Argonaute-2 (Ago2) is a key component of the RNA-induced silencing complex that mediates downregulation of mRNA by miRNAs. Its presence in extracellular vesicles (EV) has been postulated to be important for the activity of EV-carried miRNA in modulating gene expression in recipient cells. However, whether it is in fact contained within EVs or is instead an extravesicular contaminant is controversial. In this opinion piece, we argue that the ability to detect Ago2 in EVs is a result of multiple factors, including cell source, cell signaling control of Ago2 trafficking to EVs, experimental conditions, and detection methods.

Extracellular RNA comes in many forms, including cargo within extracellular vesicles (EV) and nonvesicular RNA associated with RNA-binding proteins (RBP) or lipoproteins. EV-carried miRNA has been of particular interest due to its potential to be transmitted to recipient cells and downregulate target mRNAs. As mRNA silencing and degradation is thought to require Argonaute-2 (Ago2), a key member of the RNA-induced silencing complex (RISC), whether Ago2–miRNA complexes are present in EVs has been of great interest to the field. One possibility is that miRNAs must be bound to Ago2 to efficiently assemble into functional RISCs for downregulation of mRNA targets in recipient cells. However, Ago2 has been variably reported by diverse investigators to be either inside or outside of EVs (1–9), leading to confusion as to whether Ago2–RNA complexes are transmitted via EVs. The location of Ago2–miRNA complexes to the inside of EVs has functional significance, as fusion of EVs with recipient cell membranes should deliver internal vesicle cargoes to the cytoplasm. In contrast, it is unclear how nonvesicular Ago–RNA complexes could access the cytoplasm of recipient cells to regulate gene expression. In this perspective, we argue that the ability to detect Ago2 in EVs is a result of multiple factors, including biological context, experimental conditions, and detection methods. A major confounding factor is the presence of large amounts of nonvesicular Ago2 and growth factors in serum, which is commonly used during cell culture.

EVs are small lipid-enclosed vesicles that are released from cells and carry diverse protein, lipid, and nucleic acid cargoes. They promote cellular communication in both an autocrine and paracrine fashion by multiple mechanisms, including direct induction of recipient cell signaling by ligand–receptor interactions and delivery of cargoes such as RNAs to the cytoplasm of recipient cells. Formation of EVs from endosomal membranes or the plasma membrane leads to an EV topology such that cytoplasmic cargoes such as RNAs are included inside EVs. Given this topology, EV-mediated delivery of RNAs to the cytoplasm of recipient cells should require fusion of EVs with recipient cell plasma membrane or endosomal membranes. It is not clear how RNA that is associated with the outside of EVs would be able to directly affect gene expression via miRNA-RISC-mRNA mechanisms, although such RNAs might be able to activate Toll-like receptors that are located on the inside of endosomes to induce signaling responses.

The RISC is a multiprotein complex that binds miRNAs or other siRNAs and induces downregulation of complementary strands of RNA. A key component of this complex in mammalian cells is Ago2, which of all the argonautes uniquely has RNA-slicing activity. Dicer and transactivation response RBP are two other components of the RISC-loading complex that, respectively, process pre-miRNAs to their mature form and recruit Ago2 so that it can recognize target RNAs for silencing. Because miRNAs are present in EVs and have been shown to downregulate complementary mRNAs in recipient cells, the role of Ago2 in carrying miRNAs into EVs and mediating their biological effects has been of great interest to the field. Of note, the entire RISC-loading complex was reported to be present in EVs and to mediate miRNA maturation (6). However, Ago2–RNA complexes are also present in body fluids in a nonvesicular form that appears to greatly exceed the amount contained within EVs (1, 8). The origin of the nonvesicular form is unknown but has been postulated to derive from dead cells. We posit that Ago2–RNA complexes are present in both forms, but that the relative quantities and ability to detect them depend on multiple factors that we discuss below.

Regulation of Ago2 sorting by biological context

The molecular cargoes of EVs are selected by an active process, which is subject to regulation by the signaling and metabolic state of the cell. Cellular levels and subcellular distribution of proteins are also likely to contribute to determining the molecular content of EVs. Jeppesen and colleagues recently used high resolution density gradients to separate a crude small EV pellet into nonvesicular and vesicular fractions (3). Using Western blot analysis, they found that argonautes, including Ago2, were present in both the vesicular and nonvesicular fractions of Gli36 glioblastoma EVs but only in the nonvesicular fractions of DKO1 or MDA-MB-231 EVs. As both DKO1 and MDA-MB-231 cells carry KRAS mutations, these findings are consistent with our previous finding that KRAS-MEK-ERK signaling greatly downregulates association of Ago2 with multivesicular bodies and reduces subsequent incorporation into exosome-type small EVs (5). Downregulation of Ago2 sorting to exosomes occurs both via growth factor–induced KRAS activation and through oncogenic mutations (5). Notably, most studies that have reported that Ago2 is not in EVs used KRAS-mutated cancer cells and/or growth factor–containing serum in their medium (3, 7, 9).

In contrast to signal-mediated downregulation of Ago export, several pathogens appear to upregulate Ago2 trafficking into EVs. Thus, malarial infection of primary red blood cells (RBC) or viral infection of Huh7.5 liver carcinoma cells have been shown to enhance Ago2 and miRNA trafficking into EVs (2, 4). In addition, a related “worm” argonaute protein (WAGO) was recently described to be transported into the EVs of parasitic worms in association with novel siRNAs (10). In some of those cases, the EV-associated Ago–RNA complexes were shown to contribute to pathogen infections (2, 4).

Effects of experimental conditions on Ago2 sorting and detection

Many experimental analyses of EVs are performed after purifying EVs from cultured cells. The metabolic and signaling state of those cultured cells can greatly affect the composition and extent of EV secretion. Serum is a typical cell culture additive that contains large amounts of growth factors, that would be likely to downregulate Ago2 trafficking into exosomes (5). In addition, serum contains a very large amount of nonvesicular Ago2 (1, 8), which may contaminate crude EV pellets. The high signal from this non-EV–contained Ago2 may further decrease the ability to detect lower amounts of EV-contained Ago2 in density gradient fractions by Western blots, because exposure times are often determined by the bands with the highest signal. In addition to serum, the confluency of cells in 2D tissue culture can alter the composition (7) and extent of EV secretion, as would 3D culture methods.

Role of purification and analysis methods in detection of nonvesicular and vesicular Ago2

An argument against Ago2 being a specific cargo of EVs is that it could be a contaminant of EV preparations prepared by ultracentrifugation of conditioned medium at 100,000 × g. Density gradient centrifugation is one method that has been widely used to separate vesicular from nonvesicular components. Using such methods, Ago2 has been identified in both the vesicular and nonvesicular fractions by Western blot analysis (3–5, 7, 9). While such analyses seem straightforward, such factors as whether to load the gel samples based on fraction volume, protein, or EV number may greatly affect the resulting signals. In addition, antibody choice and specificity will necessarily affect the results. For example, many of the commercially available Ago2 antibodies are less than ideal, often detecting nonspecific bands and/or lacking sensitivity. In addition, EV markers used to identify EV-containing density gradient fractions may detect different subpopulations of vesicles, which may or may not be the subpopulation that contains Ago2.

Because it is theoretically possible that even highly purified preparations may be contaminated with nonvesicular components, additional methods to determine whether protein and RNA cargoes are on the outside or inside of vesicles have been employed. These include proteinase and RNase digestion in the absence or presence of detergent. Internal EV cargo should only be susceptible to digestion in the presence of detergent to dissolve or otherwise permeabilize the lipid bilayer membrane of EVs. Using these methods, Ago2 and WAGO have been shown to be inside the EVs of cancer cells, malaria-induced RBC EVs, and parasitic nematodes (4–6, 10). Electron microscopy has also been used to visualize Ago2 localization to the inside of EVs (4). A final method to identify whether Ago2 segregates with EVs is immunoprecipitation of purified EVs with antibodies against the EV tetraspanin markers CD63 or CD81 followed by Western blot analysis; however, the results have been mixed with Ago2 detected in some cases (5) and not in others (3).

Cellular release of EVs from cells occurs by multiple biogenesis mechanisms, leading to the incorporation of diverse cargo into various types of small and large EVs. This complexity is required for effective cell–cell communication, in which EVs from different cell types and/or cell states deliver distinct cargoes to recipient cells. Ago2 is just one of these cargoes and appears to be regulated by a number of conditions, including cellular signaling and pathogen infections. On the basis of the available data, Ago2 is frequently absent from EV preparations; however, the data are also compelling for Ago2 and related argonautes being selected for inclusion in specific subsets of EVs. We propose that a more nuanced view of the role of Ago2 and other “controversial” cargoes should be taken, realizing that there is great EV heterogeneity that can depend on many factors and is often cell context specific. Indeed, identifying the cell and tissue states that lead to incorporation of those cargoes may lead to greater understanding of EV functions in cellular communication.

No potential conflicts of interest were disclosed.

Writing, review, and/or revision of the manuscript: A.M. Weaver, J.G. Patton

All authors received funding from U19CA179514 grant, which is a Center grant in the Common Fund Extracellular RNA Communication Consortium awarded to Dr. Robert Coffey.

1.
Arroyo
JD
,
Chevillet
JR
,
Kroh
EM
,
Ruf
IK
,
Pritchard
CC
,
Gibson
DF
, et al
Argonaute2 complexes carry a population of circulating microRNAs independent of vesicles in human plasma
.
Proc Natl Acad Sci U S A
2011
;
108
:
5003
8
.
2.
Bukong
TN
,
Momen-Heravi
F
,
Kodys
K
,
Bala
S
,
Szabo
G
. 
Exosomes from hepatitis C infected patients transmit HCV infection and contain replication competent viral RNA in complex with Ago2-miR122-HSP90
.
PLoS Pathog
2014
;
10
:
e1004424
.
3.
Jeppesen
DK
,
Fenix
AM
,
Franklin
JL
,
Higginbotham
JN
,
Zhang
Q
,
Zimmerman
LJ
, et al
Reassessment of exosome composition
.
Cell
2019
;
177
:
428
45
.
4.
Mantel
PY
,
Hjelmqvist
D
,
Walch
M
,
Kharoubi-Hess
S
,
Nilsson
S
,
Ravel
D
, et al
Infected erythrocyte-derived extracellular vesicles alter vascular function via regulatory Ago2-miRNA complexes in malaria
.
Nat Commun
2016
;
7
:
12727
.
5.
McKenzie
AJ
,
Hoshino
D
,
Hong
NH
,
Cha
DJ
,
Franklin
JL
,
Coffey
RJ
, et al
KRAS-MEK signaling controls ago2 sorting into exosomes
.
Cell Rep
2016
;
15
:
978
87
.
6.
Melo
SA
,
Sugimoto
H
,
O'Connell
JT
,
Kato
N
,
Villanueva
A
,
Vidal
A
, et al
Cancer exosomes perform cell-independent microRNA biogenesis and promote tumorigenesis
.
Cancer Cell
2014
;
26
:
707
21
.
7.
Shurtleff
MJ
,
Temoche-Diaz
MM
,
Karfilis
KV
,
Ri
S
,
Schekman
R
. 
Y-box protein 1 is required to sort microRNAs into exosomes in cells and in a cell-free reaction
.
Elife
2016
;
5
.
DOI: 10.7554/eLife.19276
.
8.
Turchinovich
A
,
Weiz
L
,
Langheinz
A
,
Burwinkel
B
. 
Characterization of extracellular circulating microRNA
.
Nucleic Acids Res
2011
;
39
:
7223
33
.
9.
Van Deun
J
,
Mestdagh
P
,
Sormunen
R
,
Cocquyt
V
,
Vermaelen
K
,
Vandesompele
J
, et al
The impact of disparate isolation methods for extracellular vesicles on downstream RNA profiling
.
J Extracell Vesicles
2014
;
3
.
DOI: 10.3402/jev.v3.24858
.
10.
Chow
FW
,
Koutsovoulos
G
,
Ovando-Vazquez
C
,
Neophytou
K
,
Bermudez-Barrientos
JR
,
Laetsch
DR
, et al
Secretion of an argonaute protein by a parasitic nematode and the evolution of its siRNA guides
.
Nucleic Acids Res
2019
;
47
:
3594
606
.