, the pyridine derivative risedronic acid is a ~300-fold more potent inhibitor of the human farnesyl pyrophosphate synthase than the corresponding phenyl analog.⁴⁷ Although not as significant, the 4.8-fold fold enhancement we observed between inhibitor 43 (IC₅₀ = 0.46 μM) and 17 (IC₅₀ = 2.2 μM) is very interesting considering the minor nature of the structural change. Although 43 is not the most potent inhibitor, it exhibits the best ligand efficiency, 0.39, and has a cLogP of 2.94, within the −0.5 – 3 range that is thought to be ideal to reach cytoplasmic targets of Gram-negative bacteria; with these characteristics 43 represents our best candidate to date with balanced physicochemical properties.

Discussion and Conclusions

In the development of inhibitors of the C. jejuni UDP-4-amino-sugar acetyltransferase, PglD, we initially screened both fragment-based and HTS libraries. These studies led to the identification of a thienopyrimidine-6-carboxylate core as a promising hit. Over the course of the SAR studies we found substitution at the C2-position of the core to be most impactful and that the binding mode appeared to be poorly predicted in silico. By combining data from substrate competition assays, DSF, and DLS we confirmed that the inhibitors bind exclusively to the AcCoA binding groove of the active site without disrupting the homo-trimeric structure of PglD.

Critically, crystallographic analysis reveals a 180° flip of the thienopyrimidine core to place the R₂-substituent in the AcCoA groove, as was anticipated from the SAR data. However, in contrast to previously characterized small fragments, we observed a 3.9 Å displacement of the inhibitor further along the AcCoA groove away from the known thioester binding site. Notably, in this new binding mode, the inhibitor makes several hydrogen-bonding interactions that compensate for the ones that would have been formed with the key catalytic residues (Asn118 & His134), but also precludes a stabilizing T-shaped π-interaction between the thienopyrimidine core and Phe152. Ultimately, it appears that placement of the R₂-phenyl into the adenine sub-pocket of AcCoA binding site may be the main driving force of ligand binding and indeed further optimization of this interaction resulted in to our most promising lead. Specifically, inhibitor 43 (IC₅₀ = 0.46 μM) shows an excellent ligand efficiency of 0.39 and a favorable clogP of 2.94. To the best of our knowledge this is the first competitive inhibitor with an IC₅₀ value < 1 μM for a bacterial amino-sugar-modifying enzyme. A selection of inhibitors from the series presented herein was also tested against the structurally homologous amino-sugar acetyltransferase from N. gonorrhoeae (PglC in that O-linked glycosylation pathway) and no significant inhibition was observed even at 100 μM. This observation was not surprising considering the significant differences between the AcCoA binding sites of the two enzymes.⁴⁸

With respect to biological analysis, the in vivo analysis of inhibitors that target bacterial virulence factors requires more complex experimental approaches than the simple estimation of minimal inhibitory concentrations (MICs) used in assessing bacterial viability with antibiotics that are targeted at cell survival processes. We are therefore pursuing multiple strategies to investigate in vivo efficacy both directly in C. jejuni, by assessing the impact of the inhibitors on glycoprotein biosynthesis, and in cell culture models of mammalian gut endothelia assessing C. jejuni adhesion and invasion. Currently, despite robust in vitro activity, we do not observe a reduction in glycosylation of periplasmic and cell surface proteins upon treatment of C. jejuni with the current inhibitors. This observation may be the result of insufficient inhibition potency against the target and/or may result from limited cellular uptake or efflux. In this context, C. jejuni features both a formidable cell wall including inner and outer cell membranes flanking a peptidoglycan barrier together with a dense external glycocalyx, as well as an active CmeABC efflux pump system to guard against external chemical agents.⁴⁹ With respect to potency, the X-ray analysis of PglD with compound 17 provides an excellent foundation for further structure-guided optimization of the thienopyrimidine core. Concerning cellular uptake and efflux, which are common impediments to antibiotic activity in Gram-negative pathogens, the presented thienopyrimidine core also offers considerable opportunities. In particular, we are exploring several strategies including bioisosteric replacement of the carboxylate moiety and development of pro-drug analogs in consideration of the fact that the negative charge of the carboxylic acid moiety might acting as an impediment for compound uptake into the cytoplasm, which is the site of PglD activity. Additionally, the C4-substituent of the thienopyrimidine series represents a valuable handle for modulating the cLogP of the inhibitor, since we have demonstrated that substitution at this site can be varied without adversely perturbing PglD inhibition. As elegantly demonstrated in a recent study of another Gram-negative pathogen, researchers have shown that fine-tuning the physical properties of inhibitors, in this case to mimic the zwitterionic charge distribution of amino acids, allowed novel carbapenem analogs to pass through the OccD1 membrane porin of P. aeruginosa.⁵⁰ Thus the thienopyrimidine compounds in this report are amenable to further such optimizations and current efforts are focused on overcoming the C. jejuni defenses to validate PglD as a potential antivirulence target in the biological context of infection.

Overall the studies presented herein support further development of small molecule inhibitors as tools for investigating of the role diNAcBac in bacterial glycosylation and how it relates to pathogenicity. Promising candidates targeting important virulence factors have already been documented, including agents inhibiting pilus assembly in uropathogenic E. coli⁵¹ and bacterial motility by inhibiting disulfide formation of periplasmic proteins.⁵² Progress has also been made towards modulators of quorum sensing.⁵³ Closely related to the work in this paper, the Sulea and Logan research groups have reported an inhibitor that targets the pseudaminic acid biosynthetic pathway found in H. pylori and C. jejuni.⁵⁴ The “sialic-acid like” nonulosonate sugar product is required for the glycosylation of structural flagellin proteins and exposure to inhibitors attenuated flagellin production at concentrations of ≥100 μM. The flagella are required for persistent infections of H. pylori and this work demonstrates the link between bacterial glycosylation and pathogenicity using chemical tools (as opposed to genetic modification). The most advanced antivirulence work has been accomplished with N-phenyl-4-(3-phenylthioureido)benzenesulfonamide, 56 (LED209), a prodrug agent that covalently inhibits QseC, a receptor involved in signaling between many Gram-negative pathogens and the infected host.⁵⁵ Treatment with 56 in a mouse infection model showed prophylactic efficacy against S. typhimurium and provided protection against F. tularensis both pre- and post-infection. Furthermore, 56 was not toxic at therapeutic doses and did not affect pathogen growth in vitro. These promising in vivo results are compelling evidence that killing is not necessarily required for treatment of bacterial infection and alternate therapeutic intervention strategies are worth exploring.

In conclusion, herein we disclose the hit-to-lead development of a new series of thienopyrimidine-6-carboxylate inhibitors of the C. jejuni PglD UDP-amino-sugar acetyltransferase enzyme, which, upon further optimization could represent valuable chemical tools for unraveling the intricate biological role(s) of highly modified sugars in bacteria and potentially may lead to new agents with targeted antivirulence activity.

Experimental section

General Procedures for Characterization of Compounds

All intermediate compounds were purified by normal phase flash column chromatography on silica gel using a CombiFlash instrument. The homogeneity of all final compounds was confirmed to ≥95% by analytical reverse-phase LC-MS using an Agilent Series 1100 HPLC instrument equipped with a YMC AQ12S03-1003WT C18 column and a Finnigan LCQ Deca electrospray ionization mass spectrometer, using a gradient of 5–95% in 20 min of acetonitrile in water with 0.1% TFA. Each final compound was fully characterized by ¹H and ¹³C NMR and HRMS. Chemical shifts (δ) are reported in ppm relative to the internal deuterated solvent (¹H, ¹³C), unless indicated otherwise. The high-resolution MS spectra of final products were recorded using direct analysis in real time (DART) ionization on a Bruker Daltonics APEXIV 4.7 Tesla Fourier Transform Ion Cyclotron Resonance Mass Spectrometer (FT-ICR-MS).

General procedure for the formation of thienopyrimidinones – Parent structure 6

A pressure vessel washed charged with aminothiophene (1 equivalent) and nitrile (1–1.2 equivalents) and 4M HCl in dioxane was added via syringe. The pressure vessel was tightly sealed with a Teflon screw cap stopper and the reaction was stirred at 80°C until the reaction was complete, as judged by TLC. The reaction mixture was cooled to room temperature, transferred to an Erlenmeyer flask in an ice-bath with a minimum volume of methanol and diluted with water (10-fold of solvent volume). The pH was brought to ~8 with ammonium hydroxide solution (25%) resulting in precipitate formation. The mixture was held on ice for 15–60 min and then filtered over filter paper; washing with distilled water. The residue was dried under vacuum to afford the desired thienopyrimidine in sufficient purity for the subsequent step. Due to the poor solubility of these compounds, full characterization was performed after the next synthetic step and purification.

Ethyl 2-((diethoxyphosphoryl)methyl)-5-methyl-4-oxo-3,4-dihydrothieno[2,3-d]pyrimidine-6-carboxylate (8)

A 100 mL RBF was charged with ethyl 2-(chloromethyl)-5-methyl-4-oxo-3,4-dihydrothieno[2,3-d]pyrimidine-6-carboxylate (900 mg, 3.13 mmol) (6) and triethylphosphite (10.4 g, 10.8 mL, 62.8 mmol) was added via syringe and the reaction was heated at 145 °C for 2 h. The reaction mixture was cooled down, diluted with 100 mL diethyl ether and stirred for 1 h on ice. The suspension was filtered, washed with copious amounts of ether and petroleum ether. The obtained residue was dried on high vacuum to furnish the desired the product as a fine, free-flowing white powder, 994 mg (82%). ¹H NMR (500 MHz, DMSO) δ 12.62 (s, NH), 4.28 (q, J = 7.1 Hz, 2H), 4.10 – 3.99 (m, 4H), 3.37 (d, J = 22.2 Hz, 2H), 2.79 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H), 1.21 (t, J = 7.0 Hz, 6H). ¹³C NMR (126 MHz, DMSO) δ 166.99 (d, J = 3.0 Hz), 162.98, 159.97, 154.77 (d, J = 8.6 Hz), 144.46, 122.93, 122.17, 63.42 (d, J = 6.3 Hz), 62.28, 33.99 (d, J = 131.2 Hz), 17.30 (d, J = 6.0 Hz), 15.95, 15.29. ³¹P NMR (121 MHz, DMSO) δ 21.87. MS (ESI): calcd. 389.09; found 389.07 [M+H]⁺.

General procedure for the Horner-Wadsworth-Emmons reactions

To an ice-cooled solution of phosphonate 8 (1 equivalent) and aldehyde (3 equivalents) in anhydrous THF was added sodium hydride (60% suspension in mineral oil, 3 equivalents) in small portions. The reaction was stirred at RT until complete consumption of starting material, as determined by TLC. The crude product was poured into ice-water (5 times the reaction solvent volume), filtered and washed with water. The filter-cake was washed with ice-cold ethanol, diethyl ether and pentanes. The residue was dried on high vacuum to furnish the desired product in sufficient purity for subsequent reactions.

General procedure for the hydrogenation reactions

A round bottom flask was charged with the alkene product from the HWE reaction, suspended in MeOH/THF/NH₄OH 10:2:1 and purged with nitrogen. Pearlman’s catalyst was added (20% Pd(OH)₂/C, 0.5 equivalents, CAS 12135-22-7), and the reaction mixture was purged with H₂ gas and equipped with an H₂-filled balloon. The reaction was stirred at 60°C until complete consumption of starting material, as determined by TLC, replenishing the H₂ balloon as needed. The reaction mixture was purged with nitrogen and filtered through a plug of Celite, rinsing with EtOAc. The filtrate was concentrated in vacuo and the obtained material was used without further purification.

General procedure for the formation of ester precursors (10)

To a solution of or thienopyrimidinone (1 equivalent) and (benzotriazol-1-yloxy)-tripyrrolidinophosphonium hexafluorophosphate (PyBOP, 1.3 equivalents) in anhydrous acetonitrile was added 1,8-diazabicyclo[5.4.0]undec-7-ene (1.5 equivalents, 3 equivalents in case of amine-HCl salts) via syringe. The mixture was stirred at RT for 5 min, after which the nucleophile (1.5 equivalents) was added and the reaction stirred at RT until complete consumption of starting material, as determined by TLC. The solvent was removed in vacuo and the crude residue was purified by column chromatography on silica gel using a gradient of EtOAc in 1:1 Hex/CH₂Cl₂. Note: Substituent b was introduced as a mono-Boc-protected diamine; the Boc protecting group was expediently removed by treating the compound with 4 M HCl in 1,4-dioxane followed by precipitation with cold diethyl ether and filtration.

Ethyl 5-methyl-4-(methylamino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,a)

Isolated 80 mg (77% yield) as a white solid. ¹H NMR (500 MHz, DMSO) δ 7.27 – 7.19 (m, 4H), 7.17 – 7.12 (m, 1H), 7.09 (m, NH), 4.28 (q, J = 7.1 Hz, 2H), 3.09 – 3.04 (m, 2H), 3.02 – 2.97 (m, 5H), 2.86 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.26, 167.97, 163.50, 159.84, 142.72, 141.08, 129.47, 129.39, 126.92, 120.08, 115.95, 62.16, 41.31, 34.42, 29.27, 16.74, 15.34. MS (ESI): calcd. 356.14; found 356.27 [M+H]⁺.

Ethyl 4-((2-aminoethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,b)

Isolated 70 mg (92% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 8.00 (br. s., NH₂), 7.48 (m, NH), 7.29 – 7.24 (m, 4H), 7.19 – 7.14 (m, 1H), 4.30 (q, J = 7.1 Hz, 2H), 3.81 (q, J = 5.6 Hz, 2H), 3.12 – 3.03 (m, 6H), 2.95 (s, 3H), 1.30 (t, J = 7.1 Hz, 3H).¹³C NMR (75 MHz, DMSO) δ 165.02, 162.48, 159.04, 141.38, 140.89, 129.04, 128.98, 126.75, 121.53, 116.43, 62.15, 38.60, 38.41, 33.27, 16.21, 14.79. MS (ESI): calcd. 385.170; found 385.20 [M+H]⁺.

Ethyl 4-(benzylamino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,c)

Isolated 80 mg (79% yield) as an off-white solid. ¹H NMR (500 MHz, DMSO) δ 7.70 (m, NH₂), 7.40 (d, J = 7.7 Hz, 2H), 7.29 (t, J = 7.6 Hz, 2H), 7.23 – 7.15 (m, 3H), 7.09 (m, 3H), 4.75 (d, J = 5.8 Hz, 2H), 4.27 (q, J = 7.1 Hz, 2H), 2.93 (m, 7H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.26, 168.09, 163.46, 159.24, 142.59, 141.04, 140.88, 129.38, 129.33, 128.52, 127.77, 126.85, 120.39, 115.86, 62.19, 45.07, 41.25, 34.27, 16.88, 15.31. MS (ESI): calcd. 432.17; found 432.33 [M+H]⁺.

Ethyl 5-methyl-2-phenethyl-4-(phenethylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(d,d)

Isolated 106 mg (91% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 7.32 – 7.16 (m, 8H & NH), 7.16 – 7.07 (m, 2H), 4.26 (q, J = 7.0 Hz, 2H), 3.72 (dd, J = 14.0, 6.3 Hz, 2H), 3.13 – 3.05 (m, 2H), 3.05 – 2.96 (m, 2H), 2.94 – 2.86 (m, 2H), 2.79 (s, 3H), 1.27 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.22, 168.19, 163.44, 159.23, 142.67, 140.85, 140.67, 129.82, 129.55, 129.40, 129.37, 127.29, 126.93, 120.27, 115.77, 62.14, 43.49, 41.29, 35.77, 34.47, 16.60, 15.30. MS (ESI): calcd. 446.19; found 446.40 [M+H]⁺.

Ethyl 4-((4-methoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,g)

Isolated 172 mg (83% yield) as an off-white solid. ¹H NMR (500 MHz, DMSO) δ 7.26 – 7.20 (m, 4H), 7.18 – 7.09 (m, 3H & NH), 6.86 (d, J = 8.6 Hz, 2H), 4.28 (q, J = 7.1 Hz, 2H), 3.71 – 3.66 (m, 5H), 3.12 – 3.07 (m, 2H), 3.03 – 2.99 (m, 2H), 2.86 – 2.80 (m, 5H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.07, 163.30, 158.99, 158.70, 142.08, 138.32, 130.92, 130.03, 128.70, 128.55, 126.07, 121.47, 115.33, 114.46, 61.45, 55.53, 42.47, 41.16, 34.55, 34.46, 15.70, 14.58. MS (ESI): calcd. 476.20; found 476.40 [M+H]⁺.

Ethyl 4-((3-methoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,h)

Isolated 182 mg (87% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 7.27 – 7.19 (m, 6H), 7.17 – 7.11 (m, 1H 7 NH), 6.82 (m, 2H), 4.28 (q, J = 7.1 Hz, 2H), 3.73 (m, 2H), 3.69 (s, 3H), 3.10 (m, 2H), 3.01 (m, 2H), 2.91 – 2.86 (m, 2H), 2.82 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.05, 163.28, 160.23, 158.96, 142.08, 140.67, 138.35, 130.10, 128.70, 128.56, 126.08, 121.47, 121.36, 115.34, 114.92, 112.13, 61.45, 55.39, 42.27, 41.18, 35.41, 34.56, 15.66, 14.58. MS (ESI): calcd. 476.20; found 476.47 [M+H]⁺.

Ethyl 4-((2-methoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,i)

Isolated 198 mg (95% yield) as a beige solid. ¹H NMR (500 MHz, DMSO) δ 7.25 – 7.11 (m, 7H), 7.08 (m, NH), 6.95 (d, J = 7.6 Hz, 1H), 6.86 (td, J = 7.4, 1.0 Hz, 1H), 4.28 (q, J = 7.1 Hz, 2H), 3.74 (s, 3H), 3.73 – 3.68 (m, 2H), 3.09 (m, 2H), 3.00 (m, 2H), 2.92 (t, J = 7.3 Hz, 2H), 2.81 (s, 3H), 1.30 – 1.26 (m, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.04, 163.38, 159.12, 157.72, 142.17, 138.59, 130.92, 128.72, 128.54, 128.39, 127.51, 126.04, 121.11, 115.32, 110.79, 61.40, 55.61, 41.84, 41.19, 34.53, 29.65, 15.62, 14.59. MS (ESI): calcd. 476.20; found 476.27 [M+H]⁺.

Ethyl 4-((4-fluorophenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,j)

Isolated 183 mg (90% yield) as a pale yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 7.32 – 7.23 (m, 4H), 7.21 – 7.12 (m, 3H), 7.02 (m, 2H), 5.61 (t, J = 5.5 Hz, NH), 4.34 (q, J = 7.1 Hz, 2H), 3.85 (dd, J = 12.5, 6.8 Hz, 2H), 3.22 – 3.14 (m, 4H), 2.94 (m, 2H), 2.71 (s, 3H), 1.38 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.02, 163.25, 161.95 (d, J = 245.0 Hz), 158.92, 142.03, 138.10, 134.79 (d, J = 3.3 Hz), 130.55, 130.44, 128.67, 128.57, 126.09, 121.68, 115.82 (d, J = 21.2 Hz), 115.30, 61.49, 42.44, 41.14, 34.63, 34.54, 15.71, 14.57. MS (ESI): calcd. 464.18; found 464.47 [M+H]⁺.

Ethyl 4-((3-fluorophenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,k)

Isolated 168 mg (83% yield) as a pale yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 7.34 – 7.23 (m, 5H), 7.18 (m, 1H), 7.05 – 6.89 (m, 3H), 5.63 (t, J = 5.4 Hz, NH), 4.34 (q, J = 7.1 Hz, 2H), 3.87 (dd, J = 12.5, 6.8 Hz, 2H), 3.23 – 3.15 (m, 4H), 2.96 (m, 2H), 2.72 (s, 3H), 1.38 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.01, 163.27 (d, J = 246.3 Hz), 163.23, 158.88, 142.02, 141.77 (d, J = 7.1 Hz), 138.14, 130.49 (d, J = 8.3 Hz), 128.68, 128.57, 126.09, 124.78, 124.75, 121.68, 115.90 (d, J = 20.9 Hz), 115.31, 113.88 (d, J = 21.0 Hz), 61.48, 42.17, 41.16, 35.21, 34.56, 15.69, 14.57. MS (ESI): calcd. 464.18; found 464.47 [M+H]⁺.

Ethyl 4-((2-fluorophenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,l)

Isolated 170 mg (84% yield) as an off-white solid. ¹H NMR (300 MHz, CDCl₃) δ 7.32 – 7.02 (m, 9H), 5.69 (t, J = 5.4 Hz, NH), 4.35 (q, J = 7.1 Hz, 2H), 3.89 (dd, J = 12.4, 6.6 Hz, 2H), 3.23 – 3.12 (m, 4H), 3.04 (m, 2H), 2.76 (s, 3H), 1.39 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.00, 163.30, 161.51 (d, J = 244.8 Hz), 158.99, 142.10, 138.29, 131.46 (d, J = 4.9 Hz), 128.86, 128.75, 128.70, 128.55, 126.08 (d, J = 16.0 Hz), 126.06, 124.60 (d, J = 3.5 Hz), 121.55, 115.75 (d, J = 22.0 Hz), 115.31, 61.46, 41.41, 41.16, 34.52, 29.03, 15.68, 14.58. MS (ESI): calcd. 464.18; found 464.47 [M+H]⁺.

Ethyl 5-methyl-4-((4-methylphenethyl)amino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,m)

Isolated 103 mg (77% yield) as a white solid. ¹H NMR (300 MHz, CDCl₃) δ 7.34 – 7.10 (m, 9H), 5.61 (t, J = 5.3 Hz, NH), 4.35 (q, J = 7.1 Hz, 2H), 3.88 (dd, J = 12.2, 6.6 Hz, 2H), 3.24 – 3.13 (m, 4H), 2.95 (t, J = 6.7 Hz, 2H), 2.69 (s, 3H), 2.36 (s, 3H), 1.40 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.09, 163.33, 159.00, 142.11, 138.36, 136.59, 135.89, 129.76, 128.96, 128.71, 128.56, 126.07, 121.45, 115.34, 61.44, 42.41, 41.19, 34.93, 34.57, 21.32, 15.68, 14.59. MS (ESI): calcd. 460.21; found 460.47 [M+H]⁺.

Ethyl 5-methyl-4-((3-methylphenethyl)amino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,n)

Isolated 113 mg (84% yield) as an off-white solid. ¹H NMR (300 MHz, CDCl₃) δ 7.34 – 7.16 (m, 6H), 7.07 (m, 3H), 5.61 (t, J = 5.3 Hz, NH), 4.35 (q, J = 7.1 Hz, 2H), 3.89 (dd, J = 12.2, 6.6 Hz, 2H), 3.19 (m, 4H), 2.95 (t, J = 6.7 Hz, 2H), 2.69 (s, 3H), 2.36 (s, 3H), 1.39 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.09, 163.31, 159.00, 142.11, 138.97, 138.74, 138.33, 129.89, 129.03, 128.71, 128.56, 127.79, 126.09, 126.07, 121.46, 115.36, 61.45, 42.36, 41.20, 35.29, 34.58, 21.66, 15.61, 14.58. MS (ESI): calcd. 460.21; found 460.50 [M+H]⁺.

Ethyl 5-methyl-4-((2-methylphenethyl)amino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,o)

Isolated 101 mg (75% yield) as a pale yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 7.33 – 7.14 (m, 9H), 5.68 (t, J = 5.5 Hz, NH), 4.36 (q, J = 7.1 Hz, 2H), 3.87 (dd, J = 12.8, 7.0 Hz, 2H), 3.19 (m, 4H), 3.01 (t, J = 7.1 Hz, 2H), 2.76 (s, 3H), 2.41 (s, 3H), 1.40 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.10, 163.32, 159.09, 142.07, 138.29, 137.10, 136.71, 130.94, 129.79, 128.70, 128.57, 127.16, 126.48, 126.09, 121.56, 115.35, 61.47, 41.30, 41.13, 34.69, 32.99, 19.71, 15.79, 14.59. MS (ESI): calcd. 460.21; found 460.47 [M+H]⁺.

Ethyl 5-methyl-2-phenethyl-4-((2-(pyridin-4-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylate 10(d,p)

Isolated 70 mg (72% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 8.45 (d, J = 5.9 Hz, 2H), 7.25 (d, J = 6.0 Hz, 2H), 7.22 (m, 4H), 7.18 (br. t, J = 5.8 Hz, NH), 7.13 (m, 1H), 4.28 (q, J = 7.1 Hz, 2H), 3.77 (dd, J = 13.5, 6.5 Hz, 2H), 3.10 – 3.05 (m, 2H), 3.03 – 2.98 (m, 2H), 2.92 (dd, J = 10.3, 4.0 Hz, 2H), 2.81 (s, 3H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.27, 159.30, 150.69, 149.70, 140.91, 129.45, 127.00, 125.50, 62.28, 42.30, 35.00, 34.48, 16.68, 15.37. MS (ESI): calcd. 447.185; found 447.20 [M+H]⁺.

Ethyl 5-methyl-2-phenethyl-4-((2-(pyridin-3-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylate 10(d,q)

Isolated 54 mg (41% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 8.46 (s, 1H), 8.41 (d, J = 4.5 Hz, 1H), 7.64 (d, J = 7.8 Hz, 1H), 7.31 (dd, J = 7.7, 4.8 Hz, 1H), 7.25 – 7.21 (m, 4H), 7.19 – 7.12 (m, NH, 1H), 4.28 (q, J = 7.1 Hz, 2H), 3.76 (dd, J = 13.1, 6.7 Hz, 2H), 3.10 – 3.05 (m, 2H), 3.00 (dd, J = 8.4, 6.2 Hz, 2H), 2.93 (t, J = 7.1 Hz, 2H), 2.81 (s, 3H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.19, 163.47, 159.26, 151.00, 148.56, 142.66, 140.86, 137.49, 136.20, 129.41, 126.95, 124.63, 120.37, 115.78, 62.21, 42.92, 41.25, 34.48, 32.87, 16.65, 15.32. MS (ESI): calcd. 447.185; found 447.27 [M+H]⁺.

Ethyl 5-methyl-2-phenethyl-4-((2-(pyridin-2-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylate 10(d,r)

Isolated 118 mg (90% yield) as a pale yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 8.52 (dd, J = 5.8, 1.7 Hz, 1H), 7.62 (td, J = 7.7, 1.8 Hz, 1H), 7.49 (s, NH), 7.31 – 7.23 (m, 4H), 7.21 – 7.13 (m, 3H), 4.34 (q, J = 7.1 Hz, 2H), 4.02 (dd, J = 11.5, 5.2 Hz, 2H), 3.19 – 3.06 (m, 6H), 2.97 (s, 3H), 1.38 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.08, 168.01, 163.46, 160.33, 158.80, 149.13, 142.21, 139.20, 137.12, 128.71, 128.51, 126.00, 123.76, 122.02, 120.96, 115.53, 61.35, 41.19, 40.27, 36.09, 34.56, 15.79, 14.60. MS (ESI): calcd. 447.19; found 447.40 [M+H]⁺.

Ethyl 4-((2,3-dihydro-1H-inden-2-yl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,x)

Isolated 105 mg (79% yield) as a pale yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 7.33 – 7.17 (m, 9H), 5.82 (d, J = 6.7 Hz, NH), 5.13 (m, 1H), 4.36 (q, J = 7.1 Hz, 2H), 3.52 (d, J = 7.3 Hz, 2H), 3.46 (d, J = 7.2 Hz, 2H), 2.95 (d, J = 5.4 Hz, 2H), 2.89 (d, J = 5.5 Hz, 2H), 2.84 (s, 3H), 1.40 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.03, 163.28, 158.79, 142.05, 141.03, 138.10, O128.71, 128.57, 127.14, 126.08, 125.10, 121.73, 115.31, 61.49, 52.48, 41.12, 40.45, 34.57, 15.88, 14.58. MS (ESI): calcd. 458.19; found 458.47 [M+H]⁺.

Ethyl 4-((3,5-dimethoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,y)

Isolated 77 mg (52% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 7.23 (m, 4H), 7.16 – 7.09 (m, 1H, NH), 6.41 (s, 2H), 6.33 (s, 1H), 4.28 (q, J = 7.1 Hz, 2H), 3.73 (dd, J = 13.7, 6.3 Hz, 2H), 3.67 (s, 6H), 3.10 (m, 2H), 3.04 – 2.98 (m, 2H), 2.88 – 2.79 (m, 5H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.21, 168.18, 163.44, 161.61, 159.23, 142.96, 142.67, 140.85, 129.39, 129.37, 126.94, 120.26, 115.78, 107.80, 99.12, 62.15, 56.10, 43.39, 41.34, 36.05, 34.48, 16.60, 15.29. MS (ESI): calcd. 506.21; found 506.53 [M+H]⁺

Ethyl 4-((2-(benzo[d][1,3]dioxol-5-yl)ethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylate 10(d,z)

Isolated 94 mg (66% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 7.27 – 7.20 (m, 4H), 7.10 (s, 1H, NH), 6.85 – 6.79 (m, 2H), 6.68 (d, J = 7.9 Hz, 1H), 5.95 (s, 2H), 4.28 (q, J = 7.1 Hz, 2H), 3.69 (dd, J = 14.1, 6.3 Hz, 2H), 3.12 – 3.05 (m, 2H), 3.05 – 2.95 (m, 2H), 2.85 – 2.78 (m, 5H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.19, 163.45, 159.23, 148.42, 146.70, 142.67, 140.87, 134.41, 129.40, 129.38, 126.94, 122.67, 120.25, 115.76, 110.12, 109.30, 101.84, 62.15, 43.63, 41.32, 35.45, 34.49, 16.62, 15.31. MS (ESI): calcd. 490.18; found 490.47 [M+H]⁺.

Ethyl 5-methyl-4-(methylamino)-2-(3-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylate 10(e,a)

Isolated 52 mg (33% yield) as a yellow oil. ¹H NMR (300 MHz, CDCl₃) δ 7.26 – 7.14 (m, 5H), 5.67 (m, NH), 4.33 (q, J = 7.1 Hz, 2H), 3.12 (d, J = 4.8 Hz, 3H), 2.88 (s, 3H), 2.84 (m, 2H), 2.75 – 2.67 (m, 2H), 2.22 – 2.11 (m, 2H), 1.37 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.72, 167.83, 163.32, 159.61, 142.63, 138.32, 128.75, 128.45, 125.88, 121.25, 115.33, 61.42, 39.21, 35.91, 30.26, 28.33, 15.93, 14.56. MS (ESI): calcd. 370.159; found 370.33 [M+H]⁺

Ethyl 5-methyl-4-(methylamino)-2-(4-phenylbutyl)thieno[2,3-d]pyrimidine-6-carboxylate 10(f,a)

Isolated 88 mg (43 % yield) as a pale yellow syrup. ¹H NMR (300 MHz, CDCl₃) δ 7.29 – 7.13 (m, 5H), 5.64 (br. m, NH), 4.34 (q, J = 7.1 Hz, 2H), 3.11 (d, J = 4.8 Hz, 3H), 2.89 (s, 3H), 2.88 – 2.81 (m, 2H), 2.70 – 2.63 (m, 2H), 1.95 – 1.83 (m, 2H), 1.78 – 1.67 (m, 2H), 1.38 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 168.96, 167.88, 163.35, 159.64, 142.89, 138.27, 128.65, 128.43, 125.80, 121.27, 115.32, 61.42, 39.51, 36.06, 31.51, 28.30, 15.95, 14.57. MS (ESI): calcd. 384.175; found 384.27 [M+H]⁺.

Ethyl 2-(3-methoxyphenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(h,a)

Isolated 163 mg (83% yield) as a pale yellow solid. ¹H NMR (300 MHz, CDCl₃) δ 7.16 (t, J = 7.8 Hz, 1H), 6.84 (m, 2H), 6.70 (dd, J = 8.1, 2.5 Hz, 1H), 5.66 (br. m, NH), 4.33 (q, J = 7.1 Hz, 2H), 3.75 (s, 3H), 3.15 – 3.08 (m, 7H), 2.86 (s, 3H), 1.37 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 167.96, 163.29, 159.77, 159.61, 143.75, 138.29, 129.46, 121.37, 121.09, 115.41, 114.25, 111.53, 61.44, 55.32, 41.06, 34.48, 28.32, 15.90, 14.56. MS (ESI): calcd. 386.154; found 386.27 [M+H]⁺.

Ethyl 2-(3-fluorophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(k,a)

Isolated 141 mg (54% yield) as a pale yellow oil. ¹H NMR (300 MHz, CDCl₃) δ 7.21 – 7.13 (m, 1H), 7.05 – 6.92 (m, 2H), 6.84 (td, J = 8.5, 2.6 Hz, 1H), 5.66 (br. m, NH), 4.33 (q, J = 7.1 Hz, 2H), 3.17 – 3.06 (m, 7H), 2.88 (s, 3H), 1.37 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 167.80, 167.58, 164.66, 163.28, 161.41, 159.61, 144.70 (d, J = 7.2 Hz), 138.21, 129.85 (d, J = 8.4 Hz), 124.34 (d, J = 2.6 Hz), 121.51, 115.42 (d, J = 3.6 Hz), 112.84 (d, J = 21.0 Hz), 61.46, 40.67, 34.02, 28.32, 15.92, 14.54. MS (ESI): calcd. 374.134; found 374.20 [M+H]⁺.

Ethyl 5-methyl-4-(methylamino)-2-(2-(pyridin-2-yl)ethyl)thieno[2,3-d]pyrimidine-6-carboxylate 10(q,a)

Isolated 585 mg (57% yield) as a white solid. ¹H NMR (300 MHz, CDCl₃) δ 8.54 (ddd, J = 5.0, 1.7, 0.8 Hz, 1H), 7.65 (td, J = 7.7, 1.8 Hz, 1H), 7.29 (d, J = 7.9 Hz, 1H), 7.17 (ddd, J = 7.5, 5.0, 1.1 Hz, 1H), 5.68 (br. m, NH), 4.34 (q, J = 7.1 Hz, 2H), 3.45 – 3.35 (m, 2H), 3.29 (m, 2H), 3.08 (d, J = 4.7 Hz, 3H), 2.91 (s, 3H), 1.38 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 167.41, 163.29, 161.18, 159.61, 148.18, 138.21, 137.73, 123.64, 121.71, 121.59, 115.48, 61.47, 38.68, 35.79, 28.39, 15.96, 14.55. MS (ESI): calcd. 357.139; found 357.20 [M+H]⁺.

Ethyl 2-(4-acetamidophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(s,a)

Isolated 38 mg (68% yield) as a beige solid. ¹H NMR (500 MHz, DMSO) δ 9.83 (s, NHAc), 7.41 (d, J = 8.4 Hz, 2H), 7.13 (d, J = 8.2 Hz, 2H), 7.09 (m, NH), 4.27 (q, J = 7.1 Hz, 2H), 3.02 – 2.94 (m, 7H), 2.86 (s, 3H), 1.98 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, CDCl₃ + CD₃OD) δ 174.50, 172.26, 171.43, 167.85, 164.01, 143.40, 142.04, 140.74, 133.30, 125.53, 124.67, 119.90, 65.95, 45.18, 38.26, 32.49, 28.11, 19.95, 18.70. MS (ESI): calcd. 413.16; found 413.33 [M+H]⁺.

Ethyl 2-(4-acetamidophenethyl)-4-((2-aminoethyl)amino)-5-methylthieno[2,3-d]pyrimidine-6-carboxylate 10(s,b)

Isolated 51 mg (83 % yield) as a white solid. ¹H NMR (300 MHz, DMSO) δ 9.81 (s, NHAc), 7.43 (d, J = 8.4 Hz, 2H), 7.14 (d, J = 8.4 Hz, 2H), 7.07 – 6.96 (m, NH), 4.26 (q, J = 7.0 Hz, 2H), 3.57 (dd, J = 11.0, 5.5 Hz, 2H), 3.23 (dd, J = 11.6, 6.1 Hz, 2H), 3.03 – 2.90 (m, 4H), 2.86 (s, 3H), 1.99 (s, 3H), 1.27 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, DMSO) δ 167.69, 162.98, 159.06, 156.71, 140.62, 137.83, 136.80, 129.09, 119.63, 115.35, 78.38, 61.67, 33.38, 24.59, 21.41, 16.11, 14.81. MS (ESI): calcd. 442.191; found 442.27 [M+H]⁺.

Ethyl 2-(4-acetamidophenethyl)-5-methyl-4-((2-(pyridin-4-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylate 10(s,p)

Isolated 86 mg (68% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 9.83 (s, NHAc), 8.46 (d, J = 4.7 Hz, 2H), 7.45 (d, J = 8.2 Hz, 2H), 7.24 (m, 2H, NH), 7.12 (d, J = 8.2 Hz, 2H), 4.26 (q, J = 7.0 Hz, 2H), 3.76 (dd, J = 12.7, 6.4 Hz, 2H), 3.07 – 2.90 (m, 6H), 2.79 (s, 3H), 2.00 (s, 3H), 1.28 (t, J = 7.0 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.24, 168.18, 163.44, 159.18, 150.63, 149.67, 140.79, 138.35, 137.19, 129.51, 129.43, 129.38, 125.41, 120.15, 115.75, 62.17, 46.92, 41.32, 34.98, 33.91, 25.05, 16.61, 15.28. MS (ESI): calcd. 504.207; found 504.13 [M+H]⁺.

Ethyl 2-(4-acetamidophenethyl)-5-methyl-4-((2-(pyridin-3-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylate 10(s,q)

Isolated 79 mg (63% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 9.83 (s, NHAc), 8.41 (d, J = 4.7 Hz, 1H), 7.63 (d, J = 7.7 Hz, 1H), 7.46 (m, J = 15.7, 8.4 Hz, 3H), 7.30 (dd, J = 7.7, 4.8 Hz, 1H), 7.18 – 7.10 (m, 2H, NH), 4.27 (q, J = 7.0 Hz, 2H), 3.76 (dd, J = 12.9, 6.6 Hz, 2H), 3.06 – 2.91 (m, 6H), 2.80 (s, 3H), 2.00 (s, 3H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.20, 163.45, 159.23, 151.01, 148.58, 140.83, 137.41, 137.19, 136.19, 129.62, 129.52, 124.60, 120.22, 120.13, 115.75, 62.18, 47.02, 33.93, 32.88, 27.10, 25.06, 16.63, 15.30. MS (ESI): calcd. 504.207; found 504.27 [M+H]⁺.

Ethyl 2-(4-acetamidophenethyl)-5-methyl-4-((2-(pyridin-2-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylate 10(s,r)

Isolated 55 mg (44% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 9.83 (s, NHAc), 8.52 (d, J = 4.1 Hz, 1H), 7.71 (t, J = 7.6 Hz, 1H), 7.43 (m, 2H, NH), 7.28 (d, J = 7.4 Hz, 1H), 7.24 (dd, J = 7.4, 4.9 Hz, 1H), 7.14 (d, J = 8.3 Hz, 2H), 4.27 (q, J = 7.1 Hz, 2H), 3.88 (dd, J = 12.3, 6.4 Hz, 2H), 3.09 (m, 2H), 3.05 – 2.99 (m, 2H), 2.96 (dd, J = 8.2, 6.3 Hz, 2H), 2.84 (s, 3H), 2.00 (s, 3H), 1.29 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.20, 168.29, 168.13, 163.46, 160.72, 159.20, 150.13, 140.80, 138.33, 137.87, 137.24, 129.56, 124.47, 122.79, 120.33, 120.14, 115.79, 62.17, 41.60, 41.33, 37.36, 33.85, 25.09, 16.52, 15.31. MS (ESI): calcd. 504.207; found 504.20 [M+H]⁺.

Ethyl 2-(3-acetamidophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(t,a)

Isolated 45 mg (81% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 9.84 (s, NHAc), 7.47 (s, 1H), 7.36 (d, J = 7.7 Hz, 1H), 7.14 (t, J = 7.8 Hz, 1H), 7.11 (m, J = 4.5 Hz, NH), 6.90 (d, J = 7.6 Hz, 1H), 4.28 (q, J = 7.1 Hz, 2H), 3.04 – 2.94 (m, 7H), 2.87 (s, 3H), 2.00 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, CDCl₃ + CD₃OD) δ 174.50, 172.26, 171.43, 167.85, 164.01, 143.40, 142.04, 140.74, 133.30, 125.53, 124.67, 119.90, 65.95, 45.18, 38.26, 32.49, 28.11, 19.95, 18.70. MS (ESI): calcd. 413.16; found 413.30 [M+H]⁺.

Ethyl 2-(2-acetamidophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(u,a)

Isolated 35 mg (63% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 9.43 (s, NHAc), 7.29 (d, J = 7.8 Hz, 1H), 7.24 (d, J = 7.6 Hz, 1H), 7.13 (t, J = 7.5 Hz, 1H), 7.08 (m, NH, 1H), 4.27 (q, J = 7.1 Hz, 2H), 3.03 (m, 2H), 2.97 (d, J = 4.5 Hz, 3H), 2.93 (m, 2H), 2.86 (s, 3H), 2.03 (s, 3H), 1.27 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, CDCl₃ + CD₃OD) δ 175.11, 172.05, 171.24, 167.71, 164.02, 143.67, 139.66, 139.37, 134.36, 131.18, 130.57, 130.32, 125.54, 120.05, 66.02, 44.45, 33.67, 32.49, 28.01, 19.89, 18.71. MS (ESI): calcd. 413.16; found 413.10 [M+H]⁺.

Ethyl 2-(3-chlorophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(v,a)

Isolated 122 mg (47% yield) as a pale yellow oil. ¹H NMR (300 MHz, CDCl₃) δ 7.26 (m, 1H), 7.15 – 7.08 (m, 3H), 5.65 (br. m, NH), 4.32 (q, J = 7.1 Hz, 2H), 3.13 – 3.08 (m, 7H), 2.87 (s, 3H), 1.36 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 167.77, 167.51, 163.26, 159.59, 144.15, 138.23, 134.13, 129.73, 128.89, 126.89, 126.17, 121.49, 115.43, 61.46, 40.65, 33.94, 28.33, 15.91, 14.55. MS (ESI): calcd. 390.104; found 390.20 [M+H]⁺.

Ethyl 5-methyl-4-(methylamino)-2-(3-(trifluoromethyl)phenethyl)thieno[2,3-d]pyrimidine-6-carboxylate 10(w,a)

Isolated 128 mg (50% yield) as a pale yellow semi-solid. ¹H NMR (300 MHz, CDCl₃) δ 7.52 (s, 1H), 7.37 (m, 3H), 5.66 (br. m, NH), 4.32 (q, J = 7.1 Hz, 2H), 3.25 – 3.11 (m, 4H), 3.09 (d, J = 4.8 Hz, 3H), 2.87 (s, 3H), 1.36 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 167.77, 167.36, 163.25, 159.60, 142.95, 138.20, 132.09, 130.62 (d, J = 31.9 Hz), 128.88, 125.52 (d, J = 3.8 Hz), 122.87 (d, J = 3.8 Hz), 121.55, 115.43, 61.46, 40.61, 34.04, 28.26, 15.88, 14.51. MS (ESI): calcd. 424.131; found 424.27 [M+H]⁺.

Ethyl 2-(2-(benzo[d][1,3]dioxol-5-yl)ethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylate 10(z,a)

Isolated 62 mg (41% yield) as a pale yellow solid. ¹H NMR (500 MHz, DMSO) δ 7.08 (m, NH), 6.81 (s, 1H), 6.76 (dd, J = 7.9, 2.0 Hz, 1H), 6.67 (d, J = 7.9 Hz, 1H), 5.92 (s, 2H), 4.27 (q, J = 7.1 Hz, 2H), 3.33 (s, 3H), 3.02 – 2.93 (m, 7H), 2.86 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.23, 167.95, 163.50, 159.82, 148.24, 146.36, 141.08, 136.55, 122.20, 120.06, 115.93, 109.95, 109.16, 101.71, 62.16, 41.56, 34.09, 29.27, 16.73, 15.33. MS (ESI): calcd. 400.13; found 400.40 [M+H]⁺.

Dimethyl thiophene-2,5-dicarboxylate (13)

An ice-cooled pressure vessel was charged with thiophene-2,5-carboxylic acid (2.5 g, 14.5 mmol) and 50 mL anhydrous MeOH; thionyl chloride (4.2 mL, 58 mmol) was added dropwise via syringe and once gas evolution had ceased the vessel was tightly sealed with a Teflon screw cap. The reaction was stirred 1h at RT and 3h at 60°C. The reaction mixture was cooled to RT and stirred open to air for 1h. The crude reaction mixture was diluted with 100 mL DCM, stirred for 1h and decanted. The residue was collected by filtration and washed with DCM to obtain the desired product as a whit powder, 2.9 g (99% yield). ¹H NMR (300 MHz, DMSO) δ 7.81 (s, 2H), 3.85 (s, 6H). ¹³C NMR (126 MHz, DMSO) δ 162.40, 139.25, 134.94, 53.99.

Dimethyl 3-aminothiophene-2,5-dicarboxylate (14)

An ice-cooled round-bottom flask was charged with 12 (1.0 g, 5.0 mmol), followed by 5 mL sulfuric acid. Nitric acid (0.50 mL, 6.2 mmol) was added slowly dropwise via syringe and the reaction was stirred at RT for 1 h. The crude reaction mixture was carefully poured on ice-water and stirred for 1h. The suspension was filtered, washed with water and dried on high vacuum to furnish dimethyl 3-nitrothiophene-2,5-dicarboxylate as a white solid, 1.1 g (93% yield). ¹H NMR (500 MHz, DMSO) δ 8.22 (s, 1H), 3.88 (s, 6H). ¹³C NMR (126 MHz, DMSO) δ 161.14, 159.81, 148.22, 136.58, 133.46, 129.26, 55.05, 54.53. A round-bottom flask was charged with dimethyl 3-nitrothiophene-2,5-dicarboxylate (1.0 g, 4.1 mmol), suspended in MeOH/THF 2:1 and purged with nitrogen. Pearlman’s catalyst was added (286 mg, 2.0 mmol, 20% Pd(OH)₂/C, CAS 12135-22-7), and the reaction mixture was purged with H₂ gas and equipped with an H₂-filled balloon. The reaction was stirred at RT until complete consumption of starting material, as determined by TLC. The reaction mixture was purged with nitrogen and filtered through a plug of Celite, rinsing with EtOAc. The filtrate was concentrated in vacuo and the obtained material was used as such without further purification. Isolated 900 mg (100% yield) as a yellow powder. ¹H NMR (500 MHz, DMSO) δ 7.22 (s, 1H), 6.62 (s, NH₂), 3.80 (s, 3H), 3.74 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.82, 162.69, 155.01, 136.68, 126.34, 103.33, 53.84, 52.61.

Thienopyrimidinone formation, amine coupling and ester saponification performed as described in the general procedures.

Methyl 4-(methylamino)-2-phenethylthieno[3,2-d]pyrimidine-6-carboxylate 16(d,a)

Isolated 106 mg (51% yield) as a white solid. ¹H NMR (500 MHz, DMSO) δ 7.98 (m, NH), 7.84 (s, 1H), 7.22 (m, 4H), 7.12 (m, 1H), 3.87 (s, 3H), 3.13 – 3.04 (m, 4H), 2.98 (d, J = 4.5 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.61, 163.19, 159.25, 158.42, 142.92, 137.95, 130.96, 129.43, 129.35, 126.84, 117.58, 54.06, 34.81, 26.99, 26.93. MS (ESI): calcd. 328.11; found 328.27 [M+H]⁺.

Methyl 2-(3-methoxyphenethyl)-4-(methylamino)thieno[3,2-d]pyrimidine-6-carboxylate 16(h,a)

Isolated 100 mg (64% yield) as a pale yellow oil. MS (ESI): calcd. 358.123; found 358.27 [M+H]⁺

General procedure for the saponification of esters

To solution of ester (1 equivalent) in 2:1 THF/MeOH was added 1 M NaOH (3 equivalents) and water to bring the final solvent ratio to 2:1:1 THF/MeOH/H₂O. In cases where the starting material was not fully dissolved, the solution was heated to 50 °C as necessary. The reaction was stirred at RT until complete consumption of starting material as determined by TLC. The reaction mixture was acidified to pH~3 with 1 M HCl and concentrated in vacuo. The resultant slurry was diluted with 5 mL water, filtered and washed successively with deionized water, and purification grade pentane. The obtained residue was dried on high vacuum to furnish the desired free carboxylic acids.

5-Methyl-4-(methylamino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (17)

Isolated 32 mg (87% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.27 – 7.23 (m, 4H), 7.16 (m, 1H), 7.07 (br. d, J = 4.4 Hz, NH), 3.10 – 3.06 (m, 2H), 3.02 – 2.98 (m, 5H), 2.87 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 166.82, 166.58, 164.06, 158.71, 141.62, 138.94, 128.33, 128.26, 125.79, 120.64, 115.00, 40.18, 33.32, 28.11, 15.44. HRMS (ESI): calcd. 328.11197 for C₁₇H₁₈N₃O₂S; found 328.1112 [M+H]⁺.

4-((2-Aminoethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (18)

Isolated 15 mg (35% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.28 – 7.14 (m, 4H), 7.04 (br. s, NH), 3.51 (m, 2H), 3.28 (m, 2H), 3.12 – 3.03 (m, 2H), 3.06 – 2.93 (m, 5H), 2.58 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 166.91, 166.32, 163.82, 158.03, 142.57, 139.42, 128.65, 128.97, 126.02, 120.41, 114.63, 47.69, 41.31, 33.89, 25.16, 16.46. HRMS (ESI): calcd. 357.13852 for C₁₈H₂₁N₄O₂S; found 357.1389 [M+H]⁺.

4-(Benzylamino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (19)

Isolated 40 mg (86% yield) as a white crystalline material. ¹H NMR (500 MHz, DMSO) δ 7.96 (br. s, NH), 7.41 (d, J = 7.1 Hz, 2H), 7.32 (t, J = 7.6 Hz, 2H), 7.25 – 7.21 (m, 1H), 7.19 (d, J = 7.5 Hz, 2H), 7.15 – 7.08 (m, 3H), 4.79 (d, J = 5.8 Hz, 2H), 2.96 (m, 4H), 2.95 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.72, 163.63, 157.97, 140.85, 139.07, 138.84, 128.29, 128.29, 128.25, 127.47, 126.84, 125.94, 122.31, 115.31, 44.39, 38.54, 32.66, 15.43. HRMS (ESI): calcd. 404.14327 for C₂₃H₂₂N₃O₂S; found 404.1432 [M+H]⁺.

5-Methyl-2-phenethyl-4-(phenethylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (20)

Isolated 28 mg (60% yield) as a white crystalline material. ¹H NMR (500 MHz, DMSO) δ 7.34 – 7.19 (m, 9H), 7.17 – 7.13 (m, 1H), 7.11 (br. t, J = 5.7 Hz, NH), 3.74 (dd, J = 14.6, 6.0 Hz, 2H), 3.11 (dd, J = 8.5, 6.3 Hz, 2H), 3.03 (dd, J = 8.5, 6.2 Hz, 2H), 2.95 – 2.89 (m, 2H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 166.79, 164.03, 158.14, 141.56, 139.58, 138.76, 128.73, 128.45, 128.30, 128.28, 126.19, 125.83, 120.89, 114.87, 42.33, 40.13, 34.66, 33.37, 15.33. HRMS (ESI): calcd. 418.15892 for C₂₄H₂₄N₃O₂S; found 418.1557 [M+H]⁺.

4-((4-Methoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (21)

Isolated 36 mg (55% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.52 (m, NH), 7.27 – 7.22 (m, 4H), 7.15 (m, J = 7.5 Hz, 3H), 6.86 (d, J = 8.7 Hz, 2H), 3.72 (dd, J = 14.3, 6.4 Hz, 2H), 3.70 (s, 3H), 3.13 – 3.09 (m, 2H), 3.06 (m, 2H), 2.87 – 2.82 (m, 2H), 2.82 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.45, 158.96, 158.84, 141.58, 139.77, 131.93, 130.86, 129.57, 129.40, 127.30, 116.56, 115.02, 56.16, 44.42, 34.63, 33.54, 16.15. HRMS (ESI): calcd. 448.16949 for C₂₅H₂₆N₃O₃S; found 448.1695 [M+H]⁺.

4-((3-Methoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (22)

Isolated 45 mg (68% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.31 (m, NH), 7.25 – 7.19 (m, 5H), 7.16 – 7.12 (m, 1H), 6.81 (m, 2H), 6.77 (m, 1H), 3.75 (dd, J = 14.0, 6.4 Hz, 2H), 3.69 (s, 3H), 3.13 – 3.08 (m, 2H), 3.06 – 3.01 (m, 2H), 2.88 (t, J = 7.4 Hz, 2H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.79, 160.49, 159.06, 142.11, 141.96, 139.82, 130.61, 129.49, 129.39, 127.13, 122.13, 116.29, 115.49, 112.80, 56.01, 43.78, 35.68, 34.00, 16.28. HRMS (ESI): calcd. 448.16949 for C₂₅H₂₆N₃O₃S; found 448.1684 [M+H]⁺.

4-((2-Methoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (23)

Isolated 28 mg (60% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.26 – 7.13 (m, 7H, NH), 6.95 (d, J = 7.7 Hz, 1H), 6.86 (t, J = 7.3 Hz, 1H), 3.73 (m, 5H), 3.10 (m, 2H), 3.01 (m, 2H), 2.92 (t, J = 7.1 Hz, 2H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.67, 159.08, 158.45, 141.93, 139.86, 131.43, 129.53, 129.39, 128.95, 128.07, 127.19, 121.48, 116.36, 111.85, 56.38, 42.62, 33.72, 30.29, 16.20. HRMS (ESI): calcd. 448.16949 for C₂₅H₂₆N₃O₃S; found 448.1681 [M+H]⁺.

4-((4-Fluorophenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (24)

Isolated 23 mg (35% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.28 – 7.20 (m, 5H, NH), 7.12 (m, 4H), 3.71 (dd, J = 13.8, 6.2 Hz, 2H), 3.11 – 3.06 (m, 2H), 3.03 – 2.98 (m, 2H), 2.89 (t, J = 7.4 Hz, 2H), 2.80 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 165.01, 162.00 (d, J = 241.5 Hz), 159.18, 142.45, 139.87, 136.73 (d, J = 3.0 Hz), 131.63 (d, J = 7.9 Hz), 129.44, 129.40, 127.01, 116.24 (d, J = 21.0 Hz), 116.08, 43.52, 34.82, 34.30, 16.41. HRMS (ESI): calcd. 436.14950 for C₂₄H₂₃FN₃O₂S; found 436.1473 [M+H]⁺.

4-((3-Fluorophenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (25)

Isolated 30 mg (46% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.41 (m, NH), 7.32 (dd, J = 14.4, 7.8 Hz, 1H), 7.26 – 7.21 (m, 4H), 7.16 – 7.12 (m, 1H), 7.07 (m, 2H), 7.02 (td, J = 8.8, 2.5 Hz, 1H), 3.77 (dd, J = 13.9, 6.3 Hz, 2H), 3.12 – 3.07 (m, 2H), 3.06 – 3.02 (m, 2H), 2.93 (t, J = 7.3 Hz, 2H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.44, 163.37 (d, J = 243.2 Hz), 158.87, 143.12 (d, J = 7.4 Hz), 141.59, 139.72, 131.41 (d, J = 8.3 Hz), 129.54, 129.39, 127.28, 126.12, 126.10, 124.37, 116.64 (d, J = 14.6 Hz), 116.54, 114.25 (d, J = 20.8 Hz), 43.76, 38.43, 35.12, 33.58, 16.14. HRMS (ESI): calcd. 436.14950 for C₂₄H₂₃FN₃O₂S; found 436.1479 [M+H]⁺.

4-((2-Fluorophenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (26)

Isolated 21 mg (32% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.32 – 7.19 (m, 6H, NH), 7.12 (m, 3H), 3.76 (dd, J = 13.8, 6.7 Hz, 2H), 3.08 (m, 2H), 3.02 – 2.93 (m, 4H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.90, 161.97 (d, J = 243.3 Hz), 159.18, 142.33, 139.88, 132.52 (d, J = 5.0 Hz), 129.59, 129.53, 129.47, 129.39, 127.10 (d, J = 16.1 Hz), 127.06, 125.55 (d, J = 3.3 Hz), 116.32 (d, J = 21.8 Hz), 116.16, 42.24, 34.08, 29.25, 16.34. HRMS (ESI): calcd. 436.14950 for C₂₄H₂₃FN₃O₂S; found 436.1476 [M+H]⁺.

5-Methyl-4-((4-methylphenethyl)amino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (27)

Isolated 48 mg (73% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.33 (m, NH), 7.26 – 7.21 (m, 4H), 7.16 – 7.09 (m, 5H), 3.72 (dd, J = 14.4, 6.2 Hz, 2H), 3.12 – 3.08 (m, 2H), 3.06 – 3.02 (m, 2H), 2.88 – 2.83 (m, 2H), 2.81 (s, 3H), 2.25 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.64, 158.94, 141.90, 139.78, 137.16, 136.36, 130.15, 129.73, 129.52, 129.38, 127.19, 116.37, 44.09, 35.17, 33.81, 21.80, 16.25. HRMS (ESI): calcd. 432.17457 for C₂₅H₂₆N₃O₂S; found 432.1719 [M+H]⁺.

5-Methyl-4-((3-methylphenethyl)amino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (28)

Isolated 42 mg (64% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.45 (m, NH), 7.25 – 7.22 (m, 4H), 7.20 – 7.13 (m, 2H), 7.07 (s, 1H), 7.02 (m, 2H), 3.74 (dd, J = 14.6, 6.1 Hz, 2H), 3.12 (m, 2H), 3.06 (m, 2H), 2.89 – 2.84 (m, 2H), 2.81 (s, 3H), 2.25 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.43, 158.82, 141.57, 140.02, 139.73, 138.61, 130.58, 129.56, 129.50, 129.38, 128.09, 127.30, 126.94, 116.59, 44.26, 38.35, 35.44, 33.55, 22.15, 16.12. HRMS (ESI): calcd. 432.17457 for C₂₅H₂₆N₃O₂S; found 432.1728 [M+H]⁺.

5-Methyl-4-((2-methylphenethyl)amino)-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (29)

Isolated 55 mg (84% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.42 (m, NH), 7.27 – 7.21 (m, 4H), 7.17 – 7.12 (m, 3H), 7.12 – 7.08 (m, 2H), 3.68 (dd, J = 15.5, 5.8 Hz, 2H), 3.11 – 3.06 (m, 2H), 3.04 – 2.99 (m, 2H), 2.93 – 2.88 (m, 2H), 2.85 (s, 3H), 2.40 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 164.75, 159.07, 142.05, 139.79, 138.43, 137.20, 131.24, 130.55, 129.52, 129.33, 127.51, 127.16, 127.08, 116.25, 42.76, 34.25, 33.47, 20.10, 16.39. HRMS (ESI): calcd. 432.17457 for C₂₅H₂₆N₃O₂S; found 432.1746 [M+H]⁺.

5-Methyl-2-phenethyl-4-((2-(pyridin-4-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylic acid (30)

Isolated 42 mg (64% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 8.47 (d, J = 5.8 Hz, 2H), 7.28 (d, J = 5.7 Hz, 2H), 7.25 – 7.20 (m, 4H), 7.13 (m, NH, 1H), 3.77 (dd, J = 13.4, 6.6 Hz, 2H), 3.09 – 3.05 (m, 2H), 3.00 (m, 2H), 2.93 (t, J = 7.2 Hz, 2H), 2.80 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.00, 167.89, 165.14, 159.23, 150.30, 142.65, 139.84, 129.42, 129.41, 126.95, 125.59, 122.11, 116.00, 42.20, 41.19, 35.03, 34.47, 16.48. HRMS (ESI): calcd. 419.15417 for C₂₃H₂₃N₄O₂S; found 419.1537 [M+H]⁺.

5-Methyl-2-phenethyl-4-((2-(pyridin-3-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylic acid (31)

Isolated 41 mg (87% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 8.45 (s, 1H), 8.40 (d, J = 4.7 Hz, 1H), 7.63 (d, J = 7.8 Hz, 1H), 7.30 (dd, J = 7.7, 4.8 Hz, 1H), 7.25 – 7.20 (m, 4H), 7.13 (m, 1H, NH), 3.75 (dd, J = 13.3, 6.7 Hz, 2H), 3.07 (dd, J = 8.4, 6.1 Hz, 2H), 2.99 (dd, J = 8.5, 6.3 Hz, 2H), 2.92 (t, J = 7.2 Hz, 2H), 2.80 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.00, 167.89, 165.15, 159.25, 151.01, 148.56, 142.67, 139.82, 137.49, 136.22, 129.41, 126.94, 124.63, 122.11, 115.97, 42.86, 41.23, 34.49, 32.89, 16.47. HRMS (ESI): calcd. 419.15417 for C₂₃H₂₃N₄O₂S; found 419.1537 [M+H]⁺.

5-Methyl-2-phenethyl-4-((2-(pyridin-2-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylic acid (32)

Isolated 57 mg (87% yield) as an off white solid. ¹H NMR (500 MHz, DMSO) δ 8.56 (d, J = 4.9 Hz, 1H), 7.83 (m, NH), 7.41 – 7.32 (m, 3H), 7.24 – 7.21 (m, 4H), 7.15 – 7.11 (m, 1H), 3.89 (dd, J = 12.4, 6.6 Hz, 2H), 3.12 (t, J = 6.7 Hz, 2H), 3.05 (m, 2H), 2.96 (m, 2H), 2.84 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.80, 165.14, 159.92, 159.17, 148.62, 142.70, 139.82, 139.46, 129.42, 129.38, 126.90, 125.32, 123.35, 122.12, 115.99, 41.44, 41.17, 36.77, 34.40, 16.42. MS (ESI): calcd. 419.15417 for C₂₃H₂₃N₄O₂S; found 419.1545 [M+H]⁺

2-(4-Acetamidophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (33)

Isolated 27 mg (72% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 9.85 (s, NHAc), 7.46 (m, NH), 7.44 (d, J = 8.3 Hz, 2H), 7.15 (d, J = 8.3 Hz, 2H), 3.03 (m, 7H), 2.87 (s, 3H), 1.99 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.25, 164.47, 159.15, 139.96, 138.66, 136.03, 129.62, 124.20, 120.16, 116.72, 38.55, 33.04, 30.12, 25.10, 16.22. HRMS (ESI): calcd. 385.13344 for C₁₉H₂₁N₄O₃S; found 385.1336 [M+H]⁺.

2-(3-Acetamidophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (34)

Isolated 23 mg (62% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 9.84 (s, NHAc), 7.47 (s, 1H), 7.37 (d, J = 8.3 Hz, 1H), 7.15 (t, J = 7.8 Hz, 1H), 7.08 (m, NH), 6.90 (d, J = 7.9 Hz, 1H), 3.05 – 2.94 (m, 7H), 2.86 (s, 3H), 2.00 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.33, 167.60, 165.14, 159.80, 143.12, 140.46, 140.07, 129.69, 124.16, 121.93, 120.00, 117.75, 116.19, 34.57, 29.33, 25.18, 16.55. HRMS (ESI): calcd. 385.13344 for C₁₉H₂₁N₄O₃S; found 385.1333 [M+H]⁺.

2-(2-Acetamidophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (35)

Isolated 13 mg (46% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 9.40 (s, NHAc), 7.32 (d, J = 7.8 Hz, 1H), 7.25 (d, J = 7.6 Hz, 1H), 7.14 (m, NH), 7.08 (m, 2H), 3.08 – 3.02 (m, 2H), 2.98 (d, J = 4.4 Hz, 3H), 2.96 – 2.91 (m, 2H), 2.86 (s, 3H), 2.04 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.54, 168.12, 165.17, 159.83, 140.10, 137.22, 136.98, 130.26, 127.28, 126.57, 121.82, 116.18, 30.04, 29.25, 24.42, 16.58. HRMS (ESI): calcd. 385.13344 for C₁₉H₂₁N₄O₃S; found 385.1331 [M+H]⁺.

4-((2,3-Dihydro-1H-inden-2-yl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (36)

Isolated 47 mg (72% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.25 – 7.19 (m, 5H, NH), 7.17 – 7.12 (m, 3H), 6.91 (d, J = 7.0 Hz, 1H), 5.07 – 4.98 (m, 1H), 3.36 – 3.28 (m, 2H), 3.09 – 3.04 (m, 3H), 3.01 (m, 3H), 2.87 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 168.05, 167.84, 165.15, 159.46, 142.65, 142.33, 140.06, 129.41, 129.37, 127.58, 126.91, 125.53, 122.21, 116.24, 53.03, 41.21, 39.86, 34.49, 16.39. HRMS (ESI): calcd. 430.15892 for C₂₅H₂₄N₃O₂S; found 430.1562 [M+H]⁺.

4-((3,5-Dimethoxyphenethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (37)

Isolated 31 mg (82% yield) as an off white powder. ¹H NMR (500 MHz, DMSO) δ 7.24 – 7.22 (m, 4H), 7.16 – 7.09 (m, 1H, NH), 6.41 (d, J = 2.2 Hz, 2H), 6.33 (t, J = 2.2 Hz, 1H), 3.73 (dd, J = 14.5, 6.4 Hz, 2H), 3.67 (s, 6H), 3.10 (m, 2H), 3.01 (m, 2H), 2.87 – 2.82 (m, 2H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 165.08, 161.61, 159.21, 142.94, 142.57, 139.84, 129.42, 129.38, 126.99, 116.05, 107.83, 99.15, 56.13, 43.42, 36.02, 34.40, 16.41. MS (ESI): calcd. 478.18005 for C₂₆H₂₈N₃O₄S; found 478.1782 [M+H]⁺.

4-((2-(Benzo[d][1,3]dioxol-5-yl)ethyl)amino)-5-methyl-2-phenethylthieno[2,3-d]pyrimidine-6-carboxylic acid (38)

Isolated 22 mg (47% yield) as a white solid. ¹H NMR (500 MHz, DMSO) δ 7.27 – 7.18 (m, 5H, NH), 7.16 – 7.12 (m, 1H), 6.82 (m, 2H), 6.68 (dd, J = 7.9, 1.6 Hz, 1H), 5.95 (s, 2H), 3.70 (dd, J = 14.0, 6.0 Hz, 1H), 3.13 – 3.06 (m, 2H), 3.02 (m, 2H), 2.85 – 2.78 (m, 5H). ¹³C NMR (126 MHz, DMSO) δ 164.83, 159.07, 148.43, 146.76, 142.18, 139.85, 134.16, 129.49, 129.40, 127.12, 122.76, 116.24, 110.19, 109.33, 101.86, 43.92, 35.33, 34.07, 16.31. HRMS (ESI): calcd. 462.14875 for C₂₅H₂₄N₃O₄S; found 462.1486 [M+H]⁺.

5-Methyl-4-(methylamino)-2-(3-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (39)

Isolated 36 mg (78% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.25 (m, 2H), 7.19 (m, 2H), 7.14 (m, 1H, NH), 2.97 (d, J = 4.4 Hz, 3H), 2.84 (s, 3H), 2.69 (t, J = 7.5 Hz, 2H), 2.62 (t, J = 7.6 Hz, 2H), 2.05 – 1.97 (m, 2H). ¹³C NMR (126 MHz, DMSO) δ 168.13, 165.10, 159.76, 143.09, 140.03, 129.53, 129.38, 126.84, 121.96, 116.11, 38.83, 35.90, 30.49, 29.34, 16.50. MS (ESI): calcd 342.12762 for C₁₈H₂₀N₃O₂S; found 342.1256 [M+H]⁺.

5-Methyl-4-(methylamino)-2-(4-phenylbutyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (40)

Isolated 60 mg (86% yield) as white powder. ¹H NMR (500 MHz, DMSO) δ 7.23 (t, J = 7.5 Hz, 2H), 7.13 (m, 3H), 7.03 (br. m, NH), 2.94 (d, J = 4.4 Hz, 3H), 2.84 (s, 3H), 2.69 (t, J = 7.5 Hz, 2H), 2.58 (t, J = 7.6 Hz, 2H), 1.76 – 1.68 (m, 2H), 1.63 – 1.55 (m, 2H). ¹³C NMR (126 MHz, DMSO) δ 168.72, 167.39, 165.17, 159.83, 143.33, 140.06, 129.42, 129.34, 126.74, 121.71, 116.03, 39.49, 36.12, 31.86, 29.21, 28.41, 16.53. HRMS (ESI): calcd 356.14327 for C₁₉H₂₂N₃O₂S; found 356.1413 [M+H]⁺.

2-(3-Methoxyphenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (41)

Isolated 66 mg (71% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.12 (t, J = 8.0 Hz, 1H), 7.01 (m, NH), 6.77 (m, 2H), 6.68 (dd, J = 9.0, 1.6 Hz, 1H), 3.05 – 2.93 (m, 7H), 2.82 (s, 3H). OCH₃ signal obscured by water peak. ¹³C NMR (126 MHz, DMSO) δ 167.90, 167.48, 165.15, 160.29, 159.80, 144.29, 140.03, 130.39, 121.82, 121.70, 116.09, 115.03, 112.39, 55.95, 41.03, 34.37, 29.19, 16.48. MS (ESI): calc. 358.12254 for C₁₈H₂₀N₃O₃S; found 358.1208 [M+H]⁺.

2-(3-Fluorophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (42)

Isolated 77 mg (83% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.24 (m, 1H), 7.02 (m, 2H, NH), 6.93 (m, 1H), 3.06 (t, J = 7.5 Hz, 2H), 3.00 – 2.95 (m, 2H), 2.94 (d, J = 4.5 Hz, 3H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.64, 167.50, 165.14, 163.28 (d, J = 242.8 Hz), 159.77, 145.69 (d, J = 7.3 Hz), 140.01, 131.16 (d, J = 8.4 Hz), 125.58 (d, J = 2.6 Hz), 121.85, 116.12 (d, J = 20.7 Hz), 116.09, 113.65 (d, J = 20.9 Hz), 40.65, 33.89, 29.17, 16.47. HRMS (ESI): calcd. 346.10255 for C₁₇H₁₇FN₃O₂S; found 346.1006 [M+H]⁺.

5-Methyl-4-(methylamino)-2-(2-(pyridin-2-yl)ethyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (43)

Isolated 46 mg (84% yield) as a white powder. ¹H NMR (300 MHz, DMSO) δ 8.75 (d, J = 4.9 Hz, 1H), 8.41 (td, J = 7.9, 1.5 Hz, 1H), 7.95 (d, J = 8.0 Hz, 1H), 7.82 (t, J = 6.3 Hz, 1H), 7.17 (br. m, NH), 3.47 (t, J = 7.1 Hz, 2H), 3.27 (t, J = 7.1 Hz, 2H), 2.80 (m, 6H). ¹³C NMR (75 MHz, DMSO) δ 165.80, 165.33, 164.51, 159.13, 157.64, 146.10, 142.19, 139.47, 127.38, 125.11, 121.94, 115.76, 36.57, 31.33, 28.73, 15.89. HRMS (ESI): calcd. 329.10722 for C₁₆H₁₇N₄O₂S; found 329.1067 [M+H]⁺.

2-(4-Acetamidophenethyl)-4-((2-aminoethyl)amino)-5-methylthieno[2,3-d]pyrimidine-6-carboxylic acid (44)

Isolated 17 mg (36% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 9.84 (s, NHAc), 8.50 (br. s, NH₂), 7.56 (br. s, NH), 7.44 (d, J = 8.3 Hz, 2H), 7.15 (d, J = 8.4 Hz, 2H), 3.82 (m, 2H), 3.15 (m, 2H), 3.03 (s, 3H), 3.00 (d, J = 7.2 Hz, 2H), 2.94 (d, J = 7.8 Hz, 2H), 2.00 (s, 3H). HRMS (ESI): calcd. 414.15999 for C₂₀H₂₄N₅O₃S; found 414.1591 [M+H]⁺.

2-(4-Acetamidophenethyl)-5-methyl-4-((2-(pyridin-2-yl)ethyl)amino)thieno[2,3-d]pyrimidine-6-carboxylic acid (45)

Isolated 33 mg (64% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 9.83 (s, NH), 8.51 (d, J = 4.9 Hz, 1H), 7.71 (tt, J = 7.7, 1.9 Hz, 1H), 7.42 (d, J = 6.8 Hz, 2H), 7.36 (br. t, J = 4.8 Hz, NH), 7.28 (d, J = 7.8 Hz, 1H), 7.25 – 7.20 (m, 1H), 7.13 (d, J = 6.9 Hz, 2H), 3.87 (dd, J = 12.2, 5.5 Hz, 2H), 3.08 (t, J = 6.5 Hz, 2H), 3.04 – 2.99 (m, 2H), 2.97 – 2.93 (m, 2H), 2.85 (s, 3H), 1.99 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.20, 167.80, 165.22, 160.73, 159.22, 150.13, 138.30, 137.87, 137.25, 129.57, 124.49, 122.79, 120.13, 116.03, 41.56, 41.31, 37.42, 33.89, 25.07, 16.32. HRMS (ESI): calcd. 476.17564 for C₂₅H₂₆N₅O₃S; found 476.1761 [M+H]⁺

2-(3-Chlorophenethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (46)

Isolated 62 mg (67% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.28 (s, 1H), 7.25 – 7.21 (m, 1H), 7.18 – 7.14 (m, 2H), 7.03 (m, NH), 3.07 – 3.02 (m, 2H), 2.97 (m, 2H), 2.94 (d, J = 4.5 Hz, 3H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.47, 167.21, 165.11, 159.74, 145.25, 140.00, 133.93, 131.16, 129.37, 128.21, 126.90, 121.93, 116.10, 40.51, 33.75, 29.20, 16.46. HRMS (ESI): calcd. 362.07300 for C₁₇H₁₇ClN₃O₂S; found 362.0719 [M+H]⁺.

5-Methyl-4-(methylamino)-2-(3-(trifluoromethyl)phenethyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (47)

Isolated 80 mg (86% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 7.55 (s, 1H), 7.46 (m, 3H), 7.04 (br. m, NH), 3.14 (t, J = 7.5 Hz, 2H), 3.02 (t, J = 7.5 Hz, 2H), 2.92 (d, J = 4.4 Hz, 3H), 2.81 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.36, 167.13, 165.10, 159.73, 144.05, 139.97, 133.66, 130.31, 130.16, 129.91, δ 126.06 (q, J = 3.8 Hz), 123.66 (q, J = 3.9 Hz), 121.96, 116.09, 33.79, 29.14, 16.44. HRMS (ESI): calcd. 396.09936 for C₁₈H₁₇F₃N₃O₂S; found 396.0975 [M+H]⁺.

2-(2-(Benzo[d][1,3]dioxol-5-yl)ethyl)-5-methyl-4-(methylamino)thieno[2,3-d]pyrimidine-6-carboxylic acid (48)

Isolated 25 mg (54% yield) as white solid. ¹H NMR (500 MHz, DMSO) δ 7.13 (br. s, NH), 6.82 (s, 1H), 6.77 (d, J = 7.9 Hz, 1H), 6.67 (d, J = 7.9 Hz, 1H), 5.93 (s, 2H), 3.02 – 2.93 (m, 7H), 2.86 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.45, 165.09, 159.73, 148.25, 146.39, 140.07, 136.42, 122.21, 122.05, 116.18, 109.95, 109.18, 101.72, 41.21, 34.01, 29.36, 16.53. HRMS (ESI): calcd. 372.10180 for C₁₈H₁₈N₃O₄S; found 372.1021 [M+H]⁺.

4-(methylamino)-2-phenethylthieno[3,2-d]pyrimidine-6-carboxylic acid (49)

Isolated 40 mg (84% yield) as an off-white powder. ¹H NMR (500 MHz, DMSO) δ 7.99 (m, NH), 7.81 (s, 1H), 7.27 – 7.23 (m, 4H), 7.16 (m, 1H), 3.12 – 3.07 (m, 2H), 3.05 – 3.01 (m, 2H), 2.99 (d, J = 4.0 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.38, 164.21, 158.49, 142.88, 130.23, 129.48, 129.39, 126.90, 41.32, 34.82, 28.44. HRMS (ESI): calcd. 314.09632 for C₁₆H₁₆N₃O₂S; found 314.0962 [M+H]⁺.

2-(3-Methoxyphenethyl)-4-(methylamino)thieno[3,2-d]pyrimidine-6-carboxylic acid (50)

Isolated 78 mg (81% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 8.11 (br. d, J = 4.3 Hz, 1H), 7.76 (s, 1H), 7.13 (t, J = 7.8 Hz, 1H), 6.81 – 6.75 (m, 2H), 6.68 (dd, J = 8.3, 1.7 Hz, 1H), 3.04 – 2.99 (m, 4H), 2.97 (d, J = 4.4 Hz, 3H). OCH₃ signal obscured by water peak. ¹³C NMR (126 MHz, DMSO) δ 167.10, 164.14, 160.30, 144.29, 130.40, 129.32, 121.72, 121.68, 115.03, 112.40, 55.95, 40.79, 34.69. HRMS (ESI): calcd. 344.10689 for C₁₇H₁₈N₃O₃S; found 344.1072 [M+H]⁺.

Ethyl 4-(((R)-2-methoxy-2-phenylethyl)amino)-5-methyl-2-((S)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylate – 51 (precursor)

Isolated 101 mg (53% yield) as a colorless syrup. ¹H NMR (500 MHz, DMSO) δ 7.38 (m, 2H), 7.34 – 7.28 (m, ), 7.22 (m, 5H), 7.10 (t, J = 6.8 Hz, 1H), 6.96 (br. t, J = 5.5 Hz, NH), 4.54 (dd, J = 7.5, 5.3 Hz, 1H), 4.27 (q, J = 7.1 Hz, 2H), 3.76 – 3.69 (m, 1H), 3.69 – 3.61 (m, 1H), 3.43 (m, 1H), 3.18 (s, 3H), 2.96 (m, 2H), 2.76 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H), 1.20 (d, J = 6.9 Hz, 3H). ¹³C NMR (75 MHz, DMSO) δ 167.65, 167.15, 162.87, 158.58, 147.31, 140.51, 140.03, 129.07, 128.85, 128.46, 127.41, 127.19, 126.50, 120.06, 115.15, 81.57, 79.84, 61.61, 57.16, 47.51, 39.04, 22.62, 15.94, 14.75. MS (ESI): calcd. 490.22; found 490.33 [M+H]⁺.

4-(((R)-2-methoxy-2-phenylethyl)amino)-5-methyl-2-((S)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (51)

Isolated 13 mg (28% yield) as white solid. ¹H NMR (500 MHz, DMSO) δ 7.38 (m, 2H), 7.31 (m, 3H), 7.22 (m, 4H), 7.12 – 7.08 (m, 1H), 6.85 (br. s, NH), 4.54 (m, 1H), 3.72 (m, 1H), 3.68 – 3.60 (m, 1H), 3.18 (s, 3H), 3.02 – 2.86 (m, 3H), 2.75 (s, 3H), 1.20 (d, J = 6.9 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.68, 167.10, 165.30, 159.07, 147.85, 141.06, 129.60, 129.40, 128.99, 127.97, 127.75, 127.04, 115.94, 82.06, 57.70, 47.98, 47.61, 39.57, 23.21, 16.17. HRMS (ESI): calcd. 462.18514 for C₂₆H₂₈N₃O₃S; found 462.1853 [M+H]⁺

Ethyl 4-(((S)-2-methoxy-2-phenylethyl)amino)-5-methyl-2-((S)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylate 52 (precursor)

Isolated 31 mg (16% yield) as a colorless syrup. ¹H NMR (500 MHz, DMSO) δ 7.41 – 7.36 (m, 2H), 7.31 (m, 3H), 7.25 – 7.20 (m, 4H), 7.10 (m, 1H), 6.99 (br. t, J = 5.4 Hz, NH), 4.51 (dd, J = 7.8, 4.7 Hz, 1H), 4.27 (q, J = 7.1 Hz, 2H), 3.79 – 3.73 (m, 1H), 3.62 (m, 1H), 3.44 (m, 1H), 3.17 (s, 3H), 2.95 (m, 2H), 2.78 (s, 3H), 1.28 (t, J = 7.1 Hz, 3H), 1.22 (d, J = 7.0 Hz, 3H). ¹³C NMR (75 MHz, DMSO) δ 167.62, 167.17, 162.90, 158.60, 147.27, 140.52, 140.12, 129.11, 128.88, 128.51, 127.45, 127.20, 126.54, 120.05, 115.17, 81.63, 61.68, 57.21, 47.49, 36.87, 24.96, 22.71, 16.00, 14.80. MS (ESI): calcd. 490.22; found 490.40 [M+H]⁺.

4-(((S)-2-Methoxy-2-phenylethyl)amino)-5-methyl-2-((S)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (52)

Isolated 15 mg (53% yield) as a white solid. ¹H NMR (500 MHz, DMSO) δ 7.39 (m, 2H), 7.31 (m, 3H), 7.21 (m, 4H), 7.10 (m, 1H), 6.97 (m, NH), 4.51 (dd, J = 7.7, 4.4 Hz, 1H), 3.76 (m, 1H), 3.62 (m, 1H), 3.17 (s, 3H), 2.95 (m, 3H), 2.77 (s, 3H), 1.22 (d, J = 6.9 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.30, 165.07, 159.10, 147.75, 141.02, 139.64, 129.62, 129.40, 129.02, 127.97, 127.73, 127.07, 122.30, 115.88, 82.12, 57.72, 47.89, 47.70, 39.49, 23.24, 16.32. HRMS (ESI): calcd. 462.18514 for C₂₆H₂₈N₃O₃S; found 462.1857 [M+H]⁺.

Ethyl 4-(((R)-2-methoxy-2-phenylethyl)amino)-5-methyl-2-((R)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylate 53 (precursor)

Isolated 65 mg (63% yield) as pale yellow syrup. ¹H NMR (300 MHz, CDCl₃) δ 7.46 – 7.20 (m, 9H), 7.16 – 7.09 (m, 1H), 6.11 (br. t, NH), 4.41 – 4.31 (m, 3H), 4.08 (ddd, J = 13.8, 7.2, 4.0 Hz, 1H), 3.56 – 3.44 (m, 2H), 3.30 (s, 3H), 3.17 – 2.96 (m, 2H), 2.87 (s, 3H), 1.39 (t, J = 7.1 Hz, 3H), 1.29 (d, J = 7.0 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 163.33, 158.77, 147.18, 139.30, 138.36, 128.97, 128.57, 128.47, 127.27, 126.87, 126.13, 115.37, 82.07, 61.45, 57.25, 47.11, 39.30, 24.96, 22.09, 15.71, 14.56. MS (ESI): calcd. 490.22; found 490.33 [M+H]⁺.

4-(((R)-2-Methoxy-2-phenylethyl)amino)-5-methyl-2-((R)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (53)

Isolated 43 mg (76% yield) as white solid. ¹H NMR (500 MHz, DMSO) δ 7.39 (m, 2H), 7.31 (m, 3H), 7.25 – 7.18 (m, 4H), 7.12 – 7.07 (m, 1H), 7.04 ( br. t, J = 5.4 Hz, NH), 4.50 (dd, J = 7.9, 4.7 Hz, 1H), 3.80 – 3.73 (m, 1H), 3.66 – 3.61 (m, 1H), 3.17 (s, 3H), 2.96 (m, 4H), 2.77 (s, 3H), 1.22 (d, J = 7.0 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.06, 165.01, 159.08, 147.65, 140.95, 139.63, 129.63, 129.42, 129.04, 127.96, 127.73, 127.10, 122.47, 115.92, 82.10, 68.17, 57.72, 47.73, 39.47, 23.22, 16.28. HRMS (ESI): calcd. 462.18514 for C₂₆H₂₈N₃O₃S; found 462.1847 [M+H]⁺.

Ethyl 4-(((S)-2-methoxy-2-phenylethyl)amino)-5-methyl-2-((R)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylate 54 (precursor)

Isolated 67 mg (65% yield) as a colorless syrup. ¹H NMR (300 MHz, CDCl₃) δ 7.46 – 7.32 (m, 5H), 7.30 – 7.19 (m, 4H), 7.13 (dt, J = 9.3, 4.2 Hz, 1H), 6.11 (br. t, NH), 4.39 (m, 3H), 4.06 (ddd, J = 13.7, 7.0, 4.1 Hz, 1H), 3.57 – 3.44 (m, 2H), 3.31 (s, 3H), 3.08 (m, 2H), 2.87 (s, 3H), 1.38 (t, J = 6.2 Hz, 3H), 1.27 (d, J = 6.9 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃) δ 163.33, 158.78, 147.22, 139.29, 138.35, 128.97, 128.57, 128.47, 127.26, 126.88, 126.12, 115.37, 82.06, 61.45, 57.24, 47.06, 39.28, 36.93, 22.18, 15.68, 14.55. MS (ESI): calcd. 490.22; found 490.33 [M+H]⁺.

4-(((S)-2-Methoxy-2-phenylethyl)amino)-5-methyl-2-((R)-2-phenylpropyl)thieno[2,3-d]pyrimidine-6-carboxylic acid (54)

Isolated 42 mg (74% yield) as a yellow solid. ¹H NMR (500 MHz, DMSO) δ 7.41 – 7.36 (m, 2H), 7.34 – 7.29 (m, 3H), 7.26 – 7.19 (m, 4H), 7.12 – 7.08 (m, 1H), 7.00 (br. t, J = 5.5 Hz, NH), 4.54 (dd, J = 7.6, 5.1 Hz, 1H), 3.76 – 3.70 (m, 1H), 3.69 – 3.62 (m, 1H), 3.18 (s, 3H), 2.96 (m, 3H), 2.75 (s, 3H), 1.20 (d, J = 7.0 Hz, 3H). ¹³C NMR (126 MHz, DMSO) δ 167.12, 165.02, 159.09, 147.68, 140.96, 139.62, 129.62, 129.43, 129.04, 127.96, 127.75, 127.10, 122.45, 115.91, 82.02, 68.17, 57.70, 47.71, 39.55, 23.20, 16.25. HRMS (ESI): calcd. 462.18514 for C₂₆H₂₈N₃O₃S; found 462.1847 [M+H]⁺.

Ethyl (S)-2-(4-acetamidophenethyl)-4-((2-methoxy-2-phenylethyl)amino)-5-methylthieno[2,3-d]pyrimidine-6-carboxylate 55 (precursor)

Isolated 40 mg (82% yield, over two steps) as a white powder. ¹H NMR (300 MHz, CDCl₃) δ 7.55 (s, 1H), 7.41 (m, 7H), 7.15 (d, J = 8.4 Hz, 2H), 6.16 (br. m, NH), 4.43 (dd, J = 8.6, 3.8 Hz, 1H), 4.34 (q, J = 7.1 Hz, 2H), 4.14 (ddd, J = 14.0, 7.0, 4.3 Hz, 1H), 3.54 (ddd, J = 13.7, 8.7, 3.8 Hz, 1H), 3.31 (s, 3H), 3.07 (m, 4H), 2.88 (s, 3H), 2.12 (s, 3H), 1.38 (t, J = 7.1 Hz, 4H). ¹³C NMR (75 MHz, CDCl₃) δ 163.32, 158.91, 150.73, 137.95, 129.11, 128.98, 126.87, 120.18, 115.46, 84.48, 61.49, 57.24, 47.08, 41.08, 33.87, 24.72, 15.72, 14.55. MS (ESI): calcd. 533.22; found 533.27[M+H]⁺.

(S)-2-(4-Acetamidophenethyl)-4-((2-methoxy-2-phenylethyl)amino)-5-methylthieno[2,3-d]pyrimidine-6-carboxylic acid (55)

Isolated 24 mg (63% yield) as a white powder. ¹H NMR (500 MHz, DMSO) δ 9.81 (s, NHAc), 7.43 (d, J = 8.5 Hz, 2H), 7.39 – 7.28 (m, 5H), 7.12 (d, J = 8.5 Hz, 2H), 6.98 (br. t, J = 5.2 Hz, NH), 4.54 (dd, J = 7.7, 5.0 Hz, 1H), 3.80 (m, 1H), 3.67 (m, 1H), 3.18 (s, 3H), 2.98 (m, 4H), 2.79 (s, 3H), 1.98 (s, 3H). ¹³C NMR (126 MHz, DMSO) δ 169.15, 167.82, 165.09, 159.21, 141.02, 139.66, 138.37, 137.11, 129.61, 129.55, 129.00, 127.75, 122.32, 120.10, 115.96, 82.14, 57.69, 47.66, 41.28, 33.91, 25.08, 16.33. HRMS (ESI): calcd. 505.19095 for C₂₇H₂₉N₄O₄S; found 505.1916 [M+H]⁺.

Protein expression and purification

The heterologous expression of PglD was accomplished using the E. coli BL-21(DE3) strain (Stratagene). Cells were transformed with pETNO-construct plasmids and grown to an A600 of 0.6 absorbance units at 37 °C in Lysogeny broth; the cultures were cooled to 16 °C and then induced by the addition of 0.5 mM isopropyl-D-thiogalactopyranoside (IPTG). The pETNO plasmid is based on the pET30a plasmid modified to include a His₈TEV (purification tag/protease cleavage site) in place of the existing His₆-thrombin site. Twenty hours after induction the cells were harvested by centrifugation and resuspended in ice-cold buffer composed of 50 mM HEPES pH 7.1, 10 mM imidazole, 150 mM NaCl, at1/20 the original culture volume. Maintaining a working temperature of 4 °C, the cells were lysed by sonication, and the lysate was cleared by centrifugation in a Type 45 Ti rotor (Beckman/Coulter) at 35,000 rpm. The extract was bound to Ni-NTA (Qiagen) in batch mode using 1 mL of resin per liter of culture for 1h with gentle tumbling. The protein-bound resin was washed with 25 column volumes of lysis buffer containing 50 mM imidazole, and the protein was eluted in lysis buffer containing 250 mM imidazole. The hexahistidine tag was removed by TEV cleavage, and after the digest reached completion, the reaction was dialyzed twice against 4 L 20 mM HEPES pH 7.1, 150 mM NaCl with a 10,000 MWCO Snakeskin (Thermo Scientific). Further purification was performed by size exclusion chromatography using a Superdex 200 XK16–60 column (GE Healthcare) in a running buffer of 20 mM HEPES pH 7.1, 100 mM NaCl.

High-throughput screening – Broad Institute MLPCN

To a Nunc 384-well black-clear bottom plate (Thermo Scientific) was added 200 nL of compound in neat DMSO (final assay concentration = 10 μM) and 20 μL of enzyme solution (final assay concentrations: 6 nM PglD, 50 mM HEPES pH 7.4, 0.001% Triton, 0.05% BSA, 100 μM AcCoA). Solution dispensing was accomplished utilizing a Multidrop Combi (Thermo Scientific). This mixture was allowed to incubate at room temperature for a duration of 45 minutes. The reaction was started by the addition of 10 μL reaction solution (final assay concentrations: 50 mM HEPES pH7.4, 100 μM UDP-4-aminosugar). The reaction was quenched after 30 minutes with the addition of 30 μL stop solution (final assay concentrations: 2 mM DTNB, 1 mM EDTA, 20% 1-propanol). The plates were allowed to develop for 5 minutes before reading at a λ = 405 nm.

Enzymatic inhibition assay

Kinetic characterization of PglD inhibition was carried out by monitoring CoASH release resulting from the acetyltransferase reaction with Ellman’s reagent (DTNB) in a continuous fashion. To a flat, clear bottom 96-well plate (Falcon) was added sequentially the inhibitor as a DMSO stock solution, followed by PglD in 50 mM HEPES pH 7.8, 1 mM MgCl₂, 0.05 mg/mL BSA, 0.001% Triton X-100. Inhibitors and enzyme were pre-incubated 30 min at RT, followed by addition of substrate and Ellman’s reagent cocktail to final concentrations of 3 nM PglD, 500 μM 4-amino-sugar, 300 μM AcCoA, 2 mM DTNB and 3% DMSO in 150 μL volume. Initial rates were measured in the linear portion of the reaction curve over a 5 min time period at RT, measuring absorbance at 412 nm on a Synergy H1 hybrid plate reader from Biotek.

Dynamic Light Scattering

Samples were prepared in buffer 50 mM HEPES 7.5, 100 mM NaCl and a final concentration of 3% DMSO; the concentration of C jejuni PglD was 25 μM with or without 250 μM inhibitor 57. DLS data was acquired on a Wyatt DynaPro Nanostar Light Scatterer by measuring scattering at a 90° angle with a 685 nm laser and analyzed with a Rayleigh scattering model using the Dynamics 7.1.9 software package.

Differential Scanning Fluorimetry

Samples were prepared to a final volume of 25 μL containing 10 μM PglD protein and 100 μM inhibitor in a buffer containing 50 mM HEPES 7.5, 150 mM NaCl, 0.001 % Triton X-100, 10× Sypro orange (diluted from a commercial stock solution of 5000X; Invitrogen) and a final concentration of 3% DMSO. All samples were prepared in duplicate. Fluorescence was measured using a Roche Lightcycler 480 RT-PCR instrument while increasing the temperature gradient from 30 to 95°C in increments of 4.4°C/60 s. The midpoint temperature of the unfolding protein transition (T_m) was calculated using the built-in functionality of the instrument software package.

Crystallization of C. jejuni PglD in complex with inhibitor 17

Initial crystallization screening was performed by sitting drop vapor diffusion using intelli-plate^® from Hampton Research with 50 μL/well solution, which was dispensed as a 0.15 μL drop followed 0.15 μL of 10 mg/mL PglD solution in SEC buffer. Screens used were PEG suite 1, PEG suite 2, PACT and ProComp from Qiagen, and drops were set up using an Art Robbins Phoenix robot. Diffraction quality crystals were formed via hanging drop vapor diffusion by mixing 1.5 μL of protein solution with 1.5 μL of reservoir solution (15% PEG-3350, 0.1M NaF and 10% glycerol) and incubation at 22 °C.

Ligand soaking: Wells were carefully opened and 50 μL of DMSO was added directly to the well; 0.75 μL of a solution containing 10 mM inhibitor 17 in reservoir solution (final concentration of 10% DMSO) was added directly to the drops on the slide. The crystals were then incubated for 30 min to 3 days, and then looped and flash frozen in liquid nitrogen without the need of separate cryoprotection.

Data Collection, Processing, and Structure Refinement

Diffraction data was collected either at a home source (Rigaku Saturn 944 with MicroMax-007 HF and VariMax HF generator) or at the Advanced Photon Source (APS) at Argonne National Laboratory (in Argonne, Illinois, USA) national synchrotron on the NECAT-24ID beamline. The diffraction data were indexed and scaled with either HKL2000. The initial structure model was built by applying molecular replacement methods with a ligand/solvent-omitted input model generated from the PDB model 3BSY. The initial models were further improved through iterative rounds of manual and automated refinement with COOT.⁵⁶ The final structure has been deposited into the Protein Data Bank PDB ID 5T2Y. Data collection and refinement statistics are presented in in the supporting information Table S3

Abbreviations

AcCoA

acetyl coenzyme A

BOP

(Benzotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluorophosphate

DBU

1,8-Diazabicyclo[5.4.0]undec-7-ene

diNAcBac

2,4-diacetamido-2,4,6-trideoxy-α-D-glucose

PyBOP

(Benzotriazol-1-yloxy)tripyrrolidinophosphonium hexafluorophosphate

DSF

Differential scanning fluorimetry

DLS

dynamic light scattering

PLP

Pyridoxal 5′-phosphate

STD

saturation transfer difference

UndP

undecaprenyl phosphate