Abstract

   Exosomes, biologically active nanoparticles (40–100 nm) released by
   hematopoietic and non-hematopoietic cells, contain a variety of
   proteins and small, non-coding RNA known as microRNA (miRNA). Exposure
   to various pathogens and disease states modifies the composition and
   function of exosomes, but there are no studies examining in vivo
   exosomal changes evoked by the acute stress response. The present study
   reveals that exposing male Fisher 344 rats to an acute stressor
   modulates the protein and miRNA profile of circulating plasma exosomes,
   specifically increasing surface heat shock protein 72 (Hsp72) and
   decreasing miR-142-5p and -203. The selected miRNAs and Hsp72 are
   associated with immunomodulatory functions and are likely a critical
   component of stress-evoked modulation of immunity. Further, we
   demonstrate that some of these stress-induced modifications in plasma
   exosomes are mediated by sympathetic nervous system (SNS) activation of
   alpha-1 adrenergic receptors (ADRs), since drug-mediated blockade of
   the receptors significantly attenuates the stress-induced modifications
   of exosomal Hsp72 and miR-142-5p. Together, these findings demonstrate
   that activation of the acute stress response modifies the proteomic and
   miRNA profile of exosomes released into the circulation.

Introduction

   The extracellular environment within multicellular organisms contains
   unique nanoparticles (40–100 nm) known as exosomes. Exosomes are
   secreted into the extracellular environment and contain molecules
   unique to their cellular origin [41][1], [42][2]. Interest in the
   immunological roles of exosomes has grown rapidly since the discovery
   that exosomes modulate immunity through expression of MHC class I and
   II molecules [43][3], activation of natural killer cells [44][4],
   stimulation of T cells [45][5], and induction of tolerance to oral
   antigens [46][6]. Exposure to a variety of pathogens or diseases, such
   as microbial infection [47][7] and cancer [48][8], modifies the
   composition and function of exosomes, providing insight into their
   potential uses as biomarkers and therapeutic tools. While a growing
   body of evidence demonstrates that exosomes can be modified as a
   consequence of pathological or disease states, to date there are few
   studies examining proteomic and miRNA changes in circulating plasma
   exosomes after physiological challenges such as the systemic stress
   response.

   Exposure to stressors, whether acute or chronic, can exert adverse
   consequences on immunity. Research has demonstrated that stressor
   exposure can exacerbate cardiovascular disease [49][9], diabetes
   [50][10], and cancer [51][11]. Additionally, research in stress
   physiology revealed its immunosuppressive impact through restrained T
   cell dependent antibody responses [52][12] and suppression of
   anti-viral host-defense [53][13]. In contrast, recent research
   demonstrates that exposure to acute stressors can potentiate innate
   immunity. Acute stressors can evoke exaggerated inflammatory cytokine
   and chemokine responses under sterile conditions [54][14], enhance
   dermatological immunity [55][15], and improve host defense to
   subcutaneous Escherichia coli (E. coli) challenge [56][16].
   Potentiation of innate immunity could be an adaptive feature of the
   acute stress response, possibly enhancing an organism’s response to
   injuries sustained from the stressor. The mechanisms responsible for
   stress-modified immunity are still under investigation, but recent
   research on the immunological function of exosomes during disease make
   them a novel target for immunomodulation during the acute stress
   response.

   To understand how acute stressor exposure might modify the
   immunological function of circulating exosomes, we examined components
   of the stress response that could be likely candidates for both
   associating with exosomes and modulating immunity. Of particular
   interest is the 72 kD heat shock protein (Hsp72), a molecular chaperone
   abundant in the plasma following acute stressor exposure [57][16]. When
   exposed to a stressor, cells synthesize intracellular Hsp72 to maintain
   cellular homeostasis by refolding denatured proteins and promoting
   cellular survival [58][17]. However, when cells release Hsp72 into the
   extracellular environment, it becomes immunologically active by
   stimulating leukocytes through toll like receptors (TLR) and inducing
   the secretion of pro-inflammatory molecules [59][18]. The mechanism of
   Hsp72 release into the circulation during stressor exposure is unclear,
   but several studies have demonstrated that Hsp72 can associate with
   exosomes in cell culture [60][19]–[61][22] and amniotic fluid [62][23].
   Thus far, it is unknown whether stress-induced Hsp72 circulating in the
   plasma is associated with exosomes. Given the reported immunological
   properties of both exosomes and extracellular Hsp72, it is important to
   determine if stressor exposure modifies host immunity by increasing
   Hsp72 associated with plasma exosomes.

   Exosomes also contain non-coding RNA known as microRNA (miRNA)
   [63][24], [64][25]. Exosomal miRNA can elicit activity on distal cells
   upon inclusion [65][25], regulating target genes and modulating
   translation of messenger RNA (mRNA). Various environmental stressors,
   such as heat [66][26] and oxidative stress [67][27], can modify miRNA
   associated with TLR mediated inflammation [68][28], pro-inflammatory
   cytokine expression [69][29], and macrophage differentiation [70][30],
   making miRNA another target of interest in stress-modified immunity. To
   explore the impact of acute stress on plasma exosome miRNA, we analyzed
   miR-142-5p, -150, -155, and -203 based on evidence of their
   differential presence in heat stressed rats [71][26], involvement in
   TLR-mediated immunity [72][28], and association with macrophage
   differentiation [73][30].

   We hypothesized that acute stressor exposure modulates the protein and
   miRNA character of circulating plasma exosomes. Our findings
   demonstrate that stress modifies plasma exosomes through up-regulation
   of surface Hsp72 and down-regulation of two miRNAs. Pathway enrichment
   analysis of the stress-modified exosomal miRNAs target genes reveals
   functionally enriched pathways implicated in the stress response.
   Further, we identify sympathetic nervous system (SNS) activation of the
   α[1]-adrenergic receptor (ADR) as an important signaling process to
   exosomal elevations of Hsp72 and down-regulation of miR-142-5p. These
   are the first studies to demonstrate that activation of the stress
   response modifies plasma exosomes that may be capable of regulating
   immunity.

Materials and Methods

Animals

   Adult male Fisher 344 rats (8–9 weeks old; Harlan) weighing 230–275
   grams were singly housed in Plexiglass Nalgene plastic cages (48×27×20
   cm) with microisolator tops in a pathogen-free barrier facility and
   allowed access to food and water ad libitum (Harlan Standard Lab Chow).
   The Fisher rat is a highly stress responsive inbred rat strain chosen
   for this study due to their robust and consistent stress response,
   allowing us to use fewer animals per group. All experimental groups
   started with n = 8 per experimental condition, but animals were
   considered outliers and dropped from the study if their results failed
   the Grubbs Test of Outliers. Following arrival at the University of
   Colorado’s Transgenic Animal Facility (room temperature maintained at
   23°C), rats were given two weeks to acclimate to the colony room prior
   to experimental procedures. Rats were handled briefly each day for one
   week prior to experimental procedures. The use of animals is necessary
   in this study because of the nature of information sought and the need
   for controlled conditions. This study was carried out in strict
   accordance with the recommendations in the National Research Council’s
   Guide for the Care and Use of Laboratory Animals. Care and use of the
   rats were in accordance with protocols (IACUC 1304.03: Extracellular
   heat shock protein 72 is a Danger Associated Molecular Pattern (DAMP)
   released by stress) reviewed and approved by the University of Colorado
   Institutional Care and Use Committee. All efforts were made to minimize
   suffering.

Inescapable tail shock and plasma collection

   Inescapable tail shock (stress) was performed as previously described
   [74][14]. Briefly, electrodes were placed across the tails protruding
   from the back of a restraining tube. Rats were exposed to 100, 1.5 mA,
   5-second, intermittent, (average inter-trial interval of 60+/−25 s)
   tail shocks administered by an automated shock system (Colbourn
   Instruments). This stress procedure is a well-established model of
   acute stressor exposure that has been well characterized in both stress
   and immune responses. Previous research has demonstrated that Fisher
   344 rats exposed to 15, 25, and 100 tail shocks have significantly
   elevated plasma Hsp72 in a time-dependent manner compared to
   non-stressed controls, and those concentrations rapidly decline one
   hour following stressor termination [75][31]. To analyze plasma exosome
   concentrations of Hsp72 at their peak elevation in the plasma, rats
   were sacrificed via rapid decapitation immediately following stressor
   termination. Trunk blood was collected in EDTA tubes (13×75 mm) and
   plasma was isolated at 3000×g at 4°C for 15 min. Spleens were
   aseptically removed and weighed. This stressor consistently elevates
   Hsp72 and inflammatory cytokines in the plasma [76][14].

Exosome enrichment with ExoQuick

   ExoQuick (Systems Biosciences) was used for plasma exosome isolation
   according to the manufacturer’s instructions. Plasma supernatant was
   aspirated and labeled “exosome depleted.” The precipitate was washed,
   re-suspended in 75 µl of PBS, and labeled “exosome enriched.” Exosome
   fractions used for visualization and Dynabead immunoprecipitation were
   derived from plasma pre-treated with Thromboplastin-D
   (ThermoScientific), an anti-coagulant, prior to ExoQuick precipitation
   to remove aggregating fibrins and fibrinogens. Briefly, 100 µL aliquots
   of the plasma were warmed to 37°C, treated with an equal volume of
   Thromboplastin-D, and incubated for 15 min. Fibrins and fibrinogens
   were pelleted from the plasma following centrifugation at 10,000 rpm
   for 5 min. The supernatant was fractionated with ExoQuick as described
   above.

Exosome Visualization

   Exosome enriched fractions from Thromboplastin-D treated plasma and
   untreated plasma were deposited on the surface of Formvar-carbon grids,
   air-dried, stained with uranyl acetate, and transmission election
   microscope (TEM) images were recorded with a Phillips CM100
   transmission electron microscope (FEI, Inc.).

   For nanoparticle tracking analysis (NTA), Thromboplastin-D treated
   plasma and exosome enriched samples were diluted 1∶10,000 while exosome
   depleted samples were diluted 1∶10 in sterile filtered PBS. NTA were
   performed using the LM10-HS instrument (NanoSight Ltd.) equipped with a
   green 532 nm laser in light scatter mode, with NTA 2.3 analytical
   software. The instrument was calibrated using commercially available
   100 nm polystyrene beads (Polysciences) and sterile filtered PBS.

ELISAs

   Multiple markers (CD63, A33, IL-6) were measured in the exosome
   enriched and depleted fractions with commercially available ELISAs to
   determine successful exosome isolation. CD63 is a membrane tetraspanin
   common to exosomes from a variety of cellular sources [77][2],
   [78][32]. Intestinal epithelial cells release exosomes associated with
   A33, a protein not found in soluble form in the plasma [79][33].
   Exosomes are not known to carry or transport IL-6, therefore, it should
   not be present in the exosome enriched fractions. Samples were run neat
   in the A33 (Uscn) and IL-6 ELISAs (R&D Systems). Exosome fractions were
   diluted 1∶20 for the CD63 ExoELISA (System Biosciences). Optical
   densities were measured using a SpectraMax Plus 354 plate reader
   (Molecular Devices) and analyzed using four-parameter curve fit and
   software (SoftMax 5.4.1). All values were corrected by total protein
   using a BCA ELISA (ThermoScientific) according to the manufacturer’s
   instructions. An additional assay using two common markers for
   exosomes, CD63 (BD Pharmingen; clone AD1, Cat. 551458, Lot 74411, 8
   µg/mL) and the membrane transport protein Rab5b (Santa Cruz; rabbit
   polyclonal IgG, clone A-20, Cat. sc-598, lot H1810, 4 µg/mL), provides
   additional evidence for successful plasma exosome isolation [80][34].
   Hsp72 was measured with a commercially available ELISA kit (Enzo
   Biosciences) according to the manufacturer’s instructions.

Membrane Disruption of Exosomes

   To confirm whether Hsp72 is associated with the surface membrane of
   plasma exosomes or contained within their lumen, the exosome membranes
   were ruptured prior to analysis. Following confirmation of the presence
   of exosomes with ExoELISA, the resuspended exosome enriched fraction
   was divided into a lysed group and a not lysed group. The exosome
   fraction in the lysed group was treated with an equal volume of exosome
   binding buffer (generously provided by System Biosciences) and
   subsequently warmed in a 37°C bath for 30 min. Exosome binding buffer
   is a proprietary detergent in 25 mM bicine buffer, pH 7.6. The exosome
   fraction from the not lysed group was left undisturbed but diluted in
   an equal volume of PBS. Both the lysed and not lysed exosome enriched
   fractions were immediately assayed for Hsp72 and BCA.

CD63 immunoprecipitation

   To examine CD63 depletion on Hsp72 concentrations, plasma was treated
   with Dynabeads (Invitrogen) coated in 50 µg/mL purified mouse anti-rat
   CD63 antibody (BD Pharmingen; clone AD1, Cat. 551458, Lot 74411)
   according to manufacturer’s instructions. Following CD63
   immunoprecipitation, exosomes were precipitated with ExoQuick as
   described above. Fractions were analyzed for BCA, CD63 and Hsp72.

α[1]-ADR antagonist administration

   To determine whether blockade of the α[1]-ADR impacts exosome
   associated Hsp72 and miRNA following stressor exposure, the antagonist,
   prazosin (Sigma-Aldrich), was administrated to rats as previously
   described [81][35]. 30 min prior to tail shock, rats received 2.0 mg/kg
   i.p. injection of prazosin dissolved in endotoxin-free water or no
   injection. Pilot studies indicate that saline injection has no impact
   on cytokine, CD63 exosome, or Hsp72 concentrations compared to rats
   receiving no injection. Following stressor termination, rats were
   sacrificed and plasma and spleens were collected. Lactate dehydrogenase
   (LDH), a marker of cell death, was measured in the plasma using a
   commercial assay (Bioo Scientific) according to the manufacturer’s
   instruction. Plasma exosomes were isolated with ExoQuick and assayed
   for Hsp72, BCA, and CD63. Exosomal miRNA was subsequently analyzed as
   described below.

miRNA quantification

   Plasma exosome RNA was purified using SeraMir (System Biosciences).
   Isolated miRNA was reverse transcribed for qRT-PCR using the TaqMan
   miRNA Reverse Transcription Kit and miRNA-specific stem-loop primers
   (Applied BioSystems). MiRNA from stressed, control, stress + prazosin,
   and control + prazosin rats were analyzed for changes in miR-142-5p,
   miR-150, miR-155, and miR-203. Analysis was performed on the reverse
   transcript from 1.2 ng RNA. All real time qPCR analysis was performed
   in a 96-well plate (Bio-Rad) on the CFX96 (Bio-Rad) with TaqMan probes.
   Samples were run at a 95°C melting step for 15 sec and 60°C annealing
   and elongation step for 60 sec for 40 cycles. Samples were normalized
   to miR-191 as a stable endogenous control. Ct values were obtained
   using CFX manager 2.0 (Bio-Rad) and all fell within the limit of
   detection. Data were deconvoluted using the ΔΔCt method with samples
   normalized from RNA of exosomes isolated from control rats.

Statistical analysis

   Two-way ANOVAs were used to analyze the effect of stress on individual
   proteins. If significant interactions were present, Fisher’s protected
   least significant difference (PLSD) post-hoc analyses were conducted.
   Data are presented as means ± the standard error of the mean (SEM).
   P<0.05 was considered statistically significant.

Results

Visualization and characterization of plasma exosomes

   Multiple analyses confirmed exosome enrichment with ExoQuick
   precipitation. TEM analysis of plasma from stressed rats revealed the
   presence of exosome-sized particles (∼80 nm) in exosome enriched
   fractions. The TEM image in [82]Figure 1A shows exosome-sized
   particles, but the highly stained background and presence of larger,
   aggregated particles are likely due to fibrins and fibrinogens present
   in the sample. [83]Figure 1B reveals exosome-sized particles without
   aggregation, suggesting that treating plasma with Thromboplastin-D
   prior to ExoQuick fractionation removes aggregating factors [84][36]
   for optimal visualization. NTA demonstrated successful isolation of
   particles 50 to 100 nm under identical scatter conditions for exosome
   enriched fractions ([85]Figure 1C) from stressed and non-stressed rats
   (control). These visual observations indicate that ExoQuick treatment
   of the plasma successfully enriches exosome-sized particles.

Figure 1. Confirmation of successful plasma exosome enrichment from male
Fisher 344 rats.

   [86]Figure 1
   [87]Open in a new tab

   A. TEM demonstrates successful isolation of exosome-sized particles
   (∼80 nm) from the plasma with ExoQuick precipitation. B. TEM shows how
   pre-treating plasma with Thromboplastin-D prior to exosome isolation
   with ExoQuick removes clouding factors and prevents vesicle
   aggregation. C. Exosome enriched and exosome depleted plasma fractions
   from stressed rats pre-treated with Thromboplastin-D were characterized
   for size distribution and quantitated by NanoSight LM10, using light
   scatter from a 532 nm green laser. ELISA analyses reveal that ExoQuick
   exosome isolation successfully enriches exosomes as marked by known
   exosome markers. Exposing rats to tail shock stress (Stress) had no
   impact on the rate of plasma exosome release compared to non-stressed
   rats (Control) as measured by: D. the tetraspanin CD63, E. the adhesion
   molecule Rab5b, F. and the intestinal exosome marker A33. G. Analysis
   of IL-6 confirms activation of the stress response and specificity of
   ExoQuick for exosome associated proteins. These data suggest that
   stress does not impact the rate of exosome release into the plasma.
   Proteomic results are expressed in means ± SE; 6–8 rats/condition.
   *indicates significant difference when compared to control rats
   (p<0.05). #indicates significant difference when compared to exosome
   depleted fraction (p<0.05). Two-way ANOVA with Fisher PLSD post hoc
   test was used.

   ELISA analyses reveal that the exosome associated proteins CD63, Rab5b,
   and A33 were significantly enriched in exosome enriched plasma
   fractions from stressed and control rats, compared to their
   corresponding exosome depleted fractions ([88]Figure 1D–F). IL-6
   concentrations, a stress responsive cytokine not known to associate
   with exosomes, was only significantly elevated in the exosome depleted
   plasma fraction from stressed rats ([89]Figure 1G), confirming
   activation of the stress response and successful isolation of proteins
   only known to associate with exosomes. These data suggest that stressor
   exposure does not impact the rate of exosome release.

Stressor exposure increases Hsp72 on the surface of plasma exosomes

   To determine if stressor exposure modifies the proteomic character of
   plasma exosomes, we examined the association between stress-inducible
   Hsp72 and exosomes. Tail shock exposure activated the stress response
   as indicated by reduction in spleen weights, as well as elevations in
   plasma corticosterone and blood glucose (Figure S1 in [90]File S1).
   Hsp72 is highly stress-responsive and, consistent with prior
   observations, stressed rats had marked accumulation of Hsp72 in the
   plasma compared to control rats ([91]Figure 2A) [92][14].
   Quantification of Hsp72 following exosome isolation revealed that
   exposure to stress significantly elevated Hsp72 in both the exosome
   depleted and exosome enriched plasma fractions compared to
   corresponding fractions in the control group ([93]Figure 2B), but Hsp72
   was significantly elevated in the exosome enriched control conditions
   compared to the exosome depleted control condition. Statistical
   analyses indicated a significant interaction between stress and exosome
   enrichment on Hsp72 concentrations. These data indicate that tail shock
   stress elevates Hsp72 in both the exosome depleted and exosome enriched
   fractions, but the highest concentration of Hsp72 is present in the
   exosome enriched fractions. Further analysis of the exosome enriched
   fractions revealed that stressed rats also have significantly greater
   accumulations of Hsp72 compared to control rats when corrected by CD63
   concentrations instead of total protein ([94]Figure 2C).

Figure 2. Exposure to an acute stressor modifies the proteomic profile of
circulating plasma exosomes.

   [95]Figure 2
   [96]Open in a new tab

   A. Male Fisher 344 rats exposed to inescapable tail shocks (Stress) had
   elevated plasma concentrations of Hsp72 compared to non-stressed rats
   (Control). Following exosome enrichment with ExoQuick precipitation, B.
   concentrations of Hsp72 were the highest in the exosome enriched
   fractions from stressed rats. C. Correction with CD63 particles
   confirmed that exposure to stress increases Hsp72 concentrations in the
   exosome enriched fraction compared to control conditions. D. Lysing the
   exosome enriched fraction increased Hsp72 concentrations for both the
   stress and control groups, but the impact of stressor exposure was
   restricted to the not lysed group. E. Following pre-treatment with
   Thromboplastin-D, plasma was divided and either treated with magnetic
   Dynabeads coated in CD63 antibodies (Anti-CD63) to immunoprecipitate
   CD63 particles from the plasma or left undisturbed (No Treatment) prior
   to ExoQuick precipitation. Anti-CD63 treatment attenuates CD63
   concentrations in the plasma exosome enriched fractions from stressed
   rats compared no treatment conditions. F. Anti-CD63 treatment also
   attenuates Hsp72 in the plasma exosome enriched fractions from stressed
   rats compared to no treatment conditions. Results are expressed in
   means ± SE; 6–8 rats/condition. *indicates significant difference when
   compared to control rats (p<0.05). #indicates significant difference
   when compared to exosome depleted fraction (p<0.05). Two-way ANOVA with
   Fisher PLSD post hoc test was used.

   As a first attempt to determine whether the stress-induced elevations
   of Hsp72 occurs within the lumen of the exosomes or if is restricted to
   their surface, we also measured Hsp72 after lysing exosomes in the
   exosome enriched plasma fractions. While lysing increased Hsp72
   concentrations in the exosome enriched fractions for both the stress
   and control groups, there was no impact of stress ([97]Figure 2D).
   These data suggest that tail shock stress impacts membrane associated
   Hsp72 on plasma exosomes, but not necessarily lumen-associated Hsp72.

   Immunoprecipitation with Dynabeads coated in anti-CD63 antibodies
   reduced CD63 concentrations in the exosome enriched fraction compared
   to untreated plasma ([98]Figure 2E). CD63 immunoprecipitation had no
   impact on CD63 concentration in the exosome depleted fraction. CD63
   depletion significantly decreased Hsp72 concentrations in the exosome
   enriched fraction from the plasma of stressed rats ([99]Figure 2F),
   indicating that a significant portion of stress-induced plasma Hsp72 is
   complexed with CD63 positive exosomes.

Stress reduces exosomal miR-142-5p and miR-203

   Plasma exosomes isolated from the α[1]-ADR study were analyzed for
   miRNA related to stress and inflammation (miR-142-5p, -150, -155, and
   -203) and variations were observed. Exosomal miR-142-5p and -203 were
   significantly reduced in stressed rats compared to exosomes isolated
   from control rats ([100]Figure 3). To our knowledge, this is the first
   evidence that in vivo exposure to an acute stressor, in the absence of
   pathogens or disease, modifies miRNA content in circulating plasma
   exosomes.

Figure 3. Impact of acute stressor and adrenergic regulation of plasma
exosomes miRNA.

   [101]Figure 3
   [102]Open in a new tab

   Exposure to tail shock stress (Stress) significantly reduced plasma
   exosome miR-142-52 and -203. Intraperitoneal administration of the
   α[1]-ADR antagonist prazosin (2.0 mg/kg) 30 min prior to Stress
   attenuated the stress-induced down-regulation of miR-142-5p, but not
   miR-203. *indicates significant fold change when compared to plasma
   exosomes from control rats in the absence of prazosin (p<0.05).
   #indicates significant effect of prazosin (p<0.05). Two-way ANOVA with
   Fisher PLSD post hoc test was used.

Pathway enrichment analyses of rno-miR-142-5p and rno-miR-203

   Based on miRDB.com query, 374 genes are predicted targets of
   rno-miR-142-5p and 386 genes are predicted targets of rno-miR-203.
   These targets were uploaded into the WebGestaldt Bioinformatics system
   and subjected to KEGG and Wiki pathway analyses. Of these 760 total
   predicted gene targets, 736 were unambiguously mapped and included in
   the analyses. [103]Table 1 exhibits the pathways identified by KEGG and
   Wiki modulated by the rno-miR-142-5p and rno-miR-203. The number of
   genes in each pathway and a p-value associated with each functional
   pathway category is reported. KEGG and Wiki analyses revealed that
   rno-miR-142-5p and rno-miR-203 are related to several functional
   categories, including metabolic pathways (25 genes), mRNA processing
   (12 genes), cancer pathways (10 genes), ubiquitin mediated proteolysis
   (9 genes), mitogen-activated protein kinase (MAPK) signaling (8 genes),
   transforming growth factor-β (TGF-β) signaling pathway (6 genes), and
   TGF-β receptor signaling pathway (6 genes).

Table 1. Functionally enriched pathways impacted by miR-142-5p and miR-203.

   Bioinformatics Database Functionally Enriched Pathway rno-miR-142-5p
   Gene Count Adj. P value
   KEGG
   Ubiquitin mediated proteolysis 9 0.0001
   Pathways in cancer 10 0.0127
   Spliceosome 6 0.0227
   VEGF signaling pathway 4 0.0263
   Renal cell carcinoma 4 0.0263
   Wnt signaling pathway 6 0.0263
   Carbohydrate digestion and absorption 3 0.0263
   Cell cycle 5 0.0263
   Gastric acid secretion 4 0.0263
   Phosphatidylinositol signaling system 4 0.0263
   WIKI
   Proteasome degradation 6 0.0003
   mRNA processing 7 0.0018
   Kit receptor signaling pathway 5 0.0040
   rno-miR-203 Gene Count Adj. P value
   KEGG
   Metabolic Pathways 25 0.0018
   TGF-beta signaling pathway 6 0.0034
   Aldosterone-regulated sodium reabsorption 4 0.0140
   Fc gamma R-mediated phagocytosis 5 0.0168
   Insulin signaling pathway 6 0.0168
   MAPK signaling pathway 8 0.0280
   RNA degradation 4 0.0343
   Hypertrophic cardiomyopathy (HCM) 4 0.0343
   Phosphatidylinositol signaling system 4 0.0343
   Focal adhesion 6 0.0343
   WIKI
   MAPK signaling pathway 7 0.0215
   Diurnally regulated genes with circadian orthologs 3 0.0430
   NFE2L2 2 0.0430
   P53 pathway 3 0.0430
   mRNA processing 5 0.0430
   Adipogenesis 5 0.0430
   Insulin signaling 6 0.0430
   Wnt signaling pathway 5 0.0430
   Small ligand GPCrs 2 0.0463
   TGF-beta receptor signaling pathway 6 0.0473
   [104]Open in a new tab

   KEGG and Wikipathway (Wiki) analysis revealed functionally enriched
   pathway categories generated from genes significantly differentially
   represented in plasma exosomes from rats exposed to inescapable tail
   shock.

Prazosin attenuates stress-modified changes in exosomal Hsp72 and miR-142-5p

   Consistent with previous findings, prazosin administration in stressed
   rats reduced plasma concentrations of Hsp72 compared to untreated rats
   exposed to the same stressor ([105]Figure 4A) while boosting
   stress-induced elevations in blood glucose and attenuating
   stress-induced reductions in spleen weight (Figure S2 in [106]File S1)
   [107][35]. Prazosin did not impact plasma Hsp72, spleen weight, or
   blood glucose concentrations in the control rats, confirming prazosin
   selectivity to sympathetic nervous system (SNS) activity [108][35],
   [109][37], [110][38]. Prazosin administration attenuated stress-induced
   cell death, as quantified by LDH ([111]Figure 4B). Prazosin had no
   impact on CD63 ([112]Figure 4C) or total protein (Figure S2 in
   [113]File S1); however, α[1]-ADR blockade significantly decreased Hsp72
   concentrations in both exosome fractions ([114]Figure 4D).
   Additionally, reductions in miR-142-5p were reversed with prazosin
   administration ([115]Figure 3).

Figure 4. Adrenergic regulation of exosomal Hsp72.

   [116]Figure 4
   [117]Open in a new tab

   Adult male Fisher 344 rats were either injected intraperitoneally with
   the α[1]-ADR antagonist prazosin (2.0 mg/kg) 30 min prior to exposure
   to tail shock stress (Stress) or left undisturbed. Plasma analysis
   reveals that prazosin administration A. significantly attenuates
   stress-induced elevations of Hsp72 as well as B. lactate dehydrogenase
   activity, a marker of cell death. Exosomes were fractionated with
   ExoQuick in the stressed rats, revealing that prazosin administration
   C. had no effect on CD63 particle concentrations, but D. significantly
   attenuates Hsp72 concentration in both the exosome depleted and exosome
   enriched fractions. Results are expressed in means ± SE; 6–8
   rats/condition. *indicates significant difference when compared to
   control rats (p<0.05). #indicates significant difference when compared
   to no treatment group or exosome depleted fraction (p<0.05). Two-way
   ANOVA with Fisher PLSD post hoc test was used.

Discussion

   A variety of disease states can change both the composition and
   function of exosomes. What remains less clear is if exosomal
   modifications occur during the acute stress response. In this series of
   experiments, we reveal that exposure to an intense and acute stressor
   in the absence of injury or disease alters the proteomic and miRNA
   composition of plasma exosomes. Additionally, this stress-induced
   modification of plasma exosomes is partially mediated by SNS activation
   of α[1]-ADRs.

   Circulating Hsp72 is increased following exposure to a variety of acute
   stressors [118][14], [119][35] and the release of the protein may
   contribute to stress-modulated immunity [120][16], [121][39] through
   activation of macrophages, dendritic cells, and neutrophils, inducing
   the secretion of pro-inflammatory cytokines [122][40]. Here we provide
   evidence that a significant portion of stress-induced Hsp72 is released
   into the blood via an exosome release pathway. CD63 immunoprecipitation
   of plasma from stressed rats prior to exosome isolation corresponds to
   a marked decrease in Hsp72, further supporting its association with
   circulating exosomes. Analysis of Hsp72 in the exosome enriched
   fractions following lysing revealed that while Hsp72 is present within
   the exosome lumen, the stress-induced changes in exosomal Hsp72 are
   restricted to the surface. Consistent with the exosome literature, our
   data indicate that exosome-associated Hsp72 is released, albeit at low
   levels, during basal states [123][19], [124][20], [125][23]. Stressor
   exposure also elevates Hsp72 in the exosome depleted plasma fractions,
   suggesting that a portion of circulating Hsp72 is soluble, perhaps
   released into the blood after necrotic cell death [126][14].
   Additionally, there is evidence that enzymes in the blood may cleave
   exosome associated proteins [127][41], potentially releasing soluble
   Hsp72 from stress-modified plasma exosomes. Together, these findings
   demonstrate that circulating exosomes in stressed rats have higher
   concentrations of surface Hsp72 compared to non-stressed control rats,
   and the presence of soluble Hsp72 could be the result of either
   stress-induced cell death or enzymatic cleavage of the exosomal
   proteins.

   The data presented here also demonstrate that exposure to an acute
   stressor down-regulates miR-142-5p and -203 in circulating plasma
   exosomes, likely affecting macrophage and dendritic cell-mediated
   immunity. Recent findings indicate that down-regulation of miR-142
   promotes early normal macrophage differentiation [128][30].
   Down-regulation of miR-142-5p is also known to enable cytokine-mediated
   survival signaling in response to hemodynamic stress through the IL-6
   receptor, gp130 [129][27], further supporting its immunological
   contribution during the stress response. Additionally, miR-203 is known
   to directly target suppressor of cytokine signaling 3 (SOCS3) and block
   its role as an inhibitor of cytokine signal transduction [130][42].
   Activation of SOCS3 terminates the innate immune response; therefore,
   stress-induced reductions in exosomal miR-203 could prevent activation
   of SOCS3 and its downstream suppression of pro-inflammatory cytokines.
   Down-regulation of miR-203 is also capable of inducing TNF-α and IL-6
   synthesis by enabling myeloid differentiating factor 88 (MyD88)
   activation [131][43].

   Pathway analyses of the 760 predicted mRNA targets of miR-142-5p and
   -203 revealed functionally enriched pathways that overlap with
   demonstrated targets of the stress response. For example, exposure to
   the same stressor tested here is known to modulate genes involved in
   the TGF-β pathway [132][44], which is a prime target of miR-203. In
   addition, miR-203 is known to repress the anti-inflammatory properties
   of TGF-β [133][45]. Therefore, stress-induced reductions in miR-203
   likely promote elevations in TGF-β activity, which may be crucial for
   constraining stress-induced immunomodulation. KEGG analysis also
   revealed that miR-203 targets metabolic pathways, which is in line with
   cortisol mediation of metabolic pathways during the stress response
   [134][46]. Additional KEGG analysis of predicted gene targets for
   miR-142-5p revealed involvement with ubiquitin mediated proteolysis.
   Reductions in miR-142 are associated with elevations in ubiquitin
   proteasome activity [135][47], which is a cellular strategy for
   degrading proteins denatured by stress and repairing DNA damage
   [136][48], Collectively, these observations suggest that stress-induced
   down-regulation of miR-142-5p and -203 and up-regulation of Hsp72 may
   be crucial components of exosome-based stress-induced immunomodulation
   and cytoprotection, which is consistent with previous findings
   demonstrating the immunomodulating and cellular consequences of stress
   [137][39], [138][49]. Future studies should examine the impact of
   stress-modified exosomes on these gene targets in cell lines, which
   would provide valuable insight into the role of exosomal miRNA during
   the stress response.

   This study also reveals that exosomal modification in Hsp72 and
   miR-142-5p from stressed rats rely on sympathetic nervous system
   release of norepinephrine (NE) and its subsequent activation of one of
   its target receptors, the α[1]-ADR. Consistent with previous research,
   blockade of the α[1]-ADR with prazosin prior to inescapable tail shock
   stress reduces stress-induced elevations of Hsp72 in plasma [139][35],
   [140][50], and attenuates the stress-induced down-regulation of
   miR-142-5p. Since down-regulation of miR-142-5p is known to enable
   cytokine-mediated survival [141][27], prazosin administration should
   decrease associated inflammatory cytokine activity. Indeed, previous
   research demonstrates that pre-treatment with prazosin prior to tail
   shock stress attenuates stress-induced elevations in monocyte
   chemotactic protein-1 (MCP-1) and IL-1β [142][51]. In line with these
   findings, administration of phenylephrine, an α[1]-ADR agonist, in the
   absence of stress induces an elevation of plasma Hsp72 similar to
   levels seen in rats exposed to inescapable tail shock [143][35];
   however, additional studies are needed to determine if stimulation of
   the α[1]-ADR in the absence of an acute stressor modifies exosomal
   Hsp72 and miRNAs. Interestingly, prazosin administration also decreased
   Hsp72 in the exosome depleted fraction of stressed rats. LDH assessment
   of the plasma reveals that prazosin administration attenuates cell
   death, which likely decreases the necrotic release of soluble Hsp72
   into the circulation. While both NE and epinephrine (E) bind to
   α[1]-ADRs, NE has a higher affinity for these receptors and depletion
   of E through adrenalectomy has no effect on stress-induced Hsp72 in the
   circulation [144][35].

   Based on these findings, we hypothesize that exposure to an intense,
   acute stressor modifies plasma exosome cargo, specifically Hsp72 and
   miR-142-5p, by activating the SNS and inducing the release of NE from
   sympathetic nerve terminals. Stimulation of the α[1]-ADRs by NE
   activates phospholipase C and elevates cytosolic Ca^2+. Fusion of the
   multivesicular body (MVB), the endocytic source of exosomes, to the
   plasma membrane is Ca^2+ dependent; therefore, the surge in Ca^2+ may
   facilitate exosome release. Alternatively, NE stimulation of ADRs
   increases ubiquitination through Ca^2+ flux [145][52], which is
   required for targeting cellular proteins to endosomes prior to fusing
   with the MVB. Interestingly, the presence of multiple exosome markers,
   such as the tetraspanin CD63, the membrane transport protein Rab5b, and
   the intestinal epithelial exosome marker A33, were unchanged following
   exposure to inescapable tail shock. Additionally, CD63 concentrations
   are unaffected by prazosin, therefore it is unlikely that activation of
   the α[1]-ADR impacts exosome release, but rather their composition
   through ubiquitination. Thus, exposure to an acute stressor potentially
   modifies exosome-associated Hsp72 and miRNA in the plasma by increasing
   their rate of loading onto intracellular endosomes rather than
   impacting the rate of secretion. Conversely, α[1]-ADR activation may be
   critical for Hsp72 synthesis or miRNA transcription. A recent study
   demonstrated that blockade of the α[1]-ADR in stressed rats attenuated
   stress-induced increases of intracellular Hsp72 in the spleen, liver,
   and subcutaneous adipose [146][51], however, it is unknown if these are
   the tissue source for plasma exosome Hsp72. Future studies should
   elucidate the tissue origin of these stress modified exosomes and
   examine if α[1]-ADR blockade in stressed animals affects their
   cytosolic Hsp72 and miR-142-5p, thus indicating whether α[1]-ADR
   activation is critical for their transcription and synthesis or their
   exosomal loading.

   The release of Hsp72 and miRNAs through an exosomal pathway has several
   advantages. Most notably, exosomes provide a protective lipid bi-layer
   that can facilitate long distance communication between cells. From a
   stress physiology perspective, this form of cellular communication may
   be evolutionarily advantageous. For example, if an organism is
   subjected to a harmful stressor, such as a predator, the organism’s
   cells could secrete stress-modified exosomes into the circulation prior
   to experiencing injury. When injury occurs, stress-modified exosomes
   are already in the circulation and available to facilitate the host
   immune response. Cytokine induction by resident leukocytes at the site
   of injury triggers the activation and expression of adhesion molecules
   on the adjacent vascular endothelium. Since intercellular adhesion
   molecules (ICAMs) are present on exosomes [147][5], plasma exosomes
   could bind ICAM receptors on vascular endothelial cells and
   consequently leave the circulation and migrate to the injured tissue.
   At the site of infection, stress-modified exosomes could boost innate
   immunity through Hsp72-mediated TLR4 activation of macrophages and
   neutrophils, or transfer their content through clathrin-mediated
   endocytosis with a recipient cell, subsequently stimulating a
   pro-inflammatory cytokine response and enhancing the organism’s chance
   of survival.

   There are a variety of clinical applications for stress-modified
   exosomes that could potentially modulate immunity. For example,
   exposing cells to a non-lethal stressor, such as heat, to elevate Hsp72
   and down-regulate miR-142-5p and -203 in exosomes could enhance the
   immunogenicity of exosome-based vaccines, specifically cancer vaccines
   [148][8], [149][53]. Alternatively, since exosomes are capable of
   delivering their content to recipient cells [150][25], they could
   transfer their content to target cells, where Hsp72 could translocate
   to the target cell’s cytosol and perform its cytoprotective and
   anti-apoptotic functions and miRNAs could modulate mRNA translation.

   In summary, our data indicate that in vivo exposure to an acute
   stressor modifies the proteomic and miRNA character of exosomes
   released into the plasma, likely impacting innate immune function
   through TLR association, monocyte differentiation, and cytokine
   secretion. Furthermore, our results suggest that SNS activation of
   α[1]-ADRs is a critical component of some of these exosomal
   modifications. Given the known immunomodulatory and protective
   functions of Hsp72, miRNA, and exosomes, we speculate that modulation
   of plasma exosomes is a critical component of the stress response.
   Future studies should further identify the immunomodulatory factors and
   cellular sources of stress-modified exosomes in the plasma, which will
   challenge current paradigms concerning the mechanisms of stress-evoked
   modulation of immunity and advance knowledge concerning their use in
   immunotherapy.

Supporting Information

   File S1

   Figure S1, Inescapable tail shock activates the stress response. Figure
   S2, Alpha-1 adrenergic regulation of blood glucose, spleen weight, and
   total protein.

   (DOCX)
   [151]Click here for additional data file.^ (323.9KB, docx)

Acknowledgments