Background

Adipose tissue currently is one of the most appealing sources of mesenchymal stem cells (MSCs), due to its large abundance and relatively less-invasive harvest methods (Shen et al., 2011; González-Cruz et al., 2012; Konno et al., 2013). Adipose-derived MSCs, that isolated from the stromal-vascular fraction of subcutaneous adipose tissue, have been demonstrated to display multilineage potentials both in vitro and in vivo (Anghileri et al., 2008; González et al., 2009; Gonzalez-Rey et al., 2010; Jumabay et al., 2010; Mao et al., 2017 and 2019; Darnell et al., 2018). To isolate adipose-derived MSCs, the widely-used method is to dissect the stromal-vascular fraction from the adipose tissue, and then sort the MSCs by either fluorescence-activated cell sorting (FACS) or culture selection (Aronowitz et al., 2015; Raposio et al., 2017; Gentile et al., 2019). However, heterogeneous groups of cells are contained in a stromal-vascular fraction of adipose tissue, and limited cell markers are available for MSCs selection; these make it difficult to purify adipose-derived MSCs (Gimble et al., 2011; González-Cruz et al., 2012; Konno et al., 2013).

Alternatively, the adipocytes, rather than the other types of cells in adipose tissue, can spontaneously dedifferentiate into multipotent mesenchymal cells named the dedifferentiated fat (DFAT) cells during in vitro culturing (Sugihara et al., 1986; Shen et al., 2011; Taniguchi et al., 2016). Because of the multipotency of the DFAT cells and the large abundance of the mature adipocytes, the DFAT cells have been regarded as an ideal source for human postnatal mesenchymal multipotent stem cells (Matsumoto et al., 2008; Shen et al., 2011; Côté et al., 2019). However, the current ceiling culturing for adipocyte dedifferentiation requires a long duration (typically 4 weeks) to enable the adipocytes to spontaneously lose all obvious lipid droplets (Lessard et al., 2015; Taniguchi et al., 2016). Thus, further increasing the efficiency of adipocyte dedifferentiation and shortening its processing time is attractive for its wider applications.

Adipocytes and adipose progenitor cells are also important components in tumor microenvironments (Chandler et al., 2012; Seo et al., 2015; Ling et al., 2020). Recent studies revealed that the dedifferentiation of adipocytes occurred during tumor development, which might be attributed to the activated Wnt signaling (Gustafson and Smith, 2010; Bochet et al., 2013) and Notch signaling (Bi et al., 2016). Recent studies also revealed that the dedifferentiation of adipocytes could occur in vivo in mice models (Bochet et al., 2013; Liao et al., 2015; Wang et al., 2018). Tumor progression also largely alters the local physical microenvironments, including elevated osmotic pressure, increased compressive force, and matrix stiffening (Nia et al., 2020). These physical cues largely influence the cell fates of both adipose stromal cells and cancer cells (Guo et al., 2017; Li et al., 2019 and 2020a; Han et al., 2020). Indeed, our recent study reported that the generation of osmotic stress in vitro to mimic the elevated osmolarity in in vivo tumors could also induce the dedifferentiation of adipocytes (Li et al., 2020b). Consistently, another study also reported that a tough implant in vivo drove the dedifferentiation of the local surrounding adipocytes (Ma et al., 2019). Thus, these studies inspired us to develop an alternative protocol to generate multipotent mesenchymal cells by mechanically dedifferentiating adipocytes.

The protocol described here includes the experimental set-ups to induce and verify the reprogramming of adipocytes into multipotent mesenchymal cells using our hypertonic dedifferentiation medium. We also include the procedures to generate adipocytes from preadipocytes or mesenchymal stem cells, and the differentiation assays to confirm the multilineage potentials of the CiDAs.