Pancreatic ductal adenocarcinoma (PDAC) is a devastating type of malignancy, with one of the worst prognoses of all solid cancers (Hosein et al., 2022). It develops in the exocrine compartment of the pancreas and accounts for more than 90% of pancreatic cancer cases. In contrast to other tumor types, multiple large-scale trials using targeted agents have been unsuccessful for PDAC, thus chemotherapy (largely, gemcitabine-based regimens) remains the standard of care for patients with advanced PDAC (Amrutkar and Gladhaug, 2017; Lambert et al., 2019; Hosein et al., 2022). Yet, chemotherapy only modestly improves survival, as median OS rates remain at approximately 1 year (Amrutkar and Gladhaug, 2017; Lambert et al., 2019; Hosein et al., 2022). Several mechanisms were proposed to explain PDAC aggressiveness and the limited efficacy of the existing therapies (Raimondi et al., 2009; Lambert et al., 2019). Among them, excessive deposition of desmoplastic extracellular matrix [ECM] posing physical barriers to drug delivery and nutrient/oxygen passage (Hosein et al., 2022), along with altered intracellular metabolism rewired toward increased glucose uptake/anabolic fermentation within the PDAC cells (Ying et al., 2012), attracted the most attention. In parallel with these intratumoral/intracellular phenomena, PDAC is also associated with systemically deregulated glucose homeostasis (i.e., hyperinsulinemia/hyperglycemia, which occurs in a majority of PDAC patients, and frank diabetes in up to 50% of the cases) (Permert et al., 1993; Permert et al., 1997; Stolzenberg-Solomon et al., 2005; Sah et al., 2013; Andersen et al., 2017; Risch, 2019; Deng et al., 2022). Of note, among at least six cancer types attributable to diabetes (Pearson-Stuttard et al., 2018), PDAC and diabetes exhibit a unique bi-directional relationship: epidemiological data point to PDAC as a cause of diabetes; conversely, recent reports suggested that diabetic state sustains the aggressiveness of pancreatic tumors and their resistance to therapies (Stolzenberg-Solomon et al., 2005; Sah et al., 2013; Andersen et al., 2017; Risch, 2019; Deng et al., 2022). Significant progress has been made in deciphering the role of oncogenic signaling and hypoxic/desmoplastic microenvironment in rendering PDAC highly aggressive and resistant to treatment (Ying et al., 2012; Lambert et al., 2019). Less appreciated are mechanisms linking metabolic dysregulation and ECM enzymatic remodeling in PDAC to its progression/therapy resistance.

We have recently demonstrated that a diabetic state induces overexpression of ECM-degrading enzyme heparanase in PDAC (Goldberg et al., 2019). Heparanase is the sole mammalian endoglycosidase capable of hydrolyzing glycosaminoglycan heparan sulfate [HS] in the extracellular space and on the cell surface. Heparanase enzyme is therefore critically involved in the integrity of the ECM, as well as in the modulation of interactions between HS and numerous bioactive molecules (including growth factors and their receptors) at the cell surface (Jayatilleke and Hulett, 2020; Masola et al., 2020; Vlodavsky et al., 2023). In addition, heparanase was shown to stimulate signal transduction pathways, both via its enzymatic, as well as non-enzymatic activities [reviewed in (Vlodavsky et al., 2023)]. Several studies implicated heparanase in pancreatic tumorigenesis, reporting overexpression of the enzyme in PDAC, and its correlation with poor prognosis (Koliopanos et al., 2001; Rohloff et al., 2002; Quiros et al., 2006; Hoffmann et al., 2008). Based on these findings, the inducibility of heparanase expression in PDAC by diabetic milieu components (Goldberg et al., 2019), and impairment of insulin-triggered insulin receptor (INSR) activation under conditions of heparanase deficiency/enzymatic inhibition (Cassinelli et al., 2018; Hermano et al., 2021), we hypothesized that the overexpression of the enzyme can contribute to PDAC pathogenesis by increasing insulin sensitivity of the carcinoma cells, thus mechanistically linking systemic diabetic state to the intracellular metabolic rewiring. Here we report that in the context of hyperinsulinemic conditions [a common phenomenon in PDAC, either in the setting of new onset or long-standing diabetes (Permert et al., 1993; Permert et al., 1997; Stolzenberg-Solomon et al., 2005; Sah et al., 2013)], heparanase augments insulin-INSR signaling cascade in pancreatic tumor cells. We found that elevated levels of heparanase in PDAC tumors when coupled to abundant insulin and glucose content, can empower signal transduction along the insulin-INSR-AKT pathway, promoting PDAC cell proliferation and rendering them resistant to gemcitabine. The underlying molecular events involve augmented glucose uptake by cancer cells due to insulin-dependent translocation of glucose transporter GLUT4 to the plasma membrane, highlighting a previously unknown mechanism through which heparanase enzyme acts at the interface of systemic and intracellular metabolic alterations typical for PDAC.

In light of previously reported impairment of INSR signaling under conditions of heparanase deficiency or enzymatic inhibition (Goldberg et al., 2017; Purushothaman et al., 2017; Cassinelli et al., 2018; Hermano et al., 2021), we first tested whether elevated levels of heparanase enhance INSR signal transduction in prototypic insulin-target tissue (i.e., muscle). For this purpose we utilized soleus skeletal muscle samples derived from heparanase-transgenic mice, overexpressing human heparanase in all tissues (Zcharia et al., 2004), and their wt counterparts. As shown in Figure 1, analysis of INSR signaling cascade activation 10 min after i. p. insulin administration in lysates of soleus skeletal muscle samples harvested from heparanase-overexpressing transgenic mice vs. wt mice revealed that heparanase overexpression enhanced insulin signaling, as evidenced by increased phospho-INSR [pINSR] and phospho-AKT [pAKT] (Figures 1A–C) levels in soleus skeletal muscle derived from heparanase-transgenic, as compared to wt mice. Notably, no difference between protein levels of total INSR and AKT was detected between the muscles of transgenic and wt mice prior to insulin administration (Figures 1D, E).

Augmented glucose influx is an essential prerequisite for its increased fermentation in cancer cells via the glycolytic pathway (i.e., Warburg effect), well-documented in PDAC, as well as other types of malignancies (Ying et al., 2012; Shukla et al., 2017; Ancey et al., 2018; Hoxhaj and Manning, 2020). Notably, in PDAC patients this metabolic shift often occurs in the face of systemic deterioration in glycemic control manifested by hyperinsulinemia/hyperglycemia (Permert et al., 1993; Permert et al., 1997; Stolzenberg-Solomon et al., 2005; Sah et al., 2013; Andersen et al., 2017; Risch, 2019; Deng et al., 2022). Furthermore, while several types of malignancies have been associated with impaired glucose metabolism (Pearson-Stuttard et al., 2018), relationships between PDAC and diabetes are unique due to the postulated reverse causality between these two devastating diseases. Indeed, PDAC is considered a cause of diabetes in a significant fraction of patients; in turn, diabetic conditions reportedly promote pancreatic tumorigenesis and resistance to treatment (Stolzenberg-Solomon et al., 2005; Sah et al., 2013; Andersen et al., 2017; Risch, 2019; Deng et al., 2022). Nevertheless, the mechanistic data concerning relationships between PDAC progression and systemic deterioration of glycemic control, as well as the influence of these relationships on metabolic alterations within PDAC cells are rather scarce. Moreover, the contribution of ECM-degrading enzymes to these relationships was not previously investigated. Here we show that in PDAC elevated levels of heparanase [the predominant mammalian enzyme that degrades HS glycosaminoglycan at the ECM/cell surface (Jayatilleke and Hulett, 2020)], coupled to diabetic conditions typical for PDAC (Stolzenberg-Solomon et al., 2005; Dey et al., 2012; Sah et al., 2013; Andersen et al., 2017; Cheng et al., 2018; Krapf et al., 2020; Deng et al., 2022), promote growth and therapy resistance of pancreatic carcinoma cells by fostering INSR signaling and GLUT4 mediated glucose uptake.

In fact, the involvement of heparanase in diabetes and its complications was actively investigated during the last decade (Baker et al., 2010; Gil et al., 2012; Goldberg et al., 2014; Abu El-Asrar et al., 2015; Simeonovic et al., 2020). Previous findings demonstrated essential role of the enzyme in ensuring insulin sensitivity in the insulin-target tissues [i.e., muscle, (Hermano et al., 2021)]. A number of studies also linked heparanase activity to the augmented INSR signaling in several types of malignant tumors (i.e., myeloma, breast carcinoma, synovial sarcoma) (Goldberg et al., 2017; Purushothaman et al., 2017; Cassinelli et al., 2018). Important to note that under certain experimental conditions heparanase can also exert non-enzymatic effects, playing a part in the activation of selected signaling molecules, including AKT, Src, and EGFR (Vlodavsky et al., 2023). Of those the best documented is the induction of AKT phosphorylation [reviewed in (Vlodavsky et al., 2023)]. Since AKT is a key downstream mediator of insulin-INSR signal transduction and GLUT4 translocation, it might be anticipated that augmentation of insulin signaling by heparanase is influenced by its non-enzymatic action as well. Interestingly, the previous studies focusing on heparanase-facilitated insulin responses in various experimental systems reported that enzymatic, rather than non-enzymatic action of heparanase is responsible for this phenomenon, as enzymatically-inactive forms of heparanase had no effect on signal transduction events along insulin/INSR axis, and enzymatic inhibitors of heparanase diminished heparanase-facilitated response to insulin (Goldberg et al., 2017; Purushothaman et al., 2017; Cassinelli et al., 2018). Noteworthy, we were unable to detect changes in phospho-AKT levels in heparanase-overexpressing tissue (Figure 1B) and cell (Figure 2B) lysates in the absence of insulin stimulation. Nevertheless, given the previously documented role of the signaling molecules (such as Src, EGFR) in metabolic plasticity of the cancer cell [i.e., through regulation of master glycolytic transcription factors, expression of insulin-independent glucose transporters, modulation of glycolytic enzyme activity (Sigismund et al., 2018; Pelaz and Tabernero, 2022)], along with the ability of heparanase to affect activation status of these signaling molecules in enzymatic-independent fashion, the contribution of non-enzymatic effects of heparanase to the metabolic alterations occurring in PDAC tumorigenesis could not be excluded.

As stated above, connection between PDAC and diabetes is distinctive in its reciprocality, contrasting other diabetes-influenced cancer types (Pearson-Stuttard et al., 2018). While the diabetes-promoting action of PDAC is explained by several studies as a paraneoplastic phenomenon caused by specific PDAC-secreted mediators (Sah et al., 2013; Batabyal et al., 2014; Javeed et al., 2015; Sagar et al., 2016; Andersen et al., 2017), our present findings provide an explanation for the PDAC-promoting action of diabetes, emphasizing the role of heparanase. Indeed, we documented an increase in signal transduction response to insulin along the INSR-AKT axis in heparanase-overexpressing muscle tissue (prototypic insulin target organ, Figure 1), and in a similar manner–in heparanase-overexpressing PDAC cells (Figure 2). This increase, and the resulting proliferative response of PDAC cells (Figure 4) was attenuated in the presence of specific inhibitor of the enzyme (Figures 3, 4B, C). Importantly, we detected that heparanase-driven increase in insulin-triggered signaling events within PDAC cells rendered them resistant to gemcitabine treatment (Figure 5) and revealed that underlying molecular events involve enhanced intracellular uptake of glucose (Figure 6) due to membrane translocation of insulin-dependent GLUT4 transporter (Figure 7). When taken together with the previous observations that heparanase overexpression in PDAC is induced by diabetic milieu components (Goldberg et al., 2019), these findings provide new insight into the role of the enzyme in shaping bi-directional connection between PDAC progression and intratumoral/systemic metabolic changes. Besides induction of heparanase in pancreatic tumor cells due to PDAC-triggered impairment of glucose homeostasis (Goldberg et al., 2019), this connection involves heparanase-empowered activation of the INSR-AKT signaling cascade. Together with an additional defining element of PDAC, i.e., increased insulin levels, this activation allows augmented uptake of glucose by the tumor cells through GLUT4 membrane translocation, thus promoting growth and chemoresistance of PDAC. However, the limitation of our study is that the precise mode of the enzyme action in fostering INSR signaling remained largely unexplained. The ability of heparin/HS (including heparanase-generated soluble HS fragments) to promote/stabilize growth factor tyrosine kinase receptor-ligand binding, dimerization and signal transduction (Spivak-Kroizman et al., 1994; Vlodavsky et al., 1996; Elkin et al., 2001; Derksen et al., 2002; Kemp et al., 2006; Lemmon and Schlessinger, 2010) provides one conceivable explanation, suggesting that in the similar manner, HS degradation fragments released by the enzyme facilitate insulin-INSR signaling complexes. Of note, the key extracellular HS proteoglycan perlecan was reported to modulate insulin responses, as loss of perlecan results in augmented insulin sensitivity in muscle (Yamashita et al., 2018). Given the ability of heparanase to reduce the amount of extracellular HS (Elkin et al., 2001; Hermano et al., 2014), perlecan degradation by the enzyme can explain increased insulin sensitivity in heparanase-overexpressing PDAC cells. Additionally, cell-surface HS proteoglycan glypican-4 reportedly enhances insulin signaling via interaction with the INSR (Ussar et al., 2014). As enzymatically-regulated release of glypican from the cell surface appears to be required to enable its direct interaction with INSR (Ussar et al., 2014), it is plausible that heparanase [known to drive shedding of cell surface HS proteoglycans (Ramani et al., 2013)] can facilitate this process. Clearly, further studies are warranted to fully characterize molecular events underlying heparanase-augmented insulin signaling. An additional limitation of the present study is that no in vivo model was used to validate the effect of heparanase-augmented insulin signaling on the chemo-resistance of PDAC cells to gemcitabine. This is due to the lack of immunodeficient mouse host models faithfully reflecting concurrent chronic hyperinsulinemia and PDAC progression. Yet, our present findings attest heparanase enzyme as an important and potentially modifiable contributor to progression/chemo-resistance of PDAC, acting along INSR-AKT-GLUT4 axis. Moreover, as the recent attempts to translate direct targeting of AKT pathway into clinic were proven to be highly challenging (He et al., 2021), heparanase-inhibiting approaches [which are currently under active development (Rivara et al., 2016; He et al., 2022)] may offer an appealing tactic to sensitize pancreatic tumors to gemcitabine, as well as uncouple PDAC and diabetes cancer progression in a significant fraction of patients.