i. PubChem Bioassay Name(s), AID(s), Assay-Type (Primary, DR, Counterscreen, Secondary)

The primary assay was an NMR-based chemical shift perturbation assay (AID-1033) Spectra were acquired on a 600 MHz Bruker Avance equipped with TCI cryoprobe. Ligand binding was monitored by comparing the aliphatic region of 1D ¹³C-filtered ¹H NMR spectra of 20 μM DnaK β-domain only protein solution (20 mM sodium phosphate buffer at pH 7.5 containing 90%/10% H₂O/D₂O or 99.5% D₂O; T= 300 K) in the presence or absence of 80 μM mixtures of 10 compounds, and then individual compound mixtures that caused significant perturbations in the spectrum were characterized further. Simple 1D ¹H NMR experiments of the protein measured in presence and absence of mixtures of potential ligands, observing the aliphatic region of the spectra δ<1 ppm). Compound or mixtures causing spectral perturbation above a threshold of 0.08 ppm were deconvoluted.

ii. Assay Rationale & Description

The first secondary assay was to confirm binding by NMR by obtaining K_d’s using isothermal calorimetry (ITC; AID-1495). Titrations were done using a VP-ITC from MicroCal (Northampton, MA). Full-length DnaK or SBD were used at 100 μM in 20 mM sodium phosphate buffer (pH 7.4) and 0.5–5% DMSO. Compounds were used at 10–15x in the same buffer.

The ATPase activity (AID-1494) of confirmed binders were determined using the ADAPTA™ kit from Invitrogen at 200 µM of compounds. Positive inhibition would suggest allosteric cross-talk between the domains of DnaK even though compounds bind to only the SBD. This assay cannot be considered a bona fide secondary assay because the other co-chaperone proteins are involved in the process and substrate and cofactor binding and release.

iii. Summary of Results

Using a construct of DnaK corresponding to the β-domain only (residues 393–507) whose solution structures with and without bound peptide are known (Pellecchia 2000, Stevens 2003) a simple 1D ¹H NMR based screen of an assembled a scaffold library composed of ~4,000 member was completed by monitoring the aliphatic region of the spectra δ<1 ppm) in the presence and absence pools of 10 compounds at 80 µM per compound binding to 10–20 µM of protein. Mixtures causing spectral perturbation above the threshold of 0.08 ppm were deconvoluted, and additionally reconfirmed for binding by ITC. The compound BI-88B12 emerged as a hit from the screen (Table I, in appendices).

After identifying initial binders, chemical shift mapping studies with ¹³C and ¹⁵N labeled protein suggested that these compounds do not bind to the substrate binding pocket but rather to a pocket located around loop L2,3 on the opposite side of the SBD (Figure 1, A and B). Further molecular docking studies predict binding of compounds BI-88B12 and BI-88D7 in a deep pocket near residues Leu484 and Pro419 reported to be essential for the allosteric communication between the substrate binding and the ATPase domains (Swain 2007, Vogel 2006a,b). These studies also suggested these two compounds bind in different orientations. BI-88B12 places its thiophene ring outside the pocket, this moiety is inserted into the pocket in BI-88D7. For BI-88B12, the fluorobenzene group is inserted deep into the hydrophobic pocket as evidenced by the strongest perturbation of residues there. With its thiophene group in the pocket, the indole ring of BI-88D7 is also able to interact with Asp481, located further out near the edge of the pocket. BI-88D7 causes equal perturbation of both Leu484 and Asp481, possibly due to its increased size that allows it to bridge both residues. This result possibly explains the much lower K_d of BI-88D7 as determined by ITC (see Table 1).

Interestingly, BI-88B12 was indeed found to be an inhibitor of the DnaK ATPase activity with an IC50 of 2 mM, whereas BI-88D7 did not give significant ATPase activity even at 20 mM (Table, below). Interestingly, further studies suggest that the binding of probe BI88D10 alters allosterically the affinity for ATP by ITC (please refer to table 2 of the attached pre-publication manuscript from the assay provider.

i. SAR & chemistry strategy that led to the probe

An additional 5 commercially available BI-88B12 analogs were qualitatively assessed for their ability to bind the β-domain of DnaK by the magnitude of the chemical shifts induced upon complexation on the ¹³C^δ,¹H^δ resonances of Leu484 (1:10 protein to ligand ratio, at 70 μM protein concentration) by 2D [¹³C, ¹H] correlation spectra. Hit compounds with estimated K_d values of 200 μM or less were selected for further SAR studies including K_d determination by ITC. These data were further analyzed to design 10 additional analogues that were for chemical syntheses. The various ID numbers, structures, molecular weights, activity data and source (commercial or synthesized) for the selected compounds are summarized in Table 1 below:

Table 1Selected purchased & synthesized compounds for SAR Development of confirmed hit BI-88B12

View in own window

BI88	CID [SID]	MLS – #	Structure	MW	ATPase IC50 (mM)	ITC Kd (μM)	Source
B12 Initial Hit	1244199 [56427257]	0014807		235	2	64.1 ± 8.7 (SBD)	ChemBridge
E3	743472 [56427258]	0100672		260	>20	25 (SBD) 4.9	ChemBridge
D7	2814135 [56427263]	0311872		242	>20	(SBD) 12.7 (FL)	Internally Synthesized
D10 Probe	25105719 [56427267]	0315930		273	>20	0.2 (FL)	Internally Synthesized
D5	25105722 [56427270]	0315933		267	>20	1.0 (SBD) 1.7 (FL)	Internally Synthesized
D9	25105724 [56427272]	0315935		255	>20	1.5 (SBD)	Internally Synthesized
C5	1246750 [56427262]	0315926		296	0.2–2	14.3 (SBD) 39.3 (FL)	ChemBridge

Of the commercial analogs obtained, BI-88E3 (containing a furan moiety) improved the potency of binding to the SBD of DnaK, but with a loss of ATPase activity. However, BI-88C5 uncovered a compound with improved binding affinity to the DnaK SBD and showed binding to the full-length DnaK, while also recovering a significant amount of ATPase activity, and suggested that substitution of the furan by a thiophene may be a good strategy to improve potency and ATPase activity, though the previous docking studies indicated that despite a shared thiophene ring compounds could bind in different orientations with respect to this thiophene.

Of the ten additional synthesized compounds, significant improvement in affinity to DnaK of about 10-fold was achieved with compounds BI-88D5 and BI-88D9 over the original hit BI-88B12. The most dramatic improvement (>200-fold) was achieved with BI-88D10, though with loss of significant ATPase activity.

Chemical Synthesis

To a stirred solution of free amine (1.0 equiv.) and Et₃N (2.0 equiv.) in DCM was added a solution of thiophene-2-carbonyl chloride (1.1 equiv.) in DCM at −30 °C. The resulting solution was stirred for 1h and then allowed to warm up to room temperature. After removal of the solvent, the residue was purified by flash column chromatography to provide the correspondent product (yield 75–95%).

Probe molecule (CID-25105719)

BI-88D10-N-(benzo[b]thiophen-7-ylmethyl)thiophene-2-carboxamide ¹H NMR (CDCl₃, 600MHz): δ 8.35 (s, 1H), 7.82 (m, 5H), 7.47 (s, 1H), 6.97 (s, 1H), 5.30 (s, 2H); HRESI-TOF-MS: calcd for C₁₄H₁₁NOS₂ 274.0355 [M + H]⁺, found 274.0361.

BI-88D5- N-(naphthalen-1-ylmethyl)thiophene-2-carboxamide ¹H NMR (CDCl₃, 600MHz): δ 8.17 (s, 1H), 7.87 (s, 1H), 7.73 (s, 1H), 7.46 (m, 6H), 6.99 (s, 1 H), 6.39 (s, 1H), 5.01 (s, 2H); HRESI-TOF-MS: calcd for C₁₆H₁₃NOS 268.0791 [M + H]+, found 268.0796.

BI-88D7- N-(1H-indol-5-yl)thiophene-2-carboxamide ¹H NMR (CDCl₃, 600MHz): δ 8.19 (s, 1H), 7.92 (s, 1H), 7.71 (s, 1H), 7.62 (s, 1H), 7.52 (s, 1H), 7.33 (m, 2H), 7.23 (s, 1H), 7.12 (s, 1H), 6.54 (s, 1H); HRESI-TOF-MS: calcd for C₁₃H₁₀N₂OS 243.0587 [M + H]+, found 243.0592. BI-88D6- Is an isomer and side product of BI-88D7 synthesis.

BI-88D9- N-(quinazolin-2-yl)thiophene-2-carboxamide ¹H NMR (CDCl₃, 600MHz): δ 9.35 (s, 1H), 7.90 (m, 3H), 7.61 (m, 4H), 6.89 (s, 1H); HRESI-TOF-MS: calcd for C₁₃H₉N₃OS 256.0539 [M + H]⁺, found 256.0545.