Antimycobacterial Rufomycin Analogues from Streptomyces atratus Strain MJM3502

Abstract

This study represents a systematic chemical and biological study of the rufomycin (RUF) class of cyclic heptapeptides, which our anti-TB drug discovery efforts have identified as potentially promising anti-TB agents that newly target the caseinolytic protein C1, ClpC1. Eight new RUF analogues rufomycins NBZ1–NBZ8 (1−8) as well as five known peptides (9−13) were isolated and characterized from the Streptomyces atratus strain MJM3502. Advanced Marfey’s and X-ray crystallographic analysis led to the assignment of the absolute configuration of the RUFs. Several isolates exhibited potent activity against both pathogens, M. tuberculosis H37Rv and M. abscessus, paired with favorable selectivity (selectivity index >60), which collectively underscores the promise of the rufomycins as potential anti-TB drug leads.

Graphical Abstract

Tuberculosis (TB), an infectious disease known to mankind since antiquity, still ranks as one of the top-10 causes of death worldwide. An estimated 10 million cases of active TB were reported in 2017, a year in which TB claimed the lives of 1.6 million people, including 0.3 million co-infected with HIV. In 2017, two thirds of the new TB cases occurred in eight out of the 30 countries with the highest TB burden worldwide, generating additional struggles over TB related secondary costs.^1–3 While TB incidence is falling at about 2% per year globally, this rate is still far behind the goal of 4−5% annual decline set forth in the 2020 milestones to achieve the End TB Strategy.¹ Furthermore, increasing multidrug-resistant TB (MDR-TB), extensively drug-resistant tuberculosis (XDR-TB), and even totally drug-resistant TB (TDR-TB) hinder the progress of TB treatment.⁴ Therefore, discovery and development of new anti-TB drugs with unique cellular targets is a high priority.

Rufomycins (RUFs; syn. ilamycins) comprise a family of potent anti-TB cyclic heptapeptides. They are produced by Actinomycetes, particularly Streptomyces atratus and S. macrosporeus.^5–10 Previous studies identified RUFs as a group of potential anti-TB peptides with a novel target, i.e., the caseinolytic protein C1 (ClpC1).^11–13 The clpC1 gene is evolutionarily highly conserved in Mycobacteria and essential for Mycobacterium tuberculosis cell growth.^11,14 ClpC1 can catalyze the ATP-dependent unfolding of substrate and translocation to the ClpP1/P2 complex for degradation.¹⁵ As ClpC1 is essential for protein homeostasis in mycobacteria, it is a promising new drug target. S. atratus strain MJM3502 was identified through a high-throughput screening (HTS) campaign from approximately 7,000 actinomycete cultures.¹⁶ Systematic chemical and biological investigation has now yielded eight new RUF analogues rufomycins NBZ1–NBZ8 (1−8), in addition to the five peptides (9−13)^{5−7,11−13}, characterized earlier. Advanced Marfey’s¹⁷ and X-ray crystallographic analyses ascertained the assignment of their absolute configurations. The new RUFs showed high selectivity indices and potent activities against both pathogens, M. tuberculosis and M. abscessus.

RESULTS AND DISCUSSION

The preparation isolation workflow (see Experimental Section) started with 28 g of RUF-enriched material obtained from a 240 L aliquot of MJM3502 culture broth. The procedure utilized a combination of Sephadex LH-20 and normal-phase silica gel column chromatography with orthogonal eluents, as well as semi-preparative HPLC. For the targeted isolation of new RUF species, fractionation was monitored by TLC detecting with vanillin sulfuric acid reagent. Under these conditions, RUF derivatives exhibited UV₂₅₄-absorbing bands which, upon reagent treatment and heating, developed into yellow bands and changed their color into various shades of green over time. Upon purification, their general identity as RUF analogues was evident from their 1D ¹H NMR spectra, exhibiting the typical aromatic, peptidic lower-field methine, and aliphatic complex resonances of these heptapeptides. Characteristic RUF properties were their unusually high-field N-Me leucine (NMeLeu) resonance, as well as molecular weights around 1,000 amu, as determined by (+)-HRESIMS. Among the isolates, eight compounds (1–8) were new natural products, the structures of which were elucidated as follows.

One isolate, rufomycin NBZ1 (1), was obtained as pale yellow crystals. Its molecular formula was deduced from the (+)-HRESIMS ion at m/z 1044.5760 [M + H]⁺, in conjunction with the carbon atom counts from the 1D ¹³C NMR data, to be C₅₄H₇₇N₉O₁₂. The IR spectrum of 1 exhibited a strong IR band at 1628 cm⁻¹ (Figure S9, Supporting Information), characteristic for an amide functionality. The peptide nature of 1 was corroborated via initial analysis of its ¹H and ¹³C NMR spectra (Table 1), showing seven amide carbonyl signals resonating between δ_H 169 and 175 ppm, and seven α-hydrogen and carbon resonances (δ_H 4.5~5.1 ppm, δ_C 46~61 ppm). Observation of a pair of methylene ¹H NMR signals with one unusually high field resonance (δ_H −0.65 and 1.63 ppm) was indicative of the N-MeLeu being sandwiched between the two aromatic amino acids, a common feature of RUFs, collectively suggesting that 1 is a RUF analogue.^12,13 In the aromatic chemical shift range of the ¹H NMR spectrum, seven characteristic signals were observed: one set of ABX resonances (δ_H 7.79, d, J = 2.2 Hz; δ_H 7.05, d, J = 8.6 Hz; and δ_H 7.41, dd, J = 8.6 and 2.2 Hz) revealed the presence of a mNO₂Tyr residue. Five further aromatic signals, as well as a characteristic oxiranyl resonance (δ_H 2.81, dd, J = −4.6 and 2.6 Hz; δ_H 2.86, dd, J = −4.6 and 4.1 Hz; δ_C 45.9; and δ_H 3.23, dd, J = 4.1 and 2.6 Hz; δ_C 58.9) were assigned to a 1-(1′,1′-dimethyl-2′-oxiranyl-ethyl)-Trp residue, which represents an epoxidized, reversely prenylated Trp (ErpTrp). The remaining five amino acid residues were assigned as NMeLeu₁, NMe-5-OH-Leu₂, Leu₃, Ala, and trans-crotylated Gly (TrcGly) by analysis of the TOCSY/COSY (Figure 1) and HSQC spectra (Figure S6, Supporting Information). The amino acid sequence and the position of the heptacyclic ring connection were determined via the HMBC correlations connecting the α-hydrogens to the respective carbonyl carbons of the adjacent residues (Figure 1). The two-dimensional structure of 1 was thereby assigned as cyclo-(ErpTrp-NMeLeu-mNO₂Tyr-Ala-NMe-5-OH-Leu-Leu-TrcGly), which was consistent with the fragmentation pattern observed from ESIMS/MS data (Figure 2). Application of the advanced Marfey’s method¹⁷ using the l- and d-1-fluoro-2,4-dinitrophenyl-5-leucinamide (FDLA) reagents revealed the all-l assignment of the amino acids as l-Ala, l-mNO₂Tyr, l-Trp, l-Leu, l-NMeLeu, and l-TrcGly (Figure 3 and Table 3). Partial racemization of l-NMeLeu and hydration-lactonization products of l-TrcGly were observed by LCMS (Figures 3B and 3C). All seven amino acid residues were confirmed to be l-configured by an X-ray diffraction study (Figure 4), which also established the absolute stereochemistry of an S-configuration for both C-13 and C-32. (Table S1, Supporting Information).

Table 1.

¹H (400 MHz) and ¹³C NMR (100 MHz) Data of Compounds 1−4 in CD₃OD

position	1		2		3		4
	δH, mult. (J in Hz)	δC	δH, mult. (J in Hz)	δC	δH, mult. (J in Hz)	δC	δH, mult. (J in Hz)	δC
Erp-Trp
1		173.8		173.9		173.8		173.9
2	4.90, dd (11.2, 4.1)	51.8	4.90, dd (11.0, 4.1)	51.8	4.90, dd (11.2, 4.4)	51.8	4.89, dd (11.2, 4.1)	52.0
3	3.16, dd (−13.2, 4.1)	28.2	3.16, dd (−13.3, 4.1)	28.2	3.16, dd (−13.2, 4.4)	28.2	3.19, dd (−13.2, 4.1)	28.3
	3.41, dd (−13.2, 11.2)		3.41, dd (−13.3, 11.0)		3.41, dd (−13.2, 11.2)		3.41, dd (−13.2, 11.2)
4	7.15, s	125.7	7.16, s	125.7	7.15, s	125.4	7.12, s	125.2
5		109.3		109.3		108.8		108.8
6		130.5		130.4		130.4		130.4
7	7.54, brd (7.8)	119.5	7.55, brd (7.8)	119.6	7.53, m	119.4	7.53, m	119.4
8	7.07, m	120.6	7.05, m	120.6	7.03, m	120.3	7.04, m	120.3
9	7.14, m	122.7	7.13, m	122.7	7.06, m	122.1	7.07, m	122.1
10	7.79, brd (7.3)	114.7	7.80, brd (7.3)	114.7	7.50, m	115.2	7.51, m	115.3
11		137.1		137.1		136.8		136.9
12		59.2		59.2		60.1		60.1
13	3.23, dd (4.1, 2.6)	58.9	3.25, dd (4.1, 2.6)	58.9	6.12, dd (17.5, 10.7)	145.3	6.13, dd (10.7, 17.5)	145.3
14	2.81, dd (−4.6, 2.6)	45.9	2.82, dd (−4.6, 2.6)	46.0	5.16, brd (17.5)	114.1	5.17, brd (17.5)	114.1
	2.86, dd (−4.6, 4.1)		2.87, dd (−4.6, 4.1)		5.20, brd (10.7)		5.21, brd (10.7)
15	1.51, s	23.1	1.52, s	23.1	1.70, s	28.4	1.71, s	28.4
16	1.67, s	25.1	1.68, s	25.1	1.70, s	28.5	1.71, s	28.5
TrcGly
17		173.3		173.3		173.2		173.3
18	4.60, m	53.4	4.60, m	53.4	4.59, m	53.3	4.59, m	53.4
19	2.77, m	36.2	2.76, m	36.2	2.79, m	36.2	2.86, m	36.1
	2.49, m		2.46, m		2.50, m		2.55, m
20	5.18, m	125.2	5.17, m	125.3	5.18, m	125.2	5.19, m	125.0
21	5.50, m	131.3	5.50, m	131.3	5.49, m	131.3	5.50, m	131.6
22	1.56, brd (6.2)	18.7	1.57, brd (6.2)	18.6	1.56, m	18.7	1.57, brd (6.4)	18.6
Leu
23		174.0		173.9		173.9		174.2
24	4.59, m	54.9	4.59, m	54.9	4.61, m	54.9	4.56, m	55.0
25	1.73~1.80, m	44.7	1.76, m	44.5	1.73~1.76, m	44.7	1.60, m	44.3
							1.74, m
26	1.73, m	25.9	1.75, m	26.0	1.73, m	25.9	1.61, m	25.9
27	0.94, d (6.3)	21.9	0.94, d (6.3)	22.0	0.94, d (5.8)	22.0	0.92, d (6.5)	21.6
28	0.97, d (6.2)	23.2	0.97, d (6.2)	23.2	0.97, d (5.8)	23.3	0.94, d (6.5)	23.5
NMeLeu
29		171.5		171.3		171.5		171.5
30	5.05, t (7.2)	59.8	4.99, t (7.0)	60.2	5.05, t (7.2)	59.7	4.98, dd (6.2, 8.0)	60.1
31	1.84, m	33.7	1.44, m	33.9	1.83, m	33.7	1.57, m	38.4
			2.19, m				1.94, m
32	1.47, m	33.3	1.58, m	33.5	1.47, m	33.2	1.49, m	25.7
33	0.92, d (6.9)	17.3	0.95, d (6.9)	17.8	0.92, d (6.9)	17.3	0.97, d (6.6)	23.0
34	3.33, m	68.4	3.40, m	68.3	3.33, m	68.4	0.97, d (6.6)	23.5
	3.45, m		3.47, m		3.45, m
N-Me	2.74, s	29.7	2.75, s	29.8	2.74, s	29.7	2.73, s	30.0
Ala
35		175.1		175.0		175.0		175.3
36	4.83, q (6.9)	46.7	4.87, q (6.9)	46.8	4.83, q (6.9)	46.7	4.85, q (6.8)	46.7
37	1.24, d (6.9)	17.1	1.25, d (6.9)	17.0	1.24, d (6.9)	17.2	1.27, d (6.8)	17.3
mNO2-Tyr
38		172.3		172.3		172.3		173.2
39	4.58, m	55.5	4.59, m	55.4	4.59, m	55.5	4.59, m	55.6
40	3.04, dd (−13.1, 9.9)	38.2	3.04, dd (−13.1, 9.9)	38.2	3.04, dd (−13.1, 9.8)	38.2	2.99, dd (−13.3, 9.0)	38.8
	2.72, m		2.72, m		2.71, m		2.58, m
41		129.5		129.5		129.5		128.3
42	7.79, d (2.2)	126.7	7.79, d (2.2)	126.8	7.78, d (2.2)	126.7	6.93, m	131.3
43		135.5		135.5		135.4	6.66, m	116.3
44		154.6		154.6		154.6		157.3
45	7.05, d (8.6)	121.1	7.05, d (8.6)	121.1	7.06, d (8.6)	121.1	6.66, m	116.3
46	7.41, dd (8.6, 2.2)	139.2	7.41, dd (8.6, 2.2)	139.2	7.40, dd (8.6, 2.2)	139.2	6.93, m	131.3
NMeLeu
47		169.6		169.6		169.6		169.4
48	4.66, dd (11.6, 3.2)	59.7	4.67, dd (11.6, 3.4)	59.7	4.67, dd (11.7, 3.0)	59.7	4.63, m	59.9
49	1.64, m	38.4	1.62, m	38.4	1.61, m	38.4	1.58, m	38.4
	−0.67, m		−0.66, m		−0.67, m		−0.69, m
50	1.08, m	25.4	1.06, m	25.4	1.08, m	25.4	1.08, m	25.4
51	0.16, d (6.7)	21.0	0.17, d (6.7)	21.0	0.17, d (6.7)	21.0	0.16, d (6.6)	21.0
52	0.37, d (6.7)	23.5	0.37, d (6.7)	23.5	0.38, d (6.7)	23.5	0.37, d (6.6)	23.5
N-Me	2.68, s	29.8	2.68, s	29.9	2.68, s	29.8	2.58, s	29.9

Figure 1.

Selected COSY/TOCSY (bold bonds) and HMBC correlations (arrows) of 1.

Figure 2.

ESIMS/MS of fragmentation pattern of 1.

Figure 3.

The LC-ESIMS chromatograms of the hydrolysate of 1 after derivatization with L- and D-FDLA (advanced Marfey’s analysis; excess reagent eluting between 7.4 and 7.9 min was diverted to prevent it from entering the MS detector).

Table 3.

Retention Times and MS Ions (m/z) of the Observed L/D-FDLA Amino Acid Derivatives from the Hydrolysis Products of 1

L-AA	L-FDLA-L-AA (t, min)	D-FDLA-L-AA (t, min)	MS (m/z)
Leu	10.0	16.7	426.2 [M + H]
NMeLeu	12.6	15.5	462.2 [M + Na]
Ala	6.0	8.1	384.2 [M + H]
mNO2Tyr	9.4	12.4	521.2 [M + H]
Trp	10.2	13.2	499.2 [M + H]
TrcGly	7.3	8.1	446.2 [M + Na]

Figure 4.

Perspective representation of the X-ray crystallographic model of 1.

Rufomycin NBZ2 (2) shared the molecular formula, C₅₄H₇₇N₉O₁₂, with 1 based on the (+)-HRESIMS and ¹³C NMR data. Comparison of its ¹H and ¹³C NMR data (Table 1) with those of 1 showed the close structural similarity between the two compounds: only slight differences in chemical shifts were observed around C/H-30 (δ_C 59.8 and δ_H 5.05, t, J = 7.2 Hz for 1; δ_C 60.2 and δ_H 4.99, t, J = 7.0 Hz for 2), C/H₂-31 (δ_C 33.7 and δ_H 1.84, m for 1; δ_C 33.9 and δ_H 1.44 and 2.19, m for 2), C/H-32 (δ_C 33.3 and δ_H 1.47, m for 1; δ_C 33.5 and δ_H 1.58, m for 2), and C/H₃-33 (δ_C 17.3 and δ_H 0.92, d, J = 6.9 Hz for 1; δ_C 17.8 and δ_H 0.95, d, J = 6.9 Hz for 2), suggesting that 2 is a C-32 epimer of 1. Compound 2 shared its planar structure with 1 and was fully elucidated using the HMBC and COSY/TOCSY data (Figures S13−16, Supporting Information). Analysis of the Marfey’s reaction product by LC-MS verified the presence of identical amino acid residues with all l-configuration in 1 and 2 (Figure S19, Supporting Information). Thus, 2 was assigned as the C-32 epimer of 1.

Compared to 2, 3 contained one less oxygen atom, corresponding to the molecular formula C₅₄H₇₇N₉O₁₁, which was gleaned from the (+)-HRESIMS ion at m/z 1028.5853 [M + H]⁺ and supported by the ¹³C NMR data. Analysis of the ¹H and ¹³C NMR spectra (Table 1) suggested 3 to be structurally related to 1 and 2, with the major differences being the presence of an additional double bond instead of an epoxide ring. This additional singly substituted double bond (C/H-13: δ_C 145.3, δ_H 6.12, dd, J = 17.5 and 10.7 Hz; C/H₂-14: δ_C 114.1, δ_H 5.16, d, J = 17.5 Hz and δ_H 5.20, d, J = 10.7 Hz) was assigned as Δ^13,14 based on the HMBC correlations (Figure S25, Supporting Information) from both Me-15 and Me-16 to C-13. The S-configuration of C-32 was assigned based on the similarity of the NMR chemical shifts of neighboring C/H-30 (δ_C 59.7 and δ_H 5.05), C/H₂-31 (δ_C 33.7 and δ_H 1.83), C/H-32 (δ_C 33.2 and δ_H 1.47), and C/H₃-33 (δ_C 17.3 and δ_H 0.92) in 3 and 1, which were distinct from those in 2. The remaining stereocenters of 3 were also congruent to those in 1 as determined by comparison of the NMR data and analysis of the LC-MS chromatograms of the Marfey’s reaction products (Figure S28, Supporting Information). The structure of 3, named as rufomycin NBZ3, was thus assigned as shown.

The molecular formula, C₅₄H₇₈N₈O₈, of 4 established via (+)-HRESIMS and ¹³C NMR carbon counts combined with initial interpretation of the NMR data (Table 1), revealed this compound to be a RUF analogue resembled ilamycin B1 (11).⁵ Significant differences were the presence of two aromatic AA′XX′ resonances at δ_H 6.67 and 6.93, each integrating for 2H, characteristic for a 1,4-disubstituted aromatic ring. Thus, 4 contained a Tyr rather than the typical RUF mNO₂Tyr residue. This was consistent with the molecular weight and congruent with the HMBC correlations (Figure S34, Supporting Information). Compound 4 represents a rare type of RUF that lacks the nitro group in the tyrosine residue. All the seven amino acids of 4 were assigned as l-stereoisomers based on their near identical ¹H NMR coupling and chemical shift patterns relative to those of the well-documented ilamycin B1 (11). This also considered the observation that all RUFs known to date contain exclusively l amino acids. Advanced Marfey’s analysis confirmed that all-l-configuration of the amino acids (Figure S37, Supporting Information). Collectively, this established the structure of rufomycin NBZ4 (4) as shown.

Comparison of the NMR data of 5 with those of 1 (Table 2) revealed that their structures are closely related. Combined with the molecular formula information (C₅₆H₈₁N₉O₁₃, determined by (+)-HRESIMS and ¹³C NMR data), it was evident that 5 and 1 share identical amino acid residues: ErpTrp, NMeLeu₁, mNO₂Tyr, Leu₃, Ala, and TrcGly. Two additional methoxy groups (δ_H 3.27 and 3.36, s) and additional resonances of 5 at δ_C 110.96 and δ_H 4.10 of 5 suggested the presence of a dimethyl acetal function. The presence of such a functionality can be explained via biogenetic oxidation of one of the terminal Me groups to an aldehyde and formation of the MeOH adduct; the final step could well be an artifact of the extraction and/or isolation procedure. The two OCH₃ groups were connected to C-34, as verified by the HMBC correlations from H-33 to C-34 (δ_C 111.0) and from the two methoxy groups (δ_H 3.27 and 3.36) to C-34 (Figure S34, Supporting Information). The structure of rufomycin NBZ5 (5) was thus assigned as shown.

Table 2.

¹H (400 MHz) and ¹³C NMR (100 MHz) Data of Compounds 5−8 in CD₃OD

position	5		6		7		8
	δH, mult. (J in Hz)	δC	δH, mult. (J in Hz)	δC	δH, mult. (J in Hz)	δC	δH, mult. (J in Hz)	δC
Erp-Trp
1		173.8		174.2		174.3		174.2
2	4.89, dd (11.2, 4.6)	51.8	4.40, t (6.9)	57.5	4.87, m	51.5	4.86, m	51.7
3	3.17, dd (−13.3, 4.6)	28.2	3.25, d (6.9)	27.8	3.19, m	29.0	3.21, m	29.2
	3.42, dd (−13.3, 11.2)
4	7.15, s	125.7	7.38, s	125.6	7.11, s	126.5	7.18, s	125.9
5		109.3		109.4		108.6		109.1
6		130.4		130.8		130.9		130.6
7	7.55, m	119.5	7.55, m	119.7	7.53, m	119.9	7.52, m	119.8
8	7.06, m	120.6	7.05, m	120.4	7.05, m	120.3	7.04, m	120.5
9	7.13, m	122.6	7.12, m	122.5	7.12, m	122.4	7.11, m	122.6
10	7.78, m	114.7	7.76, m	114.4	7.68, m	115.1	7.76, m	114.6
11		137.1		137.2		136.7		137.1
12		59.1		59.1		62.4		59.2
13	3.22, dd (4.1, 2.7)	58.9	3.28, dd (4.1, 2.7)	59.0	4.37, dd (8.5, 2.5)	76.5	3.24, dd (4.1, 2.7)	58.9
14	2.80, dd (−4.6, 2.7)	45.9	2.82, dd (−4.7, 2.7)	46.1	3.03, dd (−11.3, 2.5)	63.5	2.81, dd (−4.6, 2.7)	46.0
	2.85, dd (−4.6, 4.1)		2.87, dd (−4.7, 4.1)		3.30, dd (−11.3, 8.5)		2.86, dd (−4.6, 4.1)
15	1.50, s	23.1	1.55, s	22.8	1.64, s	23.9	1.51, s	23.2
16	1.66, s	25.1	1.71, s	25.0	1.75, s	26.2	1.66 s	25.0
TrcGly
17		173.1		174.2		173.4		173.5
18	4.58, m	53.3	4.34, dd (6.3, 5.1)	54.5	4.57, m	54.3	4.58, m	54.3
19	2.51, m	36.1	2.43, m	36.7	2.62, m	35.2	2.62, m	35.2
	2.81, m		2.73, m		2.81, m		2.80, m
20	5.17, m	125.1	5.36, m	126.5	5.60, m	127.6	5.60, m	127.6
21	5.49, m	131.3	5.48, m	130.6	5.64, m	129.4	5.64, m	129.4
22	1.55, m	18.7	1.55, m	18.4	1.66, m	18.3	1.65, m	18.3
Leu
23		173.9		174.6		173.1		173.4
24	4.63, m	54.6	4.52, dd (10.6, 5.1)	54.3	5.23, dd (11.1, 5.4)	55.2	5.28, dd (11.5, 4.7)	55.0
25	1.72, m	44.9	1.62, m	43.1	1.94, m	35.8	1.94, m	35.8
			1.92, m
26	1.73, m	25.9	1.68 m	26.0	1.42, m	25.9	1.75, m	24.7
27	0.94, d (6.0)	21.9	0.91, d (6.3)	21.5	0.92, d (6.5)	21.3	0.91, d (6.4)	21.3
28	0.97, d (6.0)	23.3	0.94, d (6.3)	23.5	1.02, d (6.5)	23.9	1.00, d (6.4)	24.1
NMeLeu
29		171.3		171.8		171.9		171.7
30	5.27, dd (8.2, 6.4)	59.6	4.81, m	60.3	3.79, dd (11.3, 7.1)	63.2	3.84, dd (11.4, 6.9)	60.9
31	1.72, m	32.3	1.58, m	38.3	1.88, m	26.9	1.95, dd (−12.8, 6.9)	31.9
	1.90, m		2.00, m		2.28, m		2.52, dd (−12.8, 11.4)
32	1.71, m	33.9	1.49, m	25.8	1.98, m	34.2		70.9
33	0.93, d (6.7)	16.8	0.97, d (6.5)	23.0	1.09, d (6.8)	17.5	1.35, s	26.7
34	4.10, d (4.3)	111.0	0.97, d (6.5)	23.5	4.76, d (2.2)	79.4	4.55, brs	83.1
NMe	2.71, s	29.6	2.73, s	30.0	3.25, s	38.4	3.24, s	38.6
OCH3	3.27, s/3.36, s	54.7/56.5
Ala
35		174.9		175.4		172.5		172.6
36	4.78, q (6.8)	46.7	4.82, q (6.9)	47.1	4.81, q (6.6)	47.7	4.79, q (6.6)	47.8
37	1.22, d (6.8)	17.1	1.30, d (6.9)	16.8	1.27, d (6.6)	17.8	1.27, d (6.6)	17.8
NO2-Tyr
38		172.4		172.2		171.7		171.7
39	4.60, m	55.3	4.59, dd (7.9, 5.4)	55.3	4.64, dd (10.7, 5.7)	57.1	4.65, dd (10.2, 5.9)	57.1
40	3.02, dd (13.2, 9.9)	38.3	3.16, dd (−13.3, 7.9)	37.6	2.89, dd (−14.1, 10.7)	38.1	3.09, dd (−14.1, 5.9)	38.2
	2.72, m		2.85, dd (−13.3, 5.4)		3.09, dd (−14.1, 5.7)		2.89, m
41		129.5		129.1		130.2		130.1
42	7.77, d (2.2)	126.6	7.85, d (2.2)	127.2	7.84, d (2.2)	126.5	7.83, d (2.2)	126.4
43		135.4		135.7		135.6		135.5
44		154.6		156.0		154.3		154.3
45	7.05, d (8.6)	121.1	7.00, d (8.6)	121.5	7.07, d (8.6)	121.2	7.06, d (8.6)	121.2
46	7.39, dd (8.6, 2.2)	139.1	7.47, dd (8.6, 2.2)	139.5	7.39, dd (8.6, 2.2)	138.9	7.38, dd (8.6, 2.2)	138.9
NMeLeu
47		169.5		174.1		170.0		169.9
48	4.69, dd (11.8, 3.3)	59.7	3.91, dd (10.1, 5.0)	55.3	4.28, dd (10.5, 3.8)	59.6	4.26, dd (10.8, 3.6)	59.6
49	1.61, m	38.3	1.41−1.54, m	40.4	1.54, m	37.7	1.53, m	37.7
	−0.73, m				−0.38, m		−0.50, m
50	1.08, m	25.4	1.13, m	25.6	0.96, m	25.6	0.94, m	25.6
51	0.17, d (6.7)	20.9	0.62, d (6.6)	21.5	0.10, d (6.6)	21.5	0.10, d (6.7)	21.5
52	0.35, d (6.7)	23.5	0.68, d (6.6)	23.3	0.45, d (6.6)	23.3	0.42, d (6.7)	23.3
NMe	2.68, s	29.8			2.34, s	29.3	2.33, s	29.3

Analysis of the ¹H NMR spectrum of 6 (Table 2) pointed to the typical RUF resonances of Ala, mNO₂Tyr, TrcGly, and ErpTrp. Three pairs of related doublet methyl signals, as well as one NMe signal suggested the presence of two Leu and one NMeLeu residues. The molecular formula of 6 was assigned as C₅₃H₇₅N₉O₁₁ from the (+)-HRESIMS ion at m/z 1014.5695 [M + H]⁺ and the ¹³C NMR data. This provided evidence that 6 is the first RUF analogue that lacks the Me-53 group. This was corroborated by the lack of one N-Me signal in the ¹H NMR spectrum, as well as by the absence of an HMBC correlation corresponding to the interaction between H₃-53 and the adjacent amide carbonyl group of the ErpTrp residue (Figure S51, Supporting Information). Interestingly, the only difference between 6 and ilamycin B2 (10)^6,7 was that 6 has a Leu instead of a NMeLeu between the ErpTrp and mNO₂Tyr residues. Moreover, the characteristic H₂-49 methylene resonances of 10 that give rise to a pair of signals at δ_H −0.65 and 1.63, including one prominent signal at very high field, were shifted to δ_H 1.41~1.54 ppm in 6, suggesting the N-Me group strongly affects the 3D conformational relationship of the Trp and Leu residues and, thus, the intramolecular aromatic-induced chemical shift effects. In summary, this evidence demonstrated the structure of 6 as shown and the compound was named as rufomycin NBZ6.

The structure elucidation of rufomycin NBZ7 (7) utilized the molecular formula C₅₄H₇₇N₉O₁₃ gleaned from the (+)-HRESIMS (m/z 1060.5703 [M + H]⁺) and ¹³C NMR data. A distinctive characteristic of the molecule was the observation of a spin system of hydrogens connected to the fragment C-30−C-31−C-32−(C-33)−C-34 delineated by COSY/TOCSY correlations (Figure 5A), as well as the following HMBC correlations (Figure 5A): from H₃-54 (δ_H 3.25, s) to C-30 (δ_C 63.2) and C-35 (δ_C 172.5); from H-30 (δ_H 3.79, dd, J = 11.3 and 7.1 Hz) to C-29 (δ_C 171.9) and C-32 (δ_C 34.2); from H₃-33 (δ_H 1.09, d, J = 6.8 Hz) to C-34 (δ_C 79.4); and from H-34 (δ_H 4.76, brd, J = 2.2 Hz) to C-24 (δ_C 55.2). This was consistent with the presence of a unique 6-hydroxy-5-methyl-2-piperidinone ring between a terminal methyl group of NMeLeu and the nitrogen atom of the adjacent Leu residue. Furthermore, a 2,3-dihydroxy-1,1-dimethylpropyl moiety was attached to the ring nitrogen of Trp, as elucidated by the HMBC cross-peaks (Figure S60, Supporting Information) from H₃-15 (δ_H 1.64, s) and H₃-16 (δ_H 1.75, s) to C-12 (δ_C 62.4) and C-13 (δ_C 76.5); as well as from H-13 (δ_H 4.37, dd, J = 8.5 and 2.5 Hz) and H-14 (δ_H 3.03, dd, J = −11.3 and 2.5 Hz; δ_H 3.30, dd, J = −11.3 and 8.5 Hz) to C-12. The absolute configurations of all amino acids were determined as being l, based on biosynthetic relationships and outcomes of advanced Marfey’s analyses (Figure S64, Supporting Information). The 13S configuration was assigned by biosynthetic analogy of the series of isolates. The hydroxy group and methyl groups attached to the 6-hydroxy-5-methyl-2-piperidinone ring were both α-configured based on observed NOESY correlations between H-30/H-32 and H-34/H-32 as well as H-34/H₃-33 (Figures 5B and S61, Supporting Information).

Figure 5.

COSY/TOCSY (A) and NOESY (B) correlations within the 6-hydroxy-5-methyl-2-piperidinone moiety in 7.

Finally, rufomycin NBZ8 (8), displayed a molecular formula, C₅₄H₇₅N₉O₁₃, based on the (+)-HRESIMS ion at m/z 1058.5602 [M + H]⁺ and its ¹³C NMR data, requiring 22 double bond equivalents. As 21 degrees of unsaturation could be accounted for via the mNO₂Tyr, TrcGly, ErpTrp moieties, and seven amide groups evident from the NMR data as well as a 21-membered ring, this suggested the presence of one additional ring closure. A 5,6-dihydroxy-5-methyl-2-piperidinone ring similar to that in 7 was verified by the HMBC correlations (Figure S70, Supporting Information) from H₃-54 (δ_H 3.24) to C-30 (δ_C 60.9); from H-30 (δ_H 3.84) to C-29 (δ_C 171.7), C-32 (δ_C 70.9), and C-35 (δ_C 172.6); from H₃-33 (δ_H 1.35) to C-31 (δ_C 31.9), C-32, and C-34 (δ_C 83.1); and from H-34 (δ_H 4.55) to C-24 (δ_C 55.0). C-30 (δ_C 60.9) in 8 was shifted to high field compared to 7 (δ_C 63.2), suggesting that the OH-32 group showed the same β-orientation as H-30. This was supported by the similar J values between H-30 and H-31α/β in 8 (J_30,31α = 11.4 Hz and J_30,31β = 6.9 Hz) relative to those observed in 7 (J_30,31α = 11.3 Hz and J_30,31β = 7.1 Hz). The OH-34 group in 8 was assigned to have the same α-orientation as in 7, because no NOESY cross-peak between H-34 and H-31α was observed.

In addition to compounds 1–8, the known diketopiperazine, rufomyazine (9),¹² the four known RUFs, ilamycin B2 (10),^6,7 ilamycin B1 (11),⁵ and the C-32 epimeric mixture RUFs I and II (12),^11,13 as well as the corresponding chlorohydrin, 13,^6,7 were isolated. Their structures were identified by NMR and MS analysis, as well as by comparison with the reported spectroscopic values as per the cited literature.

As compound purity is a crucial premise in bioactivity evaluation and SAR studies, the present isolates were purified by extensive chromatography to the best homogeneity practically achievable. However, the apparent relatively unfavorable impurity of the isolates, as can be gleaned from the occurrence of minor signals in the NMR spectra, arises in fact from an inherent property of these cyclic peptides that is different from purity, i.e., the presence of multiple conformers. Representing a form of Dynamic Residual Complexity, the conformational dynamics of the cyclopeptides lead to the observation of multiple isomeric species in NMR spectroscopic analysis and even (preparative) LC analysis. Collectively, this complicates both qNMR-based purity assignment of the RUFs.

However, based on their UHPLC (Figure S75. Supporting Information) and LC-MS profiles (Figures S8, S17, S26, S35, S44, S52, S62, and S72. Supporting Information), the isolates are >95% pure. The UHPLC purity profiles were consistent within one year after the initial purification of the compounds. Of all isolates 1−13, only the samples of 6 and 10 showed distinct impurities. Rather than representing impurities, the small peaks eluting before the major peaks in the UHPLC profiles of 7 and 12 can be explained by the mutarotatory equilibria of their hemiaminal functionalities. The extensive available 1D and 2D NMR data supports this tentative assignment given the homogeneity of the observed major and minor 2D cross peak patterns, and when considering the presence of intramolecular chemical shift effects. An ongoing NMR investigation dedicated to the elucidation of the RUF conformers of RUFs is aimed at a more comprehensive understanding of the conformer populations and mutarotatory equilibria of the hemiaminals vs. the actual purity of RUFs via qNMR methodology; results will be reported in due course. Based on all these considerations, from the perspective of purity, the present isolates were fit for biological and SAR studies presented here.

The biological evaluation of the isolates involved a panel of mycobacterial and related bioassays, as follows. Consistent with our previous research, the diketopiperazine, 9 did not show anti-TB activity.¹² The other twelve RUFs showed divergent in vitro antimycobacterial potencies, some with very low MICs (0.03 μM for 8 and 0.02 μM for 12), and others with higher MICs (1.7 μM for 11 and >10 μM for 4), indicating lower or even lack of potency (Table 4). An overview of the MICs of this small RUF library in combination with their structures gives a cursory idea of three structure-activity relationships (SARs), which provide important new insights: (i) the epoxide ring in the prenyl group attached to Trp contributes to bioactivity more positively than does an olefin, vicinal diol, or chlorohydrin in the same position, as seen by comparing the MICs of 1/3, 10/11/13, and 7/12; (ii) the absence of the N-Me group in NMeLeu₁ and the NO₂ group in mNO₂Tyr each led to poor activity based on the MICs of the corresponding pairs 6/10 and 4/11; (iii) RUFs with an additional 2-piperidinone ring result in prominently low MICs (7, 8, and 12). Compared to their pronounced activity against replicating M. tuberculosis, RUFs did not exhibit great activity against non-replicating M. tuberculosis in vitro: under hypoxic conditions, eight of the isolated RUFs showed MIC values of >10 μM, whereas other RUFs showed moderate MICs is the range of 4 to 8 μM.¹⁸

Table 4.

Results of a Panel of Mycobacterial and Mammalian in Vitro Bioassays (μM) and SPR (K_D, μM) Data of 1−13.

cpd	M. tb H37Rv MIC	Vero cell IC50	SI	M. abscessus MIC	day 7 MBC	day 14 MBC	KD NTD	KD FL
1	0.25	>100	>400	NT	NT	NT	0.62	0.50
2	0.44	>100	>227	NT	NT	NT	1.6	1.3
3	0.84	>100	>119	9.3	>10	9.6	1.1	1.1
4	>10	89	<9	>10	>10	5.2	1.2	5.7
5	0.11	>10	>91	2.2	>10	1.1	0.31	0.45
6	0.57	>100	>175	>10	>10	2.2	NT	1.5
7	0.10	>100	>1000	0.54	>10	0.58	0.30	0.28
8	0.030	>10	>333	0.58	>10	4.6	0.060	0.050
9	>10	>100	NC	>10	NT	NT	NB	NB
10	0.20	94	470	1.9	>10	1.0	1.7	0.54
11	1.7	100	60	>10	>10	4.5	1.2	3.2
12	0.020	77	3850	0.29	>10	0.88	0.060	0.060
13	0.23	>100	>435	2.3	>10	3.1	1.5	1.5
aRMP	0.030	>122	>4066	>4	0.24	0.12	NT	NT
bINH	0.46	>50	>109	>8	0.99	4.00	NT	NT

^aRifampin (RMP) and

^bisoniazid (INH) as positive controls. Selectivity index (SI) = IC₅₀/MIC; NT-not tested; NB-no binding observed; NC-not calculated.

The binding affinities (K_D) of all isolated RUFs with both the N-terminal domain (NTD) and the full-length (FL) caseinolytic protein C1 (ClpC1) were assessed by surface plasmon resonance (SPR), using a previously reported method (Table 4).¹¹ The K_D values of all tested RUFs varied from 0.050 μM to 5.7 μM. Not surprisingly, the RUFs with low MICs (8 and 12) showed the tightest binding affinities with both ClpC1-NTD and ClpC1-FL at 0.050−0.060 μM, supporting the prior conclusion that RUFs specifically target the ClpC1-NTD domain (Table 4). In order to establish selectivity indices, the cytotoxicity of the isolates was evaluated in a mammalian cell line (Vero cells; ATCC CRL-81). The results indicated a high selectivity of RUFs against mycobacteria. While for the first seven days, the RUF analogues inhibited M. tuberculosis growth, subsequent incubation resulted in cell death. Comparison of the minimum bactericidal concentrations (MBCs) on days 7 and 14 revealed the time-dependent bactericidal activity of the RUFs. Importantly, the RUFs also showed strong activity against M. abscessus, the cause of many drug-resistant non-tuberculous mycobacterial infections.

Concluding Remarks

With the isolation and identification of eight new compounds (1–8), this study expands the chemical space of RUFs to twelve natural congeners. This has enabled SAR studies aimed at optimizing ClpC1 binding affinities. The newly characterized RUFs, targeting the NTD of ClpC1, exhibited variable in vitro anti-TB activity and selectivity indices. While some of the new compounds exhibited similarly high potencies to 12, no new analogues with even higher potency were encountered. Considering that this study mined the minor constituents of the S. atratus strain MJU3502 metabolome, the outcomes indicate that 12 might be the most potent of the naturally produced RUFs. Semi-synthetic modifications of RUFs are one logical next step and are ongoing in our laboratory. Notably, the anti-M. abscessus activities of the RUFs highlight the potential of RUFs to serve as more universal antimycobacterial drug leads. In addition to structural modification, other ongoing work includes microsome stability, and PK/PD studies of this class of potential anti-TB leads.

EXPERIMENTAL SECTION

General Experimental Procedures

Fourier transform infrared (FTIR) spectra were recorded on a Nicolet 6700 FTIR spectrophotometer. Optical rotation values were measured on a Perkin-Elmer 241 polarimeter at room temperature; concentrations are reported in g/100 mL. All 1D/2D NMR spectra were acquired on a JEOL (JEOL Resonance Inc., Peabody, MA, USA) ECZ 400S spectrometer. For UHPLC analysis, a Shimadzu (Kyoto, Japan) Nexera UHPLC-UV system equipped with a DAD was used, employing a Kinetex 1.7 μm XB-C₁₈ 100 Å column (50 mm × 2.1 mm) and LabSolutions software for data analysis. The column oven, detector cell, and autosampler temperature were maintained at 40 °C, 40 °C and 4 °C, respectively, throughout the analysis. UHPLC-HRESIMS and ESIMS/MS spectra were carried out by using Bruker Impact II, quadrupole time of flight (qTOF) equipped with a Shimadzu UHPLC (Kyoto, Japan). The ion source was operated in the positive electrospray ionization mode using capillary voltage at 4.0 kV; nebulizer and drying gas (N₂) at 0.4 bar and 4.0 L/min, respectively; dry temperature of 225 °C; and mass scan range set from m/z 50 to 2000. The MS/MS spectra were acquired using varied collision energy ranging from 20 to 70.0 eV. 10 mM sodium formate solution introduced to the ion source as the internal calibration. The separation was performed on a CORTECS C₁₈ (100 × 3.0 mm, 2.7 μm) UHPLC column. Data were collected and processed by the Data Analysis 4.4 software (Bruker Daltonik GmbH, Germany). Sephadex LH-20 (Pharmacia, Uppsala, Sweden) and Silica gel (ICN EcoChrom 32–63, 60 Å) was used for column chromatography. Semi-preparative HPLC was performed on a Shimadzu HPLC (Kyoto, Japan) connected to a PDA detector (Shimadzu, model SPD-20A) and equipped with a Kinetex EVO C₁₈ (250 × 10 mm, S-5, 100 Å) column. TLC was analyzed by UV detector and vanillin-sulfuric acid spray (3 g vanillin, 95 mL ethanol, and 1.5 mL sulfuric acid). All solvents used were obtained from Fisher Scientific (Fair Lawn, NJ, USA) or Sigma-Aldrich (St. Louis, MO, USA). Nα-(2,4-Dinitro-5-fluorophenyl)-l-leucinamide (l-FDLA) and Nα-(2,4-Dinitro-5-fluorophenyl)-d-leucinamide (d-FDLA) were purchased from Arctom Chemicals. Amino acids standards were purchased from Sigma and AmBeed.

Strain Material

The strain MJM3502 was obtained from the Extract Collection of Useful Microorganisms (ECUM) at Myongji University, Republic of Korea, and was fermented in glucose-soybean starch (GSS) medium (rich medium). The culture medium supernatants were extracted with ethyl acetate and dried. The Streptomyces strain MJM3502 was identified as being identical with Streptomyces atratus (NRRL B-16927) through classification using the 16S rDNA sequence.¹²

Extraction and Isolation

The whole broth of a 500 L culture of MJM3502 was extracted with a 2 times volume of methanol, and then concentrated in vacuum at 40 ºC to 1/10^th volume. After adding deionized water to the concentrated solution to achieve 90% methanol, the mixture was defatted by partitioning with ethyl acetate. The methanol-water layer was extracted with one volume of ethyl acetate twice and this was cleaned by shaking with one volume of deionized water (pH 3.0). The ethyl acetate layer was evaporated to dryness to yield a yellowish solid. A RUF enriched fraction (28 g; aliquot of 41g extracted from a total of 350L of culture broth) was obtained from 3502 EtOAc extract by silica gel CC with a hexane/EtOAc (5:5) eluent. The RUF-enriched fraction was further chromatographed on silica gel using a gradient elution of hexane/acetone (7:1, 6:1, 5:1, 4:1, 3:1, 2:1, 1:1, 1:2, and 1:5) to give six fractions (A−F). From fraction B (3 g), 300 mg of rufomyazine (9) were purified by recrystallization. The remainder was separated by silica gel CC using a gradient elution of CHCl₃/MeOH (200:1, 100:1, 50:1, 30:1, 20:1, and 10:1), then further purified by semi-preparative HPLC (60% CH₃CN in H₂O, 2.5 mL/min) to give compounds 4 (20 mg) and 10 (50 mg). Similarly, ~6 g of RUF I and II mixtures (12) were purified from fraction C (10 g) by a series of columns packed with silica gel (CHCl₃/MeOH) and Sephadex LH-20 (MeOH or EtOH). Compounds 6 (3.5 mg), 11 (2.1 mg), and 13 (8.1 mg) were then purified by semi-prep HPLC (60% CH₃CN in H₂O, 2.5 mL/min) from the remaining material of fraction C. Fractions D (1.0 g), E (1.7 g), and F (1.7g) were each chromatographed on a silica gel column using a gradient elution of CHCl₃/MeOH (100:1, 50:1, 30:1, 20:1, and 10:1), then purified by semi-preparative HPLC (45−60% CH₃CN in H₂O, 2.5 mL/min) to give 1 (100 mg), 2 (10 mg), 3 (10 mg), 5 (12 mg), 7 (3 mg), and 8 (10 mg).

Rufomycin NBZ1 (1): pale yellow crystals; [α]²⁵_D −82 (c 1.2, MeOH); IR (KBr) ν_max 3275, 2955, 2872, 1628, 1539, 1512, 1456, 1327, 1244, 1084, 966, 758 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 1; (+)-HRESIMS m/z 1044.5760 [M + H]⁺ (calcd for C₅₄H₇₈N₉O₁₂, 1044.5764).

Rufomycin NBZ2 (2): pale yellow, amorphous solid; [α]²⁵_D −41 (c 0.6, MeOH); IR (KBr) ν_max 3275, 2955, 1628, 1539, 1456, 1317, 1245, 1207, 968, 750 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 1; (+)-HRESIMS m/z 1044.5782 [M + H]⁺ (calcd for C₅₄H₇₈N₉O₁₂, 1044.5764).

Rufomycin NBZ3 (3): pale yellow, amorphous solid; [α]²⁵_D −118 (c 0.3, MeOH); IR (KBr) ν_max 3271, 2927, 2856, 1630, 1537, 1456, 1317, 1205, 1084, 968, 750 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 1; (+)-HRESIMS m/z 1028.5853 [M + H]⁺ (calcd for C₅₄H₇₈N₉O₁₁, 1028.5815).

Rufomycin NBZ4 (4): white, amorphous solid; [α]²⁵_D −70 (c 0.2, MeOH); IR (KBr) ν_max 3270, 2956, 2864, 1629, 1516, 1456, 1234, 968, 922, 750 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 1; (+)-HRESIMS m/z 967.5990 [M + H]⁺ (calcd for C₅₄H₇₉N₈O₈, 967.6015).

Rufomycin NBZ5 (5): pale yellow, amorphous solid; [α]²⁵_D −70 (c 1.1, MeOH); IR (KBr) ν_max 3271, 2956, 1628, 1537, 1456, 1315, 1244, 1207, 1182, 1072, 966, 748 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 2; (+)-HRESIMS m/z 1088.6021 [M + H]⁺ (calcd for C₅₆H₈₂N₉O₁₃, 1088.6027).

Rufomycin NBZ6 (6): pale yellow, amorphous solid; [α]²⁵_D −81 (c 0.1, MeOH); IR (KBr) ν_max 3286, 1628, 1522, 1433, 1327, 1234, 1171, 1089, 752 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 2; (+)-HRESIMS m/z 1014.5695 [M + H]⁺ (calcd for C₅₃H₇₆N₉O₁₁, 1014.5659).

Rufomycin NBZ7 (7): pale yellow, amorphous solid; [α]²⁵_D −57 (c 0.1, MeOH); IR (KBr) ν_max 3304, 2929, 2872, 1539, 1456, 1323, 1257, 1082, 956, 744 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 2; (+)-HRESIMS m/z 1060.5703 [M + H]⁺ (calcd for C₅₄H₇₈N₉O₁₃, 1060.5714).

Rufomycin NBZ8 (8): pale yellow, amorphous solid; [α]²⁵_D −75 (c 0.2, MeOH); IR (KBr) ν_max 3284, 2931, 2871, 1626, 1537, 1491, 1456, 1315, 1255, 1207, 1072, 968, 751 cm⁻¹; ¹H and ¹³C NMR (CD₃OD), see Table 2; (+)-HRESIMS m/z 1058.5602 [M + H]⁺ (calcd for C₅₄H₇₆N₉O₁₃, 1058.5557).

Determination of the Absolute Configuration of the Amino Acids by the Advanced Marfey’s Method

The l-and d-FDLA derivatives of amino acid standards (l-Alanine, l-Leucine, l-mNO₂-Tyrosine, and l-NMeLeu) were prepared as reported previously1¹⁷ and documented in the Supporting Information. The derivatives were then analyzed by LC-ESIMS. The observed MS and retention times of the l/d-FDLA amino acid standards (t_R, l-/d-) were as follows: Ala (6.1/8.2 min), mNO₂Tyr (9.5/12.5 min), Leu (10.2/16.8 min), and NMeLeu (12.7/15.6 min) (Figure S1 and Table S1, Supporting Information).

Compounds 1−8 (~0.2 mg each) in 12 N HCl (1 mL) were heated to 110 °C for 20 min in a CEM Explorer 48/72/96 automated microwave synthesizer (CEM Corporation, Matthews, NC, USA) utilizing a floor-mounted IR temperature sensor controlled by an external computer loaded with Synergy application software (v 1.1). Samples were transferred to HPLC vials and then dried by air. H₂O (50 μL) and 1 M NaHCO₃ (200 μL) were added to each vial, and then divided into two equal portions. The Marfey’s solution was prepared with 2 mL of acetone and 5 mg of Marfey’s reagent. A 100 μL aliquot of Marfey’s solution (l- and d-FDLA) was added to the two corresponding portions of each sample, respectively. The vials were heated to 40 °C for 1 h then quenched with 200 μL 1 M HCl. Finally, the reaction products were diluted (200 μL CH₃CN) and filtered prior to LC-MS analysis. Aliquots (5 μL) of each sample were injected to LC-MS for analysis (Supporting Information). An l-absolute configuration for Ala, Leu, NMeLeu, Tyr, mNO₂Tyr, Trp, isoprenylTrp, and TrcGly were assigned (Figures S19, S28, S37, S46, S54, S64, and S74, Supporting Information) by analysis of the MS and retention time.

Crystal Structure Analysis

Pale yellow crystals of 1 were obtained by recrystallization in MeOH at 4 ^oC. Data collection was carried out at beam line 21-ID-D, LS-CAT, Advanced Photon Source, Argonne National Laboratory, using an Eiger 9M Detector, with a wavelength of 0.68877 Å and a crystal-to-detector distance of 100 mm. The structure of 1 was directly elucidated via the method of SHELXS and refined with SHELXL-2014. The X-ray data of 1 have been submitted to the Cambridge Crystallographic Data Center (CCDC 1957302), and are available free of charge via the Internet at www. https://www.ccdc.cam.ac.uk/.

MICs against M. tuberculosis and M. abscessus (ATCC 19977)

The MIC was defined as the minimum concentration of the compound required to achieve a reduction in fluorescence by 90% relative to the untreated bacterial controls. The anti-TB activity was determined by the microplate Alamar Blue assay (MABA) as previously described.^13,19

Cytotoxicity in Mammalian cells

Cytotoxicity was tested using Vero cells (ATCC CRL-81) based on previous reported method. The concentration of test compound effecting a reduction in fluorescence of 50% relative to untreated cells (IC₅₀) was calculated.^20,21

MBC against M. tuberculosis

Luminescent reporter strain H37Rv carrying a plasmid pMV306hsp+LuxAB+G13+CDEMBCs²² was used for assessing bacterial viability. The recombinant H37Rv strain was incubated with each test compound for two weeks. On days 7 and 14, luminescence was measured using a Berthold luminometer. The minimum bactericidal concentration (MBC) was defined as the lowest concentration reducing relative luminescence unit (RLU) by 99% relative to the RLU of bacterial inoculum at the start of the experiment. MBCs were calculated on days 7 and 14.

Analysis of Binding Affinity to Mycobacterial ClpC1-NTD and FL by SPR

SPR binding assay was performed using a previously reported method.^11,13 Kinetic rate constants (k_a and k_d) were determined by fitting the double-reference data globally to the 1:1 Langmuir model embedded in the Biacore T200 evaluation software (v3.0). K_D values were then calculated from the two rate constants (K_D = k_d/k_a). Smaller K_D values represent tighter binding affinities.