Proteomic analysis of human T cell-derived exosomes reveals differential RAS/MAPK signaling

Exosomes (Exo) are cell-derived vesicles that are present in many body fluids and cell cultures and measured between 30 and 100 nm in size. Unlike the larger microvesicles (Mv), Exo are either released from the cells when multivesicular bodies fuse with the plasma membrane or released directly from the plasma membrane. Evidence is accumulating that both Mv and Exo have specialized functions and play a key role in the pathogenesis of asthma, rheumatoid arthritis, and other inflammatory diseases [1]. Accordingly, there is a growing interest to better understand Exo biology for future therapeutic and diagnostic applications [1]. Previous work from our laboratories established the role of T cell-derived vesicles in regulating mast cells activation and function [2]. More recently, we have shown that vesicles shed from T cells induced mast cell degranulation and release of specific cytokines [2, 3]. The functional impact of Exo is conveyed by the molecular components (i.e. proteins, lipids, and nucleic acids) that they carry [4]. Although a large number of structural proteins is commonly found in these vesicles regardless of their cellular origin [5], additional knowledge about the unique cargo of T cell-derived Exo is needed. It is important to fill this gap in our knowledge due to the current role of Exo in translational medicine [5].

We took advantage of a proteomic approach to better characterize the unique set of proteins that are fond in T cell-derived Exo. Human T cells were collected and Exo were harvested before and after stimulation with anti-CD3/28 antibodies, as previously described [3]. Electron microscopy images confirmed that the size of the vesicle was less than 100 nm (Fig. 1A and Supporting Information Fig. 1A). Western blot analysis demonstrated that the Exo markers CD63 and CD81 were more abundant in Exo derived from activated, compared to resting, T cells (Fig. 1B). Interestingly, flow analysis revealed that despite lower amount of Exo collected from the same number of resting T cell (~2.5-fold reduction), the expression levels of Annexin V were comparable between the conditions (Fig. 1C and Supporting Information Fig. 1B and C). To further investigate whether the difference is merely a quantity matter or if there is also a variation in the specific protein content between the conditions, we analyzed equal amount of Exo from each group using mass spectrometry. While 1064 proteins were found at similar levels in the two conditions (Supp-porting Information Fig. 1D and Supp-porting Information Table 1), we discovered 61 proteins that were higher in Exo derived from activated T cells (Supporting Information Table 2), and only 28 proteins that were elevated in Exo collected from resting T cells (Supporting Information Table 3) (Fig. 1D). To elucidate the functions of these proteins we performed an enrichment analysis of biological processes [6]. While most of the proteins that were higher in resting T cells Exo were related to cytoskeleton organization, the proteins form activated T cells Exo (Fig. 1E) were associated with specific cellular functions such as Immune surface receptor signaling pathway and Metabolic processes. Moreover, String analysis [7] identified three specific clusters (Immune surface receptor signaling pathway; Translation; and RAS protein signaling transduction) (Fig. 1F) that correlated, at least in part, with the functional annotation (Fig. 1D).

Figure 1.

Characterization of T cell-derived exosomes. (A) Transmission electron microscopy of exosomes isolated by differential ultracentrifugation [2] from Jurkat T cells. Representative images of three independent experiments are shown. (B) Western blot analysis of Jurkat T-cell lysates and exosomes, probed for CD63 and CD81 (positive markers of exosomes) and BCL-XL (negative marker of exosomes); *p < 0.05; **p < 0.001; unpaired t-test; Data shown are representative of four independent experiments. (C) Exosomes collected from 80 × 10⁶ activated or resting Jurkat T cells were labeled with Annexin V-PE and analyzed by flow cytometry. A representative plot of two independent experiments is shown. (D) Heat map representing mass spectrometry analysis of proteins collected from human T cells; proteins that were significantly higher (red) or lower (blue) are shown; Welch t-test; p < 0.05; Data shown are pooled from three independent experiments. (E) −log₁₀ B&H (Benjamini and Hochberg’s) FDR (false discover rate) procedure of protein levels that were enriched in exosomes collected from activated T cells (D) based on functional enrichment; the numbers of specific proteins from each functional group is shown in brackets. (F) String association map of the protein clusters identified in (D).

Other have shown enriched MAPK in tumor-derived vesicles, but we were specifically intrigued by the fact that additional components of the RAS signaling pathway were enriched in Exo derived from activated T cells. Western blot analysis confirmed the presence of activated RAS in these Exo (Fig. 2A). The closely related GTPase RAP1, that shares many effector functions and signaling pathways with RAS (and is found in whole cell lysate of activated T cells), was excluded from the Exo (Fig. 2A). Moreover, additional signaling components associated with RAS, such as ZAP70, RASGRP1, and AKT were found in the same vesicular structure, while the exchange factor of RAP1, C3G was absent (Fig. 2B). To investigate whether the components of the RAS signaling pathway were functional, we cultured LAD2 mast cells with Exo derived from either resting or activated T cells (Fig. 2C). As shown, only the Exo purified from activated T cells demonstrated an increase in mast cells ERK phosphorylation (Fig. 2C). In support of our data, additional RAS-associated signaling proteins were detected in the mass-spectrometry analysis (i.e. RAF1, PI3K, PKC, RALGDS, MEK, and RAC) (Fig. 2D). Remarkably, our approaches failed to detect RAP1-specific signaling components in the Exo.

Figure 2.

RAS/MAPK signaling proteins are enriched in exosomes. (A, B) Western blot analysis of exosomes collected from 80 × 10⁶ activated or resting Jurkat T cells; *p < 0.05; Unpaired t-test. Data are representative of four independent experiments (C) LAD2 mast cells were cultured with exosomes collected from activated or resting T cells, ERK phosphorylation levels were analyzed by immunoblotting at different time points; *p < 0.05; Unpaired t-test; Data shown are representative of three independent experiments. (D) Scheme of novel (blue) and known (lavender) RAS and RAP1-assocated proteins identified in exosomes.

The immunological synapse serves as a port for budding vesicles enriched with sorted TCR [8]. An important question is whether these vesicles contain, in addition to the TCR, other downstream signaling components. Our data confirmed that this is indeed the case. But interestingly, and despite the fact that both RAS and RAP1 are activated downstream of TCR ligation, only the former was included in the vesicles. The fact that additional RAS-specific downstream effectors were also included in the vesicle further supports the concept that the cargo of the vesicles is very selective, and not a result of casual sampling of cytosolic content or receptor-associated signaling pathways. The main limitation of our study is the methods of Exo isolation, which is based on differential centrifugation, that enriches, but not complete separates different vesicular fractions. Never the less, ongoing studies in our laboratory will better distinguish the functions of the vesicular cargo, prior to designing Exo and Mv-related therapeutic approaches.

Abbreviations:

Exo

exosomes

Mv

microvesicles