Human AML cell lines LC-MS/MS
As a direct consequence of the high diversity of the aggressive blood cancer acute myeloid leukemia (AML), proteomic samples from patients are strongly heterogeneous, rendering their accurate relative quantification challenging. In the present study, we investigated the benefits of using a super-SILAC mix of AML derived cell lines as internal standard for quantitative shotgun studies. The Molm-13, NB4, MV4-11, THP-1, and OCI-AML3 cell lines were selected for their complementarity with regard to clinical, cytogenetic and molecular risk factors used for prognostication of AML patients. The resulting internal standard presents a high coverage of the AML proteome compared to single cell lines allied with high technical reproducibility, thus enabling its use for AML patient comparison. This was confirmed by comparing the protein regulation between the five cell lines and applying the internal standard to patient material.
Sample Processing Protocol
The cell lines MOLM-13, THP-1, OCL-AML3, MV4-11 and NB4 were grown in DMEM or RMPI medium supplemented with 10% fetal bovine serum (FBS). DMZO was removed by centrifugation at 900 rpm for 5 min, and lysis was performed by adding Ripa buffer (w/2x complete mini) containing benzonase (0.5U/µL) to the pellets, shortly followed by SDS (final conc. 0.5%). The same cell lines were also grown in DMEM or RPMI medium where the natural lysine and arginine were replaced by heavy isotope labeled amino acids, L-13C6-arginine (Arg6) and L-13C6,15N2-lysine (Lys8), supplemented with dialyzed 10% FBS. The metabolically labeled cell lines were cultured for at least five passages, to achieve complete incorporation of the isotope labeled amino acids, and cell lysis were performed as described for the unlabeled cell lines. An equal amount lysate from the five heavy cell lines were mixed to generate the super-SILAC mix, which was used as an internal standard. To enable comparison of the proteome of the five cell lines we combined equal amount of the super-SILAC mix (100µg) and each of the five unlabeled cell lines (100µg), resulting in five different samples of 200µg (also referred to as the 5-mix experiment). Prior to SDS-PAGE fractionation, the lysate mixtures were reduced with 20 mM DTT in 1 x LDS sample buffer at 65ºC for 20 min, and alkylated with 100 mM IAA at RT in the dark for 20 min. To adjust the pH we added 1µL 1.5M Tris (pH 8.8). The combined samples were separated by SDS-PAGE and 18 fractions was obtained from each lane, which were in-gel digested with trypsin porcine. Approximately 0.5µg of each of the 18 fractions per sample was analyzed using an Dionex Ultimate 3000 rapid separation (RS) LC nano system (Thermo Scientific) coupled online to an Orbitrap Velos Pro mass spectrometer (Thermo Scientific, Bremen, Germany). The LC-method consisted of 90 min, and the peptides were eluted at flow rate 280 nL/min. The MaxQuant software was used to analyze the raw files against the reviewed homo sapiens UniProt database.The rawfiles from the 5-mix experiment is named MV4_C12_C13_SILAC1, MV4_C12_C13_SILAC2, MV4_C12_C13_SILAC3, etc). The MaxQuant results from this dataset are found in: Results_MQ_5_cell-line-mix. Another similar experiment was performed where we mixed one heavy labeled cell line with the four remaining cell lines (light) (ex. Heavy MV4-11 + light mix of Molm13, THP-1, OCI-AML3, NB4), resulting in five samples referred to as the 4-mix experiment. These samples were processed in the same was the 5-mix experiment, but were fractionated into 16 fractions. The rawfiles are named FSOM_AML_2_1 - FSOM_AML_2_18 and FSOM_AML_3_1 - FSOM_AML_3_18 etc. The last number indicates the fraction number, while the FSMO_AML_2 is the OCI-AML3 cell line, FSMO_AML_3 is the Molm-13 cell line, FSMO_AML_4 is the MV4-11 cell line, FSMO_AML_5 is the NB4 cell line and FSMO_AML_6 is the THP-1 cell line. The MaxQuant results from this dataset are found in: Results_MQ_4_cell-line-mix. We also analyzed the super-SILAC mix with five heavy cell lines separately, also fractionated into 16 fractions. The raw files from this experiment are called FSMO_AML_1_1 - FSMO_AML_1_16. The MaxQuant results from this dataset are found in: Results_MQ_Heavy_IS_only. Technical replicates of the Molm-13 cell line analyzed in the 5-mix exp was analyzed by the Orbitrap Elite, but using the same MS method and LC gradient. The raw files from this MS analysis are called ex Molm-13_16a or Molm-13_16b, meaning run 2 and run 3 or fraction number 16. Run 1 is found in 5-mix dataset. The results can be found in: Results_MQ_Molm-13_replicates In all the result files are the experimental design provided, so that the raw files can be related to each MaxQuant search.
Data Processing Protocol
MaxQuant (version 18.104.22.168) was used to analyze the raw files. The Andromeda search engine was used to search the peak lists against the homo sapiens complement of the Uniprot/Swissprot reviewed database (downloaded Oct 30 2013, 20,278 entries). Carbamidomethylation (Cys) was set as fixed modification and oxidation (Met), acetylation (protein N-terminal) and phosphorylation (Ser, Tyr, Thr) as variable modifications. Only oxidation was taken into account for protein quantification. Quantification of SILAC pairs was performed using minimum ratio count of 2. Multiplicity was 2 (Arg6, Lys8) and the “re-quantify” and “match between runs” options were enabled. All other settings were set to default, briefly: a false discovery rate < 1% was required for peptides and proteins based on the target/decoy approach, max missed cleavages was set to two, minimum peptide sequence was seven amino acids, and enzyme specificity was trypsin, allowing N-terminal cleavage to proline. Unique and razor peptides were used for protein quantification. For qualitative comparison and visualization of the data the Perseus software (version 22.214.171.124, available in the MaxQuant environment) was used.
Aasebø E, Vaudel M, Mjaavatten O, Gausdal G, Van der Burgh A, Gjertsen BT, Døskeland SO, Bruserud O, Berven FS, Selheim F. Performance of Super-SILAC based quantitative proteomics for comparison of different acute myeloid leukemia (AML) cell lines. Proteomics. 2014 Jul 10 PubMed: 25044641