ALIX [ALG-2 (apoptosis-linked gene 2)-interacting protein X] is one of the most well-known molecular biomarkers of extracellular vesicles. Extracellular vesicles are very small vesicles released by most cells and carry in their cargo components from the donor cells, thus being potent vehicles of intercellular (horizontal) communication, influencing various physiological and pathological functions of both recipient and donor cells. The increasing interest in extracellular vesicles highlights the key importance of a reliable analysis of this protein. However, several recent studies in the extracellular vesicles field have shown discrepancies in terms of the expression pattern of apoptosis-linked gene 2-interacting protein X upon Western blot analysis, differing from the theoretical expression pattern of apoptosis-linked gene 2-interacting protein X and its predicted molecular mass. Therefore, to address and clarify this point, we analyzed total protein cell lysates from a chronic myeloid leukemia cell line (K562) for the expression of apoptosis-linked gene 2-interacting protein X by Western blot and mass spectrometry analyses, using protein samples stored at different conditions regarding freezing temperature and storage time. We found that, when stored at -20 ºC, a C-terminal specific proteolytic cleavage of apoptosis-linked gene 2-interacting protein X may occur, which depends on the length of storage time. We conclude that analysis of apoptosis-linked gene 2-interacting protein X protein expression should be only carried out when using a wide range of protease inhibitors during isolation of protein cell extract, while preferentially using fresh protein cell extracts or samples that were snap frozen in liquid nitrogen and stored at -80 ºC. The current study highlights the importance of proper handling and storage of protein cell lysates for downstream applications in pre-clinical or clinical studies.
Extracellular vesicles (EVs) are very small particles (30-1000 nm) enclosed by a phospholipid bilayer, which do not replicate [1]. EVs are released by cells from a spectrum of organisms ranging from prokaryotes to higher eukaryotes and plants [2], and carry in their cargo contents from the donor cells, such as nucleic acids, proteins, lipids, and carbohydrates [3]. These vesicles may transfer their cargo to recipient cells, thus being important mediators of intercellular (horizontal) communication [4]. Importantly, EVs not only modulate physiological processes (e.g. tissue repair or embryonic development) but also pathological processes (e.g. cancer or autoimmune diseases) and may be found in biological fluids [5], thus being a potential source of biomarkers for various human diseases [6,7].
The increasing interest in EVs provides great relevance to the proper analysis of ALIX, [ALG-2 (apoptosis-linked gene 2)-interacting protein X], since it has been defined as one of the well-established molecular markers of EVs [8,9]. ALIX is an adaptor protein first described for its capacity to bind to the calcium-binding protein, ALG-2 [10,11]. It is a ubiquitously expressed cytosolic protein that was initially identified due to its association with pro-apoptotic signaling components [12]. More recently, ALIX was found to regulate other cellular mechanisms such as endocytic membrane trafficking [13], being essential for clathrin-independent endocytosis and signaling [11]. Much of the progress in characterizing the function of ALIX has been focused on its connection to endocytic membrane trafficking. This protein interacts with several ESCRT (endosomal sorting complexes required for transport) proteins in order to participate in the budding and abscission processes that can ultimately lead to the formation of exosomes [12,14]. ALIX also increased protein content and protective functions of ex-somes shed by induced pluripotent stem cells [15]. ALIX is also associated with programmed cell death [16,17], virus egress [18], cytokinesis [19], regulation of integrin- mediated cell adhesions and extracellular matrix assembly [20] and repair of the plasma membrane [21,22].
The accession number of ALIX is Q8WUM4 and the UniProt ID is PDC6I_HUMAN. According to UniProt, the predicted molecular mass of the ALIX protein is 96.023 kDa, and upon Western blot analysis it is detected as a single band [4]. However, although some studies present only one band as expected [23], several reports identified more than a single ALIX band upon Western Blot analysis. Indeed, several ALIX-associated bands have been observed upon immunoblot analysis when using protein lysates derived from either cultured cell lines or EVs [24-29]. In some of these studies [24,25,27], ALIX appeared as two (or more) distinct bands, including a major one with the predicted molecular mass of approximately 93 kDa as well as additional bands with a smaller size. In other studies, although ALIX appeared as a single band with the expected molecular mass [26,28,29], the supplementary information (in which the respective uncropped blots are shown) revealed the presence of other band(s) which were not identified due to their surprising small molecular mass.
It is well known that a proper freezing storage temperature is one of the most important factors in order to maintain protein’s integrity over an extended period of time as well as the use of a cocktail of protease inhibitors [30-33]. Therefore, given the increased interest in ALIX in the last years [28,34] and our own need to analyze the expression of this protein, the aim of the present study was to further characterize the molecular basis for the appearance of distinct bands of ALIX.
The chronic myeloid leukemia cell line K562 was from European Collection of Authenticated Cell Cultures (ECACC). The cell line was genotyped and routinely monitored for mycoplasma contamination by PCR (VenorGeM® Advance Mycoplasma Detection Kit, Minerva). Cells were routinely grown in RPMI-1640 medium (with Ultraglutamine I and 25 mM HEPES pH 7.3; Lonza), supplemented with 10% fetal bovine serum (FBS, PAA) at 37 °C in a humidified atmosphere of 5% CO2. Cell numbers and viability were determined using the trypan blue exclusion assay. All experiments were carried out with exponentially growing cells having over 90% viability, as previously described [35,36].
Cells were harvested and washed with PBS and cell pellet was kept for different periods of 1, 2, 3 or 6 weeks at - 20 ºC or -80 ºC. Following these variable storage times cell pellets were suspended in Winman's buffer (1% NP-40, 0.1 M Tris-HCl pH 8.0, 0.15 M NaCl, and 5 mM EDTA) with EDTA-free protease inhibitor cocktail (Catalogue number 11873580001, Roche), without phosphatase inhibitor (for samples kept at -20 ºC and at -80 ºC ) or with phosphatase inhibitor (Sodium Orthovanadate, Catalogue number S6508, Sigma; for samples kept at -20 ºC ). Protein was quantified using a modified Lowry assay (Bio-Rad). Western blot analysis was performed as previously described [35,36]. Briefly, 20 μg of protein was resolved on a 10% Bis-Tris SDS-PAGE and transferred onto a nitrocellulose membrane (GE Healthcare). Membranes were then incubated with the following primary antibodies: ALIX sc-49268 from Santa Cruz Biotechnology (1 : 100); ALIX #2171 from Cell Signaling Technologies (1 : 100); anti-β-Actin from Santa Cruz Biotechnology (1 : 2000). The following secondary antibody was used: anti-goat IgG-HRP, diluted 1 : 2000, from Santa Cruz Biotechnology. Signal was detected using the ECL Western blot Detection Reagents (GE Healthcare), the Amersham Hyperfilm ECL (GE Healthcare), and the Kodak GBX developer and fixer (Sigma). Protein bands were evaluated in three independent experiments. The intensity of the bands obtained in each experiment was further analyzed using the software Quantity One-1D Analysis (Bio-Rad, USA). Two major bands of the ALIX protein were detected: a predicted one with a molecular mass of 93 kDa as well as a 75 kD species.
Protein identification was performed by mass spectrometry using previously published procedures [37,38]. Briefly, gel plugs were reduced with 25 mM DTT in 50 mM ammonium bicarbonate (ABC) for 20 min at 56 ºC and alkylated with 55 mM iodoacetamide in 50 mM ABC in the dark for 20 min at room temperature. In gel tryptic digestion was performed (20 ng trypsin per spot) at 37 ºC for 3h in the presence of surfactant at 0.01% in 50 mM ABC (Promega, Madison, WI, USA). Peptide extraction was performed using a 2.5% trifuoroacetate (TFA) solution. Protein digests were desalted and concentrated by C18 reverse phase chromatography (ZipTips, Millipore, Bedford, MA, USA) following the manufacturer’s instructions. Samples were crystallized onto a stainless steel 384-well MALDI plate using the dried droplet method. For the matrix preparation, a solution of 7-8 mg/mL α-cyano-4-hydroxycinnamic acid in 50% acteonitrile/0.1% TFA was used. Samples were analyzed using a 4800 Plus MALDI TOF/TOF Analyzer (AB SCIEX, Framingham, MA, USA) as already described [39]. Peptide mass fingerprint (PMF) data were collected in positive MS reflector mode within the ion range at m/z 700-4000 and was calibrated internally using trypsin autolysis peaks. MS spectra were analyzed using the Mascot search engine software (Matrix Science, London, UK). The UniProt protein sequence database was considered for this analysis using the human reference proteome. Cysteine carbamidomethylation and methionine oxidation were considered as constant and variable modifications, respectively. Up to two missed trypsin cleavage sites were considered.
The statistical analysis of the scanning densitometry analysis of the Western blot was carried out using the Student's t-test. Data was analyzed with the paired test. Differences were considered statistically significant when P ≤ 0.05, by comparing samples kept at -20 ºC without phosphatase inhibitor with samples kept at -20 ºC with phosphatase inhibitor or kept at -80 ºC without phosphatase inhibitor.
Total protein cell lysates from a chronic myeloid leukemia cell line (K562, ECACC) were analyzed for ALIX protein expression by Western blot, using two of the most commonly employed antibodies (ALIX sc-49268 from Santa Cruz Biotechnology and ALIX #2171 from Cell Signaling Technologies). The samples were treated with an EDTA-free protease inhibitor cocktail to avoid protein degradation during storage at -20 ºC (with or without phosphatase inhibitors) or at -80 ºC for different periods of time (1, 2, 3 and 6 weeks) [35,36].
After performing Western blot analysis of the stored protein lysates, we detected two major bands (using either antibody) as shown in Fig. 1: i) One 93 kDa band with a molecular mass compatible with the predicted size ALIX (93 kDa-ALIX) as well as ii) Another band with an approximate molecular mass of 75 kDa (75 kDa-ALIX).
Figure 1.Western blot analysis of ALIX protein levels in K562 cells following various storage conditions. Fresh total cell protein lysates were immediately stored at: -20 ºC , -20 ºC with a phosphatase inhibitor cocktail (+ PI) or at -80 ºC . Samples were kept frozen for 1, 2, 3 and 6 weeks. (A) Representative Western blot images; (B) Scanning densitometry analysis of the Western blot ALIX bands, represented as mean ± SEM from three in dependent experiments. β-Actin was used as a loading control. * P ≤ 0.05 when comparing protein levels in samples stored at -20 ºC (-PI) vs. other storage conditions (- 80 ºC or -20 ºC + PI).
In order to confirm that the antibodies were specifically detecting ALIX-based proteins, we performed mass spectrometry analysis (Supplementary Fig. S1 and 2). We detected 20 peptides related to ALIX using the 93 kDa protein band including an N-terminal peptide consisting of amino acids 24-41 as well as a C-terminal peptide encompassing amino acids 716-745. In the 75 kDa band, we still detected the N-terminal peptide, whereas the C- terminal 716-745 peptide was absent (Fig. 2). This led us to conclude that the 75 kDa protein band corresponds to a truncated form of ALIX in which the C-terminal peptide was missing presumably due to proteolysis.
Figure 2.Mass spectrometry analysis of the two ALIX bands. The sequences of both the 93 kDa (A) and 75 kDa (B) bands were analyzed and the peptides related to ALIX were detected. The C-terminal 716-745 peptide, which was present in the sequence of the intact 93 kDa protein (A, marked with a box) was absent from the sequence of the 75 kDa protein (B).
It is important to note that for both the human and mouse ALIX proteins, there are published reports about the cathepsin-mediated proteolytic cleavage that may occur at their C-termini (the residue numbers are based on Uniprot sequences) as follows: a) Regarding the human ALIX protein, there is a report based on proteomics screen for cathepsins which reveals a putative cleavage site by cathepsin L between amino acid residues A732 and G733; this occurred within the sequence QSSPA^GGHAPT (with ^ representing the putative cleavage site) where the human ALIX appears under protein IPI ID IPI00246058.7 [40]. Whereas, for the mouse ALIX protein, there is a previous paper also based on a proteomics screen that cathepsin E cleaved the murine ALIX protein between residues L705 and L706, within the sequence ERDEL^LKDLQ [41]. Thus, proteomics analysis as well as protease cleavage analysis reveals that the C-terminal ends of both the human and murine ALIX proteins are subject for proteolytic cleavage mediated by cathepsins. Thus, these C-terminal truncations of both the human and mouse ALIX proteins, are in accord with the difference observed between the native 93 kDa protein and the truncated 75 kDa protein in our present study.
When analyzing the protein cell extracts stored at -20 ºC over time (1, 2, 3 and 6 weeks), both in the presence or absence of phosphatase inhibitors, we noticed that the presumed C-terminal proteolytic degradation of ALIX depended on the duration of the storage of samples at -20 ºC (Fig. 1). In fact, over a prolonged storage time and at -20 ºC , the levels of the primary 93 kDa ALIX protein decreased, while the levels of the truncated form (75 kDa-ALIX) increased. In addition, under the same conditions (-20 ºC storage), the presence of a phosphatase inhibitor had no effect on the level of the degradation of this protein (Fig. 1).
In contrast, when analyzing the samples stored at -80 ºC for the same period of time (1, 2, 3 and 6 weeks), no truncation of ALIX was observed. In fact, the level of the 93 kDa ALIX protein remained unaltered during the 6 weeks period and no increase in the truncated form of ALIX (75 kDa-ALIX) was observed (Fig. 1).
Long-term storage of protein lysates at -80 ºC is recommended [30], but -20 ºC is commonly used for shorter storage periods. Therefore, the degradation observed in ALIX after a short storage period (1 or 2 weeks) at -20 ºC was rather unexpected. Of note, Western blot analysis was carried out at different times by two of the authors (using the same experimental conditions), which eliminates errors due to incorrect sample processing and analysis.
The obtained results demonstrate that analysis of ALIX protein upon immunoblot analysis should be carried out when using freshly isolated protein extracts or that have been properly stored at -80 ºC with the use of proper protease inhibitors, even for a short storage time. This would avoid misleading conclusions resulting from the analysis of decreased ALIX protein levels due to degradation.
This study emphasizes the importance of proper isolation, handling and storage of protein cell lysates for downstream applications such as Western blots, when quantifying ALIX protein levels.
This report is a result of project NORTE-01-0145-FEDER-000029, supported by Norte Portugal Regional Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF). This work was partially financed by FEDER-Fundo Europeu de Desenvolvimento Regional funds through the COMPETE 2020-Operacional Programme for Competitiveness and Internationalisation (POCI), Portugal 2020, and by Portuguese funds through FCT-Fundação para a Ciência e a Tecnologia/ Ministério da Ciência, Tecnologia e Inovação in the framework of the project "Institute for Research and Innovatio n in Health Sciences" (POCI-01-0145-FEDER-007274). The authors thank the Portuguese Foundation for Science and Technology (FCT) for the PhD grants of VLR and DS (SFRH/BD/87646/2012 and SFRH/BD/98054/2013 respectively) and for the post-doctoral fellow grant of RTL and CX (SFRH/BPD/68787/2010 and SFRH/BPD/122871/2016 respectively). The mass spectrometry technique was performed at the Proteomics i3S Scientific Platform. This work had support from the Portuguese Mass Spectrometry Network, integrated in the National Roadmap of Research Infrastructures of Strategic Relevance (ROTEIRO/0028/2013; LISBOA-01- 0145-FEDER-022125). This article benefited from a collaboration initiated under the framework of COST Action 17104 STRATAGEM, supported by COST (European Cooperation in Science and Technology).
The authors declare no competing interests.
Figure S1.Mass spectrometry analysis of the 93 kDa ALIX band. The analysis of the sequence of the 93 kDa protein is shown in detail (sequences and peaks of the peptides detected).
Figure S2.Mass spectrometry analysis of the 75 kDa ALIX band. The analysis of the sequence of 75 kDa protein is shown in detail (sequences and peaks of the peptides detected).