**1. Introduction**

Hepatic fibrosis is the result of chronic liver injury and is characterized by the deposition of collagen and other insoluble extracellular matrix components [1]. The major fibrosis-causing cells in the liver are hepatic stellate cells (HSC) that reside in the Space of Disse [1–3]. Under normal conditions, HSC are quiescent pericyte-like cells that store large amounts of vitamin A in cytoplasmic lipid droplets. However, in response to liver injury. HSC undergo a phenotypic activation whereby they become contractile collagen-producing myofibroblasts that function either transiently in acute injury to produce a provisional matrix that supports hepatocyte repopulation or persistently in chronic injury to unrelentingly produce fibrotic material that causes scarring and exacerbates impaired liver function. These properties arise due to increased production and/or response by HSC of molecules involved in fibrogenesis, survival and inflammation [4]. The central role of HSC in fibrosis has spawned intense effort to target either key molecular mediators of fibrosis or the activated cells themselves (e.g., using cytotoxic agents) in a quest to develop new therapies for liver fibrosis [5,6]. While hepatic fibrosis is a major cause of morbidity and mortality and is a significant contributing factor to cirrhosis, which accounts for 32,000 deaths in the US and more than 1 million deaths globally each year [7], there is a severe lack of approved anti-fibrotics for improving patient outcomes [8,9]. Rational therapeutic

strategies are being developed by leveraging our knowledge of the molecular mechanisms that regulate fibrosis and this has already led to the identification of a considerable number and variety of targets that are at various stages of pre-clinical or clinical testing [10,11]. Success in this area may most likely be achieved by combining anti-fibrotic strategies with other modalities that target hepatocyte injury or inflammation [10,12], but even so will require diligent design of clinical trials that have historically been vexed with difficulties and setbacks for reasons that include differences in fibrosis reversibility between humans and animal models, role of genetic background on fibrosis penetrance, protracted time-course of fibrosis development and influence of disease etiology on drug effectiveness [10,13,14]. Continued research of HSC biology is necessary to improve our understanding of the cellular and molecular aspects of fibrosis and to optimize development of new therapeutic options.

The biological activities of HSC occur as a result of their orchestrated interactions with other cells in the liver [12,15]. Recently, a new hepatic signaling network has been identified that involves the shuttling of molecular information between different liver cells by extracellular vesicles (EVs) [16–20]. EVs are membranous nanovesicles (50–500 nm) that comprise principally exosomes and microvesicles which are liberated from cells by, respectively, fusion of multivesicular bodies with the plasma membrane or budding off from the plasma membrane [21,22]. Despite many early reports that exosomes and microvesicles can be distinguished by size and/or the presence of marker proteins, it has recently been emphasized [23] that these features do not allow such discrimination. Thus, the use of the terms "exosome" or "microvesicle" firstly infers knowledge of specific EV biogenetic pathways that are extremely difficult to ascertain given current tools and technical limitations and, secondly, fails to take into account the innate heterogeneity of EVs in biological samples. Accordingly, since current recommendations for operationally defining EVs include reference to the cell of origin [23], in this report we have used "EVs from HSC" or 'HSC EVs" to describe the population of EVs that we recovered from HSC-conditioned culture medium using the stated enrichment methods. Irrespective of the precise EV subtypes involved, accumulating data suggest that a complex EV network exists in the liver whereby quiescent or activated HSC exchange EV cargo molecules (proteins, mRNAs, microRNAs) with one another or with cells such as hepatocytes or liver sinusoidal endothelial cells [24–32]. This bi-directional molecular transfer ensures that normal homeostatic functions or injury-associated responses in the liver are fine-tuned and synchronized at the cellular level.

An area of growing interest is the manner in which the EV cargo content varies according to the phenotypic status of their cells of origin. In view of the highly distinct functions of quiescent versus activated HSC, we hypothesized that the EVs from these cells exhibit fundamental differences in their biological properties and/or molecular payload. In this report we provide evidence for such differences in terms of their respective actions on cultured HSC and their protein cargo as assessed by mass spectrometry and proteomic analysis. We also provide a comparative proteomic analysis of EVs from activated mouse and human HSC and show substantial overlap in their constituent protein content and their predicted functional roles.
