Here we describe a protocol that couples two proteomic techniques, namely 2-dimensional Electrophoresis (2DE) and Mass Spectrometry (MS), to identify differentially expressed/post-translational modified proteins among two or more groups of primary samples. This approach, together with functional experiments, allows the identification and characterization of prognostic markers/therapeutic targets.
The identification of molecules involved in tumor initiation and progression is fundamental for understanding disease’s biology and, as a consequence, for the clinical management of patients. In the present work we will describe an optimized proteomic approach for the identification of molecules involved in the progression of Chronic Lymphocytic Leukemia (CLL). In detail, leukemic cell lysates are resolved by 2-dimensional Electrophoresis (2DE) and visualized as “spots” on the 2DE gels. Comparative analysis of proteomic maps allows the identification of differentially expressed proteins (in terms of abundance and post-translational modifications) that are picked, isolated and identified by Mass Spectrometry (MS). The biological function of the identified candidates can be tested by different assays (i.e. migration, adhesion and F-actin polymerization), that we have optimized for primary leukemic cells.
Chronic Lymphocytic Leukemia (CLL), the most common adult leukemia in the Western countries, derives from the accumulation of monoclonal neoplastic CD5+ B lymphocytes and is clinically very heterogeneous. It may be present as a pre-leukemic form, defined as monoclonal B-cell lymphocytosis (MBL) 1,2,3. In other cases the disease can appear either with an indolent clinical course, that can remain stable for years before needing treatment, or as a progressive chemorefractory disease with dismal prognosis despite therapy. Finally, it may progress into a frequently lethal high grade lymphoma (Richter’s Syndrome-RS) 2,3,4. Patients are usually classified in two main subsets: good and bad prognosis, based on a set of prognostic factors that provides complementary information on predictors of disease outcome and survival. Clinical heterogeneity likely reflects biological heterogeneity. A number of genetic, microenvironmental and cellular factors have been shown to concur to disease pathogenesis though no unifying mechanisms have been found 5. In detail, several studies have demonstrated that differences in the clinical course of the disease can be partially explained by the presence (or the absence) of some biological markers that have a prognostic value 6,7,8. These data demonstrated that a better understanding of the disease biology could provide additional hints for the clinical management of the pathology, by the identification of both prognostic markers and therapeutic targets.
The goal of the present paper is to show how a combination of different proteomic techniques can be used for the identification of proteins involved in CLL onset and progression. Our approach demonstrates that it is possible to join together basic proteomic and clinical data5.
In the present workflow, primary leukemic cells from selected patients (good vs bad prognosis) are isolated and cell lysates are then used to obtain proteomic maps. 2DE allows the characterization of the proteomic profile of a cell population, including post-translational modifications, thus giving indirect information on the biological activity of each protein. Whole cell lysates are resolved by a first dimension run based on the isoelectric point, followed by a second dimension run on a polyacrylamide gel that resolves proteins based on their molecular weight. Comparative analysis of proteomic maps allows the identification of differentially expressed proteins (both in terms of abundance and post-translational modifications) as spots on the gel that can be cut and analyzed by Mass Spectrometry. The role of each candidate can be then exploited by different assays.
This approach, restricted to CLL in the current manuscript, can be easily expanded to other diseases/samples, thus providing information about the proteomic/biological differences between two groups (i.e. pathologic vs normal, stimulated vs unstimulated, wild-type vs knockdown).
NOTE: All tissue samples were obtained with approval of the institutional review board of San Raffaele Hospital (Milan, Italy).
1. Human Tissue Samples and Cell Purification (Figure 1)
NOTE: Leukemic lymphocytes were obtained from the Peripheral Blood (PB) of CLL patients, diagnosed according to Mulligan et al. 9
1.1) Leukemic Cell Separation from PB
1.2) Purity Evaluation
NOTE: Purity of all preparations needs to be always above 99%.
1.3) Samples Storage
2. Two-dimensional Electrophoresis (2-DE) (Figure 2)
2.1) Isoelectrofocusing (IEF)
2.2) SDS-PAGE
2.3) Silver Stain for Preparative Gel
FIXER | 50% Methanol 12% Acetic Acid 0,05% Formalin |
2 hr (or overnight)* |
WASH BUFFER | 35% Ehanol | 20 min (repeat the step three times) |
SENSITIZING | 0,02% Sodium Thiosulfate | 2 min |
WASH | H2O | 5 min (repeat the step three times) |
SILVER NITRATE | 0,2% Silver Nitrate 0,076% Formalin | 20 min |
WASH | H2O | 1 min (repeat the step twice) |
DEVELOPER | 6% Sodium Carbonate 0,0004% Sodium Thiosulfate 0,05% formalin |
Until the staining is sufficient |
STOP | 50% Methanol | 30 min |
* 2 hr is the minimum time required for protein fixation |
Table 1.
NOTE: Protein spots can be visualized by staining gels with MS compatible silver stain11. All the solutions needed for silver staining are listed in Table 1.
2.4) Gel Acquisition and Analysis
3. Protein Identification by MALDI-TOF MS analysis (Figure 3)
3.1) Protein Digestion
3.2) Mass Spectrometry Analysis (Dried droplet technique)
4. Cytoskeletal Activity Assays (Figure 4)
NOTE: Resuspend cells purified in Step 1 in complete RPMI (106 cells/200 μl).
4.1) Migration Assay. This test allows quantification of the migratory capacity of the analyzed cells (lymphocytes ). Use a transwell chamber of 6.5 mm diameter and 5.0 μm pore size and perform the assay in triplicate. Optimize pore size and migration time (Step 4.1.3) in the case of different cell types.
4.2) Adhesion assay. This assay allows measuring the adhesion capacity of the cells. Perform the assay in triplicate.
4.3) Polymerization assay. This is a colorimetric assay that quantifies F-actin polymerization. Perform the assay in triplicate.
We isolated primary leukemic B cells form PB of CLL patients and we analyzed proteomic maps. Samples (n = 104) were grouped in two main subsets based on the clinical features of each patient (bad prognosis vs good prognosis) and comparative analysis of 2DE gels was performed.
The analysis allowed identification of spots differentially expressed between the two subsets (in term of presence/absence or shift on the gel, implying changes in post-translational modifications; n ≈ 16). We excised selected spots from the 2DE gels and after being reduced and alkylated, proteins were digested with trypsin and peptides in the resulting supernatant were spotted onto a MALDI plate. Spectra were acquired in a MALDi-ToF Voyager DE. The resulting peak list was submitted to MASCOT and ProFound. Masses within a certain mass tolerance are assigned to peptide sequences in the database, and then assembled into a protein, which is considered identified if it passes a certain probability score (Figure 3). By this analysis we identified proteins mainly involved in cytoskeletal activity and metabolic processes (unpublished data).
Among them, we focused on hematopoietic-lineage-cell-specific-protein-1 (HS1) whose differential phosphorylation strongly associated with the clinical course of the disease (Figure 4). In particular, we have found that patients carrying a single spot (n = 44, hyper-phosphorylated HS1) experience a bad clinical outcome, while patients with 2 spots (n = 60, hypo-phosphorylated HS1) have good prognosis13 (Figure 4). The presence of the HS1 protein in the spots was then validated by immunoblotting the 2DE gel with a monoclonal antibody against HS1 13.
Since it is known that HS1 is involved in cytoskeletal remodeling14, we performed in vitro assays to test if the HS1 phosphorylation status could differentially affect cytoskeletal activity in the two CLL subsets (good vs bad prognosis). We found that CLL cells carrying HS1 as one spot have an impaired cytoskeletal activity in terms of migration, adhesion and actin polymerization, compared to hypo-phosphorylated-HS1 samples, thus explaining a different clinical behavior15 (Figure 5).
Figure 1: Workflow of leukemic cell purification from PB. Blood from CLL patients is transferred into a 15mL tube (1), Human B-cell enrichment cocktail is added to the sample (2) and incubated for 20 minutes. Blood is then diluted with PBS in a proportion of 1:1 and laid on the top of the density gradient (3). Subsequent sample centrifugation (4) allows the formation of multiple layers: a) plasma, b) B lymphocytes, c) Ficoll, d) red cells and unwanted cells. Purified B lymphocytes (b layer) are then collected (5), washed and purity of cell preparation is analyzed by flow cytometry (6). Please click here to view a larger version of this figure.
Figure 2: Workflow of 2DE. The pellet is solubilized in 2DE buffer (1) and loaded on the IPG strip for the first dimension run (2-isoelectrofocusing). After equilibration, the strip is loaded on top of a gradient polyacrylamide gel for the second dimension run (3-SDS-PAGE). The gel is then stained to visualize spots/proteins (4). After high-resolution image acquisition, proteomic maps are analyzed (5). Please click here to view a larger version of this figure.
Figure 3: Workflow of MS analysis. (1) Spots of interest are excised from the 2D-gel with a scalpel and transferred into a clean tube. (2) After being reduced and alkylated, proteins are digested with trypsin, peptides in the resulting supernatant are spotted onto a MALDI plate (3). Spectra are acquired in a MALDi-ToF Voyager DE. (4) The resulting peak list is submitted to MASCOT and ProFound search engines and searched against a comprehensive non-redundant protein database. (5) Masses within a certain mass tolerance are assigned to peptide sequences in the database, then assembled into a protein. Please click here to view a larger version of this figure.
Figure 4: HS1 phosphorylation status correlates with prognosis of CLL patients. (1) The circle identifies two close spots with the same molecular mass (Mr) of 79 kDa and different isoelectric point (pI) of 4.83 and 4.86 respectively, which was identified by MS as HS1 protein (2). (2) Two representative gels of one bad prognosis (red square) and one good prognosis CLL patients (green square). (3) Kaplan-Meier curves show cumulative survival of CLL patients grouped according to HS1 phosphorylation pattern (1 spot, n = 44, vs 2 spots, n = 60). Patients with 2 spot (green dots) have a significantly longer survival (median survival not reached) than those with only one (red dots). Please click here to view a larger version of this figure.
Figure 5: HS1 phosphorylation status influences cytoskeletal functionality in CLL primary samples. (1) Migration on transwell of primary samples in the presence or absence of SDF-1. In the graph are displayed the means SEM of the number of cells acquired in 1 minute at the flow cytometer (n = 7 of 1 spot HS1 vs n = 12 of 2 spot HS1). (2) Spontaneous adhesion was measured after cell labeling and 1 hr incubation in 96-well plates. Displayed are the means SEM for primary samples (n = 7 of 1 spot HS1 vs n = 12 of 2 spot HS1). (3) F-actin polymerization capability of primary samples. Displayed are the means SEM of the relative F-actin content of CLL cells after stimulation with SDF-1 and staining with FITC-labeled phalloidin (n = 5 of 1 spot HS1 vs n = 5 of 2 spot HS1). MFI: Mean Fluorescence Intensity. The histogram represent as example of sample acquisition. Please click here to view a larger version of this figure.
The aim of this manuscript is to describe an optimized protocol for the identification of molecules involved in the pathogenesis of CLL, even if this approach can be extended to other diseases and/or other sample types16-18. By comparing the proteomic profiles of 2 subsets of CLL patients, good vs bad prognosis19, we demonstrated that they differ in the phosphorylation status of HS1 and that its activation affects the migration and adhesion capacity of leukemic cells.
With respect to other methods, the main advantage of the proteomic techniques presented in the manuscript, 2DE gel and MS, is that they provide a complete fingerprint of the cell proteome and most importantly they give information on the post-translational modifications of proteins. Proteins that are differentially phosphorylated or glycosylated change their position in the gel, and likely change their activity in the cells.
Moreover, we described protocols for the evaluation of cells’ migration and adhesion that were used to test cytoskeletal functionality; these protocols are still poorly described for cells growing in suspension, since they are commonly applied to adherent cells (e.g. wound-healing). The major limitation of the 2DE and MS technique is the amount of cells/proteins (≈25 x 106 cells per gel), while for functional assays it is the viability of the cells. The real challenge is to work on primary cells, but for many diseases it is not possible to reach a sufficient number of cells or live cells to perform those experiments; in this case, if available, it is possible to use cell lines.
The most critical step for the MS protocol is to work in clean (keratin free) conditions in order to avoid contamination and to obtain clear protein identification. Cytoskeletal assays are usually difficult to reproduce (especially on primary cells), thus it is suggested to perform the experiment in triplicate.
As we have demonstrated, the combination of the approaches presented in this manuscript helps in the identification of new pathways involved in different diseases, thus providing for the discovery of new therapeutic targets, as we have demonstrated for HS1 protein19.
The authors have nothing to disclose.
We thank the Lymphoid Malignancies Unit, Elisa ten Hacken, Lydia Scarfò, Massimo Alessio, Antonio Conti, Angela Bachi, and Angela Cattaneo.
This project was supported by: Associazione Italiana per la Ricerca sul Cancro AIRC (Investigator Grant and Special Program Molecular Clinical Oncology – 5 per mille #9965)
C.S. and B.A designed the study, performed the experiments, analyzed the data and wrote the manuscript. M.T.S.B., U.R., F.B. P.R. assisted in writing the manuscript. P.G. and F.C.C. designed the study provided patients’ samples and clinical data.
Name of Material/ Equipment | Company | Catalog Number | Comments/Description |
Acetonitril | MERCK | 61830025001730 | |
Ammonium Bicarbonate | SIGMA | A6141-500G | |
1,4-Dithioerythritol | SIGMA | D9680-5G | |
Iodoacetamide | SIGMA | I6125-25G | |
Calcium chloride dihydrate | SIGMA | 223506-25G | |
Trypsin, Sequencing Grade | ROCHE | 11418475001 | |
α-cyano-4-hydroxycinnamic acid | SIGMA | C2020-10G | |
Trifluoroacetic acid | SIGMA | T6580 | |
ZipTipµ-C18 | MILLIPORE | ZTC18M096 | |
RosetteSep Human B Cell Enrichment Cocktail | STEMCELL | 15064 | |
PBS | EUROCLONE | ECB4004L | |
FBS | DOMINIQUE DUTSCHER | S1810 | |
Lymphoprep | SENTINEL DIAGNOSTICS | 1114547 | |
Trypan Blue | SIGMA | T8154 | |
CD19 | BECKMAN COULTER | 082A07770 | ECD |
CD5 | BECKMAN COULTER | 082A07754 | PC5 |
CD3 | BECKMAN COULTER | 082A07746 | FITC |
CD14 | BD | 345785 | PE |
CD16 | BECKMAN COULTER | 082A07767 | PC5 |
CD56 | BECKMAN COULTER | 082A07789 | PC5 |
Ettan IPGphor 3 Isoelectric Focusing Unit | GE-Healthcare | 11-0033-64 | |
Urea | SIGMA | U6504 | |
Tris-Hcl Buffer | BIO-RAD | 161-0799 | |
CHAPS | SIGMA | C2020-10G | |
DTT | SIGMA | D9163-5G | |
IPGstrips | GE-Healthcare | 17-1233-01 | Linear pH 4-7 18cm |
IPGbuffer | GE-Healthcare | 17-6000-86 | pH 4-7 |
Glycerol | SIGMA | G6279 | |
PROTEAN II XL Cell | BIO-RAD | 165-3188 | |
SDS | INVITROGEN | NP0001 | |
Acrilamide | BIO-RAD | 161-0156 | |
Agarosio | INVITROGEN | 16500 | |
Methanol | SIGMA | 32213 | |
Acetic Acid | VWR | 631000 | |
Formalin | BIO-OPTICAL | 05-K01009 | |
Ethanol | VWR | 9832500 | |
Sodium Thiosulfate | FLUKA | 72049 | |
Silver Nitrate | FLUKA | 85228 | |
Molecular Dynamics Personal SI Laser Densitometer | Amersham Biosciences | ||
ImageMaster 2D Platinum 5.0 | Amersham Biosciences | ||
Sodium Carbonate | MERCK | A0250292 | |
MALDI-TOF Voyager-DE STR | Applied Biosystems | ||
Data Explorer software (version 3.2) | Applied Biosystems | ||
transwell chambre 6.6 mm diameter | CORNING | 3421 | |
RPMI | EUROCLONE | ECB2000L | WITH L-GLUTAMINE |
Gentamicin | SIGMA | G1397 | |
SDF-1 | PREPROTECH | 300-28A | |
Flat-bottom 96-well plate | BECTON DICKINSON | 353072 | |
BSA | SANTA CRUZ | 9048-46-8 | |
ICAM-1 | R&D Systems | 720-IC | |
CMFDA | Life Thechnology | C7025 | Green |
IgM | SOUTHERNBIOTECH | 2020-02 | |
Paraformaldehyde | SIGMA | P6148 | |
Saponine | SIGMA | S-7900 | |
Phalloidin | Life Thechnology | A12379 | Alexa fluor 488 |