Thiazolide Prodrug Esters and Derived Peptides: Synthesis and Activity

Abstract

Amino acid ester prodrugs of the thiazolides, introduced to improve the pharmacokinetic parameters of the parent drugs, proved to be stable as their salts but were unstable at pH > 5. Although some of the instability was due to simple hydrolysis, we have found that the main end products of the degradation were peptides formed by rearrangement. These peptides were stable solids: they maintained significant antiviral activity, and in general, they showed improved pharmacokinetics (better solubility and reduced clearance) compared to the parent thiazolides. We describe the preparation and evaluation of these peptides.

Introduction

The thiazolides,¹ exemplified by nitazoxanide (NTZ) 1 and RM5038 2, are broad-spectrum antiviral agents with proven activity in cell line assays²⁻⁵ and clinical trials^6,7 against a number of DNA and RNA viruses. Most recently, NTZ 1 has been shown to be effective in clinical trials against SARS-CoV-2, specifically in preventing progression of the disease from the mild to the severe phase.^8,9 In the body, these O-acetates behave as simple prodrugs for the free phenols 3 and 4, which are the effective circulating forms of the drugs in vivo.

Nevertheless, the pharmacokinetic parameters of 1, 2, and other analogues, especially their low solubility, reduce their effectiveness toward respiratory viruses where a good circulating concentration of drug is important. Amino acid prodrugs have frequently been used to address this problem:¹⁰⁻¹² consequently, we synthesized amino acid¹³ ester derivatives of 1 and 2, specifically the tert-leucine derivatives 5a and 6.¹⁴ Here, the bulky t-butyl group was necessary to impart sufficient stability to the products; 5a and 6 were stable for months at room temperature. In contrast, the corresponding valine esters (see below) decomposed at a significant rate at 20 °C.

We obtained proof of concept: the absolute bioavailability of 5a/6 was 20–25%,¹³ as measured by oral bioavailability in rats, although the aqueous solubility of 5a was still not high. By contrast, salt 6 had better solubility and could be recrystallized. An alternative approach, which we recently described,¹⁵ is offered by thiazolide amine salts such as 7, which also possess superior pharmacokinetics to 1 and 2, especially better solubility.

Since it proved difficult to obtain consistent formulations of 5a as the HCl salt, we also prepared the corresponding tosylate¹⁶ and mesylate salts 5b and 5c. These were obtained microanalytically pure and, in contrast to 5a, could be recrystallized without decomposition. It was still difficult to characterize such products by high-performance liquid chromatography (HPLC) analysis, however, as even at pH 5, solutions of these salts proved to be unstable, with apparently one major product being formed. We now report on the characterization and biological evaluation of the rearrangement products.

Results and Discussion

NMR Studies

We studied the aqueous stability further by NMR, Figures 1 and 2. Solutions of 5a in d₆-DMSO with 10% added D₂O held at 25–40 °C showed a steady decomposition: a small part of this behavior could be attributed to direct hydrolysis, regenerating 3, but a new major product was clearly being generated even without an added base. In the case of valacyclovir, which we originally used as a model for 5a and 6, it was known that some hydrolysis occurred in vivo.¹⁷ Clearly, at the 18 day time point [Figure 2g], the rearrangement and hydrolysis products accounted for ∼50% of the total analyte. In Figure 3, the aryl region is integrated and compared with the spectra of 5a, 3, and 8 (see below).

Figure 1.

¹H NMR spectrum of pure 5a in d₆-DMSO.

Figure 2.

Time-course ¹H NMR spectra of the aromatic region of 5a in d₆-DMSO + 10%D₂O. Evolution with time over an 18 day period at 298–313 K. The sample was held in an isothermal bath at a specified temperature between measurements. Spectra were recorded at the temperature stated in each case. (a) t = zero, 298 K; (b) t = 1 week, 308 K; (c) t = 8 days, 308 K; (d) t = 9 days, 308 K; (e) t = 11 days, 313 K; (f) t = 14 days, 313 K; (g) t = 18 days, 313 K.

Figure 3.

Assignment and integration of the final time-point spectrum. The figure shows the Ar region protons corresponding to: (a) tizoxanide 3, (b) prodrug ester 5a, (c) rearranged product 8, and (d) the final time-point spectrum showing by integration a virtually 50% loss of 5a.

Base-Catalyzed Rearrangement

Using initially preparative HPLC, a sample of the degradation product from 5a was purified and characterized. In fact, it was easier to obtain this new product by briefly stirring a suspension of 5a in tetrahydrofuran (THF) with one equivalent of triethylamine at 20 °C (0.5 h) followed by chromatography. This product, readily obtained in a microanalytically pure state, was shown to be the pseudotripeptide 8, corresponding to an insertion of the amino acid unit into the amide bond of 5a.

This kind of rearrangement has a precedent. Thus, Vinsova et al. obtained similar tripeptides derived from salicylanilides, e.g., 9 from amino acid ester 10.^18,19 A mechanism was proposed, and a byproduct 11 was isolated, characterized by X-ray crystallography, corresponding to dehydration of an imidazolinone intermediate along the reaction pathway. In the Supporting Information, we give a plausible mechanism for the rearrangement of 5a to 8 and highlight the “cyclol” intermediate, which could lead to a byproduct akin to 11.

In this sequence, the precursors of esters 10, which were N-protected with the benzyloxycarbonyl (Z) group, were shown to have antimicrobial activity especially against some fungal strains.²⁰ A later paper,²¹ which did not cite the work of the Vinsova group, reported tripeptide-like compounds 12, similar to 9 and formally related to niclosamide.^22,23 Such derivatives were shown to have useful activity against adenovirus: the screened analogues were prepared by linear peptide synthesis from the appropriate amino acid anilides.

The history of the (O-acyl)salicyloyl amide rearrangement actually dates back at least to the 1950s. Thus, Brenner et al. reported^24,25 the synthesis of salicyloyl di- and tripeptides such as 13 from salicylic acid derivatives by a sequence of O-acylation, deprotection, and basification, using Z-protected amino acids. An O- to N-acetyl transfer in a salicylamide derivative was reported by Titherley and Hicks in 1905.²⁶ In our series, and Vinsova’s, we believe the N-aryl/heteroaryl group in 5a and 10 may facilitate this rearrangement by increasing the acidity of the NH proton.

Synthesis and Evaluation of Peptidic Analogues

We naturally investigated the activity and pharmacokinetic parameters of 8 and analogues, not least because any activity in tripeptides such as 8 could be partly responsible for the activity of 5a and 6. Since the work of Sanchez-Cespedes et al.²¹ showed that the valine analogue of their niclosamide-derived series was the most active against adenovirus in their studies, we also prepared the corresponding Val analogues of 3 and 4. The general synthesis of all of the peptides is shown in Scheme 1.

Scheme 1. General Synthesis of Aminothiazolyl Tripeptides.

Conditions: (i) EDC, Boc-amino acid, DMAP, THF, 0–20 °C, 54–67%; (ii) 4 M HCl-dioxan, CH₂Cl₂, 20 °C, 78–94%; (iii) Et₃N, THF, 0–20 °C, 52–76%.

Acylation of 3 or 4 with Boc-Val-OH or Boc-Tle-OH afforded esters 14–17 in 54–67% yield after chromatography. Deprotection with 4 M HCl-dioxan then gave the prodrug ester HCl salts 5a, 6, 18, and 19 in good to excellent yields.¹³ Alternatively, in the case of 14, the use of toluene p-sulfonic acid or methanesulfonic acid afforded the corresponding sulfonate salts 5b and 5c. The rearrangement of salts 5a, 6, 18, and 19 to the tripeptides 8 and 20–22 was complete after treatment with Et₃N in THF for 0.5 h at 20 °C: the weaker base N-methylmorpholine gave a considerably slower reaction. Rearrangement products 8 and 20–22 were produced together with some parent thiazolides 3 or 4, which were removed by chromatography. In the case of the valine analogue 22, we streamlined the synthesis. After acylation of 4 with Boc-Val-OH, unreacted 4 was removed by chromatography and the crude product 17 was immediately progressed by deprotection as above. The resulting HCl salt 19 was subjected to base-catalyzed rearrangement as above, and direct recrystallization of the product afforded tripeptide 22 in excellent purity.

The rearranged products were screened both for antiviral activity and for their DMPK properties, Table 1. In general terms, all four compounds 8 and 20–22 showed reduced clearance compared to the parent thiazolides, as well as better solubility for the 5-Cl derivatives. It is known that glucuronidation is the major clearance pathway for 3(27) and 4,³ but we have no detail regarding the metabolites of 8 and 20–22. The significant decrease in clearance rates in hepatocytes, in contrast with the smaller effect on liver microsomes, is also consistent with a major glucuronidation metabolic pathway.

Table 1. Summarized DMPK and Antiviral Data for Thiazolides and Derived Peptides.

compound	log P	solubilitya	clearance RH μL/min/106 cells	clearance HLM μL/min/mg	IC50b	IC90b	CC50b	SIc
3	1.8	33	244	50.2	2.3	26.4	>189	>82
4	2.9	48	>300	37.5	3.9	19.6	>196	>50
8	3.3	18	31.5	36.5	1.6	79	>132	>83
20	>3.5	388	75.4	42.4	>40.8d	>40.8	40.8
21	2.9	>1000	67.7	55.9	13.7	>137	>137	>10
22	e	82	234	42.7	>42.4d	>42.4	42.4

^aMicromolar units at pH 7.4.

^bActivities vs influenza A virus and cytotoxic concentrations (CC) are in μM. IC₅₀ is 50% inhibitory concentration; IC90 is 90% inhibitory concentration; CC₅₀ is 50% cytotoxic concentration.

^cSI, selectivity index, viz. CC₅₀/IC₅₀.

^dThese compounds appear toxic.

^eFailed to create analytical method.

Clearly, the nitro-tripeptide 8 retains good activity and a high cell safety index compared to the parent tizoxanide 3. The corresponding valine analogue 21 is considerably less active and shows higher clearance figures; the influence of the t-butyl group is therefore also significant. We note that the earlier niclosamide-derived tripeptides²¹ featured only proteinogenic amino acids, with valine analogues as the closest parallel. By contrast, the chloro-tripeptides 20 and 22 were essentially inactive and appeared toxic under the assay conditions.

Conclusions

Treatment of thiazolide amino acid esters 5a, 6, 18, and 19(13) with a base under anhydrous conditions leads to rapid rearrangement; the pseudotripeptides 8 and 20–22 can be isolated in good yields. An NMR study of 5a in D₂O/d₆-DMSO showed that the same product was slowly formed under aqueous conditions, with some competing direct hydrolysis in both cases. The rearranged products of the nitro derivatives retained antiviral activity, e.g., 8 vs 5a (whose activity is equivalent to 3), and it is possible that 8 is partly responsible for the activity of 5a. The nitro valine analogue 21 retained some antiviral activity. The DMPK data of 8 and 20–22 show reduced clearance in rat hepatocyte cells and, in some cases, in human liver microsomes compared to the prodrug esters.

While a similar rearrangement has been observed from niclosamide amino acid esters by other workers,^18,19 the difference seen with the hindered amino acid t-Leu, leading to compounds 8 and 20, is noticeable. Lower rat hepatocyte clearance is seen for these derivatives compared to the valine analogues 21 and 22. Finally, chloro derivatives 20 and 22 are virtually inactive in the antiviral assay.

Experimental Section

General

Tizoxanide 3 and RM4848 4 were supplied by Romark Pharmaceuticals. “Standard neutral workup” means that the product in organic solution was washed successively with 7% aq. citric acid, satd. aq. NaHCO₃, and water. Organic extracts were finally washed with saturated brine and dried over anhydrous Na₂SO₄ prior to rotary evaporation at <30 °C. Moisture-sensitive reactions were carried out in anhydrous organic solvents (purchased from Sigma-Aldrich) under a N₂ atmosphere. Reactions were monitored by analytical thin-layer chromatography using Merck Kieselgel 60 F₂₅₄ silica plates and were viewed under ultraviolet (UV) or by staining with KMnO₄ or iodine. Preparative flash column chromatography was performed on either VWR Prolabo silica gel or Sigma-Aldrich silica gel (particle size 40–63 Å). Mass spectra were obtained in either the electrospray mode (ES) with a Micromass LCT or the chemical ionization (CI) mode with a Micromass Trio 1000 using ammonia or methane as the carrier. Elemental analyses were performed by Mrs. Jean Ellis, of this department. ¹H and ¹³C NMR spectra were obtained using a Bruker Avance-II instrument operating at 400.20 and 101.63 MHz for ¹H and ¹³C, respectively; a Bruker Avance-IIIHD operating at 400.13 and 101.61 MHz for ¹H and ¹³C, respectively; or a Bruker Avance-1 instrument operating at 400.03 and 101.59 MHz for ¹H and ¹³C, respectively; using the stated solvent. Chemical shifts are reported in ppm (δ) relative to Me₄Si. Coupling constants (J) are reported in Hz. Analytical HPLC traces were obtained using an Agilent column monitoring at 254 nm eluted with 5% MeCN-95% H₂O containing 0.5% formic acid, at a flow rate of 1 mL/min. The purity of the tested compounds 8, 20, 21, and 22 was >95% by HPLC.

(S)-2-[(5-Nitrothiazol-2-yl)carbamoyl]phenyl-2-((tert-butoxycarbonyl)amino)-3-methylbutanoate 16

As previously described¹³ for the precursor of 5a, a mixture of Boc-Val-OH (0.21 g, 0.97 mmol) and tizoxanide 3 (0.25 g, 0.94 mmol) was stirred at 20 °C in anhydrous THF (7.5 mL). N-Ethyl-N^′-3-(dimethylamino)propyl carbodiimide·HCl (EDC; 0.19 g, 1 mmol) was added in one portion, followed immediately by 4-dimethylaminopyridine (DMAP; 0.12 g, 1 mmol). After 20 h, the mixture was filtered through Celite and the precipitate was washed with further THF and then diluted with ethyl acetate (25 mL). The combined filtrate and washings were washed with 7% aq. citric acid, saturated aq. NaHCO₃, water, and brine and then dried over anhydrous Na₂SO₄ followed by evaporation, which afforded a yellow foam; this was chromatographed on silica gel, being applied in CHCl₃ and eluted with EtOAc:hexane, 1:1. Appropriate fractions were combined and evaporated to afford the title compound 16 as an off-white solid (250 mg, 54%). Found: m/z, 487.1265. C₂₀H₂₄N₄O₇S.Na requires m/z, 487.1263; δ_H [400 MHz, CDCl₃] 1.03, 1.12 (6H, 2d, J = 6.8 Hz, Me₂CH), 1.40 (9H, s, Me₃C), 2.35 (1H, m, Me₂CHCH), 4.39 (1H, m, CHCHNH), 5.20 (1H, br m, NH), 7.40 (1H, d, J = 8.0 Hz, ArH), 7.45 (1H, t, J = 8.0 Hz, ArH), 7.67 (1H, t, J = 8.0 Hz, ArH), 8.06 (1H, d, J = 8.0 Hz, ArH), 8.18 (1H, s, thiazole 4-H), and 11.10 (1H, br s, NH); δ_C 17.9, 19.3, 28.2, 30.4, 59.7, 80.7, 123.4, 124.2, 126.7, 130.9, 134.3, 140.4, 143.5, 148.5, 155.9, 161.5, 163.5, and 170.4; m/z (ES +ve mode) 487 (MNa⁺, base peak).

(S)-{2-[(5-Nitrothiazol-2-yl)carbamoyl]phenyl}-2-amino-3-methylbutanoate, Hydrochloride 18

As described for 5a,¹³ the preceding Boc derivative 16 (0.250 g, 0.54 mmol) was suspended in CH₂Cl₂ (5 mL) and 4 M HCl in dioxane (2 mL) was added with stirring at 20 °C. A solution resulted after a few minutes, but the solid soon began to precipitate. After 16 h, the reaction was diluted with ether, briefly stirred, and then cooled to 0 °C to complete precipitation; filtration afforded the title compound 18 (0.205 g, 93%) as a light-yellow solid. Found: m/z, 387.07222. C₁₅H₁₆N₄O₅SNa (MNa⁺) requires m/z, 387.0739; δ_H [(CD₃)₂SO] 1.05 (6H, 2d, Me₂CH), 2.37 (1H, m, Me₂CHCH), 4.17 (1H, m, CHCHNH), 7.49 (1H, d, ArH), 7.53 (1H, t, ArH), 7.76 (1H, t, ArH), 7.90 (1H, d, ArH), 8.74 (1H, s, thiazole 4-H), 8.80–8.90 (3H, br s, NH₃⁺), and 13.80 (1H, br s, CONH); δ_C 18.3, 18.4, 29.6, 58.0, 123.8, 126.3, 127.2, 130.3, 134.0, 142.6, 143.0, 147.9, 162.3, 165.6, and 167.9; m/z 387 (ES +ve mode, MNa⁺ for free amine, 100%).

(S)-{2-[(5-Chloro-1,3-thiazol-2-yl)carbamoyl]phenyl}-2-amino-3-methylbutanoate, Hydrochloride 19

Boc-Val-OH (0.65 g, 3.00 mmol) and RM4848 4 (0.51 g, 2.00 mmol) in THF (7.5 mL) were coupled using EDC (0.57 g, 3.00 mmol) and DMAP (0.37 g, 3.00 mmol) at 20 °C as described for the preparation of 16. After 20 h, the mixture was filtered through Celite and worked up for a neutral product. Chromatography using a gradient of 25–40% EtOAc/hexane removed unreacted 4. Product fractions were pooled and evaporated to afford the crude product 17 (0.487 g, 54%), which was immediately progressed. This material (0.47 g, 1.04 mmol) in CH₂Cl₂ (5 mL) was treated with 4 M HCl in dioxane (3 mL) at 0 °C. Further 4 M HCl in dioxane (2 mL) was added after 2.25 h; after 6 h, the mixture was stored at 0 °C overnight. Isolation of the product was completed as described for 18 to afford the title HCl salt 19 (0.317 g, 78%); δ_H [(CD₃)₂SO] 1.05 (6H, 2d, Me₂CH), 2.36 (1H, m, Me₂CHCH), 4.09 (1H, m, CHCHNH), 7.44–7.53 (2H, m, ArH), 7.60 (1H, s, thiazole 4-H), 7.69 (1H, t, ArH), 7.89 (1H, d, ArH), 8.85 (3H, br s, NH₃⁺), and 12.50 (1H, br s, NH). This product was immediately progressed by rearrangement: see below.

(S)-N-(3,3-Dimethyl-1-((5-nitrothiazol-2-yl)amino)-1-oxobutan-2-yl)-2-hydroxybenzamide 8

A suspension of prodrug ester 5a (0.100 g, 0.24 mmol)¹³ in 1,4-dioxan (3 mL) was stirred at 20 °C and treated with triethylamine (0.04 mL, 0.29 mmol). After 0.5 h, the mixture was diluted with EtOAc (20 mL) and THF (5 mL, to improve solubility), washed with 7% aq. citric acid solution and brine, dried, and then evaporated to obtain a pale-yellow solid (0.085 g), which was chromatographed, applying in CH₂Cl₂ and eluting with EtOAc/hexane, 1:3. Appropriate fractions were pooled and evaporated to give the product 8 (0.047 g, 52%), which was recrystallized from EtOAc/hexane to give an analytical sample. Found: C, 50.8; H, 4.8; N, 14.8; S, 8.5; m/z, 401.0887; C₁₆H₁₈N₄O₅S requires C, 50.8; H, 4.8; N, 14.8; S, 8.5%; C₁₆H₁₈N₄O₅SNa requires m/z, 401.0896; δ_H [(CD₃)₂SO] 1.06 (9H, s, Me₃C), 4.73 (1H, d, J = 7.6 Hz, CHNH), 6.94 (1H, t, ArH), 6.99 (1H, d, ArH), 7.40 (1H, t, ArH), 7.92 (1H, dd, ArH), 8.66 (1H, s, thiazole 4-H), 8.97 (1H, d, J = 7.6 Hz, CHNH), 11.55 (1H, br s), and 13.43 (1H, br s); δ_C 26.8, 34.6, 61.1, 117.3, 118.3, 120.0, 131.0, 133.8, 142.5, 143.1, 157.1, 161.4, 166.0, and 171.9.

(S)-N-(3,3-Dimethyl-1-((5-chlorothiazol-2-yl)amino)-1-oxobutan-2-yl)-2-hydroxybenzamide 20

This was prepared similarly to 8 from RM5064 6 (0.404 g, 1.00 mmol),¹³ affording the desired product 20 (0.279 g, 76%) as a white solid. Found: C, 52.2; H, 4.9; N, 11.5; S, 8.4; m/z, 390.0647. C₁₆H₁₈ClN₃O₃S requires C, 52.2; H, 4.9; N, 11.4; S, 8.7%; C₁₆H₁₈³⁵ClN₃O₃S.Na (MNa⁺) requires m/z, 390.0655; δ_H [(CD₃)₂SO] 1.06 (9H, s, Me₃C), 4.85 (1H, d, J = 9.2 Hz, CHNH), 6.90 (1H, t, ArH), 6.99 (1H, d, ArH), 7.09 (1H, br d, CHNH), 7.43 (1H, t, ArH), 7.49 (1H, d, ArH), 7.72 (1H, s, thiazole 4-H), 11.87 and 12.36 (2H, 2 br s, NH and OH); δ_C 26.5, 35.6, 59.9, 113.8, 118.7, 119.0, 122.0, 125.7, 134.6, 134.8, 156.4, 161.5, 169.1, and 169.8.

(S)-2-Hydroxy-N-(3-methyl-1-((5-nitrothiazol-2-yl)amino)-1-oxobutan-2-yl)benzamide 21

This was prepared similarly to 8 from the prodrug ester 18(13) (0.041 g, 0.102 mmol), affording the desired product 22 (0.025 g, 67%) as a pale-brown solid after chromatography. Found: m/z, 387.0728; C₁₅H₁₆N₄O₅S requires m/z, 387.0739; δ_H [CDCl₃] 1.15 (6H, t, Me₂CH), 2.45 (1H, m, Me₂CHCH), 4.74 (1H, t, CHCHNH), 6.90 (1H, t, ArH), 6.98 (1H, d, ArH), 7.27 (1H, d, NH), 7.44 (1H, t, ArH), 7.54 (1H, d, ArH), 8.40 (1H, s, thiazole 4-H), and 11.38 (1H, br s, NH); δ_C 18.6, 19.4, 30.6, 59.2, 113.6, 118.6, 119.4, 126.2, 135.2, 140.3, 144.0, 160.4, 160.9, 170.4, and 170.8;

(S)-2-Hydroxy-N-(3-methyl-1-((5-chlorothiazol-2-yl)amino)-1-oxobutan-2-yl)benzamide 22

This was prepared similarly to 8 from the prodrug ester 19 (0.300 g, 0.77 mmol), affording the desired product 22 (0.153 g, 56%) as a white solid. Found: C, 50.8; H, 4.6; N, 11.8; S, 8.8; m/z, 376.0490. C₁₅H₁₆ClN₃O₃S requires C, 50.9; H, 4.6; N, 11.9; S, 9.06%; C₁₅H₁₆³⁵ClN₃O₃S.Na (MNa⁺) requires m/z, 376.0499; δ_H [(CD₃)₂SO] 0.95-0.98 (6H, 2d, Me₂CH), 2.22 (1H, m, Me₂CHCH), 4.63 (1H, t, J = 7.6 Hz, CHCHNH), 6.92–6.97 (2H, m, ArH), 7.41 (1H, t, ArH), 7.55 (1H, s, thiazole 4-H), 7.98 (1H, d, ArH), 8.90 (1H, d, J = 7.6 Hz, NH), 11.77 (1H, s), and 12.68 (1H, s); δ_C 18.9, 19.6, 30.7, 58.7, 117.2, 117.5, 118.8, 119.6, 130.1, 134.0, 136.2, 155.9, 158.4, 167.6, and 171.2.

(S)-3,3-Dimethyl-1-(2-((5-nitrothiazol-2-yl)carbamoyl)phenoxy)-1-oxobutan-2-aminium 4-methylbenzenesulfonate 5b

p-Toluenesulfonic acid (15.2 g, 20.0 mmol) was added in portions to a solution of 2-((5-nitrothiazol-2-yl)carbamoyl) phenyl (S)-2-((tert-butoxycarbonyl)amino)-3,3-dimethyl butanoate (4.0 g, 8.4 mmol) 14 in ethyl acetate (80 mL), and the reaction was stirred overnight. A white precipitate began to form within ca. 2 h. The reaction was cooled to 0 °C; then, the precipitate was filtered off, washed with diethyl ether, and dried in vacuo to afford a white crystalline solid, which was recrystallized from hot ethanol to afford the title compound 5b as a pale-yellow crystalline solid (3.07 g, 66% yield). Found: C, 49.61; H, 4.86; N, 10.13; S, 11.39; m/z 379.1073. C₂₃H₂₆N₄O₈S₂ requires: C, 50.17; H, 4.76; N, 10.18; S, 11.65; C₁₆H₁₈N₄O₅S [M + H]⁺ requires m/z, 379.1071; δ_H [(CD₃)₂SO]1.11 (9H, s), 2.30 (3H, s), 4.10 (1 H, s), 7.12 (2H, d, J = 7.6 Hz), 7.41 (1H, d, J = 8.4 Hz), 7.49–7.55 (3H, m), 7.74 (1H, t, J = 7.6 Hz), 7.86 (1H, d, J = 7.6 Hz), 8.47 (3H, br s), 8.72 (1H, s), and 13.78 (1H, br s); δ_C 21.2, 26.5, 34.0, 61.6, 123.6, 126.0, 126.6, 127.3, 128.6, 130.2, 133.9, 138.3, 142.6, 143.0, 147.7, 162.4, 165.9, and 167.6; m/z 379 (ES +ve mode, MH⁺ for amine).

(S)-3,3-Dimethyl-1-(2-((5-nitrothiazol-2-yl)carbamoyl)phenoxy)-1-oxobutan-2-aminium Methanesulfonate 5c

Methanesulfonic acid (0.33 mL, 5.08 mmol) was added dropwise at 20 °C to a solution of 2-((5-nitrothiazol-2-yl)carbamoyl) phenyl (S)-2-((tert-butoxycarbonyl)amino)-3,3-dimethyl butanoate 14 (1.04 g, 2.17 mmol) in ethyl acetate (5 mL), and the reaction was stirred overnight. The resulting white precipitate was then filtered off, washed with diethyl ether, and dried in vacuo. The crude product was recrystallized from hot ethanol to afford the title compound 5c as a white crystalline solid (0.37 g, 37% yield). Found: C, 43.02; H, 4.73; N, 11.79; S, 13.56; m/z 379.1078. C₁₇H₂₂N₄O₈S₂ requires C, 43.03; H, 4.67; N, 11.81; S, 13.51; C₁₆H₁₈N₄O₅S [M + H]⁺ requires m/z, 379.1071; ¹H NMR (400 MHz, (CD₃)₂SO) δ_H 1.11 (9H, s), 2.34 (3H, s), 4.08 (1H, s), 7.44 (1H, d, J = 8.1 Hz), 7.52 (1H, t, J = 7.6 Hz), 7.75 (1H, td, J₁ = 7.5 Hz, J₂ = 1.5 Hz), 7.86 (1H, dd, J₁ = 7.7 Hz, J₂ = 1.4 Hz), 8.48 (3H, br s), 8.71 (1H, s), and 13.77 (1H, br s). ¹³C NMR (100 MHz, (CD₃)₂SO) δ_C 26.53, 33.94, 61.61, 123.69, 126.57, 127.24, 130.19, 133.85, 142.63, 142.99, 147.72, 162.32, 165.83, and 167.62; m/z 379 (ES +ve mode, MH⁺ for amine).

Antiviral Assay

Cell Culture and Treatments

Madin–Darby canine kidney (MDCK) cells (American Type Culture Collection, ATCC) were grown at 37 °C in a 5% CO₂ atmosphere in an RPMI-1640 medium (LONZA-CAMBREX, Basel, Switzerland), supplemented with 10% fetal calf serum (FCS), 2 mM glutamine, and antibiotics. Tizoxanide (Romark Laboratories, L.C.) and derived peptides, dissolved in a dimethyl sulfoxide (DMSO) stock solution, were diluted in culture medium, added to infected cells after a one-hour virus adsorption period, and maintained in the medium for the duration of the experiment. Controls received equal amounts of the DMSO vehicle, which did not affect cell viability or virus replication.

Cell viability was determined by the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) to MTT formazan conversion assay (Sigma-Aldrich), as described.^28,29 The 50% cytotoxic dose (CC₅₀) was calculated using Prism 5.0 software (Graph-Pad Software Inc.).²⁸

Virus Preparation, Infection, and Titration

The influenza A virus (IAV) strain A/PuertoRico/8/1934(H1N1) (PR8), a prototype strain of the H1N1 IAV subtype, was utilized in this study. PR8 virus was grown in the allantoic cavity of 10 day old embryonated eggs; the virus titer was determined by a plaque assay, as described previously.^30,31

Confluent cell monolayers were mock-infected or infected with PR8 virus for 1 h at 37 °C at a multiplicity of infection (MOI) of 5 HAU (hemagglutinating units)/10⁵ cells, as described.^30,31 After the adsorption period, the viral inoculum was removed and cells were treated with different concentrations (0.1, 1, 5, 10, and 50 μg/mL) of each compound or vehicle and maintained at 37 °C in an RPMI-1640 medium containing 2% FCS for 24 h; in parallel, cell viability was determined in mock-infected cells by the MTT assay, as described above. Virus yield was determined 24 h post infection (p.i.) by HA titration, as described.³¹ Compounds’ IC₅₀ (50% inhibitory concentration) and IC₉₀ (90% inhibitory concentration) were calculated using Prism 5.0 software.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsbiomedchemau.2c00083.

• Photocopy ¹H and ¹³C NMR spectra for compounds 8, 20, 21, and 22; full time-course NMR spectra for the degradation of 5a; proposed mechanism for the rearrangement of 5a to 8; and analytical HPLC traces for compounds 8, 20, 21, and 22 (PDF)

Author Contributions

CRediT: Andrew Valentine Stachulski conceptualization (equal), investigation (equal), methodology (equal), project administration (equal), supervision (lead), writing-original draft (lead); Jean-Francois Rossignol conceptualization (equal), funding acquisition (equal), project administration (equal), supervision (equal), writing-review & editing (equal); Sophie Pate investigation (equal), methodology (equal); Joshua Taujanskas investigation (equal), methodology (equal), writing-review & editing (equal); Jonathan A. Iggo data curation (equal), formal analysis (equal), investigation (equal), methodology (equal), writing-review & editing (equal); Rudi Aerts data curation (equal), investigation (equal), methodology (equal), supervision (equal); Etienne Pascal data curation (equal), investigation (equal), methodology (equal), writing-review & editing (equal); Sara Piacentini data curation (equal), formal analysis (equal), investigation (equal), methodology (equal), writing-review & editing (equal); Simone La Frazia data curation (equal), formal analysis (equal), investigation (equal), methodology (equal), writing-review & editing (equal); M. Gabriella Santoro investigation (equal), methodology (equal), project administration (equal), supervision (equal), writing-original draft (equal), writing-review & editing (equal); Lieven Van Vooren formal analysis (equal), project administration (equal), supervision (equal), validation (equal), writing-review & editing (equal); Liesje Sintubin data curation (equal), investigation (equal), methodology (equal), software (equal), writing-review & editing (equal); Mark S Cooper data curation (equal), formal analysis (equal), investigation (equal), methodology (equal), project administration (equal), writing-review & editing (equal); Karl Swift investigation (equal), methodology (equal), project administration (equal), supervision (equal), writing-review & editing (equal); Paul M. O’Neill project administration (equal), supervision (equal), writing-review & editing (equal).

The authors declare no competing financial interest.