Cyclic Nucleotide-Gated Channel Block by Hydrolysis-Resistant Tetracaine Derivatives

Abstract

To meet a pressing need for better cyclic nucleotide-gated (CNG) channel antagonists, we have increased the biological stability of tetracaine-based blockers by synthesizing amide and thioamide linkage substitutions of tetracaine (1) and a higher affinity octyl tail derivative (5). We report the apparent K_D values, the mechanism of block, and the in vitro hydrolysis rates for these compounds. The ester linkage substitutions did not adversely affect CNG channel block; unexpectedly, thioamide substitution in 1 (compound 8) improved block significantly. Furthermore, the ester linkage substitutions did not appear to affect the mechanism of block in terms of the strong state preference for closed channels. All ester substituted compounds, especially the thioamide substitutions, were more resistant to hydrolysis by serum cholinesterase than their ester counterparts. These findings have implications for dissecting the physiological roles of CNG channels, treating certain forms of retinal degeneration, and possibly the current clinical uses of compound 1.

Introduction

Cyclic nucleotide-gated (CNG)^a ion channels are known for their role in phototransduction in retinal photoreceptors and in odorant transduction in the olfactory epithelium.^1,2 CNG channels are also present in other brain regions and non-sensory tissues, but their physiological roles are much less clear.^3–7 CNG channel activation in photoreceptors is regulated by the cytoplasmic concentration of cGMP, which binds to and opens the channel to allow influx of Na⁺ and Ca²⁺ ions. Alterations of CNG channel activity have been observed in some forms of retinitis pigmentosa, a group of inherited diseases that cause progressive degeneration of rod and cone photoreceptors.^8–14 Mutations that cause elevated cGMP levels lead to prolonged channel activation and Ca²⁺-triggered cell death.^{10,12,14–17} In mouse models, reduction of CNG channel activity strongly correlated with improvements in the overall progression of the disease^18–20 (see also^21,22), but unfortunately, there are no clinically approved drugs that target CNG channels.

Compared to voltage-gated channels, CNG channel pharmacology is quite unsophisticated.^20,23 The most widely used CNG channel antagonist in research, l-cis-diltiazem, is an incomplete blocker and relatively non-specific.^24–27 CNG channels are also blocked by some local anesthetics, one example being tetracaine [2-(dimethylamino)ethyl 4-(butylamino)benzoate] (1).^28–30 Compound 1 blocks CNG channels with relatively high affinity, although differently than voltage-gated sodium channels. Like sodium channels, the interaction of 1 with CNG channels is thought to be in the selectivity filter and the pore region.^31–35 However, 1 has a greater affinity for the sodium channel in its open, inactivated-state;³⁶ whereas for CNG channels, it binds with higher affinity to the closed state.³⁷ The inability of 1 and l-cis-diltiazem to distinguish between CNG channels and other channels leads to the conclusion that much research remains to be done on better CNG channel antagonists.

In recent years, our group has developed a number of high affinity CNG channel blockers using 1 as a scaffold.^38–40 Compound 1 is clinically approved for temporary anesthesia in various surgical procedures, including those involving the eye.^41,42 Its effects are localized and shortlived due to its rapid degradation by esterases.⁴³ One major improvement in the design of a tetracaine-based CNG channel blocker would be to increase its biological stability against hydrolysis. In this study, we have synthesized amide and thioamide linkage versions of 1, as well as its relative 5 that has a higher affinity for CNG channels.⁴⁰ We report the apparent K_D values and in vitro hydrolysis rates by serum cholinesterase (butyrylcholinesterase) for these novel compounds. We believe we have uncovered promising compounds not only for CNG channel research, but for the treatment of specific forms of retinal degeneration as well. Furthermore, these improvements may have implications for current clinical usage of 1 in general.

Chemistry

Compound 1 derivatives were prepared according to Scheme 1. An alkyl substituent was added to the amino end of 4-aminobenzoic acid (2) via reductive amination using a synthesis adapted from Sato et al.⁴⁴ The resulting alkylated benzoic acid derivatives (3 and 4) were then activated at the carboxylic acid with 1,1'-carbonyldiimidazole (CDI) and subsequently esterified or amidated using 2-(dimethylamino)ethanol or N′,N′-dimethylethane-1,2-diamine, respectively, to yield target compounds 5–7.⁴⁵ Compounds 6 and 7 were further treated with Lawesson’s reagent to yield target thioamide compounds 8 and 9.⁴⁶

Scheme 1. Synthesis of Tetracaine Derivativesa.

^aReagents and conditions: (i) butanal or octanal, α-picoline-borane, MeOH, room temp, 16–24 h (88%, 94%); (ii) CDI, DME, 60 °C, 2 h, followed by 2-(dimethylamino)ethanol, NaH, 60 °C to room temp, 16–24 h (95%); (iii) CDI, DME, 60 °C, 2 h, followed by N′,N′-dimethylethane-1,2-diamine, NaH, 60 °C to room temp, 16–24 h (93%, 87%); (iv) Lawesson’s reagent, toluene, reflux, 2.5 h (26% 73%).

Results

CNG Channel Block at Saturating cGMP

The effectiveness of retinal rod CNG channel current block by the ester linkage-substituted tetracaine derivatives was tested in Xenopus oocyte preparations. Excised, inside-out patches pulled from oocytes expressed heteromeric rod CNG channels consisting of CNGA1 and CNGB1 subunits. This was verified by substantial block of 2 mM cGMP-induced currents with 20 µM l-cis-diltiazem (74.2 ± 5.8 % at V_m = +50 mV).⁴⁷

Each compound’s apparent affinity for the heteromeric CNG channel was determined under maximal channel activation (2 mM cGMP). CNG channel currents were elicited by a voltage step protocol to −50 mV and +50 mV (Figures 1A and 2A, insets). The apparent K_D value at each membrane potential was estimated first by determining I_+B and I_−B at steady-state, where I_+B is the current in the presence of blocker and I_−B is current in the absence of blocker, for different blocker concentrations ([B]). The following equation for block at a single binding site was fit to the data to obtain K_D:

$$I_{+\mathrm{B}}=I_{-\mathrm{B}}{K}_{\mathrm{D}}/(K_{\mathrm{D}}+[B])$$

Figure 1.

Heteromeric rod CNG channel block by compounds 1 and 8. Leak-subtracted currents from a representative excised inside-out patch from oocytes expressing heteromeric rod CNG channels (A). Currents were elicited by a voltage step protocol from 0 to −50 to +50 mV (inset) in the presence of 2 mM cGMP or 2 mM cGMP and compound (5 µM). Time scale is shown in the lower right hand corner of the panel and inset, and the zero current level is indicated by the dotted line. (B) Currents obtained from a concentration series of compounds 1 (white) and 8 (black) are plotted against compound concentration. Solid line indicates the fit of the equation for block at a single binding site (see text). K_D values determined from the fit of the equation were 4.2 µM at +50 mV and 15.6 µM at −50 mV for compound 1, and 0.5 µM at +50 mV and 3.7 µM at −50 mV, for compound 8.

Figure 2.

Heteromeric rod CNG channel block by compounds 5 and 9. Leak-subtracted currents from a representative excised inside-out patch from oocytes expressing heteromeric rod CNG channels (A). Currents were elicited by a voltage step protocol from 0 to −50 to +50 mV (inset) in the presence of 2 mM cGMP or 2 mM cGMP and compound (2 µM). Time scale is shown in the lower right hand corner of the panel and inset, and the zero current level is indicated by the dotted line. (B) Currents obtained from a concentration series of compounds 5 (white) and 9 (black) are plotted against compound concentration. Solid line indicates the fit of the equation for block at a single binding site (see text). K_D values determined from the fit of the equation were 1.5 µM at +50 mV and 1.8 µM at −50 mV for compound 5, and 1.3 µM at +50 mV and 1.8 µM at −50 mV, for compound 9.

Representative traces for CNG channel currents at positive and negative membrane potentials activated by 2 mM cGMP, as well as graphs with fitted curves are shown for compounds 1, 8, 5 and 9 in Figures 1 and 2. As reported previously,⁴⁰ 5 had a higher affinity for CNG channels than 1 (Figures 1 and 2). Compound 8 proved to be a higher affinity CNG channel blocker compared to compound 1 at both positive and negative membrane potentials (Figure 1). In contrast, compound 9 with the same head group linkage as compound 8 had a similar CNG channel affinity compared to compound 5 (Figure 2). From Figures 1A and 2A, it is also possible to note the characteristic voltage-dependence of block of compounds 1 and 5, which improves with positive membrane potentials. All compounds tested exhibited voltage-dependent block, although the voltage-dependence of 5 and its derivatives was appreciably less than that of 1 and its derivatives.

A small transient decay in current attributed to an ion accumulation effect was seen with each voltage step with large currents (typically >1 nA; Figures 1 and 2).⁴⁸ Corrections for ion accumulation did not change substantially previous K_D estimates for 1 (6.8 µM corrected for ion accumulation,³⁹ and 6.7 µM not corrected,⁴⁰ both determined at +40 mV). We found that the K_D value estimates taking this effect into account did not change the relationship amongst the compounds for block, hence it was not corrected for in the values reported here.

The K_D values at both +50 mV and −50 mV for all compounds tested are plotted in Figure 3. The mean K_D values are summarized in Table 1, along with estimated log P values for each compound. Based on these results, it is apparent that the head group linkage of 5 has a limited role in CNG channel current block. However, substitutions of the head group linkage in 1, in particular the carbonyl oxygen, play a direct role in CNG channel interaction. While the amide substitution of the ester linkage of 1 (compound 6) has little effect on the K_D values, the thioamide substitution unexpectedly improves the effectiveness of block (compound 8).

Figure 3.

K_D values determined from all experimental patches. Plots show all K_D values determined at +50 mV (upper panel) and −50 mV (lower panel). Solid horizontal brackets with asterisks indicate groups significantly different than compound 1 using the Holm-Sidak method for multiple pair-wise comparisons, P-values <0.01.

Table 1.

K_D Value Determinations, Estimated Log P Values, and In Vitro Serum Cholinesterase Hydrolysis Rates for Tetracaine (1) and Derivatives.

Compounda	Structure	KD(+50) (µM)	KD(−50) (µM)	n	Log Pb	Hydrolysis rate nmols.min−1.mg−1
1		4.9 ± 1.8	21.8 ±8.6	16	3.1	132 ±10
6		4.4 ±2.0	28.6 ± 13.1	7	2.3	8.4 ±3.5
8		0.6 ±0.3	3.9± 1.8	4	2.8	NDc
5		1.0 ±0.6	2.4 ±1.4	5	5.1	213 ± 31
7		1.6 ± 1.5	3.7 ±1.5	4	4.3	1.2 ±0.5
9		0.7 ±0.5	1.8 ±0.8	5	4.8	NDc

^aThe compounds are shown here in what is expected to be the predominant protonation state at pH 7.6.

^bCalculated for the unprotonated forms using ALOGPS 2.1 (Virtual Computational Chemistry Laboratory).

^cNot Detected.

State Dependence of Block

Compound 1 has been reported to preferentially block CNG channels in the closed conformation, and the effectiveness of block improves with half-maximal channel activation.³⁷ Although we previously established that 5 has a greater affinity for CNG channels under saturating concentrations of cGMP than 1, we did not examine if the mechanism of block was different. We sought to determine if the ester linkage-substituted compounds, as well as compound 5, exhibit state-preference for block. We measured the apparent K_D values at both positive and negative potentials for CNG channel currents activated by saturating (2 mM) and sub-saturating concentrations of cGMP (50 and 100 µM). Under these conditions, with no exceptions, the determined K_D values for all compounds were smaller at sub-saturating cGMP than at saturating cGMP. The K_D values at sub-saturating cGMP (at +50 mV) normalized to the K_D values at saturating cGMP are plotted against 1 – I/I_max, which is related to the fraction of closed channels, for compounds 1, 8, 5, and 9 in Figure 4. The solid lines are simulations for the expected relationship between K_D and fraction of closed channels for an exclusive closed channel blocker, using three previously determined estimates for the probability of closed heteromeric retinal rod channels at saturating cGMP (a–c).^49–51 The dotted line is a simulation for a blocker with no preference for state, while the dashed line is a simulation for an exclusive open channel blocker. Despite inherent variability in the relationships between K_D values and fractional current, the data for all compounds track much better with an expected closed channel blocker. Thus, increasing the butyl tail of compound 1 to eight carbons (compound 5), as well as substituting the ester linkages of both 1 and 5, does not appear to alter the essential mechanism of CNG channel block.

Figure 4.

State dependence of block. Plots showing relationship of all K_D values determined for compounds 1 (2 µM), 8 (2 µM), 5 (1 µM), and 9 (1 µM) at sub-saturating cGMP (50 or 100 µM) at +50 mV normalized to K_D values determined at saturating cGMP (2 mM), versus 1 - I/I_max, which is related to the fraction of closed channels. K_D values at saturating cGMP were corrected for ion accumulation. Solid lines indicate simulations for exclusive closed channel blockers, using K_D/K_Dsat = (1 - F_sat)/(1 - F_satI/I_max); where F_sat is the estimated fraction of open heteromeric rod channels in saturating cGMP assuming F_sat = 0.3 (a),⁴⁹ 0.56 (b),⁵⁰ or 0.78 (c).⁵¹ The dotted line is a simulation for a blocker with no preference for state, or K_D/K_Dsat = 1. The dashed line is a simulation for an exclusive open channel blocker, using K_D/K_Dsat = (I/I_max)⁻¹.

Resistance to In Vitro Hydrolysis

In addition to the unexpected increase in apparent affinity for CNG channels by compound 8, the amide and thioamide linkage substitutions should provide an improvement for many in vivo applications or tissue preparations to a tetracaine-based CNG channel blocker in terms of biological stability. Compound 1 is rapidly hydrolyzed by butyrylcholinesterase in the bloodstream;⁴³ hence its limitation to local targets in clinical usage. Butyrylcholinesterase is also the predominant esterase present in ocular tissues.^52–54 Similarly, in vivo use of a tetracaine-based CNG channel blocker would be short-lived. Amide and thioamide linkages are more resistant to hydrolysis than ester linkages. We therefore tested the resistance of the ester linkage-substituted compounds to in vitro hydrolysis using butyrylcholinesterase purified from human blood serum (Table 1). The concentrations of compounds used in the assays were well in excess of butyrylcholinesterase’s K_m for 1 and other local anesthetics.⁴³ The amide linkage substitutions of compounds 6 and 7 increased the resistance to hydrolysis substantially. The increase in hydrolysis resistance was even more dramatic for the thioamide linkage substitutions (compounds 8 and 9), which proved to be much too stable for our system to detect any hydrolysis product, even after 24 h incubations.

One concern relating to the amide linkage (compounds 6 and 7) is that it could increase the susceptibility of hydrolysis by proteases present in the eye.^55,56 However this was not the case. Using the broad-spectrum, nonspecific endopeptidases chymotrypsin and proteinase K, we were not able to detect any hydrolysis products for all compounds, while hydrolysis products of 4-nitrophenyl acetate and BSA were detected as positive controls.

Discussion

We have synthesized and determined the apparent affinities and mechanism of CNG channel block, as well as the hydrolysis rates, for 4 novel tetracaine-based compounds with potentially increased biological stabilities (compounds 6, 7, 8, and 9). The K_D values obtained for 1 and 5 (Table 1) were similar to those previously reported.^39,40 As expected, 5 had a higher apparent affinity for CNG channels than 1. The substituted amide linkages (compounds 6 and 7) did not adversely affect the apparent affinity for CNG channel block (Table 1). In addition, these compounds were more resistant to esterase hydrolysis than their ester counterparts (compounds 1 and 5). This poses a significant advantage for their use in in vivo research or in the clinic.

Compounds 8 and 9 were even more resistant to esterase hydrolysis than compounds 6 and 7, attributed to the thioamide linkage substitution (Table 1). For compound 9 the thioamide substitution did not enhance the apparent affinity for CNG channels, but it did for compound 8. It is unknown why the thioamide substitution in compound 8 improved CNG channel block, as the precise structure of the channel pore remains unresolved. These results suggest that 1 and its derivatives bind at a different site in the channel than 5 and its derivatives. The observation is further supported by the exhibited differences in voltage-dependence of block. It is possible that the lower electronegativity or increased size of sulfur in lieu of the carbonyl oxygen enhances the interaction of compound 8, improving the apparent affinity for the channel pore. It is also possible that the presence of sulfur instead if oxygen increases the partial double bond character of the carbon-nitrogen bond, affecting the orientation of the aromatic and head group moieties.

Compound 1 has been reported to be a closed-channel blocker.³⁷ Although we have previously determined the K_D values for 5, its mechanism of block had not yet been examined. Compound 5, as well as all compounds in this study, exhibited preference for the closed state over the open state (Figure 4). The physiological concentration of free cGMP in dark-adapted rod outer segments has been estimated to be about 3.5 µM.^57,58 Given the very large fraction of closed channels, the potency for CNG channel block in the intact retina is likely to be even greater than the values reported here.

It was a concern that the amide bond in compounds 6 and 7 could be susceptible to hydrolysis by proteases; however we found that they were resistant to both proteinase K and chymotrypsin. Chymotrypsin preferentially catalyzes the hydrolysis of peptide bonds involving L-isomers of tyrosine, phenylalanine, and tryptophan on the carboxy side.⁵⁹ As the amide bond adjacent to the aromatic ring in compounds 6 and 7 resembles a peptide bond of an aromatic amino acid residue, it was plausible that it would be recognized by the hydrophobic pocket of the chymotrypsin catalytic site. The resistance of compounds 6 and 7 to chymotrypsin may be due to the lack of a ‘carbon backbone’ in the structure. A similar argument can be applied to explain the resistance to hydrolysis by proteinase K: the aromatic ring directly adjacent to the equivalent of a peptide bond in compounds 6 and 7 renders it unrecognizable by the endopeptidase. Not surprisingly, compounds 8 and 9 with thioamide linkages were also completely resistant to protease action.

Compound 1’s actions are very short-lived, being rapidly degraded mainly by butyrylcholinesterase in the blood. The greater stability of the amide and thioamide linkages should enhance the half-life. Particularly for compounds 8 and 9, the clearance rate would likely depend on other metabolic pathways, but this would need to be determined in a proper pharmacokinetic study. One question that emerges from this study is which compound appears to have the greatest potential for clinical use? We are contemplating only local tissue uses of these compounds. Nonetheless, the stability of compounds 8 and 9 against hydrolysis is so dramatic that their safety for clinical use might come into question. The high stability of the thioamide linkage in a tetracaine-based CNG channel blocker may increase the chances for toxic systemic reactions. Large systemic doses of thiobenzamide and its para-substituted derivatives have been shown to be hepato- and nephrotoxic,^60–62 and certain thioamides such as propylthiouracil inhibit thyroid peroxidase and are used to treat hyperthyroidism.⁶³ These issues remain to be examined at the low doses needed to achieve CNG channel block. Lastly, compounds 7 and 9 have larger estimated log P values than compounds 6 and 8 (Table 1), and as such would theoretically have better accessibility to reach channel targets.

This study may have broader implications than for CNG channel blockers alone. There are situations in which longer lasting anesthetics are desirable, and there have been several attempts to achieve this.^64–66 The amino amide type of local anesthetics are well recognized for their better biological stability than amino esters; but the thioamide linkage, to our knowledge, is the first of its kind. Indeed, there have been some attempts to increase the half-life of 1 in biological preparations.^67–70 The hydrolysis resistant versions of 1 reported here may be attractive options for longer lasting local anesthetics; however their effects on nerve conductance need to be further examined.

Experimental Section

Retinal Rod CNG Channel Expression in Oocytes

Ovaries were surgically removed from adult Xenopus laevis females (Xenopus Express; Brooksville, FL) anesthetized with ice-cold 0.1% tricaine and 0.1% NaHCO₃ solution. Oocytes were chemically released from ovarian follicles in Ca²⁺-free Barth’s solution (in mM: 88 NaCl, 1 KCl, 2.4 NaHCO₃, 0.82 MgSO₄, 7.5 Tris, 2.5 sodium pyruvate, and 100 U/mL penicillin and 100 µg/mL streptomycin, pH 7.4) containing 0.1 U/mL Liberase Blendzymes (Roche, Indianapolis, IN). Stage IV and V oocytes were visually sorted and stored in ND-96 (in mM: 96 NaCl, 2 KCl, 1.8 CaCl₂, 1 MgCl₂, 5 HEPES, 2.5 sodium pyruvate, and 100 U/mL penicillin and 100 µg/mL streptomycin, pH 7.4) at 16 °C. Oocytes were co-injected the following day with 33 ng of CNGA1 and 67 ng of CNGB1 cRNA (2:1) synthesized from linearized pGEM-HE expression vectors containing the channel subunit cDNA sequences⁷¹ using T7 mMESSAGE mMACHINE (Ambion, Austin, TX). Injected oocytes were incubated at 18 °C the first day, and 16 °C for the remaining days.

Electrophysiological Recordings

Recordings from inside-out excised patches were made 3–7 days after oocyte injection on an Axopatch 1D amplifier (Axon Instruments, Foster City, CA). Briefly, oocyte vitelline membranes were removed in solution containing (in mM) 200 K aspartate, 20 KCl, 1 MgCl₂, 10 EGTA, 10 HEPES, pH 7.4. Oocytes were placed in a recording chamber in solution containing (in mM) 130 NaCl, 2 HEPES, 0.02 EDTA, 1 EGTA, pH 7.6, and borosilicate glass electrodes (1–3 MΩ) were filled with identical solution. Macroscopic currents were filtered at 1 kHz and sampled at 2 kHz using pCLAMP 8.0 software (Axon Instruments). Channels were either fully-activated with 2 mM or partially activated with 50 and 100 µM cGMP. Solutions containing different tetracaine derivatives were exchanged using a RSC-100 rapid solution changer, (Molecular Kinetics, Pullman, WA). Glass syringes and Teflon tubing were used to minimize the binding of compounds to surfaces; concentrations passing through the perfusion system were verified by absorbance. Current traces were digitally filtered at 300 Hz (Gaussian) and averaged using Clampfit 8.2 software. Currents in the absence of cGMP were subtracted from all currents analyzed. K_D’s were determined and expressed as the mean ± SD.

In Vitro Ester Hydrolysis

All enzymatic assays were performed with 50 µM of compound in 0.1 M phosphate buffer, pH 7.4, at 37 °C with stirring, unless otherwise noted. Butyrylcholinesterase stock solutions from human serum (Sigma, St Louis, MO) were prepared at 100 U/mL in 0.1 M phosphate buffer pH 7.4, and stored at −20 °C until use. Compound hydrolysis was monitored on an 8452A Diode Array Spectrophotometer (Hewlett Packard). Product peak wavelength absorption was immediately monitored upon addition of compound. Absorbances for complete hydrolysis were determined when there were no further detectable changes. Hydrolysis rates were calculated based on the changes in absorbance during the first 1 min of hydrolysis (1 and 5) or first 9 min (6 and 7), and were adjusted to the absorbance at the completion of the reaction. All hydrolysis assays were performed in triplicate, and final rate constants are expressed as the mean ± SD of the three individually determined rate constants. In some experiments with compound 1, samples were taken at the beginning and end of the assay to verify the hydrolysis product and the completion of the reaction by HPLC. These samples were compared to 1 and 3 on a C18 column eluted with 0.1% TFA in water/acetonitrile.

Activity of chymotrypsin (bovine, Worthington Biochemical Co.; 190 µg/mL) was verified with 90 µM 4-nitrophenyl acetate in 0.1 M HEPES solution, pH 6.5, at 37 °C with stirring. Hydrolysis product was monitored at 404 nm for 10 min. Activity of proteinase K (from Engyodontium album, Sigma; 165 µg/mL) was verified with 16.5 mg/mL bovine serum albumin in 0.1 M phosphate buffer, pH 7.4, at 37 °C. Samples were taken at time point increments up to 2 h when the reaction reached completion, and were analyzed with 12% SDS-PAGE, stained with 0.1% Coomassie Blue.

Statistical Analysis

Statistical comparison between groups was made using a one-way repeated-measures ANOVA and the Holm-Sidak post hoc method for multiple pair-wise comparisons. Statistical significance was accepted at P-values <0.01.

Chemistry

Reagents, including 1 and 2, were obtained from Sigma-Aldrich and were used without further purification. TLC was performed on glass backed silica plates and eluted in a mixture of 5–10% methanol and 90–95% dichloromethane. Plates were visualized using short wave UV light and KMnO₄. Crude compounds were initially purified using column chromatography, which was packed with normal phase silica gel and eluted using either ethyl acetate/hexane or methanol/dichloromethane mixtures. Trace impurities were removed by reversed-phase HPLC on an Xterra Prep RP8 column, 19 × 100 mm, 5 µm (Waters, Milford, MA) with a water-methanol gradient (5 mM ammonium acetate, pH 5), monitored at 214 and 310 nm to yield final products. Purity was assessed to be greater than 95% with an Xterra Analytical RP8 column, 4.6 × 250 mm, 5 µm, under similar conditions and monitoring. ¹H and ¹³C NMR spectra were obtained using a Bruker 500 MHz FT-NMR spectrometer. ESI-MS was performed on a Thermo Finnigan TSQ Classic mass spectrometer.

4-(Butylamino)benzoic Acid (3)

Compound 2 (~3 mmol) was dissolved in 15 mL methanol with α-picoline-borane (1.1 mol eq) and butanal (1.1 mol eq). The reaction was stoppered with a vent needle and stirred overnight at room temperature. After 16–24 h, solvent was removed in vacuo, 10 mL 1 M HCl was added to the flask, and the reaction was stirred at room temperature for an additional 30 min. The pH was adjusted to neutral using NaHCO₃ and the intermediate product was extracted with ethyl acetate (2 × 60 mL). The organic layer was washed with brine (1 × 45 mL), dried with magnesium sulfate, filtered, removed in vacuo, and subsequently purified via column chromatography with 30% ethyl acetate in hexane to yield 3 (88%) as a white powder. ¹H NMR (500 MHz) (CD₃OD): δ 7.92 (d, J = 8.8 Hz, 2H), 6.55 (d, J = 8.8 Hz, 2H), 3.18 (t, J = 7.2 Hz, 2H), 1.63 (m, 2H), 1.44 (m, 2H), 0.97 (t, J = 7.4 Hz, 3H); ¹³C NMR (125 MHz) (CD₃OD): δ 172.6, 153.1, 132.7, 117.3, 111.6, 43.4, 31.7, 20.5, 14.1.

4-(Octylamino)benzoic Acid (4)

Prepared as described for 3 with octanal to yield 4 (94%) as a white powder. ¹H NMR (500 MHz) (CD₃OD): δ 7.92 (d, J = 8.9 Hz, 2H), 6.55 (d, J = 8.9 Hz, 2H), 3.17 (t, J = 7.2 Hz, 2H), 1.63 (m, 2H), 1.25–1.41 (m, 10H), 0.89 (t, J = 7.0 Hz, 3H); ¹³C NMR (125 MHz) (CD₃OD):δ 172.6, 153.1, 132.7, 117.3, 111.6, 43.7, 32.1, 29.7, 29.6, 29.5, 27.4, 23.0, 14.4.

2-(Dimethylamino)ethyl 4-(Octylamino)benzoate (5)

In a flame sealed flask, 4 (~0.50 mmol) and CDI (1.5 mol eq) were dissolved in 3.0 mL of 1,2-dimethoxyethane (DME). The solution was stirred at 60 °C for approximately 2 h under argon. 2-(Dimethylamino)ethanol (3 mol eq) was added to the solution followed by a small quantity of NaH (~ 2 mg). The flask was allowed to cool to room temperature and react for an additional 16–24 h. The reaction was worked up by dissolving it in 100 mL of chloroform and washing with water (2 × 60 mL) and brine (1 × 60 mL). The organic layer was dried with sodium sulfate, filtered, removed in vacuo, and subsequently purified via column chromatography with 30% ethyl acetate in hexane to yield 5 (95%) as a white powder. This compound was reported previously although synthesized by a different method.^{40 1}H NMR (500 MHz) (CD₃OD): δ 7.83 (d, J = 8.9 Hz, 2H), 6.59 (d, J = 8.7 Hz, 2H), 4.58 (t, J = 5.0 Hz, 2H), 3.57 (t, J = 5.0 Hz, 2H), 3.14 ( t, J = 7.1 Hz, 2H), 3.00 (s, 6H), 1.63 (m, 2H), 1.25–1.5 (m, 10 H), 0.90 (t, J = 6.8 Hz, 3H); ¹³C NMR (125 MHz) (CDCl₃): δ 166.8, 152.2, 131.6, 117.9, 111.3, 62.3, 58.0, 45.9, 43.4, 31.8, 29.4, 29.3, 29.2, 27.1, 22.7, 14.1; ESI-MS: m/z 321.1 MH⁺; mp: 40–41°C.

4-(Butylamino)-N-(2-(dimethylamino)ethyl)benzamide (6)

In a flame sealed flask, 3 (~0.50 mmol) and CDI (1.5 mol eq) were dissolved in 3.0 mL of DME and stirred at 60 °C for approximately 2 h under argon. N′,N′-Dimethylethane-1,2-diamine (3 mol eq) was added to the solution followed by a small quantity of NaH (~ 2 mg). The flask was allowed to cool to room temperature and react for an additional 16–24 h. The reaction was worked up by dissolving it in 100 mL of chloroform and washing with water (2 × 60 mL) and brine (1 × 60 mL). The organic layer was dried over sodium sulfate, filtered, removed in vacuo, and subsequently purified via column chromatography with 30% ethyl acetate in hexane to yield 6 (93%) as pale brown oil. ¹H NMR (500 MHz) (CD₃OD): δ 8.13 (d, J = 8.8 Hz, 2H), 7.61 (d, J = 8.8 Hz, 2H), 3.82 (t, J = 5.9 Hz, 2H), 3.44 (m, 4H), 3.02 (s, 6H), 1.76 (m, 2H), 1.50 (m, 2H), 1.02 (t, J = 7.4 Hz, 3H); ¹³C NMR (125 MHz) (CD₃OD): δ 170.2, 141.8, 135.8, 131.7, 124.0, 59.5, 53.1, 44.8, 37.3, 30.3, 21.6, 14.8; ESI-MS: m/z 264.22 MH⁺.

4-(Octylamino)-N-(2-(dimethylamino)ethyl)benzamide (7)

Prepared as described for 6 using compound 4 to yield 7 (87%) as pale brown oil. ¹H NMR (500 MHz) (CDCl₃): δ 7.66 (d, J = 8.8 Hz, 2H), 6.99 (br, 1H), 6.55 (d, J = 8.8 Hz, 2H), 3.98 (br, 1H), 3.57 (m, 2H), 3.13 (t, J = 7.1 Hz, 2H), 2.68 (t, J = 5.7 Hz, 2H), 2.40 (s, 6H), 1.61 (m, 2H), 1.25–1.40 (m, 10H), 0.88 (t, J = 6.9 Hz, 3H); ¹³C NMR (125 MHz) (CDCl₃): δ 167.8, 151.4, 129.1, 122.4, 111.8, 58.4, 51.1, 45.2, 43.8, 36.9, 32.1, 29.7, 29.6, 27.4, 23.0, 14.4; ESI-MS: m/z 320.08 MH⁺.

4-(Butylamino)-N-(2-(dimethylamino)ethyl)benzothioamide (8)

In a flame sealed flask, 6 (~ 0.25 mmol) was dissolved in 5 mL of dry toluene with Lawesson’s reagent (1 mol eq) and refluxed under argon for 2.5 h. The reaction was dissolved in 20 mL of ethyl acetate, washed with water (2 × 20 mL), dried over sodium sulfate and subsequently purified via column chromatography with 10% methanol in dichloromethane to yield 8 (26%) as yellow oil. ¹H NMR (500 MHz) (CDCl₃): δ 8.64 (br, 1H), 7.79 (d, J = 9 Hz, 2H), 6.51 (d, J = 9 Hz, 2H), 4.03 (br, 1H), 3.97 (m, 2H), 3.14 (t, J = 7.1 Hz, 2H), 2.85 (m, 2H), 2.44 (s, 6H), 1.60 (m, 2H), 1.42 (m, 2H), 0.96 (t, J = 7.4 Hz, 3H); ¹³C NMR (125 MHz) (CDCl₃): δ 197.6, 151.7, 129.3, 129.2, 111.7, 57.0, 45.1, 43.5, 43.2, 31.7, 20.5, 14.2; ESI-MS: m/z 280.1 MH⁺.

4-(Octylamino)-N-(2-(dimethylamino)ethyl)benzothioamide (9)

Prepared as described for 8 using compound 7 to yield 9 (73%) as yellow oil. ¹H NMR (500 MHz) (CDCl₃): δ 8.41 (br, 1H), 7.76 (d, J = 8.8 Hz, 2H), 6.52 (d, J = 8.8 Hz, 2H), 4.02 (br, 1H), 3.90 (m, 2H), 3.14 (t, J = 7.1 Hz, 2H), 2.72 (t, J = 5.4 Hz, 2H), 2.35 (s, 6H), 1.62 (m, 2H), 1.25–1.40 (m, 10H), 0.88 (t, J = 7.0 Hz, 3H); ¹³C NMR (125 MHz) (CDCl₃): δ 197.4, 151.6, 129.4, 129.2, 111.8, 56.9, 51.2, 45.2, 43.8, 43.6, 32.2, 29.7, 29.6, 27.4, 23.0, 14.5; ESI-MS: m/z 336.13 MH⁺.