SIRT2 deletion enhances KRAS-induced tumorigenesis in vivo by regulating K147 acetylation status

The observation that cellular transformation depends on breaching a crucial KRAS activity threshold, along with the finding that only a small percentage of cellsharboring KRAS mutations are transformed, support the idea that additional, not fully uncovered, regulatory mechanisms may contribute to KRAS activation. Here we report that KrasG12D mice lacking Sirt2 show an aggressive tumorigenic phenotype as compared to KrasG12D mice. This phenotype includes increased proliferation, KRAS acetylation, and activation of RAS downstream signaling markers. Mechanistically, KRAS K147 is identified as a novel SIRT2-specific deacetylation target by mass spectrometry, whereas its acetylation status directly regulates KRAS activity, ultimately exerting an impact on cellular behavior as revealed by cell proliferation, colony formation, and tumor growth. Given the significance of KRAS activity as a driver in tumorigenesis, identification of K147 acetylation as a novel post-translational modification directed by SIRT2 in vivo may provide a better understanding of the mechanistic link regarding the crosstalk between non-genetic and genetic factors in KRAS driven tumors.

are deleted and Kras G12D is expressed in the pancreas simultaneously. Mice were housed, fed, and treated in accordance with the guidelines approved by the Northwestern University IACUC.

Immunohistochemistry
For immunohistochemistry (IHC), sections were stained using the protocol described below. Briefly, slides were heated via pressure cooker in DAKO retrieval buffer, and endogenous peroxidases quenched in 3% hydrogen peroxide in methanol for 30 min. Tissues were blocked with 5% BSA in PBS for 30 min and exposed to primary antibodies against BrdU

Cell treatments
To stimulate KRAS signaling, cells were first starved for 24 h in 0.1% FBS medium followed by treatment with EGF (100 ng/mL) (Sigma). To inhibit the activity of endogenous sirtuins, cells were treated with 2 µM nicotinamide (NAM) (Sigma) for 12 h before lysing the cells. To specifically inhibit SIRT2, cells were treated with 5 µM AGK2 (Santa Cruz) for 12 h before cell lysis.

Intranasal infection
AdenoCRE (2.5x10 7 particles, Gene Transfer Vector Core, University of Iowa) was first mixed with Minimum Essential Medium (MEM) and 2 M CaCl 2 (final concentration 10 mM) in a final volume of 125 µL/mouse infected. Using a protein gel loading tip, half of the virus mixture (62.5 µL) was administered by placing the pipette tip at the opening of one nostril. The virus solution was slowly expelled from the tip while the mouse inhaled the drop that was forming. Mice were left to recover for 10-15 minutes. After breathing returned to normal, the procedure was repeated with the remaining 62.5 µL of the virus mixture. Mice were monitored for up to 6 months. All procedures were performed in accordance with approved Northwestern University IACUC protocol.

Western blotting
Cells or tissue samples were lysed using buffer A (25 mM Tris-HCl pH 7.5, 150 mM NaCl, 1 mM EDTA, 0.5% NP-40, and 5% glycerol), followed by sonication and centrifugation at 14,000 rpm for 25 min at 4 o C. After protein quantification using the Bradford assay, equal amounts of protein (20-40 µg) were mixed with 2 Laemmli lysis buffer, boiled for 5 min, separated on a denaturing SDS-polyacrylamide gel, and transferred to a PVDF membrane. The membrane was blocked in 5% milk/PBS/0.05% Tween for 1 h and incubated with antibodies against pERK,

Cellular transformation
Transformation ability of NIH3T3 cells was assessed by checking colonies formed by confluent cells. A total of 5 x 10 5 cells/well were seeded in 6-well tissue culture plates in complete medium which was replaced every 3-4 days. Fourteen days post-confluence, the cells were fixed and stained with crystal violet (Sigma), and the number of colonies composed of >50 cells was determined. All assays were performed three times. The mean ± SD was calculated from three independent experiments. Cellular transformation was also assessed by observing anchorageindependent growth using the soft agar assay. Briefly, 10 4 cells/well were seeded into 0.35% low melting point agarose (Lonza) on top of a 0.5% agarose layer in 6-well plates. After 3 weeks, colonies were fixed in methanol and stained with 0.1% crystal violet. For quantification, crystal violet-positive colonies were counted. All assays were performed three times. The mean ± SD was calculated from three independent experiments.

Deacetylation assays
Both in vitro and in vivo deacetylation assays were performed as described previously (5).

Dot blot assay
Samples (2 µL containing 10 ng -0.5 µg protein) were spotted onto a nitrocellulose membrane at the center of the grid using a narrow-mouth pipette tip. Membrane was left to dry followed by blocking of non-specific sites after incubation with 5% milk in Tris-Buffered Saline/Tween 20 (TBST) for 0.5-1 h at room temperature. Next, the membrane was incubated with the primary antibody diluted in 1% milk in TBST for 30 min at room temperature. After washing three times with TBST (3 x 5 min), the membrane was incubated for 30 min at room temperature with a secondary antibody conjugated to HRP. Finally, membranes were exposed to x-ray films following incubation with a chemiluminiscence reagent.

Custom anti-Ac-K147 antibody development
The anti-Ac-K147 antibody was generated by Eurogentec following the Speedy 28-day protocol.
For immunization of rabbits, a peptide carrying acetylated K147 (N-IETSAK(Ac)TRQGVD-C) as well as a control non-acetylated peptide were synthesized and attached to keyhole limpet hemocyaninand (KLH) carrier. After purification, the raised antibody, together with the synthesized peptides used for immunization, was shipped to our laboratory where they were used in additional validation experiments.

GDP/GTP exchange
Cells were lysed following the same procedure as described for the Ras activity assay.

Mass spectrometry
KRAS G12V was purified from 293T cells expressing either Flag-KRAS G12V alone or Flag-KRAS G12V and SIRT2. For purification, an anti-Flag antibody covalently attached to agarose (ANTI-FLAG ® M2 affinity gel, Sigma) was used to immunoprecipitate KRAS G12V , followed by elution with excess free Flag peptide according to the procedure described in (6). Eluted samples were run on a 4-20% polyacrylamide gel, followed by silver staining (ThermoFisher Scientific).
Protein bands were cut and sent to the MSRC Proteomics Laboratory at Vanderbilt University for mass spectrometry analysis as described before (6).

Cloning, expression, and purification of KRAS proteins
KRAS, KRAS K147Q , KRAS G12V , and KRAS G12V/K147Q gBlocks ® gene fragments (IDT) were cloned into pMCSG7 vector. Each plasmid was digested with SspI restriction enzymes and the linearized vector was gel-purified. Ten μL of Gibson assembly ® master mix (New England Biolabs), 1 pmol pMCSG7 linearized vector, and 3 pmol of each gBlock product were incubated for 1 h at 50 °C. One μL of the Gibson assembly ® reaction mix was transformed into TOP10 electro-competent cells (Sigma). All the sequences were validated by DNA sequencing. To improve expression and solubility of the KRAS proteins, the catalytic domain of each KRAS construct (residues 1-169) was amplified from purified pMSCG7, using the following primers:   were determined after immunoprecipitation using an anti-Ac-K antibody followed by western blotting using an anti-HA antibody. Interaction of both KRAS and KRAS G12V with SIRT2 was confirmed after immunoprecipitation using an anti-HA antibody followed by western blotting using an anti-Flag antibody, whereas successful immunoprecipitation of both KRAS and KRAS G12V was confirmed by western blotting using an anti-HA antibody. Levels of expressed SIRT2 are shown after immunoblotting using an anti-Flag antibody. (J) 293T cells were transfected with either Flag-KRAS (upper) or Flag-KRAS G12V (lower) and cells were subsequently treated with nicotinamide and AGK2. The relative amounts of GTP-bound active Ras were determined through a specific protein interaction with Raf1-RBD. Levels of expressed KRAS and KRAS G12V are shown after immunoblotting using an anti-Flag antibody. (K) 293T cells were co-transfected with Flag-KRAS, HATs, and either HA-SIRT2 wt (wild-type SIRT2) or HA-SIRT2 dn (deacetylation null mutant SIRT2). KRAS activity was determined as described in (J). Levels of expressed KRAS and SIRT2 are shown after immunoblotting using anti-Flag and anti-HA antibodies, respectively. Tubulin is used as loading control. After immunoprecipitation using an anti-Flag antibody, samples were separated on a gel, and bands corresponding to Flag-tagged KRAS G12V were excised, in-gel trypsin digested, and analyzed by mass spectrometry. Identified peptides are highlighted in yellow, and acetylated lysines are indicated by red circles. K104 and K147 were found to be acetylated in sample 1, whereas K147 acetylation was not detected in sample 2. (B, C) Spectra for both sites found to be acetylated in sample 1 are shown.