Methionine-specific nonreversible bioconjugation: Advancing precision protein modification

Significance

We developed a method using activated allylic bromides for selective, nonreversible modification of methionine residues under generally biocompatible conditions. Our strategy features a moderate reaction rate, high chemoselectivity, and great conjugate stability, making it well-suited for late-stage protein modification and analysis. This method offers a practical and alternative approach to current methionine bioconjugation techniques, expanding opportunities for studying and manipulating proteins in complex biological systems.

Abstract

We present a methionine-selective, nonreversible bioconjugation strategy that employs activated allylic bromides under mild, aqueous reaction conditions compatible with various peptides and proteins. Compared with conventional allylic bromides, our method improves conjugate stability and suppresses nonspecific reactivity under the examined reaction conditions. This method enables methionine-preferred labeling of peptides and proteins, and provides proof-of-concept applications in covalent inhibitor design and protein functionalization. As a complementary addition to existing methionine bioconjugation strategies, this chemistry expands the toolkit available for protein modification and chemical biology research.

Recent advances in small-molecule–protein conjugates have driven the development of bioconjugation methods targeting native amino acids (1–4). While significant progress has been made in lysine and cysteine bioconjugation (5–11), challenges persist due to lysine’s high surface abundance, which complicates selective modification, and the need for prereduction of cysteine disulfide bonds to generate free thiols. These limitations have shifted research toward less common amino acids, such as methionine, tyrosine, tryptophan, and histidine, which offer unique opportunities for site-specific modification (12–16).

Methionine, a biologically essential yet relatively rare amino acid, has emerged as a promising target due to its integral roles in protein synthesis, antioxidant defense, and other cellular processes (17). Its chemical properties make it suitable for the development of next generation therapeutic and diagnostic tools (18). Despite its potential, methionine bioconjugation methods remain significantly underdeveloped compared to cysteine, largely due to methionine’s weak nucleophilicity, which poses challenges for achieving both high reactivity and selectivity (19, 20).

Nevertheless, methionine’s susceptibility to oxidation has enabled innovative redox-based bioconjugation methods. Chang and Toste introduced the ReACT (redox-activated chemical tagging) strategy, employing strained oxaziridine reagents for selective methionine oxidation, with applications in pharmaceuticals and diagnostics (21–23). MacMillan advanced this field with a photoexcited lumiflavin system that facilitates single-electron sulfur oxidation and α-deprotonation, generating α-thio radicals for precise modifications on the methyl group instead of the sulfur atom (24, 25). Recently, Ball and colleagues broadened the toolkit by using (diacetoxyiodo)benzene to form sulfoximines in partially aqueous media (26).

Despite these advancements, challenges in selectivity, efficiency, and biocompatibility persist, emphasizing the need for further innovation. Among the simpler strategies, alkylation of nucleophilic amino acid side chains—such as using allylic bromides to form methionine sulfonium salts—remains widely used and attractive for its effectiveness (27, 28). Gaunt’s innovative hypervalent iodine reagent selectively targets methionine, efficiently generating stable diazo sulfonium adducts that enable further functionalization (29). Similarly, Deming’s chemoselective alkylation methods addressed certain limitations but highlighted the need for nonreversible strategies, particularly in applications like antibody–drug conjugates, where preventing premature payload release is essential (30–33). (Fig. 1A). Persistent issues, including slow reaction rates, poor selectivity, and stability in cysteine-rich environments, as well as limited biocompatibility, continue to restrict broader applicability in protein labeling and underscore the urgency for improved methodologies.

Fig. 1.

The development of methionine-selective bioconjugation methods. (A) Existing methionine functionalization methods. (B) Activated allylic bromides as methionine-selective biocompatible click reagents for peptide and protein modification.

With our continuing interest in amino acid-selective bioconjugation and to enrich the toolbox of methionine-selective labeling via simple alkylation strategies, we investigated activated allylic bromides as methionine-specific reagents under biocompatible conditions. This approach achieves efficient and selective methionine bioconjugation and yields highly stable conjugates, addressing the challenge of reversibility observed with conventional alkylation methods. By fine-tuning the electronic properties of allylic bromides, we successfully suppressed reversible substitution and minimized interference from competing biomolecules such as thiols. These advancements complement existing redox- and photo-mediated methodologies and offer a practical, mild, and modular platform for stable protein functionalization relevant to biochemical and therapeutic research.

Results

To mimic the chemical environment of methionine within peptide and proteins and to facilitate product analysis and purification, we initially employed N-Cbz-methionine methyl ester (1a, 0.1 mmol) as a suitable methionine mimic and examined its reaction with activated allylic bromide (2a, 1 equiv.) in a phosphate buffer (PB) solution (pH 7), using acetonitrile (CH₃CN) as a cosolvent in a 5:2 ratio at 25 °C (Table 1). This reaction afforded the conjugated product (3aa) in good yield. To enhance efficiency and assess the reaction’s compatibility, we systematically investigated various solvent systems, reaction times, temperatures, and atmospheric conditions. Notably, the desired product was also obtained in high yields in aqueous media, with or without additives such as potassium phosphate (K₃PO₄) or formic acid (FA). However, when the reaction was conducted in buffer alone or when CH₃CN was replaced with organic solvents, such as MeOH, EtOH, IPA, DMSO, and 1,4-dioxane, the product yield decreased significantly (Table 1). Upon careful analysis of these low-yield entries, no obvious by-products were detected. The reduced efficiency observed with alternative solvent systems is therefore attributed primarily to insufficient solubility of the allylic bromide substrates, whereas acetonitrile serves as an effective cosolvent to maintain adequate electrophile solubility under aqueous reaction conditions. Notably, extending the reaction time for these entries to 4 h (8 h for entry 5) and increasing the reaction temperature to 37 °C afforded moderate to good yields of 3aa (SI Appendix, Fig. S1), accompanied by a slight increase in the proportion of the Z-isomer of 2a, likely due to partial isomerization of the allylic bromide under prolonged or higher temperature conditions. These results further underscore the critical role of CH₃CN as a cosolvent in promoting the reaction efficiency. By increasing the reaction temperature to 37 °C, we established the optimal conditions, achieving excellent yields of up to 87%.

Table 1.

Optimization of the reaction conditions

^*Reactions were conducted under the indicated conditions on a 0.1 mmol scale, using 1 equiv. of 2a and 1 mL solvent (corresponding to a 0.1 M concentration of 1a).

^†Reactions were monitored by ¹H NMR by using DMF as the internal standard and the yield were reported as the mean ± SD of three independent experiments.

With the optimized reaction conditions established, we first examined the reactivity of methionine (1a) with various allylic bromides, as summarized in Table 2. When the simple allylic bromide (2b) was used as the substrate, the corresponding product (3ab) was obtained in only 37% yield. In contrast, substrates such as simple allylic esters, amides, and ketones afforded excellent yields (3ac–3ae). When the β-substituent of activated allylic bromide was changed from an isopropyl to a methyl or phenyl group, the reaction also proceeded smoothly, giving the desired products (3af–3ag) in good yields. However, for substrate 2g, the low yield likely due to its poor solubility in the reaction solvent. Furthermore, preliminary stability studies revealed that the presence of a β-substituent on the activated allylic bromide significantly enhances the stability of the corresponding products toward various nucleophiles (SI Appendix, Table S1 and Fig. S2). In particular, 3aa, bearing an isopropyl group at the β-position, exhibits notably high stability across different nucleophiles.

Table 2.

Reactivity of various allylic bromide

^*Reactions were conducted under the indicated conditions on a 0.1 mmol scale, using 1 equiv. of 2# and 1 mL solvent (corresponding to a 0.1 M concentration of 1a).

^†Reactions were monitored by ¹H NMR by using DMF as the internal standard and the yield were reported as the mean ± SD of three independent experiments.

To further demonstrate the practical potential of the methionine-selective bioconjugation methodology, we investigated the chemoselectivity of the reagent and the stability of the model substrate 2a as well as the resulting products under various conditions. The results are summarized in Fig. 2. Initially, under the optimized reaction conditions, we successfully scaled up the reaction to a 5 mmol scale, affording product 3aa in consistently high isolated yield (Fig. 2A). Subsequently, chemoselectivity studies revealed that the methionine conjugate could be efficiently obtained even in the presence of other protected nucleophilic amino acids such as serine, tyrosine, tryptophan, and glutamic acid, with no significant decrease in yield. In particular, in the cases of lysine and cysteine, we further examined the reactivity of 2a toward N-terminal protected cysteine and lysine under different conditions (SI Appendix, Fig. S7 and S8). The results showed that at pH 7, 2a reacted only slightly with Cys (Boc-Cys-OMe), and did not react with Lys (Ac-Lys-OMe); when 1a (Cbz-Met-OMe) was simultaneously present, extending the reaction time to 4 h still afforded 3aa in good yields (75 and 82% respectively). In contrast, when the solvent system H₂O:CH₃CN = 5:2 was used with 1 equiv. of FA as an additive (pH = 2.7), 2a hardly reacted with Cys, and after 2 h, 3aa was obtained in high yield (86 and 88% respectively) with only trace (<1%) Cys-related by-product, corresponding to a >80-fold selectivity for 1a over Cys.

Fig. 2.

(A) Gram-scale synthesis of 3aa (isolated yield from three independent experiments, concentration of the reactants: 0.1 M). (B) Chemoselectivity of 2a in the presence of various nucleophilic amino acids. The reaction was conducted on a scale of 0.1 M in H₂O:CH₃CN = 5:2, FA (1 equiv, pH = 2.7), air, 2 h. (C) Stability of 3aa in diverse solutions (10 mM 3aa, 50 mM H₂O₂; FBS, fetal bovine serum; GSH: glutathione, 20 mM in PB pH = 7 buffer; 24 h) and nucleophiles (1 equiv. of nucleophiles as additive in pH = 7 buffer for 12 h. THU: thiourea; DTT: dithiothreitol. Amino acids Cys, Trp, Glu, Ser, Tyr, Arg, and Lys were used in their free amino acid forms). (D) Second-order rate kinetics, k = (5.8 ± 0.089) × 10⁻³ M⁻¹s⁻¹. (E) Reaction pathway studies. Data are reported as mean ± SD, calculated from at least three independent experiments. All reactions were performed at 37 °C. Statistical significance in (B–D) was calculated via one-way ANOVA; ns indicates no significance, P > 0.05; *P < 0.05; **P < 0.01.

Regarding product stability, 3aa was exposed to various conditions including buffer solutions at different pH values, H₂O₂ solution, FBS, and GSH solution. After stirring for 24 h, 3aa remained remarkably stable, with no significant degradation observed (Fig. 2C). Moreover, 3aa retained high stability in the presence of strong nucleophiles or nucleophilic amino acids, with no appreciable formation of side products observed—except in the case of 2-pyridinethione (PyS), where slow degradation occurred; nevertheless, the conjugate still exhibited a half-life of over 8 h (Fig. 2C and SI Appendix, Fig. S3). Finally, kinetic studies indicated that the reaction follows second-order rate kinetics, with a rate constant of (5.8 ± 0.089) × 10⁻³ M⁻¹s⁻¹ (Fig. 2D). Although moderate in magnitude, the reaction also proceeds smoothly at lower concentrations; for example, at 0.01 M, good conversion was still achieved by extending the reaction time to 4 h or using 2 equiv. of 2a (SI Appendix, Fig. S4).

Then, to elucidate the reaction mechanism (Fig. 2E and SI Appendix, Figs. S9 and S10), we examined deuterium-labeled substrate 2d and compared the reaction outcomes with β-methyl (1f) and β-isopropyl (1a) substituted analogs. The isotopic distribution of 2d-D1 and 2d-D2 remained unchanged after reaction, indicating a direct SN2 displacement rather than an allylic addition (SN2′) pathway. Trace (<1%) SN2′ products were detected only for the β-methyl substrate, while none were observed for the β-isopropyl analog (Fig. 2E), confirming that the reaction proceeds via direct SN2 substitution, especially when sterically hindered substituents are present at the β-position.

We next investigated the reactivity of methionine substrate 1a with a diverse set of 2-amido allylic bromides, as illustrated in Fig. 3. Initially, we fixed the β-substituent as an isopropyl group and explored the substrate scope by varying the amide (R¹) group. As shown in Fig. 3A, substrates bearing simple alkyl groups (3ah–3aj) underwent efficient conjugation, affording the desired products in good yields. To further assess the compatibility of β-substituted allylic bromides with different functional groups and expand their potential as biocompatible linkers, we explored a broader substrate scope incorporating diverse biocompatible functionalities. Substituents such as carbamate, ester, and vinyl groups (3ak–3am), along with biocompatible click handles such as alkynyl and azido groups (3an–3ap), were well tolerated, yielding the corresponding conjugates in good to excellent yields (73 to 93%). Notably, when dopamine was introduced as a substituent, the reaction proceeded smoothly, affording the corresponding product in up to 82% yield, demonstrating good compatibility with this endogenous molecule. A similar trend was observed when aryl-substituted substrates (3ar–3aw) were examined. Notably, substrates containing a Suzuki coupling handle (3at and 3au) or an aldehyde group (3av) successfully yielded the desired conjugation products in moderate to good yields, further highlighting the versatility of this methodology.

Fig. 3.

Scope of β-substituted allylic bromide. All the reactions were conducted in a scale of 0.1 M, and the typical reaction condition is 1a (0.1 mmol), 2# (0.1 mmol), PB pH = 7:CH₃CN = 5:2 (1 mL), air, 37 °C, 2 to 4 h. Data are reported as the average isolated yields from three independent experiments.

Next, we expanded our investigation of the R² substrate scope (Fig. 3B) and sought to extend the linker into a multifunctional handle (Fig. 3C). The results shown in Fig. 3B indicate that the reaction exhibits good tolerance toward various β-substituents. Substituent groups such as methyl, cyclohexyl, naphthyl, and phenyl rings, as well as substituted phenyl groups (3ax–3aac), were well accommodated, yielding the corresponding methionine conjugates in 73 to 86% yields. Additionally, the reaction tolerated functional groups such as carboxyl and alkynyl at the R² position, affording conjugation products 3aad and 3aae in 66 and 64% yields, respectively.

Building on our findings, we aimed to introduce distinct functional groups at the R¹ and R² positions to develop multifunctional handles capable of addressing complex applications. As expected, substrates incorporating two functional groups within a single molecule underwent smooth conjugation, yielding the desired products with efficiencies ranging from 66 to 71% (3aaf–3aah).

To demonstrate the potential of this methodology in bioconjugation, we further examined methionine-containing peptide and protein substrates, as illustrated in Fig. 4. We initially focused on simple dipeptides and tripeptides, which, under the previously optimized reaction conditions, yielded the desired conjugates with high efficiency (3ba, 84%; 3ca, 87%). Next, we evaluated the reactivity of a hexapeptide, argireline, which contains a mid-chain methionine and is widely recognized as the first acetyl hexapeptide-8 introduced in the cosmetic industry to mimic the effects of botulinum toxin. Under the standard reaction conditions, the conversion rate was limited to 28%. However, extending the reaction time to 12 h significantly improved the conversion to 72%. Further optimization (SI Appendix, Table S2) revealed that using a solvent mixture of H₂O:CH₃CN (5:2) with 5 mM FA as an additive facilitated efficient conjugation, affording the argireline-2a conjugate as the main product with a conversion rate up to 80%. Subsequently, stability studies of 2a in different buffers and aqueous media (SI Appendix, Fig. S5) revealed that, while 2a remained stable under neutral conditions, gradual hydrolysis occurred in mildly acidic environments. In particular, incubation in 5 mM FA for 12 h resulted in approximately 30% hydrolysis of 2a, consistent with the need to use excess reagent (5 equiv.) in peptide conjugation reactions. Furthermore, the stability profile of 2a toward GSH under near-neutral conditions suggests that, in certain intracellular or thiol-rich environments, careful consideration of buffer composition and pH is required to minimize undesired side reactions.

Fig. 4.

Scope of methionine-containing peptides and proteins. Standard condition is 1# (200 µM), 2a (1 mM, 10 to 20 equiv.), H₂O:CH₃CN = 5:2 (1 mL), FA (5 mM, pH = 3.6), 37 °C, air, 12 to 18 h; ^aFor entries 1b and 1c, reaction condition is: 1# (0.1 mmol), 2a (1 equiv.), PB pH = 7:CH₃CN = 5:2 (1 mL) 37 °C, air, 2 h; ^bFor entries 1i–1m, the concentration of proteins were indicated in each entry, reaction solvent was H₂O:CH₃CN = 9:1 (1 mL), 37 °C, 18 h. ^cConversion rate refers to the fraction of protein molecules detected by MS as modified under the specified reaction conditions. ^dThe LAR value, calculated as described in the SI Appendix, is reported as the average LAR derived from the MS peak distribution, with ~34% unmodified antibody and ~66% bearing one to four ligands. Data are reported as the average isolated yields or conversion rate from three independent experiments.

Under these optimized conditions, the reaction also proceeded smoothly with various peptides, including tetrapeptide 1e, pentapeptide 1g, octapeptide 1f, and nonapeptide 1h, affording the desired methionine-selective conjugates 3ea–3ha (Fig. 4A). For substrates 1e, 1g, and 1h without N-terminal acetyl protection, we also evaluated the reaction at pH 7. For substrates 1e and 1g, no significant overalkylated products were observed after 24 h. In the case of 1h, approximately 7% of an overalkylated product was detected after 24 h, while the desired conjugate 3ha remained the major product, obtained in about 6 - fold higher yield than the overalkylated species (SI Appendix, Figs. S15, S19, and S22).

To broaden the scope of our methionine-selective modification methodology, we examined a range of biofunctional proteins as substrates (Fig. 4B). Myoglobin, a key oxygen-binding protein (34, 35), was efficiently conjugated with 2a in a methionine-selective manner, affording the myoglobin–2a conjugate (3ia) in 72% yield with a mono/di ratio of 67:33 (SI Appendix, Figs. S24–S27). CD and LC–MS analyses confirmed that the protein structure remained intact, and its peroxidase-like activity was largely preserved (>80% of the native enzyme, SI Appendix, Fig. S32), indicating mild and biocompatible conditions. This strategy was further applied to RNase A and Cytochrome C, providing conjugates 3ja and 3ka in 58 to 67% yields. Both retained their native folds as verified by CD and LC–MS analyses (SI Appendix, Figs. S33–S39), and RNase A maintained comparable RNA-cleaving activity to the unmodified enzyme.

Encouraged by these results, we extended the study to larger proteins such as transferrin (3la) and trastuzumab (3ma). After 18 h, transferrin was nearly fully converted (Fig. 4), and tandem MS revealed selective modification at several solvent-accessible methionine residues near its active pocket, with CD data confirming structural integrity (SI Appendix, Figs. S40–S44). In these protein systems, CH₃CN served as an effective cosolvent, enhancing reaction efficiency through improved substrate solubility, while maintaining high selectivity under mild conditions. The trastuzumab conjugate showed a ligand-to-antibody ratio (LAR) of 1.95, demonstrating good control of the conjugation process under biocompatible conditions. Collectively, these results highlight that our method enables efficient and chemoselective methionine modification across diverse proteins—from small models to complex biomacromolecules—while preserving their native structure and function, underscoring its robustness and broad potential for late-stage bioconjugation.

Extensive methodological studies have established the foundation for applying this strategy to a wider range of biological and pharmaceutical contexts. To demonstrate its practical utility (Fig. 5), we first applied our methodology to medicinal chemistry. Methionine-targeted covalent inhibitors represent a promising avenue for drug discovery and functional protein studies, particularly in complex biological systems. Previous studies have shown that covalent targeting of the BRD4 BD1 Met149 residue effectively disrupts BRD4-mediated cell signaling and significantly inhibits the proliferation of associated cancer cells (36). In our study, the allylic bromide-based compound 2ai exhibited concentration-dependent inhibitory activity toward BRD4 BD1 high-expression tumor cells (Fig. 5 B and C). Specifically, 2ai showed an IC₅₀ of 2.1 ± 0.3 μM against the H2228 lung cancer cell line, while its precursor 2ai–3 (IC₅₀ = 11.1 ± 2.7 μM) and model compound 2a (IC₅₀ = 24.8 ± 2.4 μM) displayed weaker effects (SI Appendix, Fig. S52), highlighting the significant potential of 2ai for further development as a lead compound.

Fig. 5.

Application of Methionine-selective bioconjugation methods. (A) The depiction of BRD4 BD1 covalent inhibitor. (B) Cell viability assay of BRD4 BD1 covalent inhibitor 2ai on BRD4 high expression cancer cell lines H2228, Molm 13 and MV4-11, IC₅₀= mean ± SD, n = 3. (C) Colony formation assay of 2ai on cancer cell line H2228. (D) Application of the methionine-selective modification methodology in protein fluorescent labeling; Myo: Myoglobin; “+:” low concentration; “++:” high concentration; “−:” not added.

Protein fluorescence labeling is widely employed in protein research, molecular biology, and drug screening due to its high sensitivity and versatility. Our method exhibits excellent functional group compatibility, allowing myoglobin to serve as a model substrate for fluorescence labeling with chemical dyes possessing diverse photophysical properties (Fig. 5D). This strategy enables rapid fluorescence detection via simple protein gel electrophoresis, regardless of whether low or high dye concentrations are employed. These findings further validate our methodology as a valuable tool for protein fluorescence labeling and broader applications in biomolecular research.

Discussion

In conclusion, we have developed a methionine-specific, nonreversible bioconjugation method using activated allylic bromides under biocompatible conditions. This method overcomes the limitations of conventional allylic bromide–based strategies, including poor biocompatibility, slow reaction rates, and limited product stability, by enabling efficient conjugation under mild conditions with excellent functional group tolerance. It is applicable to diverse contexts such as covalent inhibitor design and protein fluorescent labeling. Moreover, compared with random lysine modification, methionine-targeted conjugation offers greater site control and lower heterogeneity due to the low abundance and defined structural positioning of methionine residues. Our method also preserves protein integrity and activity, as demonstrated for myoglobin, RNase A, and transferrin. Overall, this work establishes a robust and broadly applicable platform for selective and biocompatible protein modification, providing tools and strategies for methionine-targeted bioconjugation chemistry.

Materials and Methods

All chemicals were purchased from BLDpharm, TCI Chemicals, Sigma-Aldrich, or MedChem Express unless otherwise specified. Deuterated solvents were purchased from Cambridge Isotope Laboratories, Inc. and directly used. Unless otherwise noted, reagents were commercially available and used without further purification. All reactions were conducted under indicated atmosphere by using standard Schlenk techniques. All glassware and vials were oven-dried prior to use. All work-up and purification procedures were carried out with reagent-grade solvents.

Analytical thin-layer chromatography (TLC) was performed using glass plates precoated with 0.25 mm Yantai silica plates (GF-254). The developed chromatography was analyzed by UV lamp (254 nm and 365 nm). Flash chromatography was performed on 300 to 400 mesh Yantai silica gel with the indicated eluents. All the reactions were monitored by TLC (for bromide substrates), ¹H NMR or LCMS (Shimadzu LC2030, solvent system: A, H₂O with 0.1% FA; B, CH₃CN with 0.1% FA; Column: Shim-pack Scepter, C 18 to 120, 5 µm, 4.6 × 250 mm; Method: 5 to 90% B over 15 min, then hold 7 min, then 90 to 5% B over 2 min, hold 1 min, 1 mL/min) as appropriate. The peptide conjugates were purified by prep-HPLC (Shimadzu LC20AR, solvent system: A, H₂O with 0.1% FA; B, CH₃CN with 0.1% FA; Column: Shim-pack Scepter, C 18 to 120, 5 µm, 10.0 × 250 mm; Method: 5 to 95% B over 100 min, then hold 10 min, then 95 to 5% B over 5 min, hold 5 min, 2.5 mL/min) as appropriate. MS analysis was conducted on an Agilent 7890A GC System with an Agilent 5975C MSD or Bruker SCION TQ with Bruker 456 GC System.

NMR (¹H NMR, ¹³C NMR) spectra were recorded respectively at 400 MHz and 101 MHz in indicated deuterated solvents [CDCl₃, DMSO-d₆, CD₃OD, CD₃CN, (CD₃)₂CO]. Fluorine NMR spectra (¹⁹F NMR) were recorded at 376 MHz, and the chemical shifts were accurate to one decimal place or two decimal places to help distinguish overlapping peaks. Chemical shifts were expressed in parts per million (ppm) downfield from tetramethylsilane and refer to the solvent signals [D₂O: δ H 4.79 ppm; CDCl₃: δ H 7.26 and δ C 77.16 ppm; DMSO-d₆: δ H 2.50 and δ C 39.50 ppm, CD₃OD: δ H 3.31 and δ C 49.00 ppm, CD₃CN: δ H 1.94 and δ C 118.26 ppm, (CD₃)₂CO: δ H 2.05 and δ C 29.84 ppm]. The signals of water were observed at about 1.58 ppm in CDCl₃ and 3.33 ppm in DMSO-d₆, respectively. Coupling constants (J) were reported in Hertz (Hz). Splitting patterns were designated as s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet; br, broad; dd, doublet of doublets, etc. High-resolution mass spectrometry (HRMS) data were obtained on a Waters LC-TOF mass spectrometer (Xevo G2-XS Q-Tof) using electrospray ionization (ESI) in positive mode (for more detailed data, see the SI Appendix).

Supplementary Material

Appendix 01 (PDF)

Data, Materials, and Software Availability

All study data are included in the article and/or SI Appendix.