Target discovery of acivicin in cancer cells elucidates its mechanism of growth inhibition

Using a chemical proteomic strategy we analyzed the targets of acivicin and provided a mechanistic explanation for its inhibition of cancer cell growth.

The solvent was removed under reduced pressure and gave 15 mg compound 4 as a yellow oil (97%). The product was used without further purification.  HPLC t R = 6.7 min (2-50% MeCN, 10 min) Preparative: Cells were grown in Petri dishes (150 mm) until they reached 70% confluency.
Then the media was exchanged with 10 mL media with appropriate probe concentration and media lacking the probe as control. After incubation (24 h, 48 h, 72 h, 96 h and 120 h) the cells were washed with 10 mL PBS and harvested by scraping in 20 mL PBS. Cells were pelleted by centrifugation at 800 g for 5 min. The cells were lyzed by resuspending in 500 µL lysis buffer and incubation for 10 min on ice. Soluble and insoluble fraction were isolated by centrifugation at 21000 g for 20 min at 5 °C and insoluble fraction was resuspended in 50 µL lysis buffer by sonication under ice cooling. The samples then underwent subsequently Click reaction and preparative gel-based analysis (see below).

In situ ABPP labeling experiments
Analytical labeling in PBS: Cells were grown in 6 well plates until a confluency of 70% was Las-4000 luminescent image analyser containing a VRF43LMD3 lens and a 575DF20 filter.

Mouse liver lysate
300 mg of mouse liver were cut into smaller pieces and added to a 2 mL tube containing ceramic beads (91-PCS-CKM, peqlab) and 1 mL cold PBS was added. The liver was homogenized by a Precellys 24 homogenizer (peqlab) with 5000 rpm for 10 sec in two cycles.

9
Soluble and insoluble fraction were separated by spinning at 21000 g for 20 min at 5 °C. The insoluble pellet was resuspended in 1 mL PBS by sonication. Protein concentration was assayed (Rotiquant universal, Carl Roth Laborbedarf) and adjusted to 2 mg/mL in PBS.

In vitro labeling experiments
Analytical: 1 µl probe stock was added to 44 µL proteome to adjust the desired concentration and incubated at RT for 1 h. After incubation Click reaction and analytical gel-based analysis (see above) was performed.
Preparative: 1 µL probe stock was added to 946 µL proteome solution to get the appropriate probe concentration (50 µM for ACVL probes and 100 µM ACV2 and 20 µM ACV1) with one sample lacking the probe as control. The solution was incubated for 1 h at RT.
Subsequently Click reaction and preparative gel-based analysis (see above) was performed.

Competitive labeling experiments
43 uL of proteome solution were incubated with 1 µL of acivicin stock to adjust the desired concentration (see gel below) and incubated for 15 min at RT. After pre-incubation 1 µL of probe stock was added to adjust the desired probe concentration and incubated for 1 h at RT.
After that Click reaction and analytical gel-based analysis (see above) was performed. After completion Click reaction and analytical gel-based analysis (see above) was performed.

Labeling of recombinant CES1 for 24 h
Three samples were prepared by adding 1 µL of CES1 stock (5 µg/µL) to 43 µl PBS and two samples by adding 1 µL CES1 stock to 43 µL A549 lysate. One sample of each set served as control lacking probe and was incubated with 1 µL DMSO. Another sample from each set was incubated with ACV2 at a concentration of 20 µM and incubated for 24 h at RT. As a positive control one sample of CES1 in PBS was labeled with 10 µM FP probe for 1 h at RT.
After completion of incubation Click reaction and analytical gel-based analysis (see above) was performed.

ALDH-Assay
For

Expression level determination
HepG2 cells were cultivated in 6 wells until 70% confluency. Media was exchanged and 25 µM ACV1 and 10 µM ACV2 were added 1:1000 from DMSO stocks and one 1 mL media

Overexpression for target validation
Expression clones ALDH1A1, ALDH1B1, ALDH2, ALDH4A1 and ACAA2 were grown in 10 mL ampicillin LB at 37 °C with two cultures for each gene until OD 600 of 0.5 and subsequently one sample was induced with anhydrotetracycline and cultivated for additional 3 h the corresponding sample was cultivated the additional 3 h without induction. The bacterial cell pellets were washed with PBS and resuspended in 100 µL PBS containing 50 µM ACVL1 for ALDH1A1, ALDH1B1, ALDH2; 50 µM ACVL2a for ACAA2 and 50 µM ACV1 for ALDH4A1. The cells were incubated for 2 h at RT and lyzed by sonication.
Soluble and insoluble fraction were separated by centrifugation (21000 g for 20 min at 5 °C).
Insoluble pellets were resuspended in 100 µL PBS using sonication. The samples underwent Click reaction and analytical gel-based analysis (see above).

Overexpression and purification of ALDH4A1
Cells were grown in ampicillin M9 media at 37 °C until an OD600 of 0.7, then induced with

Overexpression and purification of ALDH1A1
ALDH1A1 was purified as described previously. 4

Metabolic labeling
In

Mutagenesis of CES1 and ALDH4A1
To analyze the binding site of ACV2 in CES1 the active site serine was converted to alanine by mutagenesis using the following primers to create the mutant S221A: Forward primer: 5'-ctttctcctcccgctgcctctccaaagatggtc-3'

14
Reverse primer: 5'-gaccatctttggagaggcagcgggaggagaaag-3' For ALDH4A1 the nucleophilic cysteine was exchanged by alanine to generate the mutant C348A: Forward primer: 5'-cgagcacgcggaagccttctggccaccg-3' Reverse primer: 5'-cggtggccagaaggcttccgcgtgctcg-3' The point mutations were incorporated by PCR on the corresponding plasmid created before (CES1 on pT-Rex™-DEST30 vector and ALDH4A1 on pDEST 007 vector). After PCR a DpnI digest was performed according to manufacturer's instructions and plasmids were transformed and amplified in chemically competent E. coli XL1 as described above.

Labeling of CES1 and mutant S221A
In a 6 well plate with HEK293T cells two wells were transfected with CES1(wt) and one well

Labeling of ALDH4A1 and mutant C348A
pDEST 007 plasmid carrying ALDH4A1 (wt) and C348A were transformed into competent SoluBL21™ Competent E. coli cells (Amsbio) and overexpressed as described above To evaluate knockdown efficiency 40 µg lysate were labeled with 10 µM FP for CES1 and 50 µM ACV1 for ALDH4A1 for 1 h and Click reaction and analytical gel-based analysis and Immunoblotting was carried out with actin as loading control (see above).

In gel digestion
Protein bands were excised from the gel and subsequently washed with ddH 2 O (100 µL, and added to the pre equilibrated filters and centrifuged (21000 g, 1 min, RT). The filtrate was transferred into a vial.

Mass spectrometry and bioinformatics
Measurements were performed using an Orbitrap XL coupled online to an Ultimate 3000