During breast cancer progression, the extracellular matrix (ECM) undergoes extensive remodeling, characterized by disruption of the basal lamina, vascular endothelial invasion, and tissue stiffening (Insua-Rodríguez and Oskarsson, 2016, Madu et al., 2020). Secretion of degradative matricellular enzymes like matrix metalloproteinases break down the existing basal membrane (Przybylo and Radisky, 2007), while increased deposition and assembly of fibrous components form a de novo desmoplastic tumor ECM which promotes tumor progression and invasion (Huang et al., 2021).

A primary component of the desmoplastic tumor ECM is fibronectin (FN) (Huang et al., 2021, Yoneda, 2015). FN is secreted as a soluble dimer that binds primarily to α5β1 integrins on the cell surface, which transmit actomyosin contractile forces to FN. This force stretches FN into an open conformation that facilitates FN-FN binding (Schwarzbauer and DeSimone, 2011, Dalton and Lemmon, 2021). This cycle of stretching to expose new binding sites and subsequent incorporation of soluble FN leads to the assembly of insoluble fibrils, in a process known as fibrillogenesis (Schwarzbauer and DeSimone, 2011, Dalton and Lemmon, 2021, Weinberg et al., 2017). While secretion of soluble FN is predominantly from hepatocytes under physiological conditions, secretion of FN from local sources is often observed in cancer (Libring et al., 2020, Tomasini-Johansson et al., 2018). In addition to serving a structural role in the desmoplastic ECM, assembled FN fibrils are also capable of binding upwards of 40 growth factors, including platelet-derived growth factors (PDGFs), fibroblast growth factors (FGFs), and of particular interest, latent transforming growth factor-β1 (TGF-β1) (Martino and Hubbell, 2010), making it a crucial regulator of cellular signal transduction in the tumor ECM. FN fibrils also serve as scaffolds for the assembly of a variety other ECM components, including collagens and elastins, facilitating the formation and maturation of new tissue (Schwarzbauer and DeSimone, 2011, Dalton and Lemmon, 2021). As such, FN fibrils not only dictate the mechanical structure of the tumor ECM, but also play a key role in regulating signaling cues to resident cells.

One of the most notable inducers of FN fibrillogenesis is TGF-β1 (Griggs et al., 2017, Hinz, 2013). TGF-β1 is produced as a complex containing the active dimer, held latent by the latency- associated peptide (LAP) to form the small latent complex (SLC). This complex is then secreted with latent TGF- β binding protein (LTBP), forming the large latent complex (LLC), which facilitates the docking of latent TGF-β1 to ECM proteins like FN (Hinz, 2013, Shi et al., 2011). TGF-β1 is incapable of signaling in its latent form, but can be activated via decreases in microenvironment pH, enzymatic cleavage, or mechanical activation via integrins (Hinz, 2013, Taylor, 2009, Buscemi et al., 2011). Integrin-based activation of latent TGF-β1 is achieved through the binding of integrin receptors (most notably those containing an αv subunit) to RGD sequences on LAP. Integrins transduce actomyosin forces from the cytoskeleton to stretch open the matrix-bound LAP, freeing the active TGF-β1 ligand (Shi et al., 2011, Buscemi et al., 2011).

FN and TGF-β1 are heavily interconnected in the tumor ECM: deposition of FN fibrils is driven by TGF-β1, while tethering of TGF-β1 to FN fibrils has been shown to play a critical role in TGF-β1 activation and signaling in epithelial cells (Griggs et al., 2017). While the effect of adding the active TGF-β1 ligand to epithelial cells has been well-studied and has been shown to drive EMT, fibrosis, and tumor ECM production (Taylor, 2009, Hargadon, 2016, Liu et al., 2018), the role of autocrine TGF-β1 signaling is less well characterized. Prior research has demonstrated that exogenous TGF-β1 stimulates the production of endogenous TGF-β1, establishing an autocrine feedback loop that poten-tiates tumor ECM signaling (Gregory et al., 2011). In contrast to this, other studies have demonstrated an antagonistic effect of autocrine TGF-β1 to exogenous TGF-β1, inhibiting cell motility (Ungefroren et al., 2020).

In the current work, we sought to investigate the role of au-tocrine TGF-β1 in FN fibrillogenesis and downstream TGF- β1 signaling. We hypothesized that deletion of autocrine TGF-β1 signaling would not affect FN fibrillogenesis, but would impair downstream signaling due to a lack of tethered TGF-β1 in the FN matrix. We probed FN fibrillogenesis in both MCF10A human mammary epithelial cells, which have been established as an in vitro model of benign epithelial cells (Griggs et al., 2017, Magdaleno et al., 2021), and MDA-MB-231 human triple-negative breast cancer cells, which form malignant tumors in mouse xenograft models (Libring et al., 2020, Park and Helfman, 2019). Using CRISPR-Cas9, we disrupted the TGFB1 gene in both cell lines, thus preventing any autocrine TGF-β1 signaling, and investigated the subsequent effects on FN fibrillogenesis in both the absence and presence of exogenous active TGF-β1.