Exosomes or Microvesicles, a Secreted Subcellular Organelle Contributing to Inflammation and Diabetes

Type 1 and type 2 diabetes (T1D and T2D) are generally recognized as distinct disease entities based on the mechanism of induction of disease: T1D results from low or no insulin production due to the gradual loss of cells that produce insulin, whereas T2D is generally thought to occur as a result of the body’s development of resistance to insulin. Although the events initiating T1D and T2D might be very different, the progression to diabetes and complications are similar and involve interactions between the immune system and the metabolic system. Notably, chronic inflammation is a shared manifestation of the two types of diabetes (1,2). Recently, secreted microvesicles, particularly exosomes, were suggested to be intermediates linking inflammation to diabetes (3). The cellular origins of the vesicles and the genetic and environmental factors that cause an abnormal production of the vesicles are likely to be central in linking the inflammation to tissue-specific damage. For example, islet mesenchymal stem cells from autoimmune-prone animals were shown to release highly proinflammatory exosomes that contributed to T1D (4), and the low-grade inflammation in obesity may stimulate release of proinflammatory microvesicles and exosomes to accelerate T2D (5,6).

“Extracellular vesicles” (EVs) is a collective term for different types of small-sized membrane vesicles or particles that are secreted by cells, including exosomes, microvesicles, apoptotic bodies, and virus-like particles. Figure 1 shows a generalization of their similarities and differences in biogenesis and immune-stimulatory function. There are many reasons that such organelles may be produced, including packaging of paracrine factors for intercellular communications, apoptosis resulting from breakdown of cells, and packaging of material for ingestion and disposal by scavenger cells (7). Viruses too have learned to take advantage of the vesicle biogenesis machinery to hide from immune surveillance to establish infections at immune-privileged sites (8,9). Therefore, EV preparations from culture media, or biological fluids such as serum , semen, urine, etc., represent a pool of mixed EVs, with each subspecies having a different cargo and function. The cargo of the EVs contains important complex molecules such as RNA, DNA, proteins, lipids, and carbohydrates. The ratio of each subspecies and their molecular content may vary depending on the physiological/pathological status of the parent cells and tissues that release the EVs. Thus, EVs may potentially be a powerful source for noninvasive diagnosis of diseases (10).

The study by Freeman et al. (11) in this issue of Diabetes is unique in that it uses an ex vivo approach to study the role and mechanism of action of EVs derived from blood plasma of normal patients, patients with prediabetes, and patients with diabetes in the development of disease (11). The availability of plasma EVs from cohorts of individuals at two time points, separated by 5 years, allowed monitoring of changes associated with disease progression, since it was possible to compare individuals without diabetes with those who progressed to overt diabetes. The ex vivo approach is useful because it allows the investigation of the role of vesicles in disease in a quasi-prospective manner to monitor disease progression. Several of their findings were somewhat expected and actually served as a validation of published findings using in vitro systems. For example, they noted that the numbers of exosomes in the plasma correlated with disease severity (11). This has been observed in previous studies of diabetes (12,13). Thus, the more EVs present in the plasma, the more advanced the disease stage. Another interesting observation by Freeman et al. is that EVs from individuals with diabetes were preferentially internalized by monocytes and B cells, in contrast to EVs from normal euglycemic individuals. This led to an increase in cytokine secretion and activation of cell survival and oxidation signals (11), suggesting that the diabetes-related EVs may act as a proinflammatory trigger.

The study revealed a few rather unexpected findings. For example, the authors observed that patients with diabetes had significantly higher levels of erythrocyte-derived EVs as characterized by positivity for CD235. In addition, their observation that the levels of insulin signaling proteins in the vesicles reflected the levels of insulin resistance and β-cell dysfunction merits further investigation. Furthermore, the authors argued that a hyperinsulinemic condition could increase an autophagy pathway resulting in the release of EVs; however, testing this hypothesis experimentally may require the use of an in vivo model of insulin resistance. Finally, the authors claimed that ultracentrifugation was a less reproducible method than the commercial kit for EV isolation. The approach for isolating EVs has been a controversial topic. Generally, the ultracentrifugation method is considered the gold standard for isolating exosomes (14). More recent studies have suggested that the use of commercial kits gives better yield and/or purity of EVs (15,16); however, Rider et al. (17) have shown that a newer approach described as ExtraPEG may even be better than the ultracentrifugation method and existing commercial kits and perhaps more economical. Ultimately, the choice of techniques for EV isolation may be determined by the desired EV subsets and/or molecular contents.

Future studies are required to dissect the antigenic materials in the EVs and to identify the innate and adaptive immune cells and molecular pathways that can react to the antigens. Heat-shock protein Hsp72, expressed in exosomes, can activate the Stat3 signaling pathway by inducing IL-6 (18). Exosomes containing viral miRNAs may activate antiviral immune responses in dendritic cells (19). Depending on the types of antigens, exosomes may activate macrophages to secrete either TNF-α or IL-10, a functional polarization of the macrophages toward type 1 (M1) or type 2 (M2) cells, respectively. Adaptive immunity can be more potent than innate immunity in driving the polarization of the macrophages. It has been shown that dendritic cells function independently from macrophages to drive inflammation in adipose tissue, possibly by activating antigen-specific T cells (20). Innate lymphoid cells in adipose tissue (21) or islets (22) may respond to early antigens or inflammatory signals to further polarize the immune responses into different profiles. Sheng et al. (23) found that exosomes contain unique antigens that activate autoreactive T cells to produce type 1 T helper (Th1) cytokine IFN-γ, possibly by first activating marginal zone-like B cells to present the antigens (24). Furthermore, retroviral antigens expressed in the vesicles may specifically induce antiviral immunity that could contribute to autoimmunity (25). Finally, genetic factors may determine the types of antigens expressed in the EVs and thus the types of immune responses that are induced. For example, adipose-derived stem cells release exosomes that drive macrophage differentiation into IL-10–secreting M2 cells (26). However, mesenchymal stem cells derived from autoimmune-prone mouse strain, NOD, release exosomes that contain antigens that activate autoreactive T cells to release the Th1 cytokine IFN-γ (4). Therefore, a polarization of the immune responses toward a type 1 reaction to the EVs could be the driving force of β-cell damage and insulin resistance.

In summary, EVs are complex not only in shape, size, and molecular content but also in cell/tissue origin and function. As illustrated by Freeman et al. (11), the study of these vesicles will immensely enhance our understanding of the mechanisms of development of diseases and may provide novel methods for monitoring disease progression. One point that is clear is that EVs are produced in abundance in diseases such as cancers, autoimmune diseases, and viral infections. Unfortunately, there are currently no available technologies that can segregate EVs based on their origin or function. The ability to segregate EVs into subspecies based on their origin or function will significantly enhance our understanding of the pathology of several diseases, particularly autoimmune diseases.